Adding upstream version 5.10.209.upstream/5.10.209upstream

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

118 files changed, 40297 insertions, 0 deletions

diff --git a/arch/mips/kernel/.gitignore b/arch/mips/kernel/.gitignore
new file mode 100644
index 000000000..bbb90f92d
--- /dev/null
+++ b/arch/mips/kernel/.gitignore
@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
+vmlinux.lds
diff --git a/arch/mips/kernel/Makefile b/arch/mips/kernel/Makefile
new file mode 100644
index 000000000..2a05b923f
--- /dev/null
+++ b/arch/mips/kernel/Makefile
@@ -0,0 +1,119 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Makefile for the Linux/MIPS kernel.
+#
+
+extra-y		:= head.o vmlinux.lds
+
+obj-y		+= branch.o cmpxchg.o elf.o entry.o genex.o idle.o irq.o \
+		   process.o prom.o ptrace.o reset.o setup.o signal.o \
+		   syscall.o time.o topology.o traps.o unaligned.o watch.o \
+		   vdso.o cacheinfo.o
+
+ifdef CONFIG_CPU_R3K_TLB
+obj-y		+= cpu-r3k-probe.o
+else
+obj-y		+= cpu-probe.o
+endif
+
+ifdef CONFIG_FUNCTION_TRACER
+CFLAGS_REMOVE_ftrace.o = -pg
+CFLAGS_REMOVE_early_printk.o = -pg
+CFLAGS_REMOVE_perf_event.o = -pg
+CFLAGS_REMOVE_perf_event_mipsxx.o = -pg
+endif
+
+obj-$(CONFIG_CEVT_BCM1480)	+= cevt-bcm1480.o
+obj-$(CONFIG_CEVT_R4K)		+= cevt-r4k.o
+obj-$(CONFIG_CEVT_DS1287)	+= cevt-ds1287.o
+obj-$(CONFIG_CEVT_GT641XX)	+= cevt-gt641xx.o
+obj-$(CONFIG_CEVT_SB1250)	+= cevt-sb1250.o
+obj-$(CONFIG_CEVT_TXX9)		+= cevt-txx9.o
+obj-$(CONFIG_CSRC_BCM1480)	+= csrc-bcm1480.o
+obj-$(CONFIG_CSRC_IOASIC)	+= csrc-ioasic.o
+obj-$(CONFIG_CSRC_R4K)		+= csrc-r4k.o
+obj-$(CONFIG_CSRC_SB1250)	+= csrc-sb1250.o
+obj-$(CONFIG_SYNC_R4K)		+= sync-r4k.o
+
+obj-$(CONFIG_DEBUG_FS)		+= segment.o
+obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
+obj-$(CONFIG_MODULES)		+= module.o
+
+obj-$(CONFIG_FTRACE_SYSCALLS)	+= ftrace.o
+obj-$(CONFIG_FUNCTION_TRACER)	+= mcount.o ftrace.o
+
+sw-y				:= r4k_switch.o
+sw-$(CONFIG_CPU_R3000)		:= r2300_switch.o
+sw-$(CONFIG_CPU_TX39XX)		:= r2300_switch.o
+sw-$(CONFIG_CPU_CAVIUM_OCTEON)	:= octeon_switch.o
+obj-y				+= $(sw-y)
+
+obj-$(CONFIG_MIPS_FP_SUPPORT)	+= fpu-probe.o
+obj-$(CONFIG_CPU_R2300_FPU)	+= r2300_fpu.o
+obj-$(CONFIG_CPU_R4K_FPU)	+= r4k_fpu.o
+
+obj-$(CONFIG_SMP)		+= smp.o
+obj-$(CONFIG_SMP_UP)		+= smp-up.o
+obj-$(CONFIG_CPU_BMIPS)		+= smp-bmips.o bmips_vec.o bmips_5xxx_init.o
+
+obj-$(CONFIG_MIPS_MT)		+= mips-mt.o
+obj-$(CONFIG_MIPS_MT_FPAFF)	+= mips-mt-fpaff.o
+obj-$(CONFIG_MIPS_MT_SMP)	+= smp-mt.o
+obj-$(CONFIG_MIPS_CMP)		+= smp-cmp.o
+obj-$(CONFIG_MIPS_CPS)		+= smp-cps.o cps-vec.o
+obj-$(CONFIG_MIPS_CPS_NS16550)	+= cps-vec-ns16550.o
+obj-$(CONFIG_MIPS_SPRAM)	+= spram.o
+
+obj-$(CONFIG_MIPS_VPE_LOADER)	+= vpe.o
+obj-$(CONFIG_MIPS_VPE_LOADER_CMP) += vpe-cmp.o
+obj-$(CONFIG_MIPS_VPE_LOADER_MT) += vpe-mt.o
+obj-$(CONFIG_MIPS_VPE_APSP_API) += rtlx.o
+obj-$(CONFIG_MIPS_VPE_APSP_API_CMP) += rtlx-cmp.o
+obj-$(CONFIG_MIPS_VPE_APSP_API_MT) += rtlx-mt.o
+
+obj-$(CONFIG_IRQ_CPU_RM7K)	+= irq-rm7000.o
+obj-$(CONFIG_MIPS_MSC)		+= irq-msc01.o
+obj-$(CONFIG_IRQ_TXX9)		+= irq_txx9.o
+obj-$(CONFIG_IRQ_GT641XX)	+= irq-gt641xx.o
+
+obj-$(CONFIG_KPROBES)		+= kprobes.o
+obj-$(CONFIG_32BIT)		+= scall32-o32.o
+obj-$(CONFIG_64BIT)		+= scall64-n64.o
+obj-$(CONFIG_MIPS32_COMPAT)	+= linux32.o ptrace32.o signal32.o
+obj-$(CONFIG_MIPS32_N32)	+= binfmt_elfn32.o scall64-n32.o signal_n32.o
+obj-$(CONFIG_MIPS32_O32)	+= binfmt_elfo32.o scall64-o32.o signal_o32.o
+
+obj-$(CONFIG_KGDB)		+= kgdb.o
+obj-$(CONFIG_PROC_FS)		+= proc.o
+obj-$(CONFIG_MAGIC_SYSRQ)	+= sysrq.o
+
+obj-$(CONFIG_CPU_R4X00_BUGS64)	+= r4k-bugs64.o
+
+obj-$(CONFIG_I8253)		+= i8253.o
+
+obj-$(CONFIG_GPIO_TXX9)		+= gpio_txx9.o
+
+obj-$(CONFIG_RELOCATABLE)	+= relocate.o
+
+obj-$(CONFIG_KEXEC)		+= machine_kexec.o relocate_kernel.o crash.o
+obj-$(CONFIG_CRASH_DUMP)	+= crash_dump.o
+obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
+obj-$(CONFIG_EARLY_PRINTK_8250)	+= early_printk_8250.o
+obj-$(CONFIG_SPINLOCK_TEST)	+= spinlock_test.o
+obj-$(CONFIG_MIPSR2_TO_R6_EMULATOR)	+= mips-r2-to-r6-emul.o
+
+CFLAGS_cpu-bugs64.o	= $(shell if $(CC) $(KBUILD_CFLAGS) -Wa,-mdaddi -c -o /dev/null -x c /dev/null >/dev/null 2>&1; then echo "-DHAVE_AS_SET_DADDI"; fi)
+
+obj-$(CONFIG_PERF_EVENTS)	+= perf_event.o
+obj-$(CONFIG_HW_PERF_EVENTS)	+= perf_event_mipsxx.o
+
+obj-$(CONFIG_JUMP_LABEL)	+= jump_label.o
+obj-$(CONFIG_UPROBES)		+= uprobes.o
+
+obj-$(CONFIG_MIPS_CM)		+= mips-cm.o
+obj-$(CONFIG_MIPS_CPC)		+= mips-cpc.o
+
+obj-$(CONFIG_CPU_PM)		+= pm.o
+obj-$(CONFIG_MIPS_CPS_PM)	+= pm-cps.o
+
+CPPFLAGS_vmlinux.lds		:= $(KBUILD_CFLAGS)
diff --git a/arch/mips/kernel/asm-offsets.c b/arch/mips/kernel/asm-offsets.c
new file mode 100644
index 000000000..aebfda811
--- /dev/null
+++ b/arch/mips/kernel/asm-offsets.c
@@ -0,0 +1,405 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * asm-offsets.c: Calculate pt_regs and task_struct offsets.
+ *
+ * Copyright (C) 1996 David S. Miller
+ * Copyright (C) 1997, 1998, 1999, 2000, 2001, 2002, 2003 Ralf Baechle
+ * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ *
+ * Kevin Kissell, kevink@mips.com and Carsten Langgaard, carstenl@mips.com
+ * Copyright (C) 2000 MIPS Technologies, Inc.
+ */
+#include <linux/compat.h>
+#include <linux/types.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/kbuild.h>
+#include <linux/suspend.h>
+#include <asm/cpu-info.h>
+#include <asm/pm.h>
+#include <asm/ptrace.h>
+#include <asm/processor.h>
+#include <asm/smp-cps.h>
+
+#include <linux/kvm_host.h>
+
+void output_ptreg_defines(void)
+{
+	COMMENT("MIPS pt_regs offsets.");
+	OFFSET(PT_R0, pt_regs, regs[0]);
+	OFFSET(PT_R1, pt_regs, regs[1]);
+	OFFSET(PT_R2, pt_regs, regs[2]);
+	OFFSET(PT_R3, pt_regs, regs[3]);
+	OFFSET(PT_R4, pt_regs, regs[4]);
+	OFFSET(PT_R5, pt_regs, regs[5]);
+	OFFSET(PT_R6, pt_regs, regs[6]);
+	OFFSET(PT_R7, pt_regs, regs[7]);
+	OFFSET(PT_R8, pt_regs, regs[8]);
+	OFFSET(PT_R9, pt_regs, regs[9]);
+	OFFSET(PT_R10, pt_regs, regs[10]);
+	OFFSET(PT_R11, pt_regs, regs[11]);
+	OFFSET(PT_R12, pt_regs, regs[12]);
+	OFFSET(PT_R13, pt_regs, regs[13]);
+	OFFSET(PT_R14, pt_regs, regs[14]);
+	OFFSET(PT_R15, pt_regs, regs[15]);
+	OFFSET(PT_R16, pt_regs, regs[16]);
+	OFFSET(PT_R17, pt_regs, regs[17]);
+	OFFSET(PT_R18, pt_regs, regs[18]);
+	OFFSET(PT_R19, pt_regs, regs[19]);
+	OFFSET(PT_R20, pt_regs, regs[20]);
+	OFFSET(PT_R21, pt_regs, regs[21]);
+	OFFSET(PT_R22, pt_regs, regs[22]);
+	OFFSET(PT_R23, pt_regs, regs[23]);
+	OFFSET(PT_R24, pt_regs, regs[24]);
+	OFFSET(PT_R25, pt_regs, regs[25]);
+	OFFSET(PT_R26, pt_regs, regs[26]);
+	OFFSET(PT_R27, pt_regs, regs[27]);
+	OFFSET(PT_R28, pt_regs, regs[28]);
+	OFFSET(PT_R29, pt_regs, regs[29]);
+	OFFSET(PT_R30, pt_regs, regs[30]);
+	OFFSET(PT_R31, pt_regs, regs[31]);
+	OFFSET(PT_LO, pt_regs, lo);
+	OFFSET(PT_HI, pt_regs, hi);
+#ifdef CONFIG_CPU_HAS_SMARTMIPS
+	OFFSET(PT_ACX, pt_regs, acx);
+#endif
+	OFFSET(PT_EPC, pt_regs, cp0_epc);
+	OFFSET(PT_BVADDR, pt_regs, cp0_badvaddr);
+	OFFSET(PT_STATUS, pt_regs, cp0_status);
+	OFFSET(PT_CAUSE, pt_regs, cp0_cause);
+#ifdef CONFIG_CPU_CAVIUM_OCTEON
+	OFFSET(PT_MPL, pt_regs, mpl);
+	OFFSET(PT_MTP, pt_regs, mtp);
+#endif /* CONFIG_CPU_CAVIUM_OCTEON */
+	DEFINE(PT_SIZE, sizeof(struct pt_regs));
+	BLANK();
+}
+
+void output_task_defines(void)
+{
+	COMMENT("MIPS task_struct offsets.");
+	OFFSET(TASK_STATE, task_struct, state);
+	OFFSET(TASK_THREAD_INFO, task_struct, stack);
+	OFFSET(TASK_FLAGS, task_struct, flags);
+	OFFSET(TASK_MM, task_struct, mm);
+	OFFSET(TASK_PID, task_struct, pid);
+#if defined(CONFIG_STACKPROTECTOR)
+	OFFSET(TASK_STACK_CANARY, task_struct, stack_canary);
+#endif
+	DEFINE(TASK_STRUCT_SIZE, sizeof(struct task_struct));
+	BLANK();
+}
+
+void output_thread_info_defines(void)
+{
+	COMMENT("MIPS thread_info offsets.");
+	OFFSET(TI_TASK, thread_info, task);
+	OFFSET(TI_FLAGS, thread_info, flags);
+	OFFSET(TI_TP_VALUE, thread_info, tp_value);
+	OFFSET(TI_CPU, thread_info, cpu);
+	OFFSET(TI_PRE_COUNT, thread_info, preempt_count);
+	OFFSET(TI_ADDR_LIMIT, thread_info, addr_limit);
+	OFFSET(TI_REGS, thread_info, regs);
+	DEFINE(_THREAD_SIZE, THREAD_SIZE);
+	DEFINE(_THREAD_MASK, THREAD_MASK);
+	DEFINE(_IRQ_STACK_SIZE, IRQ_STACK_SIZE);
+	DEFINE(_IRQ_STACK_START, IRQ_STACK_START);
+	BLANK();
+}
+
+void output_thread_defines(void)
+{
+	COMMENT("MIPS specific thread_struct offsets.");
+	OFFSET(THREAD_REG16, task_struct, thread.reg16);
+	OFFSET(THREAD_REG17, task_struct, thread.reg17);
+	OFFSET(THREAD_REG18, task_struct, thread.reg18);
+	OFFSET(THREAD_REG19, task_struct, thread.reg19);
+	OFFSET(THREAD_REG20, task_struct, thread.reg20);
+	OFFSET(THREAD_REG21, task_struct, thread.reg21);
+	OFFSET(THREAD_REG22, task_struct, thread.reg22);
+	OFFSET(THREAD_REG23, task_struct, thread.reg23);
+	OFFSET(THREAD_REG29, task_struct, thread.reg29);
+	OFFSET(THREAD_REG30, task_struct, thread.reg30);
+	OFFSET(THREAD_REG31, task_struct, thread.reg31);
+	OFFSET(THREAD_STATUS, task_struct,
+	       thread.cp0_status);
+
+	OFFSET(THREAD_BVADDR, task_struct, \
+	       thread.cp0_badvaddr);
+	OFFSET(THREAD_BUADDR, task_struct, \
+	       thread.cp0_baduaddr);
+	OFFSET(THREAD_ECODE, task_struct, \
+	       thread.error_code);
+	OFFSET(THREAD_TRAPNO, task_struct, thread.trap_nr);
+	BLANK();
+}
+
+#ifdef CONFIG_MIPS_FP_SUPPORT
+void output_thread_fpu_defines(void)
+{
+	OFFSET(THREAD_FPU, task_struct, thread.fpu);
+
+	OFFSET(THREAD_FPR0, task_struct, thread.fpu.fpr[0]);
+	OFFSET(THREAD_FPR1, task_struct, thread.fpu.fpr[1]);
+	OFFSET(THREAD_FPR2, task_struct, thread.fpu.fpr[2]);
+	OFFSET(THREAD_FPR3, task_struct, thread.fpu.fpr[3]);
+	OFFSET(THREAD_FPR4, task_struct, thread.fpu.fpr[4]);
+	OFFSET(THREAD_FPR5, task_struct, thread.fpu.fpr[5]);
+	OFFSET(THREAD_FPR6, task_struct, thread.fpu.fpr[6]);
+	OFFSET(THREAD_FPR7, task_struct, thread.fpu.fpr[7]);
+	OFFSET(THREAD_FPR8, task_struct, thread.fpu.fpr[8]);
+	OFFSET(THREAD_FPR9, task_struct, thread.fpu.fpr[9]);
+	OFFSET(THREAD_FPR10, task_struct, thread.fpu.fpr[10]);
+	OFFSET(THREAD_FPR11, task_struct, thread.fpu.fpr[11]);
+	OFFSET(THREAD_FPR12, task_struct, thread.fpu.fpr[12]);
+	OFFSET(THREAD_FPR13, task_struct, thread.fpu.fpr[13]);
+	OFFSET(THREAD_FPR14, task_struct, thread.fpu.fpr[14]);
+	OFFSET(THREAD_FPR15, task_struct, thread.fpu.fpr[15]);
+	OFFSET(THREAD_FPR16, task_struct, thread.fpu.fpr[16]);
+	OFFSET(THREAD_FPR17, task_struct, thread.fpu.fpr[17]);
+	OFFSET(THREAD_FPR18, task_struct, thread.fpu.fpr[18]);
+	OFFSET(THREAD_FPR19, task_struct, thread.fpu.fpr[19]);
+	OFFSET(THREAD_FPR20, task_struct, thread.fpu.fpr[20]);
+	OFFSET(THREAD_FPR21, task_struct, thread.fpu.fpr[21]);
+	OFFSET(THREAD_FPR22, task_struct, thread.fpu.fpr[22]);
+	OFFSET(THREAD_FPR23, task_struct, thread.fpu.fpr[23]);
+	OFFSET(THREAD_FPR24, task_struct, thread.fpu.fpr[24]);
+	OFFSET(THREAD_FPR25, task_struct, thread.fpu.fpr[25]);
+	OFFSET(THREAD_FPR26, task_struct, thread.fpu.fpr[26]);
+	OFFSET(THREAD_FPR27, task_struct, thread.fpu.fpr[27]);
+	OFFSET(THREAD_FPR28, task_struct, thread.fpu.fpr[28]);
+	OFFSET(THREAD_FPR29, task_struct, thread.fpu.fpr[29]);
+	OFFSET(THREAD_FPR30, task_struct, thread.fpu.fpr[30]);
+	OFFSET(THREAD_FPR31, task_struct, thread.fpu.fpr[31]);
+
+	OFFSET(THREAD_FCR31, task_struct, thread.fpu.fcr31);
+	OFFSET(THREAD_MSA_CSR, task_struct, thread.fpu.msacsr);
+	BLANK();
+}
+#endif
+
+void output_mm_defines(void)
+{
+	COMMENT("Size of struct page");
+	DEFINE(STRUCT_PAGE_SIZE, sizeof(struct page));
+	BLANK();
+	COMMENT("Linux mm_struct offsets.");
+	OFFSET(MM_USERS, mm_struct, mm_users);
+	OFFSET(MM_PGD, mm_struct, pgd);
+	OFFSET(MM_CONTEXT, mm_struct, context);
+	BLANK();
+	DEFINE(_PGD_T_SIZE, sizeof(pgd_t));
+	DEFINE(_PMD_T_SIZE, sizeof(pmd_t));
+	DEFINE(_PTE_T_SIZE, sizeof(pte_t));
+	BLANK();
+	DEFINE(_PGD_T_LOG2, PGD_T_LOG2);
+#ifndef __PAGETABLE_PMD_FOLDED
+	DEFINE(_PMD_T_LOG2, PMD_T_LOG2);
+#endif
+	DEFINE(_PTE_T_LOG2, PTE_T_LOG2);
+	BLANK();
+	DEFINE(_PGD_ORDER, PGD_ORDER);
+#ifndef __PAGETABLE_PMD_FOLDED
+	DEFINE(_PMD_ORDER, PMD_ORDER);
+#endif
+	DEFINE(_PTE_ORDER, PTE_ORDER);
+	BLANK();
+	DEFINE(_PMD_SHIFT, PMD_SHIFT);
+	DEFINE(_PGDIR_SHIFT, PGDIR_SHIFT);
+	BLANK();
+	DEFINE(_PTRS_PER_PGD, PTRS_PER_PGD);
+	DEFINE(_PTRS_PER_PMD, PTRS_PER_PMD);
+	DEFINE(_PTRS_PER_PTE, PTRS_PER_PTE);
+	BLANK();
+	DEFINE(_PAGE_SHIFT, PAGE_SHIFT);
+	DEFINE(_PAGE_SIZE, PAGE_SIZE);
+	BLANK();
+}
+
+#ifdef CONFIG_32BIT
+void output_sc_defines(void)
+{
+	COMMENT("Linux sigcontext offsets.");
+	OFFSET(SC_REGS, sigcontext, sc_regs);
+	OFFSET(SC_FPREGS, sigcontext, sc_fpregs);
+	OFFSET(SC_ACX, sigcontext, sc_acx);
+	OFFSET(SC_MDHI, sigcontext, sc_mdhi);
+	OFFSET(SC_MDLO, sigcontext, sc_mdlo);
+	OFFSET(SC_PC, sigcontext, sc_pc);
+	OFFSET(SC_FPC_CSR, sigcontext, sc_fpc_csr);
+	OFFSET(SC_FPC_EIR, sigcontext, sc_fpc_eir);
+	OFFSET(SC_HI1, sigcontext, sc_hi1);
+	OFFSET(SC_LO1, sigcontext, sc_lo1);
+	OFFSET(SC_HI2, sigcontext, sc_hi2);
+	OFFSET(SC_LO2, sigcontext, sc_lo2);
+	OFFSET(SC_HI3, sigcontext, sc_hi3);
+	OFFSET(SC_LO3, sigcontext, sc_lo3);
+	BLANK();
+}
+#endif
+
+#ifdef CONFIG_64BIT
+void output_sc_defines(void)
+{
+	COMMENT("Linux sigcontext offsets.");
+	OFFSET(SC_REGS, sigcontext, sc_regs);
+	OFFSET(SC_FPREGS, sigcontext, sc_fpregs);
+	OFFSET(SC_MDHI, sigcontext, sc_mdhi);
+	OFFSET(SC_MDLO, sigcontext, sc_mdlo);
+	OFFSET(SC_PC, sigcontext, sc_pc);
+	OFFSET(SC_FPC_CSR, sigcontext, sc_fpc_csr);
+	BLANK();
+}
+#endif
+
+void output_signal_defined(void)
+{
+	COMMENT("Linux signal numbers.");
+	DEFINE(_SIGHUP, SIGHUP);
+	DEFINE(_SIGINT, SIGINT);
+	DEFINE(_SIGQUIT, SIGQUIT);
+	DEFINE(_SIGILL, SIGILL);
+	DEFINE(_SIGTRAP, SIGTRAP);
+	DEFINE(_SIGIOT, SIGIOT);
+	DEFINE(_SIGABRT, SIGABRT);
+	DEFINE(_SIGEMT, SIGEMT);
+	DEFINE(_SIGFPE, SIGFPE);
+	DEFINE(_SIGKILL, SIGKILL);
+	DEFINE(_SIGBUS, SIGBUS);
+	DEFINE(_SIGSEGV, SIGSEGV);
+	DEFINE(_SIGSYS, SIGSYS);
+	DEFINE(_SIGPIPE, SIGPIPE);
+	DEFINE(_SIGALRM, SIGALRM);
+	DEFINE(_SIGTERM, SIGTERM);
+	DEFINE(_SIGUSR1, SIGUSR1);
+	DEFINE(_SIGUSR2, SIGUSR2);
+	DEFINE(_SIGCHLD, SIGCHLD);
+	DEFINE(_SIGPWR, SIGPWR);
+	DEFINE(_SIGWINCH, SIGWINCH);
+	DEFINE(_SIGURG, SIGURG);
+	DEFINE(_SIGIO, SIGIO);
+	DEFINE(_SIGSTOP, SIGSTOP);
+	DEFINE(_SIGTSTP, SIGTSTP);
+	DEFINE(_SIGCONT, SIGCONT);
+	DEFINE(_SIGTTIN, SIGTTIN);
+	DEFINE(_SIGTTOU, SIGTTOU);
+	DEFINE(_SIGVTALRM, SIGVTALRM);
+	DEFINE(_SIGPROF, SIGPROF);
+	DEFINE(_SIGXCPU, SIGXCPU);
+	DEFINE(_SIGXFSZ, SIGXFSZ);
+	BLANK();
+}
+
+#ifdef CONFIG_CPU_CAVIUM_OCTEON
+void output_octeon_cop2_state_defines(void)
+{
+	COMMENT("Octeon specific octeon_cop2_state offsets.");
+	OFFSET(OCTEON_CP2_CRC_IV,	octeon_cop2_state, cop2_crc_iv);
+	OFFSET(OCTEON_CP2_CRC_LENGTH,	octeon_cop2_state, cop2_crc_length);
+	OFFSET(OCTEON_CP2_CRC_POLY,	octeon_cop2_state, cop2_crc_poly);
+	OFFSET(OCTEON_CP2_LLM_DAT,	octeon_cop2_state, cop2_llm_dat);
+	OFFSET(OCTEON_CP2_3DES_IV,	octeon_cop2_state, cop2_3des_iv);
+	OFFSET(OCTEON_CP2_3DES_KEY,	octeon_cop2_state, cop2_3des_key);
+	OFFSET(OCTEON_CP2_3DES_RESULT,	octeon_cop2_state, cop2_3des_result);
+	OFFSET(OCTEON_CP2_AES_INP0,	octeon_cop2_state, cop2_aes_inp0);
+	OFFSET(OCTEON_CP2_AES_IV,	octeon_cop2_state, cop2_aes_iv);
+	OFFSET(OCTEON_CP2_AES_KEY,	octeon_cop2_state, cop2_aes_key);
+	OFFSET(OCTEON_CP2_AES_KEYLEN,	octeon_cop2_state, cop2_aes_keylen);
+	OFFSET(OCTEON_CP2_AES_RESULT,	octeon_cop2_state, cop2_aes_result);
+	OFFSET(OCTEON_CP2_GFM_MULT,	octeon_cop2_state, cop2_gfm_mult);
+	OFFSET(OCTEON_CP2_GFM_POLY,	octeon_cop2_state, cop2_gfm_poly);
+	OFFSET(OCTEON_CP2_GFM_RESULT,	octeon_cop2_state, cop2_gfm_result);
+	OFFSET(OCTEON_CP2_HSH_DATW,	octeon_cop2_state, cop2_hsh_datw);
+	OFFSET(OCTEON_CP2_HSH_IVW,	octeon_cop2_state, cop2_hsh_ivw);
+	OFFSET(OCTEON_CP2_SHA3,		octeon_cop2_state, cop2_sha3);
+	OFFSET(THREAD_CP2,	task_struct, thread.cp2);
+	OFFSET(THREAD_CVMSEG,	task_struct, thread.cvmseg.cvmseg);
+	BLANK();
+}
+#endif
+
+#ifdef CONFIG_HIBERNATION
+void output_pbe_defines(void)
+{
+	COMMENT(" Linux struct pbe offsets. ");
+	OFFSET(PBE_ADDRESS, pbe, address);
+	OFFSET(PBE_ORIG_ADDRESS, pbe, orig_address);
+	OFFSET(PBE_NEXT, pbe, next);
+	DEFINE(PBE_SIZE, sizeof(struct pbe));
+	BLANK();
+}
+#endif
+
+#ifdef CONFIG_CPU_PM
+void output_pm_defines(void)
+{
+	COMMENT(" PM offsets. ");
+#ifdef CONFIG_EVA
+	OFFSET(SSS_SEGCTL0,	mips_static_suspend_state, segctl[0]);
+	OFFSET(SSS_SEGCTL1,	mips_static_suspend_state, segctl[1]);
+	OFFSET(SSS_SEGCTL2,	mips_static_suspend_state, segctl[2]);
+#endif
+	OFFSET(SSS_SP,		mips_static_suspend_state, sp);
+	BLANK();
+}
+#endif
+
+#ifdef CONFIG_MIPS_FP_SUPPORT
+void output_kvm_defines(void)
+{
+	COMMENT(" KVM/MIPS Specific offsets. ");
+
+	OFFSET(VCPU_FPR0, kvm_vcpu_arch, fpu.fpr[0]);
+	OFFSET(VCPU_FPR1, kvm_vcpu_arch, fpu.fpr[1]);
+	OFFSET(VCPU_FPR2, kvm_vcpu_arch, fpu.fpr[2]);
+	OFFSET(VCPU_FPR3, kvm_vcpu_arch, fpu.fpr[3]);
+	OFFSET(VCPU_FPR4, kvm_vcpu_arch, fpu.fpr[4]);
+	OFFSET(VCPU_FPR5, kvm_vcpu_arch, fpu.fpr[5]);
+	OFFSET(VCPU_FPR6, kvm_vcpu_arch, fpu.fpr[6]);
+	OFFSET(VCPU_FPR7, kvm_vcpu_arch, fpu.fpr[7]);
+	OFFSET(VCPU_FPR8, kvm_vcpu_arch, fpu.fpr[8]);
+	OFFSET(VCPU_FPR9, kvm_vcpu_arch, fpu.fpr[9]);
+	OFFSET(VCPU_FPR10, kvm_vcpu_arch, fpu.fpr[10]);
+	OFFSET(VCPU_FPR11, kvm_vcpu_arch, fpu.fpr[11]);
+	OFFSET(VCPU_FPR12, kvm_vcpu_arch, fpu.fpr[12]);
+	OFFSET(VCPU_FPR13, kvm_vcpu_arch, fpu.fpr[13]);
+	OFFSET(VCPU_FPR14, kvm_vcpu_arch, fpu.fpr[14]);
+	OFFSET(VCPU_FPR15, kvm_vcpu_arch, fpu.fpr[15]);
+	OFFSET(VCPU_FPR16, kvm_vcpu_arch, fpu.fpr[16]);
+	OFFSET(VCPU_FPR17, kvm_vcpu_arch, fpu.fpr[17]);
+	OFFSET(VCPU_FPR18, kvm_vcpu_arch, fpu.fpr[18]);
+	OFFSET(VCPU_FPR19, kvm_vcpu_arch, fpu.fpr[19]);
+	OFFSET(VCPU_FPR20, kvm_vcpu_arch, fpu.fpr[20]);
+	OFFSET(VCPU_FPR21, kvm_vcpu_arch, fpu.fpr[21]);
+	OFFSET(VCPU_FPR22, kvm_vcpu_arch, fpu.fpr[22]);
+	OFFSET(VCPU_FPR23, kvm_vcpu_arch, fpu.fpr[23]);
+	OFFSET(VCPU_FPR24, kvm_vcpu_arch, fpu.fpr[24]);
+	OFFSET(VCPU_FPR25, kvm_vcpu_arch, fpu.fpr[25]);
+	OFFSET(VCPU_FPR26, kvm_vcpu_arch, fpu.fpr[26]);
+	OFFSET(VCPU_FPR27, kvm_vcpu_arch, fpu.fpr[27]);
+	OFFSET(VCPU_FPR28, kvm_vcpu_arch, fpu.fpr[28]);
+	OFFSET(VCPU_FPR29, kvm_vcpu_arch, fpu.fpr[29]);
+	OFFSET(VCPU_FPR30, kvm_vcpu_arch, fpu.fpr[30]);
+	OFFSET(VCPU_FPR31, kvm_vcpu_arch, fpu.fpr[31]);
+
+	OFFSET(VCPU_FCR31, kvm_vcpu_arch, fpu.fcr31);
+	OFFSET(VCPU_MSA_CSR, kvm_vcpu_arch, fpu.msacsr);
+	BLANK();
+}
+#endif
+
+#ifdef CONFIG_MIPS_CPS
+void output_cps_defines(void)
+{
+	COMMENT(" MIPS CPS offsets. ");
+
+	OFFSET(COREBOOTCFG_VPEMASK, core_boot_config, vpe_mask);
+	OFFSET(COREBOOTCFG_VPECONFIG, core_boot_config, vpe_config);
+	DEFINE(COREBOOTCFG_SIZE, sizeof(struct core_boot_config));
+
+	OFFSET(VPEBOOTCFG_PC, vpe_boot_config, pc);
+	OFFSET(VPEBOOTCFG_SP, vpe_boot_config, sp);
+	OFFSET(VPEBOOTCFG_GP, vpe_boot_config, gp);
+	DEFINE(VPEBOOTCFG_SIZE, sizeof(struct vpe_boot_config));
+}
+#endif
diff --git a/arch/mips/kernel/binfmt_elfn32.c b/arch/mips/kernel/binfmt_elfn32.c
new file mode 100644
index 000000000..c4441416e
--- /dev/null
+++ b/arch/mips/kernel/binfmt_elfn32.c
@@ -0,0 +1,113 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Support for n32 Linux/MIPS ELF binaries.
+ * Author: Ralf Baechle (ralf@linux-mips.org)
+ *
+ * Copyright (C) 1999, 2001 Ralf Baechle
+ * Copyright (C) 1999, 2001 Silicon Graphics, Inc.
+ *
+ * Heavily inspired by the 32-bit Sparc compat code which is
+ * Copyright (C) 1995, 1996, 1997, 1998 David S. Miller (davem@redhat.com)
+ * Copyright (C) 1995, 1996, 1997, 1998 Jakub Jelinek	(jj@ultra.linux.cz)
+ */
+
+#define ELF_ARCH		EM_MIPS
+#define ELF_CLASS		ELFCLASS32
+#ifdef __MIPSEB__
+#define ELF_DATA		ELFDATA2MSB;
+#else /* __MIPSEL__ */
+#define ELF_DATA		ELFDATA2LSB;
+#endif
+
+/* ELF register definitions */
+#define ELF_NGREG	45
+#define ELF_NFPREG	33
+
+typedef unsigned long elf_greg_t;
+typedef elf_greg_t elf_gregset_t[ELF_NGREG];
+
+typedef double elf_fpreg_t;
+typedef elf_fpreg_t elf_fpregset_t[ELF_NFPREG];
+
+/*
+ * This is used to ensure we don't load something for the wrong architecture.
+ */
+#define elf_check_arch elfn32_check_arch
+
+#define TASK32_SIZE		0x7fff8000UL
+#undef ELF_ET_DYN_BASE
+#define ELF_ET_DYN_BASE		(TASK32_SIZE / 3 * 2)
+
+#include <asm/processor.h>
+#include <linux/elfcore.h>
+#include <linux/compat.h>
+#include <linux/math64.h>
+
+#define elf_prstatus elf_prstatus32
+struct elf_prstatus32
+{
+	struct elf_siginfo pr_info;	/* Info associated with signal */
+	short	pr_cursig;		/* Current signal */
+	unsigned int pr_sigpend;	/* Set of pending signals */
+	unsigned int pr_sighold;	/* Set of held signals */
+	pid_t	pr_pid;
+	pid_t	pr_ppid;
+	pid_t	pr_pgrp;
+	pid_t	pr_sid;
+	struct old_timeval32 pr_utime; /* User time */
+	struct old_timeval32 pr_stime; /* System time */
+	struct old_timeval32 pr_cutime;/* Cumulative user time */
+	struct old_timeval32 pr_cstime;/* Cumulative system time */
+	elf_gregset_t pr_reg;	/* GP registers */
+	int pr_fpvalid;		/* True if math co-processor being used.  */
+};
+
+#define elf_prpsinfo elf_prpsinfo32
+struct elf_prpsinfo32
+{
+	char	pr_state;	/* numeric process state */
+	char	pr_sname;	/* char for pr_state */
+	char	pr_zomb;	/* zombie */
+	char	pr_nice;	/* nice val */
+	unsigned int pr_flag;	/* flags */
+	__kernel_uid_t	pr_uid;
+	__kernel_gid_t	pr_gid;
+	pid_t	pr_pid, pr_ppid, pr_pgrp, pr_sid;
+	/* Lots missing */
+	char	pr_fname[16];	/* filename of executable */
+	char	pr_psargs[ELF_PRARGSZ]; /* initial part of arg list */
+};
+
+#define elf_caddr_t	u32
+#define init_elf_binfmt init_elfn32_binfmt
+
+#define jiffies_to_timeval jiffies_to_old_timeval32
+static __inline__ void
+jiffies_to_old_timeval32(unsigned long jiffies, struct old_timeval32 *value)
+{
+	/*
+	 * Convert jiffies to nanoseconds and separate with
+	 * one divide.
+	 */
+	u64 nsec = (u64)jiffies * TICK_NSEC;
+	u32 rem;
+	value->tv_sec = div_u64_rem(nsec, NSEC_PER_SEC, &rem);
+	value->tv_usec = rem / NSEC_PER_USEC;
+}
+
+#define ELF_CORE_EFLAGS EF_MIPS_ABI2
+
+#undef TASK_SIZE
+#define TASK_SIZE TASK_SIZE32
+
+#undef ns_to_kernel_old_timeval
+#define ns_to_kernel_old_timeval ns_to_old_timeval32
+
+/*
+ * Some data types as stored in coredump.
+ */
+#define user_long_t             compat_long_t
+#define user_siginfo_t          compat_siginfo_t
+#define copy_siginfo_to_external        copy_siginfo_to_external32
+
+#include "../../../fs/binfmt_elf.c"
diff --git a/arch/mips/kernel/binfmt_elfo32.c b/arch/mips/kernel/binfmt_elfo32.c
new file mode 100644
index 000000000..7b2a23f48
--- /dev/null
+++ b/arch/mips/kernel/binfmt_elfo32.c
@@ -0,0 +1,116 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Support for o32 Linux/MIPS ELF binaries.
+ * Author: Ralf Baechle (ralf@linux-mips.org)
+ *
+ * Copyright (C) 1999, 2001 Ralf Baechle
+ * Copyright (C) 1999, 2001 Silicon Graphics, Inc.
+ *
+ * Heavily inspired by the 32-bit Sparc compat code which is
+ * Copyright (C) 1995, 1996, 1997, 1998 David S. Miller (davem@redhat.com)
+ * Copyright (C) 1995, 1996, 1997, 1998 Jakub Jelinek	(jj@ultra.linux.cz)
+ */
+
+#define ELF_ARCH		EM_MIPS
+#define ELF_CLASS		ELFCLASS32
+#ifdef __MIPSEB__
+#define ELF_DATA		ELFDATA2MSB;
+#else /* __MIPSEL__ */
+#define ELF_DATA		ELFDATA2LSB;
+#endif
+
+/* ELF register definitions */
+#define ELF_NGREG	45
+#define ELF_NFPREG	33
+
+typedef unsigned int elf_greg_t;
+typedef elf_greg_t elf_gregset_t[ELF_NGREG];
+
+typedef double elf_fpreg_t;
+typedef elf_fpreg_t elf_fpregset_t[ELF_NFPREG];
+
+/*
+ * This is used to ensure we don't load something for the wrong architecture.
+ */
+#define elf_check_arch elfo32_check_arch
+
+#ifdef CONFIG_KVM_GUEST
+#define TASK32_SIZE		0x3fff8000UL
+#else
+#define TASK32_SIZE		0x7fff8000UL
+#endif
+#undef ELF_ET_DYN_BASE
+#define ELF_ET_DYN_BASE		(TASK32_SIZE / 3 * 2)
+
+#include <asm/processor.h>
+
+#include <linux/elfcore.h>
+#include <linux/compat.h>
+#include <linux/math64.h>
+
+#define elf_prstatus elf_prstatus32
+struct elf_prstatus32
+{
+	struct elf_siginfo pr_info;	/* Info associated with signal */
+	short	pr_cursig;		/* Current signal */
+	unsigned int pr_sigpend;	/* Set of pending signals */
+	unsigned int pr_sighold;	/* Set of held signals */
+	pid_t	pr_pid;
+	pid_t	pr_ppid;
+	pid_t	pr_pgrp;
+	pid_t	pr_sid;
+	struct old_timeval32 pr_utime; /* User time */
+	struct old_timeval32 pr_stime; /* System time */
+	struct old_timeval32 pr_cutime;/* Cumulative user time */
+	struct old_timeval32 pr_cstime;/* Cumulative system time */
+	elf_gregset_t pr_reg;	/* GP registers */
+	int pr_fpvalid;		/* True if math co-processor being used.  */
+};
+
+#define elf_prpsinfo elf_prpsinfo32
+struct elf_prpsinfo32
+{
+	char	pr_state;	/* numeric process state */
+	char	pr_sname;	/* char for pr_state */
+	char	pr_zomb;	/* zombie */
+	char	pr_nice;	/* nice val */
+	unsigned int pr_flag;	/* flags */
+	__kernel_uid_t	pr_uid;
+	__kernel_gid_t	pr_gid;
+	pid_t	pr_pid, pr_ppid, pr_pgrp, pr_sid;
+	/* Lots missing */
+	char	pr_fname[16];	/* filename of executable */
+	char	pr_psargs[ELF_PRARGSZ]; /* initial part of arg list */
+};
+
+#define elf_caddr_t	u32
+#define init_elf_binfmt init_elf32_binfmt
+
+#define jiffies_to_timeval jiffies_to_old_timeval32
+static inline void
+jiffies_to_old_timeval32(unsigned long jiffies, struct old_timeval32 *value)
+{
+	/*
+	 * Convert jiffies to nanoseconds and separate with
+	 * one divide.
+	 */
+	u64 nsec = (u64)jiffies * TICK_NSEC;
+	u32 rem;
+	value->tv_sec = div_u64_rem(nsec, NSEC_PER_SEC, &rem);
+	value->tv_usec = rem / NSEC_PER_USEC;
+}
+
+#undef TASK_SIZE
+#define TASK_SIZE TASK_SIZE32
+
+#undef ns_to_kernel_old_timeval
+#define ns_to_kernel_old_timeval ns_to_old_timeval32
+
+/*
+ * Some data types as stored in coredump.
+ */
+#define user_long_t             compat_long_t
+#define user_siginfo_t          compat_siginfo_t
+#define copy_siginfo_to_external        copy_siginfo_to_external32
+
+#include "../../../fs/binfmt_elf.c"
diff --git a/arch/mips/kernel/bmips_5xxx_init.S b/arch/mips/kernel/bmips_5xxx_init.S
new file mode 100644
index 000000000..9e422d186
--- /dev/null
+++ b/arch/mips/kernel/bmips_5xxx_init.S
@@ -0,0 +1,747 @@
+
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2011-2012 by Broadcom Corporation
+ *
+ * Init for bmips 5000.
+ * Used to init second core in dual core 5000's.
+ */
+
+#include <linux/init.h>
+
+#include <asm/asm.h>
+#include <asm/asmmacro.h>
+#include <asm/cacheops.h>
+#include <asm/regdef.h>
+#include <asm/mipsregs.h>
+#include <asm/stackframe.h>
+#include <asm/addrspace.h>
+#include <asm/hazards.h>
+#include <asm/bmips.h>
+
+#ifdef CONFIG_CPU_BMIPS5000
+
+
+#define cacheop(kva, size, linesize, op) 	\
+	.set noreorder			;	\
+	addu		t1, kva, size	;	\
+	subu		t2, linesize, 1	;	\
+	not		t2		;	\
+	and		t0, kva, t2	;	\
+	addiu		t1, t1, -1	;	\
+	and		t1, t2		;	\
+9:	cache		op, 0(t0)	;	\
+	bne		t0, t1, 9b	;	\
+	 addu		t0, linesize	;	\
+	.set reorder			;
+
+
+
+#define	IS_SHIFT	22
+#define	IL_SHIFT	19
+#define	IA_SHIFT	16
+#define	DS_SHIFT	13
+#define	DL_SHIFT	10
+#define	DA_SHIFT	 7
+#define	IS_MASK		 7
+#define	IL_MASK		 7
+#define	IA_MASK		 7
+#define	DS_MASK		 7
+#define	DL_MASK		 7
+#define	DA_MASK		 7
+#define	ICE_MASK	0x80000000
+#define	DCE_MASK	0x40000000
+
+#define CP0_BRCM_CONFIG0	$22, 0
+#define CP0_BRCM_MODE		$22, 1
+#define	CP0_CONFIG_K0_MASK	7
+
+#define CP0_ICACHE_TAG_LO	$28
+#define CP0_ICACHE_DATA_LO	$28, 1
+#define CP0_DCACHE_TAG_LO	$28, 2
+#define CP0_D_SEC_CACHE_DATA_LO	$28, 3
+#define CP0_ICACHE_TAG_HI	$29
+#define CP0_ICACHE_DATA_HI	$29, 1
+#define CP0_DCACHE_TAG_HI	$29, 2
+
+#define CP0_BRCM_MODE_Luc_MASK		(1 << 11)
+#define	CP0_BRCM_CONFIG0_CWF_MASK	(1 << 20)
+#define	CP0_BRCM_CONFIG0_TSE_MASK	(1 << 19)
+#define CP0_BRCM_MODE_SET_MASK		(1 << 7)
+#define CP0_BRCM_MODE_ClkRATIO_MASK	(7 << 4)
+#define CP0_BRCM_MODE_BrPRED_MASK 	(3 << 24)
+#define CP0_BRCM_MODE_BrPRED_SHIFT	24
+#define CP0_BRCM_MODE_BrHIST_MASK 	(0x1f << 20)
+#define CP0_BRCM_MODE_BrHIST_SHIFT	20
+
+/* ZSC L2 Cache Register Access Register Definitions */
+#define BRCM_ZSC_ALL_REGS_SELECT		0x7 << 24
+
+#define BRCM_ZSC_CONFIG_REG			0 << 3
+#define BRCM_ZSC_REQ_BUFFER_REG			2 << 3
+#define BRCM_ZSC_RBUS_ADDR_MAPPING_REG0		4 << 3
+#define BRCM_ZSC_RBUS_ADDR_MAPPING_REG1		6 << 3
+#define BRCM_ZSC_RBUS_ADDR_MAPPING_REG2		8 << 3
+
+#define BRCM_ZSC_SCB0_ADDR_MAPPING_REG0		0xa << 3
+#define BRCM_ZSC_SCB0_ADDR_MAPPING_REG1		0xc << 3
+
+#define BRCM_ZSC_SCB1_ADDR_MAPPING_REG0		0xe << 3
+#define BRCM_ZSC_SCB1_ADDR_MAPPING_REG1		0x10 << 3
+
+#define BRCM_ZSC_CONFIG_LMB1En			1 << (15)
+#define BRCM_ZSC_CONFIG_LMB0En			1 << (14)
+
+/* branch predition values */
+
+#define BRCM_BrPRED_ALL_TAKEN		(0x0)
+#define BRCM_BrPRED_ALL_NOT_TAKEN	(0x1)
+#define BRCM_BrPRED_BHT_ENABLE		(0x2)
+#define BRCM_BrPRED_PREDICT_BACKWARD	(0x3)
+
+
+
+.align 2
+/*
+ * Function: 	size_i_cache
+ * Arguments: 	None
+ * Returns:	v0 = i cache size, v1 = I cache line size
+ * Description: compute the I-cache size and I-cache line size
+ * Trashes:	v0, v1, a0, t0
+ *
+ *	pseudo code:
+ *
+ */
+
+LEAF(size_i_cache)
+	.set	noreorder
+
+	mfc0	a0, CP0_CONFIG, 1
+	move	t0, a0
+
+	/*
+	 * Determine sets per way: IS
+	 *
+	 * This field contains the number of sets (i.e., indices) per way of
+	 * the instruction cache:
+	 * i) 0x0: 64, ii) 0x1: 128, iii) 0x2: 256, iv) 0x3: 512, v) 0x4: 1k
+	 * vi) 0x5 - 0x7: Reserved.
+	 */
+
+	srl	a0, a0, IS_SHIFT
+	and	a0, a0, IS_MASK
+
+	/* sets per way = (64<<IS) */
+
+	li	v0, 0x40
+	sllv	v0, v0, a0
+
+	/*
+	 * Determine line size
+	 *
+	 * This field contains the line size of the instruction cache:
+	 * i) 0x0: No I-cache present, i) 0x3: 16 bytes, ii) 0x4: 32 bytes, iii)
+	 * 0x5: 64 bytes, iv) the rest: Reserved.
+	 */
+
+	move	a0, t0
+
+	srl	a0, a0, IL_SHIFT
+	and	a0, a0, IL_MASK
+
+	beqz	a0, no_i_cache
+	nop
+
+	/* line size = 2 ^ (IL+1) */
+
+	addi	a0, a0, 1
+	li	v1, 1
+	sll	v1, v1, a0
+
+	/* v0 now have sets per way, multiply it by line size now
+	 * that will give the set size
+	 */
+
+	sll	v0, v0, a0
+
+	/*
+	 * Determine set associativity
+	 *
+	 * This field contains the set associativity of the instruction cache.
+	 * i) 0x0: Direct mapped, ii) 0x1: 2-way, iii) 0x2: 3-way, iv) 0x3:
+	 * 4-way, v) 0x4 - 0x7: Reserved.
+	 */
+
+	move	a0, t0
+
+	srl	a0, a0, IA_SHIFT
+	and	a0, a0, IA_MASK
+	addi	a0, a0, 0x1
+
+	/* v0 has the set size, multiply it by
+	 * set associativiy, to get the cache size
+	 */
+
+	multu	v0, a0	/*multu is interlocked, so no need to insert nops */
+	mflo	v0
+	b	1f
+	nop
+
+no_i_cache:
+	move	v0, zero
+	move	v1, zero
+1:
+	jr	ra
+	nop
+	.set	reorder
+
+END(size_i_cache)
+
+/*
+ * Function: 	size_d_cache
+ * Arguments: 	None
+ * Returns:	v0 = d cache size, v1 = d cache line size
+ * Description: compute the D-cache size and D-cache line size.
+ * Trashes:	v0, v1, a0, t0
+ *
+ */
+
+LEAF(size_d_cache)
+	.set	noreorder
+
+	mfc0	a0, CP0_CONFIG, 1
+	move	t0, a0
+
+	/*
+	 * Determine sets per way: IS
+	 *
+	 * This field contains the number of sets (i.e., indices) per way of
+	 * the instruction cache:
+	 * i) 0x0: 64, ii) 0x1: 128, iii) 0x2: 256, iv) 0x3: 512, v) 0x4: 1k
+	 * vi) 0x5 - 0x7: Reserved.
+	 */
+
+	srl	a0, a0, DS_SHIFT
+	and	a0, a0, DS_MASK
+
+	/* sets per way = (64<<IS) */
+
+	li	v0, 0x40
+	sllv	v0, v0, a0
+
+	/*
+	 * Determine line size
+	 *
+	 * This field contains the line size of the instruction cache:
+	 * i) 0x0: No I-cache present, i) 0x3: 16 bytes, ii) 0x4: 32 bytes, iii)
+	 * 0x5: 64 bytes, iv) the rest: Reserved.
+	 */
+	move	a0, t0
+
+	srl	a0, a0, DL_SHIFT
+	and	a0, a0, DL_MASK
+
+	beqz	a0, no_d_cache
+	nop
+
+	/* line size = 2 ^ (IL+1) */
+
+	addi	a0, a0, 1
+	li	v1, 1
+	sll	v1, v1, a0
+
+	/* v0 now have sets per way, multiply it by line size now
+	 * that will give the set size
+	 */
+
+	sll	v0, v0, a0
+
+	/* determine set associativity
+	 *
+	 * This field contains the set associativity of the instruction cache.
+	 * i) 0x0: Direct mapped, ii) 0x1: 2-way, iii) 0x2: 3-way, iv) 0x3:
+	 * 4-way, v) 0x4 - 0x7: Reserved.
+	 */
+
+	move	a0, t0
+
+	srl	a0, a0, DA_SHIFT
+	and	a0, a0, DA_MASK
+	addi	a0, a0, 0x1
+
+	/* v0 has the set size, multiply it by
+	 * set associativiy, to get the cache size
+	 */
+
+	multu	v0, a0	/*multu is interlocked, so no need to insert nops */
+	mflo	v0
+
+	b	1f
+	nop
+
+no_d_cache:
+	move	v0, zero
+	move	v1, zero
+1:
+	jr	ra
+	nop
+	.set	reorder
+
+END(size_d_cache)
+
+
+/*
+ * Function: enable_ID
+ * Arguments: 	None
+ * Returns:	None
+ * Description: Enable I and D caches, initialize I and D-caches, also set
+ *		hardware delay for d-cache (TP0).
+ * Trashes:	t0
+ *
+ */
+	.global	enable_ID
+	.ent	enable_ID
+	.set	noreorder
+enable_ID:
+	mfc0	t0, CP0_BRCM_CONFIG0
+	or	t0, t0, (ICE_MASK | DCE_MASK)
+	mtc0	t0, CP0_BRCM_CONFIG0
+	jr	ra
+	nop
+
+	.end	enable_ID
+	.set	reorder
+
+
+/*
+ * Function: l1_init
+ * Arguments: 	None
+ * Returns:	None
+ * Description: Enable I and D caches, and initialize I and D-caches
+ * Trashes:	a0, v0, v1, t0, t1, t2, t8
+ *
+ */
+	.globl	l1_init
+	.ent	l1_init
+	.set	noreorder
+l1_init:
+
+	/* save return address */
+	move	t8, ra
+
+
+	/* initialize I and D cache Data and Tag registers.  */
+	mtc0	zero, CP0_ICACHE_TAG_LO
+	mtc0	zero, CP0_ICACHE_TAG_HI
+	mtc0	zero, CP0_ICACHE_DATA_LO
+	mtc0	zero, CP0_ICACHE_DATA_HI
+	mtc0	zero, CP0_DCACHE_TAG_LO
+	mtc0	zero, CP0_DCACHE_TAG_HI
+
+	/* Enable Caches before Clearing. If the caches are disabled
+	 * then the cache operations to clear the cache will be ignored
+	 */
+
+	jal	enable_ID
+	nop
+
+	jal	size_i_cache	/* v0 = i-cache size, v1 = i-cache line size */
+	nop
+
+	/* run uncached in kseg 1 */
+	la	k0, 1f
+	lui	k1, 0x2000
+	or	k0, k1, k0
+	jr	k0
+	nop
+1:
+
+	/*
+	 * set K0 cache mode
+	 */
+
+	mfc0	t0, CP0_CONFIG
+	and	t0, t0, ~CP0_CONFIG_K0_MASK
+	or	t0, t0, 3	/* Write Back mode */
+	mtc0	t0, CP0_CONFIG
+
+	/*
+	 * Initialize instruction cache.
+	 */
+
+	li	a0, KSEG0
+	cacheop(a0, v0, v1, Index_Store_Tag_I)
+
+	/*
+	 * Now we can run from I-$, kseg 0
+	 */
+	la	k0, 1f
+	lui	k1, 0x2000
+	or	k0, k1, k0
+	xor	k0, k1, k0
+	jr	k0
+	nop
+1:
+	/*
+	 * Initialize data cache.
+	 */
+
+	jal	size_d_cache	/* v0 = d-cache size, v1 = d-cache line size */
+	nop
+
+
+	li	a0, KSEG0
+	cacheop(a0, v0, v1, Index_Store_Tag_D)
+
+	jr	t8
+	nop
+
+	.end 	l1_init
+	.set	reorder
+
+
+/*
+ * Function: 	set_other_config
+ * Arguments:	none
+ * Returns:	None
+ * Description: initialize other remainder configuration to defaults.
+ * Trashes:	t0, t1
+ *
+ *	pseudo code:
+ *
+ */
+LEAF(set_other_config)
+	.set noreorder
+
+	/* enable Bus error for I-fetch */
+	mfc0	t0, CP0_CACHEERR, 0
+	li	t1, 0x4
+	or	t0, t1
+	mtc0	t0, CP0_CACHEERR, 0
+
+	/* enable Bus error for Load */
+	mfc0	t0, CP0_CACHEERR, 1
+	li	t1, 0x4
+	or	t0, t1
+	mtc0	t0, CP0_CACHEERR, 1
+
+	/* enable Bus Error for Store */
+	mfc0	t0, CP0_CACHEERR, 2
+	li	t1, 0x4
+	or	t0, t1
+	mtc0	t0, CP0_CACHEERR, 2
+
+	jr	ra
+	nop
+	.set reorder
+END(set_other_config)
+
+/*
+ * Function: 	set_branch_pred
+ * Arguments:	none
+ * Returns:	None
+ * Description:
+ * Trashes:	t0, t1
+ *
+ *	pseudo code:
+ *
+ */
+
+LEAF(set_branch_pred)
+	.set noreorder
+	mfc0	t0, CP0_BRCM_MODE
+	li	t1, ~(CP0_BRCM_MODE_BrPRED_MASK | CP0_BRCM_MODE_BrHIST_MASK )
+	and	t0, t0, t1
+
+	/* enable Branch prediction */
+	li	t1, BRCM_BrPRED_BHT_ENABLE
+	sll	t1, CP0_BRCM_MODE_BrPRED_SHIFT
+	or	t0, t0, t1
+
+	/* set history count to 8 */
+	li	t1, 8
+	sll	t1, CP0_BRCM_MODE_BrHIST_SHIFT
+	or	t0, t0, t1
+
+	mtc0	t0, CP0_BRCM_MODE
+	jr	ra
+	nop
+	.set	reorder
+END(set_branch_pred)
+
+
+/*
+ * Function: 	set_luc
+ * Arguments:	set link uncached.
+ * Returns:	None
+ * Description:
+ * Trashes:	t0, t1
+ *
+ */
+LEAF(set_luc)
+	.set noreorder
+	mfc0	t0, CP0_BRCM_MODE
+	li	t1, ~(CP0_BRCM_MODE_Luc_MASK)
+	and	t0, t0, t1
+
+	/* set Luc */
+	ori	t0, t0, CP0_BRCM_MODE_Luc_MASK
+
+	mtc0	t0, CP0_BRCM_MODE
+	jr	ra
+	nop
+	.set	reorder
+END(set_luc)
+
+/*
+ * Function: 	set_cwf_tse
+ * Arguments:	set CWF and TSE bits
+ * Returns:	None
+ * Description:
+ * Trashes:	t0, t1
+ *
+ */
+LEAF(set_cwf_tse)
+	.set noreorder
+	mfc0	t0, CP0_BRCM_CONFIG0
+	li	t1, (CP0_BRCM_CONFIG0_CWF_MASK | CP0_BRCM_CONFIG0_TSE_MASK)
+	or	t0, t0, t1
+
+	mtc0	t0, CP0_BRCM_CONFIG0
+	jr	ra
+	nop
+	.set	reorder
+END(set_cwf_tse)
+
+/*
+ * Function: 	set_clock_ratio
+ * Arguments:	set clock ratio specified by a0
+ * Returns:	None
+ * Description:
+ * Trashes:	v0, v1, a0, a1
+ *
+ *	pseudo code:
+ *
+ */
+LEAF(set_clock_ratio)
+	.set noreorder
+
+	mfc0	t0, CP0_BRCM_MODE
+	li	t1, ~(CP0_BRCM_MODE_SET_MASK | CP0_BRCM_MODE_ClkRATIO_MASK)
+	and	t0, t0, t1
+	li	t1, CP0_BRCM_MODE_SET_MASK
+	or	t0, t0, t1
+	or	t0, t0, a0
+	mtc0	t0, CP0_BRCM_MODE
+	jr	ra
+	nop
+	.set	reorder
+END(set_clock_ratio)
+/*
+ * Function: set_zephyr
+ * Arguments:	None
+ * Returns:	None
+ * Description: Set any zephyr bits
+ * Trashes:	t0 & t1
+ *
+ */
+LEAF(set_zephyr)
+	.set	noreorder
+
+	/* enable read/write of CP0 #22 sel. 8 */
+	li	t0, 0x5a455048
+	.word	0x4088b00f	/* mtc0    t0, $22, 15 */
+
+	.word	0x4008b008	/* mfc0    t0, $22, 8 */
+	li	t1, 0x09008000	/* turn off pref, jtb */
+	or	t0, t0, t1
+	.word	0x4088b008	/* mtc0    t0, $22, 8 */
+	sync
+
+	/* disable read/write of CP0 #22 sel 8 */
+	li	t0, 0x0
+	.word	0x4088b00f	/* mtc0    t0, $22, 15 */
+
+
+	jr	ra
+	nop
+	.set reorder
+
+END(set_zephyr)
+
+
+/*
+ * Function:	set_llmb
+ * Arguments:	a0=0 disable llmb, a0=1 enables llmb
+ * Returns:	None
+ * Description:
+ * Trashes:	t0, t1, t2
+ *
+ *      pseudo code:
+ *
+ */
+LEAF(set_llmb)
+	.set noreorder
+
+	li	t2, 0x90000000 | BRCM_ZSC_ALL_REGS_SELECT | BRCM_ZSC_CONFIG_REG
+	sync
+	cache	0x7, 0x0(t2)
+	sync
+	mfc0	t0, CP0_D_SEC_CACHE_DATA_LO
+	li	t1, ~(BRCM_ZSC_CONFIG_LMB1En | BRCM_ZSC_CONFIG_LMB0En)
+	and	t0, t0, t1
+
+	beqz	a0, svlmb
+	nop
+
+enable_lmb:
+	li	t1, (BRCM_ZSC_CONFIG_LMB1En | BRCM_ZSC_CONFIG_LMB0En)
+	or	t0, t0, t1
+
+svlmb:
+	mtc0	t0, CP0_D_SEC_CACHE_DATA_LO
+	sync
+	cache	0xb, 0x0(t2)
+	sync
+
+	jr	ra
+	nop
+	.set reorder
+
+END(set_llmb)
+/*
+ * Function: 	core_init
+ * Arguments:	none
+ * Returns:	None
+ * Description: initialize core related configuration
+ * Trashes:	v0,v1,a0,a1,t8
+ *
+ *	pseudo code:
+ *
+ */
+	.globl	core_init
+	.ent	core_init
+	.set	noreorder
+core_init:
+	move	t8, ra
+
+	/* set Zephyr bits. */
+	bal	set_zephyr
+	nop
+
+	/* set low latency memory bus */
+	li	a0, 1
+	bal	set_llmb
+	nop
+
+	/* set branch prediction (TP0 only) */
+	bal	set_branch_pred
+	nop
+
+	/* set link uncached */
+	bal	set_luc
+	nop
+
+	/* set CWF and TSE */
+	bal	set_cwf_tse
+	nop
+
+	/*
+	 *set clock ratio by setting 1 to 'set'
+	 * and 0 to ClkRatio, (TP0 only)
+	 */
+	li	a0, 0
+	bal	set_clock_ratio
+	nop
+
+	/* set other configuration to defaults */
+	bal	set_other_config
+	nop
+
+	move	ra, t8
+	jr	ra
+	nop
+
+	.set reorder
+	.end	core_init
+
+/*
+ * Function: 	clear_jump_target_buffer
+ * Arguments:	None
+ * Returns:	None
+ * Description:
+ * Trashes:	t0, t1, t2
+ *
+ */
+#define RESET_CALL_RETURN_STACK_THIS_THREAD		(0x06<<16)
+#define RESET_JUMP_TARGET_BUFFER_THIS_THREAD		(0x04<<16)
+#define JTB_CS_CNTL_MASK				(0xFF<<16)
+
+	.globl	clear_jump_target_buffer
+	.ent	clear_jump_target_buffer
+	.set	noreorder
+clear_jump_target_buffer:
+
+	mfc0	t0, $22, 2
+	nop
+	nop
+
+	li	t1, ~JTB_CS_CNTL_MASK
+	and	t0, t0, t1
+	li	t2, RESET_CALL_RETURN_STACK_THIS_THREAD
+	or	t0, t0, t2
+	mtc0	t0, $22, 2
+	nop
+	nop
+
+	and	t0, t0, t1
+	li	t2, RESET_JUMP_TARGET_BUFFER_THIS_THREAD
+	or	t0, t0, t2
+	mtc0	t0, $22, 2
+	nop
+	nop
+	jr	ra
+	nop
+
+	.end	clear_jump_target_buffer
+	.set	reorder
+/*
+ * Function: 	bmips_cache_init
+ * Arguments: 	None
+ * Returns:	None
+ * Description: Enable I and D caches, and initialize I and D-caches
+ * Trashes:	v0, v1, t0, t1, t2, t5, t7, t8
+ *
+ */
+	.globl	bmips_5xxx_init
+	.ent	bmips_5xxx_init
+	.set	noreorder
+bmips_5xxx_init:
+
+	/* save return address and A0 */
+	move	t7, ra
+	move	t5, a0
+
+	jal	l1_init
+	nop
+
+	jal	core_init
+	nop
+
+	jal	clear_jump_target_buffer
+	nop
+
+	mtc0	zero, CP0_CAUSE
+
+	move 	a0, t5
+	jr	t7
+	nop
+
+	.end 	bmips_5xxx_init
+	.set	reorder
+
+
+#endif
diff --git a/arch/mips/kernel/bmips_vec.S b/arch/mips/kernel/bmips_vec.S
new file mode 100644
index 000000000..921a5fa55
--- /dev/null
+++ b/arch/mips/kernel/bmips_vec.S
@@ -0,0 +1,322 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2011 by Kevin Cernekee (cernekee@gmail.com)
+ *
+ * Reset/NMI/re-entry vectors for BMIPS processors
+ */
+
+
+#include <asm/asm.h>
+#include <asm/asmmacro.h>
+#include <asm/cacheops.h>
+#include <asm/cpu.h>
+#include <asm/regdef.h>
+#include <asm/mipsregs.h>
+#include <asm/stackframe.h>
+#include <asm/addrspace.h>
+#include <asm/hazards.h>
+#include <asm/bmips.h>
+
+	.macro	BARRIER
+	.set	mips32
+	_ssnop
+	_ssnop
+	_ssnop
+	.set	mips0
+	.endm
+
+/***********************************************************************
+ * Alternate CPU1 startup vector for BMIPS4350
+ *
+ * On some systems the bootloader has already started CPU1 and configured
+ * it to resume execution at 0x8000_0200 (!BEV IV vector) when it is
+ * triggered by the SW1 interrupt.  If that is the case we try to move
+ * it to a more convenient place: BMIPS_WARM_RESTART_VEC @ 0x8000_0380.
+ ***********************************************************************/
+
+LEAF(bmips_smp_movevec)
+	la	k0, 1f
+	li	k1, CKSEG1
+	or	k0, k1
+	jr	k0
+
+1:
+	/* clear IV, pending IPIs */
+	mtc0	zero, CP0_CAUSE
+
+	/* re-enable IRQs to wait for SW1 */
+	li	k0, ST0_IE | ST0_BEV | STATUSF_IP1
+	mtc0	k0, CP0_STATUS
+
+	/* set up CPU1 CBR; move BASE to 0xa000_0000 */
+	li	k0, 0xff400000
+	mtc0	k0, $22, 6
+	/* set up relocation vector address based on thread ID */
+	mfc0	k1, $22, 3
+	srl	k1, 16
+	andi	k1, 0x8000
+	or	k1, CKSEG1 | BMIPS_RELO_VECTOR_CONTROL_0
+	or	k0, k1
+	li	k1, 0xa0080000
+	sw	k1, 0(k0)
+
+	/* wait here for SW1 interrupt from bmips_boot_secondary() */
+	wait
+
+	la	k0, bmips_reset_nmi_vec
+	li	k1, CKSEG1
+	or	k0, k1
+	jr	k0
+END(bmips_smp_movevec)
+
+/***********************************************************************
+ * Reset/NMI vector
+ * For BMIPS processors that can relocate their exception vectors, this
+ * entire function gets copied to 0x8000_0000.
+ ***********************************************************************/
+
+NESTED(bmips_reset_nmi_vec, PT_SIZE, sp)
+	.set	push
+	.set	noat
+	.align	4
+
+#ifdef CONFIG_SMP
+	/* if the NMI bit is clear, assume this is a CPU1 reset instead */
+	li	k1, (1 << 19)
+	mfc0	k0, CP0_STATUS
+	and	k0, k1
+	beqz	k0, soft_reset
+
+#if defined(CONFIG_CPU_BMIPS5000)
+	mfc0	k0, CP0_PRID
+	li	k1, PRID_IMP_BMIPS5000
+	/* mask with PRID_IMP_BMIPS5000 to cover both variants */
+	andi	k0, PRID_IMP_BMIPS5000
+	bne	k0, k1, 1f
+
+	/* if we're not on core 0, this must be the SMP boot signal */
+	li	k1, (3 << 25)
+	mfc0	k0, $22
+	and	k0, k1
+	bnez	k0, bmips_smp_entry
+1:
+#endif /* CONFIG_CPU_BMIPS5000 */
+#endif /* CONFIG_SMP */
+
+	/* nope, it's just a regular NMI */
+	SAVE_ALL
+	move	a0, sp
+
+	/* clear EXL, ERL, BEV so that TLB refills still work */
+	mfc0	k0, CP0_STATUS
+	li	k1, ST0_ERL | ST0_EXL | ST0_BEV | ST0_IE
+	or	k0, k1
+	xor	k0, k1
+	mtc0	k0, CP0_STATUS
+	BARRIER
+
+	/* jump to the NMI handler function */
+	la	k0, nmi_handler
+	jr	k0
+
+	RESTORE_ALL
+	.set	arch=r4000
+	eret
+
+#ifdef CONFIG_SMP
+soft_reset:
+
+#if defined(CONFIG_CPU_BMIPS5000)
+	mfc0	k0, CP0_PRID
+	andi	k0, 0xff00
+	li	k1, PRID_IMP_BMIPS5200
+	bne	k0, k1, bmips_smp_entry
+
+        /* if running on TP 1, jump  to  bmips_smp_entry */
+        mfc0    k0, $22
+        li      k1, (1 << 24)
+        and     k1, k0
+        bnez    k1, bmips_smp_entry
+        nop
+
+        /*
+         * running on TP0, can not be core 0 (the boot core).
+         * Check for soft reset.  Indicates a warm boot
+         */
+        mfc0    k0, $12
+        li      k1, (1 << 20)
+        and     k0, k1
+        beqz    k0, bmips_smp_entry
+
+        /*
+         * Warm boot.
+         * Cache init is only done on TP0
+         */
+        la      k0, bmips_5xxx_init
+        jalr    k0
+        nop
+
+        b       bmips_smp_entry
+        nop
+#endif
+
+/***********************************************************************
+ * CPU1 reset vector (used for the initial boot only)
+ * This is still part of bmips_reset_nmi_vec().
+ ***********************************************************************/
+
+bmips_smp_entry:
+
+	/* set up CP0 STATUS; enable FPU */
+	li	k0, 0x30000000
+	mtc0	k0, CP0_STATUS
+	BARRIER
+
+	/* set local CP0 CONFIG to make kseg0 cacheable, write-back */
+	mfc0	k0, CP0_CONFIG
+	ori	k0, 0x07
+	xori	k0, 0x04
+	mtc0	k0, CP0_CONFIG
+
+	mfc0	k0, CP0_PRID
+	andi	k0, 0xff00
+#if defined(CONFIG_CPU_BMIPS4350) || defined(CONFIG_CPU_BMIPS4380)
+	li	k1, PRID_IMP_BMIPS43XX
+	bne	k0, k1, 2f
+
+	/* initialize CPU1's local I-cache */
+	li	k0, 0x80000000
+	li	k1, 0x80010000
+	mtc0	zero, $28
+	mtc0	zero, $28, 1
+	BARRIER
+
+1:	cache	Index_Store_Tag_I, 0(k0)
+	addiu	k0, 16
+	bne	k0, k1, 1b
+
+	b	3f
+2:
+#endif /* CONFIG_CPU_BMIPS4350 || CONFIG_CPU_BMIPS4380 */
+#if defined(CONFIG_CPU_BMIPS5000)
+	/* mask with PRID_IMP_BMIPS5000 to cover both variants */
+	li	k1, PRID_IMP_BMIPS5000
+	andi	k0, PRID_IMP_BMIPS5000
+	bne	k0, k1, 3f
+
+	/* set exception vector base */
+	la	k0, ebase
+	lw	k0, 0(k0)
+	mtc0	k0, $15, 1
+	BARRIER
+#endif /* CONFIG_CPU_BMIPS5000 */
+3:
+	/* jump back to kseg0 in case we need to remap the kseg1 area */
+	la	k0, 1f
+	jr	k0
+1:
+	la	k0, bmips_enable_xks01
+	jalr	k0
+
+	/* use temporary stack to set up upper memory TLB */
+	li	sp, BMIPS_WARM_RESTART_VEC
+	la	k0, plat_wired_tlb_setup
+	jalr	k0
+
+	/* switch to permanent stack and continue booting */
+
+	.global bmips_secondary_reentry
+bmips_secondary_reentry:
+	la	k0, bmips_smp_boot_sp
+	lw	sp, 0(k0)
+	la	k0, bmips_smp_boot_gp
+	lw	gp, 0(k0)
+	la	k0, start_secondary
+	jr	k0
+
+#endif /* CONFIG_SMP */
+
+	.align	4
+	.global bmips_reset_nmi_vec_end
+bmips_reset_nmi_vec_end:
+
+END(bmips_reset_nmi_vec)
+
+	.set	pop
+
+/***********************************************************************
+ * CPU1 warm restart vector (used for second and subsequent boots).
+ * Also used for S2 standby recovery (PM).
+ * This entire function gets copied to (BMIPS_WARM_RESTART_VEC)
+ ***********************************************************************/
+
+LEAF(bmips_smp_int_vec)
+
+	.align	4
+	mfc0	k0, CP0_STATUS
+	ori	k0, 0x01
+	xori	k0, 0x01
+	mtc0	k0, CP0_STATUS
+	eret
+
+	.align	4
+	.global bmips_smp_int_vec_end
+bmips_smp_int_vec_end:
+
+END(bmips_smp_int_vec)
+
+/***********************************************************************
+ * XKS01 support
+ * Certain CPUs support extending kseg0 to 1024MB.
+ ***********************************************************************/
+
+LEAF(bmips_enable_xks01)
+
+#if defined(CONFIG_XKS01)
+	mfc0	t0, CP0_PRID
+	andi	t2, t0, 0xff00
+#if defined(CONFIG_CPU_BMIPS4380)
+	li	t1, PRID_IMP_BMIPS43XX
+	bne	t2, t1, 1f
+
+	andi	t0, 0xff
+	addiu	t1, t0, -PRID_REV_BMIPS4380_HI
+	bgtz	t1, 2f
+	addiu	t0, -PRID_REV_BMIPS4380_LO
+	bltz	t0, 2f
+
+	mfc0	t0, $22, 3
+	li	t1, 0x1ff0
+	li	t2, (1 << 12) | (1 << 9)
+	or	t0, t1
+	xor	t0, t1
+	or	t0, t2
+	mtc0	t0, $22, 3
+	BARRIER
+	b	2f
+1:
+#endif /* CONFIG_CPU_BMIPS4380 */
+#if defined(CONFIG_CPU_BMIPS5000)
+	li	t1, PRID_IMP_BMIPS5000
+	/* mask with PRID_IMP_BMIPS5000 to cover both variants */
+	andi	t2, PRID_IMP_BMIPS5000
+	bne	t2, t1, 2f
+
+	mfc0	t0, $22, 5
+	li	t1, 0x01ff
+	li	t2, (1 << 8) | (1 << 5)
+	or	t0, t1
+	xor	t0, t1
+	or	t0, t2
+	mtc0	t0, $22, 5
+	BARRIER
+#endif /* CONFIG_CPU_BMIPS5000 */
+2:
+#endif /* defined(CONFIG_XKS01) */
+
+	jr	ra
+
+END(bmips_enable_xks01)
diff --git a/arch/mips/kernel/branch.c b/arch/mips/kernel/branch.c
new file mode 100644
index 000000000..0216ff24c
--- /dev/null
+++ b/arch/mips/kernel/branch.c
@@ -0,0 +1,908 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1996, 97, 2000, 2001 by Ralf Baechle
+ * Copyright (C) 2001 MIPS Technologies, Inc.
+ */
+#include <linux/kernel.h>
+#include <linux/sched/signal.h>
+#include <linux/signal.h>
+#include <linux/export.h>
+#include <asm/branch.h>
+#include <asm/cpu.h>
+#include <asm/cpu-features.h>
+#include <asm/fpu.h>
+#include <asm/fpu_emulator.h>
+#include <asm/inst.h>
+#include <asm/mips-r2-to-r6-emul.h>
+#include <asm/ptrace.h>
+#include <linux/uaccess.h>
+
+#include "probes-common.h"
+
+/*
+ * Calculate and return exception PC in case of branch delay slot
+ * for microMIPS and MIPS16e. It does not clear the ISA mode bit.
+ */
+int __isa_exception_epc(struct pt_regs *regs)
+{
+	unsigned short inst;
+	long epc = regs->cp0_epc;
+
+	/* Calculate exception PC in branch delay slot. */
+	if (__get_user(inst, (u16 __user *) msk_isa16_mode(epc))) {
+		/* This should never happen because delay slot was checked. */
+		force_sig(SIGSEGV);
+		return epc;
+	}
+	if (cpu_has_mips16) {
+		union mips16e_instruction inst_mips16e;
+
+		inst_mips16e.full = inst;
+		if (inst_mips16e.ri.opcode == MIPS16e_jal_op)
+			epc += 4;
+		else
+			epc += 2;
+	} else if (mm_insn_16bit(inst))
+		epc += 2;
+	else
+		epc += 4;
+
+	return epc;
+}
+
+/* (microMIPS) Convert 16-bit register encoding to 32-bit register encoding. */
+static const unsigned int reg16to32map[8] = {16, 17, 2, 3, 4, 5, 6, 7};
+
+int __mm_isBranchInstr(struct pt_regs *regs, struct mm_decoded_insn dec_insn,
+		       unsigned long *contpc)
+{
+	union mips_instruction insn = (union mips_instruction)dec_insn.insn;
+	int __maybe_unused bc_false = 0;
+
+	if (!cpu_has_mmips)
+		return 0;
+
+	switch (insn.mm_i_format.opcode) {
+	case mm_pool32a_op:
+		if ((insn.mm_i_format.simmediate & MM_POOL32A_MINOR_MASK) ==
+		    mm_pool32axf_op) {
+			switch (insn.mm_i_format.simmediate >>
+				MM_POOL32A_MINOR_SHIFT) {
+			case mm_jalr_op:
+			case mm_jalrhb_op:
+			case mm_jalrs_op:
+			case mm_jalrshb_op:
+				if (insn.mm_i_format.rt != 0)	/* Not mm_jr */
+					regs->regs[insn.mm_i_format.rt] =
+						regs->cp0_epc +
+						dec_insn.pc_inc +
+						dec_insn.next_pc_inc;
+				*contpc = regs->regs[insn.mm_i_format.rs];
+				return 1;
+			}
+		}
+		break;
+	case mm_pool32i_op:
+		switch (insn.mm_i_format.rt) {
+		case mm_bltzals_op:
+		case mm_bltzal_op:
+			regs->regs[31] = regs->cp0_epc +
+				dec_insn.pc_inc +
+				dec_insn.next_pc_inc;
+			fallthrough;
+		case mm_bltz_op:
+			if ((long)regs->regs[insn.mm_i_format.rs] < 0)
+				*contpc = regs->cp0_epc +
+					dec_insn.pc_inc +
+					(insn.mm_i_format.simmediate << 1);
+			else
+				*contpc = regs->cp0_epc +
+					dec_insn.pc_inc +
+					dec_insn.next_pc_inc;
+			return 1;
+		case mm_bgezals_op:
+		case mm_bgezal_op:
+			regs->regs[31] = regs->cp0_epc +
+					dec_insn.pc_inc +
+					dec_insn.next_pc_inc;
+			fallthrough;
+		case mm_bgez_op:
+			if ((long)regs->regs[insn.mm_i_format.rs] >= 0)
+				*contpc = regs->cp0_epc +
+					dec_insn.pc_inc +
+					(insn.mm_i_format.simmediate << 1);
+			else
+				*contpc = regs->cp0_epc +
+					dec_insn.pc_inc +
+					dec_insn.next_pc_inc;
+			return 1;
+		case mm_blez_op:
+			if ((long)regs->regs[insn.mm_i_format.rs] <= 0)
+				*contpc = regs->cp0_epc +
+					dec_insn.pc_inc +
+					(insn.mm_i_format.simmediate << 1);
+			else
+				*contpc = regs->cp0_epc +
+					dec_insn.pc_inc +
+					dec_insn.next_pc_inc;
+			return 1;
+		case mm_bgtz_op:
+			if ((long)regs->regs[insn.mm_i_format.rs] <= 0)
+				*contpc = regs->cp0_epc +
+					dec_insn.pc_inc +
+					(insn.mm_i_format.simmediate << 1);
+			else
+				*contpc = regs->cp0_epc +
+					dec_insn.pc_inc +
+					dec_insn.next_pc_inc;
+			return 1;
+#ifdef CONFIG_MIPS_FP_SUPPORT
+		case mm_bc2f_op:
+		case mm_bc1f_op: {
+			unsigned int fcr31;
+			unsigned int bit;
+
+			bc_false = 1;
+			fallthrough;
+		case mm_bc2t_op:
+		case mm_bc1t_op:
+			preempt_disable();
+			if (is_fpu_owner())
+			        fcr31 = read_32bit_cp1_register(CP1_STATUS);
+			else
+				fcr31 = current->thread.fpu.fcr31;
+			preempt_enable();
+
+			if (bc_false)
+				fcr31 = ~fcr31;
+
+			bit = (insn.mm_i_format.rs >> 2);
+			bit += (bit != 0);
+			bit += 23;
+			if (fcr31 & (1 << bit))
+				*contpc = regs->cp0_epc +
+					dec_insn.pc_inc +
+					(insn.mm_i_format.simmediate << 1);
+			else
+				*contpc = regs->cp0_epc +
+					dec_insn.pc_inc + dec_insn.next_pc_inc;
+			return 1;
+		}
+#endif /* CONFIG_MIPS_FP_SUPPORT */
+		}
+		break;
+	case mm_pool16c_op:
+		switch (insn.mm_i_format.rt) {
+		case mm_jalr16_op:
+		case mm_jalrs16_op:
+			regs->regs[31] = regs->cp0_epc +
+				dec_insn.pc_inc + dec_insn.next_pc_inc;
+			fallthrough;
+		case mm_jr16_op:
+			*contpc = regs->regs[insn.mm_i_format.rs];
+			return 1;
+		}
+		break;
+	case mm_beqz16_op:
+		if ((long)regs->regs[reg16to32map[insn.mm_b1_format.rs]] == 0)
+			*contpc = regs->cp0_epc +
+				dec_insn.pc_inc +
+				(insn.mm_b1_format.simmediate << 1);
+		else
+			*contpc = regs->cp0_epc +
+				dec_insn.pc_inc + dec_insn.next_pc_inc;
+		return 1;
+	case mm_bnez16_op:
+		if ((long)regs->regs[reg16to32map[insn.mm_b1_format.rs]] != 0)
+			*contpc = regs->cp0_epc +
+				dec_insn.pc_inc +
+				(insn.mm_b1_format.simmediate << 1);
+		else
+			*contpc = regs->cp0_epc +
+				dec_insn.pc_inc + dec_insn.next_pc_inc;
+		return 1;
+	case mm_b16_op:
+		*contpc = regs->cp0_epc + dec_insn.pc_inc +
+			 (insn.mm_b0_format.simmediate << 1);
+		return 1;
+	case mm_beq32_op:
+		if (regs->regs[insn.mm_i_format.rs] ==
+		    regs->regs[insn.mm_i_format.rt])
+			*contpc = regs->cp0_epc +
+				dec_insn.pc_inc +
+				(insn.mm_i_format.simmediate << 1);
+		else
+			*contpc = regs->cp0_epc +
+				dec_insn.pc_inc +
+				dec_insn.next_pc_inc;
+		return 1;
+	case mm_bne32_op:
+		if (regs->regs[insn.mm_i_format.rs] !=
+		    regs->regs[insn.mm_i_format.rt])
+			*contpc = regs->cp0_epc +
+				dec_insn.pc_inc +
+				(insn.mm_i_format.simmediate << 1);
+		else
+			*contpc = regs->cp0_epc +
+				dec_insn.pc_inc + dec_insn.next_pc_inc;
+		return 1;
+	case mm_jalx32_op:
+		regs->regs[31] = regs->cp0_epc +
+			dec_insn.pc_inc + dec_insn.next_pc_inc;
+		*contpc = regs->cp0_epc + dec_insn.pc_inc;
+		*contpc >>= 28;
+		*contpc <<= 28;
+		*contpc |= (insn.j_format.target << 2);
+		return 1;
+	case mm_jals32_op:
+	case mm_jal32_op:
+		regs->regs[31] = regs->cp0_epc +
+			dec_insn.pc_inc + dec_insn.next_pc_inc;
+		fallthrough;
+	case mm_j32_op:
+		*contpc = regs->cp0_epc + dec_insn.pc_inc;
+		*contpc >>= 27;
+		*contpc <<= 27;
+		*contpc |= (insn.j_format.target << 1);
+		set_isa16_mode(*contpc);
+		return 1;
+	}
+	return 0;
+}
+
+/*
+ * Compute return address and emulate branch in microMIPS mode after an
+ * exception only. It does not handle compact branches/jumps and cannot
+ * be used in interrupt context. (Compact branches/jumps do not cause
+ * exceptions.)
+ */
+int __microMIPS_compute_return_epc(struct pt_regs *regs)
+{
+	u16 __user *pc16;
+	u16 halfword;
+	unsigned int word;
+	unsigned long contpc;
+	struct mm_decoded_insn mminsn = { 0 };
+
+	mminsn.micro_mips_mode = 1;
+
+	/* This load never faults. */
+	pc16 = (unsigned short __user *)msk_isa16_mode(regs->cp0_epc);
+	__get_user(halfword, pc16);
+	pc16++;
+	contpc = regs->cp0_epc + 2;
+	word = ((unsigned int)halfword << 16);
+	mminsn.pc_inc = 2;
+
+	if (!mm_insn_16bit(halfword)) {
+		__get_user(halfword, pc16);
+		pc16++;
+		contpc = regs->cp0_epc + 4;
+		mminsn.pc_inc = 4;
+		word |= halfword;
+	}
+	mminsn.insn = word;
+
+	if (get_user(halfword, pc16))
+		goto sigsegv;
+	mminsn.next_pc_inc = 2;
+	word = ((unsigned int)halfword << 16);
+
+	if (!mm_insn_16bit(halfword)) {
+		pc16++;
+		if (get_user(halfword, pc16))
+			goto sigsegv;
+		mminsn.next_pc_inc = 4;
+		word |= halfword;
+	}
+	mminsn.next_insn = word;
+
+	mm_isBranchInstr(regs, mminsn, &contpc);
+
+	regs->cp0_epc = contpc;
+
+	return 0;
+
+sigsegv:
+	force_sig(SIGSEGV);
+	return -EFAULT;
+}
+
+/*
+ * Compute return address and emulate branch in MIPS16e mode after an
+ * exception only. It does not handle compact branches/jumps and cannot
+ * be used in interrupt context. (Compact branches/jumps do not cause
+ * exceptions.)
+ */
+int __MIPS16e_compute_return_epc(struct pt_regs *regs)
+{
+	u16 __user *addr;
+	union mips16e_instruction inst;
+	u16 inst2;
+	u32 fullinst;
+	long epc;
+
+	epc = regs->cp0_epc;
+
+	/* Read the instruction. */
+	addr = (u16 __user *)msk_isa16_mode(epc);
+	if (__get_user(inst.full, addr)) {
+		force_sig(SIGSEGV);
+		return -EFAULT;
+	}
+
+	switch (inst.ri.opcode) {
+	case MIPS16e_extend_op:
+		regs->cp0_epc += 4;
+		return 0;
+
+		/*
+		 *  JAL and JALX in MIPS16e mode
+		 */
+	case MIPS16e_jal_op:
+		addr += 1;
+		if (__get_user(inst2, addr)) {
+			force_sig(SIGSEGV);
+			return -EFAULT;
+		}
+		fullinst = ((unsigned)inst.full << 16) | inst2;
+		regs->regs[31] = epc + 6;
+		epc += 4;
+		epc >>= 28;
+		epc <<= 28;
+		/*
+		 * JAL:5 X:1 TARGET[20-16]:5 TARGET[25:21]:5 TARGET[15:0]:16
+		 *
+		 * ......TARGET[15:0].................TARGET[20:16]...........
+		 * ......TARGET[25:21]
+		 */
+		epc |=
+		    ((fullinst & 0xffff) << 2) | ((fullinst & 0x3e00000) >> 3) |
+		    ((fullinst & 0x1f0000) << 7);
+		if (!inst.jal.x)
+			set_isa16_mode(epc);	/* Set ISA mode bit. */
+		regs->cp0_epc = epc;
+		return 0;
+
+		/*
+		 *  J(AL)R(C)
+		 */
+	case MIPS16e_rr_op:
+		if (inst.rr.func == MIPS16e_jr_func) {
+
+			if (inst.rr.ra)
+				regs->cp0_epc = regs->regs[31];
+			else
+				regs->cp0_epc =
+				    regs->regs[reg16to32[inst.rr.rx]];
+
+			if (inst.rr.l) {
+				if (inst.rr.nd)
+					regs->regs[31] = epc + 2;
+				else
+					regs->regs[31] = epc + 4;
+			}
+			return 0;
+		}
+		break;
+	}
+
+	/*
+	 * All other cases have no branch delay slot and are 16-bits.
+	 * Branches do not cause an exception.
+	 */
+	regs->cp0_epc += 2;
+
+	return 0;
+}
+
+/**
+ * __compute_return_epc_for_insn - Computes the return address and do emulate
+ *				    branch simulation, if required.
+ *
+ * @regs:	Pointer to pt_regs
+ * @insn:	branch instruction to decode
+ * Return:	-EFAULT on error and forces SIGILL, and on success
+ *		returns 0 or BRANCH_LIKELY_TAKEN as appropriate after
+ *		evaluating the branch.
+ *
+ * MIPS R6 Compact branches and forbidden slots:
+ *	Compact branches do not throw exceptions because they do
+ *	not have delay slots. The forbidden slot instruction ($PC+4)
+ *	is only executed if the branch was not taken. Otherwise the
+ *	forbidden slot is skipped entirely. This means that the
+ *	only possible reason to be here because of a MIPS R6 compact
+ *	branch instruction is that the forbidden slot has thrown one.
+ *	In that case the branch was not taken, so the EPC can be safely
+ *	set to EPC + 8.
+ */
+int __compute_return_epc_for_insn(struct pt_regs *regs,
+				   union mips_instruction insn)
+{
+	long epc = regs->cp0_epc;
+	unsigned int dspcontrol;
+	int ret = 0;
+
+	switch (insn.i_format.opcode) {
+	/*
+	 * jr and jalr are in r_format format.
+	 */
+	case spec_op:
+		switch (insn.r_format.func) {
+		case jalr_op:
+			regs->regs[insn.r_format.rd] = epc + 8;
+			fallthrough;
+		case jr_op:
+			if (NO_R6EMU && insn.r_format.func == jr_op)
+				goto sigill_r2r6;
+			regs->cp0_epc = regs->regs[insn.r_format.rs];
+			break;
+		}
+		break;
+
+	/*
+	 * This group contains:
+	 * bltz_op, bgez_op, bltzl_op, bgezl_op,
+	 * bltzal_op, bgezal_op, bltzall_op, bgezall_op.
+	 */
+	case bcond_op:
+		switch (insn.i_format.rt) {
+		case bltzl_op:
+			if (NO_R6EMU)
+				goto sigill_r2r6;
+			fallthrough;
+		case bltz_op:
+			if ((long)regs->regs[insn.i_format.rs] < 0) {
+				epc = epc + 4 + (insn.i_format.simmediate << 2);
+				if (insn.i_format.rt == bltzl_op)
+					ret = BRANCH_LIKELY_TAKEN;
+			} else
+				epc += 8;
+			regs->cp0_epc = epc;
+			break;
+
+		case bgezl_op:
+			if (NO_R6EMU)
+				goto sigill_r2r6;
+			fallthrough;
+		case bgez_op:
+			if ((long)regs->regs[insn.i_format.rs] >= 0) {
+				epc = epc + 4 + (insn.i_format.simmediate << 2);
+				if (insn.i_format.rt == bgezl_op)
+					ret = BRANCH_LIKELY_TAKEN;
+			} else
+				epc += 8;
+			regs->cp0_epc = epc;
+			break;
+
+		case bltzal_op:
+		case bltzall_op:
+			if (NO_R6EMU && (insn.i_format.rs ||
+			    insn.i_format.rt == bltzall_op))
+				goto sigill_r2r6;
+			regs->regs[31] = epc + 8;
+			/*
+			 * OK we are here either because we hit a NAL
+			 * instruction or because we are emulating an
+			 * old bltzal{,l} one. Let's figure out what the
+			 * case really is.
+			 */
+			if (!insn.i_format.rs) {
+				/*
+				 * NAL or BLTZAL with rs == 0
+				 * Doesn't matter if we are R6 or not. The
+				 * result is the same
+				 */
+				regs->cp0_epc += 4 +
+					(insn.i_format.simmediate << 2);
+				break;
+			}
+			/* Now do the real thing for non-R6 BLTZAL{,L} */
+			if ((long)regs->regs[insn.i_format.rs] < 0) {
+				epc = epc + 4 + (insn.i_format.simmediate << 2);
+				if (insn.i_format.rt == bltzall_op)
+					ret = BRANCH_LIKELY_TAKEN;
+			} else
+				epc += 8;
+			regs->cp0_epc = epc;
+			break;
+
+		case bgezal_op:
+		case bgezall_op:
+			if (NO_R6EMU && (insn.i_format.rs ||
+			    insn.i_format.rt == bgezall_op))
+				goto sigill_r2r6;
+			regs->regs[31] = epc + 8;
+			/*
+			 * OK we are here either because we hit a BAL
+			 * instruction or because we are emulating an
+			 * old bgezal{,l} one. Let's figure out what the
+			 * case really is.
+			 */
+			if (!insn.i_format.rs) {
+				/*
+				 * BAL or BGEZAL with rs == 0
+				 * Doesn't matter if we are R6 or not. The
+				 * result is the same
+				 */
+				regs->cp0_epc += 4 +
+					(insn.i_format.simmediate << 2);
+				break;
+			}
+			/* Now do the real thing for non-R6 BGEZAL{,L} */
+			if ((long)regs->regs[insn.i_format.rs] >= 0) {
+				epc = epc + 4 + (insn.i_format.simmediate << 2);
+				if (insn.i_format.rt == bgezall_op)
+					ret = BRANCH_LIKELY_TAKEN;
+			} else
+				epc += 8;
+			regs->cp0_epc = epc;
+			break;
+
+		case bposge32_op:
+			if (!cpu_has_dsp)
+				goto sigill_dsp;
+
+			dspcontrol = rddsp(0x01);
+
+			if (dspcontrol >= 32) {
+				epc = epc + 4 + (insn.i_format.simmediate << 2);
+			} else
+				epc += 8;
+			regs->cp0_epc = epc;
+			break;
+		}
+		break;
+
+	/*
+	 * These are unconditional and in j_format.
+	 */
+	case jalx_op:
+	case jal_op:
+		regs->regs[31] = regs->cp0_epc + 8;
+		fallthrough;
+	case j_op:
+		epc += 4;
+		epc >>= 28;
+		epc <<= 28;
+		epc |= (insn.j_format.target << 2);
+		regs->cp0_epc = epc;
+		if (insn.i_format.opcode == jalx_op)
+			set_isa16_mode(regs->cp0_epc);
+		break;
+
+	/*
+	 * These are conditional and in i_format.
+	 */
+	case beql_op:
+		if (NO_R6EMU)
+			goto sigill_r2r6;
+		fallthrough;
+	case beq_op:
+		if (regs->regs[insn.i_format.rs] ==
+		    regs->regs[insn.i_format.rt]) {
+			epc = epc + 4 + (insn.i_format.simmediate << 2);
+			if (insn.i_format.opcode == beql_op)
+				ret = BRANCH_LIKELY_TAKEN;
+		} else
+			epc += 8;
+		regs->cp0_epc = epc;
+		break;
+
+	case bnel_op:
+		if (NO_R6EMU)
+			goto sigill_r2r6;
+		fallthrough;
+	case bne_op:
+		if (regs->regs[insn.i_format.rs] !=
+		    regs->regs[insn.i_format.rt]) {
+			epc = epc + 4 + (insn.i_format.simmediate << 2);
+			if (insn.i_format.opcode == bnel_op)
+				ret = BRANCH_LIKELY_TAKEN;
+		} else
+			epc += 8;
+		regs->cp0_epc = epc;
+		break;
+
+	case blezl_op: /* not really i_format */
+		if (!insn.i_format.rt && NO_R6EMU)
+			goto sigill_r2r6;
+		fallthrough;
+	case blez_op:
+		/*
+		 * Compact branches for R6 for the
+		 * blez and blezl opcodes.
+		 * BLEZ  | rs = 0 | rt != 0  == BLEZALC
+		 * BLEZ  | rs = rt != 0      == BGEZALC
+		 * BLEZ  | rs != 0 | rt != 0 == BGEUC
+		 * BLEZL | rs = 0 | rt != 0  == BLEZC
+		 * BLEZL | rs = rt != 0      == BGEZC
+		 * BLEZL | rs != 0 | rt != 0 == BGEC
+		 *
+		 * For real BLEZ{,L}, rt is always 0.
+		 */
+
+		if (cpu_has_mips_r6 && insn.i_format.rt) {
+			if ((insn.i_format.opcode == blez_op) &&
+			    ((!insn.i_format.rs && insn.i_format.rt) ||
+			     (insn.i_format.rs == insn.i_format.rt)))
+				regs->regs[31] = epc + 4;
+			regs->cp0_epc += 8;
+			break;
+		}
+		/* rt field assumed to be zero */
+		if ((long)regs->regs[insn.i_format.rs] <= 0) {
+			epc = epc + 4 + (insn.i_format.simmediate << 2);
+			if (insn.i_format.opcode == blezl_op)
+				ret = BRANCH_LIKELY_TAKEN;
+		} else
+			epc += 8;
+		regs->cp0_epc = epc;
+		break;
+
+	case bgtzl_op:
+		if (!insn.i_format.rt && NO_R6EMU)
+			goto sigill_r2r6;
+		fallthrough;
+	case bgtz_op:
+		/*
+		 * Compact branches for R6 for the
+		 * bgtz and bgtzl opcodes.
+		 * BGTZ  | rs = 0 | rt != 0  == BGTZALC
+		 * BGTZ  | rs = rt != 0      == BLTZALC
+		 * BGTZ  | rs != 0 | rt != 0 == BLTUC
+		 * BGTZL | rs = 0 | rt != 0  == BGTZC
+		 * BGTZL | rs = rt != 0      == BLTZC
+		 * BGTZL | rs != 0 | rt != 0 == BLTC
+		 *
+		 * *ZALC varint for BGTZ &&& rt != 0
+		 * For real GTZ{,L}, rt is always 0.
+		 */
+		if (cpu_has_mips_r6 && insn.i_format.rt) {
+			if ((insn.i_format.opcode == blez_op) &&
+			    ((!insn.i_format.rs && insn.i_format.rt) ||
+			    (insn.i_format.rs == insn.i_format.rt)))
+				regs->regs[31] = epc + 4;
+			regs->cp0_epc += 8;
+			break;
+		}
+
+		/* rt field assumed to be zero */
+		if ((long)regs->regs[insn.i_format.rs] > 0) {
+			epc = epc + 4 + (insn.i_format.simmediate << 2);
+			if (insn.i_format.opcode == bgtzl_op)
+				ret = BRANCH_LIKELY_TAKEN;
+		} else
+			epc += 8;
+		regs->cp0_epc = epc;
+		break;
+
+#ifdef CONFIG_MIPS_FP_SUPPORT
+	/*
+	 * And now the FPA/cp1 branch instructions.
+	 */
+	case cop1_op: {
+		unsigned int bit, fcr31, reg;
+
+		if (cpu_has_mips_r6 &&
+		    ((insn.i_format.rs == bc1eqz_op) ||
+		     (insn.i_format.rs == bc1nez_op))) {
+			if (!init_fp_ctx(current))
+				lose_fpu(1);
+			reg = insn.i_format.rt;
+			bit = get_fpr32(&current->thread.fpu.fpr[reg], 0) & 0x1;
+			if (insn.i_format.rs == bc1eqz_op)
+				bit = !bit;
+			own_fpu(1);
+			if (bit)
+				epc = epc + 4 +
+					(insn.i_format.simmediate << 2);
+			else
+				epc += 8;
+			regs->cp0_epc = epc;
+
+			break;
+		} else {
+
+			preempt_disable();
+			if (is_fpu_owner())
+			        fcr31 = read_32bit_cp1_register(CP1_STATUS);
+			else
+				fcr31 = current->thread.fpu.fcr31;
+			preempt_enable();
+
+			bit = (insn.i_format.rt >> 2);
+			bit += (bit != 0);
+			bit += 23;
+			switch (insn.i_format.rt & 3) {
+			case 0: /* bc1f */
+			case 2: /* bc1fl */
+				if (~fcr31 & (1 << bit)) {
+					epc = epc + 4 +
+						(insn.i_format.simmediate << 2);
+					if (insn.i_format.rt == 2)
+						ret = BRANCH_LIKELY_TAKEN;
+				} else
+					epc += 8;
+				regs->cp0_epc = epc;
+				break;
+
+			case 1: /* bc1t */
+			case 3: /* bc1tl */
+				if (fcr31 & (1 << bit)) {
+					epc = epc + 4 +
+						(insn.i_format.simmediate << 2);
+					if (insn.i_format.rt == 3)
+						ret = BRANCH_LIKELY_TAKEN;
+				} else
+					epc += 8;
+				regs->cp0_epc = epc;
+				break;
+			}
+			break;
+		}
+	}
+#endif /* CONFIG_MIPS_FP_SUPPORT */
+
+#ifdef CONFIG_CPU_CAVIUM_OCTEON
+	case lwc2_op: /* This is bbit0 on Octeon */
+		if ((regs->regs[insn.i_format.rs] & (1ull<<insn.i_format.rt))
+		     == 0)
+			epc = epc + 4 + (insn.i_format.simmediate << 2);
+		else
+			epc += 8;
+		regs->cp0_epc = epc;
+		break;
+	case ldc2_op: /* This is bbit032 on Octeon */
+		if ((regs->regs[insn.i_format.rs] &
+		    (1ull<<(insn.i_format.rt+32))) == 0)
+			epc = epc + 4 + (insn.i_format.simmediate << 2);
+		else
+			epc += 8;
+		regs->cp0_epc = epc;
+		break;
+	case swc2_op: /* This is bbit1 on Octeon */
+		if (regs->regs[insn.i_format.rs] & (1ull<<insn.i_format.rt))
+			epc = epc + 4 + (insn.i_format.simmediate << 2);
+		else
+			epc += 8;
+		regs->cp0_epc = epc;
+		break;
+	case sdc2_op: /* This is bbit132 on Octeon */
+		if (regs->regs[insn.i_format.rs] &
+		    (1ull<<(insn.i_format.rt+32)))
+			epc = epc + 4 + (insn.i_format.simmediate << 2);
+		else
+			epc += 8;
+		regs->cp0_epc = epc;
+		break;
+#else
+	case bc6_op:
+		/* Only valid for MIPS R6 */
+		if (!cpu_has_mips_r6)
+			goto sigill_r6;
+		regs->cp0_epc += 8;
+		break;
+	case balc6_op:
+		if (!cpu_has_mips_r6)
+			goto sigill_r6;
+		/* Compact branch: BALC */
+		regs->regs[31] = epc + 4;
+		epc += 4 + (insn.i_format.simmediate << 2);
+		regs->cp0_epc = epc;
+		break;
+	case pop66_op:
+		if (!cpu_has_mips_r6)
+			goto sigill_r6;
+		/* Compact branch: BEQZC || JIC */
+		regs->cp0_epc += 8;
+		break;
+	case pop76_op:
+		if (!cpu_has_mips_r6)
+			goto sigill_r6;
+		/* Compact branch: BNEZC || JIALC */
+		if (!insn.i_format.rs) {
+			/* JIALC: set $31/ra */
+			regs->regs[31] = epc + 4;
+		}
+		regs->cp0_epc += 8;
+		break;
+#endif
+	case pop10_op:
+	case pop30_op:
+		/* Only valid for MIPS R6 */
+		if (!cpu_has_mips_r6)
+			goto sigill_r6;
+		/*
+		 * Compact branches:
+		 * bovc, beqc, beqzalc, bnvc, bnec, bnezlac
+		 */
+		if (insn.i_format.rt && !insn.i_format.rs)
+			regs->regs[31] = epc + 4;
+		regs->cp0_epc += 8;
+		break;
+	}
+
+	return ret;
+
+sigill_dsp:
+	pr_debug("%s: DSP branch but not DSP ASE - sending SIGILL.\n",
+		 current->comm);
+	force_sig(SIGILL);
+	return -EFAULT;
+sigill_r2r6:
+	pr_debug("%s: R2 branch but r2-to-r6 emulator is not present - sending SIGILL.\n",
+		 current->comm);
+	force_sig(SIGILL);
+	return -EFAULT;
+sigill_r6:
+	pr_debug("%s: R6 branch but no MIPSr6 ISA support - sending SIGILL.\n",
+		 current->comm);
+	force_sig(SIGILL);
+	return -EFAULT;
+}
+EXPORT_SYMBOL_GPL(__compute_return_epc_for_insn);
+
+int __compute_return_epc(struct pt_regs *regs)
+{
+	unsigned int __user *addr;
+	long epc;
+	union mips_instruction insn;
+
+	epc = regs->cp0_epc;
+	if (epc & 3)
+		goto unaligned;
+
+	/*
+	 * Read the instruction
+	 */
+	addr = (unsigned int __user *) epc;
+	if (__get_user(insn.word, addr)) {
+		force_sig(SIGSEGV);
+		return -EFAULT;
+	}
+
+	return __compute_return_epc_for_insn(regs, insn);
+
+unaligned:
+	printk("%s: unaligned epc - sending SIGBUS.\n", current->comm);
+	force_sig(SIGBUS);
+	return -EFAULT;
+}
+
+#if (defined CONFIG_KPROBES) || (defined CONFIG_UPROBES)
+
+int __insn_is_compact_branch(union mips_instruction insn)
+{
+	if (!cpu_has_mips_r6)
+		return 0;
+
+	switch (insn.i_format.opcode) {
+	case blezl_op:
+	case bgtzl_op:
+	case blez_op:
+	case bgtz_op:
+		/*
+		 * blez[l] and bgtz[l] opcodes with non-zero rt
+		 * are MIPS R6 compact branches
+		 */
+		if (insn.i_format.rt)
+			return 1;
+		break;
+	case bc6_op:
+	case balc6_op:
+	case pop10_op:
+	case pop30_op:
+	case pop66_op:
+	case pop76_op:
+		return 1;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(__insn_is_compact_branch);
+
+#endif  /* CONFIG_KPROBES || CONFIG_UPROBES */
diff --git a/arch/mips/kernel/cacheinfo.c b/arch/mips/kernel/cacheinfo.c
new file mode 100644
index 000000000..529dab855
--- /dev/null
+++ b/arch/mips/kernel/cacheinfo.c
@@ -0,0 +1,100 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * MIPS cacheinfo support
+ */
+#include <linux/cacheinfo.h>
+
+/* Populates leaf and increments to next leaf */
+#define populate_cache(cache, leaf, c_level, c_type)		\
+do {								\
+	leaf->type = c_type;					\
+	leaf->level = c_level;					\
+	leaf->coherency_line_size = c->cache.linesz;		\
+	leaf->number_of_sets = c->cache.sets;			\
+	leaf->ways_of_associativity = c->cache.ways;		\
+	leaf->size = c->cache.linesz * c->cache.sets *		\
+		c->cache.ways;					\
+	leaf++;							\
+} while (0)
+
+int init_cache_level(unsigned int cpu)
+{
+	struct cpuinfo_mips *c = &current_cpu_data;
+	struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu);
+	int levels = 0, leaves = 0;
+
+	/*
+	 * If Dcache is not set, we assume the cache structures
+	 * are not properly initialized.
+	 */
+	if (c->dcache.waysize)
+		levels += 1;
+	else
+		return -ENOENT;
+
+
+	leaves += (c->icache.waysize) ? 2 : 1;
+
+	if (c->scache.waysize) {
+		levels++;
+		leaves++;
+	}
+
+	if (c->tcache.waysize) {
+		levels++;
+		leaves++;
+	}
+
+	this_cpu_ci->num_levels = levels;
+	this_cpu_ci->num_leaves = leaves;
+	return 0;
+}
+
+static void fill_cpumask_siblings(int cpu, cpumask_t *cpu_map)
+{
+	int cpu1;
+
+	for_each_possible_cpu(cpu1)
+		if (cpus_are_siblings(cpu, cpu1))
+			cpumask_set_cpu(cpu1, cpu_map);
+}
+
+static void fill_cpumask_cluster(int cpu, cpumask_t *cpu_map)
+{
+	int cpu1;
+	int cluster = cpu_cluster(&cpu_data[cpu]);
+
+	for_each_possible_cpu(cpu1)
+		if (cpu_cluster(&cpu_data[cpu1]) == cluster)
+			cpumask_set_cpu(cpu1, cpu_map);
+}
+
+int populate_cache_leaves(unsigned int cpu)
+{
+	struct cpuinfo_mips *c = &current_cpu_data;
+	struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu);
+	struct cacheinfo *this_leaf = this_cpu_ci->info_list;
+
+	if (c->icache.waysize) {
+		/* L1 caches are per core */
+		fill_cpumask_siblings(cpu, &this_leaf->shared_cpu_map);
+		populate_cache(dcache, this_leaf, 1, CACHE_TYPE_DATA);
+		fill_cpumask_siblings(cpu, &this_leaf->shared_cpu_map);
+		populate_cache(icache, this_leaf, 1, CACHE_TYPE_INST);
+	} else {
+		populate_cache(dcache, this_leaf, 1, CACHE_TYPE_UNIFIED);
+	}
+
+	if (c->scache.waysize) {
+		/* L2 cache is per cluster */
+		fill_cpumask_cluster(cpu, &this_leaf->shared_cpu_map);
+		populate_cache(scache, this_leaf, 2, CACHE_TYPE_UNIFIED);
+	}
+
+	if (c->tcache.waysize)
+		populate_cache(tcache, this_leaf, 3, CACHE_TYPE_UNIFIED);
+
+	this_cpu_ci->cpu_map_populated = true;
+
+	return 0;
+}
diff --git a/arch/mips/kernel/cevt-bcm1480.c b/arch/mips/kernel/cevt-bcm1480.c
new file mode 100644
index 000000000..d39a2963b
--- /dev/null
+++ b/arch/mips/kernel/cevt-bcm1480.c
@@ -0,0 +1,138 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2000,2001,2004 Broadcom Corporation
+ */
+#include <linux/clockchips.h>
+#include <linux/interrupt.h>
+#include <linux/percpu.h>
+#include <linux/smp.h>
+#include <linux/irq.h>
+
+#include <asm/addrspace.h>
+#include <asm/io.h>
+#include <asm/time.h>
+
+#include <asm/sibyte/bcm1480_regs.h>
+#include <asm/sibyte/sb1250_regs.h>
+#include <asm/sibyte/bcm1480_int.h>
+#include <asm/sibyte/bcm1480_scd.h>
+
+#include <asm/sibyte/sb1250.h>
+
+#define IMR_IP2_VAL	K_BCM1480_INT_MAP_I0
+#define IMR_IP3_VAL	K_BCM1480_INT_MAP_I1
+#define IMR_IP4_VAL	K_BCM1480_INT_MAP_I2
+
+/*
+ * The general purpose timer ticks at 1MHz independent if
+ * the rest of the system
+ */
+
+static int sibyte_set_periodic(struct clock_event_device *evt)
+{
+	unsigned int cpu = smp_processor_id();
+	void __iomem *cfg, *init;
+
+	cfg = IOADDR(A_SCD_TIMER_REGISTER(cpu, R_SCD_TIMER_CFG));
+	init = IOADDR(A_SCD_TIMER_REGISTER(cpu, R_SCD_TIMER_INIT));
+
+	__raw_writeq(0, cfg);
+	__raw_writeq((V_SCD_TIMER_FREQ / HZ) - 1, init);
+	__raw_writeq(M_SCD_TIMER_ENABLE | M_SCD_TIMER_MODE_CONTINUOUS, cfg);
+	return 0;
+}
+
+static int sibyte_shutdown(struct clock_event_device *evt)
+{
+	unsigned int cpu = smp_processor_id();
+	void __iomem *cfg;
+
+	cfg = IOADDR(A_SCD_TIMER_REGISTER(cpu, R_SCD_TIMER_CFG));
+
+	/* Stop the timer until we actually program a shot */
+	__raw_writeq(0, cfg);
+	return 0;
+}
+
+static int sibyte_next_event(unsigned long delta, struct clock_event_device *cd)
+{
+	unsigned int cpu = smp_processor_id();
+	void __iomem *cfg, *init;
+
+	cfg = IOADDR(A_SCD_TIMER_REGISTER(cpu, R_SCD_TIMER_CFG));
+	init = IOADDR(A_SCD_TIMER_REGISTER(cpu, R_SCD_TIMER_INIT));
+
+	__raw_writeq(0, cfg);
+	__raw_writeq(delta - 1, init);
+	__raw_writeq(M_SCD_TIMER_ENABLE, cfg);
+
+	return 0;
+}
+
+static irqreturn_t sibyte_counter_handler(int irq, void *dev_id)
+{
+	unsigned int cpu = smp_processor_id();
+	struct clock_event_device *cd = dev_id;
+	void __iomem *cfg;
+	unsigned long tmode;
+
+	if (clockevent_state_periodic(cd))
+		tmode = M_SCD_TIMER_ENABLE | M_SCD_TIMER_MODE_CONTINUOUS;
+	else
+		tmode = 0;
+
+	/* ACK interrupt */
+	cfg = IOADDR(A_SCD_TIMER_REGISTER(cpu, R_SCD_TIMER_CFG));
+	____raw_writeq(tmode, cfg);
+
+	cd->event_handler(cd);
+
+	return IRQ_HANDLED;
+}
+
+static DEFINE_PER_CPU(struct clock_event_device, sibyte_hpt_clockevent);
+static DEFINE_PER_CPU(char [18], sibyte_hpt_name);
+
+void sb1480_clockevent_init(void)
+{
+	unsigned int cpu = smp_processor_id();
+	unsigned int irq = K_BCM1480_INT_TIMER_0 + cpu;
+	struct clock_event_device *cd = &per_cpu(sibyte_hpt_clockevent, cpu);
+	unsigned char *name = per_cpu(sibyte_hpt_name, cpu);
+	unsigned long flags =  IRQF_PERCPU | IRQF_TIMER;
+
+	BUG_ON(cpu > 3);	/* Only have 4 general purpose timers */
+
+	sprintf(name, "bcm1480-counter-%d", cpu);
+	cd->name		= name;
+	cd->features		= CLOCK_EVT_FEAT_PERIODIC |
+				  CLOCK_EVT_FEAT_ONESHOT;
+	clockevent_set_clock(cd, V_SCD_TIMER_FREQ);
+	cd->max_delta_ns	= clockevent_delta2ns(0x7fffff, cd);
+	cd->max_delta_ticks	= 0x7fffff;
+	cd->min_delta_ns	= clockevent_delta2ns(2, cd);
+	cd->min_delta_ticks	= 2;
+	cd->rating		= 200;
+	cd->irq			= irq;
+	cd->cpumask		= cpumask_of(cpu);
+	cd->set_next_event	= sibyte_next_event;
+	cd->set_state_shutdown	= sibyte_shutdown;
+	cd->set_state_periodic	= sibyte_set_periodic;
+	cd->set_state_oneshot	= sibyte_shutdown;
+	clockevents_register_device(cd);
+
+	bcm1480_mask_irq(cpu, irq);
+
+	/*
+	 * Map the timer interrupt to IP[4] of this cpu
+	 */
+	__raw_writeq(IMR_IP4_VAL,
+		     IOADDR(A_BCM1480_IMR_REGISTER(cpu,
+			R_BCM1480_IMR_INTERRUPT_MAP_BASE_H) + (irq << 3)));
+
+	bcm1480_unmask_irq(cpu, irq);
+
+	irq_set_affinity(irq, cpumask_of(cpu));
+	if (request_irq(irq, sibyte_counter_handler, flags, name, cd))
+		pr_err("Failed to request irq %d (%s)\n", irq, name);
+}
diff --git a/arch/mips/kernel/cevt-ds1287.c b/arch/mips/kernel/cevt-ds1287.c
new file mode 100644
index 000000000..9a47fbcd4
--- /dev/null
+++ b/arch/mips/kernel/cevt-ds1287.c
@@ -0,0 +1,121 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ *  DS1287 clockevent driver
+ *
+ *  Copyright (C) 2008	Yoichi Yuasa <yuasa@linux-mips.org>
+ */
+#include <linux/clockchips.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/mc146818rtc.h>
+#include <linux/irq.h>
+
+#include <asm/time.h>
+
+int ds1287_timer_state(void)
+{
+	return (CMOS_READ(RTC_REG_C) & RTC_PF) != 0;
+}
+
+int ds1287_set_base_clock(unsigned int hz)
+{
+	u8 rate;
+
+	switch (hz) {
+	case 128:
+		rate = 0x9;
+		break;
+	case 256:
+		rate = 0x8;
+		break;
+	case 1024:
+		rate = 0x6;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	CMOS_WRITE(RTC_REF_CLCK_32KHZ | rate, RTC_REG_A);
+
+	return 0;
+}
+
+static int ds1287_set_next_event(unsigned long delta,
+				 struct clock_event_device *evt)
+{
+	return -EINVAL;
+}
+
+static int ds1287_shutdown(struct clock_event_device *evt)
+{
+	u8 val;
+
+	spin_lock(&rtc_lock);
+
+	val = CMOS_READ(RTC_REG_B);
+	val &= ~RTC_PIE;
+	CMOS_WRITE(val, RTC_REG_B);
+
+	spin_unlock(&rtc_lock);
+	return 0;
+}
+
+static int ds1287_set_periodic(struct clock_event_device *evt)
+{
+	u8 val;
+
+	spin_lock(&rtc_lock);
+
+	val = CMOS_READ(RTC_REG_B);
+	val |= RTC_PIE;
+	CMOS_WRITE(val, RTC_REG_B);
+
+	spin_unlock(&rtc_lock);
+	return 0;
+}
+
+static void ds1287_event_handler(struct clock_event_device *dev)
+{
+}
+
+static struct clock_event_device ds1287_clockevent = {
+	.name			= "ds1287",
+	.features		= CLOCK_EVT_FEAT_PERIODIC,
+	.set_next_event		= ds1287_set_next_event,
+	.set_state_shutdown	= ds1287_shutdown,
+	.set_state_periodic	= ds1287_set_periodic,
+	.tick_resume		= ds1287_shutdown,
+	.event_handler		= ds1287_event_handler,
+};
+
+static irqreturn_t ds1287_interrupt(int irq, void *dev_id)
+{
+	struct clock_event_device *cd = &ds1287_clockevent;
+
+	/* Ack the RTC interrupt. */
+	CMOS_READ(RTC_REG_C);
+
+	cd->event_handler(cd);
+
+	return IRQ_HANDLED;
+}
+
+int __init ds1287_clockevent_init(int irq)
+{
+	unsigned long flags = IRQF_PERCPU | IRQF_TIMER;
+	struct clock_event_device *cd;
+
+	cd = &ds1287_clockevent;
+	cd->rating = 100;
+	cd->irq = irq;
+	clockevent_set_clock(cd, 32768);
+	cd->max_delta_ns = clockevent_delta2ns(0x7fffffff, cd);
+	cd->max_delta_ticks = 0x7fffffff;
+	cd->min_delta_ns = clockevent_delta2ns(0x300, cd);
+	cd->min_delta_ticks = 0x300;
+	cd->cpumask = cpumask_of(0);
+
+	clockevents_register_device(&ds1287_clockevent);
+
+	return request_irq(irq, ds1287_interrupt, flags, "ds1287", NULL);
+}
diff --git a/arch/mips/kernel/cevt-gt641xx.c b/arch/mips/kernel/cevt-gt641xx.c
new file mode 100644
index 000000000..5b132e8c5
--- /dev/null
+++ b/arch/mips/kernel/cevt-gt641xx.c
@@ -0,0 +1,146 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ *  GT641xx clockevent routines.
+ *
+ *  Copyright (C) 2007	Yoichi Yuasa <yuasa@linux-mips.org>
+ */
+#include <linux/clockchips.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/irq.h>
+
+#include <asm/gt64120.h>
+#include <asm/time.h>
+
+static DEFINE_RAW_SPINLOCK(gt641xx_timer_lock);
+static unsigned int gt641xx_base_clock;
+
+void gt641xx_set_base_clock(unsigned int clock)
+{
+	gt641xx_base_clock = clock;
+}
+
+int gt641xx_timer0_state(void)
+{
+	if (GT_READ(GT_TC0_OFS))
+		return 0;
+
+	GT_WRITE(GT_TC0_OFS, gt641xx_base_clock / HZ);
+	GT_WRITE(GT_TC_CONTROL_OFS, GT_TC_CONTROL_ENTC0_MSK);
+
+	return 1;
+}
+
+static int gt641xx_timer0_set_next_event(unsigned long delta,
+					 struct clock_event_device *evt)
+{
+	u32 ctrl;
+
+	raw_spin_lock(&gt641xx_timer_lock);
+
+	ctrl = GT_READ(GT_TC_CONTROL_OFS);
+	ctrl &= ~(GT_TC_CONTROL_ENTC0_MSK | GT_TC_CONTROL_SELTC0_MSK);
+	ctrl |= GT_TC_CONTROL_ENTC0_MSK;
+
+	GT_WRITE(GT_TC0_OFS, delta);
+	GT_WRITE(GT_TC_CONTROL_OFS, ctrl);
+
+	raw_spin_unlock(&gt641xx_timer_lock);
+
+	return 0;
+}
+
+static int gt641xx_timer0_shutdown(struct clock_event_device *evt)
+{
+	u32 ctrl;
+
+	raw_spin_lock(&gt641xx_timer_lock);
+
+	ctrl = GT_READ(GT_TC_CONTROL_OFS);
+	ctrl &= ~(GT_TC_CONTROL_ENTC0_MSK | GT_TC_CONTROL_SELTC0_MSK);
+	GT_WRITE(GT_TC_CONTROL_OFS, ctrl);
+
+	raw_spin_unlock(&gt641xx_timer_lock);
+	return 0;
+}
+
+static int gt641xx_timer0_set_oneshot(struct clock_event_device *evt)
+{
+	u32 ctrl;
+
+	raw_spin_lock(&gt641xx_timer_lock);
+
+	ctrl = GT_READ(GT_TC_CONTROL_OFS);
+	ctrl &= ~GT_TC_CONTROL_SELTC0_MSK;
+	ctrl |= GT_TC_CONTROL_ENTC0_MSK;
+	GT_WRITE(GT_TC_CONTROL_OFS, ctrl);
+
+	raw_spin_unlock(&gt641xx_timer_lock);
+	return 0;
+}
+
+static int gt641xx_timer0_set_periodic(struct clock_event_device *evt)
+{
+	u32 ctrl;
+
+	raw_spin_lock(&gt641xx_timer_lock);
+
+	ctrl = GT_READ(GT_TC_CONTROL_OFS);
+	ctrl |= GT_TC_CONTROL_ENTC0_MSK | GT_TC_CONTROL_SELTC0_MSK;
+	GT_WRITE(GT_TC_CONTROL_OFS, ctrl);
+
+	raw_spin_unlock(&gt641xx_timer_lock);
+	return 0;
+}
+
+static void gt641xx_timer0_event_handler(struct clock_event_device *dev)
+{
+}
+
+static struct clock_event_device gt641xx_timer0_clockevent = {
+	.name			= "gt641xx-timer0",
+	.features		= CLOCK_EVT_FEAT_PERIODIC |
+				  CLOCK_EVT_FEAT_ONESHOT,
+	.irq			= GT641XX_TIMER0_IRQ,
+	.set_next_event		= gt641xx_timer0_set_next_event,
+	.set_state_shutdown	= gt641xx_timer0_shutdown,
+	.set_state_periodic	= gt641xx_timer0_set_periodic,
+	.set_state_oneshot	= gt641xx_timer0_set_oneshot,
+	.tick_resume		= gt641xx_timer0_shutdown,
+	.event_handler		= gt641xx_timer0_event_handler,
+};
+
+static irqreturn_t gt641xx_timer0_interrupt(int irq, void *dev_id)
+{
+	struct clock_event_device *cd = &gt641xx_timer0_clockevent;
+
+	cd->event_handler(cd);
+
+	return IRQ_HANDLED;
+}
+
+static int __init gt641xx_timer0_clockevent_init(void)
+{
+	struct clock_event_device *cd;
+
+	if (!gt641xx_base_clock)
+		return 0;
+
+	GT_WRITE(GT_TC0_OFS, gt641xx_base_clock / HZ);
+
+	cd = &gt641xx_timer0_clockevent;
+	cd->rating = 200 + gt641xx_base_clock / 10000000;
+	clockevent_set_clock(cd, gt641xx_base_clock);
+	cd->max_delta_ns = clockevent_delta2ns(0x7fffffff, cd);
+	cd->max_delta_ticks = 0x7fffffff;
+	cd->min_delta_ns = clockevent_delta2ns(0x300, cd);
+	cd->min_delta_ticks = 0x300;
+	cd->cpumask = cpumask_of(0);
+
+	clockevents_register_device(&gt641xx_timer0_clockevent);
+
+	return request_irq(GT641XX_TIMER0_IRQ, gt641xx_timer0_interrupt,
+			   IRQF_PERCPU | IRQF_TIMER, "gt641xx_timer0", NULL);
+}
+arch_initcall(gt641xx_timer0_clockevent_init);
diff --git a/arch/mips/kernel/cevt-r4k.c b/arch/mips/kernel/cevt-r4k.c
new file mode 100644
index 000000000..995ad9e69
--- /dev/null
+++ b/arch/mips/kernel/cevt-r4k.c
@@ -0,0 +1,345 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2007 MIPS Technologies, Inc.
+ * Copyright (C) 2007 Ralf Baechle <ralf@linux-mips.org>
+ */
+#include <linux/clockchips.h>
+#include <linux/interrupt.h>
+#include <linux/cpufreq.h>
+#include <linux/percpu.h>
+#include <linux/smp.h>
+#include <linux/irq.h>
+
+#include <asm/time.h>
+#include <asm/cevt-r4k.h>
+
+static int mips_next_event(unsigned long delta,
+			   struct clock_event_device *evt)
+{
+	unsigned int cnt;
+	int res;
+
+	cnt = read_c0_count();
+	cnt += delta;
+	write_c0_compare(cnt);
+	res = ((int)(read_c0_count() - cnt) >= 0) ? -ETIME : 0;
+	return res;
+}
+
+/**
+ * calculate_min_delta() - Calculate a good minimum delta for mips_next_event().
+ *
+ * Running under virtualisation can introduce overhead into mips_next_event() in
+ * the form of hypervisor emulation of CP0_Count/CP0_Compare registers,
+ * potentially with an unnatural frequency, which makes a fixed min_delta_ns
+ * value inappropriate as it may be too small.
+ *
+ * It can also introduce occasional latency from the guest being descheduled.
+ *
+ * This function calculates a good minimum delta based roughly on the 75th
+ * percentile of the time taken to do the mips_next_event() sequence, in order
+ * to handle potentially higher overhead while also eliminating outliers due to
+ * unpredictable hypervisor latency (which can be handled by retries).
+ *
+ * Return:	An appropriate minimum delta for the clock event device.
+ */
+static unsigned int calculate_min_delta(void)
+{
+	unsigned int cnt, i, j, k, l;
+	unsigned int buf1[4], buf2[3];
+	unsigned int min_delta;
+
+	/*
+	 * Calculate the median of 5 75th percentiles of 5 samples of how long
+	 * it takes to set CP0_Compare = CP0_Count + delta.
+	 */
+	for (i = 0; i < 5; ++i) {
+		for (j = 0; j < 5; ++j) {
+			/*
+			 * This is like the code in mips_next_event(), and
+			 * directly measures the borderline "safe" delta.
+			 */
+			cnt = read_c0_count();
+			write_c0_compare(cnt);
+			cnt = read_c0_count() - cnt;
+
+			/* Sorted insert into buf1 */
+			for (k = 0; k < j; ++k) {
+				if (cnt < buf1[k]) {
+					l = min_t(unsigned int,
+						  j, ARRAY_SIZE(buf1) - 1);
+					for (; l > k; --l)
+						buf1[l] = buf1[l - 1];
+					break;
+				}
+			}
+			if (k < ARRAY_SIZE(buf1))
+				buf1[k] = cnt;
+		}
+
+		/* Sorted insert of 75th percentile into buf2 */
+		for (k = 0; k < i && k < ARRAY_SIZE(buf2); ++k) {
+			if (buf1[ARRAY_SIZE(buf1) - 1] < buf2[k]) {
+				l = min_t(unsigned int,
+					  i, ARRAY_SIZE(buf2) - 1);
+				for (; l > k; --l)
+					buf2[l] = buf2[l - 1];
+				break;
+			}
+		}
+		if (k < ARRAY_SIZE(buf2))
+			buf2[k] = buf1[ARRAY_SIZE(buf1) - 1];
+	}
+
+	/* Use 2 * median of 75th percentiles */
+	min_delta = buf2[ARRAY_SIZE(buf2) - 1] * 2;
+
+	/* Don't go too low */
+	if (min_delta < 0x300)
+		min_delta = 0x300;
+
+	pr_debug("%s: median 75th percentile=%#x, min_delta=%#x\n",
+		 __func__, buf2[ARRAY_SIZE(buf2) - 1], min_delta);
+	return min_delta;
+}
+
+DEFINE_PER_CPU(struct clock_event_device, mips_clockevent_device);
+int cp0_timer_irq_installed;
+
+/*
+ * Possibly handle a performance counter interrupt.
+ * Return true if the timer interrupt should not be checked
+ */
+static inline int handle_perf_irq(int r2)
+{
+	/*
+	 * The performance counter overflow interrupt may be shared with the
+	 * timer interrupt (cp0_perfcount_irq < 0). If it is and a
+	 * performance counter has overflowed (perf_irq() == IRQ_HANDLED)
+	 * and we can't reliably determine if a counter interrupt has also
+	 * happened (!r2) then don't check for a timer interrupt.
+	 */
+	return (cp0_perfcount_irq < 0) &&
+		perf_irq() == IRQ_HANDLED &&
+		!r2;
+}
+
+irqreturn_t c0_compare_interrupt(int irq, void *dev_id)
+{
+	const int r2 = cpu_has_mips_r2_r6;
+	struct clock_event_device *cd;
+	int cpu = smp_processor_id();
+
+	/*
+	 * Suckage alert:
+	 * Before R2 of the architecture there was no way to see if a
+	 * performance counter interrupt was pending, so we have to run
+	 * the performance counter interrupt handler anyway.
+	 */
+	if (handle_perf_irq(r2))
+		return IRQ_HANDLED;
+
+	/*
+	 * The same applies to performance counter interrupts.	But with the
+	 * above we now know that the reason we got here must be a timer
+	 * interrupt.  Being the paranoiacs we are we check anyway.
+	 */
+	if (!r2 || (read_c0_cause() & CAUSEF_TI)) {
+		/* Clear Count/Compare Interrupt */
+		write_c0_compare(read_c0_compare());
+		cd = &per_cpu(mips_clockevent_device, cpu);
+		cd->event_handler(cd);
+
+		return IRQ_HANDLED;
+	}
+
+	return IRQ_NONE;
+}
+
+struct irqaction c0_compare_irqaction = {
+	.handler = c0_compare_interrupt,
+	/*
+	 * IRQF_SHARED: The timer interrupt may be shared with other interrupts
+	 * such as perf counter and FDC interrupts.
+	 */
+	.flags = IRQF_PERCPU | IRQF_TIMER | IRQF_SHARED,
+	.name = "timer",
+};
+
+
+void mips_event_handler(struct clock_event_device *dev)
+{
+}
+
+/*
+ * FIXME: This doesn't hold for the relocated E9000 compare interrupt.
+ */
+static int c0_compare_int_pending(void)
+{
+	/* When cpu_has_mips_r2, this checks Cause.TI instead of Cause.IP7 */
+	return (read_c0_cause() >> cp0_compare_irq_shift) & (1ul << CAUSEB_IP);
+}
+
+/*
+ * Compare interrupt can be routed and latched outside the core,
+ * so wait up to worst case number of cycle counter ticks for timer interrupt
+ * changes to propagate to the cause register.
+ */
+#define COMPARE_INT_SEEN_TICKS 50
+
+int c0_compare_int_usable(void)
+{
+	unsigned int delta;
+	unsigned int cnt;
+
+#ifdef CONFIG_KVM_GUEST
+    return 1;
+#endif
+
+	/*
+	 * IP7 already pending?	 Try to clear it by acking the timer.
+	 */
+	if (c0_compare_int_pending()) {
+		cnt = read_c0_count();
+		write_c0_compare(cnt);
+		back_to_back_c0_hazard();
+		while (read_c0_count() < (cnt  + COMPARE_INT_SEEN_TICKS))
+			if (!c0_compare_int_pending())
+				break;
+		if (c0_compare_int_pending())
+			return 0;
+	}
+
+	for (delta = 0x10; delta <= 0x400000; delta <<= 1) {
+		cnt = read_c0_count();
+		cnt += delta;
+		write_c0_compare(cnt);
+		back_to_back_c0_hazard();
+		if ((int)(read_c0_count() - cnt) < 0)
+		    break;
+		/* increase delta if the timer was already expired */
+	}
+
+	while ((int)(read_c0_count() - cnt) <= 0)
+		;	/* Wait for expiry  */
+
+	while (read_c0_count() < (cnt + COMPARE_INT_SEEN_TICKS))
+		if (c0_compare_int_pending())
+			break;
+	if (!c0_compare_int_pending())
+		return 0;
+	cnt = read_c0_count();
+	write_c0_compare(cnt);
+	back_to_back_c0_hazard();
+	while (read_c0_count() < (cnt + COMPARE_INT_SEEN_TICKS))
+		if (!c0_compare_int_pending())
+			break;
+	if (c0_compare_int_pending())
+		return 0;
+
+	/*
+	 * Feels like a real count / compare timer.
+	 */
+	return 1;
+}
+
+unsigned int __weak get_c0_compare_int(void)
+{
+	return MIPS_CPU_IRQ_BASE + cp0_compare_irq;
+}
+
+#ifdef CONFIG_CPU_FREQ
+
+static unsigned long mips_ref_freq;
+
+static int r4k_cpufreq_callback(struct notifier_block *nb,
+				unsigned long val, void *data)
+{
+	struct cpufreq_freqs *freq = data;
+	struct clock_event_device *cd;
+	unsigned long rate;
+	int cpu;
+
+	if (!mips_ref_freq)
+		mips_ref_freq = freq->old;
+
+	if (val == CPUFREQ_POSTCHANGE) {
+		rate = cpufreq_scale(mips_hpt_frequency, mips_ref_freq,
+				     freq->new);
+
+		for_each_cpu(cpu, freq->policy->cpus) {
+			cd = &per_cpu(mips_clockevent_device, cpu);
+
+			clockevents_update_freq(cd, rate);
+		}
+	}
+
+	return 0;
+}
+
+static struct notifier_block r4k_cpufreq_notifier = {
+	.notifier_call  = r4k_cpufreq_callback,
+};
+
+static int __init r4k_register_cpufreq_notifier(void)
+{
+	return cpufreq_register_notifier(&r4k_cpufreq_notifier,
+					 CPUFREQ_TRANSITION_NOTIFIER);
+
+}
+core_initcall(r4k_register_cpufreq_notifier);
+
+#endif /* !CONFIG_CPU_FREQ */
+
+int r4k_clockevent_init(void)
+{
+	unsigned long flags = IRQF_PERCPU | IRQF_TIMER | IRQF_SHARED;
+	unsigned int cpu = smp_processor_id();
+	struct clock_event_device *cd;
+	unsigned int irq, min_delta;
+
+	if (!cpu_has_counter || !mips_hpt_frequency)
+		return -ENXIO;
+
+	if (!c0_compare_int_usable())
+		return -ENXIO;
+
+	/*
+	 * With vectored interrupts things are getting platform specific.
+	 * get_c0_compare_int is a hook to allow a platform to return the
+	 * interrupt number of its liking.
+	 */
+	irq = get_c0_compare_int();
+
+	cd = &per_cpu(mips_clockevent_device, cpu);
+
+	cd->name		= "MIPS";
+	cd->features		= CLOCK_EVT_FEAT_ONESHOT |
+				  CLOCK_EVT_FEAT_C3STOP |
+				  CLOCK_EVT_FEAT_PERCPU;
+
+	min_delta		= calculate_min_delta();
+
+	cd->rating		= 300;
+	cd->irq			= irq;
+	cd->cpumask		= cpumask_of(cpu);
+	cd->set_next_event	= mips_next_event;
+	cd->event_handler	= mips_event_handler;
+
+	clockevents_config_and_register(cd, mips_hpt_frequency, min_delta, 0x7fffffff);
+
+	if (cp0_timer_irq_installed)
+		return 0;
+
+	cp0_timer_irq_installed = 1;
+
+	if (request_irq(irq, c0_compare_interrupt, flags, "timer",
+			c0_compare_interrupt))
+		pr_err("Failed to request irq %d (timer)\n", irq);
+
+	return 0;
+}
+
diff --git a/arch/mips/kernel/cevt-sb1250.c b/arch/mips/kernel/cevt-sb1250.c
new file mode 100644
index 000000000..0451273fa
--- /dev/null
+++ b/arch/mips/kernel/cevt-sb1250.c
@@ -0,0 +1,138 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2000, 2001 Broadcom Corporation
+ */
+#include <linux/clockchips.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/percpu.h>
+#include <linux/smp.h>
+
+#include <asm/addrspace.h>
+#include <asm/io.h>
+#include <asm/time.h>
+
+#include <asm/sibyte/sb1250.h>
+#include <asm/sibyte/sb1250_regs.h>
+#include <asm/sibyte/sb1250_int.h>
+#include <asm/sibyte/sb1250_scd.h>
+
+#define IMR_IP2_VAL	K_INT_MAP_I0
+#define IMR_IP3_VAL	K_INT_MAP_I1
+#define IMR_IP4_VAL	K_INT_MAP_I2
+
+/*
+ * The general purpose timer ticks at 1MHz independent if
+ * the rest of the system
+ */
+
+static int sibyte_shutdown(struct clock_event_device *evt)
+{
+	void __iomem *cfg;
+
+	cfg = IOADDR(A_SCD_TIMER_REGISTER(smp_processor_id(), R_SCD_TIMER_CFG));
+
+	/* Stop the timer until we actually program a shot */
+	__raw_writeq(0, cfg);
+
+	return 0;
+}
+
+static int sibyte_set_periodic(struct clock_event_device *evt)
+{
+	unsigned int cpu = smp_processor_id();
+	void __iomem *cfg, *init;
+
+	cfg = IOADDR(A_SCD_TIMER_REGISTER(cpu, R_SCD_TIMER_CFG));
+	init = IOADDR(A_SCD_TIMER_REGISTER(cpu, R_SCD_TIMER_INIT));
+
+	__raw_writeq(0, cfg);
+	__raw_writeq((V_SCD_TIMER_FREQ / HZ) - 1, init);
+	__raw_writeq(M_SCD_TIMER_ENABLE | M_SCD_TIMER_MODE_CONTINUOUS, cfg);
+
+	return 0;
+}
+
+static int sibyte_next_event(unsigned long delta, struct clock_event_device *cd)
+{
+	unsigned int cpu = smp_processor_id();
+	void __iomem *cfg, *init;
+
+	cfg = IOADDR(A_SCD_TIMER_REGISTER(cpu, R_SCD_TIMER_CFG));
+	init = IOADDR(A_SCD_TIMER_REGISTER(cpu, R_SCD_TIMER_INIT));
+
+	__raw_writeq(0, cfg);
+	__raw_writeq(delta - 1, init);
+	__raw_writeq(M_SCD_TIMER_ENABLE, cfg);
+
+	return 0;
+}
+
+static irqreturn_t sibyte_counter_handler(int irq, void *dev_id)
+{
+	unsigned int cpu = smp_processor_id();
+	struct clock_event_device *cd = dev_id;
+	void __iomem *cfg;
+	unsigned long tmode;
+
+	if (clockevent_state_periodic(cd))
+		tmode = M_SCD_TIMER_ENABLE | M_SCD_TIMER_MODE_CONTINUOUS;
+	else
+		tmode = 0;
+
+	/* ACK interrupt */
+	cfg = IOADDR(A_SCD_TIMER_REGISTER(cpu, R_SCD_TIMER_CFG));
+	____raw_writeq(tmode, cfg);
+
+	cd->event_handler(cd);
+
+	return IRQ_HANDLED;
+}
+
+static DEFINE_PER_CPU(struct clock_event_device, sibyte_hpt_clockevent);
+static DEFINE_PER_CPU(char [18], sibyte_hpt_name);
+
+void sb1250_clockevent_init(void)
+{
+	unsigned int cpu = smp_processor_id();
+	unsigned int irq = K_INT_TIMER_0 + cpu;
+	struct clock_event_device *cd = &per_cpu(sibyte_hpt_clockevent, cpu);
+	unsigned char *name = per_cpu(sibyte_hpt_name, cpu);
+	unsigned long flags = IRQF_PERCPU | IRQF_TIMER;
+
+	/* Only have 4 general purpose timers, and we use last one as hpt */
+	BUG_ON(cpu > 2);
+
+	sprintf(name, "sb1250-counter-%d", cpu);
+	cd->name		= name;
+	cd->features		= CLOCK_EVT_FEAT_PERIODIC |
+				  CLOCK_EVT_FEAT_ONESHOT;
+	clockevent_set_clock(cd, V_SCD_TIMER_FREQ);
+	cd->max_delta_ns	= clockevent_delta2ns(0x7fffff, cd);
+	cd->max_delta_ticks	= 0x7fffff;
+	cd->min_delta_ns	= clockevent_delta2ns(2, cd);
+	cd->min_delta_ticks	= 2;
+	cd->rating		= 200;
+	cd->irq			= irq;
+	cd->cpumask		= cpumask_of(cpu);
+	cd->set_next_event	= sibyte_next_event;
+	cd->set_state_shutdown	= sibyte_shutdown;
+	cd->set_state_periodic	= sibyte_set_periodic;
+	cd->set_state_oneshot	= sibyte_shutdown;
+	clockevents_register_device(cd);
+
+	sb1250_mask_irq(cpu, irq);
+
+	/*
+	 * Map the timer interrupt to IP[4] of this cpu
+	 */
+	__raw_writeq(IMR_IP4_VAL,
+		     IOADDR(A_IMR_REGISTER(cpu, R_IMR_INTERRUPT_MAP_BASE) +
+			    (irq << 3)));
+
+	sb1250_unmask_irq(cpu, irq);
+
+	irq_set_affinity(irq, cpumask_of(cpu));
+	if (request_irq(irq, sibyte_counter_handler, flags, name, cd))
+		pr_err("Failed to request irq %d (%s)\n", irq, name);
+}
diff --git a/arch/mips/kernel/cevt-txx9.c b/arch/mips/kernel/cevt-txx9.c
new file mode 100644
index 000000000..5709469c2
--- /dev/null
+++ b/arch/mips/kernel/cevt-txx9.c
@@ -0,0 +1,220 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Based on linux/arch/mips/kernel/cevt-r4k.c,
+ *	    linux/arch/mips/jmr3927/rbhma3100/setup.c
+ *
+ * Copyright 2001 MontaVista Software Inc.
+ * Copyright (C) 2000-2001 Toshiba Corporation
+ * Copyright (C) 2007 MIPS Technologies, Inc.
+ * Copyright (C) 2007 Ralf Baechle <ralf@linux-mips.org>
+ */
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/sched_clock.h>
+#include <asm/time.h>
+#include <asm/txx9tmr.h>
+
+#define TCR_BASE (TXx9_TMTCR_CCDE | TXx9_TMTCR_CRE | TXx9_TMTCR_TMODE_ITVL)
+#define TIMER_CCD	0	/* 1/2 */
+#define TIMER_CLK(imclk)	((imclk) / (2 << TIMER_CCD))
+
+struct txx9_clocksource {
+	struct clocksource cs;
+	struct txx9_tmr_reg __iomem *tmrptr;
+};
+
+static u64 txx9_cs_read(struct clocksource *cs)
+{
+	struct txx9_clocksource *txx9_cs =
+		container_of(cs, struct txx9_clocksource, cs);
+	return __raw_readl(&txx9_cs->tmrptr->trr);
+}
+
+/* Use 1 bit smaller width to use full bits in that width */
+#define TXX9_CLOCKSOURCE_BITS (TXX9_TIMER_BITS - 1)
+
+static struct txx9_clocksource txx9_clocksource = {
+	.cs = {
+		.name		= "TXx9",
+		.rating		= 200,
+		.read		= txx9_cs_read,
+		.mask		= CLOCKSOURCE_MASK(TXX9_CLOCKSOURCE_BITS),
+		.flags		= CLOCK_SOURCE_IS_CONTINUOUS,
+	},
+};
+
+static u64 notrace txx9_read_sched_clock(void)
+{
+	return __raw_readl(&txx9_clocksource.tmrptr->trr);
+}
+
+void __init txx9_clocksource_init(unsigned long baseaddr,
+				  unsigned int imbusclk)
+{
+	struct txx9_tmr_reg __iomem *tmrptr;
+
+	clocksource_register_hz(&txx9_clocksource.cs, TIMER_CLK(imbusclk));
+
+	tmrptr = ioremap(baseaddr, sizeof(struct txx9_tmr_reg));
+	__raw_writel(TCR_BASE, &tmrptr->tcr);
+	__raw_writel(0, &tmrptr->tisr);
+	__raw_writel(TIMER_CCD, &tmrptr->ccdr);
+	__raw_writel(TXx9_TMITMR_TZCE, &tmrptr->itmr);
+	__raw_writel(1 << TXX9_CLOCKSOURCE_BITS, &tmrptr->cpra);
+	__raw_writel(TCR_BASE | TXx9_TMTCR_TCE, &tmrptr->tcr);
+	txx9_clocksource.tmrptr = tmrptr;
+
+	sched_clock_register(txx9_read_sched_clock, TXX9_CLOCKSOURCE_BITS,
+			     TIMER_CLK(imbusclk));
+}
+
+struct txx9_clock_event_device {
+	struct clock_event_device cd;
+	struct txx9_tmr_reg __iomem *tmrptr;
+};
+
+static void txx9tmr_stop_and_clear(struct txx9_tmr_reg __iomem *tmrptr)
+{
+	/* stop and reset counter */
+	__raw_writel(TCR_BASE, &tmrptr->tcr);
+	/* clear pending interrupt */
+	__raw_writel(0, &tmrptr->tisr);
+}
+
+static int txx9tmr_set_state_periodic(struct clock_event_device *evt)
+{
+	struct txx9_clock_event_device *txx9_cd =
+		container_of(evt, struct txx9_clock_event_device, cd);
+	struct txx9_tmr_reg __iomem *tmrptr = txx9_cd->tmrptr;
+
+	txx9tmr_stop_and_clear(tmrptr);
+
+	__raw_writel(TXx9_TMITMR_TIIE | TXx9_TMITMR_TZCE, &tmrptr->itmr);
+	/* start timer */
+	__raw_writel(((u64)(NSEC_PER_SEC / HZ) * evt->mult) >> evt->shift,
+		     &tmrptr->cpra);
+	__raw_writel(TCR_BASE | TXx9_TMTCR_TCE, &tmrptr->tcr);
+	return 0;
+}
+
+static int txx9tmr_set_state_oneshot(struct clock_event_device *evt)
+{
+	struct txx9_clock_event_device *txx9_cd =
+		container_of(evt, struct txx9_clock_event_device, cd);
+	struct txx9_tmr_reg __iomem *tmrptr = txx9_cd->tmrptr;
+
+	txx9tmr_stop_and_clear(tmrptr);
+	__raw_writel(TXx9_TMITMR_TIIE, &tmrptr->itmr);
+	return 0;
+}
+
+static int txx9tmr_set_state_shutdown(struct clock_event_device *evt)
+{
+	struct txx9_clock_event_device *txx9_cd =
+		container_of(evt, struct txx9_clock_event_device, cd);
+	struct txx9_tmr_reg __iomem *tmrptr = txx9_cd->tmrptr;
+
+	txx9tmr_stop_and_clear(tmrptr);
+	__raw_writel(0, &tmrptr->itmr);
+	return 0;
+}
+
+static int txx9tmr_tick_resume(struct clock_event_device *evt)
+{
+	struct txx9_clock_event_device *txx9_cd =
+		container_of(evt, struct txx9_clock_event_device, cd);
+	struct txx9_tmr_reg __iomem *tmrptr = txx9_cd->tmrptr;
+
+	txx9tmr_stop_and_clear(tmrptr);
+	__raw_writel(TIMER_CCD, &tmrptr->ccdr);
+	__raw_writel(0, &tmrptr->itmr);
+	return 0;
+}
+
+static int txx9tmr_set_next_event(unsigned long delta,
+				  struct clock_event_device *evt)
+{
+	struct txx9_clock_event_device *txx9_cd =
+		container_of(evt, struct txx9_clock_event_device, cd);
+	struct txx9_tmr_reg __iomem *tmrptr = txx9_cd->tmrptr;
+
+	txx9tmr_stop_and_clear(tmrptr);
+	/* start timer */
+	__raw_writel(delta, &tmrptr->cpra);
+	__raw_writel(TCR_BASE | TXx9_TMTCR_TCE, &tmrptr->tcr);
+	return 0;
+}
+
+static struct txx9_clock_event_device txx9_clock_event_device = {
+	.cd = {
+		.name			= "TXx9",
+		.features		= CLOCK_EVT_FEAT_PERIODIC |
+					  CLOCK_EVT_FEAT_ONESHOT,
+		.rating			= 200,
+		.set_state_shutdown	= txx9tmr_set_state_shutdown,
+		.set_state_periodic	= txx9tmr_set_state_periodic,
+		.set_state_oneshot	= txx9tmr_set_state_oneshot,
+		.tick_resume		= txx9tmr_tick_resume,
+		.set_next_event		= txx9tmr_set_next_event,
+	},
+};
+
+static irqreturn_t txx9tmr_interrupt(int irq, void *dev_id)
+{
+	struct txx9_clock_event_device *txx9_cd = dev_id;
+	struct clock_event_device *cd = &txx9_cd->cd;
+	struct txx9_tmr_reg __iomem *tmrptr = txx9_cd->tmrptr;
+
+	__raw_writel(0, &tmrptr->tisr); /* ack interrupt */
+	cd->event_handler(cd);
+	return IRQ_HANDLED;
+}
+
+void __init txx9_clockevent_init(unsigned long baseaddr, int irq,
+				 unsigned int imbusclk)
+{
+	struct clock_event_device *cd = &txx9_clock_event_device.cd;
+	struct txx9_tmr_reg __iomem *tmrptr;
+
+	tmrptr = ioremap(baseaddr, sizeof(struct txx9_tmr_reg));
+	txx9tmr_stop_and_clear(tmrptr);
+	__raw_writel(TIMER_CCD, &tmrptr->ccdr);
+	__raw_writel(0, &tmrptr->itmr);
+	txx9_clock_event_device.tmrptr = tmrptr;
+
+	clockevent_set_clock(cd, TIMER_CLK(imbusclk));
+	cd->max_delta_ns =
+		clockevent_delta2ns(0xffffffff >> (32 - TXX9_TIMER_BITS), cd);
+	cd->max_delta_ticks = 0xffffffff >> (32 - TXX9_TIMER_BITS);
+	cd->min_delta_ns = clockevent_delta2ns(0xf, cd);
+	cd->min_delta_ticks = 0xf;
+	cd->irq = irq;
+	cd->cpumask = cpumask_of(0),
+	clockevents_register_device(cd);
+	if (request_irq(irq, txx9tmr_interrupt, IRQF_PERCPU | IRQF_TIMER,
+			"txx9tmr", &txx9_clock_event_device))
+		pr_err("Failed to request irq %d (txx9tmr)\n", irq);
+	printk(KERN_INFO "TXx9: clockevent device at 0x%lx, irq %d\n",
+	       baseaddr, irq);
+}
+
+void __init txx9_tmr_init(unsigned long baseaddr)
+{
+	struct txx9_tmr_reg __iomem *tmrptr;
+
+	tmrptr = ioremap(baseaddr, sizeof(struct txx9_tmr_reg));
+	/* Start once to make CounterResetEnable effective */
+	__raw_writel(TXx9_TMTCR_CRE | TXx9_TMTCR_TCE, &tmrptr->tcr);
+	/* Stop and reset the counter */
+	__raw_writel(TXx9_TMTCR_CRE, &tmrptr->tcr);
+	__raw_writel(0, &tmrptr->tisr);
+	__raw_writel(0xffffffff, &tmrptr->cpra);
+	__raw_writel(0, &tmrptr->itmr);
+	__raw_writel(0, &tmrptr->ccdr);
+	__raw_writel(0, &tmrptr->pgmr);
+	iounmap(tmrptr);
+}
diff --git a/arch/mips/kernel/cmpxchg.c b/arch/mips/kernel/cmpxchg.c
new file mode 100644
index 000000000..89107deb0
--- /dev/null
+++ b/arch/mips/kernel/cmpxchg.c
@@ -0,0 +1,104 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2017 Imagination Technologies
+ * Author: Paul Burton <paul.burton@mips.com>
+ */
+
+#include <linux/bitops.h>
+#include <asm/cmpxchg.h>
+
+unsigned long __xchg_small(volatile void *ptr, unsigned long val, unsigned int size)
+{
+	u32 old32, new32, load32, mask;
+	volatile u32 *ptr32;
+	unsigned int shift;
+
+	/* Check that ptr is naturally aligned */
+	WARN_ON((unsigned long)ptr & (size - 1));
+
+	/* Mask value to the correct size. */
+	mask = GENMASK((size * BITS_PER_BYTE) - 1, 0);
+	val &= mask;
+
+	/*
+	 * Calculate a shift & mask that correspond to the value we wish to
+	 * exchange within the naturally aligned 4 byte integerthat includes
+	 * it.
+	 */
+	shift = (unsigned long)ptr & 0x3;
+	if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN))
+		shift ^= sizeof(u32) - size;
+	shift *= BITS_PER_BYTE;
+	mask <<= shift;
+
+	/*
+	 * Calculate a pointer to the naturally aligned 4 byte integer that
+	 * includes our byte of interest, and load its value.
+	 */
+	ptr32 = (volatile u32 *)((unsigned long)ptr & ~0x3);
+	load32 = *ptr32;
+
+	do {
+		old32 = load32;
+		new32 = (load32 & ~mask) | (val << shift);
+		load32 = cmpxchg(ptr32, old32, new32);
+	} while (load32 != old32);
+
+	return (load32 & mask) >> shift;
+}
+
+unsigned long __cmpxchg_small(volatile void *ptr, unsigned long old,
+			      unsigned long new, unsigned int size)
+{
+	u32 mask, old32, new32, load32, load;
+	volatile u32 *ptr32;
+	unsigned int shift;
+
+	/* Check that ptr is naturally aligned */
+	WARN_ON((unsigned long)ptr & (size - 1));
+
+	/* Mask inputs to the correct size. */
+	mask = GENMASK((size * BITS_PER_BYTE) - 1, 0);
+	old &= mask;
+	new &= mask;
+
+	/*
+	 * Calculate a shift & mask that correspond to the value we wish to
+	 * compare & exchange within the naturally aligned 4 byte integer
+	 * that includes it.
+	 */
+	shift = (unsigned long)ptr & 0x3;
+	if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN))
+		shift ^= sizeof(u32) - size;
+	shift *= BITS_PER_BYTE;
+	mask <<= shift;
+
+	/*
+	 * Calculate a pointer to the naturally aligned 4 byte integer that
+	 * includes our byte of interest, and load its value.
+	 */
+	ptr32 = (volatile u32 *)((unsigned long)ptr & ~0x3);
+	load32 = *ptr32;
+
+	while (true) {
+		/*
+		 * Ensure the byte we want to exchange matches the expected
+		 * old value, and if not then bail.
+		 */
+		load = (load32 & mask) >> shift;
+		if (load != old)
+			return load;
+
+		/*
+		 * Calculate the old & new values of the naturally aligned
+		 * 4 byte integer that include the byte we want to exchange.
+		 * Attempt to exchange the old value for the new value, and
+		 * return if we succeed.
+		 */
+		old32 = (load32 & ~mask) | (old << shift);
+		new32 = (load32 & ~mask) | (new << shift);
+		load32 = cmpxchg(ptr32, old32, new32);
+		if (load32 == old32)
+			return old;
+	}
+}
diff --git a/arch/mips/kernel/cps-vec-ns16550.S b/arch/mips/kernel/cps-vec-ns16550.S
new file mode 100644
index 000000000..30725e1df
--- /dev/null
+++ b/arch/mips/kernel/cps-vec-ns16550.S
@@ -0,0 +1,212 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Copyright (C) 2015 Imagination Technologies
+ * Author: Paul Burton <paul.burton@mips.com>
+ */
+
+#include <asm/addrspace.h>
+#include <asm/asm.h>
+#include <asm/asm-offsets.h>
+#include <asm/mipsregs.h>
+#include <asm/regdef.h>
+#include <linux/serial_reg.h>
+
+#define UART_TX_OFS	(UART_TX << CONFIG_MIPS_CPS_NS16550_SHIFT)
+#define UART_LSR_OFS	(UART_LSR << CONFIG_MIPS_CPS_NS16550_SHIFT)
+
+#if CONFIG_MIPS_CPS_NS16550_WIDTH == 1
+# define UART_L		lb
+# define UART_S		sb
+#elif CONFIG_MIPS_CPS_NS16550_WIDTH == 2
+# define UART_L		lh
+# define UART_S		sh
+#elif CONFIG_MIPS_CPS_NS16550_WIDTH == 4
+# define UART_L		lw
+# define UART_S		sw
+#else
+# define UART_L		lb
+# define UART_S		sb
+#endif
+
+/**
+ * _mips_cps_putc() - write a character to the UART
+ * @a0: ASCII character to write
+ * @t9: UART base address
+ */
+LEAF(_mips_cps_putc)
+1:	UART_L		t0, UART_LSR_OFS(t9)
+	andi		t0, t0, UART_LSR_TEMT
+	beqz		t0, 1b
+	UART_S		a0, UART_TX_OFS(t9)
+	jr		ra
+	END(_mips_cps_putc)
+
+/**
+ * _mips_cps_puts() - write a string to the UART
+ * @a0: pointer to NULL-terminated ASCII string
+ * @t9: UART base address
+ *
+ * Write a null-terminated ASCII string to the UART.
+ */
+NESTED(_mips_cps_puts, 0, ra)
+	move		s7, ra
+	move		s6, a0
+
+1:	lb		a0, 0(s6)
+	beqz		a0, 2f
+	jal		_mips_cps_putc
+	PTR_ADDIU	s6, s6, 1
+	b		1b
+
+2:	jr		s7
+	END(_mips_cps_puts)
+
+/**
+ * _mips_cps_putx4 - write a 4b hex value to the UART
+ * @a0: the 4b value to write to the UART
+ * @t9: UART base address
+ *
+ * Write a single hexadecimal character to the UART.
+ */
+NESTED(_mips_cps_putx4, 0, ra)
+	andi		a0, a0, 0xf
+	li		t0, '0'
+	blt		a0, 10, 1f
+	li		t0, 'a'
+	addiu		a0, a0, -10
+1:	addu		a0, a0, t0
+	b		_mips_cps_putc
+	END(_mips_cps_putx4)
+
+/**
+ * _mips_cps_putx8 - write an 8b hex value to the UART
+ * @a0: the 8b value to write to the UART
+ * @t9: UART base address
+ *
+ * Write an 8 bit value (ie. 2 hexadecimal characters) to the UART.
+ */
+NESTED(_mips_cps_putx8, 0, ra)
+	move		s3, ra
+	move		s2, a0
+	srl		a0, a0, 4
+	jal		_mips_cps_putx4
+	move		a0, s2
+	move		ra, s3
+	b		_mips_cps_putx4
+	END(_mips_cps_putx8)
+
+/**
+ * _mips_cps_putx16 - write a 16b hex value to the UART
+ * @a0: the 16b value to write to the UART
+ * @t9: UART base address
+ *
+ * Write a 16 bit value (ie. 4 hexadecimal characters) to the UART.
+ */
+NESTED(_mips_cps_putx16, 0, ra)
+	move		s5, ra
+	move		s4, a0
+	srl		a0, a0, 8
+	jal		_mips_cps_putx8
+	move		a0, s4
+	move		ra, s5
+	b		_mips_cps_putx8
+	END(_mips_cps_putx16)
+
+/**
+ * _mips_cps_putx32 - write a 32b hex value to the UART
+ * @a0: the 32b value to write to the UART
+ * @t9: UART base address
+ *
+ * Write a 32 bit value (ie. 8 hexadecimal characters) to the UART.
+ */
+NESTED(_mips_cps_putx32, 0, ra)
+	move		s7, ra
+	move		s6, a0
+	srl		a0, a0, 16
+	jal		_mips_cps_putx16
+	move		a0, s6
+	move		ra, s7
+	b		_mips_cps_putx16
+	END(_mips_cps_putx32)
+
+#ifdef CONFIG_64BIT
+
+/**
+ * _mips_cps_putx64 - write a 64b hex value to the UART
+ * @a0: the 64b value to write to the UART
+ * @t9: UART base address
+ *
+ * Write a 64 bit value (ie. 16 hexadecimal characters) to the UART.
+ */
+NESTED(_mips_cps_putx64, 0, ra)
+	move		sp, ra
+	move		s8, a0
+	dsrl32		a0, a0, 0
+	jal		_mips_cps_putx32
+	move		a0, s8
+	move		ra, sp
+	b		_mips_cps_putx32
+	END(_mips_cps_putx64)
+
+#define _mips_cps_putxlong _mips_cps_putx64
+
+#else /* !CONFIG_64BIT */
+
+#define _mips_cps_putxlong _mips_cps_putx32
+
+#endif /* !CONFIG_64BIT */
+
+/**
+ * mips_cps_bev_dump() - dump relevant exception state to UART
+ * @a0: pointer to NULL-terminated ASCII string naming the exception
+ *
+ * Write information that may be useful in debugging an exception to the
+ * UART configured by CONFIG_MIPS_CPS_NS16550_*. As this BEV exception
+ * will only be run if something goes horribly wrong very early during
+ * the bringup of a core and it is very likely to be unsafe to perform
+ * memory accesses at that point (cache state indeterminate, EVA may not
+ * be configured, coherence may be disabled) let alone have a stack,
+ * this is all written in assembly using only registers & unmapped
+ * uncached access to the UART registers.
+ */
+LEAF(mips_cps_bev_dump)
+	move		s0, ra
+	move		s1, a0
+
+	li		t9, CKSEG1ADDR(CONFIG_MIPS_CPS_NS16550_BASE)
+
+	PTR_LA		a0, str_newline
+	jal		_mips_cps_puts
+	PTR_LA		a0, str_bev
+	jal		_mips_cps_puts
+	move		a0, s1
+	jal		_mips_cps_puts
+	PTR_LA		a0, str_newline
+	jal		_mips_cps_puts
+	PTR_LA		a0, str_newline
+	jal		_mips_cps_puts
+
+#define DUMP_COP0_REG(reg, name, sz, _mfc0)		\
+	PTR_LA		a0, 8f;				\
+	jal		_mips_cps_puts;			\
+	_mfc0		a0, reg;			\
+	jal		_mips_cps_putx##sz;		\
+	PTR_LA		a0, str_newline;		\
+	jal		_mips_cps_puts;			\
+	TEXT(name)
+
+	DUMP_COP0_REG(CP0_CAUSE,    "Cause:    0x", 32, mfc0)
+	DUMP_COP0_REG(CP0_STATUS,   "Status:   0x", 32, mfc0)
+	DUMP_COP0_REG(CP0_EBASE,    "EBase:    0x", long, MFC0)
+	DUMP_COP0_REG(CP0_BADVADDR, "BadVAddr: 0x", long, MFC0)
+	DUMP_COP0_REG(CP0_BADINSTR, "BadInstr: 0x", 32, mfc0)
+
+	PTR_LA		a0, str_newline
+	jal		_mips_cps_puts
+	jr		s0
+	END(mips_cps_bev_dump)
+
+.pushsection	.data
+str_bev: .asciiz "BEV Exception: "
+str_newline: .asciiz "\r\n"
+.popsection
diff --git a/arch/mips/kernel/cps-vec.S b/arch/mips/kernel/cps-vec.S
new file mode 100644
index 000000000..4db7ff055
--- /dev/null
+++ b/arch/mips/kernel/cps-vec.S
@@ -0,0 +1,630 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Copyright (C) 2013 Imagination Technologies
+ * Author: Paul Burton <paul.burton@mips.com>
+ */
+
+#include <asm/addrspace.h>
+#include <asm/asm.h>
+#include <asm/asm-offsets.h>
+#include <asm/asmmacro.h>
+#include <asm/cacheops.h>
+#include <asm/eva.h>
+#include <asm/mipsregs.h>
+#include <asm/mipsmtregs.h>
+#include <asm/pm.h>
+
+#define GCR_CPC_BASE_OFS	0x0088
+#define GCR_CL_COHERENCE_OFS	0x2008
+#define GCR_CL_ID_OFS		0x2028
+
+#define CPC_CL_VC_STOP_OFS	0x2020
+#define CPC_CL_VC_RUN_OFS	0x2028
+
+.extern mips_cm_base
+
+.set noreorder
+
+#ifdef CONFIG_64BIT
+# define STATUS_BITDEPS		ST0_KX
+#else
+# define STATUS_BITDEPS		0
+#endif
+
+#ifdef CONFIG_MIPS_CPS_NS16550
+
+#define DUMP_EXCEP(name)		\
+	PTR_LA	a0, 8f;			\
+	jal	mips_cps_bev_dump;	\
+	 nop;				\
+	TEXT(name)
+
+#else /* !CONFIG_MIPS_CPS_NS16550 */
+
+#define DUMP_EXCEP(name)
+
+#endif /* !CONFIG_MIPS_CPS_NS16550 */
+
+	/*
+	 * Set dest to non-zero if the core supports the MT ASE, else zero. If
+	 * MT is not supported then branch to nomt.
+	 */
+	.macro	has_mt	dest, nomt
+	mfc0	\dest, CP0_CONFIG, 1
+	bgez	\dest, \nomt
+	 mfc0	\dest, CP0_CONFIG, 2
+	bgez	\dest, \nomt
+	 mfc0	\dest, CP0_CONFIG, 3
+	andi	\dest, \dest, MIPS_CONF3_MT
+	beqz	\dest, \nomt
+	 nop
+	.endm
+
+	/*
+	 * Set dest to non-zero if the core supports MIPSr6 multithreading
+	 * (ie. VPs), else zero. If MIPSr6 multithreading is not supported then
+	 * branch to nomt.
+	 */
+	.macro	has_vp	dest, nomt
+	mfc0	\dest, CP0_CONFIG, 1
+	bgez	\dest, \nomt
+	 mfc0	\dest, CP0_CONFIG, 2
+	bgez	\dest, \nomt
+	 mfc0	\dest, CP0_CONFIG, 3
+	bgez	\dest, \nomt
+	 mfc0	\dest, CP0_CONFIG, 4
+	bgez	\dest, \nomt
+	 mfc0	\dest, CP0_CONFIG, 5
+	andi	\dest, \dest, MIPS_CONF5_VP
+	beqz	\dest, \nomt
+	 nop
+	.endm
+
+	/* Calculate an uncached address for the CM GCRs */
+	.macro	cmgcrb	dest
+	.set	push
+	.set	noat
+	MFC0	$1, CP0_CMGCRBASE
+	PTR_SLL	$1, $1, 4
+	PTR_LI	\dest, UNCAC_BASE
+	PTR_ADDU \dest, \dest, $1
+	.set	pop
+	.endm
+
+.section .text.cps-vec
+.balign 0x1000
+
+LEAF(mips_cps_core_entry)
+	/*
+	 * These first 4 bytes will be patched by cps_smp_setup to load the
+	 * CCA to use into register s0.
+	 */
+	.word	0
+
+	/* Check whether we're here due to an NMI */
+	mfc0	k0, CP0_STATUS
+	and	k0, k0, ST0_NMI
+	beqz	k0, not_nmi
+	 nop
+
+	/* This is an NMI */
+	PTR_LA	k0, nmi_handler
+	jr	k0
+	 nop
+
+not_nmi:
+	/* Setup Cause */
+	li	t0, CAUSEF_IV
+	mtc0	t0, CP0_CAUSE
+
+	/* Setup Status */
+	li	t0, ST0_CU1 | ST0_CU0 | ST0_BEV | STATUS_BITDEPS
+	mtc0	t0, CP0_STATUS
+
+	/* Skip cache & coherence setup if we're already coherent */
+	cmgcrb	v1
+	lw	s7, GCR_CL_COHERENCE_OFS(v1)
+	bnez	s7, 1f
+	 nop
+
+	/* Initialize the L1 caches */
+	jal	mips_cps_cache_init
+	 nop
+
+	/* Enter the coherent domain */
+	li	t0, 0xff
+	sw	t0, GCR_CL_COHERENCE_OFS(v1)
+	ehb
+
+	/* Set Kseg0 CCA to that in s0 */
+1:	mfc0	t0, CP0_CONFIG
+	ori	t0, 0x7
+	xori	t0, 0x7
+	or	t0, t0, s0
+	mtc0	t0, CP0_CONFIG
+	ehb
+
+	/* Jump to kseg0 */
+	PTR_LA	t0, 1f
+	jr	t0
+	 nop
+
+	/*
+	 * We're up, cached & coherent. Perform any EVA initialization necessary
+	 * before we access memory.
+	 */
+1:	eva_init
+
+	/* Retrieve boot configuration pointers */
+	jal	mips_cps_get_bootcfg
+	 nop
+
+	/* Skip core-level init if we started up coherent */
+	bnez	s7, 1f
+	 nop
+
+	/* Perform any further required core-level initialisation */
+	jal	mips_cps_core_init
+	 nop
+
+	/*
+	 * Boot any other VPEs within this core that should be online, and
+	 * deactivate this VPE if it should be offline.
+	 */
+	move	a1, t9
+	jal	mips_cps_boot_vpes
+	 move	a0, v0
+
+	/* Off we go! */
+1:	PTR_L	t1, VPEBOOTCFG_PC(v1)
+	PTR_L	gp, VPEBOOTCFG_GP(v1)
+	PTR_L	sp, VPEBOOTCFG_SP(v1)
+	jr	t1
+	 nop
+	END(mips_cps_core_entry)
+
+.org 0x200
+LEAF(excep_tlbfill)
+	DUMP_EXCEP("TLB Fill")
+	b	.
+	 nop
+	END(excep_tlbfill)
+
+.org 0x280
+LEAF(excep_xtlbfill)
+	DUMP_EXCEP("XTLB Fill")
+	b	.
+	 nop
+	END(excep_xtlbfill)
+
+.org 0x300
+LEAF(excep_cache)
+	DUMP_EXCEP("Cache")
+	b	.
+	 nop
+	END(excep_cache)
+
+.org 0x380
+LEAF(excep_genex)
+	DUMP_EXCEP("General")
+	b	.
+	 nop
+	END(excep_genex)
+
+.org 0x400
+LEAF(excep_intex)
+	DUMP_EXCEP("Interrupt")
+	b	.
+	 nop
+	END(excep_intex)
+
+.org 0x480
+LEAF(excep_ejtag)
+	PTR_LA	k0, ejtag_debug_handler
+	jr	k0
+	 nop
+	END(excep_ejtag)
+
+LEAF(mips_cps_core_init)
+#ifdef CONFIG_MIPS_MT_SMP
+	/* Check that the core implements the MT ASE */
+	has_mt	t0, 3f
+
+	.set	push
+	.set	MIPS_ISA_LEVEL_RAW
+	.set	mt
+
+	/* Only allow 1 TC per VPE to execute... */
+	dmt
+
+	/* ...and for the moment only 1 VPE */
+	dvpe
+	PTR_LA	t1, 1f
+	jr.hb	t1
+	 nop
+
+	/* Enter VPE configuration state */
+1:	mfc0	t0, CP0_MVPCONTROL
+	ori	t0, t0, MVPCONTROL_VPC
+	mtc0	t0, CP0_MVPCONTROL
+
+	/* Retrieve the number of VPEs within the core */
+	mfc0	t0, CP0_MVPCONF0
+	srl	t0, t0, MVPCONF0_PVPE_SHIFT
+	andi	t0, t0, (MVPCONF0_PVPE >> MVPCONF0_PVPE_SHIFT)
+	addiu	ta3, t0, 1
+
+	/* If there's only 1, we're done */
+	beqz	t0, 2f
+	 nop
+
+	/* Loop through each VPE within this core */
+	li	ta1, 1
+
+1:	/* Operate on the appropriate TC */
+	mtc0	ta1, CP0_VPECONTROL
+	ehb
+
+	/* Bind TC to VPE (1:1 TC:VPE mapping) */
+	mttc0	ta1, CP0_TCBIND
+
+	/* Set exclusive TC, non-active, master */
+	li	t0, VPECONF0_MVP
+	sll	t1, ta1, VPECONF0_XTC_SHIFT
+	or	t0, t0, t1
+	mttc0	t0, CP0_VPECONF0
+
+	/* Set TC non-active, non-allocatable */
+	mttc0	zero, CP0_TCSTATUS
+
+	/* Set TC halted */
+	li	t0, TCHALT_H
+	mttc0	t0, CP0_TCHALT
+
+	/* Next VPE */
+	addiu	ta1, ta1, 1
+	slt	t0, ta1, ta3
+	bnez	t0, 1b
+	 nop
+
+	/* Leave VPE configuration state */
+2:	mfc0	t0, CP0_MVPCONTROL
+	xori	t0, t0, MVPCONTROL_VPC
+	mtc0	t0, CP0_MVPCONTROL
+
+3:	.set	pop
+#endif
+	jr	ra
+	 nop
+	END(mips_cps_core_init)
+
+/**
+ * mips_cps_get_bootcfg() - retrieve boot configuration pointers
+ *
+ * Returns: pointer to struct core_boot_config in v0, pointer to
+ *          struct vpe_boot_config in v1, VPE ID in t9
+ */
+LEAF(mips_cps_get_bootcfg)
+	/* Calculate a pointer to this cores struct core_boot_config */
+	cmgcrb	t0
+	lw	t0, GCR_CL_ID_OFS(t0)
+	li	t1, COREBOOTCFG_SIZE
+	mul	t0, t0, t1
+	PTR_LA	t1, mips_cps_core_bootcfg
+	PTR_L	t1, 0(t1)
+	PTR_ADDU v0, t0, t1
+
+	/* Calculate this VPEs ID. If the core doesn't support MT use 0 */
+	li	t9, 0
+#if defined(CONFIG_CPU_MIPSR6)
+	has_vp	ta2, 1f
+
+	/*
+	 * Assume non-contiguous numbering. Perhaps some day we'll need
+	 * to handle contiguous VP numbering, but no such systems yet
+	 * exist.
+	 */
+	mfc0	t9, CP0_GLOBALNUMBER
+	andi	t9, t9, MIPS_GLOBALNUMBER_VP
+#elif defined(CONFIG_MIPS_MT_SMP)
+	has_mt	ta2, 1f
+
+	/* Find the number of VPEs present in the core */
+	mfc0	t1, CP0_MVPCONF0
+	srl	t1, t1, MVPCONF0_PVPE_SHIFT
+	andi	t1, t1, MVPCONF0_PVPE >> MVPCONF0_PVPE_SHIFT
+	addiu	t1, t1, 1
+
+	/* Calculate a mask for the VPE ID from EBase.CPUNum */
+	clz	t1, t1
+	li	t2, 31
+	subu	t1, t2, t1
+	li	t2, 1
+	sll	t1, t2, t1
+	addiu	t1, t1, -1
+
+	/* Retrieve the VPE ID from EBase.CPUNum */
+	mfc0	t9, $15, 1
+	and	t9, t9, t1
+#endif
+
+1:	/* Calculate a pointer to this VPEs struct vpe_boot_config */
+	li	t1, VPEBOOTCFG_SIZE
+	mul	v1, t9, t1
+	PTR_L	ta3, COREBOOTCFG_VPECONFIG(v0)
+	PTR_ADDU v1, v1, ta3
+
+	jr	ra
+	 nop
+	END(mips_cps_get_bootcfg)
+
+LEAF(mips_cps_boot_vpes)
+	lw	ta2, COREBOOTCFG_VPEMASK(a0)
+	PTR_L	ta3, COREBOOTCFG_VPECONFIG(a0)
+
+#if defined(CONFIG_CPU_MIPSR6)
+
+	has_vp	t0, 5f
+
+	/* Find base address of CPC */
+	cmgcrb	t3
+	PTR_L	t1, GCR_CPC_BASE_OFS(t3)
+	PTR_LI	t2, ~0x7fff
+	and	t1, t1, t2
+	PTR_LI	t2, UNCAC_BASE
+	PTR_ADD	t1, t1, t2
+
+	/* Start any other VPs that ought to be running */
+	PTR_S	ta2, CPC_CL_VC_RUN_OFS(t1)
+
+	/* Ensure this VP stops running if it shouldn't be */
+	not	ta2
+	PTR_S	ta2, CPC_CL_VC_STOP_OFS(t1)
+	ehb
+
+#elif defined(CONFIG_MIPS_MT)
+
+	/* If the core doesn't support MT then return */
+	has_mt	t0, 5f
+
+	/* Enter VPE configuration state */
+	.set	push
+	.set	MIPS_ISA_LEVEL_RAW
+	.set	mt
+	dvpe
+	.set	pop
+
+	PTR_LA	t1, 1f
+	jr.hb	t1
+	 nop
+1:	mfc0	t1, CP0_MVPCONTROL
+	ori	t1, t1, MVPCONTROL_VPC
+	mtc0	t1, CP0_MVPCONTROL
+	ehb
+
+	/* Loop through each VPE */
+	move	t8, ta2
+	li	ta1, 0
+
+	/* Check whether the VPE should be running. If not, skip it */
+1:	andi	t0, ta2, 1
+	beqz	t0, 2f
+	 nop
+
+	/* Operate on the appropriate TC */
+	mfc0	t0, CP0_VPECONTROL
+	ori	t0, t0, VPECONTROL_TARGTC
+	xori	t0, t0, VPECONTROL_TARGTC
+	or	t0, t0, ta1
+	mtc0	t0, CP0_VPECONTROL
+	ehb
+
+	.set	push
+	.set	MIPS_ISA_LEVEL_RAW
+	.set	mt
+
+	/* Skip the VPE if its TC is not halted */
+	mftc0	t0, CP0_TCHALT
+	beqz	t0, 2f
+	 nop
+
+	/* Calculate a pointer to the VPEs struct vpe_boot_config */
+	li	t0, VPEBOOTCFG_SIZE
+	mul	t0, t0, ta1
+	addu	t0, t0, ta3
+
+	/* Set the TC restart PC */
+	lw	t1, VPEBOOTCFG_PC(t0)
+	mttc0	t1, CP0_TCRESTART
+
+	/* Set the TC stack pointer */
+	lw	t1, VPEBOOTCFG_SP(t0)
+	mttgpr	t1, sp
+
+	/* Set the TC global pointer */
+	lw	t1, VPEBOOTCFG_GP(t0)
+	mttgpr	t1, gp
+
+	/* Copy config from this VPE */
+	mfc0	t0, CP0_CONFIG
+	mttc0	t0, CP0_CONFIG
+
+	/*
+	 * Copy the EVA config from this VPE if the CPU supports it.
+	 * CONFIG3 must exist to be running MT startup - just read it.
+	 */
+	mfc0	t0, CP0_CONFIG, 3
+	and	t0, t0, MIPS_CONF3_SC
+	beqz	t0, 3f
+	 nop
+	mfc0    t0, CP0_SEGCTL0
+	mttc0	t0, CP0_SEGCTL0
+	mfc0    t0, CP0_SEGCTL1
+	mttc0	t0, CP0_SEGCTL1
+	mfc0    t0, CP0_SEGCTL2
+	mttc0	t0, CP0_SEGCTL2
+3:
+	/* Ensure no software interrupts are pending */
+	mttc0	zero, CP0_CAUSE
+	mttc0	zero, CP0_STATUS
+
+	/* Set TC active, not interrupt exempt */
+	mftc0	t0, CP0_TCSTATUS
+	li	t1, ~TCSTATUS_IXMT
+	and	t0, t0, t1
+	ori	t0, t0, TCSTATUS_A
+	mttc0	t0, CP0_TCSTATUS
+
+	/* Clear the TC halt bit */
+	mttc0	zero, CP0_TCHALT
+
+	/* Set VPE active */
+	mftc0	t0, CP0_VPECONF0
+	ori	t0, t0, VPECONF0_VPA
+	mttc0	t0, CP0_VPECONF0
+
+	/* Next VPE */
+2:	srl	ta2, ta2, 1
+	addiu	ta1, ta1, 1
+	bnez	ta2, 1b
+	 nop
+
+	/* Leave VPE configuration state */
+	mfc0	t1, CP0_MVPCONTROL
+	xori	t1, t1, MVPCONTROL_VPC
+	mtc0	t1, CP0_MVPCONTROL
+	ehb
+	evpe
+
+	.set	pop
+
+	/* Check whether this VPE is meant to be running */
+	li	t0, 1
+	sll	t0, t0, a1
+	and	t0, t0, t8
+	bnez	t0, 2f
+	 nop
+
+	/* This VPE should be offline, halt the TC */
+	li	t0, TCHALT_H
+	mtc0	t0, CP0_TCHALT
+	PTR_LA	t0, 1f
+1:	jr.hb	t0
+	 nop
+
+2:
+
+#endif /* CONFIG_MIPS_MT_SMP */
+
+	/* Return */
+5:	jr	ra
+	 nop
+	END(mips_cps_boot_vpes)
+
+LEAF(mips_cps_cache_init)
+	/*
+	 * Clear the bits used to index the caches. Note that the architecture
+	 * dictates that writing to any of TagLo or TagHi selects 0 or 2 should
+	 * be valid for all MIPS32 CPUs, even those for which said writes are
+	 * unnecessary.
+	 */
+	mtc0	zero, CP0_TAGLO, 0
+	mtc0	zero, CP0_TAGHI, 0
+	mtc0	zero, CP0_TAGLO, 2
+	mtc0	zero, CP0_TAGHI, 2
+	ehb
+
+	/* Primary cache configuration is indicated by Config1 */
+	mfc0	v0, CP0_CONFIG, 1
+
+	/* Detect I-cache line size */
+	_EXT	t0, v0, MIPS_CONF1_IL_SHF, MIPS_CONF1_IL_SZ
+	beqz	t0, icache_done
+	 li	t1, 2
+	sllv	t0, t1, t0
+
+	/* Detect I-cache size */
+	_EXT	t1, v0, MIPS_CONF1_IS_SHF, MIPS_CONF1_IS_SZ
+	xori	t2, t1, 0x7
+	beqz	t2, 1f
+	 li	t3, 32
+	addiu	t1, t1, 1
+	sllv	t1, t3, t1
+1:	/* At this point t1 == I-cache sets per way */
+	_EXT	t2, v0, MIPS_CONF1_IA_SHF, MIPS_CONF1_IA_SZ
+	addiu	t2, t2, 1
+	mul	t1, t1, t0
+	mul	t1, t1, t2
+
+	li	a0, CKSEG0
+	PTR_ADD	a1, a0, t1
+1:	cache	Index_Store_Tag_I, 0(a0)
+	PTR_ADD	a0, a0, t0
+	bne	a0, a1, 1b
+	 nop
+icache_done:
+
+	/* Detect D-cache line size */
+	_EXT	t0, v0, MIPS_CONF1_DL_SHF, MIPS_CONF1_DL_SZ
+	beqz	t0, dcache_done
+	 li	t1, 2
+	sllv	t0, t1, t0
+
+	/* Detect D-cache size */
+	_EXT	t1, v0, MIPS_CONF1_DS_SHF, MIPS_CONF1_DS_SZ
+	xori	t2, t1, 0x7
+	beqz	t2, 1f
+	 li	t3, 32
+	addiu	t1, t1, 1
+	sllv	t1, t3, t1
+1:	/* At this point t1 == D-cache sets per way */
+	_EXT	t2, v0, MIPS_CONF1_DA_SHF, MIPS_CONF1_DA_SZ
+	addiu	t2, t2, 1
+	mul	t1, t1, t0
+	mul	t1, t1, t2
+
+	li	a0, CKSEG0
+	PTR_ADDU a1, a0, t1
+	PTR_SUBU a1, a1, t0
+1:	cache	Index_Store_Tag_D, 0(a0)
+	bne	a0, a1, 1b
+	 PTR_ADD a0, a0, t0
+dcache_done:
+
+	jr	ra
+	 nop
+	END(mips_cps_cache_init)
+
+#if defined(CONFIG_MIPS_CPS_PM) && defined(CONFIG_CPU_PM)
+
+	/* Calculate a pointer to this CPUs struct mips_static_suspend_state */
+	.macro	psstate	dest
+	.set	push
+	.set	noat
+	lw	$1, TI_CPU(gp)
+	sll	$1, $1, LONGLOG
+	PTR_LA	\dest, __per_cpu_offset
+	addu	$1, $1, \dest
+	lw	$1, 0($1)
+	PTR_LA	\dest, cps_cpu_state
+	addu	\dest, \dest, $1
+	.set	pop
+	.endm
+
+LEAF(mips_cps_pm_save)
+	/* Save CPU state */
+	SUSPEND_SAVE_REGS
+	psstate	t1
+	SUSPEND_SAVE_STATIC
+	jr	v0
+	 nop
+	END(mips_cps_pm_save)
+
+LEAF(mips_cps_pm_restore)
+	/* Restore CPU state */
+	psstate	t1
+	RESUME_RESTORE_STATIC
+	RESUME_RESTORE_REGS_RETURN
+	END(mips_cps_pm_restore)
+
+#endif /* CONFIG_MIPS_CPS_PM && CONFIG_CPU_PM */
diff --git a/arch/mips/kernel/cpu-probe.c b/arch/mips/kernel/cpu-probe.c
new file mode 100644
index 000000000..f8d1933bf
--- /dev/null
+++ b/arch/mips/kernel/cpu-probe.c
@@ -0,0 +1,2170 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Processor capabilities determination functions.
+ *
+ * Copyright (C) xxxx  the Anonymous
+ * Copyright (C) 1994 - 2006 Ralf Baechle
+ * Copyright (C) 2003, 2004  Maciej W. Rozycki
+ * Copyright (C) 2001, 2004, 2011, 2012	 MIPS Technologies, Inc.
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/ptrace.h>
+#include <linux/smp.h>
+#include <linux/stddef.h>
+#include <linux/export.h>
+
+#include <asm/bugs.h>
+#include <asm/cpu.h>
+#include <asm/cpu-features.h>
+#include <asm/cpu-type.h>
+#include <asm/fpu.h>
+#include <asm/mipsregs.h>
+#include <asm/mipsmtregs.h>
+#include <asm/msa.h>
+#include <asm/watch.h>
+#include <asm/elf.h>
+#include <asm/pgtable-bits.h>
+#include <asm/spram.h>
+#include <linux/uaccess.h>
+
+#include "fpu-probe.h"
+
+#include <asm/mach-loongson64/cpucfg-emul.h>
+
+/* Hardware capabilities */
+unsigned int elf_hwcap __read_mostly;
+EXPORT_SYMBOL_GPL(elf_hwcap);
+
+static inline unsigned long cpu_get_msa_id(void)
+{
+	unsigned long status, msa_id;
+
+	status = read_c0_status();
+	__enable_fpu(FPU_64BIT);
+	enable_msa();
+	msa_id = read_msa_ir();
+	disable_msa();
+	write_c0_status(status);
+	return msa_id;
+}
+
+static int mips_dsp_disabled;
+
+static int __init dsp_disable(char *s)
+{
+	cpu_data[0].ases &= ~(MIPS_ASE_DSP | MIPS_ASE_DSP2P);
+	mips_dsp_disabled = 1;
+
+	return 1;
+}
+
+__setup("nodsp", dsp_disable);
+
+static int mips_htw_disabled;
+
+static int __init htw_disable(char *s)
+{
+	mips_htw_disabled = 1;
+	cpu_data[0].options &= ~MIPS_CPU_HTW;
+	write_c0_pwctl(read_c0_pwctl() &
+		       ~(1 << MIPS_PWCTL_PWEN_SHIFT));
+
+	return 1;
+}
+
+__setup("nohtw", htw_disable);
+
+static int mips_ftlb_disabled;
+static int mips_has_ftlb_configured;
+
+enum ftlb_flags {
+	FTLB_EN		= 1 << 0,
+	FTLB_SET_PROB	= 1 << 1,
+};
+
+static int set_ftlb_enable(struct cpuinfo_mips *c, enum ftlb_flags flags);
+
+static int __init ftlb_disable(char *s)
+{
+	unsigned int config4, mmuextdef;
+
+	/*
+	 * If the core hasn't done any FTLB configuration, there is nothing
+	 * for us to do here.
+	 */
+	if (!mips_has_ftlb_configured)
+		return 1;
+
+	/* Disable it in the boot cpu */
+	if (set_ftlb_enable(&cpu_data[0], 0)) {
+		pr_warn("Can't turn FTLB off\n");
+		return 1;
+	}
+
+	config4 = read_c0_config4();
+
+	/* Check that FTLB has been disabled */
+	mmuextdef = config4 & MIPS_CONF4_MMUEXTDEF;
+	/* MMUSIZEEXT == VTLB ON, FTLB OFF */
+	if (mmuextdef == MIPS_CONF4_MMUEXTDEF_FTLBSIZEEXT) {
+		/* This should never happen */
+		pr_warn("FTLB could not be disabled!\n");
+		return 1;
+	}
+
+	mips_ftlb_disabled = 1;
+	mips_has_ftlb_configured = 0;
+
+	/*
+	 * noftlb is mainly used for debug purposes so print
+	 * an informative message instead of using pr_debug()
+	 */
+	pr_info("FTLB has been disabled\n");
+
+	/*
+	 * Some of these bits are duplicated in the decode_config4.
+	 * MIPS_CONF4_MMUEXTDEF_MMUSIZEEXT is the only possible case
+	 * once FTLB has been disabled so undo what decode_config4 did.
+	 */
+	cpu_data[0].tlbsize -= cpu_data[0].tlbsizeftlbways *
+			       cpu_data[0].tlbsizeftlbsets;
+	cpu_data[0].tlbsizeftlbsets = 0;
+	cpu_data[0].tlbsizeftlbways = 0;
+
+	return 1;
+}
+
+__setup("noftlb", ftlb_disable);
+
+/*
+ * Check if the CPU has per tc perf counters
+ */
+static inline void cpu_set_mt_per_tc_perf(struct cpuinfo_mips *c)
+{
+	if (read_c0_config7() & MTI_CONF7_PTC)
+		c->options |= MIPS_CPU_MT_PER_TC_PERF_COUNTERS;
+}
+
+static inline void check_errata(void)
+{
+	struct cpuinfo_mips *c = &current_cpu_data;
+
+	switch (current_cpu_type()) {
+	case CPU_34K:
+		/*
+		 * Erratum "RPS May Cause Incorrect Instruction Execution"
+		 * This code only handles VPE0, any SMP/RTOS code
+		 * making use of VPE1 will be responsable for that VPE.
+		 */
+		if ((c->processor_id & PRID_REV_MASK) <= PRID_REV_34K_V1_0_2)
+			write_c0_config7(read_c0_config7() | MIPS_CONF7_RPS);
+		break;
+	default:
+		break;
+	}
+}
+
+void __init check_bugs32(void)
+{
+	check_errata();
+}
+
+/*
+ * Probe whether cpu has config register by trying to play with
+ * alternate cache bit and see whether it matters.
+ * It's used by cpu_probe to distinguish between R3000A and R3081.
+ */
+static inline int cpu_has_confreg(void)
+{
+#ifdef CONFIG_CPU_R3000
+	extern unsigned long r3k_cache_size(unsigned long);
+	unsigned long size1, size2;
+	unsigned long cfg = read_c0_conf();
+
+	size1 = r3k_cache_size(ST0_ISC);
+	write_c0_conf(cfg ^ R30XX_CONF_AC);
+	size2 = r3k_cache_size(ST0_ISC);
+	write_c0_conf(cfg);
+	return size1 != size2;
+#else
+	return 0;
+#endif
+}
+
+static inline void set_elf_platform(int cpu, const char *plat)
+{
+	if (cpu == 0)
+		__elf_platform = plat;
+}
+
+static inline void set_elf_base_platform(const char *plat)
+{
+	if (__elf_base_platform == NULL) {
+		__elf_base_platform = plat;
+	}
+}
+
+static inline void cpu_probe_vmbits(struct cpuinfo_mips *c)
+{
+#ifdef __NEED_VMBITS_PROBE
+	write_c0_entryhi(0x3fffffffffffe000ULL);
+	back_to_back_c0_hazard();
+	c->vmbits = fls64(read_c0_entryhi() & 0x3fffffffffffe000ULL);
+#endif
+}
+
+static void set_isa(struct cpuinfo_mips *c, unsigned int isa)
+{
+	switch (isa) {
+	case MIPS_CPU_ISA_M64R5:
+		c->isa_level |= MIPS_CPU_ISA_M32R5 | MIPS_CPU_ISA_M64R5;
+		set_elf_base_platform("mips64r5");
+		fallthrough;
+	case MIPS_CPU_ISA_M64R2:
+		c->isa_level |= MIPS_CPU_ISA_M32R2 | MIPS_CPU_ISA_M64R2;
+		set_elf_base_platform("mips64r2");
+		fallthrough;
+	case MIPS_CPU_ISA_M64R1:
+		c->isa_level |= MIPS_CPU_ISA_M32R1 | MIPS_CPU_ISA_M64R1;
+		set_elf_base_platform("mips64");
+		fallthrough;
+	case MIPS_CPU_ISA_V:
+		c->isa_level |= MIPS_CPU_ISA_V;
+		set_elf_base_platform("mips5");
+		fallthrough;
+	case MIPS_CPU_ISA_IV:
+		c->isa_level |= MIPS_CPU_ISA_IV;
+		set_elf_base_platform("mips4");
+		fallthrough;
+	case MIPS_CPU_ISA_III:
+		c->isa_level |= MIPS_CPU_ISA_II | MIPS_CPU_ISA_III;
+		set_elf_base_platform("mips3");
+		break;
+
+	/* R6 incompatible with everything else */
+	case MIPS_CPU_ISA_M64R6:
+		c->isa_level |= MIPS_CPU_ISA_M32R6 | MIPS_CPU_ISA_M64R6;
+		set_elf_base_platform("mips64r6");
+		fallthrough;
+	case MIPS_CPU_ISA_M32R6:
+		c->isa_level |= MIPS_CPU_ISA_M32R6;
+		set_elf_base_platform("mips32r6");
+		/* Break here so we don't add incompatible ISAs */
+		break;
+	case MIPS_CPU_ISA_M32R5:
+		c->isa_level |= MIPS_CPU_ISA_M32R5;
+		set_elf_base_platform("mips32r5");
+		fallthrough;
+	case MIPS_CPU_ISA_M32R2:
+		c->isa_level |= MIPS_CPU_ISA_M32R2;
+		set_elf_base_platform("mips32r2");
+		fallthrough;
+	case MIPS_CPU_ISA_M32R1:
+		c->isa_level |= MIPS_CPU_ISA_M32R1;
+		set_elf_base_platform("mips32");
+		fallthrough;
+	case MIPS_CPU_ISA_II:
+		c->isa_level |= MIPS_CPU_ISA_II;
+		set_elf_base_platform("mips2");
+		break;
+	}
+}
+
+static char unknown_isa[] = KERN_ERR \
+	"Unsupported ISA type, c0.config0: %d.";
+
+static unsigned int calculate_ftlb_probability(struct cpuinfo_mips *c)
+{
+
+	unsigned int probability = c->tlbsize / c->tlbsizevtlb;
+
+	/*
+	 * 0 = All TLBWR instructions go to FTLB
+	 * 1 = 15:1: For every 16 TBLWR instructions, 15 go to the
+	 * FTLB and 1 goes to the VTLB.
+	 * 2 = 7:1: As above with 7:1 ratio.
+	 * 3 = 3:1: As above with 3:1 ratio.
+	 *
+	 * Use the linear midpoint as the probability threshold.
+	 */
+	if (probability >= 12)
+		return 1;
+	else if (probability >= 6)
+		return 2;
+	else
+		/*
+		 * So FTLB is less than 4 times bigger than VTLB.
+		 * A 3:1 ratio can still be useful though.
+		 */
+		return 3;
+}
+
+static int set_ftlb_enable(struct cpuinfo_mips *c, enum ftlb_flags flags)
+{
+	unsigned int config;
+
+	/* It's implementation dependent how the FTLB can be enabled */
+	switch (c->cputype) {
+	case CPU_PROAPTIV:
+	case CPU_P5600:
+	case CPU_P6600:
+		/* proAptiv & related cores use Config6 to enable the FTLB */
+		config = read_c0_config6();
+
+		if (flags & FTLB_EN)
+			config |= MTI_CONF6_FTLBEN;
+		else
+			config &= ~MTI_CONF6_FTLBEN;
+
+		if (flags & FTLB_SET_PROB) {
+			config &= ~(3 << MTI_CONF6_FTLBP_SHIFT);
+			config |= calculate_ftlb_probability(c)
+				  << MTI_CONF6_FTLBP_SHIFT;
+		}
+
+		write_c0_config6(config);
+		back_to_back_c0_hazard();
+		break;
+	case CPU_I6400:
+	case CPU_I6500:
+		/* There's no way to disable the FTLB */
+		if (!(flags & FTLB_EN))
+			return 1;
+		return 0;
+	case CPU_LOONGSON64:
+		/* Flush ITLB, DTLB, VTLB and FTLB */
+		write_c0_diag(LOONGSON_DIAG_ITLB | LOONGSON_DIAG_DTLB |
+			      LOONGSON_DIAG_VTLB | LOONGSON_DIAG_FTLB);
+		/* Loongson-3 cores use Config6 to enable the FTLB */
+		config = read_c0_config6();
+		if (flags & FTLB_EN)
+			/* Enable FTLB */
+			write_c0_config6(config & ~LOONGSON_CONF6_FTLBDIS);
+		else
+			/* Disable FTLB */
+			write_c0_config6(config | LOONGSON_CONF6_FTLBDIS);
+		break;
+	default:
+		return 1;
+	}
+
+	return 0;
+}
+
+static int mm_config(struct cpuinfo_mips *c)
+{
+	unsigned int config0, update, mm;
+
+	config0 = read_c0_config();
+	mm = config0 & MIPS_CONF_MM;
+
+	/*
+	 * It's implementation dependent what type of write-merge is supported
+	 * and whether it can be enabled/disabled. If it is settable lets make
+	 * the merging allowed by default. Some platforms might have
+	 * write-through caching unsupported. In this case just ignore the
+	 * CP0.Config.MM bit field value.
+	 */
+	switch (c->cputype) {
+	case CPU_24K:
+	case CPU_34K:
+	case CPU_74K:
+	case CPU_P5600:
+	case CPU_P6600:
+		c->options |= MIPS_CPU_MM_FULL;
+		update = MIPS_CONF_MM_FULL;
+		break;
+	case CPU_1004K:
+	case CPU_1074K:
+	case CPU_INTERAPTIV:
+	case CPU_PROAPTIV:
+		mm = 0;
+		fallthrough;
+	default:
+		update = 0;
+		break;
+	}
+
+	if (update) {
+		config0 = (config0 & ~MIPS_CONF_MM) | update;
+		write_c0_config(config0);
+	} else if (mm == MIPS_CONF_MM_SYSAD) {
+		c->options |= MIPS_CPU_MM_SYSAD;
+	} else if (mm == MIPS_CONF_MM_FULL) {
+		c->options |= MIPS_CPU_MM_FULL;
+	}
+
+	return 0;
+}
+
+static inline unsigned int decode_config0(struct cpuinfo_mips *c)
+{
+	unsigned int config0;
+	int isa, mt;
+
+	config0 = read_c0_config();
+
+	/*
+	 * Look for Standard TLB or Dual VTLB and FTLB
+	 */
+	mt = config0 & MIPS_CONF_MT;
+	if (mt == MIPS_CONF_MT_TLB)
+		c->options |= MIPS_CPU_TLB;
+	else if (mt == MIPS_CONF_MT_FTLB)
+		c->options |= MIPS_CPU_TLB | MIPS_CPU_FTLB;
+
+	isa = (config0 & MIPS_CONF_AT) >> 13;
+	switch (isa) {
+	case 0:
+		switch ((config0 & MIPS_CONF_AR) >> 10) {
+		case 0:
+			set_isa(c, MIPS_CPU_ISA_M32R1);
+			break;
+		case 1:
+			set_isa(c, MIPS_CPU_ISA_M32R2);
+			break;
+		case 2:
+			set_isa(c, MIPS_CPU_ISA_M32R6);
+			break;
+		default:
+			goto unknown;
+		}
+		break;
+	case 2:
+		switch ((config0 & MIPS_CONF_AR) >> 10) {
+		case 0:
+			set_isa(c, MIPS_CPU_ISA_M64R1);
+			break;
+		case 1:
+			set_isa(c, MIPS_CPU_ISA_M64R2);
+			break;
+		case 2:
+			set_isa(c, MIPS_CPU_ISA_M64R6);
+			break;
+		default:
+			goto unknown;
+		}
+		break;
+	default:
+		goto unknown;
+	}
+
+	return config0 & MIPS_CONF_M;
+
+unknown:
+	panic(unknown_isa, config0);
+}
+
+static inline unsigned int decode_config1(struct cpuinfo_mips *c)
+{
+	unsigned int config1;
+
+	config1 = read_c0_config1();
+
+	if (config1 & MIPS_CONF1_MD)
+		c->ases |= MIPS_ASE_MDMX;
+	if (config1 & MIPS_CONF1_PC)
+		c->options |= MIPS_CPU_PERF;
+	if (config1 & MIPS_CONF1_WR)
+		c->options |= MIPS_CPU_WATCH;
+	if (config1 & MIPS_CONF1_CA)
+		c->ases |= MIPS_ASE_MIPS16;
+	if (config1 & MIPS_CONF1_EP)
+		c->options |= MIPS_CPU_EJTAG;
+	if (config1 & MIPS_CONF1_FP) {
+		c->options |= MIPS_CPU_FPU;
+		c->options |= MIPS_CPU_32FPR;
+	}
+	if (cpu_has_tlb) {
+		c->tlbsize = ((config1 & MIPS_CONF1_TLBS) >> 25) + 1;
+		c->tlbsizevtlb = c->tlbsize;
+		c->tlbsizeftlbsets = 0;
+	}
+
+	return config1 & MIPS_CONF_M;
+}
+
+static inline unsigned int decode_config2(struct cpuinfo_mips *c)
+{
+	unsigned int config2;
+
+	config2 = read_c0_config2();
+
+	if (config2 & MIPS_CONF2_SL)
+		c->scache.flags &= ~MIPS_CACHE_NOT_PRESENT;
+
+	return config2 & MIPS_CONF_M;
+}
+
+static inline unsigned int decode_config3(struct cpuinfo_mips *c)
+{
+	unsigned int config3;
+
+	config3 = read_c0_config3();
+
+	if (config3 & MIPS_CONF3_SM) {
+		c->ases |= MIPS_ASE_SMARTMIPS;
+		c->options |= MIPS_CPU_RIXI | MIPS_CPU_CTXTC;
+	}
+	if (config3 & MIPS_CONF3_RXI)
+		c->options |= MIPS_CPU_RIXI;
+	if (config3 & MIPS_CONF3_CTXTC)
+		c->options |= MIPS_CPU_CTXTC;
+	if (config3 & MIPS_CONF3_DSP)
+		c->ases |= MIPS_ASE_DSP;
+	if (config3 & MIPS_CONF3_DSP2P) {
+		c->ases |= MIPS_ASE_DSP2P;
+		if (cpu_has_mips_r6)
+			c->ases |= MIPS_ASE_DSP3;
+	}
+	if (config3 & MIPS_CONF3_VINT)
+		c->options |= MIPS_CPU_VINT;
+	if (config3 & MIPS_CONF3_VEIC)
+		c->options |= MIPS_CPU_VEIC;
+	if (config3 & MIPS_CONF3_LPA)
+		c->options |= MIPS_CPU_LPA;
+	if (config3 & MIPS_CONF3_MT)
+		c->ases |= MIPS_ASE_MIPSMT;
+	if (config3 & MIPS_CONF3_ULRI)
+		c->options |= MIPS_CPU_ULRI;
+	if (config3 & MIPS_CONF3_ISA)
+		c->options |= MIPS_CPU_MICROMIPS;
+	if (config3 & MIPS_CONF3_VZ)
+		c->ases |= MIPS_ASE_VZ;
+	if (config3 & MIPS_CONF3_SC)
+		c->options |= MIPS_CPU_SEGMENTS;
+	if (config3 & MIPS_CONF3_BI)
+		c->options |= MIPS_CPU_BADINSTR;
+	if (config3 & MIPS_CONF3_BP)
+		c->options |= MIPS_CPU_BADINSTRP;
+	if (config3 & MIPS_CONF3_MSA)
+		c->ases |= MIPS_ASE_MSA;
+	if (config3 & MIPS_CONF3_PW) {
+		c->htw_seq = 0;
+		c->options |= MIPS_CPU_HTW;
+	}
+	if (config3 & MIPS_CONF3_CDMM)
+		c->options |= MIPS_CPU_CDMM;
+	if (config3 & MIPS_CONF3_SP)
+		c->options |= MIPS_CPU_SP;
+
+	return config3 & MIPS_CONF_M;
+}
+
+static inline unsigned int decode_config4(struct cpuinfo_mips *c)
+{
+	unsigned int config4;
+	unsigned int newcf4;
+	unsigned int mmuextdef;
+	unsigned int ftlb_page = MIPS_CONF4_FTLBPAGESIZE;
+	unsigned long asid_mask;
+
+	config4 = read_c0_config4();
+
+	if (cpu_has_tlb) {
+		if (((config4 & MIPS_CONF4_IE) >> 29) == 2)
+			c->options |= MIPS_CPU_TLBINV;
+
+		/*
+		 * R6 has dropped the MMUExtDef field from config4.
+		 * On R6 the fields always describe the FTLB, and only if it is
+		 * present according to Config.MT.
+		 */
+		if (!cpu_has_mips_r6)
+			mmuextdef = config4 & MIPS_CONF4_MMUEXTDEF;
+		else if (cpu_has_ftlb)
+			mmuextdef = MIPS_CONF4_MMUEXTDEF_VTLBSIZEEXT;
+		else
+			mmuextdef = 0;
+
+		switch (mmuextdef) {
+		case MIPS_CONF4_MMUEXTDEF_MMUSIZEEXT:
+			c->tlbsize += (config4 & MIPS_CONF4_MMUSIZEEXT) * 0x40;
+			c->tlbsizevtlb = c->tlbsize;
+			break;
+		case MIPS_CONF4_MMUEXTDEF_VTLBSIZEEXT:
+			c->tlbsizevtlb +=
+				((config4 & MIPS_CONF4_VTLBSIZEEXT) >>
+				  MIPS_CONF4_VTLBSIZEEXT_SHIFT) * 0x40;
+			c->tlbsize = c->tlbsizevtlb;
+			ftlb_page = MIPS_CONF4_VFTLBPAGESIZE;
+			fallthrough;
+		case MIPS_CONF4_MMUEXTDEF_FTLBSIZEEXT:
+			if (mips_ftlb_disabled)
+				break;
+			newcf4 = (config4 & ~ftlb_page) |
+				(page_size_ftlb(mmuextdef) <<
+				 MIPS_CONF4_FTLBPAGESIZE_SHIFT);
+			write_c0_config4(newcf4);
+			back_to_back_c0_hazard();
+			config4 = read_c0_config4();
+			if (config4 != newcf4) {
+				pr_err("PAGE_SIZE 0x%lx is not supported by FTLB (config4=0x%x)\n",
+				       PAGE_SIZE, config4);
+				/* Switch FTLB off */
+				set_ftlb_enable(c, 0);
+				mips_ftlb_disabled = 1;
+				break;
+			}
+			c->tlbsizeftlbsets = 1 <<
+				((config4 & MIPS_CONF4_FTLBSETS) >>
+				 MIPS_CONF4_FTLBSETS_SHIFT);
+			c->tlbsizeftlbways = ((config4 & MIPS_CONF4_FTLBWAYS) >>
+					      MIPS_CONF4_FTLBWAYS_SHIFT) + 2;
+			c->tlbsize += c->tlbsizeftlbways * c->tlbsizeftlbsets;
+			mips_has_ftlb_configured = 1;
+			break;
+		}
+	}
+
+	c->kscratch_mask = (config4 & MIPS_CONF4_KSCREXIST)
+				>> MIPS_CONF4_KSCREXIST_SHIFT;
+
+	asid_mask = MIPS_ENTRYHI_ASID;
+	if (config4 & MIPS_CONF4_AE)
+		asid_mask |= MIPS_ENTRYHI_ASIDX;
+	set_cpu_asid_mask(c, asid_mask);
+
+	/*
+	 * Warn if the computed ASID mask doesn't match the mask the kernel
+	 * is built for. This may indicate either a serious problem or an
+	 * easy optimisation opportunity, but either way should be addressed.
+	 */
+	WARN_ON(asid_mask != cpu_asid_mask(c));
+
+	return config4 & MIPS_CONF_M;
+}
+
+static inline unsigned int decode_config5(struct cpuinfo_mips *c)
+{
+	unsigned int config5, max_mmid_width;
+	unsigned long asid_mask;
+
+	config5 = read_c0_config5();
+	config5 &= ~(MIPS_CONF5_UFR | MIPS_CONF5_UFE);
+
+	if (cpu_has_mips_r6) {
+		if (!__builtin_constant_p(cpu_has_mmid) || cpu_has_mmid)
+			config5 |= MIPS_CONF5_MI;
+		else
+			config5 &= ~MIPS_CONF5_MI;
+	}
+
+	write_c0_config5(config5);
+
+	if (config5 & MIPS_CONF5_EVA)
+		c->options |= MIPS_CPU_EVA;
+	if (config5 & MIPS_CONF5_MRP)
+		c->options |= MIPS_CPU_MAAR;
+	if (config5 & MIPS_CONF5_LLB)
+		c->options |= MIPS_CPU_RW_LLB;
+	if (config5 & MIPS_CONF5_MVH)
+		c->options |= MIPS_CPU_MVH;
+	if (cpu_has_mips_r6 && (config5 & MIPS_CONF5_VP))
+		c->options |= MIPS_CPU_VP;
+	if (config5 & MIPS_CONF5_CA2)
+		c->ases |= MIPS_ASE_MIPS16E2;
+
+	if (config5 & MIPS_CONF5_CRCP)
+		elf_hwcap |= HWCAP_MIPS_CRC32;
+
+	if (cpu_has_mips_r6) {
+		/* Ensure the write to config5 above takes effect */
+		back_to_back_c0_hazard();
+
+		/* Check whether we successfully enabled MMID support */
+		config5 = read_c0_config5();
+		if (config5 & MIPS_CONF5_MI)
+			c->options |= MIPS_CPU_MMID;
+
+		/*
+		 * Warn if we've hardcoded cpu_has_mmid to a value unsuitable
+		 * for the CPU we're running on, or if CPUs in an SMP system
+		 * have inconsistent MMID support.
+		 */
+		WARN_ON(!!cpu_has_mmid != !!(config5 & MIPS_CONF5_MI));
+
+		if (cpu_has_mmid) {
+			write_c0_memorymapid(~0ul);
+			back_to_back_c0_hazard();
+			asid_mask = read_c0_memorymapid();
+
+			/*
+			 * We maintain a bitmap to track MMID allocation, and
+			 * need a sensible upper bound on the size of that
+			 * bitmap. The initial CPU with MMID support (I6500)
+			 * supports 16 bit MMIDs, which gives us an 8KiB
+			 * bitmap. The architecture recommends that hardware
+			 * support 32 bit MMIDs, which would give us a 512MiB
+			 * bitmap - that's too big in most cases.
+			 *
+			 * Cap MMID width at 16 bits for now & we can revisit
+			 * this if & when hardware supports anything wider.
+			 */
+			max_mmid_width = 16;
+			if (asid_mask > GENMASK(max_mmid_width - 1, 0)) {
+				pr_info("Capping MMID width at %d bits",
+					max_mmid_width);
+				asid_mask = GENMASK(max_mmid_width - 1, 0);
+			}
+
+			set_cpu_asid_mask(c, asid_mask);
+		}
+	}
+
+	return config5 & MIPS_CONF_M;
+}
+
+static void decode_configs(struct cpuinfo_mips *c)
+{
+	int ok;
+
+	/* MIPS32 or MIPS64 compliant CPU.  */
+	c->options = MIPS_CPU_4KEX | MIPS_CPU_4K_CACHE | MIPS_CPU_COUNTER |
+		     MIPS_CPU_DIVEC | MIPS_CPU_LLSC | MIPS_CPU_MCHECK;
+
+	c->scache.flags = MIPS_CACHE_NOT_PRESENT;
+
+	/* Enable FTLB if present and not disabled */
+	set_ftlb_enable(c, mips_ftlb_disabled ? 0 : FTLB_EN);
+
+	ok = decode_config0(c);			/* Read Config registers.  */
+	BUG_ON(!ok);				/* Arch spec violation!	 */
+	if (ok)
+		ok = decode_config1(c);
+	if (ok)
+		ok = decode_config2(c);
+	if (ok)
+		ok = decode_config3(c);
+	if (ok)
+		ok = decode_config4(c);
+	if (ok)
+		ok = decode_config5(c);
+
+	/* Probe the EBase.WG bit */
+	if (cpu_has_mips_r2_r6) {
+		u64 ebase;
+		unsigned int status;
+
+		/* {read,write}_c0_ebase_64() may be UNDEFINED prior to r6 */
+		ebase = cpu_has_mips64r6 ? read_c0_ebase_64()
+					 : (s32)read_c0_ebase();
+		if (ebase & MIPS_EBASE_WG) {
+			/* WG bit already set, we can avoid the clumsy probe */
+			c->options |= MIPS_CPU_EBASE_WG;
+		} else {
+			/* Its UNDEFINED to change EBase while BEV=0 */
+			status = read_c0_status();
+			write_c0_status(status | ST0_BEV);
+			irq_enable_hazard();
+			/*
+			 * On pre-r6 cores, this may well clobber the upper bits
+			 * of EBase. This is hard to avoid without potentially
+			 * hitting UNDEFINED dm*c0 behaviour if EBase is 32-bit.
+			 */
+			if (cpu_has_mips64r6)
+				write_c0_ebase_64(ebase | MIPS_EBASE_WG);
+			else
+				write_c0_ebase(ebase | MIPS_EBASE_WG);
+			back_to_back_c0_hazard();
+			/* Restore BEV */
+			write_c0_status(status);
+			if (read_c0_ebase() & MIPS_EBASE_WG) {
+				c->options |= MIPS_CPU_EBASE_WG;
+				write_c0_ebase(ebase);
+			}
+		}
+	}
+
+	/* configure the FTLB write probability */
+	set_ftlb_enable(c, (mips_ftlb_disabled ? 0 : FTLB_EN) | FTLB_SET_PROB);
+
+	mips_probe_watch_registers(c);
+
+#ifndef CONFIG_MIPS_CPS
+	if (cpu_has_mips_r2_r6) {
+		unsigned int core;
+
+		core = get_ebase_cpunum();
+		if (cpu_has_mipsmt)
+			core >>= fls(core_nvpes()) - 1;
+		cpu_set_core(c, core);
+	}
+#endif
+}
+
+/*
+ * Probe for certain guest capabilities by writing config bits and reading back.
+ * Finally write back the original value.
+ */
+#define probe_gc0_config(name, maxconf, bits)				\
+do {									\
+	unsigned int tmp;						\
+	tmp = read_gc0_##name();					\
+	write_gc0_##name(tmp | (bits));					\
+	back_to_back_c0_hazard();					\
+	maxconf = read_gc0_##name();					\
+	write_gc0_##name(tmp);						\
+} while (0)
+
+/*
+ * Probe for dynamic guest capabilities by changing certain config bits and
+ * reading back to see if they change. Finally write back the original value.
+ */
+#define probe_gc0_config_dyn(name, maxconf, dynconf, bits)		\
+do {									\
+	maxconf = read_gc0_##name();					\
+	write_gc0_##name(maxconf ^ (bits));				\
+	back_to_back_c0_hazard();					\
+	dynconf = maxconf ^ read_gc0_##name();				\
+	write_gc0_##name(maxconf);					\
+	maxconf |= dynconf;						\
+} while (0)
+
+static inline unsigned int decode_guest_config0(struct cpuinfo_mips *c)
+{
+	unsigned int config0;
+
+	probe_gc0_config(config, config0, MIPS_CONF_M);
+
+	if (config0 & MIPS_CONF_M)
+		c->guest.conf |= BIT(1);
+	return config0 & MIPS_CONF_M;
+}
+
+static inline unsigned int decode_guest_config1(struct cpuinfo_mips *c)
+{
+	unsigned int config1, config1_dyn;
+
+	probe_gc0_config_dyn(config1, config1, config1_dyn,
+			     MIPS_CONF_M | MIPS_CONF1_PC | MIPS_CONF1_WR |
+			     MIPS_CONF1_FP);
+
+	if (config1 & MIPS_CONF1_FP)
+		c->guest.options |= MIPS_CPU_FPU;
+	if (config1_dyn & MIPS_CONF1_FP)
+		c->guest.options_dyn |= MIPS_CPU_FPU;
+
+	if (config1 & MIPS_CONF1_WR)
+		c->guest.options |= MIPS_CPU_WATCH;
+	if (config1_dyn & MIPS_CONF1_WR)
+		c->guest.options_dyn |= MIPS_CPU_WATCH;
+
+	if (config1 & MIPS_CONF1_PC)
+		c->guest.options |= MIPS_CPU_PERF;
+	if (config1_dyn & MIPS_CONF1_PC)
+		c->guest.options_dyn |= MIPS_CPU_PERF;
+
+	if (config1 & MIPS_CONF_M)
+		c->guest.conf |= BIT(2);
+	return config1 & MIPS_CONF_M;
+}
+
+static inline unsigned int decode_guest_config2(struct cpuinfo_mips *c)
+{
+	unsigned int config2;
+
+	probe_gc0_config(config2, config2, MIPS_CONF_M);
+
+	if (config2 & MIPS_CONF_M)
+		c->guest.conf |= BIT(3);
+	return config2 & MIPS_CONF_M;
+}
+
+static inline unsigned int decode_guest_config3(struct cpuinfo_mips *c)
+{
+	unsigned int config3, config3_dyn;
+
+	probe_gc0_config_dyn(config3, config3, config3_dyn,
+			     MIPS_CONF_M | MIPS_CONF3_MSA | MIPS_CONF3_ULRI |
+			     MIPS_CONF3_CTXTC);
+
+	if (config3 & MIPS_CONF3_CTXTC)
+		c->guest.options |= MIPS_CPU_CTXTC;
+	if (config3_dyn & MIPS_CONF3_CTXTC)
+		c->guest.options_dyn |= MIPS_CPU_CTXTC;
+
+	if (config3 & MIPS_CONF3_PW)
+		c->guest.options |= MIPS_CPU_HTW;
+
+	if (config3 & MIPS_CONF3_ULRI)
+		c->guest.options |= MIPS_CPU_ULRI;
+
+	if (config3 & MIPS_CONF3_SC)
+		c->guest.options |= MIPS_CPU_SEGMENTS;
+
+	if (config3 & MIPS_CONF3_BI)
+		c->guest.options |= MIPS_CPU_BADINSTR;
+	if (config3 & MIPS_CONF3_BP)
+		c->guest.options |= MIPS_CPU_BADINSTRP;
+
+	if (config3 & MIPS_CONF3_MSA)
+		c->guest.ases |= MIPS_ASE_MSA;
+	if (config3_dyn & MIPS_CONF3_MSA)
+		c->guest.ases_dyn |= MIPS_ASE_MSA;
+
+	if (config3 & MIPS_CONF_M)
+		c->guest.conf |= BIT(4);
+	return config3 & MIPS_CONF_M;
+}
+
+static inline unsigned int decode_guest_config4(struct cpuinfo_mips *c)
+{
+	unsigned int config4;
+
+	probe_gc0_config(config4, config4,
+			 MIPS_CONF_M | MIPS_CONF4_KSCREXIST);
+
+	c->guest.kscratch_mask = (config4 & MIPS_CONF4_KSCREXIST)
+				>> MIPS_CONF4_KSCREXIST_SHIFT;
+
+	if (config4 & MIPS_CONF_M)
+		c->guest.conf |= BIT(5);
+	return config4 & MIPS_CONF_M;
+}
+
+static inline unsigned int decode_guest_config5(struct cpuinfo_mips *c)
+{
+	unsigned int config5, config5_dyn;
+
+	probe_gc0_config_dyn(config5, config5, config5_dyn,
+			 MIPS_CONF_M | MIPS_CONF5_MVH | MIPS_CONF5_MRP);
+
+	if (config5 & MIPS_CONF5_MRP)
+		c->guest.options |= MIPS_CPU_MAAR;
+	if (config5_dyn & MIPS_CONF5_MRP)
+		c->guest.options_dyn |= MIPS_CPU_MAAR;
+
+	if (config5 & MIPS_CONF5_LLB)
+		c->guest.options |= MIPS_CPU_RW_LLB;
+
+	if (config5 & MIPS_CONF5_MVH)
+		c->guest.options |= MIPS_CPU_MVH;
+
+	if (config5 & MIPS_CONF_M)
+		c->guest.conf |= BIT(6);
+	return config5 & MIPS_CONF_M;
+}
+
+static inline void decode_guest_configs(struct cpuinfo_mips *c)
+{
+	unsigned int ok;
+
+	ok = decode_guest_config0(c);
+	if (ok)
+		ok = decode_guest_config1(c);
+	if (ok)
+		ok = decode_guest_config2(c);
+	if (ok)
+		ok = decode_guest_config3(c);
+	if (ok)
+		ok = decode_guest_config4(c);
+	if (ok)
+		decode_guest_config5(c);
+}
+
+static inline void cpu_probe_guestctl0(struct cpuinfo_mips *c)
+{
+	unsigned int guestctl0, temp;
+
+	guestctl0 = read_c0_guestctl0();
+
+	if (guestctl0 & MIPS_GCTL0_G0E)
+		c->options |= MIPS_CPU_GUESTCTL0EXT;
+	if (guestctl0 & MIPS_GCTL0_G1)
+		c->options |= MIPS_CPU_GUESTCTL1;
+	if (guestctl0 & MIPS_GCTL0_G2)
+		c->options |= MIPS_CPU_GUESTCTL2;
+	if (!(guestctl0 & MIPS_GCTL0_RAD)) {
+		c->options |= MIPS_CPU_GUESTID;
+
+		/*
+		 * Probe for Direct Root to Guest (DRG). Set GuestCtl1.RID = 0
+		 * first, otherwise all data accesses will be fully virtualised
+		 * as if they were performed by guest mode.
+		 */
+		write_c0_guestctl1(0);
+		tlbw_use_hazard();
+
+		write_c0_guestctl0(guestctl0 | MIPS_GCTL0_DRG);
+		back_to_back_c0_hazard();
+		temp = read_c0_guestctl0();
+
+		if (temp & MIPS_GCTL0_DRG) {
+			write_c0_guestctl0(guestctl0);
+			c->options |= MIPS_CPU_DRG;
+		}
+	}
+}
+
+static inline void cpu_probe_guestctl1(struct cpuinfo_mips *c)
+{
+	if (cpu_has_guestid) {
+		/* determine the number of bits of GuestID available */
+		write_c0_guestctl1(MIPS_GCTL1_ID);
+		back_to_back_c0_hazard();
+		c->guestid_mask = (read_c0_guestctl1() & MIPS_GCTL1_ID)
+						>> MIPS_GCTL1_ID_SHIFT;
+		write_c0_guestctl1(0);
+	}
+}
+
+static inline void cpu_probe_gtoffset(struct cpuinfo_mips *c)
+{
+	/* determine the number of bits of GTOffset available */
+	write_c0_gtoffset(0xffffffff);
+	back_to_back_c0_hazard();
+	c->gtoffset_mask = read_c0_gtoffset();
+	write_c0_gtoffset(0);
+}
+
+static inline void cpu_probe_vz(struct cpuinfo_mips *c)
+{
+	cpu_probe_guestctl0(c);
+	if (cpu_has_guestctl1)
+		cpu_probe_guestctl1(c);
+
+	cpu_probe_gtoffset(c);
+
+	decode_guest_configs(c);
+}
+
+#define R4K_OPTS (MIPS_CPU_TLB | MIPS_CPU_4KEX | MIPS_CPU_4K_CACHE \
+		| MIPS_CPU_COUNTER)
+
+static inline void cpu_probe_legacy(struct cpuinfo_mips *c, unsigned int cpu)
+{
+	switch (c->processor_id & PRID_IMP_MASK) {
+	case PRID_IMP_R2000:
+		c->cputype = CPU_R2000;
+		__cpu_name[cpu] = "R2000";
+		c->fpu_msk31 |= FPU_CSR_CONDX | FPU_CSR_FS;
+		c->options = MIPS_CPU_TLB | MIPS_CPU_3K_CACHE |
+			     MIPS_CPU_NOFPUEX;
+		if (__cpu_has_fpu())
+			c->options |= MIPS_CPU_FPU;
+		c->tlbsize = 64;
+		break;
+	case PRID_IMP_R3000:
+		if ((c->processor_id & PRID_REV_MASK) == PRID_REV_R3000A) {
+			if (cpu_has_confreg()) {
+				c->cputype = CPU_R3081E;
+				__cpu_name[cpu] = "R3081";
+			} else {
+				c->cputype = CPU_R3000A;
+				__cpu_name[cpu] = "R3000A";
+			}
+		} else {
+			c->cputype = CPU_R3000;
+			__cpu_name[cpu] = "R3000";
+		}
+		c->fpu_msk31 |= FPU_CSR_CONDX | FPU_CSR_FS;
+		c->options = MIPS_CPU_TLB | MIPS_CPU_3K_CACHE |
+			     MIPS_CPU_NOFPUEX;
+		if (__cpu_has_fpu())
+			c->options |= MIPS_CPU_FPU;
+		c->tlbsize = 64;
+		break;
+	case PRID_IMP_R4000:
+		if (read_c0_config() & CONF_SC) {
+			if ((c->processor_id & PRID_REV_MASK) >=
+			    PRID_REV_R4400) {
+				c->cputype = CPU_R4400PC;
+				__cpu_name[cpu] = "R4400PC";
+			} else {
+				c->cputype = CPU_R4000PC;
+				__cpu_name[cpu] = "R4000PC";
+			}
+		} else {
+			int cca = read_c0_config() & CONF_CM_CMASK;
+			int mc;
+
+			/*
+			 * SC and MC versions can't be reliably told apart,
+			 * but only the latter support coherent caching
+			 * modes so assume the firmware has set the KSEG0
+			 * coherency attribute reasonably (if uncached, we
+			 * assume SC).
+			 */
+			switch (cca) {
+			case CONF_CM_CACHABLE_CE:
+			case CONF_CM_CACHABLE_COW:
+			case CONF_CM_CACHABLE_CUW:
+				mc = 1;
+				break;
+			default:
+				mc = 0;
+				break;
+			}
+			if ((c->processor_id & PRID_REV_MASK) >=
+			    PRID_REV_R4400) {
+				c->cputype = mc ? CPU_R4400MC : CPU_R4400SC;
+				__cpu_name[cpu] = mc ? "R4400MC" : "R4400SC";
+			} else {
+				c->cputype = mc ? CPU_R4000MC : CPU_R4000SC;
+				__cpu_name[cpu] = mc ? "R4000MC" : "R4000SC";
+			}
+		}
+
+		set_isa(c, MIPS_CPU_ISA_III);
+		c->fpu_msk31 |= FPU_CSR_CONDX;
+		c->options = R4K_OPTS | MIPS_CPU_FPU | MIPS_CPU_32FPR |
+			     MIPS_CPU_WATCH | MIPS_CPU_VCE |
+			     MIPS_CPU_LLSC;
+		c->tlbsize = 48;
+		break;
+	case PRID_IMP_VR41XX:
+		set_isa(c, MIPS_CPU_ISA_III);
+		c->fpu_msk31 |= FPU_CSR_CONDX;
+		c->options = R4K_OPTS;
+		c->tlbsize = 32;
+		switch (c->processor_id & 0xf0) {
+		case PRID_REV_VR4111:
+			c->cputype = CPU_VR4111;
+			__cpu_name[cpu] = "NEC VR4111";
+			break;
+		case PRID_REV_VR4121:
+			c->cputype = CPU_VR4121;
+			__cpu_name[cpu] = "NEC VR4121";
+			break;
+		case PRID_REV_VR4122:
+			if ((c->processor_id & 0xf) < 0x3) {
+				c->cputype = CPU_VR4122;
+				__cpu_name[cpu] = "NEC VR4122";
+			} else {
+				c->cputype = CPU_VR4181A;
+				__cpu_name[cpu] = "NEC VR4181A";
+			}
+			break;
+		case PRID_REV_VR4130:
+			if ((c->processor_id & 0xf) < 0x4) {
+				c->cputype = CPU_VR4131;
+				__cpu_name[cpu] = "NEC VR4131";
+			} else {
+				c->cputype = CPU_VR4133;
+				c->options |= MIPS_CPU_LLSC;
+				__cpu_name[cpu] = "NEC VR4133";
+			}
+			break;
+		default:
+			printk(KERN_INFO "Unexpected CPU of NEC VR4100 series\n");
+			c->cputype = CPU_VR41XX;
+			__cpu_name[cpu] = "NEC Vr41xx";
+			break;
+		}
+		break;
+	case PRID_IMP_R4600:
+		c->cputype = CPU_R4600;
+		__cpu_name[cpu] = "R4600";
+		set_isa(c, MIPS_CPU_ISA_III);
+		c->fpu_msk31 |= FPU_CSR_CONDX;
+		c->options = R4K_OPTS | MIPS_CPU_FPU | MIPS_CPU_32FPR |
+			     MIPS_CPU_LLSC;
+		c->tlbsize = 48;
+		break;
+	#if 0
+	case PRID_IMP_R4650:
+		/*
+		 * This processor doesn't have an MMU, so it's not
+		 * "real easy" to run Linux on it. It is left purely
+		 * for documentation.  Commented out because it shares
+		 * it's c0_prid id number with the TX3900.
+		 */
+		c->cputype = CPU_R4650;
+		__cpu_name[cpu] = "R4650";
+		set_isa(c, MIPS_CPU_ISA_III);
+		c->fpu_msk31 |= FPU_CSR_CONDX;
+		c->options = R4K_OPTS | MIPS_CPU_FPU | MIPS_CPU_LLSC;
+		c->tlbsize = 48;
+		break;
+	#endif
+	case PRID_IMP_TX39:
+		c->fpu_msk31 |= FPU_CSR_CONDX | FPU_CSR_FS;
+		c->options = MIPS_CPU_TLB | MIPS_CPU_TX39_CACHE;
+
+		if ((c->processor_id & 0xf0) == (PRID_REV_TX3927 & 0xf0)) {
+			c->cputype = CPU_TX3927;
+			__cpu_name[cpu] = "TX3927";
+			c->tlbsize = 64;
+		} else {
+			switch (c->processor_id & PRID_REV_MASK) {
+			case PRID_REV_TX3912:
+				c->cputype = CPU_TX3912;
+				__cpu_name[cpu] = "TX3912";
+				c->tlbsize = 32;
+				break;
+			case PRID_REV_TX3922:
+				c->cputype = CPU_TX3922;
+				__cpu_name[cpu] = "TX3922";
+				c->tlbsize = 64;
+				break;
+			}
+		}
+		break;
+	case PRID_IMP_R4700:
+		c->cputype = CPU_R4700;
+		__cpu_name[cpu] = "R4700";
+		set_isa(c, MIPS_CPU_ISA_III);
+		c->fpu_msk31 |= FPU_CSR_CONDX;
+		c->options = R4K_OPTS | MIPS_CPU_FPU | MIPS_CPU_32FPR |
+			     MIPS_CPU_LLSC;
+		c->tlbsize = 48;
+		break;
+	case PRID_IMP_TX49:
+		c->cputype = CPU_TX49XX;
+		__cpu_name[cpu] = "R49XX";
+		set_isa(c, MIPS_CPU_ISA_III);
+		c->fpu_msk31 |= FPU_CSR_CONDX;
+		c->options = R4K_OPTS | MIPS_CPU_LLSC;
+		if (!(c->processor_id & 0x08))
+			c->options |= MIPS_CPU_FPU | MIPS_CPU_32FPR;
+		c->tlbsize = 48;
+		break;
+	case PRID_IMP_R5000:
+		c->cputype = CPU_R5000;
+		__cpu_name[cpu] = "R5000";
+		set_isa(c, MIPS_CPU_ISA_IV);
+		c->options = R4K_OPTS | MIPS_CPU_FPU | MIPS_CPU_32FPR |
+			     MIPS_CPU_LLSC;
+		c->tlbsize = 48;
+		break;
+	case PRID_IMP_R5500:
+		c->cputype = CPU_R5500;
+		__cpu_name[cpu] = "R5500";
+		set_isa(c, MIPS_CPU_ISA_IV);
+		c->options = R4K_OPTS | MIPS_CPU_FPU | MIPS_CPU_32FPR |
+			     MIPS_CPU_WATCH | MIPS_CPU_LLSC;
+		c->tlbsize = 48;
+		break;
+	case PRID_IMP_NEVADA:
+		c->cputype = CPU_NEVADA;
+		__cpu_name[cpu] = "Nevada";
+		set_isa(c, MIPS_CPU_ISA_IV);
+		c->options = R4K_OPTS | MIPS_CPU_FPU | MIPS_CPU_32FPR |
+			     MIPS_CPU_DIVEC | MIPS_CPU_LLSC;
+		c->tlbsize = 48;
+		break;
+	case PRID_IMP_RM7000:
+		c->cputype = CPU_RM7000;
+		__cpu_name[cpu] = "RM7000";
+		set_isa(c, MIPS_CPU_ISA_IV);
+		c->options = R4K_OPTS | MIPS_CPU_FPU | MIPS_CPU_32FPR |
+			     MIPS_CPU_LLSC;
+		/*
+		 * Undocumented RM7000:	 Bit 29 in the info register of
+		 * the RM7000 v2.0 indicates if the TLB has 48 or 64
+		 * entries.
+		 *
+		 * 29	   1 =>	   64 entry JTLB
+		 *	   0 =>	   48 entry JTLB
+		 */
+		c->tlbsize = (read_c0_info() & (1 << 29)) ? 64 : 48;
+		break;
+	case PRID_IMP_R10000:
+		c->cputype = CPU_R10000;
+		__cpu_name[cpu] = "R10000";
+		set_isa(c, MIPS_CPU_ISA_IV);
+		c->options = MIPS_CPU_TLB | MIPS_CPU_4K_CACHE | MIPS_CPU_4KEX |
+			     MIPS_CPU_FPU | MIPS_CPU_32FPR |
+			     MIPS_CPU_COUNTER | MIPS_CPU_WATCH |
+			     MIPS_CPU_LLSC;
+		c->tlbsize = 64;
+		break;
+	case PRID_IMP_R12000:
+		c->cputype = CPU_R12000;
+		__cpu_name[cpu] = "R12000";
+		set_isa(c, MIPS_CPU_ISA_IV);
+		c->options = MIPS_CPU_TLB | MIPS_CPU_4K_CACHE | MIPS_CPU_4KEX |
+			     MIPS_CPU_FPU | MIPS_CPU_32FPR |
+			     MIPS_CPU_COUNTER | MIPS_CPU_WATCH |
+			     MIPS_CPU_LLSC;
+		c->tlbsize = 64;
+		write_c0_r10k_diag(read_c0_r10k_diag() | R10K_DIAG_E_GHIST);
+		break;
+	case PRID_IMP_R14000:
+		if (((c->processor_id >> 4) & 0x0f) > 2) {
+			c->cputype = CPU_R16000;
+			__cpu_name[cpu] = "R16000";
+		} else {
+			c->cputype = CPU_R14000;
+			__cpu_name[cpu] = "R14000";
+		}
+		set_isa(c, MIPS_CPU_ISA_IV);
+		c->options = MIPS_CPU_TLB | MIPS_CPU_4K_CACHE | MIPS_CPU_4KEX |
+			     MIPS_CPU_FPU | MIPS_CPU_32FPR |
+			     MIPS_CPU_COUNTER | MIPS_CPU_WATCH |
+			     MIPS_CPU_LLSC;
+		c->tlbsize = 64;
+		write_c0_r10k_diag(read_c0_r10k_diag() | R10K_DIAG_E_GHIST);
+		break;
+	case PRID_IMP_LOONGSON_64C:  /* Loongson-2/3 */
+		switch (c->processor_id & PRID_REV_MASK) {
+		case PRID_REV_LOONGSON2E:
+			c->cputype = CPU_LOONGSON2EF;
+			__cpu_name[cpu] = "ICT Loongson-2";
+			set_elf_platform(cpu, "loongson2e");
+			set_isa(c, MIPS_CPU_ISA_III);
+			c->fpu_msk31 |= FPU_CSR_CONDX;
+			break;
+		case PRID_REV_LOONGSON2F:
+			c->cputype = CPU_LOONGSON2EF;
+			__cpu_name[cpu] = "ICT Loongson-2";
+			set_elf_platform(cpu, "loongson2f");
+			set_isa(c, MIPS_CPU_ISA_III);
+			c->fpu_msk31 |= FPU_CSR_CONDX;
+			break;
+		case PRID_REV_LOONGSON3A_R1:
+			c->cputype = CPU_LOONGSON64;
+			__cpu_name[cpu] = "ICT Loongson-3";
+			set_elf_platform(cpu, "loongson3a");
+			set_isa(c, MIPS_CPU_ISA_M64R1);
+			c->ases |= (MIPS_ASE_LOONGSON_MMI | MIPS_ASE_LOONGSON_CAM |
+				MIPS_ASE_LOONGSON_EXT);
+			break;
+		case PRID_REV_LOONGSON3B_R1:
+		case PRID_REV_LOONGSON3B_R2:
+			c->cputype = CPU_LOONGSON64;
+			__cpu_name[cpu] = "ICT Loongson-3";
+			set_elf_platform(cpu, "loongson3b");
+			set_isa(c, MIPS_CPU_ISA_M64R1);
+			c->ases |= (MIPS_ASE_LOONGSON_MMI | MIPS_ASE_LOONGSON_CAM |
+				MIPS_ASE_LOONGSON_EXT);
+			break;
+		}
+
+		c->options = R4K_OPTS |
+			     MIPS_CPU_FPU | MIPS_CPU_LLSC |
+			     MIPS_CPU_32FPR;
+		c->tlbsize = 64;
+		set_cpu_asid_mask(c, MIPS_ENTRYHI_ASID);
+		c->writecombine = _CACHE_UNCACHED_ACCELERATED;
+		break;
+	case PRID_IMP_LOONGSON_32:  /* Loongson-1 */
+		decode_configs(c);
+
+		c->cputype = CPU_LOONGSON32;
+
+		switch (c->processor_id & PRID_REV_MASK) {
+		case PRID_REV_LOONGSON1B:
+			__cpu_name[cpu] = "Loongson 1B";
+			break;
+		}
+
+		break;
+	}
+}
+
+static inline void cpu_probe_mips(struct cpuinfo_mips *c, unsigned int cpu)
+{
+	c->writecombine = _CACHE_UNCACHED_ACCELERATED;
+	switch (c->processor_id & PRID_IMP_MASK) {
+	case PRID_IMP_QEMU_GENERIC:
+		c->writecombine = _CACHE_UNCACHED;
+		c->cputype = CPU_QEMU_GENERIC;
+		__cpu_name[cpu] = "MIPS GENERIC QEMU";
+		break;
+	case PRID_IMP_4KC:
+		c->cputype = CPU_4KC;
+		c->writecombine = _CACHE_UNCACHED;
+		__cpu_name[cpu] = "MIPS 4Kc";
+		break;
+	case PRID_IMP_4KEC:
+	case PRID_IMP_4KECR2:
+		c->cputype = CPU_4KEC;
+		c->writecombine = _CACHE_UNCACHED;
+		__cpu_name[cpu] = "MIPS 4KEc";
+		break;
+	case PRID_IMP_4KSC:
+	case PRID_IMP_4KSD:
+		c->cputype = CPU_4KSC;
+		c->writecombine = _CACHE_UNCACHED;
+		__cpu_name[cpu] = "MIPS 4KSc";
+		break;
+	case PRID_IMP_5KC:
+		c->cputype = CPU_5KC;
+		c->writecombine = _CACHE_UNCACHED;
+		__cpu_name[cpu] = "MIPS 5Kc";
+		break;
+	case PRID_IMP_5KE:
+		c->cputype = CPU_5KE;
+		c->writecombine = _CACHE_UNCACHED;
+		__cpu_name[cpu] = "MIPS 5KE";
+		break;
+	case PRID_IMP_20KC:
+		c->cputype = CPU_20KC;
+		c->writecombine = _CACHE_UNCACHED;
+		__cpu_name[cpu] = "MIPS 20Kc";
+		break;
+	case PRID_IMP_24K:
+		c->cputype = CPU_24K;
+		c->writecombine = _CACHE_UNCACHED;
+		__cpu_name[cpu] = "MIPS 24Kc";
+		break;
+	case PRID_IMP_24KE:
+		c->cputype = CPU_24K;
+		c->writecombine = _CACHE_UNCACHED;
+		__cpu_name[cpu] = "MIPS 24KEc";
+		break;
+	case PRID_IMP_25KF:
+		c->cputype = CPU_25KF;
+		c->writecombine = _CACHE_UNCACHED;
+		__cpu_name[cpu] = "MIPS 25Kc";
+		break;
+	case PRID_IMP_34K:
+		c->cputype = CPU_34K;
+		c->writecombine = _CACHE_UNCACHED;
+		__cpu_name[cpu] = "MIPS 34Kc";
+		cpu_set_mt_per_tc_perf(c);
+		break;
+	case PRID_IMP_74K:
+		c->cputype = CPU_74K;
+		c->writecombine = _CACHE_UNCACHED;
+		__cpu_name[cpu] = "MIPS 74Kc";
+		break;
+	case PRID_IMP_M14KC:
+		c->cputype = CPU_M14KC;
+		c->writecombine = _CACHE_UNCACHED;
+		__cpu_name[cpu] = "MIPS M14Kc";
+		break;
+	case PRID_IMP_M14KEC:
+		c->cputype = CPU_M14KEC;
+		c->writecombine = _CACHE_UNCACHED;
+		__cpu_name[cpu] = "MIPS M14KEc";
+		break;
+	case PRID_IMP_1004K:
+		c->cputype = CPU_1004K;
+		c->writecombine = _CACHE_UNCACHED;
+		__cpu_name[cpu] = "MIPS 1004Kc";
+		cpu_set_mt_per_tc_perf(c);
+		break;
+	case PRID_IMP_1074K:
+		c->cputype = CPU_1074K;
+		c->writecombine = _CACHE_UNCACHED;
+		__cpu_name[cpu] = "MIPS 1074Kc";
+		break;
+	case PRID_IMP_INTERAPTIV_UP:
+		c->cputype = CPU_INTERAPTIV;
+		__cpu_name[cpu] = "MIPS interAptiv";
+		cpu_set_mt_per_tc_perf(c);
+		break;
+	case PRID_IMP_INTERAPTIV_MP:
+		c->cputype = CPU_INTERAPTIV;
+		__cpu_name[cpu] = "MIPS interAptiv (multi)";
+		cpu_set_mt_per_tc_perf(c);
+		break;
+	case PRID_IMP_PROAPTIV_UP:
+		c->cputype = CPU_PROAPTIV;
+		__cpu_name[cpu] = "MIPS proAptiv";
+		break;
+	case PRID_IMP_PROAPTIV_MP:
+		c->cputype = CPU_PROAPTIV;
+		__cpu_name[cpu] = "MIPS proAptiv (multi)";
+		break;
+	case PRID_IMP_P5600:
+		c->cputype = CPU_P5600;
+		__cpu_name[cpu] = "MIPS P5600";
+		break;
+	case PRID_IMP_P6600:
+		c->cputype = CPU_P6600;
+		__cpu_name[cpu] = "MIPS P6600";
+		break;
+	case PRID_IMP_I6400:
+		c->cputype = CPU_I6400;
+		__cpu_name[cpu] = "MIPS I6400";
+		break;
+	case PRID_IMP_I6500:
+		c->cputype = CPU_I6500;
+		__cpu_name[cpu] = "MIPS I6500";
+		break;
+	case PRID_IMP_M5150:
+		c->cputype = CPU_M5150;
+		__cpu_name[cpu] = "MIPS M5150";
+		break;
+	case PRID_IMP_M6250:
+		c->cputype = CPU_M6250;
+		__cpu_name[cpu] = "MIPS M6250";
+		break;
+	}
+
+	decode_configs(c);
+
+	spram_config();
+
+	mm_config(c);
+
+	switch (__get_cpu_type(c->cputype)) {
+	case CPU_M5150:
+	case CPU_P5600:
+		set_isa(c, MIPS_CPU_ISA_M32R5);
+		break;
+	case CPU_I6500:
+		c->options |= MIPS_CPU_SHARED_FTLB_ENTRIES;
+		fallthrough;
+	case CPU_I6400:
+		c->options |= MIPS_CPU_SHARED_FTLB_RAM;
+		fallthrough;
+	default:
+		break;
+	}
+
+	/* Recent MIPS cores use the implementation-dependent ExcCode 16 for
+	 * cache/FTLB parity exceptions.
+	 */
+	switch (__get_cpu_type(c->cputype)) {
+	case CPU_PROAPTIV:
+	case CPU_P5600:
+	case CPU_P6600:
+	case CPU_I6400:
+	case CPU_I6500:
+		c->options |= MIPS_CPU_FTLBPAREX;
+		break;
+	}
+}
+
+static inline void cpu_probe_alchemy(struct cpuinfo_mips *c, unsigned int cpu)
+{
+	decode_configs(c);
+	switch (c->processor_id & PRID_IMP_MASK) {
+	case PRID_IMP_AU1_REV1:
+	case PRID_IMP_AU1_REV2:
+		c->cputype = CPU_ALCHEMY;
+		switch ((c->processor_id >> 24) & 0xff) {
+		case 0:
+			__cpu_name[cpu] = "Au1000";
+			break;
+		case 1:
+			__cpu_name[cpu] = "Au1500";
+			break;
+		case 2:
+			__cpu_name[cpu] = "Au1100";
+			break;
+		case 3:
+			__cpu_name[cpu] = "Au1550";
+			break;
+		case 4:
+			__cpu_name[cpu] = "Au1200";
+			if ((c->processor_id & PRID_REV_MASK) == 2)
+				__cpu_name[cpu] = "Au1250";
+			break;
+		case 5:
+			__cpu_name[cpu] = "Au1210";
+			break;
+		default:
+			__cpu_name[cpu] = "Au1xxx";
+			break;
+		}
+		break;
+	}
+}
+
+static inline void cpu_probe_sibyte(struct cpuinfo_mips *c, unsigned int cpu)
+{
+	decode_configs(c);
+
+	c->writecombine = _CACHE_UNCACHED_ACCELERATED;
+	switch (c->processor_id & PRID_IMP_MASK) {
+	case PRID_IMP_SB1:
+		c->cputype = CPU_SB1;
+		__cpu_name[cpu] = "SiByte SB1";
+		/* FPU in pass1 is known to have issues. */
+		if ((c->processor_id & PRID_REV_MASK) < 0x02)
+			c->options &= ~(MIPS_CPU_FPU | MIPS_CPU_32FPR);
+		break;
+	case PRID_IMP_SB1A:
+		c->cputype = CPU_SB1A;
+		__cpu_name[cpu] = "SiByte SB1A";
+		break;
+	}
+}
+
+static inline void cpu_probe_sandcraft(struct cpuinfo_mips *c, unsigned int cpu)
+{
+	decode_configs(c);
+	switch (c->processor_id & PRID_IMP_MASK) {
+	case PRID_IMP_SR71000:
+		c->cputype = CPU_SR71000;
+		__cpu_name[cpu] = "Sandcraft SR71000";
+		c->scache.ways = 8;
+		c->tlbsize = 64;
+		break;
+	}
+}
+
+static inline void cpu_probe_nxp(struct cpuinfo_mips *c, unsigned int cpu)
+{
+	decode_configs(c);
+	switch (c->processor_id & PRID_IMP_MASK) {
+	case PRID_IMP_PR4450:
+		c->cputype = CPU_PR4450;
+		__cpu_name[cpu] = "Philips PR4450";
+		set_isa(c, MIPS_CPU_ISA_M32R1);
+		break;
+	}
+}
+
+static inline void cpu_probe_broadcom(struct cpuinfo_mips *c, unsigned int cpu)
+{
+	decode_configs(c);
+	switch (c->processor_id & PRID_IMP_MASK) {
+	case PRID_IMP_BMIPS32_REV4:
+	case PRID_IMP_BMIPS32_REV8:
+		c->cputype = CPU_BMIPS32;
+		__cpu_name[cpu] = "Broadcom BMIPS32";
+		set_elf_platform(cpu, "bmips32");
+		break;
+	case PRID_IMP_BMIPS3300:
+	case PRID_IMP_BMIPS3300_ALT:
+	case PRID_IMP_BMIPS3300_BUG:
+		c->cputype = CPU_BMIPS3300;
+		__cpu_name[cpu] = "Broadcom BMIPS3300";
+		set_elf_platform(cpu, "bmips3300");
+		break;
+	case PRID_IMP_BMIPS43XX: {
+		int rev = c->processor_id & PRID_REV_MASK;
+
+		if (rev >= PRID_REV_BMIPS4380_LO &&
+				rev <= PRID_REV_BMIPS4380_HI) {
+			c->cputype = CPU_BMIPS4380;
+			__cpu_name[cpu] = "Broadcom BMIPS4380";
+			set_elf_platform(cpu, "bmips4380");
+			c->options |= MIPS_CPU_RIXI;
+		} else {
+			c->cputype = CPU_BMIPS4350;
+			__cpu_name[cpu] = "Broadcom BMIPS4350";
+			set_elf_platform(cpu, "bmips4350");
+		}
+		break;
+	}
+	case PRID_IMP_BMIPS5000:
+	case PRID_IMP_BMIPS5200:
+		c->cputype = CPU_BMIPS5000;
+		if ((c->processor_id & PRID_IMP_MASK) == PRID_IMP_BMIPS5200)
+			__cpu_name[cpu] = "Broadcom BMIPS5200";
+		else
+			__cpu_name[cpu] = "Broadcom BMIPS5000";
+		set_elf_platform(cpu, "bmips5000");
+		c->options |= MIPS_CPU_ULRI | MIPS_CPU_RIXI;
+		break;
+	}
+}
+
+static inline void cpu_probe_cavium(struct cpuinfo_mips *c, unsigned int cpu)
+{
+	decode_configs(c);
+	switch (c->processor_id & PRID_IMP_MASK) {
+	case PRID_IMP_CAVIUM_CN38XX:
+	case PRID_IMP_CAVIUM_CN31XX:
+	case PRID_IMP_CAVIUM_CN30XX:
+		c->cputype = CPU_CAVIUM_OCTEON;
+		__cpu_name[cpu] = "Cavium Octeon";
+		goto platform;
+	case PRID_IMP_CAVIUM_CN58XX:
+	case PRID_IMP_CAVIUM_CN56XX:
+	case PRID_IMP_CAVIUM_CN50XX:
+	case PRID_IMP_CAVIUM_CN52XX:
+		c->cputype = CPU_CAVIUM_OCTEON_PLUS;
+		__cpu_name[cpu] = "Cavium Octeon+";
+platform:
+		set_elf_platform(cpu, "octeon");
+		break;
+	case PRID_IMP_CAVIUM_CN61XX:
+	case PRID_IMP_CAVIUM_CN63XX:
+	case PRID_IMP_CAVIUM_CN66XX:
+	case PRID_IMP_CAVIUM_CN68XX:
+	case PRID_IMP_CAVIUM_CNF71XX:
+		c->cputype = CPU_CAVIUM_OCTEON2;
+		__cpu_name[cpu] = "Cavium Octeon II";
+		set_elf_platform(cpu, "octeon2");
+		break;
+	case PRID_IMP_CAVIUM_CN70XX:
+	case PRID_IMP_CAVIUM_CN73XX:
+	case PRID_IMP_CAVIUM_CNF75XX:
+	case PRID_IMP_CAVIUM_CN78XX:
+		c->cputype = CPU_CAVIUM_OCTEON3;
+		__cpu_name[cpu] = "Cavium Octeon III";
+		set_elf_platform(cpu, "octeon3");
+		break;
+	default:
+		printk(KERN_INFO "Unknown Octeon chip!\n");
+		c->cputype = CPU_UNKNOWN;
+		break;
+	}
+}
+
+#ifdef CONFIG_CPU_LOONGSON64
+#include <loongson_regs.h>
+
+static inline void decode_cpucfg(struct cpuinfo_mips *c)
+{
+	u32 cfg1 = read_cpucfg(LOONGSON_CFG1);
+	u32 cfg2 = read_cpucfg(LOONGSON_CFG2);
+	u32 cfg3 = read_cpucfg(LOONGSON_CFG3);
+
+	if (cfg1 & LOONGSON_CFG1_MMI)
+		c->ases |= MIPS_ASE_LOONGSON_MMI;
+
+	if (cfg2 & LOONGSON_CFG2_LEXT1)
+		c->ases |= MIPS_ASE_LOONGSON_EXT;
+
+	if (cfg2 & LOONGSON_CFG2_LEXT2)
+		c->ases |= MIPS_ASE_LOONGSON_EXT2;
+
+	if (cfg2 & LOONGSON_CFG2_LSPW) {
+		c->options |= MIPS_CPU_LDPTE;
+		c->guest.options |= MIPS_CPU_LDPTE;
+	}
+
+	if (cfg3 & LOONGSON_CFG3_LCAMP)
+		c->ases |= MIPS_ASE_LOONGSON_CAM;
+}
+
+static inline void cpu_probe_loongson(struct cpuinfo_mips *c, unsigned int cpu)
+{
+	c->cputype = CPU_LOONGSON64;
+
+	/* All Loongson processors covered here define ExcCode 16 as GSExc. */
+	decode_configs(c);
+	c->options |= MIPS_CPU_GSEXCEX;
+
+	switch (c->processor_id & PRID_IMP_MASK) {
+	case PRID_IMP_LOONGSON_64R: /* Loongson-64 Reduced */
+		switch (c->processor_id & PRID_REV_MASK) {
+		case PRID_REV_LOONGSON2K_R1_0:
+		case PRID_REV_LOONGSON2K_R1_1:
+		case PRID_REV_LOONGSON2K_R1_2:
+		case PRID_REV_LOONGSON2K_R1_3:
+			__cpu_name[cpu] = "Loongson-2K";
+			set_elf_platform(cpu, "gs264e");
+			set_isa(c, MIPS_CPU_ISA_M64R2);
+			break;
+		}
+		c->ases |= (MIPS_ASE_LOONGSON_MMI | MIPS_ASE_LOONGSON_EXT |
+				MIPS_ASE_LOONGSON_EXT2);
+		break;
+	case PRID_IMP_LOONGSON_64C:  /* Loongson-3 Classic */
+		switch (c->processor_id & PRID_REV_MASK) {
+		case PRID_REV_LOONGSON3A_R2_0:
+		case PRID_REV_LOONGSON3A_R2_1:
+			__cpu_name[cpu] = "ICT Loongson-3";
+			set_elf_platform(cpu, "loongson3a");
+			set_isa(c, MIPS_CPU_ISA_M64R2);
+			break;
+		case PRID_REV_LOONGSON3A_R3_0:
+		case PRID_REV_LOONGSON3A_R3_1:
+			__cpu_name[cpu] = "ICT Loongson-3";
+			set_elf_platform(cpu, "loongson3a");
+			set_isa(c, MIPS_CPU_ISA_M64R2);
+			break;
+		}
+		/*
+		 * Loongson-3 Classic did not implement MIPS standard TLBINV
+		 * but implemented TLBINVF and EHINV. As currently we're only
+		 * using these two features, enable MIPS_CPU_TLBINV as well.
+		 *
+		 * Also some early Loongson-3A2000 had wrong TLB type in Config
+		 * register, we correct it here.
+		 */
+		c->options |= MIPS_CPU_FTLB | MIPS_CPU_TLBINV | MIPS_CPU_LDPTE;
+		c->ases |= (MIPS_ASE_LOONGSON_MMI | MIPS_ASE_LOONGSON_CAM |
+			MIPS_ASE_LOONGSON_EXT | MIPS_ASE_LOONGSON_EXT2);
+		c->ases &= ~MIPS_ASE_VZ; /* VZ of Loongson-3A2000/3000 is incomplete */
+		break;
+	case PRID_IMP_LOONGSON_64G:
+		__cpu_name[cpu] = "ICT Loongson-3";
+		set_elf_platform(cpu, "loongson3a");
+		set_isa(c, MIPS_CPU_ISA_M64R2);
+		decode_cpucfg(c);
+		break;
+	default:
+		panic("Unknown Loongson Processor ID!");
+		break;
+	}
+}
+#else
+static inline void cpu_probe_loongson(struct cpuinfo_mips *c, unsigned int cpu) { }
+#endif
+
+static inline void cpu_probe_ingenic(struct cpuinfo_mips *c, unsigned int cpu)
+{
+	decode_configs(c);
+
+	/*
+	 * XBurst misses a config2 register, so config3 decode was skipped in
+	 * decode_configs().
+	 */
+	decode_config3(c);
+
+	/* XBurst does not implement the CP0 counter. */
+	c->options &= ~MIPS_CPU_COUNTER;
+	BUG_ON(__builtin_constant_p(cpu_has_counter) && cpu_has_counter);
+
+	/* XBurst has virtually tagged icache */
+	c->icache.flags |= MIPS_CACHE_VTAG;
+
+	switch (c->processor_id & PRID_IMP_MASK) {
+
+	/* XBurst®1 with MXU1.0/MXU1.1 SIMD ISA */
+	case PRID_IMP_XBURST_REV1:
+
+		/*
+		 * The XBurst core by default attempts to avoid branch target
+		 * buffer lookups by detecting & special casing loops. This
+		 * feature will cause BogoMIPS and lpj calculate in error.
+		 * Set cp0 config7 bit 4 to disable this feature.
+		 */
+		set_c0_config7(MIPS_CONF7_BTB_LOOP_EN);
+
+		switch (c->processor_id & PRID_COMP_MASK) {
+
+		/*
+		 * The config0 register in the XBurst CPUs with a processor ID of
+		 * PRID_COMP_INGENIC_D0 report themselves as MIPS32r2 compatible,
+		 * but they don't actually support this ISA.
+		 */
+		case PRID_COMP_INGENIC_D0:
+			c->isa_level &= ~MIPS_CPU_ISA_M32R2;
+
+			/* FPU is not properly detected on JZ4760(B). */
+			if (c->processor_id == 0x2ed0024f)
+				c->options |= MIPS_CPU_FPU;
+
+			fallthrough;
+
+		/*
+		 * The config0 register in the XBurst CPUs with a processor ID of
+		 * PRID_COMP_INGENIC_D0 or PRID_COMP_INGENIC_D1 has an abandoned
+		 * huge page tlb mode, this mode is not compatible with the MIPS
+		 * standard, it will cause tlbmiss and into an infinite loop
+		 * (line 21 in the tlb-funcs.S) when starting the init process.
+		 * After chip reset, the default is HPTLB mode, Write 0xa9000000
+		 * to cp0 register 5 sel 4 to switch back to VTLB mode to prevent
+		 * getting stuck.
+		 */
+		case PRID_COMP_INGENIC_D1:
+			write_c0_page_ctrl(XBURST_PAGECTRL_HPTLB_DIS);
+			break;
+
+		default:
+			break;
+		}
+		fallthrough;
+
+	/* XBurst®1 with MXU2.0 SIMD ISA */
+	case PRID_IMP_XBURST_REV2:
+		/* Ingenic uses the WA bit to achieve write-combine memory writes */
+		c->writecombine = _CACHE_CACHABLE_WA;
+		c->cputype = CPU_XBURST;
+		__cpu_name[cpu] = "Ingenic XBurst";
+		break;
+
+	/* XBurst®2 with MXU2.1 SIMD ISA */
+	case PRID_IMP_XBURST2:
+		c->cputype = CPU_XBURST;
+		__cpu_name[cpu] = "Ingenic XBurst II";
+		break;
+
+	default:
+		panic("Unknown Ingenic Processor ID!");
+		break;
+	}
+}
+
+static inline void cpu_probe_netlogic(struct cpuinfo_mips *c, int cpu)
+{
+	decode_configs(c);
+
+	if ((c->processor_id & PRID_IMP_MASK) == PRID_IMP_NETLOGIC_AU13XX) {
+		c->cputype = CPU_ALCHEMY;
+		__cpu_name[cpu] = "Au1300";
+		/* following stuff is not for Alchemy */
+		return;
+	}
+
+	c->options = (MIPS_CPU_TLB	 |
+			MIPS_CPU_4KEX	 |
+			MIPS_CPU_COUNTER |
+			MIPS_CPU_DIVEC	 |
+			MIPS_CPU_WATCH	 |
+			MIPS_CPU_EJTAG	 |
+			MIPS_CPU_LLSC);
+
+	switch (c->processor_id & PRID_IMP_MASK) {
+	case PRID_IMP_NETLOGIC_XLP2XX:
+	case PRID_IMP_NETLOGIC_XLP9XX:
+	case PRID_IMP_NETLOGIC_XLP5XX:
+		c->cputype = CPU_XLP;
+		__cpu_name[cpu] = "Broadcom XLPII";
+		break;
+
+	case PRID_IMP_NETLOGIC_XLP8XX:
+	case PRID_IMP_NETLOGIC_XLP3XX:
+		c->cputype = CPU_XLP;
+		__cpu_name[cpu] = "Netlogic XLP";
+		break;
+
+	case PRID_IMP_NETLOGIC_XLR732:
+	case PRID_IMP_NETLOGIC_XLR716:
+	case PRID_IMP_NETLOGIC_XLR532:
+	case PRID_IMP_NETLOGIC_XLR308:
+	case PRID_IMP_NETLOGIC_XLR532C:
+	case PRID_IMP_NETLOGIC_XLR516C:
+	case PRID_IMP_NETLOGIC_XLR508C:
+	case PRID_IMP_NETLOGIC_XLR308C:
+		c->cputype = CPU_XLR;
+		__cpu_name[cpu] = "Netlogic XLR";
+		break;
+
+	case PRID_IMP_NETLOGIC_XLS608:
+	case PRID_IMP_NETLOGIC_XLS408:
+	case PRID_IMP_NETLOGIC_XLS404:
+	case PRID_IMP_NETLOGIC_XLS208:
+	case PRID_IMP_NETLOGIC_XLS204:
+	case PRID_IMP_NETLOGIC_XLS108:
+	case PRID_IMP_NETLOGIC_XLS104:
+	case PRID_IMP_NETLOGIC_XLS616B:
+	case PRID_IMP_NETLOGIC_XLS608B:
+	case PRID_IMP_NETLOGIC_XLS416B:
+	case PRID_IMP_NETLOGIC_XLS412B:
+	case PRID_IMP_NETLOGIC_XLS408B:
+	case PRID_IMP_NETLOGIC_XLS404B:
+		c->cputype = CPU_XLR;
+		__cpu_name[cpu] = "Netlogic XLS";
+		break;
+
+	default:
+		pr_info("Unknown Netlogic chip id [%02x]!\n",
+		       c->processor_id);
+		c->cputype = CPU_XLR;
+		break;
+	}
+
+	if (c->cputype == CPU_XLP) {
+		set_isa(c, MIPS_CPU_ISA_M64R2);
+		c->options |= (MIPS_CPU_FPU | MIPS_CPU_ULRI | MIPS_CPU_MCHECK);
+		/* This will be updated again after all threads are woken up */
+		c->tlbsize = ((read_c0_config6() >> 16) & 0xffff) + 1;
+	} else {
+		set_isa(c, MIPS_CPU_ISA_M64R1);
+		c->tlbsize = ((read_c0_config1() >> 25) & 0x3f) + 1;
+	}
+	c->kscratch_mask = 0xf;
+}
+
+#ifdef CONFIG_64BIT
+/* For use by uaccess.h */
+u64 __ua_limit;
+EXPORT_SYMBOL(__ua_limit);
+#endif
+
+const char *__cpu_name[NR_CPUS];
+const char *__elf_platform;
+const char *__elf_base_platform;
+
+void cpu_probe(void)
+{
+	struct cpuinfo_mips *c = &current_cpu_data;
+	unsigned int cpu = smp_processor_id();
+
+	/*
+	 * Set a default elf platform, cpu probe may later
+	 * overwrite it with a more precise value
+	 */
+	set_elf_platform(cpu, "mips");
+
+	c->processor_id = PRID_IMP_UNKNOWN;
+	c->fpu_id	= FPIR_IMP_NONE;
+	c->cputype	= CPU_UNKNOWN;
+	c->writecombine = _CACHE_UNCACHED;
+
+	c->fpu_csr31	= FPU_CSR_RN;
+	c->fpu_msk31	= FPU_CSR_RSVD | FPU_CSR_ABS2008 | FPU_CSR_NAN2008;
+
+	c->processor_id = read_c0_prid();
+	switch (c->processor_id & PRID_COMP_MASK) {
+	case PRID_COMP_LEGACY:
+		cpu_probe_legacy(c, cpu);
+		break;
+	case PRID_COMP_MIPS:
+		cpu_probe_mips(c, cpu);
+		break;
+	case PRID_COMP_ALCHEMY:
+		cpu_probe_alchemy(c, cpu);
+		break;
+	case PRID_COMP_SIBYTE:
+		cpu_probe_sibyte(c, cpu);
+		break;
+	case PRID_COMP_BROADCOM:
+		cpu_probe_broadcom(c, cpu);
+		break;
+	case PRID_COMP_SANDCRAFT:
+		cpu_probe_sandcraft(c, cpu);
+		break;
+	case PRID_COMP_NXP:
+		cpu_probe_nxp(c, cpu);
+		break;
+	case PRID_COMP_CAVIUM:
+		cpu_probe_cavium(c, cpu);
+		break;
+	case PRID_COMP_LOONGSON:
+		cpu_probe_loongson(c, cpu);
+		break;
+	case PRID_COMP_INGENIC_13:
+	case PRID_COMP_INGENIC_D0:
+	case PRID_COMP_INGENIC_D1:
+	case PRID_COMP_INGENIC_E1:
+		cpu_probe_ingenic(c, cpu);
+		break;
+	case PRID_COMP_NETLOGIC:
+		cpu_probe_netlogic(c, cpu);
+		break;
+	}
+
+	BUG_ON(!__cpu_name[cpu]);
+	BUG_ON(c->cputype == CPU_UNKNOWN);
+
+	/*
+	 * Platform code can force the cpu type to optimize code
+	 * generation. In that case be sure the cpu type is correctly
+	 * manually setup otherwise it could trigger some nasty bugs.
+	 */
+	BUG_ON(current_cpu_type() != c->cputype);
+
+	if (cpu_has_rixi) {
+		/* Enable the RIXI exceptions */
+		set_c0_pagegrain(PG_IEC);
+		back_to_back_c0_hazard();
+		/* Verify the IEC bit is set */
+		if (read_c0_pagegrain() & PG_IEC)
+			c->options |= MIPS_CPU_RIXIEX;
+	}
+
+	if (mips_fpu_disabled)
+		c->options &= ~MIPS_CPU_FPU;
+
+	if (mips_dsp_disabled)
+		c->ases &= ~(MIPS_ASE_DSP | MIPS_ASE_DSP2P);
+
+	if (mips_htw_disabled) {
+		c->options &= ~MIPS_CPU_HTW;
+		write_c0_pwctl(read_c0_pwctl() &
+			       ~(1 << MIPS_PWCTL_PWEN_SHIFT));
+	}
+
+	if (c->options & MIPS_CPU_FPU)
+		cpu_set_fpu_opts(c);
+	else
+		cpu_set_nofpu_opts(c);
+
+	if (cpu_has_mips_r2_r6) {
+		c->srsets = ((read_c0_srsctl() >> 26) & 0x0f) + 1;
+		/* R2 has Performance Counter Interrupt indicator */
+		c->options |= MIPS_CPU_PCI;
+	}
+	else
+		c->srsets = 1;
+
+	if (cpu_has_mips_r6)
+		elf_hwcap |= HWCAP_MIPS_R6;
+
+	if (cpu_has_msa) {
+		c->msa_id = cpu_get_msa_id();
+		WARN(c->msa_id & MSA_IR_WRPF,
+		     "Vector register partitioning unimplemented!");
+		elf_hwcap |= HWCAP_MIPS_MSA;
+	}
+
+	if (cpu_has_mips16)
+		elf_hwcap |= HWCAP_MIPS_MIPS16;
+
+	if (cpu_has_mdmx)
+		elf_hwcap |= HWCAP_MIPS_MDMX;
+
+	if (cpu_has_mips3d)
+		elf_hwcap |= HWCAP_MIPS_MIPS3D;
+
+	if (cpu_has_smartmips)
+		elf_hwcap |= HWCAP_MIPS_SMARTMIPS;
+
+	if (cpu_has_dsp)
+		elf_hwcap |= HWCAP_MIPS_DSP;
+
+	if (cpu_has_dsp2)
+		elf_hwcap |= HWCAP_MIPS_DSP2;
+
+	if (cpu_has_dsp3)
+		elf_hwcap |= HWCAP_MIPS_DSP3;
+
+	if (cpu_has_mips16e2)
+		elf_hwcap |= HWCAP_MIPS_MIPS16E2;
+
+	if (cpu_has_loongson_mmi)
+		elf_hwcap |= HWCAP_LOONGSON_MMI;
+
+	if (cpu_has_loongson_ext)
+		elf_hwcap |= HWCAP_LOONGSON_EXT;
+
+	if (cpu_has_loongson_ext2)
+		elf_hwcap |= HWCAP_LOONGSON_EXT2;
+
+	if (cpu_has_vz)
+		cpu_probe_vz(c);
+
+	cpu_probe_vmbits(c);
+
+	/* Synthesize CPUCFG data if running on Loongson processors;
+	 * no-op otherwise.
+	 *
+	 * This looks at previously probed features, so keep this at bottom.
+	 */
+	loongson3_cpucfg_synthesize_data(c);
+
+#ifdef CONFIG_64BIT
+	if (cpu == 0)
+		__ua_limit = ~((1ull << cpu_vmbits) - 1);
+#endif
+}
+
+void cpu_report(void)
+{
+	struct cpuinfo_mips *c = &current_cpu_data;
+
+	pr_info("CPU%d revision is: %08x (%s)\n",
+		smp_processor_id(), c->processor_id, cpu_name_string());
+	if (c->options & MIPS_CPU_FPU)
+		printk(KERN_INFO "FPU revision is: %08x\n", c->fpu_id);
+	if (cpu_has_msa)
+		pr_info("MSA revision is: %08x\n", c->msa_id);
+}
+
+void cpu_set_cluster(struct cpuinfo_mips *cpuinfo, unsigned int cluster)
+{
+	/* Ensure the core number fits in the field */
+	WARN_ON(cluster > (MIPS_GLOBALNUMBER_CLUSTER >>
+			   MIPS_GLOBALNUMBER_CLUSTER_SHF));
+
+	cpuinfo->globalnumber &= ~MIPS_GLOBALNUMBER_CLUSTER;
+	cpuinfo->globalnumber |= cluster << MIPS_GLOBALNUMBER_CLUSTER_SHF;
+}
+
+void cpu_set_core(struct cpuinfo_mips *cpuinfo, unsigned int core)
+{
+	/* Ensure the core number fits in the field */
+	WARN_ON(core > (MIPS_GLOBALNUMBER_CORE >> MIPS_GLOBALNUMBER_CORE_SHF));
+
+	cpuinfo->globalnumber &= ~MIPS_GLOBALNUMBER_CORE;
+	cpuinfo->globalnumber |= core << MIPS_GLOBALNUMBER_CORE_SHF;
+}
+
+void cpu_set_vpe_id(struct cpuinfo_mips *cpuinfo, unsigned int vpe)
+{
+	/* Ensure the VP(E) ID fits in the field */
+	WARN_ON(vpe > (MIPS_GLOBALNUMBER_VP >> MIPS_GLOBALNUMBER_VP_SHF));
+
+	/* Ensure we're not using VP(E)s without support */
+	WARN_ON(vpe && !IS_ENABLED(CONFIG_MIPS_MT_SMP) &&
+		!IS_ENABLED(CONFIG_CPU_MIPSR6));
+
+	cpuinfo->globalnumber &= ~MIPS_GLOBALNUMBER_VP;
+	cpuinfo->globalnumber |= vpe << MIPS_GLOBALNUMBER_VP_SHF;
+}
diff --git a/arch/mips/kernel/cpu-r3k-probe.c b/arch/mips/kernel/cpu-r3k-probe.c
new file mode 100644
index 000000000..abdbbe8c5
--- /dev/null
+++ b/arch/mips/kernel/cpu-r3k-probe.c
@@ -0,0 +1,171 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Processor capabilities determination functions.
+ *
+ * Copyright (C) xxxx  the Anonymous
+ * Copyright (C) 1994 - 2006 Ralf Baechle
+ * Copyright (C) 2003, 2004  Maciej W. Rozycki
+ * Copyright (C) 2001, 2004, 2011, 2012	 MIPS Technologies, Inc.
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/ptrace.h>
+#include <linux/smp.h>
+#include <linux/stddef.h>
+#include <linux/export.h>
+
+#include <asm/bugs.h>
+#include <asm/cpu.h>
+#include <asm/cpu-features.h>
+#include <asm/cpu-type.h>
+#include <asm/fpu.h>
+#include <asm/mipsregs.h>
+#include <asm/elf.h>
+
+#include "fpu-probe.h"
+
+/* Hardware capabilities */
+unsigned int elf_hwcap __read_mostly;
+EXPORT_SYMBOL_GPL(elf_hwcap);
+
+void __init check_bugs32(void)
+{
+
+}
+
+/*
+ * Probe whether cpu has config register by trying to play with
+ * alternate cache bit and see whether it matters.
+ * It's used by cpu_probe to distinguish between R3000A and R3081.
+ */
+static inline int cpu_has_confreg(void)
+{
+#ifdef CONFIG_CPU_R3000
+	extern unsigned long r3k_cache_size(unsigned long);
+	unsigned long size1, size2;
+	unsigned long cfg = read_c0_conf();
+
+	size1 = r3k_cache_size(ST0_ISC);
+	write_c0_conf(cfg ^ R30XX_CONF_AC);
+	size2 = r3k_cache_size(ST0_ISC);
+	write_c0_conf(cfg);
+	return size1 != size2;
+#else
+	return 0;
+#endif
+}
+
+static inline void set_elf_platform(int cpu, const char *plat)
+{
+	if (cpu == 0)
+		__elf_platform = plat;
+}
+
+const char *__cpu_name[NR_CPUS];
+const char *__elf_platform;
+const char *__elf_base_platform;
+
+void cpu_probe(void)
+{
+	struct cpuinfo_mips *c = &current_cpu_data;
+	unsigned int cpu = smp_processor_id();
+
+	/*
+	 * Set a default elf platform, cpu probe may later
+	 * overwrite it with a more precise value
+	 */
+	set_elf_platform(cpu, "mips");
+
+	c->processor_id = PRID_IMP_UNKNOWN;
+	c->fpu_id	= FPIR_IMP_NONE;
+	c->cputype	= CPU_UNKNOWN;
+	c->writecombine = _CACHE_UNCACHED;
+
+	c->fpu_csr31	= FPU_CSR_RN;
+	c->fpu_msk31	= FPU_CSR_RSVD | FPU_CSR_ABS2008 | FPU_CSR_NAN2008 |
+			  FPU_CSR_CONDX | FPU_CSR_FS;
+
+	c->srsets = 1;
+
+	c->processor_id = read_c0_prid();
+	switch (c->processor_id & (PRID_COMP_MASK | PRID_IMP_MASK)) {
+	case PRID_COMP_LEGACY | PRID_IMP_R2000:
+		c->cputype = CPU_R2000;
+		__cpu_name[cpu] = "R2000";
+		c->options = MIPS_CPU_TLB | MIPS_CPU_3K_CACHE |
+			     MIPS_CPU_NOFPUEX;
+		if (__cpu_has_fpu())
+			c->options |= MIPS_CPU_FPU;
+		c->tlbsize = 64;
+		break;
+	case PRID_COMP_LEGACY | PRID_IMP_R3000:
+		if ((c->processor_id & PRID_REV_MASK) == PRID_REV_R3000A) {
+			if (cpu_has_confreg()) {
+				c->cputype = CPU_R3081E;
+				__cpu_name[cpu] = "R3081";
+			} else {
+				c->cputype = CPU_R3000A;
+				__cpu_name[cpu] = "R3000A";
+			}
+		} else {
+			c->cputype = CPU_R3000;
+			__cpu_name[cpu] = "R3000";
+		}
+		c->options = MIPS_CPU_TLB | MIPS_CPU_3K_CACHE |
+			     MIPS_CPU_NOFPUEX;
+		if (__cpu_has_fpu())
+			c->options |= MIPS_CPU_FPU;
+		c->tlbsize = 64;
+		break;
+	case PRID_COMP_LEGACY | PRID_IMP_TX39:
+		c->options = MIPS_CPU_TLB | MIPS_CPU_TX39_CACHE;
+
+		if ((c->processor_id & 0xf0) == (PRID_REV_TX3927 & 0xf0)) {
+			c->cputype = CPU_TX3927;
+			__cpu_name[cpu] = "TX3927";
+			c->tlbsize = 64;
+		} else {
+			switch (c->processor_id & PRID_REV_MASK) {
+			case PRID_REV_TX3912:
+				c->cputype = CPU_TX3912;
+				__cpu_name[cpu] = "TX3912";
+				c->tlbsize = 32;
+				break;
+			case PRID_REV_TX3922:
+				c->cputype = CPU_TX3922;
+				__cpu_name[cpu] = "TX3922";
+				c->tlbsize = 64;
+				break;
+			}
+		}
+		break;
+	}
+
+	BUG_ON(!__cpu_name[cpu]);
+	BUG_ON(c->cputype == CPU_UNKNOWN);
+
+	/*
+	 * Platform code can force the cpu type to optimize code
+	 * generation. In that case be sure the cpu type is correctly
+	 * manually setup otherwise it could trigger some nasty bugs.
+	 */
+	BUG_ON(current_cpu_type() != c->cputype);
+
+	if (mips_fpu_disabled)
+		c->options &= ~MIPS_CPU_FPU;
+
+	if (c->options & MIPS_CPU_FPU)
+		cpu_set_fpu_opts(c);
+	else
+		cpu_set_nofpu_opts(c);
+}
+
+void cpu_report(void)
+{
+	struct cpuinfo_mips *c = &current_cpu_data;
+
+	pr_info("CPU%d revision is: %08x (%s)\n",
+		smp_processor_id(), c->processor_id, cpu_name_string());
+	if (c->options & MIPS_CPU_FPU)
+		pr_info("FPU revision is: %08x\n", c->fpu_id);
+}
diff --git a/arch/mips/kernel/crash.c b/arch/mips/kernel/crash.c
new file mode 100644
index 000000000..81845ba04
--- /dev/null
+++ b/arch/mips/kernel/crash.c
@@ -0,0 +1,103 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/kernel.h>
+#include <linux/smp.h>
+#include <linux/reboot.h>
+#include <linux/kexec.h>
+#include <linux/memblock.h>
+#include <linux/crash_dump.h>
+#include <linux/delay.h>
+#include <linux/irq.h>
+#include <linux/types.h>
+#include <linux/sched.h>
+#include <linux/sched/task_stack.h>
+
+/* This keeps a track of which one is crashing cpu. */
+static int crashing_cpu = -1;
+static cpumask_t cpus_in_crash = CPU_MASK_NONE;
+
+#ifdef CONFIG_SMP
+static void crash_shutdown_secondary(void *passed_regs)
+{
+	struct pt_regs *regs = passed_regs;
+	int cpu = smp_processor_id();
+
+	/*
+	 * If we are passed registers, use those.  Otherwise get the
+	 * regs from the last interrupt, which should be correct, as
+	 * we are in an interrupt.  But if the regs are not there,
+	 * pull them from the top of the stack.  They are probably
+	 * wrong, but we need something to keep from crashing again.
+	 */
+	if (!regs)
+		regs = get_irq_regs();
+	if (!regs)
+		regs = task_pt_regs(current);
+
+	if (!cpu_online(cpu))
+		return;
+
+	/* We won't be sent IPIs any more. */
+	set_cpu_online(cpu, false);
+
+	local_irq_disable();
+	if (!cpumask_test_cpu(cpu, &cpus_in_crash))
+		crash_save_cpu(regs, cpu);
+	cpumask_set_cpu(cpu, &cpus_in_crash);
+
+	while (!atomic_read(&kexec_ready_to_reboot))
+		cpu_relax();
+
+	kexec_reboot();
+
+	/* NOTREACHED */
+}
+
+static void crash_kexec_prepare_cpus(void)
+{
+	static int cpus_stopped;
+	unsigned int msecs;
+	unsigned int ncpus;
+
+	if (cpus_stopped)
+		return;
+
+	ncpus = num_online_cpus() - 1;/* Excluding the panic cpu */
+
+	smp_call_function(crash_shutdown_secondary, NULL, 0);
+	smp_wmb();
+
+	/*
+	 * The crash CPU sends an IPI and wait for other CPUs to
+	 * respond. Delay of at least 10 seconds.
+	 */
+	pr_emerg("Sending IPI to other cpus...\n");
+	msecs = 10000;
+	while ((cpumask_weight(&cpus_in_crash) < ncpus) && (--msecs > 0)) {
+		cpu_relax();
+		mdelay(1);
+	}
+
+	cpus_stopped = 1;
+}
+
+/* Override the weak function in kernel/panic.c */
+void crash_smp_send_stop(void)
+{
+	if (_crash_smp_send_stop)
+		_crash_smp_send_stop();
+
+	crash_kexec_prepare_cpus();
+}
+
+#else /* !defined(CONFIG_SMP)  */
+static void crash_kexec_prepare_cpus(void) {}
+#endif /* !defined(CONFIG_SMP)	*/
+
+void default_machine_crash_shutdown(struct pt_regs *regs)
+{
+	local_irq_disable();
+	crashing_cpu = smp_processor_id();
+	crash_save_cpu(regs, crashing_cpu);
+	crash_kexec_prepare_cpus();
+	cpumask_set_cpu(crashing_cpu, &cpus_in_crash);
+}
diff --git a/arch/mips/kernel/crash_dump.c b/arch/mips/kernel/crash_dump.c
new file mode 100644
index 000000000..01b2bd95b
--- /dev/null
+++ b/arch/mips/kernel/crash_dump.c
@@ -0,0 +1,67 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/highmem.h>
+#include <linux/memblock.h>
+#include <linux/crash_dump.h>
+#include <linux/uaccess.h>
+#include <linux/slab.h>
+
+static void *kdump_buf_page;
+
+/**
+ * copy_oldmem_page - copy one page from "oldmem"
+ * @pfn: page frame number to be copied
+ * @buf: target memory address for the copy; this can be in kernel address
+ *	space or user address space (see @userbuf)
+ * @csize: number of bytes to copy
+ * @offset: offset in bytes into the page (based on pfn) to begin the copy
+ * @userbuf: if set, @buf is in user address space, use copy_to_user(),
+ *	otherwise @buf is in kernel address space, use memcpy().
+ *
+ * Copy a page from "oldmem". For this page, there is no pte mapped
+ * in the current kernel.
+ *
+ * Calling copy_to_user() in atomic context is not desirable. Hence first
+ * copying the data to a pre-allocated kernel page and then copying to user
+ * space in non-atomic context.
+ */
+ssize_t copy_oldmem_page(unsigned long pfn, char *buf,
+			 size_t csize, unsigned long offset, int userbuf)
+{
+	void  *vaddr;
+
+	if (!csize)
+		return 0;
+
+	vaddr = kmap_atomic_pfn(pfn);
+
+	if (!userbuf) {
+		memcpy(buf, (vaddr + offset), csize);
+		kunmap_atomic(vaddr);
+	} else {
+		if (!kdump_buf_page) {
+			pr_warn("Kdump: Kdump buffer page not allocated\n");
+
+			return -EFAULT;
+		}
+		copy_page(kdump_buf_page, vaddr);
+		kunmap_atomic(vaddr);
+		if (copy_to_user(buf, (kdump_buf_page + offset), csize))
+			return -EFAULT;
+	}
+
+	return csize;
+}
+
+static int __init kdump_buf_page_init(void)
+{
+	int ret = 0;
+
+	kdump_buf_page = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!kdump_buf_page) {
+		pr_warn("Kdump: Failed to allocate kdump buffer page\n");
+		ret = -ENOMEM;
+	}
+
+	return ret;
+}
+arch_initcall(kdump_buf_page_init);
diff --git a/arch/mips/kernel/csrc-bcm1480.c b/arch/mips/kernel/csrc-bcm1480.c
new file mode 100644
index 000000000..6c18a138f
--- /dev/null
+++ b/arch/mips/kernel/csrc-bcm1480.c
@@ -0,0 +1,48 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2000,2001,2004 Broadcom Corporation
+ */
+#include <linux/clocksource.h>
+#include <linux/sched_clock.h>
+
+#include <asm/addrspace.h>
+#include <asm/io.h>
+#include <asm/time.h>
+
+#include <asm/sibyte/bcm1480_regs.h>
+#include <asm/sibyte/sb1250_regs.h>
+#include <asm/sibyte/bcm1480_int.h>
+#include <asm/sibyte/bcm1480_scd.h>
+
+#include <asm/sibyte/sb1250.h>
+
+static u64 bcm1480_hpt_read(struct clocksource *cs)
+{
+	return (u64) __raw_readq(IOADDR(A_SCD_ZBBUS_CYCLE_COUNT));
+}
+
+struct clocksource bcm1480_clocksource = {
+	.name	= "zbbus-cycles",
+	.rating = 200,
+	.read	= bcm1480_hpt_read,
+	.mask	= CLOCKSOURCE_MASK(64),
+	.flags	= CLOCK_SOURCE_IS_CONTINUOUS,
+};
+
+static u64 notrace sb1480_read_sched_clock(void)
+{
+	return __raw_readq(IOADDR(A_SCD_ZBBUS_CYCLE_COUNT));
+}
+
+void __init sb1480_clocksource_init(void)
+{
+	struct clocksource *cs = &bcm1480_clocksource;
+	unsigned int plldiv;
+	unsigned long zbbus;
+
+	plldiv = G_BCM1480_SYS_PLL_DIV(__raw_readq(IOADDR(A_SCD_SYSTEM_CFG)));
+	zbbus = ((plldiv >> 1) * 50000000) + ((plldiv & 1) * 25000000);
+	clocksource_register_hz(cs, zbbus);
+
+	sched_clock_register(sb1480_read_sched_clock, 64, zbbus);
+}
diff --git a/arch/mips/kernel/csrc-ioasic.c b/arch/mips/kernel/csrc-ioasic.c
new file mode 100644
index 000000000..bad740ad3
--- /dev/null
+++ b/arch/mips/kernel/csrc-ioasic.c
@@ -0,0 +1,65 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ *  DEC I/O ASIC's counter clocksource
+ *
+ *  Copyright (C) 2008	Yoichi Yuasa <yuasa@linux-mips.org>
+ */
+#include <linux/clocksource.h>
+#include <linux/sched_clock.h>
+#include <linux/init.h>
+
+#include <asm/ds1287.h>
+#include <asm/time.h>
+#include <asm/dec/ioasic.h>
+#include <asm/dec/ioasic_addrs.h>
+
+static u64 dec_ioasic_hpt_read(struct clocksource *cs)
+{
+	return ioasic_read(IO_REG_FCTR);
+}
+
+static struct clocksource clocksource_dec = {
+	.name		= "dec-ioasic",
+	.read		= dec_ioasic_hpt_read,
+	.mask		= CLOCKSOURCE_MASK(32),
+	.flags		= CLOCK_SOURCE_IS_CONTINUOUS,
+};
+
+static u64 notrace dec_ioasic_read_sched_clock(void)
+{
+	return ioasic_read(IO_REG_FCTR);
+}
+
+int __init dec_ioasic_clocksource_init(void)
+{
+	unsigned int freq;
+	u32 start, end;
+	int i = HZ / 8;
+
+	ds1287_timer_state();
+	while (!ds1287_timer_state())
+		;
+
+	start = dec_ioasic_hpt_read(&clocksource_dec);
+
+	while (i--)
+		while (!ds1287_timer_state())
+			;
+
+	end = dec_ioasic_hpt_read(&clocksource_dec);
+
+	freq = (end - start) * 8;
+
+	/* An early revision of the I/O ASIC didn't have the counter.  */
+	if (!freq)
+		return -ENXIO;
+
+	printk(KERN_INFO "I/O ASIC clock frequency %dHz\n", freq);
+
+	clocksource_dec.rating = 200 + freq / 10000000;
+	clocksource_register_hz(&clocksource_dec, freq);
+
+	sched_clock_register(dec_ioasic_read_sched_clock, 32, freq);
+
+	return 0;
+}
diff --git a/arch/mips/kernel/csrc-r4k.c b/arch/mips/kernel/csrc-r4k.c
new file mode 100644
index 000000000..edc4afc08
--- /dev/null
+++ b/arch/mips/kernel/csrc-r4k.c
@@ -0,0 +1,130 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2007 by Ralf Baechle
+ */
+#include <linux/clocksource.h>
+#include <linux/cpufreq.h>
+#include <linux/init.h>
+#include <linux/sched_clock.h>
+
+#include <asm/time.h>
+
+static u64 c0_hpt_read(struct clocksource *cs)
+{
+	return read_c0_count();
+}
+
+static struct clocksource clocksource_mips = {
+	.name		= "MIPS",
+	.read		= c0_hpt_read,
+	.mask		= CLOCKSOURCE_MASK(32),
+	.flags		= CLOCK_SOURCE_IS_CONTINUOUS,
+};
+
+static u64 __maybe_unused notrace r4k_read_sched_clock(void)
+{
+	return read_c0_count();
+}
+
+static inline unsigned int rdhwr_count(void)
+{
+	unsigned int count;
+
+	__asm__ __volatile__(
+	"	.set push\n"
+	"	.set mips32r2\n"
+	"	rdhwr	%0, $2\n"
+	"	.set pop\n"
+	: "=r" (count));
+
+	return count;
+}
+
+static bool rdhwr_count_usable(void)
+{
+	unsigned int prev, curr, i;
+
+	/*
+	 * Older QEMUs have a broken implementation of RDHWR for the CP0 count
+	 * which always returns a constant value. Try to identify this and don't
+	 * use it in the VDSO if it is broken. This workaround can be removed
+	 * once the fix has been in QEMU stable for a reasonable amount of time.
+	 */
+	for (i = 0, prev = rdhwr_count(); i < 100; i++) {
+		curr = rdhwr_count();
+
+		if (curr != prev)
+			return true;
+
+		prev = curr;
+	}
+
+	pr_warn("Not using R4K clocksource in VDSO due to broken RDHWR\n");
+	return false;
+}
+
+#ifdef CONFIG_CPU_FREQ
+
+static bool __read_mostly r4k_clock_unstable;
+
+static void r4k_clocksource_unstable(char *reason)
+{
+	if (r4k_clock_unstable)
+		return;
+
+	r4k_clock_unstable = true;
+
+	pr_info("R4K timer is unstable due to %s\n", reason);
+
+	clocksource_mark_unstable(&clocksource_mips);
+}
+
+static int r4k_cpufreq_callback(struct notifier_block *nb,
+				unsigned long val, void *data)
+{
+	if (val == CPUFREQ_POSTCHANGE)
+		r4k_clocksource_unstable("CPU frequency change");
+
+	return 0;
+}
+
+static struct notifier_block r4k_cpufreq_notifier = {
+	.notifier_call  = r4k_cpufreq_callback,
+};
+
+static int __init r4k_register_cpufreq_notifier(void)
+{
+	return cpufreq_register_notifier(&r4k_cpufreq_notifier,
+					 CPUFREQ_TRANSITION_NOTIFIER);
+
+}
+core_initcall(r4k_register_cpufreq_notifier);
+
+#endif /* !CONFIG_CPU_FREQ */
+
+int __init init_r4k_clocksource(void)
+{
+	if (!cpu_has_counter || !mips_hpt_frequency)
+		return -ENXIO;
+
+	/* Calculate a somewhat reasonable rating value */
+	clocksource_mips.rating = 200 + mips_hpt_frequency / 10000000;
+
+	/*
+	 * R2 onwards makes the count accessible to user mode so it can be used
+	 * by the VDSO (HWREna is configured by configure_hwrena()).
+	 */
+	if (cpu_has_mips_r2_r6 && rdhwr_count_usable())
+		clocksource_mips.vdso_clock_mode = VDSO_CLOCKMODE_R4K;
+
+	clocksource_register_hz(&clocksource_mips, mips_hpt_frequency);
+
+#ifndef CONFIG_CPU_FREQ
+	sched_clock_register(r4k_read_sched_clock, 32, mips_hpt_frequency);
+#endif
+
+	return 0;
+}
diff --git a/arch/mips/kernel/csrc-sb1250.c b/arch/mips/kernel/csrc-sb1250.c
new file mode 100644
index 000000000..fa2fa3e10
--- /dev/null
+++ b/arch/mips/kernel/csrc-sb1250.c
@@ -0,0 +1,71 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2000, 2001 Broadcom Corporation
+ */
+#include <linux/clocksource.h>
+#include <linux/sched_clock.h>
+
+#include <asm/addrspace.h>
+#include <asm/io.h>
+#include <asm/time.h>
+
+#include <asm/sibyte/sb1250.h>
+#include <asm/sibyte/sb1250_regs.h>
+#include <asm/sibyte/sb1250_int.h>
+#include <asm/sibyte/sb1250_scd.h>
+
+#define SB1250_HPT_NUM		3
+#define SB1250_HPT_VALUE	M_SCD_TIMER_CNT /* max value */
+
+/*
+ * The HPT is free running from SB1250_HPT_VALUE down to 0 then starts over
+ * again.
+ */
+static inline u64 sb1250_hpt_get_cycles(void)
+{
+	unsigned int count;
+	void __iomem *addr;
+
+	addr = IOADDR(A_SCD_TIMER_REGISTER(SB1250_HPT_NUM, R_SCD_TIMER_CNT));
+	count = G_SCD_TIMER_CNT(__raw_readq(addr));
+
+	return SB1250_HPT_VALUE - count;
+}
+
+static u64 sb1250_hpt_read(struct clocksource *cs)
+{
+	return sb1250_hpt_get_cycles();
+}
+
+struct clocksource bcm1250_clocksource = {
+	.name	= "bcm1250-counter-3",
+	.rating = 200,
+	.read	= sb1250_hpt_read,
+	.mask	= CLOCKSOURCE_MASK(23),
+	.flags	= CLOCK_SOURCE_IS_CONTINUOUS,
+};
+
+static u64 notrace sb1250_read_sched_clock(void)
+{
+	return sb1250_hpt_get_cycles();
+}
+
+void __init sb1250_clocksource_init(void)
+{
+	struct clocksource *cs = &bcm1250_clocksource;
+
+	/* Setup hpt using timer #3 but do not enable irq for it */
+	__raw_writeq(0,
+		     IOADDR(A_SCD_TIMER_REGISTER(SB1250_HPT_NUM,
+						 R_SCD_TIMER_CFG)));
+	__raw_writeq(SB1250_HPT_VALUE,
+		     IOADDR(A_SCD_TIMER_REGISTER(SB1250_HPT_NUM,
+						 R_SCD_TIMER_INIT)));
+	__raw_writeq(M_SCD_TIMER_ENABLE | M_SCD_TIMER_MODE_CONTINUOUS,
+		     IOADDR(A_SCD_TIMER_REGISTER(SB1250_HPT_NUM,
+						 R_SCD_TIMER_CFG)));
+
+	clocksource_register_hz(cs, V_SCD_TIMER_FREQ);
+
+	sched_clock_register(sb1250_read_sched_clock, 23, V_SCD_TIMER_FREQ);
+}
diff --git a/arch/mips/kernel/early_printk.c b/arch/mips/kernel/early_printk.c
new file mode 100644
index 000000000..4a1647ddf
--- /dev/null
+++ b/arch/mips/kernel/early_printk.c
@@ -0,0 +1,41 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2002, 2003, 06, 07 Ralf Baechle (ralf@linux-mips.org)
+ * Copyright (C) 2007 MIPS Technologies, Inc.
+ *   written by Ralf Baechle (ralf@linux-mips.org)
+ */
+#include <linux/kernel.h>
+#include <linux/console.h>
+#include <linux/printk.h>
+#include <linux/init.h>
+
+#include <asm/setup.h>
+
+static void early_console_write(struct console *con, const char *s, unsigned n)
+{
+	while (n-- && *s) {
+		if (*s == '\n')
+			prom_putchar('\r');
+		prom_putchar(*s);
+		s++;
+	}
+}
+
+static struct console early_console_prom = {
+	.name	= "early",
+	.write	= early_console_write,
+	.flags	= CON_PRINTBUFFER | CON_BOOT,
+	.index	= -1
+};
+
+void __init setup_early_printk(void)
+{
+	if (early_console)
+		return;
+	early_console = &early_console_prom;
+
+	register_console(&early_console_prom);
+}
diff --git a/arch/mips/kernel/early_printk_8250.c b/arch/mips/kernel/early_printk_8250.c
new file mode 100644
index 000000000..567c6ec0c
--- /dev/null
+++ b/arch/mips/kernel/early_printk_8250.c
@@ -0,0 +1,54 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ *  8250/16550-type serial ports prom_putchar()
+ *
+ *  Copyright (C) 2010  Yoichi Yuasa <yuasa@linux-mips.org>
+ */
+#include <linux/io.h>
+#include <linux/serial_core.h>
+#include <linux/serial_reg.h>
+#include <asm/setup.h>
+
+static void __iomem *serial8250_base;
+static unsigned int serial8250_reg_shift;
+static unsigned int serial8250_tx_timeout;
+
+void setup_8250_early_printk_port(unsigned long base, unsigned int reg_shift,
+				  unsigned int timeout)
+{
+	serial8250_base = (void __iomem *)base;
+	serial8250_reg_shift = reg_shift;
+	serial8250_tx_timeout = timeout;
+}
+
+static inline u8 serial_in(int offset)
+{
+	return readb(serial8250_base + (offset << serial8250_reg_shift));
+}
+
+static inline void serial_out(int offset, char value)
+{
+	writeb(value, serial8250_base + (offset << serial8250_reg_shift));
+}
+
+void prom_putchar(char c)
+{
+	unsigned int timeout;
+	int status, bits;
+
+	if (!serial8250_base)
+		return;
+
+	timeout = serial8250_tx_timeout;
+	bits = UART_LSR_TEMT | UART_LSR_THRE;
+
+	do {
+		status = serial_in(UART_LSR);
+
+		if (--timeout == 0)
+			break;
+	} while ((status & bits) != bits);
+
+	if (timeout)
+		serial_out(UART_TX, c);
+}
diff --git a/arch/mips/kernel/elf.c b/arch/mips/kernel/elf.c
new file mode 100644
index 000000000..7b045d2a0
--- /dev/null
+++ b/arch/mips/kernel/elf.c
@@ -0,0 +1,343 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2014 Imagination Technologies
+ * Author: Paul Burton <paul.burton@mips.com>
+ */
+
+#include <linux/binfmts.h>
+#include <linux/elf.h>
+#include <linux/export.h>
+#include <linux/sched.h>
+
+#include <asm/cpu-features.h>
+#include <asm/cpu-info.h>
+
+#ifdef CONFIG_MIPS_FP_SUPPORT
+
+/* Whether to accept legacy-NaN and 2008-NaN user binaries.  */
+bool mips_use_nan_legacy;
+bool mips_use_nan_2008;
+
+/* FPU modes */
+enum {
+	FP_FRE,
+	FP_FR0,
+	FP_FR1,
+};
+
+/**
+ * struct mode_req - ABI FPU mode requirements
+ * @single:	The program being loaded needs an FPU but it will only issue
+ *		single precision instructions meaning that it can execute in
+ *		either FR0 or FR1.
+ * @soft:	The soft(-float) requirement means that the program being
+ *		loaded needs has no FPU dependency at all (i.e. it has no
+ *		FPU instructions).
+ * @fr1:	The program being loaded depends on FPU being in FR=1 mode.
+ * @frdefault:	The program being loaded depends on the default FPU mode.
+ *		That is FR0 for O32 and FR1 for N32/N64.
+ * @fre:	The program being loaded depends on FPU with FRE=1. This mode is
+ *		a bridge which uses FR=1 whilst still being able to maintain
+ *		full compatibility with pre-existing code using the O32 FP32
+ *		ABI.
+ *
+ * More information about the FP ABIs can be found here:
+ *
+ * https://dmz-portal.mips.com/wiki/MIPS_O32_ABI_-_FR0_and_FR1_Interlinking#10.4.1._Basic_mode_set-up
+ *
+ */
+
+struct mode_req {
+	bool single;
+	bool soft;
+	bool fr1;
+	bool frdefault;
+	bool fre;
+};
+
+static const struct mode_req fpu_reqs[] = {
+	[MIPS_ABI_FP_ANY]    = { true,  true,  true,  true,  true  },
+	[MIPS_ABI_FP_DOUBLE] = { false, false, false, true,  true  },
+	[MIPS_ABI_FP_SINGLE] = { true,  false, false, false, false },
+	[MIPS_ABI_FP_SOFT]   = { false, true,  false, false, false },
+	[MIPS_ABI_FP_OLD_64] = { false, false, false, false, false },
+	[MIPS_ABI_FP_XX]     = { false, false, true,  true,  true  },
+	[MIPS_ABI_FP_64]     = { false, false, true,  false, false },
+	[MIPS_ABI_FP_64A]    = { false, false, true,  false, true  }
+};
+
+/*
+ * Mode requirements when .MIPS.abiflags is not present in the ELF.
+ * Not present means that everything is acceptable except FR1.
+ */
+static struct mode_req none_req = { true, true, false, true, true };
+
+int arch_elf_pt_proc(void *_ehdr, void *_phdr, struct file *elf,
+		     bool is_interp, struct arch_elf_state *state)
+{
+	union {
+		struct elf32_hdr e32;
+		struct elf64_hdr e64;
+	} *ehdr = _ehdr;
+	struct elf32_phdr *phdr32 = _phdr;
+	struct elf64_phdr *phdr64 = _phdr;
+	struct mips_elf_abiflags_v0 abiflags;
+	bool elf32;
+	u32 flags;
+	int ret;
+	loff_t pos;
+
+	elf32 = ehdr->e32.e_ident[EI_CLASS] == ELFCLASS32;
+	flags = elf32 ? ehdr->e32.e_flags : ehdr->e64.e_flags;
+
+	/* Let's see if this is an O32 ELF */
+	if (elf32) {
+		if (flags & EF_MIPS_FP64) {
+			/*
+			 * Set MIPS_ABI_FP_OLD_64 for EF_MIPS_FP64. We will override it
+			 * later if needed
+			 */
+			if (is_interp)
+				state->interp_fp_abi = MIPS_ABI_FP_OLD_64;
+			else
+				state->fp_abi = MIPS_ABI_FP_OLD_64;
+		}
+		if (phdr32->p_type != PT_MIPS_ABIFLAGS)
+			return 0;
+
+		if (phdr32->p_filesz < sizeof(abiflags))
+			return -EINVAL;
+		pos = phdr32->p_offset;
+	} else {
+		if (phdr64->p_type != PT_MIPS_ABIFLAGS)
+			return 0;
+		if (phdr64->p_filesz < sizeof(abiflags))
+			return -EINVAL;
+		pos = phdr64->p_offset;
+	}
+
+	ret = kernel_read(elf, &abiflags, sizeof(abiflags), &pos);
+	if (ret < 0)
+		return ret;
+	if (ret != sizeof(abiflags))
+		return -EIO;
+
+	/* Record the required FP ABIs for use by mips_check_elf */
+	if (is_interp)
+		state->interp_fp_abi = abiflags.fp_abi;
+	else
+		state->fp_abi = abiflags.fp_abi;
+
+	return 0;
+}
+
+int arch_check_elf(void *_ehdr, bool has_interpreter, void *_interp_ehdr,
+		   struct arch_elf_state *state)
+{
+	union {
+		struct elf32_hdr e32;
+		struct elf64_hdr e64;
+	} *ehdr = _ehdr;
+	union {
+		struct elf32_hdr e32;
+		struct elf64_hdr e64;
+	} *iehdr = _interp_ehdr;
+	struct mode_req prog_req, interp_req;
+	int fp_abi, interp_fp_abi, abi0, abi1, max_abi;
+	bool elf32;
+	u32 flags;
+
+	elf32 = ehdr->e32.e_ident[EI_CLASS] == ELFCLASS32;
+	flags = elf32 ? ehdr->e32.e_flags : ehdr->e64.e_flags;
+
+	/*
+	 * Determine the NaN personality, reject the binary if not allowed.
+	 * Also ensure that any interpreter matches the executable.
+	 */
+	if (flags & EF_MIPS_NAN2008) {
+		if (mips_use_nan_2008)
+			state->nan_2008 = 1;
+		else
+			return -ENOEXEC;
+	} else {
+		if (mips_use_nan_legacy)
+			state->nan_2008 = 0;
+		else
+			return -ENOEXEC;
+	}
+	if (has_interpreter) {
+		bool ielf32;
+		u32 iflags;
+
+		ielf32 = iehdr->e32.e_ident[EI_CLASS] == ELFCLASS32;
+		iflags = ielf32 ? iehdr->e32.e_flags : iehdr->e64.e_flags;
+
+		if ((flags ^ iflags) & EF_MIPS_NAN2008)
+			return -ELIBBAD;
+	}
+
+	if (!IS_ENABLED(CONFIG_MIPS_O32_FP64_SUPPORT))
+		return 0;
+
+	fp_abi = state->fp_abi;
+
+	if (has_interpreter) {
+		interp_fp_abi = state->interp_fp_abi;
+
+		abi0 = min(fp_abi, interp_fp_abi);
+		abi1 = max(fp_abi, interp_fp_abi);
+	} else {
+		abi0 = abi1 = fp_abi;
+	}
+
+	if (elf32 && !(flags & EF_MIPS_ABI2)) {
+		/* Default to a mode capable of running code expecting FR=0 */
+		state->overall_fp_mode = cpu_has_mips_r6 ? FP_FRE : FP_FR0;
+
+		/* Allow all ABIs we know about */
+		max_abi = MIPS_ABI_FP_64A;
+	} else {
+		/* MIPS64 code always uses FR=1, thus the default is easy */
+		state->overall_fp_mode = FP_FR1;
+
+		/* Disallow access to the various FPXX & FP64 ABIs */
+		max_abi = MIPS_ABI_FP_SOFT;
+	}
+
+	if ((abi0 > max_abi && abi0 != MIPS_ABI_FP_UNKNOWN) ||
+	    (abi1 > max_abi && abi1 != MIPS_ABI_FP_UNKNOWN))
+		return -ELIBBAD;
+
+	/* It's time to determine the FPU mode requirements */
+	prog_req = (abi0 == MIPS_ABI_FP_UNKNOWN) ? none_req : fpu_reqs[abi0];
+	interp_req = (abi1 == MIPS_ABI_FP_UNKNOWN) ? none_req : fpu_reqs[abi1];
+
+	/*
+	 * Check whether the program's and interp's ABIs have a matching FPU
+	 * mode requirement.
+	 */
+	prog_req.single = interp_req.single && prog_req.single;
+	prog_req.soft = interp_req.soft && prog_req.soft;
+	prog_req.fr1 = interp_req.fr1 && prog_req.fr1;
+	prog_req.frdefault = interp_req.frdefault && prog_req.frdefault;
+	prog_req.fre = interp_req.fre && prog_req.fre;
+
+	/*
+	 * Determine the desired FPU mode
+	 *
+	 * Decision making:
+	 *
+	 * - We want FR_FRE if FRE=1 and both FR=1 and FR=0 are false. This
+	 *   means that we have a combination of program and interpreter
+	 *   that inherently require the hybrid FP mode.
+	 * - If FR1 and FRDEFAULT is true, that means we hit the any-abi or
+	 *   fpxx case. This is because, in any-ABI (or no-ABI) we have no FPU
+	 *   instructions so we don't care about the mode. We will simply use
+	 *   the one preferred by the hardware. In fpxx case, that ABI can
+	 *   handle both FR=1 and FR=0, so, again, we simply choose the one
+	 *   preferred by the hardware. Next, if we only use single-precision
+	 *   FPU instructions, and the default ABI FPU mode is not good
+	 *   (ie single + any ABI combination), we set again the FPU mode to the
+	 *   one is preferred by the hardware. Next, if we know that the code
+	 *   will only use single-precision instructions, shown by single being
+	 *   true but frdefault being false, then we again set the FPU mode to
+	 *   the one that is preferred by the hardware.
+	 * - We want FP_FR1 if that's the only matching mode and the default one
+	 *   is not good.
+	 * - Return with -ELIBADD if we can't find a matching FPU mode.
+	 */
+	if (prog_req.fre && !prog_req.frdefault && !prog_req.fr1)
+		state->overall_fp_mode = FP_FRE;
+	else if ((prog_req.fr1 && prog_req.frdefault) ||
+		 (prog_req.single && !prog_req.frdefault))
+		/* Make sure 64-bit MIPS III/IV/64R1 will not pick FR1 */
+		state->overall_fp_mode = ((raw_current_cpu_data.fpu_id & MIPS_FPIR_F64) &&
+					  cpu_has_mips_r2_r6) ?
+					  FP_FR1 : FP_FR0;
+	else if (prog_req.fr1)
+		state->overall_fp_mode = FP_FR1;
+	else  if (!prog_req.fre && !prog_req.frdefault &&
+		  !prog_req.fr1 && !prog_req.single && !prog_req.soft)
+		return -ELIBBAD;
+
+	return 0;
+}
+
+static inline void set_thread_fp_mode(int hybrid, int regs32)
+{
+	if (hybrid)
+		set_thread_flag(TIF_HYBRID_FPREGS);
+	else
+		clear_thread_flag(TIF_HYBRID_FPREGS);
+	if (regs32)
+		set_thread_flag(TIF_32BIT_FPREGS);
+	else
+		clear_thread_flag(TIF_32BIT_FPREGS);
+}
+
+void mips_set_personality_fp(struct arch_elf_state *state)
+{
+	/*
+	 * This function is only ever called for O32 ELFs so we should
+	 * not be worried about N32/N64 binaries.
+	 */
+
+	if (!IS_ENABLED(CONFIG_MIPS_O32_FP64_SUPPORT))
+		return;
+
+	switch (state->overall_fp_mode) {
+	case FP_FRE:
+		set_thread_fp_mode(1, 0);
+		break;
+	case FP_FR0:
+		set_thread_fp_mode(0, 1);
+		break;
+	case FP_FR1:
+		set_thread_fp_mode(0, 0);
+		break;
+	default:
+		BUG();
+	}
+}
+
+/*
+ * Select the IEEE 754 NaN encoding and ABS.fmt/NEG.fmt execution mode
+ * in FCSR according to the ELF NaN personality.
+ */
+void mips_set_personality_nan(struct arch_elf_state *state)
+{
+	struct cpuinfo_mips *c = &boot_cpu_data;
+	struct task_struct *t = current;
+
+	t->thread.fpu.fcr31 = c->fpu_csr31;
+	switch (state->nan_2008) {
+	case 0:
+		break;
+	case 1:
+		if (!(c->fpu_msk31 & FPU_CSR_NAN2008))
+			t->thread.fpu.fcr31 |= FPU_CSR_NAN2008;
+		if (!(c->fpu_msk31 & FPU_CSR_ABS2008))
+			t->thread.fpu.fcr31 |= FPU_CSR_ABS2008;
+		break;
+	default:
+		BUG();
+	}
+}
+
+#endif /* CONFIG_MIPS_FP_SUPPORT */
+
+int mips_elf_read_implies_exec(void *elf_ex, int exstack)
+{
+	if (exstack != EXSTACK_DISABLE_X) {
+		/* The binary doesn't request a non-executable stack */
+		return 1;
+	}
+
+	if (!cpu_has_rixi) {
+		/* The CPU doesn't support non-executable memory */
+		return 1;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(mips_elf_read_implies_exec);
diff --git a/arch/mips/kernel/entry.S b/arch/mips/kernel/entry.S
new file mode 100644
index 000000000..4b896f502
--- /dev/null
+++ b/arch/mips/kernel/entry.S
@@ -0,0 +1,186 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1994 - 2000, 2001, 2003 Ralf Baechle
+ * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ * Copyright (C) 2001 MIPS Technologies, Inc.
+ */
+
+#include <asm/asm.h>
+#include <asm/asmmacro.h>
+#include <asm/compiler.h>
+#include <asm/irqflags.h>
+#include <asm/regdef.h>
+#include <asm/mipsregs.h>
+#include <asm/stackframe.h>
+#include <asm/isadep.h>
+#include <asm/thread_info.h>
+#include <asm/war.h>
+
+#ifndef CONFIG_PREEMPTION
+#define resume_kernel	restore_all
+#else
+#define __ret_from_irq	ret_from_exception
+#endif
+
+	.text
+	.align	5
+#ifndef CONFIG_PREEMPTION
+FEXPORT(ret_from_exception)
+	local_irq_disable			# preempt stop
+	b	__ret_from_irq
+#endif
+FEXPORT(ret_from_irq)
+	LONG_S	s0, TI_REGS($28)
+FEXPORT(__ret_from_irq)
+/*
+ * We can be coming here from a syscall done in the kernel space,
+ * e.g. a failed kernel_execve().
+ */
+resume_userspace_check:
+	LONG_L	t0, PT_STATUS(sp)		# returning to kernel mode?
+	andi	t0, t0, KU_USER
+	beqz	t0, resume_kernel
+
+resume_userspace:
+	local_irq_disable		# make sure we dont miss an
+					# interrupt setting need_resched
+					# between sampling and return
+	LONG_L	a2, TI_FLAGS($28)	# current->work
+	andi	t0, a2, _TIF_WORK_MASK	# (ignoring syscall_trace)
+	bnez	t0, work_pending
+	j	restore_all
+
+#ifdef CONFIG_PREEMPTION
+resume_kernel:
+	local_irq_disable
+	lw	t0, TI_PRE_COUNT($28)
+	bnez	t0, restore_all
+	LONG_L	t0, TI_FLAGS($28)
+	andi	t1, t0, _TIF_NEED_RESCHED
+	beqz	t1, restore_all
+	LONG_L	t0, PT_STATUS(sp)		# Interrupts off?
+	andi	t0, 1
+	beqz	t0, restore_all
+	PTR_LA	ra, restore_all
+	j	preempt_schedule_irq
+#endif
+
+FEXPORT(ret_from_kernel_thread)
+	jal	schedule_tail		# a0 = struct task_struct *prev
+	move	a0, s1
+	jal	s0
+	j	syscall_exit
+
+FEXPORT(ret_from_fork)
+	jal	schedule_tail		# a0 = struct task_struct *prev
+
+FEXPORT(syscall_exit)
+#ifdef CONFIG_DEBUG_RSEQ
+	move	a0, sp
+	jal	rseq_syscall
+#endif
+	local_irq_disable		# make sure need_resched and
+					# signals dont change between
+					# sampling and return
+	LONG_L	a2, TI_FLAGS($28)	# current->work
+	li	t0, _TIF_ALLWORK_MASK
+	and	t0, a2, t0
+	bnez	t0, syscall_exit_work
+
+restore_all:				# restore full frame
+	.set	noat
+	RESTORE_TEMP
+	RESTORE_AT
+	RESTORE_STATIC
+restore_partial:		# restore partial frame
+#ifdef CONFIG_TRACE_IRQFLAGS
+	SAVE_STATIC
+	SAVE_AT
+	SAVE_TEMP
+	LONG_L	v0, PT_STATUS(sp)
+#if defined(CONFIG_CPU_R3000) || defined(CONFIG_CPU_TX39XX)
+	and	v0, ST0_IEP
+#else
+	and	v0, ST0_IE
+#endif
+	beqz	v0, 1f
+	jal	trace_hardirqs_on
+	b	2f
+1:	jal	trace_hardirqs_off
+2:
+	RESTORE_TEMP
+	RESTORE_AT
+	RESTORE_STATIC
+#endif
+	RESTORE_SOME
+	RESTORE_SP_AND_RET
+	.set	at
+
+work_pending:
+	andi	t0, a2, _TIF_NEED_RESCHED # a2 is preloaded with TI_FLAGS
+	beqz	t0, work_notifysig
+work_resched:
+	TRACE_IRQS_OFF
+	jal	schedule
+
+	local_irq_disable		# make sure need_resched and
+					# signals dont change between
+					# sampling and return
+	LONG_L	a2, TI_FLAGS($28)
+	andi	t0, a2, _TIF_WORK_MASK	# is there any work to be done
+					# other than syscall tracing?
+	beqz	t0, restore_all
+	andi	t0, a2, _TIF_NEED_RESCHED
+	bnez	t0, work_resched
+
+work_notifysig:				# deal with pending signals and
+					# notify-resume requests
+	move	a0, sp
+	li	a1, 0
+	jal	do_notify_resume	# a2 already loaded
+	j	resume_userspace_check
+
+FEXPORT(syscall_exit_partial)
+#ifdef CONFIG_DEBUG_RSEQ
+	move	a0, sp
+	jal	rseq_syscall
+#endif
+	local_irq_disable		# make sure need_resched doesn't
+					# change between and return
+	LONG_L	a2, TI_FLAGS($28)	# current->work
+	li	t0, _TIF_ALLWORK_MASK
+	and	t0, a2
+	beqz	t0, restore_partial
+	SAVE_STATIC
+syscall_exit_work:
+	LONG_L	t0, PT_STATUS(sp)		# returning to kernel mode?
+	andi	t0, t0, KU_USER
+	beqz	t0, resume_kernel
+	li	t0, _TIF_WORK_SYSCALL_EXIT
+	and	t0, a2			# a2 is preloaded with TI_FLAGS
+	beqz	t0, work_pending	# trace bit set?
+	local_irq_enable		# could let syscall_trace_leave()
+					# call schedule() instead
+	TRACE_IRQS_ON
+	move	a0, sp
+	jal	syscall_trace_leave
+	b	resume_userspace
+
+#if defined(CONFIG_CPU_MIPSR2) || defined(CONFIG_CPU_MIPSR5) || \
+    defined(CONFIG_CPU_MIPSR6) || defined(CONFIG_MIPS_MT)
+
+/*
+ * MIPS32R2 Instruction Hazard Barrier - must be called
+ *
+ * For C code use the inline version named instruction_hazard().
+ */
+LEAF(mips_ihb)
+	.set	MIPS_ISA_LEVEL_RAW
+	jr.hb	ra
+	nop
+	END(mips_ihb)
+
+#endif /* CONFIG_CPU_MIPSR2 - CONFIG_CPU_MIPSR6 or CONFIG_MIPS_MT */
diff --git a/arch/mips/kernel/fpu-probe.c b/arch/mips/kernel/fpu-probe.c
new file mode 100644
index 000000000..e689d6a83
--- /dev/null
+++ b/arch/mips/kernel/fpu-probe.c
@@ -0,0 +1,321 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Processor capabilities determination functions.
+ *
+ * Copyright (C) xxxx  the Anonymous
+ * Copyright (C) 1994 - 2006 Ralf Baechle
+ * Copyright (C) 2003, 2004  Maciej W. Rozycki
+ * Copyright (C) 2001, 2004, 2011, 2012	 MIPS Technologies, Inc.
+ */
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+
+#include <asm/bugs.h>
+#include <asm/cpu.h>
+#include <asm/cpu-features.h>
+#include <asm/cpu-type.h>
+#include <asm/elf.h>
+#include <asm/fpu.h>
+#include <asm/mipsregs.h>
+
+#include "fpu-probe.h"
+
+/*
+ * Get the FPU Implementation/Revision.
+ */
+static inline unsigned long cpu_get_fpu_id(void)
+{
+	unsigned long tmp, fpu_id;
+
+	tmp = read_c0_status();
+	__enable_fpu(FPU_AS_IS);
+	fpu_id = read_32bit_cp1_register(CP1_REVISION);
+	write_c0_status(tmp);
+	return fpu_id;
+}
+
+/*
+ * Check if the CPU has an external FPU.
+ */
+int __cpu_has_fpu(void)
+{
+	return (cpu_get_fpu_id() & FPIR_IMP_MASK) != FPIR_IMP_NONE;
+}
+
+/*
+ * Determine the FCSR mask for FPU hardware.
+ */
+static inline void cpu_set_fpu_fcsr_mask(struct cpuinfo_mips *c)
+{
+	unsigned long sr, mask, fcsr, fcsr0, fcsr1;
+
+	fcsr = c->fpu_csr31;
+	mask = FPU_CSR_ALL_X | FPU_CSR_ALL_E | FPU_CSR_ALL_S | FPU_CSR_RM;
+
+	sr = read_c0_status();
+	__enable_fpu(FPU_AS_IS);
+
+	fcsr0 = fcsr & mask;
+	write_32bit_cp1_register(CP1_STATUS, fcsr0);
+	fcsr0 = read_32bit_cp1_register(CP1_STATUS);
+
+	fcsr1 = fcsr | ~mask;
+	write_32bit_cp1_register(CP1_STATUS, fcsr1);
+	fcsr1 = read_32bit_cp1_register(CP1_STATUS);
+
+	write_32bit_cp1_register(CP1_STATUS, fcsr);
+
+	write_c0_status(sr);
+
+	c->fpu_msk31 = ~(fcsr0 ^ fcsr1) & ~mask;
+}
+
+/*
+ * Determine the IEEE 754 NaN encodings and ABS.fmt/NEG.fmt execution modes
+ * supported by FPU hardware.
+ */
+static void cpu_set_fpu_2008(struct cpuinfo_mips *c)
+{
+	if (c->isa_level & (MIPS_CPU_ISA_M32R1 | MIPS_CPU_ISA_M64R1 |
+			    MIPS_CPU_ISA_M32R2 | MIPS_CPU_ISA_M64R2 |
+			    MIPS_CPU_ISA_M32R5 | MIPS_CPU_ISA_M64R5 |
+			    MIPS_CPU_ISA_M32R6 | MIPS_CPU_ISA_M64R6)) {
+		unsigned long sr, fir, fcsr, fcsr0, fcsr1;
+
+		sr = read_c0_status();
+		__enable_fpu(FPU_AS_IS);
+
+		fir = read_32bit_cp1_register(CP1_REVISION);
+		if (fir & MIPS_FPIR_HAS2008) {
+			fcsr = read_32bit_cp1_register(CP1_STATUS);
+
+			/*
+			 * MAC2008 toolchain never landed in real world, so
+			 * we're only testing whether it can be disabled and
+			 *  don't try to enabled it.
+			 */
+			fcsr0 = fcsr & ~(FPU_CSR_ABS2008 | FPU_CSR_NAN2008 |
+					 FPU_CSR_MAC2008);
+			write_32bit_cp1_register(CP1_STATUS, fcsr0);
+			fcsr0 = read_32bit_cp1_register(CP1_STATUS);
+
+			fcsr1 = fcsr | FPU_CSR_ABS2008 | FPU_CSR_NAN2008;
+			write_32bit_cp1_register(CP1_STATUS, fcsr1);
+			fcsr1 = read_32bit_cp1_register(CP1_STATUS);
+
+			write_32bit_cp1_register(CP1_STATUS, fcsr);
+
+			if (c->isa_level & (MIPS_CPU_ISA_M32R2 |
+					    MIPS_CPU_ISA_M64R2)) {
+				/*
+				 * The bit for MAC2008 might be reused by R6
+				 * in future, so we only test for R2-R5.
+				 */
+				if (fcsr0 & FPU_CSR_MAC2008)
+					c->options |= MIPS_CPU_MAC_2008_ONLY;
+			}
+
+			if (!(fcsr0 & FPU_CSR_NAN2008))
+				c->options |= MIPS_CPU_NAN_LEGACY;
+			if (fcsr1 & FPU_CSR_NAN2008)
+				c->options |= MIPS_CPU_NAN_2008;
+
+			if ((fcsr0 ^ fcsr1) & FPU_CSR_ABS2008)
+				c->fpu_msk31 &= ~FPU_CSR_ABS2008;
+			else
+				c->fpu_csr31 |= fcsr & FPU_CSR_ABS2008;
+
+			if ((fcsr0 ^ fcsr1) & FPU_CSR_NAN2008)
+				c->fpu_msk31 &= ~FPU_CSR_NAN2008;
+			else
+				c->fpu_csr31 |= fcsr & FPU_CSR_NAN2008;
+		} else {
+			c->options |= MIPS_CPU_NAN_LEGACY;
+		}
+
+		write_c0_status(sr);
+	} else {
+		c->options |= MIPS_CPU_NAN_LEGACY;
+	}
+}
+
+/*
+ * IEEE 754 conformance mode to use.  Affects the NaN encoding and the
+ * ABS.fmt/NEG.fmt execution mode.
+ */
+static enum { STRICT, LEGACY, STD2008, RELAXED } ieee754 = STRICT;
+
+/*
+ * Set the IEEE 754 NaN encodings and the ABS.fmt/NEG.fmt execution modes
+ * to support by the FPU emulator according to the IEEE 754 conformance
+ * mode selected.  Note that "relaxed" straps the emulator so that it
+ * allows 2008-NaN binaries even for legacy processors.
+ */
+static void cpu_set_nofpu_2008(struct cpuinfo_mips *c)
+{
+	c->options &= ~(MIPS_CPU_NAN_2008 | MIPS_CPU_NAN_LEGACY);
+	c->fpu_csr31 &= ~(FPU_CSR_ABS2008 | FPU_CSR_NAN2008);
+	c->fpu_msk31 &= ~(FPU_CSR_ABS2008 | FPU_CSR_NAN2008);
+
+	switch (ieee754) {
+	case STRICT:
+		if (c->isa_level & (MIPS_CPU_ISA_M32R1 | MIPS_CPU_ISA_M64R1 |
+				    MIPS_CPU_ISA_M32R2 | MIPS_CPU_ISA_M64R2 |
+				    MIPS_CPU_ISA_M32R5 | MIPS_CPU_ISA_M64R5 |
+				    MIPS_CPU_ISA_M32R6 | MIPS_CPU_ISA_M64R6)) {
+			c->options |= MIPS_CPU_NAN_2008 | MIPS_CPU_NAN_LEGACY;
+		} else {
+			c->options |= MIPS_CPU_NAN_LEGACY;
+			c->fpu_msk31 |= FPU_CSR_ABS2008 | FPU_CSR_NAN2008;
+		}
+		break;
+	case LEGACY:
+		c->options |= MIPS_CPU_NAN_LEGACY;
+		c->fpu_msk31 |= FPU_CSR_ABS2008 | FPU_CSR_NAN2008;
+		break;
+	case STD2008:
+		c->options |= MIPS_CPU_NAN_2008;
+		c->fpu_csr31 |= FPU_CSR_ABS2008 | FPU_CSR_NAN2008;
+		c->fpu_msk31 |= FPU_CSR_ABS2008 | FPU_CSR_NAN2008;
+		break;
+	case RELAXED:
+		c->options |= MIPS_CPU_NAN_2008 | MIPS_CPU_NAN_LEGACY;
+		break;
+	}
+}
+
+/*
+ * Override the IEEE 754 NaN encoding and ABS.fmt/NEG.fmt execution mode
+ * according to the "ieee754=" parameter.
+ */
+static void cpu_set_nan_2008(struct cpuinfo_mips *c)
+{
+	switch (ieee754) {
+	case STRICT:
+		mips_use_nan_legacy = !!cpu_has_nan_legacy;
+		mips_use_nan_2008 = !!cpu_has_nan_2008;
+		break;
+	case LEGACY:
+		mips_use_nan_legacy = !!cpu_has_nan_legacy;
+		mips_use_nan_2008 = !cpu_has_nan_legacy;
+		break;
+	case STD2008:
+		mips_use_nan_legacy = !cpu_has_nan_2008;
+		mips_use_nan_2008 = !!cpu_has_nan_2008;
+		break;
+	case RELAXED:
+		mips_use_nan_legacy = true;
+		mips_use_nan_2008 = true;
+		break;
+	}
+}
+
+/*
+ * IEEE 754 NaN encoding and ABS.fmt/NEG.fmt execution mode override
+ * settings:
+ *
+ * strict:  accept binaries that request a NaN encoding supported by the FPU
+ * legacy:  only accept legacy-NaN binaries
+ * 2008:    only accept 2008-NaN binaries
+ * relaxed: accept any binaries regardless of whether supported by the FPU
+ */
+static int __init ieee754_setup(char *s)
+{
+	if (!s)
+		return -1;
+	else if (!strcmp(s, "strict"))
+		ieee754 = STRICT;
+	else if (!strcmp(s, "legacy"))
+		ieee754 = LEGACY;
+	else if (!strcmp(s, "2008"))
+		ieee754 = STD2008;
+	else if (!strcmp(s, "relaxed"))
+		ieee754 = RELAXED;
+	else
+		return -1;
+
+	if (!(boot_cpu_data.options & MIPS_CPU_FPU))
+		cpu_set_nofpu_2008(&boot_cpu_data);
+	cpu_set_nan_2008(&boot_cpu_data);
+
+	return 0;
+}
+
+early_param("ieee754", ieee754_setup);
+
+/*
+ * Set the FIR feature flags for the FPU emulator.
+ */
+static void cpu_set_nofpu_id(struct cpuinfo_mips *c)
+{
+	u32 value;
+
+	value = 0;
+	if (c->isa_level & (MIPS_CPU_ISA_M32R1 | MIPS_CPU_ISA_M64R1 |
+			    MIPS_CPU_ISA_M32R2 | MIPS_CPU_ISA_M64R2 |
+			    MIPS_CPU_ISA_M32R5 | MIPS_CPU_ISA_M64R5 |
+			    MIPS_CPU_ISA_M32R6 | MIPS_CPU_ISA_M64R6))
+		value |= MIPS_FPIR_D | MIPS_FPIR_S;
+	if (c->isa_level & (MIPS_CPU_ISA_M32R2 | MIPS_CPU_ISA_M64R2 |
+			    MIPS_CPU_ISA_M32R5 | MIPS_CPU_ISA_M64R5 |
+			    MIPS_CPU_ISA_M32R6 | MIPS_CPU_ISA_M64R6))
+		value |= MIPS_FPIR_F64 | MIPS_FPIR_L | MIPS_FPIR_W;
+	if (c->options & MIPS_CPU_NAN_2008)
+		value |= MIPS_FPIR_HAS2008;
+	c->fpu_id = value;
+}
+
+/* Determined FPU emulator mask to use for the boot CPU with "nofpu".  */
+static unsigned int mips_nofpu_msk31;
+
+/*
+ * Set options for FPU hardware.
+ */
+void cpu_set_fpu_opts(struct cpuinfo_mips *c)
+{
+	c->fpu_id = cpu_get_fpu_id();
+	mips_nofpu_msk31 = c->fpu_msk31;
+
+	if (c->isa_level & (MIPS_CPU_ISA_M32R1 | MIPS_CPU_ISA_M64R1 |
+			    MIPS_CPU_ISA_M32R2 | MIPS_CPU_ISA_M64R2 |
+			    MIPS_CPU_ISA_M32R5 | MIPS_CPU_ISA_M64R5 |
+			    MIPS_CPU_ISA_M32R6 | MIPS_CPU_ISA_M64R6)) {
+		if (c->fpu_id & MIPS_FPIR_3D)
+			c->ases |= MIPS_ASE_MIPS3D;
+		if (c->fpu_id & MIPS_FPIR_UFRP)
+			c->options |= MIPS_CPU_UFR;
+		if (c->fpu_id & MIPS_FPIR_FREP)
+			c->options |= MIPS_CPU_FRE;
+	}
+
+	cpu_set_fpu_fcsr_mask(c);
+	cpu_set_fpu_2008(c);
+	cpu_set_nan_2008(c);
+}
+
+/*
+ * Set options for the FPU emulator.
+ */
+void cpu_set_nofpu_opts(struct cpuinfo_mips *c)
+{
+	c->options &= ~MIPS_CPU_FPU;
+	c->fpu_msk31 = mips_nofpu_msk31;
+
+	cpu_set_nofpu_2008(c);
+	cpu_set_nan_2008(c);
+	cpu_set_nofpu_id(c);
+}
+
+int mips_fpu_disabled;
+
+static int __init fpu_disable(char *s)
+{
+	cpu_set_nofpu_opts(&boot_cpu_data);
+	mips_fpu_disabled = 1;
+
+	return 1;
+}
+
+__setup("nofpu", fpu_disable);
+
diff --git a/arch/mips/kernel/fpu-probe.h b/arch/mips/kernel/fpu-probe.h
new file mode 100644
index 000000000..951ce5089
--- /dev/null
+++ b/arch/mips/kernel/fpu-probe.h
@@ -0,0 +1,40 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+
+#include <linux/kernel.h>
+
+#include <asm/cpu.h>
+#include <asm/cpu-info.h>
+
+#ifdef CONFIG_MIPS_FP_SUPPORT
+
+extern int mips_fpu_disabled;
+
+int __cpu_has_fpu(void);
+void cpu_set_fpu_opts(struct cpuinfo_mips *c);
+void cpu_set_nofpu_opts(struct cpuinfo_mips *c);
+
+#else /* !CONFIG_MIPS_FP_SUPPORT */
+
+#define mips_fpu_disabled 1
+
+static inline unsigned long cpu_get_fpu_id(void)
+{
+	return FPIR_IMP_NONE;
+}
+
+static inline int __cpu_has_fpu(void)
+{
+	return 0;
+}
+
+static inline void cpu_set_fpu_opts(struct cpuinfo_mips *c)
+{
+	/* no-op */
+}
+
+static inline void cpu_set_nofpu_opts(struct cpuinfo_mips *c)
+{
+	/* no-op */
+}
+
+#endif /* CONFIG_MIPS_FP_SUPPORT */
diff --git a/arch/mips/kernel/ftrace.c b/arch/mips/kernel/ftrace.c
new file mode 100644
index 000000000..f57e68f40
--- /dev/null
+++ b/arch/mips/kernel/ftrace.c
@@ -0,0 +1,414 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Code for replacing ftrace calls with jumps.
+ *
+ * Copyright (C) 2007-2008 Steven Rostedt <srostedt@redhat.com>
+ * Copyright (C) 2009, 2010 DSLab, Lanzhou University, China
+ * Author: Wu Zhangjin <wuzhangjin@gmail.com>
+ *
+ * Thanks goes to Steven Rostedt for writing the original x86 version.
+ */
+
+#include <linux/uaccess.h>
+#include <linux/init.h>
+#include <linux/ftrace.h>
+#include <linux/syscalls.h>
+
+#include <asm/asm.h>
+#include <asm/asm-offsets.h>
+#include <asm/cacheflush.h>
+#include <asm/syscall.h>
+#include <asm/uasm.h>
+#include <asm/unistd.h>
+
+#include <asm-generic/sections.h>
+
+#if defined(KBUILD_MCOUNT_RA_ADDRESS) && defined(CONFIG_32BIT)
+#define MCOUNT_OFFSET_INSNS 5
+#else
+#define MCOUNT_OFFSET_INSNS 4
+#endif
+
+#ifdef CONFIG_DYNAMIC_FTRACE
+
+/* Arch override because MIPS doesn't need to run this from stop_machine() */
+void arch_ftrace_update_code(int command)
+{
+	ftrace_modify_all_code(command);
+}
+
+#define JAL 0x0c000000		/* jump & link: ip --> ra, jump to target */
+#define ADDR_MASK 0x03ffffff	/*  op_code|addr : 31...26|25 ....0 */
+#define JUMP_RANGE_MASK ((1UL << 28) - 1)
+
+#define INSN_NOP 0x00000000	/* nop */
+#define INSN_JAL(addr)	\
+	((unsigned int)(JAL | (((addr) >> 2) & ADDR_MASK)))
+
+static unsigned int insn_jal_ftrace_caller __read_mostly;
+static unsigned int insn_la_mcount[2] __read_mostly;
+static unsigned int insn_j_ftrace_graph_caller __maybe_unused __read_mostly;
+
+static inline void ftrace_dyn_arch_init_insns(void)
+{
+	u32 *buf;
+	unsigned int v1;
+
+	/* la v1, _mcount */
+	v1 = 3;
+	buf = (u32 *)&insn_la_mcount[0];
+	UASM_i_LA(&buf, v1, MCOUNT_ADDR);
+
+	/* jal (ftrace_caller + 8), jump over the first two instruction */
+	buf = (u32 *)&insn_jal_ftrace_caller;
+	uasm_i_jal(&buf, (FTRACE_ADDR + 8) & JUMP_RANGE_MASK);
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+	/* j ftrace_graph_caller */
+	buf = (u32 *)&insn_j_ftrace_graph_caller;
+	uasm_i_j(&buf, (unsigned long)ftrace_graph_caller & JUMP_RANGE_MASK);
+#endif
+}
+
+static int ftrace_modify_code(unsigned long ip, unsigned int new_code)
+{
+	int faulted;
+	mm_segment_t old_fs;
+
+	/* *(unsigned int *)ip = new_code; */
+	safe_store_code(new_code, ip, faulted);
+
+	if (unlikely(faulted))
+		return -EFAULT;
+
+	old_fs = get_fs();
+	set_fs(KERNEL_DS);
+	flush_icache_range(ip, ip + 8);
+	set_fs(old_fs);
+
+	return 0;
+}
+
+#ifndef CONFIG_64BIT
+static int ftrace_modify_code_2(unsigned long ip, unsigned int new_code1,
+				unsigned int new_code2)
+{
+	int faulted;
+	mm_segment_t old_fs;
+
+	safe_store_code(new_code1, ip, faulted);
+	if (unlikely(faulted))
+		return -EFAULT;
+
+	ip += 4;
+	safe_store_code(new_code2, ip, faulted);
+	if (unlikely(faulted))
+		return -EFAULT;
+
+	ip -= 4;
+	old_fs = get_fs();
+	set_fs(KERNEL_DS);
+	flush_icache_range(ip, ip + 8);
+	set_fs(old_fs);
+
+	return 0;
+}
+
+static int ftrace_modify_code_2r(unsigned long ip, unsigned int new_code1,
+				 unsigned int new_code2)
+{
+	int faulted;
+	mm_segment_t old_fs;
+
+	ip += 4;
+	safe_store_code(new_code2, ip, faulted);
+	if (unlikely(faulted))
+		return -EFAULT;
+
+	ip -= 4;
+	safe_store_code(new_code1, ip, faulted);
+	if (unlikely(faulted))
+		return -EFAULT;
+
+	old_fs = get_fs();
+	set_fs(KERNEL_DS);
+	flush_icache_range(ip, ip + 8);
+	set_fs(old_fs);
+
+	return 0;
+}
+#endif
+
+/*
+ * The details about the calling site of mcount on MIPS
+ *
+ * 1. For kernel:
+ *
+ * move at, ra
+ * jal _mcount		--> nop
+ *  sub sp, sp, 8	--> nop  (CONFIG_32BIT)
+ *
+ * 2. For modules:
+ *
+ * 2.1 For KBUILD_MCOUNT_RA_ADDRESS and CONFIG_32BIT
+ *
+ * lui v1, hi_16bit_of_mcount	     --> b 1f (0x10000005)
+ * addiu v1, v1, low_16bit_of_mcount --> nop  (CONFIG_32BIT)
+ * move at, ra
+ * move $12, ra_address
+ * jalr v1
+ *  sub sp, sp, 8
+ *				    1: offset = 5 instructions
+ * 2.2 For the Other situations
+ *
+ * lui v1, hi_16bit_of_mcount	     --> b 1f (0x10000004)
+ * addiu v1, v1, low_16bit_of_mcount --> nop  (CONFIG_32BIT)
+ * move at, ra
+ * jalr v1
+ *  nop | move $12, ra_address | sub sp, sp, 8
+ *				    1: offset = 4 instructions
+ */
+
+#define INSN_B_1F (0x10000000 | MCOUNT_OFFSET_INSNS)
+
+int ftrace_make_nop(struct module *mod,
+		    struct dyn_ftrace *rec, unsigned long addr)
+{
+	unsigned int new;
+	unsigned long ip = rec->ip;
+
+	/*
+	 * If ip is in kernel space, no long call, otherwise, long call is
+	 * needed.
+	 */
+	new = core_kernel_text(ip) ? INSN_NOP : INSN_B_1F;
+#ifdef CONFIG_64BIT
+	return ftrace_modify_code(ip, new);
+#else
+	/*
+	 * On 32 bit MIPS platforms, gcc adds a stack adjust
+	 * instruction in the delay slot after the branch to
+	 * mcount and expects mcount to restore the sp on return.
+	 * This is based on a legacy API and does nothing but
+	 * waste instructions so it's being removed at runtime.
+	 */
+	return ftrace_modify_code_2(ip, new, INSN_NOP);
+#endif
+}
+
+int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr)
+{
+	unsigned int new;
+	unsigned long ip = rec->ip;
+
+	new = core_kernel_text(ip) ? insn_jal_ftrace_caller : insn_la_mcount[0];
+
+#ifdef CONFIG_64BIT
+	return ftrace_modify_code(ip, new);
+#else
+	return ftrace_modify_code_2r(ip, new, core_kernel_text(ip) ?
+						INSN_NOP : insn_la_mcount[1]);
+#endif
+}
+
+#define FTRACE_CALL_IP ((unsigned long)(&ftrace_call))
+
+int ftrace_update_ftrace_func(ftrace_func_t func)
+{
+	unsigned int new;
+
+	new = INSN_JAL((unsigned long)func);
+
+	return ftrace_modify_code(FTRACE_CALL_IP, new);
+}
+
+int __init ftrace_dyn_arch_init(void)
+{
+	/* Encode the instructions when booting */
+	ftrace_dyn_arch_init_insns();
+
+	/* Remove "b ftrace_stub" to ensure ftrace_caller() is executed */
+	ftrace_modify_code(MCOUNT_ADDR, INSN_NOP);
+
+	return 0;
+}
+#endif	/* CONFIG_DYNAMIC_FTRACE */
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+
+#ifdef CONFIG_DYNAMIC_FTRACE
+
+extern void ftrace_graph_call(void);
+#define FTRACE_GRAPH_CALL_IP	((unsigned long)(&ftrace_graph_call))
+
+int ftrace_enable_ftrace_graph_caller(void)
+{
+	return ftrace_modify_code(FTRACE_GRAPH_CALL_IP,
+			insn_j_ftrace_graph_caller);
+}
+
+int ftrace_disable_ftrace_graph_caller(void)
+{
+	return ftrace_modify_code(FTRACE_GRAPH_CALL_IP, INSN_NOP);
+}
+
+#endif	/* CONFIG_DYNAMIC_FTRACE */
+
+#ifndef KBUILD_MCOUNT_RA_ADDRESS
+
+#define S_RA_SP (0xafbf << 16)	/* s{d,w} ra, offset(sp) */
+#define S_R_SP	(0xafb0 << 16)	/* s{d,w} R, offset(sp) */
+#define OFFSET_MASK	0xffff	/* stack offset range: 0 ~ PT_SIZE */
+
+unsigned long ftrace_get_parent_ra_addr(unsigned long self_ra, unsigned long
+		old_parent_ra, unsigned long parent_ra_addr, unsigned long fp)
+{
+	unsigned long sp, ip, tmp;
+	unsigned int code;
+	int faulted;
+
+	/*
+	 * For module, move the ip from the return address after the
+	 * instruction "lui v1, hi_16bit_of_mcount"(offset is 24), but for
+	 * kernel, move after the instruction "move ra, at"(offset is 16)
+	 */
+	ip = self_ra - (core_kernel_text(self_ra) ? 16 : 24);
+
+	/*
+	 * search the text until finding the non-store instruction or "s{d,w}
+	 * ra, offset(sp)" instruction
+	 */
+	do {
+		/* get the code at "ip": code = *(unsigned int *)ip; */
+		safe_load_code(code, ip, faulted);
+
+		if (unlikely(faulted))
+			return 0;
+		/*
+		 * If we hit the non-store instruction before finding where the
+		 * ra is stored, then this is a leaf function and it does not
+		 * store the ra on the stack
+		 */
+		if ((code & S_R_SP) != S_R_SP)
+			return parent_ra_addr;
+
+		/* Move to the next instruction */
+		ip -= 4;
+	} while ((code & S_RA_SP) != S_RA_SP);
+
+	sp = fp + (code & OFFSET_MASK);
+
+	/* tmp = *(unsigned long *)sp; */
+	safe_load_stack(tmp, sp, faulted);
+	if (unlikely(faulted))
+		return 0;
+
+	if (tmp == old_parent_ra)
+		return sp;
+	return 0;
+}
+
+#endif	/* !KBUILD_MCOUNT_RA_ADDRESS */
+
+/*
+ * Hook the return address and push it in the stack of return addrs
+ * in current thread info.
+ */
+void prepare_ftrace_return(unsigned long *parent_ra_addr, unsigned long self_ra,
+			   unsigned long fp)
+{
+	unsigned long old_parent_ra;
+	unsigned long return_hooker = (unsigned long)
+	    &return_to_handler;
+	int faulted, insns;
+
+	if (unlikely(ftrace_graph_is_dead()))
+		return;
+
+	if (unlikely(atomic_read(&current->tracing_graph_pause)))
+		return;
+
+	/*
+	 * "parent_ra_addr" is the stack address where the return address of
+	 * the caller of _mcount is saved.
+	 *
+	 * If gcc < 4.5, a leaf function does not save the return address
+	 * in the stack address, so we "emulate" one in _mcount's stack space,
+	 * and hijack it directly.
+	 * For a non-leaf function, it does save the return address to its own
+	 * stack space, so we can not hijack it directly, but need to find the
+	 * real stack address, which is done by ftrace_get_parent_addr().
+	 *
+	 * If gcc >= 4.5, with the new -mmcount-ra-address option, for a
+	 * non-leaf function, the location of the return address will be saved
+	 * to $12 for us.
+	 * For a leaf function, it just puts a zero into $12, so we handle
+	 * it in ftrace_graph_caller() of mcount.S.
+	 */
+
+	/* old_parent_ra = *parent_ra_addr; */
+	safe_load_stack(old_parent_ra, parent_ra_addr, faulted);
+	if (unlikely(faulted))
+		goto out;
+#ifndef KBUILD_MCOUNT_RA_ADDRESS
+	parent_ra_addr = (unsigned long *)ftrace_get_parent_ra_addr(self_ra,
+			old_parent_ra, (unsigned long)parent_ra_addr, fp);
+	/*
+	 * If fails when getting the stack address of the non-leaf function's
+	 * ra, stop function graph tracer and return
+	 */
+	if (parent_ra_addr == NULL)
+		goto out;
+#endif
+	/* *parent_ra_addr = return_hooker; */
+	safe_store_stack(return_hooker, parent_ra_addr, faulted);
+	if (unlikely(faulted))
+		goto out;
+
+	/*
+	 * Get the recorded ip of the current mcount calling site in the
+	 * __mcount_loc section, which will be used to filter the function
+	 * entries configured through the tracing/set_graph_function interface.
+	 */
+
+	insns = core_kernel_text(self_ra) ? 2 : MCOUNT_OFFSET_INSNS + 1;
+	self_ra -= (MCOUNT_INSN_SIZE * insns);
+
+	if (function_graph_enter(old_parent_ra, self_ra, fp, NULL))
+		*parent_ra_addr = old_parent_ra;
+	return;
+out:
+	ftrace_graph_stop();
+	WARN_ON(1);
+}
+#endif	/* CONFIG_FUNCTION_GRAPH_TRACER */
+
+#ifdef CONFIG_FTRACE_SYSCALLS
+
+#ifdef CONFIG_32BIT
+unsigned long __init arch_syscall_addr(int nr)
+{
+	return (unsigned long)sys_call_table[nr - __NR_O32_Linux];
+}
+#endif
+
+#ifdef CONFIG_64BIT
+
+unsigned long __init arch_syscall_addr(int nr)
+{
+#ifdef CONFIG_MIPS32_N32
+	if (nr >= __NR_N32_Linux && nr < __NR_N32_Linux + __NR_N32_Linux_syscalls)
+		return (unsigned long)sysn32_call_table[nr - __NR_N32_Linux];
+#endif
+	if (nr >= __NR_64_Linux  && nr < __NR_64_Linux + __NR_64_Linux_syscalls)
+		return (unsigned long)sys_call_table[nr - __NR_64_Linux];
+#ifdef CONFIG_MIPS32_O32
+	if (nr >= __NR_O32_Linux && nr < __NR_O32_Linux + __NR_O32_Linux_syscalls)
+		return (unsigned long)sys32_call_table[nr - __NR_O32_Linux];
+#endif
+
+	return (unsigned long) &sys_ni_syscall;
+}
+#endif
+
+#endif /* CONFIG_FTRACE_SYSCALLS */
diff --git a/arch/mips/kernel/genex.S b/arch/mips/kernel/genex.S
new file mode 100644
index 000000000..bcce32a3d
--- /dev/null
+++ b/arch/mips/kernel/genex.S
@@ -0,0 +1,682 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1994 - 2000, 2001, 2003 Ralf Baechle
+ * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ * Copyright (C) 2002, 2007  Maciej W. Rozycki
+ * Copyright (C) 2001, 2012 MIPS Technologies, Inc.  All rights reserved.
+ */
+#include <linux/init.h>
+
+#include <asm/asm.h>
+#include <asm/asmmacro.h>
+#include <asm/cacheops.h>
+#include <asm/irqflags.h>
+#include <asm/regdef.h>
+#include <asm/fpregdef.h>
+#include <asm/mipsregs.h>
+#include <asm/stackframe.h>
+#include <asm/sync.h>
+#include <asm/war.h>
+#include <asm/thread_info.h>
+
+	__INIT
+
+/*
+ * General exception vector for all other CPUs.
+ *
+ * Be careful when changing this, it has to be at most 128 bytes
+ * to fit into space reserved for the exception handler.
+ */
+NESTED(except_vec3_generic, 0, sp)
+	.set	push
+	.set	noat
+	mfc0	k1, CP0_CAUSE
+	andi	k1, k1, 0x7c
+#ifdef CONFIG_64BIT
+	dsll	k1, k1, 1
+#endif
+	PTR_L	k0, exception_handlers(k1)
+	jr	k0
+	.set	pop
+	END(except_vec3_generic)
+
+/*
+ * General exception handler for CPUs with virtual coherency exception.
+ *
+ * Be careful when changing this, it has to be at most 256 (as a special
+ * exception) bytes to fit into space reserved for the exception handler.
+ */
+NESTED(except_vec3_r4000, 0, sp)
+	.set	push
+	.set	arch=r4000
+	.set	noat
+	mfc0	k1, CP0_CAUSE
+	li	k0, 31<<2
+	andi	k1, k1, 0x7c
+	.set	push
+	.set	noreorder
+	.set	nomacro
+	beq	k1, k0, handle_vced
+	 li	k0, 14<<2
+	beq	k1, k0, handle_vcei
+#ifdef CONFIG_64BIT
+	 dsll	k1, k1, 1
+#endif
+	.set	pop
+	PTR_L	k0, exception_handlers(k1)
+	jr	k0
+
+	/*
+	 * Big shit, we now may have two dirty primary cache lines for the same
+	 * physical address.  We can safely invalidate the line pointed to by
+	 * c0_badvaddr because after return from this exception handler the
+	 * load / store will be re-executed.
+	 */
+handle_vced:
+	MFC0	k0, CP0_BADVADDR
+	li	k1, -4					# Is this ...
+	and	k0, k1					# ... really needed?
+	mtc0	zero, CP0_TAGLO
+	cache	Index_Store_Tag_D, (k0)
+	cache	Hit_Writeback_Inv_SD, (k0)
+#ifdef CONFIG_PROC_FS
+	PTR_LA	k0, vced_count
+	lw	k1, (k0)
+	addiu	k1, 1
+	sw	k1, (k0)
+#endif
+	eret
+
+handle_vcei:
+	MFC0	k0, CP0_BADVADDR
+	cache	Hit_Writeback_Inv_SD, (k0)		# also cleans pi
+#ifdef CONFIG_PROC_FS
+	PTR_LA	k0, vcei_count
+	lw	k1, (k0)
+	addiu	k1, 1
+	sw	k1, (k0)
+#endif
+	eret
+	.set	pop
+	END(except_vec3_r4000)
+
+	__FINIT
+
+	.align	5	/* 32 byte rollback region */
+LEAF(__r4k_wait)
+	.set	push
+	.set	noreorder
+	/* start of rollback region */
+	LONG_L	t0, TI_FLAGS($28)
+	nop
+	andi	t0, _TIF_NEED_RESCHED
+	bnez	t0, 1f
+	 nop
+	nop
+	nop
+#ifdef CONFIG_CPU_MICROMIPS
+	nop
+	nop
+	nop
+	nop
+#endif
+	.set	MIPS_ISA_ARCH_LEVEL_RAW
+	wait
+	/* end of rollback region (the region size must be power of two) */
+1:
+	jr	ra
+	 nop
+	.set	pop
+	END(__r4k_wait)
+
+	.macro	BUILD_ROLLBACK_PROLOGUE handler
+	FEXPORT(rollback_\handler)
+	.set	push
+	.set	noat
+	MFC0	k0, CP0_EPC
+	PTR_LA	k1, __r4k_wait
+	ori	k0, 0x1f	/* 32 byte rollback region */
+	xori	k0, 0x1f
+	bne	k0, k1, \handler
+	MTC0	k0, CP0_EPC
+	.set pop
+	.endm
+
+	.align	5
+BUILD_ROLLBACK_PROLOGUE handle_int
+NESTED(handle_int, PT_SIZE, sp)
+	.cfi_signal_frame
+#ifdef CONFIG_TRACE_IRQFLAGS
+	/*
+	 * Check to see if the interrupted code has just disabled
+	 * interrupts and ignore this interrupt for now if so.
+	 *
+	 * local_irq_disable() disables interrupts and then calls
+	 * trace_hardirqs_off() to track the state. If an interrupt is taken
+	 * after interrupts are disabled but before the state is updated
+	 * it will appear to restore_all that it is incorrectly returning with
+	 * interrupts disabled
+	 */
+	.set	push
+	.set	noat
+	mfc0	k0, CP0_STATUS
+#if defined(CONFIG_CPU_R3000) || defined(CONFIG_CPU_TX39XX)
+	and	k0, ST0_IEP
+	bnez	k0, 1f
+
+	mfc0	k0, CP0_EPC
+	.set	noreorder
+	j	k0
+	 rfe
+#else
+	and	k0, ST0_IE
+	bnez	k0, 1f
+
+	eret
+#endif
+1:
+	.set pop
+#endif
+	SAVE_ALL docfi=1
+	CLI
+	TRACE_IRQS_OFF
+
+	LONG_L	s0, TI_REGS($28)
+	LONG_S	sp, TI_REGS($28)
+
+	/*
+	 * SAVE_ALL ensures we are using a valid kernel stack for the thread.
+	 * Check if we are already using the IRQ stack.
+	 */
+	move	s1, sp # Preserve the sp
+
+	/* Get IRQ stack for this CPU */
+	ASM_CPUID_MFC0	k0, ASM_SMP_CPUID_REG
+#if defined(CONFIG_32BIT) || defined(KBUILD_64BIT_SYM32)
+	lui	k1, %hi(irq_stack)
+#else
+	lui	k1, %highest(irq_stack)
+	daddiu	k1, %higher(irq_stack)
+	dsll	k1, 16
+	daddiu	k1, %hi(irq_stack)
+	dsll	k1, 16
+#endif
+	LONG_SRL	k0, SMP_CPUID_PTRSHIFT
+	LONG_ADDU	k1, k0
+	LONG_L	t0, %lo(irq_stack)(k1)
+
+	# Check if already on IRQ stack
+	PTR_LI	t1, ~(_THREAD_SIZE-1)
+	and	t1, t1, sp
+	beq	t0, t1, 2f
+
+	/* Switch to IRQ stack */
+	li	t1, _IRQ_STACK_START
+	PTR_ADD sp, t0, t1
+
+	/* Save task's sp on IRQ stack so that unwinding can follow it */
+	LONG_S	s1, 0(sp)
+2:
+	jal	plat_irq_dispatch
+
+	/* Restore sp */
+	move	sp, s1
+
+	j	ret_from_irq
+#ifdef CONFIG_CPU_MICROMIPS
+	nop
+#endif
+	END(handle_int)
+
+	__INIT
+
+/*
+ * Special interrupt vector for MIPS64 ISA & embedded MIPS processors.
+ * This is a dedicated interrupt exception vector which reduces the
+ * interrupt processing overhead.  The jump instruction will be replaced
+ * at the initialization time.
+ *
+ * Be careful when changing this, it has to be at most 128 bytes
+ * to fit into space reserved for the exception handler.
+ */
+NESTED(except_vec4, 0, sp)
+1:	j	1b			/* Dummy, will be replaced */
+	END(except_vec4)
+
+/*
+ * EJTAG debug exception handler.
+ * The EJTAG debug exception entry point is 0xbfc00480, which
+ * normally is in the boot PROM, so the boot PROM must do an
+ * unconditional jump to this vector.
+ */
+NESTED(except_vec_ejtag_debug, 0, sp)
+	j	ejtag_debug_handler
+#ifdef CONFIG_CPU_MICROMIPS
+	 nop
+#endif
+	END(except_vec_ejtag_debug)
+
+	__FINIT
+
+/*
+ * Vectored interrupt handler.
+ * This prototype is copied to ebase + n*IntCtl.VS and patched
+ * to invoke the handler
+ */
+BUILD_ROLLBACK_PROLOGUE except_vec_vi
+NESTED(except_vec_vi, 0, sp)
+	SAVE_SOME docfi=1
+	SAVE_AT docfi=1
+	.set	push
+	.set	noreorder
+	PTR_LA	v1, except_vec_vi_handler
+FEXPORT(except_vec_vi_lui)
+	lui	v0, 0		/* Patched */
+	jr	v1
+FEXPORT(except_vec_vi_ori)
+	 ori	v0, 0		/* Patched */
+	.set	pop
+	END(except_vec_vi)
+EXPORT(except_vec_vi_end)
+
+/*
+ * Common Vectored Interrupt code
+ * Complete the register saves and invoke the handler which is passed in $v0
+ */
+NESTED(except_vec_vi_handler, 0, sp)
+	SAVE_TEMP
+	SAVE_STATIC
+	CLI
+#ifdef CONFIG_TRACE_IRQFLAGS
+	move	s0, v0
+	TRACE_IRQS_OFF
+	move	v0, s0
+#endif
+
+	LONG_L	s0, TI_REGS($28)
+	LONG_S	sp, TI_REGS($28)
+
+	/*
+	 * SAVE_ALL ensures we are using a valid kernel stack for the thread.
+	 * Check if we are already using the IRQ stack.
+	 */
+	move	s1, sp # Preserve the sp
+
+	/* Get IRQ stack for this CPU */
+	ASM_CPUID_MFC0	k0, ASM_SMP_CPUID_REG
+#if defined(CONFIG_32BIT) || defined(KBUILD_64BIT_SYM32)
+	lui	k1, %hi(irq_stack)
+#else
+	lui	k1, %highest(irq_stack)
+	daddiu	k1, %higher(irq_stack)
+	dsll	k1, 16
+	daddiu	k1, %hi(irq_stack)
+	dsll	k1, 16
+#endif
+	LONG_SRL	k0, SMP_CPUID_PTRSHIFT
+	LONG_ADDU	k1, k0
+	LONG_L	t0, %lo(irq_stack)(k1)
+
+	# Check if already on IRQ stack
+	PTR_LI	t1, ~(_THREAD_SIZE-1)
+	and	t1, t1, sp
+	beq	t0, t1, 2f
+
+	/* Switch to IRQ stack */
+	li	t1, _IRQ_STACK_START
+	PTR_ADD sp, t0, t1
+
+	/* Save task's sp on IRQ stack so that unwinding can follow it */
+	LONG_S	s1, 0(sp)
+2:
+	jalr	v0
+
+	/* Restore sp */
+	move	sp, s1
+
+	j	ret_from_irq
+	END(except_vec_vi_handler)
+
+/*
+ * EJTAG debug exception handler.
+ */
+NESTED(ejtag_debug_handler, PT_SIZE, sp)
+	.set	push
+	.set	noat
+	MTC0	k0, CP0_DESAVE
+	mfc0	k0, CP0_DEBUG
+
+	sll	k0, k0, 30	# Check for SDBBP.
+	bgez	k0, ejtag_return
+
+#ifdef CONFIG_SMP
+1:	PTR_LA	k0, ejtag_debug_buffer_spinlock
+	__SYNC(full, loongson3_war)
+2:	ll	k0, 0(k0)
+	bnez	k0, 2b
+	PTR_LA	k0, ejtag_debug_buffer_spinlock
+	sc	k0, 0(k0)
+	beqz	k0, 1b
+# ifdef CONFIG_WEAK_REORDERING_BEYOND_LLSC
+	sync
+# endif
+
+	PTR_LA	k0, ejtag_debug_buffer
+	LONG_S	k1, 0(k0)
+
+	ASM_CPUID_MFC0 k1, ASM_SMP_CPUID_REG
+	PTR_SRL	k1, SMP_CPUID_PTRSHIFT
+	PTR_SLL	k1, LONGLOG
+	PTR_LA	k0, ejtag_debug_buffer_per_cpu
+	PTR_ADDU k0, k1
+
+	PTR_LA	k1, ejtag_debug_buffer
+	LONG_L	k1, 0(k1)
+	LONG_S	k1, 0(k0)
+
+	PTR_LA	k0, ejtag_debug_buffer_spinlock
+	sw	zero, 0(k0)
+#else
+	PTR_LA	k0, ejtag_debug_buffer
+	LONG_S	k1, 0(k0)
+#endif
+
+	SAVE_ALL
+	move	a0, sp
+	jal	ejtag_exception_handler
+	RESTORE_ALL
+
+#ifdef CONFIG_SMP
+	ASM_CPUID_MFC0 k1, ASM_SMP_CPUID_REG
+	PTR_SRL	k1, SMP_CPUID_PTRSHIFT
+	PTR_SLL	k1, LONGLOG
+	PTR_LA	k0, ejtag_debug_buffer_per_cpu
+	PTR_ADDU k0, k1
+	LONG_L	k1, 0(k0)
+#else
+	PTR_LA	k0, ejtag_debug_buffer
+	LONG_L	k1, 0(k0)
+#endif
+
+ejtag_return:
+	back_to_back_c0_hazard
+	MFC0	k0, CP0_DESAVE
+	.set	mips32
+	deret
+	.set	pop
+	END(ejtag_debug_handler)
+
+/*
+ * This buffer is reserved for the use of the EJTAG debug
+ * handler.
+ */
+	.data
+EXPORT(ejtag_debug_buffer)
+	.fill	LONGSIZE
+#ifdef CONFIG_SMP
+EXPORT(ejtag_debug_buffer_spinlock)
+	.fill	LONGSIZE
+EXPORT(ejtag_debug_buffer_per_cpu)
+	.fill	LONGSIZE * NR_CPUS
+#endif
+	.previous
+
+	__INIT
+
+/*
+ * NMI debug exception handler for MIPS reference boards.
+ * The NMI debug exception entry point is 0xbfc00000, which
+ * normally is in the boot PROM, so the boot PROM must do a
+ * unconditional jump to this vector.
+ */
+NESTED(except_vec_nmi, 0, sp)
+	j	nmi_handler
+#ifdef CONFIG_CPU_MICROMIPS
+	 nop
+#endif
+	END(except_vec_nmi)
+
+	__FINIT
+
+NESTED(nmi_handler, PT_SIZE, sp)
+	.cfi_signal_frame
+	.set	push
+	.set	noat
+	/*
+	 * Clear ERL - restore segment mapping
+	 * Clear BEV - required for page fault exception handler to work
+	 */
+	mfc0	k0, CP0_STATUS
+	ori	k0, k0, ST0_EXL
+	li	k1, ~(ST0_BEV | ST0_ERL)
+	and	k0, k0, k1
+	mtc0	k0, CP0_STATUS
+	_ehb
+	SAVE_ALL
+	move	a0, sp
+	jal	nmi_exception_handler
+	/* nmi_exception_handler never returns */
+	.set	pop
+	END(nmi_handler)
+
+	.macro	__build_clear_none
+	.endm
+
+	.macro	__build_clear_sti
+	TRACE_IRQS_ON
+	STI
+	.endm
+
+	.macro	__build_clear_cli
+	CLI
+	TRACE_IRQS_OFF
+	.endm
+
+	.macro	__build_clear_fpe
+	CLI
+	TRACE_IRQS_OFF
+	.set	push
+	/* gas fails to assemble cfc1 for some archs (octeon).*/ \
+	.set	mips1
+	SET_HARDFLOAT
+	cfc1	a1, fcr31
+	.set	pop
+	.endm
+
+	.macro	__build_clear_msa_fpe
+	CLI
+	TRACE_IRQS_OFF
+	_cfcmsa	a1, MSA_CSR
+	.endm
+
+	.macro	__build_clear_ade
+	MFC0	t0, CP0_BADVADDR
+	PTR_S	t0, PT_BVADDR(sp)
+	KMODE
+	.endm
+
+	.macro __build_clear_gsexc
+	.set	push
+	/*
+	 * We need to specify a selector to access the CP0.Diag1 (GSCause)
+	 * register. All GSExc-equipped processors have MIPS32.
+	 */
+	.set	mips32
+	mfc0	a1, CP0_DIAGNOSTIC1
+	.set	pop
+	TRACE_IRQS_ON
+	STI
+	.endm
+
+	.macro	__BUILD_silent exception
+	.endm
+
+	/* Gas tries to parse the ASM_PRINT argument as a string containing
+	   string escapes and emits bogus warnings if it believes to
+	   recognize an unknown escape code.  So make the arguments
+	   start with an n and gas will believe \n is ok ...  */
+	.macro	__BUILD_verbose nexception
+	LONG_L	a1, PT_EPC(sp)
+#ifdef CONFIG_32BIT
+	ASM_PRINT("Got \nexception at %08lx\012")
+#endif
+#ifdef CONFIG_64BIT
+	ASM_PRINT("Got \nexception at %016lx\012")
+#endif
+	.endm
+
+	.macro	__BUILD_count exception
+	LONG_L	t0,exception_count_\exception
+	LONG_ADDIU	t0, 1
+	LONG_S	t0,exception_count_\exception
+	.comm	exception_count\exception, 8, 8
+	.endm
+
+	.macro	__BUILD_HANDLER exception handler clear verbose ext
+	.align	5
+	NESTED(handle_\exception, PT_SIZE, sp)
+	.cfi_signal_frame
+	.set	noat
+	SAVE_ALL
+	FEXPORT(handle_\exception\ext)
+	__build_clear_\clear
+	.set	at
+	__BUILD_\verbose \exception
+	move	a0, sp
+	jal	do_\handler
+	j	ret_from_exception
+	END(handle_\exception)
+	.endm
+
+	.macro	BUILD_HANDLER exception handler clear verbose
+	__BUILD_HANDLER \exception \handler \clear \verbose _int
+	.endm
+
+	BUILD_HANDLER adel ade ade silent		/* #4  */
+	BUILD_HANDLER ades ade ade silent		/* #5  */
+	BUILD_HANDLER ibe be cli silent			/* #6  */
+	BUILD_HANDLER dbe be cli silent			/* #7  */
+	BUILD_HANDLER bp bp sti silent			/* #9  */
+	BUILD_HANDLER ri ri sti silent			/* #10 */
+	BUILD_HANDLER cpu cpu sti silent		/* #11 */
+	BUILD_HANDLER ov ov sti silent			/* #12 */
+	BUILD_HANDLER tr tr sti silent			/* #13 */
+	BUILD_HANDLER msa_fpe msa_fpe msa_fpe silent	/* #14 */
+#ifdef CONFIG_MIPS_FP_SUPPORT
+	BUILD_HANDLER fpe fpe fpe silent		/* #15 */
+#endif
+	BUILD_HANDLER ftlb ftlb none silent		/* #16 */
+	BUILD_HANDLER gsexc gsexc gsexc silent		/* #16 */
+	BUILD_HANDLER msa msa sti silent		/* #21 */
+	BUILD_HANDLER mdmx mdmx sti silent		/* #22 */
+#ifdef	CONFIG_HARDWARE_WATCHPOINTS
+	/*
+	 * For watch, interrupts will be enabled after the watch
+	 * registers are read.
+	 */
+	BUILD_HANDLER watch watch cli silent		/* #23 */
+#else
+	BUILD_HANDLER watch watch sti verbose		/* #23 */
+#endif
+	BUILD_HANDLER mcheck mcheck cli verbose		/* #24 */
+	BUILD_HANDLER mt mt sti silent			/* #25 */
+	BUILD_HANDLER dsp dsp sti silent		/* #26 */
+	BUILD_HANDLER reserved reserved sti verbose	/* others */
+
+	.align	5
+	LEAF(handle_ri_rdhwr_tlbp)
+	.set	push
+	.set	noat
+	.set	noreorder
+	/* check if TLB contains a entry for EPC */
+	MFC0	k1, CP0_ENTRYHI
+	andi	k1, MIPS_ENTRYHI_ASID | MIPS_ENTRYHI_ASIDX
+	MFC0	k0, CP0_EPC
+	PTR_SRL	k0, _PAGE_SHIFT + 1
+	PTR_SLL	k0, _PAGE_SHIFT + 1
+	or	k1, k0
+	MTC0	k1, CP0_ENTRYHI
+	mtc0_tlbw_hazard
+	tlbp
+	tlb_probe_hazard
+	mfc0	k1, CP0_INDEX
+	.set	pop
+	bltz	k1, handle_ri	/* slow path */
+	/* fall thru */
+	END(handle_ri_rdhwr_tlbp)
+
+	LEAF(handle_ri_rdhwr)
+	.set	push
+	.set	noat
+	.set	noreorder
+	/* MIPS32:    0x7c03e83b: rdhwr v1,$29 */
+	/* microMIPS: 0x007d6b3c: rdhwr v1,$29 */
+	MFC0	k1, CP0_EPC
+#if defined(CONFIG_CPU_MICROMIPS) || defined(CONFIG_CPU_MIPS32_R2) || defined(CONFIG_CPU_MIPS64_R2)
+	and	k0, k1, 1
+	beqz	k0, 1f
+	 xor	k1, k0
+	lhu	k0, (k1)
+	lhu	k1, 2(k1)
+	ins	k1, k0, 16, 16
+	lui	k0, 0x007d
+	b	docheck
+	 ori	k0, 0x6b3c
+1:
+	lui	k0, 0x7c03
+	lw	k1, (k1)
+	ori	k0, 0xe83b
+#else
+	andi	k0, k1, 1
+	bnez	k0, handle_ri
+	 lui	k0, 0x7c03
+	lw	k1, (k1)
+	ori	k0, 0xe83b
+#endif
+	.set	reorder
+docheck:
+	bne	k0, k1, handle_ri	/* if not ours */
+
+isrdhwr:
+	/* The insn is rdhwr.  No need to check CAUSE.BD here. */
+	get_saved_sp	/* k1 := current_thread_info */
+	.set	noreorder
+	MFC0	k0, CP0_EPC
+#if defined(CONFIG_CPU_R3000) || defined(CONFIG_CPU_TX39XX)
+	ori	k1, _THREAD_MASK
+	xori	k1, _THREAD_MASK
+	LONG_L	v1, TI_TP_VALUE(k1)
+	LONG_ADDIU	k0, 4
+	jr	k0
+	 rfe
+#else
+#ifndef CONFIG_CPU_DADDI_WORKAROUNDS
+	LONG_ADDIU	k0, 4		/* stall on $k0 */
+#else
+	.set	at=v1
+	LONG_ADDIU	k0, 4
+	.set	noat
+#endif
+	MTC0	k0, CP0_EPC
+	/* I hope three instructions between MTC0 and ERET are enough... */
+	ori	k1, _THREAD_MASK
+	xori	k1, _THREAD_MASK
+	LONG_L	v1, TI_TP_VALUE(k1)
+	.set	push
+	.set	arch=r4000
+	eret
+	.set	pop
+#endif
+	.set	pop
+	END(handle_ri_rdhwr)
+
+#ifdef CONFIG_CPU_R4X00_BUGS64
+/* A temporary overflow handler used by check_daddi(). */
+
+	__INIT
+
+	BUILD_HANDLER  daddi_ov daddi_ov none silent	/* #12 */
+#endif
diff --git a/arch/mips/kernel/gpio_txx9.c b/arch/mips/kernel/gpio_txx9.c
new file mode 100644
index 000000000..8c083612d
--- /dev/null
+++ b/arch/mips/kernel/gpio_txx9.c
@@ -0,0 +1,86 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * A gpio chip driver for TXx9 SoCs
+ *
+ * Copyright (C) 2008 Atsushi Nemoto <anemo@mba.ocn.ne.jp>
+ */
+
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/gpio/driver.h>
+#include <linux/errno.h>
+#include <linux/io.h>
+#include <asm/txx9pio.h>
+
+static DEFINE_SPINLOCK(txx9_gpio_lock);
+
+static struct txx9_pio_reg __iomem *txx9_pioptr;
+
+static int txx9_gpio_get(struct gpio_chip *chip, unsigned int offset)
+{
+	return !!(__raw_readl(&txx9_pioptr->din) & (1 << offset));
+}
+
+static void txx9_gpio_set_raw(unsigned int offset, int value)
+{
+	u32 val;
+	val = __raw_readl(&txx9_pioptr->dout);
+	if (value)
+		val |= 1 << offset;
+	else
+		val &= ~(1 << offset);
+	__raw_writel(val, &txx9_pioptr->dout);
+}
+
+static void txx9_gpio_set(struct gpio_chip *chip, unsigned int offset,
+			  int value)
+{
+	unsigned long flags;
+	spin_lock_irqsave(&txx9_gpio_lock, flags);
+	txx9_gpio_set_raw(offset, value);
+	mmiowb();
+	spin_unlock_irqrestore(&txx9_gpio_lock, flags);
+}
+
+static int txx9_gpio_dir_in(struct gpio_chip *chip, unsigned int offset)
+{
+	unsigned long flags;
+	spin_lock_irqsave(&txx9_gpio_lock, flags);
+	__raw_writel(__raw_readl(&txx9_pioptr->dir) & ~(1 << offset),
+		     &txx9_pioptr->dir);
+	mmiowb();
+	spin_unlock_irqrestore(&txx9_gpio_lock, flags);
+	return 0;
+}
+
+static int txx9_gpio_dir_out(struct gpio_chip *chip, unsigned int offset,
+			     int value)
+{
+	unsigned long flags;
+	spin_lock_irqsave(&txx9_gpio_lock, flags);
+	txx9_gpio_set_raw(offset, value);
+	__raw_writel(__raw_readl(&txx9_pioptr->dir) | (1 << offset),
+		     &txx9_pioptr->dir);
+	mmiowb();
+	spin_unlock_irqrestore(&txx9_gpio_lock, flags);
+	return 0;
+}
+
+static struct gpio_chip txx9_gpio_chip = {
+	.get = txx9_gpio_get,
+	.set = txx9_gpio_set,
+	.direction_input = txx9_gpio_dir_in,
+	.direction_output = txx9_gpio_dir_out,
+	.label = "TXx9",
+};
+
+int __init txx9_gpio_init(unsigned long baseaddr,
+			  unsigned int base, unsigned int num)
+{
+	txx9_pioptr = ioremap(baseaddr, sizeof(struct txx9_pio_reg));
+	if (!txx9_pioptr)
+		return -ENODEV;
+	txx9_gpio_chip.base = base;
+	txx9_gpio_chip.ngpio = num;
+	return gpiochip_add_data(&txx9_gpio_chip, NULL);
+}
diff --git a/arch/mips/kernel/head.S b/arch/mips/kernel/head.S
new file mode 100644
index 000000000..61b73580b
--- /dev/null
+++ b/arch/mips/kernel/head.S
@@ -0,0 +1,185 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1994, 1995 Waldorf Electronics
+ * Written by Ralf Baechle and Andreas Busse
+ * Copyright (C) 1994 - 99, 2003, 06 Ralf Baechle
+ * Copyright (C) 1996 Paul M. Antoine
+ * Modified for DECStation and hence R3000 support by Paul M. Antoine
+ * Further modifications by David S. Miller and Harald Koerfgen
+ * Copyright (C) 1999 Silicon Graphics, Inc.
+ * Kevin Kissell, kevink@mips.com and Carsten Langgaard, carstenl@mips.com
+ * Copyright (C) 2000 MIPS Technologies, Inc.  All rights reserved.
+ */
+#include <linux/init.h>
+#include <linux/threads.h>
+
+#include <asm/addrspace.h>
+#include <asm/asm.h>
+#include <asm/asmmacro.h>
+#include <asm/irqflags.h>
+#include <asm/regdef.h>
+#include <asm/mipsregs.h>
+#include <asm/stackframe.h>
+
+#include <kernel-entry-init.h>
+
+	/*
+	 * For the moment disable interrupts, mark the kernel mode and
+	 * set ST0_KX so that the CPU does not spit fire when using
+	 * 64-bit addresses.  A full initialization of the CPU's status
+	 * register is done later in per_cpu_trap_init().
+	 */
+	.macro	setup_c0_status set clr
+	.set	push
+	mfc0	t0, CP0_STATUS
+	or	t0, ST0_KERNEL_CUMASK|\set|0x1f|\clr
+	xor	t0, 0x1f|\clr
+	mtc0	t0, CP0_STATUS
+	.set	noreorder
+	sll	zero,3				# ehb
+	.set	pop
+	.endm
+
+	.macro	setup_c0_status_pri
+#ifdef CONFIG_64BIT
+	setup_c0_status ST0_KX 0
+#else
+	setup_c0_status 0 0
+#endif
+	.endm
+
+	.macro	setup_c0_status_sec
+#ifdef CONFIG_64BIT
+	setup_c0_status ST0_KX ST0_BEV
+#else
+	setup_c0_status 0 ST0_BEV
+#endif
+	.endm
+
+#ifndef CONFIG_NO_EXCEPT_FILL
+	/*
+	 * Reserved space for exception handlers.
+	 * Necessary for machines which link their kernels at KSEG0.
+	 */
+	.fill	0x400
+#endif
+
+EXPORT(_stext)
+
+#ifdef CONFIG_BOOT_RAW
+	/*
+	 * Give us a fighting chance of running if execution beings at the
+	 * kernel load address.	 This is needed because this platform does
+	 * not have a ELF loader yet.
+	 */
+FEXPORT(__kernel_entry)
+	j	kernel_entry
+#endif /* CONFIG_BOOT_RAW */
+
+	__REF
+
+NESTED(kernel_entry, 16, sp)			# kernel entry point
+
+	kernel_entry_setup			# cpu specific setup
+
+	setup_c0_status_pri
+
+	/* We might not get launched at the address the kernel is linked to,
+	   so we jump there.  */
+	PTR_LA	t0, 0f
+	jr	t0
+0:
+
+#ifdef CONFIG_USE_OF
+#if defined(CONFIG_MIPS_RAW_APPENDED_DTB) || \
+	defined(CONFIG_MIPS_ELF_APPENDED_DTB)
+
+	PTR_LA		t2, __appended_dtb
+
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	li		t1, 0xd00dfeed
+#else  /* !CONFIG_CPU_BIG_ENDIAN */
+	li		t1, 0xedfe0dd0
+#endif /* !CONFIG_CPU_BIG_ENDIAN */
+	lw		t0, (t2)
+	beq		t0, t1, dtb_found
+#endif /* CONFIG_MIPS_RAW_APPENDED_DTB || CONFIG_MIPS_ELF_APPENDED_DTB */
+	li		t1, -2
+	move		t2, a1
+	beq		a0, t1, dtb_found
+
+#ifdef CONFIG_BUILTIN_DTB
+	PTR_LA	t2, __dtb_start
+	PTR_LA	t1, __dtb_end
+	bne		t1, t2, dtb_found
+#endif /* CONFIG_BUILTIN_DTB */
+
+	li		t2, 0
+dtb_found:
+#endif /* CONFIG_USE_OF */
+	PTR_LA		t0, __bss_start		# clear .bss
+	LONG_S		zero, (t0)
+	PTR_LA		t1, __bss_stop - LONGSIZE
+1:
+	PTR_ADDIU	t0, LONGSIZE
+	LONG_S		zero, (t0)
+	bne		t0, t1, 1b
+
+	LONG_S		a0, fw_arg0		# firmware arguments
+	LONG_S		a1, fw_arg1
+	LONG_S		a2, fw_arg2
+	LONG_S		a3, fw_arg3
+
+#ifdef CONFIG_USE_OF
+	LONG_S		t2, fw_passed_dtb
+#endif
+
+	MTC0		zero, CP0_CONTEXT	# clear context register
+#ifdef CONFIG_64BIT
+	MTC0		zero, CP0_XCONTEXT
+#endif
+	PTR_LA		$28, init_thread_union
+	/* Set the SP after an empty pt_regs.  */
+	PTR_LI		sp, _THREAD_SIZE - 32 - PT_SIZE
+	PTR_ADDU	sp, $28
+	back_to_back_c0_hazard
+	set_saved_sp	sp, t0, t1
+	PTR_SUBU	sp, 4 * SZREG		# init stack pointer
+
+#ifdef CONFIG_RELOCATABLE
+	/* Copy kernel and apply the relocations */
+	jal		relocate_kernel
+
+	/* Repoint the sp into the new kernel image */
+	PTR_LI		sp, _THREAD_SIZE - 32 - PT_SIZE
+	PTR_ADDU	sp, $28
+	set_saved_sp	sp, t0, t1
+	PTR_SUBU	sp, 4 * SZREG		# init stack pointer
+
+	/*
+	 * relocate_kernel returns the entry point either
+	 * in the relocated kernel or the original if for
+	 * some reason relocation failed - jump there now
+	 * with instruction hazard barrier because of the
+	 * newly sync'd icache.
+	 */
+	jr.hb		v0
+#else  /* !CONFIG_RELOCATABLE */
+	j		start_kernel
+#endif /* !CONFIG_RELOCATABLE */
+	END(kernel_entry)
+
+#ifdef CONFIG_SMP
+/*
+ * SMP slave cpus entry point.	Board specific code for bootstrap calls this
+ * function after setting up the stack and gp registers.
+ */
+NESTED(smp_bootstrap, 16, sp)
+	smp_slave_setup
+	setup_c0_status_sec
+	j	start_secondary
+	END(smp_bootstrap)
+#endif /* CONFIG_SMP */
diff --git a/arch/mips/kernel/i8253.c b/arch/mips/kernel/i8253.c
new file mode 100644
index 000000000..ca21210e0
--- /dev/null
+++ b/arch/mips/kernel/i8253.c
@@ -0,0 +1,38 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * i8253.c  8253/PIT functions
+ *
+ */
+#include <linux/clockchips.h>
+#include <linux/i8253.h>
+#include <linux/export.h>
+#include <linux/smp.h>
+#include <linux/irq.h>
+
+#include <asm/time.h>
+
+static irqreturn_t timer_interrupt(int irq, void *dev_id)
+{
+	i8253_clockevent.event_handler(&i8253_clockevent);
+
+	return IRQ_HANDLED;
+}
+
+void __init setup_pit_timer(void)
+{
+	unsigned long flags = IRQF_NOBALANCING | IRQF_TIMER;
+
+	clockevent_i8253_init(true);
+	if (request_irq(0, timer_interrupt, flags, "timer", NULL))
+		pr_err("Failed to request irq 0 (timer)\n");
+}
+
+static int __init init_pit_clocksource(void)
+{
+	if (num_possible_cpus() > 1 || /* PIT does not scale! */
+	    !clockevent_state_periodic(&i8253_clockevent))
+		return 0;
+
+	return clocksource_i8253_init();
+}
+arch_initcall(init_pit_clocksource);
diff --git a/arch/mips/kernel/idle.c b/arch/mips/kernel/idle.c
new file mode 100644
index 000000000..18e69ebf5
--- /dev/null
+++ b/arch/mips/kernel/idle.c
@@ -0,0 +1,272 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * MIPS idle loop and WAIT instruction support.
+ *
+ * Copyright (C) xxxx  the Anonymous
+ * Copyright (C) 1994 - 2006 Ralf Baechle
+ * Copyright (C) 2003, 2004  Maciej W. Rozycki
+ * Copyright (C) 2001, 2004, 2011, 2012	 MIPS Technologies, Inc.
+ */
+#include <linux/cpu.h>
+#include <linux/export.h>
+#include <linux/init.h>
+#include <linux/irqflags.h>
+#include <linux/printk.h>
+#include <linux/sched.h>
+#include <asm/cpu.h>
+#include <asm/cpu-info.h>
+#include <asm/cpu-type.h>
+#include <asm/idle.h>
+#include <asm/mipsregs.h>
+
+/*
+ * Not all of the MIPS CPUs have the "wait" instruction available. Moreover,
+ * the implementation of the "wait" feature differs between CPU families. This
+ * points to the function that implements CPU specific wait.
+ * The wait instruction stops the pipeline and reduces the power consumption of
+ * the CPU very much.
+ */
+void (*cpu_wait)(void);
+EXPORT_SYMBOL(cpu_wait);
+
+static void __cpuidle r3081_wait(void)
+{
+	unsigned long cfg = read_c0_conf();
+	write_c0_conf(cfg | R30XX_CONF_HALT);
+	raw_local_irq_enable();
+}
+
+static void __cpuidle r39xx_wait(void)
+{
+	if (!need_resched())
+		write_c0_conf(read_c0_conf() | TX39_CONF_HALT);
+	raw_local_irq_enable();
+}
+
+void __cpuidle r4k_wait(void)
+{
+	raw_local_irq_enable();
+	__r4k_wait();
+}
+
+/*
+ * This variant is preferable as it allows testing need_resched and going to
+ * sleep depending on the outcome atomically.  Unfortunately the "It is
+ * implementation-dependent whether the pipeline restarts when a non-enabled
+ * interrupt is requested" restriction in the MIPS32/MIPS64 architecture makes
+ * using this version a gamble.
+ */
+void __cpuidle r4k_wait_irqoff(void)
+{
+	if (!need_resched())
+		__asm__(
+		"	.set	push		\n"
+		"	.set	arch=r4000	\n"
+		"	wait			\n"
+		"	.set	pop		\n");
+	raw_local_irq_enable();
+}
+
+/*
+ * The RM7000 variant has to handle erratum 38.	 The workaround is to not
+ * have any pending stores when the WAIT instruction is executed.
+ */
+static void __cpuidle rm7k_wait_irqoff(void)
+{
+	if (!need_resched())
+		__asm__(
+		"	.set	push					\n"
+		"	.set	arch=r4000				\n"
+		"	.set	noat					\n"
+		"	mfc0	$1, $12					\n"
+		"	sync						\n"
+		"	mtc0	$1, $12		# stalls until W stage	\n"
+		"	wait						\n"
+		"	mtc0	$1, $12		# stalls until W stage	\n"
+		"	.set	pop					\n");
+	raw_local_irq_enable();
+}
+
+/*
+ * Au1 'wait' is only useful when the 32kHz counter is used as timer,
+ * since coreclock (and the cp0 counter) stops upon executing it. Only an
+ * interrupt can wake it, so they must be enabled before entering idle modes.
+ */
+static void __cpuidle au1k_wait(void)
+{
+	unsigned long c0status = read_c0_status() | 1;	/* irqs on */
+
+	__asm__(
+	"	.set	push			\n"
+	"	.set	arch=r4000		\n"
+	"	cache	0x14, 0(%0)		\n"
+	"	cache	0x14, 32(%0)		\n"
+	"	sync				\n"
+	"	mtc0	%1, $12			\n" /* wr c0status */
+	"	wait				\n"
+	"	nop				\n"
+	"	nop				\n"
+	"	nop				\n"
+	"	nop				\n"
+	"	.set	pop			\n"
+	: : "r" (au1k_wait), "r" (c0status));
+}
+
+static int __initdata nowait;
+
+static int __init wait_disable(char *s)
+{
+	nowait = 1;
+
+	return 1;
+}
+
+__setup("nowait", wait_disable);
+
+void __init check_wait(void)
+{
+	struct cpuinfo_mips *c = &current_cpu_data;
+
+	if (nowait) {
+		printk("Wait instruction disabled.\n");
+		return;
+	}
+
+	/*
+	 * MIPSr6 specifies that masked interrupts should unblock an executing
+	 * wait instruction, and thus that it is safe for us to use
+	 * r4k_wait_irqoff. Yippee!
+	 */
+	if (cpu_has_mips_r6) {
+		cpu_wait = r4k_wait_irqoff;
+		return;
+	}
+
+	switch (current_cpu_type()) {
+	case CPU_R3081:
+	case CPU_R3081E:
+		cpu_wait = r3081_wait;
+		break;
+	case CPU_TX3927:
+		cpu_wait = r39xx_wait;
+		break;
+	case CPU_R4200:
+	case CPU_R4600:
+	case CPU_R4640:
+	case CPU_R4650:
+	case CPU_R4700:
+	case CPU_R5000:
+	case CPU_R5500:
+	case CPU_NEVADA:
+	case CPU_4KC:
+	case CPU_4KEC:
+	case CPU_4KSC:
+	case CPU_5KC:
+	case CPU_5KE:
+	case CPU_25KF:
+	case CPU_PR4450:
+	case CPU_BMIPS3300:
+	case CPU_BMIPS4350:
+	case CPU_BMIPS4380:
+	case CPU_CAVIUM_OCTEON:
+	case CPU_CAVIUM_OCTEON_PLUS:
+	case CPU_CAVIUM_OCTEON2:
+	case CPU_CAVIUM_OCTEON3:
+	case CPU_XBURST:
+	case CPU_LOONGSON32:
+	case CPU_XLR:
+	case CPU_XLP:
+		cpu_wait = r4k_wait;
+		break;
+	case CPU_LOONGSON64:
+		if ((c->processor_id & (PRID_IMP_MASK | PRID_REV_MASK)) >=
+				(PRID_IMP_LOONGSON_64C | PRID_REV_LOONGSON3A_R2_0) ||
+				(c->processor_id & PRID_IMP_MASK) == PRID_IMP_LOONGSON_64R)
+			cpu_wait = r4k_wait;
+		break;
+
+	case CPU_BMIPS5000:
+		cpu_wait = r4k_wait_irqoff;
+		break;
+	case CPU_RM7000:
+		cpu_wait = rm7k_wait_irqoff;
+		break;
+
+	case CPU_PROAPTIV:
+	case CPU_P5600:
+		/*
+		 * Incoming Fast Debug Channel (FDC) data during a wait
+		 * instruction causes the wait never to resume, even if an
+		 * interrupt is received. Avoid using wait at all if FDC data is
+		 * likely to be received.
+		 */
+		if (IS_ENABLED(CONFIG_MIPS_EJTAG_FDC_TTY))
+			break;
+		fallthrough;
+	case CPU_M14KC:
+	case CPU_M14KEC:
+	case CPU_24K:
+	case CPU_34K:
+	case CPU_1004K:
+	case CPU_1074K:
+	case CPU_INTERAPTIV:
+	case CPU_M5150:
+	case CPU_QEMU_GENERIC:
+		cpu_wait = r4k_wait;
+		if (read_c0_config7() & MIPS_CONF7_WII)
+			cpu_wait = r4k_wait_irqoff;
+		break;
+
+	case CPU_74K:
+		cpu_wait = r4k_wait;
+		if ((c->processor_id & 0xff) >= PRID_REV_ENCODE_332(2, 1, 0))
+			cpu_wait = r4k_wait_irqoff;
+		break;
+
+	case CPU_TX49XX:
+		cpu_wait = r4k_wait_irqoff;
+		break;
+	case CPU_ALCHEMY:
+		cpu_wait = au1k_wait;
+		break;
+	case CPU_20KC:
+		/*
+		 * WAIT on Rev1.0 has E1, E2, E3 and E16.
+		 * WAIT on Rev2.0 and Rev3.0 has E16.
+		 * Rev3.1 WAIT is nop, why bother
+		 */
+		if ((c->processor_id & 0xff) <= 0x64)
+			break;
+
+		/*
+		 * Another rev is incremeting c0_count at a reduced clock
+		 * rate while in WAIT mode.  So we basically have the choice
+		 * between using the cp0 timer as clocksource or avoiding
+		 * the WAIT instruction.  Until more details are known,
+		 * disable the use of WAIT for 20Kc entirely.
+		   cpu_wait = r4k_wait;
+		 */
+		break;
+	default:
+		break;
+	}
+}
+
+void arch_cpu_idle(void)
+{
+	if (cpu_wait)
+		cpu_wait();
+	else
+		raw_local_irq_enable();
+}
+
+#ifdef CONFIG_CPU_IDLE
+
+int mips_cpuidle_wait_enter(struct cpuidle_device *dev,
+			    struct cpuidle_driver *drv, int index)
+{
+	arch_cpu_idle();
+	return index;
+}
+
+#endif
diff --git a/arch/mips/kernel/irq-gt641xx.c b/arch/mips/kernel/irq-gt641xx.c
new file mode 100644
index 000000000..93bcf5736
--- /dev/null
+++ b/arch/mips/kernel/irq-gt641xx.c
@@ -0,0 +1,118 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ *  GT641xx IRQ routines.
+ *
+ *  Copyright (C) 2007	Yoichi Yuasa <yuasa@linux-mips.org>
+ */
+#include <linux/hardirq.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <linux/spinlock.h>
+#include <linux/types.h>
+
+#include <asm/gt64120.h>
+
+#define GT641XX_IRQ_TO_BIT(irq) (1U << (irq - GT641XX_IRQ_BASE))
+
+static DEFINE_RAW_SPINLOCK(gt641xx_irq_lock);
+
+static void ack_gt641xx_irq(struct irq_data *d)
+{
+	unsigned long flags;
+	u32 cause;
+
+	raw_spin_lock_irqsave(&gt641xx_irq_lock, flags);
+	cause = GT_READ(GT_INTRCAUSE_OFS);
+	cause &= ~GT641XX_IRQ_TO_BIT(d->irq);
+	GT_WRITE(GT_INTRCAUSE_OFS, cause);
+	raw_spin_unlock_irqrestore(&gt641xx_irq_lock, flags);
+}
+
+static void mask_gt641xx_irq(struct irq_data *d)
+{
+	unsigned long flags;
+	u32 mask;
+
+	raw_spin_lock_irqsave(&gt641xx_irq_lock, flags);
+	mask = GT_READ(GT_INTRMASK_OFS);
+	mask &= ~GT641XX_IRQ_TO_BIT(d->irq);
+	GT_WRITE(GT_INTRMASK_OFS, mask);
+	raw_spin_unlock_irqrestore(&gt641xx_irq_lock, flags);
+}
+
+static void mask_ack_gt641xx_irq(struct irq_data *d)
+{
+	unsigned long flags;
+	u32 cause, mask;
+
+	raw_spin_lock_irqsave(&gt641xx_irq_lock, flags);
+	mask = GT_READ(GT_INTRMASK_OFS);
+	mask &= ~GT641XX_IRQ_TO_BIT(d->irq);
+	GT_WRITE(GT_INTRMASK_OFS, mask);
+
+	cause = GT_READ(GT_INTRCAUSE_OFS);
+	cause &= ~GT641XX_IRQ_TO_BIT(d->irq);
+	GT_WRITE(GT_INTRCAUSE_OFS, cause);
+	raw_spin_unlock_irqrestore(&gt641xx_irq_lock, flags);
+}
+
+static void unmask_gt641xx_irq(struct irq_data *d)
+{
+	unsigned long flags;
+	u32 mask;
+
+	raw_spin_lock_irqsave(&gt641xx_irq_lock, flags);
+	mask = GT_READ(GT_INTRMASK_OFS);
+	mask |= GT641XX_IRQ_TO_BIT(d->irq);
+	GT_WRITE(GT_INTRMASK_OFS, mask);
+	raw_spin_unlock_irqrestore(&gt641xx_irq_lock, flags);
+}
+
+static struct irq_chip gt641xx_irq_chip = {
+	.name		= "GT641xx",
+	.irq_ack	= ack_gt641xx_irq,
+	.irq_mask	= mask_gt641xx_irq,
+	.irq_mask_ack	= mask_ack_gt641xx_irq,
+	.irq_unmask	= unmask_gt641xx_irq,
+};
+
+void gt641xx_irq_dispatch(void)
+{
+	u32 cause, mask;
+	int i;
+
+	cause = GT_READ(GT_INTRCAUSE_OFS);
+	mask = GT_READ(GT_INTRMASK_OFS);
+	cause &= mask;
+
+	/*
+	 * bit0 : logical or of all the interrupt bits.
+	 * bit30: logical or of bits[29:26,20:1].
+	 * bit31: logical or of bits[25:1].
+	 */
+	for (i = 1; i < 30; i++) {
+		if (cause & (1U << i)) {
+			do_IRQ(GT641XX_IRQ_BASE + i);
+			return;
+		}
+	}
+
+	atomic_inc(&irq_err_count);
+}
+
+void __init gt641xx_irq_init(void)
+{
+	int i;
+
+	GT_WRITE(GT_INTRMASK_OFS, 0);
+	GT_WRITE(GT_INTRCAUSE_OFS, 0);
+
+	/*
+	 * bit0 : logical or of all the interrupt bits.
+	 * bit30: logical or of bits[29:26,20:1].
+	 * bit31: logical or of bits[25:1].
+	 */
+	for (i = 1; i < 30; i++)
+		irq_set_chip_and_handler(GT641XX_IRQ_BASE + i,
+					 &gt641xx_irq_chip, handle_level_irq);
+}
diff --git a/arch/mips/kernel/irq-msc01.c b/arch/mips/kernel/irq-msc01.c
new file mode 100644
index 000000000..ab511b64a
--- /dev/null
+++ b/arch/mips/kernel/irq-msc01.c
@@ -0,0 +1,156 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ *
+ * Copyright (c) 2004 MIPS Inc
+ * Author: chris@mips.com
+ *
+ * Copyright (C) 2004, 06 Ralf Baechle <ralf@linux-mips.org>
+ */
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/kernel_stat.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/msc01_ic.h>
+#include <asm/traps.h>
+
+static unsigned long _icctrl_msc;
+#define MSC01_IC_REG_BASE	_icctrl_msc
+
+#define MSCIC_WRITE(reg, data)	do { *(volatile u32 *)(reg) = data; } while (0)
+#define MSCIC_READ(reg, data)	do { data = *(volatile u32 *)(reg); } while (0)
+
+static unsigned int irq_base;
+
+/* mask off an interrupt */
+static inline void mask_msc_irq(struct irq_data *d)
+{
+	unsigned int irq = d->irq;
+
+	if (irq < (irq_base + 32))
+		MSCIC_WRITE(MSC01_IC_DISL, 1<<(irq - irq_base));
+	else
+		MSCIC_WRITE(MSC01_IC_DISH, 1<<(irq - irq_base - 32));
+}
+
+/* unmask an interrupt */
+static inline void unmask_msc_irq(struct irq_data *d)
+{
+	unsigned int irq = d->irq;
+
+	if (irq < (irq_base + 32))
+		MSCIC_WRITE(MSC01_IC_ENAL, 1<<(irq - irq_base));
+	else
+		MSCIC_WRITE(MSC01_IC_ENAH, 1<<(irq - irq_base - 32));
+}
+
+/*
+ * Masks and ACKs an IRQ
+ */
+static void level_mask_and_ack_msc_irq(struct irq_data *d)
+{
+	mask_msc_irq(d);
+	if (!cpu_has_veic)
+		MSCIC_WRITE(MSC01_IC_EOI, 0);
+}
+
+/*
+ * Masks and ACKs an IRQ
+ */
+static void edge_mask_and_ack_msc_irq(struct irq_data *d)
+{
+	unsigned int irq = d->irq;
+
+	mask_msc_irq(d);
+	if (!cpu_has_veic)
+		MSCIC_WRITE(MSC01_IC_EOI, 0);
+	else {
+		u32 r;
+		MSCIC_READ(MSC01_IC_SUP+irq*8, r);
+		MSCIC_WRITE(MSC01_IC_SUP+irq*8, r | ~MSC01_IC_SUP_EDGE_BIT);
+		MSCIC_WRITE(MSC01_IC_SUP+irq*8, r);
+	}
+}
+
+/*
+ * Interrupt handler for interrupts coming from SOC-it.
+ */
+void ll_msc_irq(void)
+{
+	unsigned int irq;
+
+	/* read the interrupt vector register */
+	MSCIC_READ(MSC01_IC_VEC, irq);
+	if (irq < 64)
+		do_IRQ(irq + irq_base);
+	else {
+		/* Ignore spurious interrupt */
+	}
+}
+
+static void msc_bind_eic_interrupt(int irq, int set)
+{
+	MSCIC_WRITE(MSC01_IC_RAMW,
+		    (irq<<MSC01_IC_RAMW_ADDR_SHF) | (set<<MSC01_IC_RAMW_DATA_SHF));
+}
+
+static struct irq_chip msc_levelirq_type = {
+	.name = "SOC-it-Level",
+	.irq_ack = level_mask_and_ack_msc_irq,
+	.irq_mask = mask_msc_irq,
+	.irq_mask_ack = level_mask_and_ack_msc_irq,
+	.irq_unmask = unmask_msc_irq,
+	.irq_eoi = unmask_msc_irq,
+};
+
+static struct irq_chip msc_edgeirq_type = {
+	.name = "SOC-it-Edge",
+	.irq_ack = edge_mask_and_ack_msc_irq,
+	.irq_mask = mask_msc_irq,
+	.irq_mask_ack = edge_mask_and_ack_msc_irq,
+	.irq_unmask = unmask_msc_irq,
+	.irq_eoi = unmask_msc_irq,
+};
+
+
+void __init init_msc_irqs(unsigned long icubase, unsigned int irqbase, msc_irqmap_t *imp, int nirq)
+{
+	_icctrl_msc = (unsigned long) ioremap(icubase, 0x40000);
+
+	/* Reset interrupt controller - initialises all registers to 0 */
+	MSCIC_WRITE(MSC01_IC_RST, MSC01_IC_RST_RST_BIT);
+
+	board_bind_eic_interrupt = &msc_bind_eic_interrupt;
+
+	for (; nirq > 0; nirq--, imp++) {
+		int n = imp->im_irq;
+
+		switch (imp->im_type) {
+		case MSC01_IRQ_EDGE:
+			irq_set_chip_and_handler_name(irqbase + n,
+						      &msc_edgeirq_type,
+						      handle_edge_irq,
+						      "edge");
+			if (cpu_has_veic)
+				MSCIC_WRITE(MSC01_IC_SUP+n*8, MSC01_IC_SUP_EDGE_BIT);
+			else
+				MSCIC_WRITE(MSC01_IC_SUP+n*8, MSC01_IC_SUP_EDGE_BIT | imp->im_lvl);
+			break;
+		case MSC01_IRQ_LEVEL:
+			irq_set_chip_and_handler_name(irqbase + n,
+						      &msc_levelirq_type,
+						      handle_level_irq,
+						      "level");
+			if (cpu_has_veic)
+				MSCIC_WRITE(MSC01_IC_SUP+n*8, 0);
+			else
+				MSCIC_WRITE(MSC01_IC_SUP+n*8, imp->im_lvl);
+		}
+	}
+
+	irq_base = irqbase;
+
+	MSCIC_WRITE(MSC01_IC_GENA, MSC01_IC_GENA_GENA_BIT);	/* Enable interrupt generation */
+
+}
diff --git a/arch/mips/kernel/irq-rm7000.c b/arch/mips/kernel/irq-rm7000.c
new file mode 100644
index 000000000..e1a497f63
--- /dev/null
+++ b/arch/mips/kernel/irq-rm7000.c
@@ -0,0 +1,45 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2003 Ralf Baechle
+ *
+ * Handler for RM7000 extended interrupts.  These are a non-standard
+ * feature so we handle them separately from standard interrupts.
+ */
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/kernel.h>
+
+#include <asm/irq_cpu.h>
+#include <asm/mipsregs.h>
+
+static inline void unmask_rm7k_irq(struct irq_data *d)
+{
+	set_c0_intcontrol(0x100 << (d->irq - RM7K_CPU_IRQ_BASE));
+}
+
+static inline void mask_rm7k_irq(struct irq_data *d)
+{
+	clear_c0_intcontrol(0x100 << (d->irq - RM7K_CPU_IRQ_BASE));
+}
+
+static struct irq_chip rm7k_irq_controller = {
+	.name = "RM7000",
+	.irq_ack = mask_rm7k_irq,
+	.irq_mask = mask_rm7k_irq,
+	.irq_mask_ack = mask_rm7k_irq,
+	.irq_unmask = unmask_rm7k_irq,
+	.irq_eoi = unmask_rm7k_irq
+};
+
+void __init rm7k_cpu_irq_init(void)
+{
+	int base = RM7K_CPU_IRQ_BASE;
+	int i;
+
+	clear_c0_intcontrol(0x00000f00);		/* Mask all */
+
+	for (i = base; i < base + 4; i++)
+		irq_set_chip_and_handler(i, &rm7k_irq_controller,
+					 handle_percpu_irq);
+}
diff --git a/arch/mips/kernel/irq.c b/arch/mips/kernel/irq.c
new file mode 100644
index 000000000..85b6c60f2
--- /dev/null
+++ b/arch/mips/kernel/irq.c
@@ -0,0 +1,109 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Code to handle x86 style IRQs plus some generic interrupt stuff.
+ *
+ * Copyright (C) 1992 Linus Torvalds
+ * Copyright (C) 1994 - 2000 Ralf Baechle
+ */
+#include <linux/kernel.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/kernel_stat.h>
+#include <linux/proc_fs.h>
+#include <linux/mm.h>
+#include <linux/random.h>
+#include <linux/sched.h>
+#include <linux/seq_file.h>
+#include <linux/kallsyms.h>
+#include <linux/kgdb.h>
+#include <linux/ftrace.h>
+
+#include <linux/atomic.h>
+#include <linux/uaccess.h>
+
+void *irq_stack[NR_CPUS];
+
+/*
+ * 'what should we do if we get a hw irq event on an illegal vector'.
+ * each architecture has to answer this themselves.
+ */
+void ack_bad_irq(unsigned int irq)
+{
+	printk("unexpected IRQ # %d\n", irq);
+}
+
+atomic_t irq_err_count;
+
+int arch_show_interrupts(struct seq_file *p, int prec)
+{
+	seq_printf(p, "%*s: %10u\n", prec, "ERR", atomic_read(&irq_err_count));
+	return 0;
+}
+
+asmlinkage void spurious_interrupt(void)
+{
+	atomic_inc(&irq_err_count);
+}
+
+void __init init_IRQ(void)
+{
+	int i;
+	unsigned int order = get_order(IRQ_STACK_SIZE);
+
+	for (i = 0; i < NR_IRQS; i++)
+		irq_set_noprobe(i);
+
+	if (cpu_has_veic)
+		clear_c0_status(ST0_IM);
+
+	arch_init_irq();
+
+	for_each_possible_cpu(i) {
+		void *s = (void *)__get_free_pages(GFP_KERNEL, order);
+
+		irq_stack[i] = s;
+		pr_debug("CPU%d IRQ stack at 0x%p - 0x%p\n", i,
+			irq_stack[i], irq_stack[i] + IRQ_STACK_SIZE);
+	}
+}
+
+#ifdef CONFIG_DEBUG_STACKOVERFLOW
+static inline void check_stack_overflow(void)
+{
+	unsigned long sp;
+
+	__asm__ __volatile__("move %0, $sp" : "=r" (sp));
+	sp &= THREAD_MASK;
+
+	/*
+	 * Check for stack overflow: is there less than STACK_WARN free?
+	 * STACK_WARN is defined as 1/8 of THREAD_SIZE by default.
+	 */
+	if (unlikely(sp < (sizeof(struct thread_info) + STACK_WARN))) {
+		printk("do_IRQ: stack overflow: %ld\n",
+		       sp - sizeof(struct thread_info));
+		dump_stack();
+	}
+}
+#else
+static inline void check_stack_overflow(void) {}
+#endif
+
+
+/*
+ * do_IRQ handles all normal device IRQ's (the special
+ * SMP cross-CPU interrupts have their own specific
+ * handlers).
+ */
+void __irq_entry do_IRQ(unsigned int irq)
+{
+	irq_enter();
+	check_stack_overflow();
+	generic_handle_irq(irq);
+	irq_exit();
+}
+
diff --git a/arch/mips/kernel/irq_txx9.c b/arch/mips/kernel/irq_txx9.c
new file mode 100644
index 000000000..ab00e4904
--- /dev/null
+++ b/arch/mips/kernel/irq_txx9.c
@@ -0,0 +1,191 @@
+/*
+ * Based on linux/arch/mips/jmr3927/rbhma3100/irq.c,
+ *	    linux/arch/mips/tx4927/common/tx4927_irq.c,
+ *	    linux/arch/mips/tx4938/common/irq.c
+ *
+ * Copyright 2001, 2003-2005 MontaVista Software Inc.
+ * Author: MontaVista Software, Inc.
+ *	   ahennessy@mvista.com
+ *	   source@mvista.com
+ * Copyright (C) 2000-2001 Toshiba Corporation
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/types.h>
+#include <linux/irq.h>
+#include <asm/txx9irq.h>
+
+struct txx9_irc_reg {
+	u32 cer;
+	u32 cr[2];
+	u32 unused0;
+	u32 ilr[8];
+	u32 unused1[4];
+	u32 imr;
+	u32 unused2[7];
+	u32 scr;
+	u32 unused3[7];
+	u32 ssr;
+	u32 unused4[7];
+	u32 csr;
+};
+
+/* IRCER : Int. Control Enable */
+#define TXx9_IRCER_ICE	0x00000001
+
+/* IRCR : Int. Control */
+#define TXx9_IRCR_LOW	0x00000000
+#define TXx9_IRCR_HIGH	0x00000001
+#define TXx9_IRCR_DOWN	0x00000002
+#define TXx9_IRCR_UP	0x00000003
+#define TXx9_IRCR_EDGE(cr)	((cr) & 0x00000002)
+
+/* IRSCR : Int. Status Control */
+#define TXx9_IRSCR_EIClrE	0x00000100
+#define TXx9_IRSCR_EIClr_MASK	0x0000000f
+
+/* IRCSR : Int. Current Status */
+#define TXx9_IRCSR_IF	0x00010000
+#define TXx9_IRCSR_ILV_MASK	0x00000700
+#define TXx9_IRCSR_IVL_MASK	0x0000001f
+
+#define irc_dlevel	0
+#define irc_elevel	1
+
+static struct txx9_irc_reg __iomem *txx9_ircptr __read_mostly;
+
+static struct {
+	unsigned char level;
+	unsigned char mode;
+} txx9irq[TXx9_MAX_IR] __read_mostly;
+
+static void txx9_irq_unmask(struct irq_data *d)
+{
+	unsigned int irq_nr = d->irq - TXX9_IRQ_BASE;
+	u32 __iomem *ilrp = &txx9_ircptr->ilr[(irq_nr % 16 ) / 2];
+	int ofs = irq_nr / 16 * 16 + (irq_nr & 1) * 8;
+
+	__raw_writel((__raw_readl(ilrp) & ~(0xff << ofs))
+		     | (txx9irq[irq_nr].level << ofs),
+		     ilrp);
+#ifdef CONFIG_CPU_TX39XX
+	/* update IRCSR */
+	__raw_writel(0, &txx9_ircptr->imr);
+	__raw_writel(irc_elevel, &txx9_ircptr->imr);
+#endif
+}
+
+static inline void txx9_irq_mask(struct irq_data *d)
+{
+	unsigned int irq_nr = d->irq - TXX9_IRQ_BASE;
+	u32 __iomem *ilrp = &txx9_ircptr->ilr[(irq_nr % 16) / 2];
+	int ofs = irq_nr / 16 * 16 + (irq_nr & 1) * 8;
+
+	__raw_writel((__raw_readl(ilrp) & ~(0xff << ofs))
+		     | (irc_dlevel << ofs),
+		     ilrp);
+#ifdef CONFIG_CPU_TX39XX
+	/* update IRCSR */
+	__raw_writel(0, &txx9_ircptr->imr);
+	__raw_writel(irc_elevel, &txx9_ircptr->imr);
+	/* flush write buffer */
+	__raw_readl(&txx9_ircptr->ssr);
+#else
+	mmiowb();
+#endif
+}
+
+static void txx9_irq_mask_ack(struct irq_data *d)
+{
+	unsigned int irq_nr = d->irq - TXX9_IRQ_BASE;
+
+	txx9_irq_mask(d);
+	/* clear edge detection */
+	if (unlikely(TXx9_IRCR_EDGE(txx9irq[irq_nr].mode)))
+		__raw_writel(TXx9_IRSCR_EIClrE | irq_nr, &txx9_ircptr->scr);
+}
+
+static int txx9_irq_set_type(struct irq_data *d, unsigned int flow_type)
+{
+	unsigned int irq_nr = d->irq - TXX9_IRQ_BASE;
+	u32 cr;
+	u32 __iomem *crp;
+	int ofs;
+	int mode;
+
+	if (flow_type & IRQF_TRIGGER_PROBE)
+		return 0;
+	switch (flow_type & IRQF_TRIGGER_MASK) {
+	case IRQF_TRIGGER_RISING:	mode = TXx9_IRCR_UP;	break;
+	case IRQF_TRIGGER_FALLING:	mode = TXx9_IRCR_DOWN;	break;
+	case IRQF_TRIGGER_HIGH: mode = TXx9_IRCR_HIGH;	break;
+	case IRQF_TRIGGER_LOW:	mode = TXx9_IRCR_LOW;	break;
+	default:
+		return -EINVAL;
+	}
+	crp = &txx9_ircptr->cr[(unsigned int)irq_nr / 8];
+	cr = __raw_readl(crp);
+	ofs = (irq_nr & (8 - 1)) * 2;
+	cr &= ~(0x3 << ofs);
+	cr |= (mode & 0x3) << ofs;
+	__raw_writel(cr, crp);
+	txx9irq[irq_nr].mode = mode;
+	return 0;
+}
+
+static struct irq_chip txx9_irq_chip = {
+	.name		= "TXX9",
+	.irq_ack	= txx9_irq_mask_ack,
+	.irq_mask	= txx9_irq_mask,
+	.irq_mask_ack	= txx9_irq_mask_ack,
+	.irq_unmask	= txx9_irq_unmask,
+	.irq_set_type	= txx9_irq_set_type,
+};
+
+void __init txx9_irq_init(unsigned long baseaddr)
+{
+	int i;
+
+	txx9_ircptr = ioremap(baseaddr, sizeof(struct txx9_irc_reg));
+	for (i = 0; i < TXx9_MAX_IR; i++) {
+		txx9irq[i].level = 4; /* middle level */
+		txx9irq[i].mode = TXx9_IRCR_LOW;
+		irq_set_chip_and_handler(TXX9_IRQ_BASE + i, &txx9_irq_chip,
+					 handle_level_irq);
+	}
+
+	/* mask all IRC interrupts */
+	__raw_writel(0, &txx9_ircptr->imr);
+	for (i = 0; i < 8; i++)
+		__raw_writel(0, &txx9_ircptr->ilr[i]);
+	/* setup IRC interrupt mode (Low Active) */
+	for (i = 0; i < 2; i++)
+		__raw_writel(0, &txx9_ircptr->cr[i]);
+	/* enable interrupt control */
+	__raw_writel(TXx9_IRCER_ICE, &txx9_ircptr->cer);
+	__raw_writel(irc_elevel, &txx9_ircptr->imr);
+}
+
+int __init txx9_irq_set_pri(int irc_irq, int new_pri)
+{
+	int old_pri;
+
+	if ((unsigned int)irc_irq >= TXx9_MAX_IR)
+		return 0;
+	old_pri = txx9irq[irc_irq].level;
+	txx9irq[irc_irq].level = new_pri;
+	return old_pri;
+}
+
+int txx9_irq(void)
+{
+	u32 csr = __raw_readl(&txx9_ircptr->csr);
+
+	if (likely(!(csr & TXx9_IRCSR_IF)))
+		return TXX9_IRQ_BASE + (csr & (TXx9_MAX_IR - 1));
+	return -1;
+}
diff --git a/arch/mips/kernel/jump_label.c b/arch/mips/kernel/jump_label.c
new file mode 100644
index 000000000..9f5b1247b
--- /dev/null
+++ b/arch/mips/kernel/jump_label.c
@@ -0,0 +1,90 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (c) 2010 Cavium Networks, Inc.
+ */
+
+#include <linux/jump_label.h>
+#include <linux/kernel.h>
+#include <linux/memory.h>
+#include <linux/mutex.h>
+#include <linux/types.h>
+#include <linux/cpu.h>
+
+#include <asm/cacheflush.h>
+#include <asm/inst.h>
+
+/*
+ * Define parameters for the standard MIPS and the microMIPS jump
+ * instruction encoding respectively:
+ *
+ * - the ISA bit of the target, either 0 or 1 respectively,
+ *
+ * - the amount the jump target address is shifted right to fit in the
+ *   immediate field of the machine instruction, either 2 or 1,
+ *
+ * - the mask determining the size of the jump region relative to the
+ *   delay-slot instruction, either 256MB or 128MB,
+ *
+ * - the jump target alignment, either 4 or 2 bytes.
+ */
+#define J_ISA_BIT	IS_ENABLED(CONFIG_CPU_MICROMIPS)
+#define J_RANGE_SHIFT	(2 - J_ISA_BIT)
+#define J_RANGE_MASK	((1ul << (26 + J_RANGE_SHIFT)) - 1)
+#define J_ALIGN_MASK	((1ul << J_RANGE_SHIFT) - 1)
+
+void arch_jump_label_transform(struct jump_entry *e,
+			       enum jump_label_type type)
+{
+	union mips_instruction *insn_p;
+	union mips_instruction insn;
+	long offset;
+
+	insn_p = (union mips_instruction *)msk_isa16_mode(e->code);
+
+	/* Target must have the right alignment and ISA must be preserved. */
+	BUG_ON((e->target & J_ALIGN_MASK) != J_ISA_BIT);
+
+	if (type == JUMP_LABEL_JMP) {
+		if (!IS_ENABLED(CONFIG_CPU_MICROMIPS) && MIPS_ISA_REV >= 6) {
+			offset = e->target - ((unsigned long)insn_p + 4);
+			offset >>= 2;
+
+			/*
+			 * The branch offset must fit in the instruction's 26
+			 * bit field.
+			 */
+			WARN_ON((offset >= (long)BIT(25)) ||
+				(offset < -(long)BIT(25)));
+
+			insn.j_format.opcode = bc6_op;
+			insn.j_format.target = offset;
+		} else {
+			/*
+			 * Jump only works within an aligned region its delay
+			 * slot is in.
+			 */
+			WARN_ON((e->target & ~J_RANGE_MASK) !=
+				((e->code + 4) & ~J_RANGE_MASK));
+
+			insn.j_format.opcode = J_ISA_BIT ? mm_j32_op : j_op;
+			insn.j_format.target = e->target >> J_RANGE_SHIFT;
+		}
+	} else {
+		insn.word = 0; /* nop */
+	}
+
+	mutex_lock(&text_mutex);
+	if (IS_ENABLED(CONFIG_CPU_MICROMIPS)) {
+		insn_p->halfword[0] = insn.word >> 16;
+		insn_p->halfword[1] = insn.word;
+	} else
+		*insn_p = insn;
+
+	flush_icache_range((unsigned long)insn_p,
+			   (unsigned long)insn_p + sizeof(*insn_p));
+
+	mutex_unlock(&text_mutex);
+}
diff --git a/arch/mips/kernel/kgdb.c b/arch/mips/kernel/kgdb.c
new file mode 100644
index 000000000..ea781b29f
--- /dev/null
+++ b/arch/mips/kernel/kgdb.c
@@ -0,0 +1,415 @@
+/*
+ *  Originally written by Glenn Engel, Lake Stevens Instrument Division
+ *
+ *  Contributed by HP Systems
+ *
+ *  Modified for Linux/MIPS (and MIPS in general) by Andreas Busse
+ *  Send complaints, suggestions etc. to <andy@waldorf-gmbh.de>
+ *
+ *  Copyright (C) 1995 Andreas Busse
+ *
+ *  Copyright (C) 2003 MontaVista Software Inc.
+ *  Author: Jun Sun, jsun@mvista.com or jsun@junsun.net
+ *
+ *  Copyright (C) 2004-2005 MontaVista Software Inc.
+ *  Author: Manish Lachwani, mlachwani@mvista.com or manish@koffee-break.com
+ *
+ *  Copyright (C) 2007-2008 Wind River Systems, Inc.
+ *  Author/Maintainer: Jason Wessel, jason.wessel@windriver.com
+ *
+ *  This file is licensed under the terms of the GNU General Public License
+ *  version 2. This program is licensed "as is" without any warranty of any
+ *  kind, whether express or implied.
+ */
+
+#include <linux/ptrace.h>		/* for linux pt_regs struct */
+#include <linux/kgdb.h>
+#include <linux/kdebug.h>
+#include <linux/sched.h>
+#include <linux/smp.h>
+#include <asm/inst.h>
+#include <asm/fpu.h>
+#include <asm/cacheflush.h>
+#include <asm/processor.h>
+#include <asm/sigcontext.h>
+#include <linux/uaccess.h>
+#include <asm/irq_regs.h>
+
+static struct hard_trap_info {
+	unsigned char tt;	/* Trap type code for MIPS R3xxx and R4xxx */
+	unsigned char signo;	/* Signal that we map this trap into */
+} hard_trap_info[] = {
+	{ 6, SIGBUS },		/* instruction bus error */
+	{ 7, SIGBUS },		/* data bus error */
+	{ 9, SIGTRAP },		/* break */
+/*	{ 11, SIGILL }, */	/* CPU unusable */
+	{ 12, SIGFPE },		/* overflow */
+	{ 13, SIGTRAP },	/* trap */
+	{ 14, SIGSEGV },	/* virtual instruction cache coherency */
+	{ 15, SIGFPE },		/* floating point exception */
+	{ 23, SIGSEGV },	/* watch */
+	{ 31, SIGSEGV },	/* virtual data cache coherency */
+	{ 0, 0}			/* Must be last */
+};
+
+struct dbg_reg_def_t dbg_reg_def[DBG_MAX_REG_NUM] =
+{
+	{ "zero", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[0]) },
+	{ "at", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[1]) },
+	{ "v0", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[2]) },
+	{ "v1", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[3]) },
+	{ "a0", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[4]) },
+	{ "a1", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[5]) },
+	{ "a2", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[6]) },
+	{ "a3", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[7]) },
+	{ "t0", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[8]) },
+	{ "t1", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[9]) },
+	{ "t2", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[10]) },
+	{ "t3", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[11]) },
+	{ "t4", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[12]) },
+	{ "t5", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[13]) },
+	{ "t6", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[14]) },
+	{ "t7", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[15]) },
+	{ "s0", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[16]) },
+	{ "s1", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[17]) },
+	{ "s2", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[18]) },
+	{ "s3", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[19]) },
+	{ "s4", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[20]) },
+	{ "s5", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[21]) },
+	{ "s6", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[22]) },
+	{ "s7", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[23]) },
+	{ "t8", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[24]) },
+	{ "t9", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[25]) },
+	{ "k0", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[26]) },
+	{ "k1", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[27]) },
+	{ "gp", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[28]) },
+	{ "sp", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[29]) },
+	{ "s8", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[30]) },
+	{ "ra", GDB_SIZEOF_REG, offsetof(struct pt_regs, regs[31]) },
+	{ "sr", GDB_SIZEOF_REG, offsetof(struct pt_regs, cp0_status) },
+	{ "lo", GDB_SIZEOF_REG, offsetof(struct pt_regs, lo) },
+	{ "hi", GDB_SIZEOF_REG, offsetof(struct pt_regs, hi) },
+	{ "bad", GDB_SIZEOF_REG, offsetof(struct pt_regs, cp0_badvaddr) },
+	{ "cause", GDB_SIZEOF_REG, offsetof(struct pt_regs, cp0_cause) },
+	{ "pc", GDB_SIZEOF_REG, offsetof(struct pt_regs, cp0_epc) },
+	{ "f0", GDB_SIZEOF_REG, 0 },
+	{ "f1", GDB_SIZEOF_REG, 1 },
+	{ "f2", GDB_SIZEOF_REG, 2 },
+	{ "f3", GDB_SIZEOF_REG, 3 },
+	{ "f4", GDB_SIZEOF_REG, 4 },
+	{ "f5", GDB_SIZEOF_REG, 5 },
+	{ "f6", GDB_SIZEOF_REG, 6 },
+	{ "f7", GDB_SIZEOF_REG, 7 },
+	{ "f8", GDB_SIZEOF_REG, 8 },
+	{ "f9", GDB_SIZEOF_REG, 9 },
+	{ "f10", GDB_SIZEOF_REG, 10 },
+	{ "f11", GDB_SIZEOF_REG, 11 },
+	{ "f12", GDB_SIZEOF_REG, 12 },
+	{ "f13", GDB_SIZEOF_REG, 13 },
+	{ "f14", GDB_SIZEOF_REG, 14 },
+	{ "f15", GDB_SIZEOF_REG, 15 },
+	{ "f16", GDB_SIZEOF_REG, 16 },
+	{ "f17", GDB_SIZEOF_REG, 17 },
+	{ "f18", GDB_SIZEOF_REG, 18 },
+	{ "f19", GDB_SIZEOF_REG, 19 },
+	{ "f20", GDB_SIZEOF_REG, 20 },
+	{ "f21", GDB_SIZEOF_REG, 21 },
+	{ "f22", GDB_SIZEOF_REG, 22 },
+	{ "f23", GDB_SIZEOF_REG, 23 },
+	{ "f24", GDB_SIZEOF_REG, 24 },
+	{ "f25", GDB_SIZEOF_REG, 25 },
+	{ "f26", GDB_SIZEOF_REG, 26 },
+	{ "f27", GDB_SIZEOF_REG, 27 },
+	{ "f28", GDB_SIZEOF_REG, 28 },
+	{ "f29", GDB_SIZEOF_REG, 29 },
+	{ "f30", GDB_SIZEOF_REG, 30 },
+	{ "f31", GDB_SIZEOF_REG, 31 },
+	{ "fsr", GDB_SIZEOF_REG, 0 },
+	{ "fir", GDB_SIZEOF_REG, 0 },
+};
+
+int dbg_set_reg(int regno, void *mem, struct pt_regs *regs)
+{
+	int fp_reg;
+
+	if (regno < 0 || regno >= DBG_MAX_REG_NUM)
+		return -EINVAL;
+
+	if (dbg_reg_def[regno].offset != -1 && regno < 38) {
+		memcpy((void *)regs + dbg_reg_def[regno].offset, mem,
+		       dbg_reg_def[regno].size);
+	} else if (current && dbg_reg_def[regno].offset != -1 && regno < 72) {
+		/* FP registers 38 -> 69 */
+		if (!(regs->cp0_status & ST0_CU1))
+			return 0;
+		if (regno == 70) {
+			/* Process the fcr31/fsr (register 70) */
+			memcpy((void *)&current->thread.fpu.fcr31, mem,
+			       dbg_reg_def[regno].size);
+			goto out_save;
+		} else if (regno == 71) {
+			/* Ignore the fir (register 71) */
+			goto out_save;
+		}
+		fp_reg = dbg_reg_def[regno].offset;
+		memcpy((void *)&current->thread.fpu.fpr[fp_reg], mem,
+		       dbg_reg_def[regno].size);
+out_save:
+		restore_fp(current);
+	}
+
+	return 0;
+}
+
+char *dbg_get_reg(int regno, void *mem, struct pt_regs *regs)
+{
+	int fp_reg;
+
+	if (regno >= DBG_MAX_REG_NUM || regno < 0)
+		return NULL;
+
+	if (dbg_reg_def[regno].offset != -1 && regno < 38) {
+		/* First 38 registers */
+		memcpy(mem, (void *)regs + dbg_reg_def[regno].offset,
+		       dbg_reg_def[regno].size);
+	} else if (current && dbg_reg_def[regno].offset != -1 && regno < 72) {
+		/* FP registers 38 -> 69 */
+		if (!(regs->cp0_status & ST0_CU1))
+			goto out;
+		save_fp(current);
+		if (regno == 70) {
+			/* Process the fcr31/fsr (register 70) */
+			memcpy(mem, (void *)&current->thread.fpu.fcr31,
+			       dbg_reg_def[regno].size);
+			goto out;
+		} else if (regno == 71) {
+			/* Ignore the fir (register 71) */
+			memset(mem, 0, dbg_reg_def[regno].size);
+			goto out;
+		}
+		fp_reg = dbg_reg_def[regno].offset;
+		memcpy(mem, (void *)&current->thread.fpu.fpr[fp_reg],
+		       dbg_reg_def[regno].size);
+	}
+
+out:
+	return dbg_reg_def[regno].name;
+
+}
+
+void arch_kgdb_breakpoint(void)
+{
+	__asm__ __volatile__(
+		".globl breakinst\n\t"
+		".set\tnoreorder\n\t"
+		"nop\n"
+		"breakinst:\tbreak\n\t"
+		"nop\n\t"
+		".set\treorder");
+}
+
+void kgdb_call_nmi_hook(void *ignored)
+{
+	mm_segment_t old_fs;
+
+	old_fs = get_fs();
+	set_fs(KERNEL_DS);
+
+	kgdb_nmicallback(raw_smp_processor_id(), get_irq_regs());
+
+	set_fs(old_fs);
+}
+
+static int compute_signal(int tt)
+{
+	struct hard_trap_info *ht;
+
+	for (ht = hard_trap_info; ht->tt && ht->signo; ht++)
+		if (ht->tt == tt)
+			return ht->signo;
+
+	return SIGHUP;		/* default for things we don't know about */
+}
+
+/*
+ * Similar to regs_to_gdb_regs() except that process is sleeping and so
+ * we may not be able to get all the info.
+ */
+void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p)
+{
+	int reg;
+#if (KGDB_GDB_REG_SIZE == 32)
+	u32 *ptr = (u32 *)gdb_regs;
+#else
+	u64 *ptr = (u64 *)gdb_regs;
+#endif
+
+	for (reg = 0; reg < 16; reg++)
+		*(ptr++) = 0;
+
+	/* S0 - S7 */
+	*(ptr++) = p->thread.reg16;
+	*(ptr++) = p->thread.reg17;
+	*(ptr++) = p->thread.reg18;
+	*(ptr++) = p->thread.reg19;
+	*(ptr++) = p->thread.reg20;
+	*(ptr++) = p->thread.reg21;
+	*(ptr++) = p->thread.reg22;
+	*(ptr++) = p->thread.reg23;
+
+	for (reg = 24; reg < 28; reg++)
+		*(ptr++) = 0;
+
+	/* GP, SP, FP, RA */
+	*(ptr++) = (long)p;
+	*(ptr++) = p->thread.reg29;
+	*(ptr++) = p->thread.reg30;
+	*(ptr++) = p->thread.reg31;
+
+	*(ptr++) = p->thread.cp0_status;
+
+	/* lo, hi */
+	*(ptr++) = 0;
+	*(ptr++) = 0;
+
+	/*
+	 * BadVAddr, Cause
+	 * Ideally these would come from the last exception frame up the stack
+	 * but that requires unwinding, otherwise we can't know much for sure.
+	 */
+	*(ptr++) = 0;
+	*(ptr++) = 0;
+
+	/*
+	 * PC
+	 * use return address (RA), i.e. the moment after return from resume()
+	 */
+	*(ptr++) = p->thread.reg31;
+}
+
+void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long pc)
+{
+	regs->cp0_epc = pc;
+}
+
+/*
+ * Calls linux_debug_hook before the kernel dies. If KGDB is enabled,
+ * then try to fall into the debugger
+ */
+static int kgdb_mips_notify(struct notifier_block *self, unsigned long cmd,
+			    void *ptr)
+{
+	struct die_args *args = (struct die_args *)ptr;
+	struct pt_regs *regs = args->regs;
+	int trap = (regs->cp0_cause & 0x7c) >> 2;
+	mm_segment_t old_fs;
+
+#ifdef CONFIG_KPROBES
+	/*
+	 * Return immediately if the kprobes fault notifier has set
+	 * DIE_PAGE_FAULT.
+	 */
+	if (cmd == DIE_PAGE_FAULT)
+		return NOTIFY_DONE;
+#endif /* CONFIG_KPROBES */
+
+	/* Userspace events, ignore. */
+	if (user_mode(regs))
+		return NOTIFY_DONE;
+
+	/* Kernel mode. Set correct address limit */
+	old_fs = get_fs();
+	set_fs(KERNEL_DS);
+
+	if (atomic_read(&kgdb_active) != -1)
+		kgdb_nmicallback(smp_processor_id(), regs);
+
+	if (kgdb_handle_exception(trap, compute_signal(trap), cmd, regs)) {
+		set_fs(old_fs);
+		return NOTIFY_DONE;
+	}
+
+	if (atomic_read(&kgdb_setting_breakpoint))
+		if ((trap == 9) && (regs->cp0_epc == (unsigned long)breakinst))
+			regs->cp0_epc += 4;
+
+	/* In SMP mode, __flush_cache_all does IPI */
+	local_irq_enable();
+	__flush_cache_all();
+
+	set_fs(old_fs);
+	return NOTIFY_STOP;
+}
+
+#ifdef CONFIG_KGDB_LOW_LEVEL_TRAP
+int kgdb_ll_trap(int cmd, const char *str,
+		 struct pt_regs *regs, long err, int trap, int sig)
+{
+	struct die_args args = {
+		.regs	= regs,
+		.str	= str,
+		.err	= err,
+		.trapnr = trap,
+		.signr	= sig,
+
+	};
+
+	if (!kgdb_io_module_registered)
+		return NOTIFY_DONE;
+
+	return kgdb_mips_notify(NULL, cmd, &args);
+}
+#endif /* CONFIG_KGDB_LOW_LEVEL_TRAP */
+
+static struct notifier_block kgdb_notifier = {
+	.notifier_call = kgdb_mips_notify,
+};
+
+/*
+ * Handle the 'c' command
+ */
+int kgdb_arch_handle_exception(int vector, int signo, int err_code,
+			       char *remcom_in_buffer, char *remcom_out_buffer,
+			       struct pt_regs *regs)
+{
+	char *ptr;
+	unsigned long address;
+
+	switch (remcom_in_buffer[0]) {
+	case 'c':
+		/* handle the optional parameter */
+		ptr = &remcom_in_buffer[1];
+		if (kgdb_hex2long(&ptr, &address))
+			regs->cp0_epc = address;
+
+		return 0;
+	}
+
+	return -1;
+}
+
+const struct kgdb_arch arch_kgdb_ops = {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	.gdb_bpt_instr = { spec_op << 2, 0x00, 0x00, break_op },
+#else
+	.gdb_bpt_instr = { break_op, 0x00, 0x00, spec_op << 2 },
+#endif
+};
+
+int kgdb_arch_init(void)
+{
+	register_die_notifier(&kgdb_notifier);
+
+	return 0;
+}
+
+/*
+ *	kgdb_arch_exit - Perform any architecture specific uninitalization.
+ *
+ *	This function will handle the uninitalization of any architecture
+ *	specific callbacks, for dynamic registration and unregistration.
+ */
+void kgdb_arch_exit(void)
+{
+	unregister_die_notifier(&kgdb_notifier);
+}
diff --git a/arch/mips/kernel/kprobes.c b/arch/mips/kernel/kprobes.c
new file mode 100644
index 000000000..54dfba8fa
--- /dev/null
+++ b/arch/mips/kernel/kprobes.c
@@ -0,0 +1,518 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *  Kernel Probes (KProbes)
+ *  arch/mips/kernel/kprobes.c
+ *
+ *  Copyright 2006 Sony Corp.
+ *  Copyright 2010 Cavium Networks
+ *
+ *  Some portions copied from the powerpc version.
+ *
+ *   Copyright (C) IBM Corporation, 2002, 2004
+ */
+
+#include <linux/kprobes.h>
+#include <linux/preempt.h>
+#include <linux/uaccess.h>
+#include <linux/kdebug.h>
+#include <linux/slab.h>
+
+#include <asm/ptrace.h>
+#include <asm/branch.h>
+#include <asm/break.h>
+
+#include "probes-common.h"
+
+static const union mips_instruction breakpoint_insn = {
+	.b_format = {
+		.opcode = spec_op,
+		.code = BRK_KPROBE_BP,
+		.func = break_op
+	}
+};
+
+static const union mips_instruction breakpoint2_insn = {
+	.b_format = {
+		.opcode = spec_op,
+		.code = BRK_KPROBE_SSTEPBP,
+		.func = break_op
+	}
+};
+
+DEFINE_PER_CPU(struct kprobe *, current_kprobe);
+DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
+
+static int __kprobes insn_has_delayslot(union mips_instruction insn)
+{
+	return __insn_has_delay_slot(insn);
+}
+
+/*
+ * insn_has_ll_or_sc function checks whether instruction is ll or sc
+ * one; putting breakpoint on top of atomic ll/sc pair is bad idea;
+ * so we need to prevent it and refuse kprobes insertion for such
+ * instructions; cannot do much about breakpoint in the middle of
+ * ll/sc pair; it is upto user to avoid those places
+ */
+static int __kprobes insn_has_ll_or_sc(union mips_instruction insn)
+{
+	int ret = 0;
+
+	switch (insn.i_format.opcode) {
+	case ll_op:
+	case lld_op:
+	case sc_op:
+	case scd_op:
+		ret = 1;
+		break;
+	default:
+		break;
+	}
+	return ret;
+}
+
+int __kprobes arch_prepare_kprobe(struct kprobe *p)
+{
+	union mips_instruction insn;
+	union mips_instruction prev_insn;
+	int ret = 0;
+
+	insn = p->addr[0];
+
+	if (insn_has_ll_or_sc(insn)) {
+		pr_notice("Kprobes for ll and sc instructions are not"
+			  "supported\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (copy_from_kernel_nofault(&prev_insn, p->addr - 1,
+			sizeof(mips_instruction)) == 0 &&
+	    insn_has_delayslot(prev_insn)) {
+		pr_notice("Kprobes for branch delayslot are not supported\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (__insn_is_compact_branch(insn)) {
+		pr_notice("Kprobes for compact branches are not supported\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	/* insn: must be on special executable page on mips. */
+	p->ainsn.insn = get_insn_slot();
+	if (!p->ainsn.insn) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	/*
+	 * In the kprobe->ainsn.insn[] array we store the original
+	 * instruction at index zero and a break trap instruction at
+	 * index one.
+	 *
+	 * On MIPS arch if the instruction at probed address is a
+	 * branch instruction, we need to execute the instruction at
+	 * Branch Delayslot (BD) at the time of probe hit. As MIPS also
+	 * doesn't have single stepping support, the BD instruction can
+	 * not be executed in-line and it would be executed on SSOL slot
+	 * using a normal breakpoint instruction in the next slot.
+	 * So, read the instruction and save it for later execution.
+	 */
+	if (insn_has_delayslot(insn))
+		memcpy(&p->ainsn.insn[0], p->addr + 1, sizeof(kprobe_opcode_t));
+	else
+		memcpy(&p->ainsn.insn[0], p->addr, sizeof(kprobe_opcode_t));
+
+	p->ainsn.insn[1] = breakpoint2_insn;
+	p->opcode = *p->addr;
+
+out:
+	return ret;
+}
+
+void __kprobes arch_arm_kprobe(struct kprobe *p)
+{
+	*p->addr = breakpoint_insn;
+	flush_insn_slot(p);
+}
+
+void __kprobes arch_disarm_kprobe(struct kprobe *p)
+{
+	*p->addr = p->opcode;
+	flush_insn_slot(p);
+}
+
+void __kprobes arch_remove_kprobe(struct kprobe *p)
+{
+	if (p->ainsn.insn) {
+		free_insn_slot(p->ainsn.insn, 0);
+		p->ainsn.insn = NULL;
+	}
+}
+
+static void save_previous_kprobe(struct kprobe_ctlblk *kcb)
+{
+	kcb->prev_kprobe.kp = kprobe_running();
+	kcb->prev_kprobe.status = kcb->kprobe_status;
+	kcb->prev_kprobe.old_SR = kcb->kprobe_old_SR;
+	kcb->prev_kprobe.saved_SR = kcb->kprobe_saved_SR;
+	kcb->prev_kprobe.saved_epc = kcb->kprobe_saved_epc;
+}
+
+static void restore_previous_kprobe(struct kprobe_ctlblk *kcb)
+{
+	__this_cpu_write(current_kprobe, kcb->prev_kprobe.kp);
+	kcb->kprobe_status = kcb->prev_kprobe.status;
+	kcb->kprobe_old_SR = kcb->prev_kprobe.old_SR;
+	kcb->kprobe_saved_SR = kcb->prev_kprobe.saved_SR;
+	kcb->kprobe_saved_epc = kcb->prev_kprobe.saved_epc;
+}
+
+static void set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
+			       struct kprobe_ctlblk *kcb)
+{
+	__this_cpu_write(current_kprobe, p);
+	kcb->kprobe_saved_SR = kcb->kprobe_old_SR = (regs->cp0_status & ST0_IE);
+	kcb->kprobe_saved_epc = regs->cp0_epc;
+}
+
+/**
+ * evaluate_branch_instrucion -
+ *
+ * Evaluate the branch instruction at probed address during probe hit. The
+ * result of evaluation would be the updated epc. The insturction in delayslot
+ * would actually be single stepped using a normal breakpoint) on SSOL slot.
+ *
+ * The result is also saved in the kprobe control block for later use,
+ * in case we need to execute the delayslot instruction. The latter will be
+ * false for NOP instruction in dealyslot and the branch-likely instructions
+ * when the branch is taken. And for those cases we set a flag as
+ * SKIP_DELAYSLOT in the kprobe control block
+ */
+static int evaluate_branch_instruction(struct kprobe *p, struct pt_regs *regs,
+					struct kprobe_ctlblk *kcb)
+{
+	union mips_instruction insn = p->opcode;
+	long epc;
+	int ret = 0;
+
+	epc = regs->cp0_epc;
+	if (epc & 3)
+		goto unaligned;
+
+	if (p->ainsn.insn->word == 0)
+		kcb->flags |= SKIP_DELAYSLOT;
+	else
+		kcb->flags &= ~SKIP_DELAYSLOT;
+
+	ret = __compute_return_epc_for_insn(regs, insn);
+	if (ret < 0)
+		return ret;
+
+	if (ret == BRANCH_LIKELY_TAKEN)
+		kcb->flags |= SKIP_DELAYSLOT;
+
+	kcb->target_epc = regs->cp0_epc;
+
+	return 0;
+
+unaligned:
+	pr_notice("%s: unaligned epc - sending SIGBUS.\n", current->comm);
+	force_sig(SIGBUS);
+	return -EFAULT;
+
+}
+
+static void prepare_singlestep(struct kprobe *p, struct pt_regs *regs,
+						struct kprobe_ctlblk *kcb)
+{
+	int ret = 0;
+
+	regs->cp0_status &= ~ST0_IE;
+
+	/* single step inline if the instruction is a break */
+	if (p->opcode.word == breakpoint_insn.word ||
+	    p->opcode.word == breakpoint2_insn.word)
+		regs->cp0_epc = (unsigned long)p->addr;
+	else if (insn_has_delayslot(p->opcode)) {
+		ret = evaluate_branch_instruction(p, regs, kcb);
+		if (ret < 0) {
+			pr_notice("Kprobes: Error in evaluating branch\n");
+			return;
+		}
+	}
+	regs->cp0_epc = (unsigned long)&p->ainsn.insn[0];
+}
+
+/*
+ * Called after single-stepping.  p->addr is the address of the
+ * instruction whose first byte has been replaced by the "break 0"
+ * instruction.	 To avoid the SMP problems that can occur when we
+ * temporarily put back the original opcode to single-step, we
+ * single-stepped a copy of the instruction.  The address of this
+ * copy is p->ainsn.insn.
+ *
+ * This function prepares to return from the post-single-step
+ * breakpoint trap. In case of branch instructions, the target
+ * epc to be restored.
+ */
+static void __kprobes resume_execution(struct kprobe *p,
+				       struct pt_regs *regs,
+				       struct kprobe_ctlblk *kcb)
+{
+	if (insn_has_delayslot(p->opcode))
+		regs->cp0_epc = kcb->target_epc;
+	else {
+		unsigned long orig_epc = kcb->kprobe_saved_epc;
+		regs->cp0_epc = orig_epc + 4;
+	}
+}
+
+static int __kprobes kprobe_handler(struct pt_regs *regs)
+{
+	struct kprobe *p;
+	int ret = 0;
+	kprobe_opcode_t *addr;
+	struct kprobe_ctlblk *kcb;
+
+	addr = (kprobe_opcode_t *) regs->cp0_epc;
+
+	/*
+	 * We don't want to be preempted for the entire
+	 * duration of kprobe processing
+	 */
+	preempt_disable();
+	kcb = get_kprobe_ctlblk();
+
+	/* Check we're not actually recursing */
+	if (kprobe_running()) {
+		p = get_kprobe(addr);
+		if (p) {
+			if (kcb->kprobe_status == KPROBE_HIT_SS &&
+			    p->ainsn.insn->word == breakpoint_insn.word) {
+				regs->cp0_status &= ~ST0_IE;
+				regs->cp0_status |= kcb->kprobe_saved_SR;
+				goto no_kprobe;
+			}
+			/*
+			 * We have reentered the kprobe_handler(), since
+			 * another probe was hit while within the handler.
+			 * We here save the original kprobes variables and
+			 * just single step on the instruction of the new probe
+			 * without calling any user handlers.
+			 */
+			save_previous_kprobe(kcb);
+			set_current_kprobe(p, regs, kcb);
+			kprobes_inc_nmissed_count(p);
+			prepare_singlestep(p, regs, kcb);
+			kcb->kprobe_status = KPROBE_REENTER;
+			if (kcb->flags & SKIP_DELAYSLOT) {
+				resume_execution(p, regs, kcb);
+				restore_previous_kprobe(kcb);
+				preempt_enable_no_resched();
+			}
+			return 1;
+		} else if (addr->word != breakpoint_insn.word) {
+			/*
+			 * The breakpoint instruction was removed by
+			 * another cpu right after we hit, no further
+			 * handling of this interrupt is appropriate
+			 */
+			ret = 1;
+		}
+		goto no_kprobe;
+	}
+
+	p = get_kprobe(addr);
+	if (!p) {
+		if (addr->word != breakpoint_insn.word) {
+			/*
+			 * The breakpoint instruction was removed right
+			 * after we hit it.  Another cpu has removed
+			 * either a probepoint or a debugger breakpoint
+			 * at this address.  In either case, no further
+			 * handling of this interrupt is appropriate.
+			 */
+			ret = 1;
+		}
+		/* Not one of ours: let kernel handle it */
+		goto no_kprobe;
+	}
+
+	set_current_kprobe(p, regs, kcb);
+	kcb->kprobe_status = KPROBE_HIT_ACTIVE;
+
+	if (p->pre_handler && p->pre_handler(p, regs)) {
+		/* handler has already set things up, so skip ss setup */
+		reset_current_kprobe();
+		preempt_enable_no_resched();
+		return 1;
+	}
+
+	prepare_singlestep(p, regs, kcb);
+	if (kcb->flags & SKIP_DELAYSLOT) {
+		kcb->kprobe_status = KPROBE_HIT_SSDONE;
+		if (p->post_handler)
+			p->post_handler(p, regs, 0);
+		resume_execution(p, regs, kcb);
+		preempt_enable_no_resched();
+	} else
+		kcb->kprobe_status = KPROBE_HIT_SS;
+
+	return 1;
+
+no_kprobe:
+	preempt_enable_no_resched();
+	return ret;
+
+}
+
+static inline int post_kprobe_handler(struct pt_regs *regs)
+{
+	struct kprobe *cur = kprobe_running();
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+	if (!cur)
+		return 0;
+
+	if ((kcb->kprobe_status != KPROBE_REENTER) && cur->post_handler) {
+		kcb->kprobe_status = KPROBE_HIT_SSDONE;
+		cur->post_handler(cur, regs, 0);
+	}
+
+	resume_execution(cur, regs, kcb);
+
+	regs->cp0_status |= kcb->kprobe_saved_SR;
+
+	/* Restore back the original saved kprobes variables and continue. */
+	if (kcb->kprobe_status == KPROBE_REENTER) {
+		restore_previous_kprobe(kcb);
+		goto out;
+	}
+	reset_current_kprobe();
+out:
+	preempt_enable_no_resched();
+
+	return 1;
+}
+
+int kprobe_fault_handler(struct pt_regs *regs, int trapnr)
+{
+	struct kprobe *cur = kprobe_running();
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+	if (cur->fault_handler && cur->fault_handler(cur, regs, trapnr))
+		return 1;
+
+	if (kcb->kprobe_status & KPROBE_HIT_SS) {
+		resume_execution(cur, regs, kcb);
+		regs->cp0_status |= kcb->kprobe_old_SR;
+
+		reset_current_kprobe();
+		preempt_enable_no_resched();
+	}
+	return 0;
+}
+
+/*
+ * Wrapper routine for handling exceptions.
+ */
+int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
+				       unsigned long val, void *data)
+{
+
+	struct die_args *args = (struct die_args *)data;
+	int ret = NOTIFY_DONE;
+
+	switch (val) {
+	case DIE_BREAK:
+		if (kprobe_handler(args->regs))
+			ret = NOTIFY_STOP;
+		break;
+	case DIE_SSTEPBP:
+		if (post_kprobe_handler(args->regs))
+			ret = NOTIFY_STOP;
+		break;
+
+	case DIE_PAGE_FAULT:
+		/* kprobe_running() needs smp_processor_id() */
+		preempt_disable();
+
+		if (kprobe_running()
+		    && kprobe_fault_handler(args->regs, args->trapnr))
+			ret = NOTIFY_STOP;
+		preempt_enable();
+		break;
+	default:
+		break;
+	}
+	return ret;
+}
+
+/*
+ * Function return probe trampoline:
+ *	- init_kprobes() establishes a probepoint here
+ *	- When the probed function returns, this probe causes the
+ *	  handlers to fire
+ */
+static void __used kretprobe_trampoline_holder(void)
+{
+	asm volatile(
+		".set push\n\t"
+		/* Keep the assembler from reordering and placing JR here. */
+		".set noreorder\n\t"
+		"nop\n\t"
+		".global kretprobe_trampoline\n"
+		"kretprobe_trampoline:\n\t"
+		"nop\n\t"
+		".set pop"
+		: : : "memory");
+}
+
+void kretprobe_trampoline(void);
+
+void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
+				      struct pt_regs *regs)
+{
+	ri->ret_addr = (kprobe_opcode_t *) regs->regs[31];
+	ri->fp = NULL;
+
+	/* Replace the return addr with trampoline addr */
+	regs->regs[31] = (unsigned long)kretprobe_trampoline;
+}
+
+/*
+ * Called when the probe at kretprobe trampoline is hit
+ */
+static int __kprobes trampoline_probe_handler(struct kprobe *p,
+						struct pt_regs *regs)
+{
+	instruction_pointer(regs) = __kretprobe_trampoline_handler(regs,
+						kretprobe_trampoline, NULL);
+	/*
+	 * By returning a non-zero value, we are telling
+	 * kprobe_handler() that we don't want the post_handler
+	 * to run (and have re-enabled preemption)
+	 */
+	return 1;
+}
+
+int __kprobes arch_trampoline_kprobe(struct kprobe *p)
+{
+	if (p->addr == (kprobe_opcode_t *)kretprobe_trampoline)
+		return 1;
+
+	return 0;
+}
+
+static struct kprobe trampoline_p = {
+	.addr = (kprobe_opcode_t *)kretprobe_trampoline,
+	.pre_handler = trampoline_probe_handler
+};
+
+int __init arch_init_kprobes(void)
+{
+	return register_kprobe(&trampoline_p);
+}
diff --git a/arch/mips/kernel/linux32.c b/arch/mips/kernel/linux32.c
new file mode 100644
index 000000000..6b61be486
--- /dev/null
+++ b/arch/mips/kernel/linux32.c
@@ -0,0 +1,133 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Conversion between 32-bit and 64-bit native system calls.
+ *
+ * Copyright (C) 2000 Silicon Graphics, Inc.
+ * Written by Ulf Carlsson (ulfc@engr.sgi.com)
+ */
+#include <linux/compiler.h>
+#include <linux/mm.h>
+#include <linux/errno.h>
+#include <linux/file.h>
+#include <linux/highuid.h>
+#include <linux/resource.h>
+#include <linux/highmem.h>
+#include <linux/time.h>
+#include <linux/times.h>
+#include <linux/poll.h>
+#include <linux/skbuff.h>
+#include <linux/filter.h>
+#include <linux/shm.h>
+#include <linux/sem.h>
+#include <linux/msg.h>
+#include <linux/icmpv6.h>
+#include <linux/syscalls.h>
+#include <linux/sysctl.h>
+#include <linux/utime.h>
+#include <linux/utsname.h>
+#include <linux/personality.h>
+#include <linux/dnotify.h>
+#include <linux/binfmts.h>
+#include <linux/security.h>
+#include <linux/compat.h>
+#include <linux/vfs.h>
+#include <linux/ipc.h>
+#include <linux/slab.h>
+
+#include <net/sock.h>
+#include <net/scm.h>
+
+#include <asm/compat-signal.h>
+#include <asm/sim.h>
+#include <linux/uaccess.h>
+#include <asm/mmu_context.h>
+#include <asm/mman.h>
+
+#ifdef __MIPSEB__
+#define merge_64(r1, r2) ((((r1) & 0xffffffffUL) << 32) + ((r2) & 0xffffffffUL))
+#endif
+#ifdef __MIPSEL__
+#define merge_64(r1, r2) ((((r2) & 0xffffffffUL) << 32) + ((r1) & 0xffffffffUL))
+#endif
+
+SYSCALL_DEFINE4(32_truncate64, const char __user *, path,
+	unsigned long, __dummy, unsigned long, a2, unsigned long, a3)
+{
+	return ksys_truncate(path, merge_64(a2, a3));
+}
+
+SYSCALL_DEFINE4(32_ftruncate64, unsigned long, fd, unsigned long, __dummy,
+	unsigned long, a2, unsigned long, a3)
+{
+	return ksys_ftruncate(fd, merge_64(a2, a3));
+}
+
+SYSCALL_DEFINE5(32_llseek, unsigned int, fd, unsigned int, offset_high,
+		unsigned int, offset_low, loff_t __user *, result,
+		unsigned int, origin)
+{
+	return sys_llseek(fd, offset_high, offset_low, result, origin);
+}
+
+/* From the Single Unix Spec: pread & pwrite act like lseek to pos + op +
+   lseek back to original location.  They fail just like lseek does on
+   non-seekable files.	*/
+
+SYSCALL_DEFINE6(32_pread, unsigned long, fd, char __user *, buf, size_t, count,
+	unsigned long, unused, unsigned long, a4, unsigned long, a5)
+{
+	return ksys_pread64(fd, buf, count, merge_64(a4, a5));
+}
+
+SYSCALL_DEFINE6(32_pwrite, unsigned int, fd, const char __user *, buf,
+	size_t, count, u32, unused, u64, a4, u64, a5)
+{
+	return ksys_pwrite64(fd, buf, count, merge_64(a4, a5));
+}
+
+SYSCALL_DEFINE1(32_personality, unsigned long, personality)
+{
+	unsigned int p = personality & 0xffffffff;
+	int ret;
+
+	if (personality(current->personality) == PER_LINUX32 &&
+	    personality(p) == PER_LINUX)
+		p = (p & ~PER_MASK) | PER_LINUX32;
+	ret = sys_personality(p);
+	if (ret != -1 && personality(ret) == PER_LINUX32)
+		ret = (ret & ~PER_MASK) | PER_LINUX;
+	return ret;
+}
+
+asmlinkage ssize_t sys32_readahead(int fd, u32 pad0, u64 a2, u64 a3,
+				   size_t count)
+{
+	return ksys_readahead(fd, merge_64(a2, a3), count);
+}
+
+asmlinkage long sys32_sync_file_range(int fd, int __pad,
+	unsigned long a2, unsigned long a3,
+	unsigned long a4, unsigned long a5,
+	int flags)
+{
+	return ksys_sync_file_range(fd,
+			merge_64(a2, a3), merge_64(a4, a5),
+			flags);
+}
+
+asmlinkage long sys32_fadvise64_64(int fd, int __pad,
+	unsigned long a2, unsigned long a3,
+	unsigned long a4, unsigned long a5,
+	int flags)
+{
+	return ksys_fadvise64_64(fd,
+			merge_64(a2, a3), merge_64(a4, a5),
+			flags);
+}
+
+asmlinkage long sys32_fallocate(int fd, int mode, unsigned offset_a2,
+	unsigned offset_a3, unsigned len_a4, unsigned len_a5)
+{
+	return ksys_fallocate(fd, mode, merge_64(offset_a2, offset_a3),
+			      merge_64(len_a4, len_a5));
+}
diff --git a/arch/mips/kernel/machine_kexec.c b/arch/mips/kernel/machine_kexec.c
new file mode 100644
index 000000000..432bfd3e7
--- /dev/null
+++ b/arch/mips/kernel/machine_kexec.c
@@ -0,0 +1,264 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * machine_kexec.c for kexec
+ * Created by <nschichan@corp.free.fr> on Thu Oct 12 15:15:06 2006
+ */
+#include <linux/compiler.h>
+#include <linux/kexec.h>
+#include <linux/mm.h>
+#include <linux/delay.h>
+#include <linux/libfdt.h>
+
+#include <asm/cacheflush.h>
+#include <asm/page.h>
+
+extern const unsigned char relocate_new_kernel[];
+extern const size_t relocate_new_kernel_size;
+
+extern unsigned long kexec_start_address;
+extern unsigned long kexec_indirection_page;
+
+static unsigned long reboot_code_buffer;
+
+#ifdef CONFIG_SMP
+static void (*relocated_kexec_smp_wait)(void *);
+
+atomic_t kexec_ready_to_reboot = ATOMIC_INIT(0);
+void (*_crash_smp_send_stop)(void) = NULL;
+#endif
+
+void (*_machine_kexec_shutdown)(void) = NULL;
+void (*_machine_crash_shutdown)(struct pt_regs *regs) = NULL;
+
+static void kexec_image_info(const struct kimage *kimage)
+{
+	unsigned long i;
+
+	pr_debug("kexec kimage info:\n");
+	pr_debug("  type:        %d\n", kimage->type);
+	pr_debug("  start:       %lx\n", kimage->start);
+	pr_debug("  head:        %lx\n", kimage->head);
+	pr_debug("  nr_segments: %lu\n", kimage->nr_segments);
+
+	for (i = 0; i < kimage->nr_segments; i++) {
+		pr_debug("    segment[%lu]: %016lx - %016lx, 0x%lx bytes, %lu pages\n",
+			i,
+			kimage->segment[i].mem,
+			kimage->segment[i].mem + kimage->segment[i].memsz,
+			(unsigned long)kimage->segment[i].memsz,
+			(unsigned long)kimage->segment[i].memsz /  PAGE_SIZE);
+	}
+}
+
+#ifdef CONFIG_UHI_BOOT
+
+static int uhi_machine_kexec_prepare(struct kimage *kimage)
+{
+	int i;
+
+	/*
+	 * In case DTB file is not passed to the new kernel, a flat device
+	 * tree will be created by kexec tool. It holds modified command
+	 * line for the new kernel.
+	 */
+	for (i = 0; i < kimage->nr_segments; i++) {
+		struct fdt_header fdt;
+
+		if (kimage->segment[i].memsz <= sizeof(fdt))
+			continue;
+
+		if (copy_from_user(&fdt, kimage->segment[i].buf, sizeof(fdt)))
+			continue;
+
+		if (fdt_check_header(&fdt))
+			continue;
+
+		kexec_args[0] = -2;
+		kexec_args[1] = (unsigned long)
+			phys_to_virt((unsigned long)kimage->segment[i].mem);
+		break;
+	}
+
+	return 0;
+}
+
+int (*_machine_kexec_prepare)(struct kimage *) = uhi_machine_kexec_prepare;
+
+#else
+
+int (*_machine_kexec_prepare)(struct kimage *) = NULL;
+
+#endif /* CONFIG_UHI_BOOT */
+
+int
+machine_kexec_prepare(struct kimage *kimage)
+{
+#ifdef CONFIG_SMP
+	if (!kexec_nonboot_cpu_func())
+		return -EINVAL;
+#endif
+
+	kexec_image_info(kimage);
+
+	if (_machine_kexec_prepare)
+		return _machine_kexec_prepare(kimage);
+
+	return 0;
+}
+
+void
+machine_kexec_cleanup(struct kimage *kimage)
+{
+}
+
+#ifdef CONFIG_SMP
+static void kexec_shutdown_secondary(void *param)
+{
+	int cpu = smp_processor_id();
+
+	if (!cpu_online(cpu))
+		return;
+
+	/* We won't be sent IPIs any more. */
+	set_cpu_online(cpu, false);
+
+	local_irq_disable();
+	while (!atomic_read(&kexec_ready_to_reboot))
+		cpu_relax();
+
+	kexec_reboot();
+
+	/* NOTREACHED */
+}
+#endif
+
+void
+machine_shutdown(void)
+{
+	if (_machine_kexec_shutdown)
+		_machine_kexec_shutdown();
+
+#ifdef CONFIG_SMP
+	smp_call_function(kexec_shutdown_secondary, NULL, 0);
+
+	while (num_online_cpus() > 1) {
+		cpu_relax();
+		mdelay(1);
+	}
+#endif
+}
+
+void
+machine_crash_shutdown(struct pt_regs *regs)
+{
+	if (_machine_crash_shutdown)
+		_machine_crash_shutdown(regs);
+	else
+		default_machine_crash_shutdown(regs);
+}
+
+#ifdef CONFIG_SMP
+void kexec_nonboot_cpu_jump(void)
+{
+	local_flush_icache_range((unsigned long)relocated_kexec_smp_wait,
+				 reboot_code_buffer + relocate_new_kernel_size);
+
+	relocated_kexec_smp_wait(NULL);
+}
+#endif
+
+void kexec_reboot(void)
+{
+	void (*do_kexec)(void) __noreturn;
+
+	/*
+	 * We know we were online, and there will be no incoming IPIs at
+	 * this point. Mark online again before rebooting so that the crash
+	 * analysis tool will see us correctly.
+	 */
+	set_cpu_online(smp_processor_id(), true);
+
+	/* Ensure remote CPUs observe that we're online before rebooting. */
+	smp_mb__after_atomic();
+
+#ifdef CONFIG_SMP
+	if (smp_processor_id() > 0) {
+		/*
+		 * Instead of cpu_relax() or wait, this is needed for kexec
+		 * smp reboot. Kdump usually doesn't require an smp new
+		 * kernel, but kexec may do.
+		 */
+		kexec_nonboot_cpu();
+
+		/* NOTREACHED */
+	}
+#endif
+
+	/*
+	 * Make sure we get correct instructions written by the
+	 * machine_kexec() CPU.
+	 */
+	local_flush_icache_range(reboot_code_buffer,
+				 reboot_code_buffer + relocate_new_kernel_size);
+
+	do_kexec = (void *)reboot_code_buffer;
+	do_kexec();
+}
+
+void
+machine_kexec(struct kimage *image)
+{
+	unsigned long entry;
+	unsigned long *ptr;
+
+	reboot_code_buffer =
+	  (unsigned long)page_address(image->control_code_page);
+
+	kexec_start_address =
+		(unsigned long) phys_to_virt(image->start);
+
+	if (image->type == KEXEC_TYPE_DEFAULT) {
+		kexec_indirection_page =
+			(unsigned long) phys_to_virt(image->head & PAGE_MASK);
+	} else {
+		kexec_indirection_page = (unsigned long)&image->head;
+	}
+
+	memcpy((void*)reboot_code_buffer, relocate_new_kernel,
+	       relocate_new_kernel_size);
+
+	/*
+	 * The generic kexec code builds a page list with physical
+	 * addresses. they are directly accessible through KSEG0 (or
+	 * CKSEG0 or XPHYS if on 64bit system), hence the
+	 * phys_to_virt() call.
+	 */
+	for (ptr = &image->head; (entry = *ptr) && !(entry &IND_DONE);
+	     ptr = (entry & IND_INDIRECTION) ?
+	       phys_to_virt(entry & PAGE_MASK) : ptr + 1) {
+		if (*ptr & IND_SOURCE || *ptr & IND_INDIRECTION ||
+		    *ptr & IND_DESTINATION)
+			*ptr = (unsigned long) phys_to_virt(*ptr);
+	}
+
+	/* Mark offline BEFORE disabling local irq. */
+	set_cpu_online(smp_processor_id(), false);
+
+	/*
+	 * we do not want to be bothered.
+	 */
+	local_irq_disable();
+
+	printk("Will call new kernel at %08lx\n", image->start);
+	printk("Bye ...\n");
+	/* Make reboot code buffer available to the boot CPU. */
+	__flush_cache_all();
+#ifdef CONFIG_SMP
+	/* All secondary cpus now may jump to kexec_wait cycle */
+	relocated_kexec_smp_wait = reboot_code_buffer +
+		(void *)(kexec_smp_wait - relocate_new_kernel);
+	smp_wmb();
+	atomic_set(&kexec_ready_to_reboot, 1);
+#endif
+	kexec_reboot();
+}
diff --git a/arch/mips/kernel/mcount.S b/arch/mips/kernel/mcount.S
new file mode 100644
index 000000000..cff52b283
--- /dev/null
+++ b/arch/mips/kernel/mcount.S
@@ -0,0 +1,220 @@
+/*
+ * MIPS specific _mcount support
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive for
+ * more details.
+ *
+ * Copyright (C) 2009 Lemote Inc. & DSLab, Lanzhou University, China
+ * Copyright (C) 2010 DSLab, Lanzhou University, China
+ * Author: Wu Zhangjin <wuzhangjin@gmail.com>
+ */
+
+#include <asm/export.h>
+#include <asm/regdef.h>
+#include <asm/stackframe.h>
+#include <asm/ftrace.h>
+
+	.text
+	.set noreorder
+	.set noat
+
+	.macro MCOUNT_SAVE_REGS
+	PTR_SUBU	sp, PT_SIZE
+	PTR_S	ra, PT_R31(sp)
+	PTR_S	AT, PT_R1(sp)
+	PTR_S	a0, PT_R4(sp)
+	PTR_S	a1, PT_R5(sp)
+	PTR_S	a2, PT_R6(sp)
+	PTR_S	a3, PT_R7(sp)
+#ifdef CONFIG_64BIT
+	PTR_S	a4, PT_R8(sp)
+	PTR_S	a5, PT_R9(sp)
+	PTR_S	a6, PT_R10(sp)
+	PTR_S	a7, PT_R11(sp)
+#endif
+	.endm
+
+	.macro MCOUNT_RESTORE_REGS
+	PTR_L	ra, PT_R31(sp)
+	PTR_L	AT, PT_R1(sp)
+	PTR_L	a0, PT_R4(sp)
+	PTR_L	a1, PT_R5(sp)
+	PTR_L	a2, PT_R6(sp)
+	PTR_L	a3, PT_R7(sp)
+#ifdef CONFIG_64BIT
+	PTR_L	a4, PT_R8(sp)
+	PTR_L	a5, PT_R9(sp)
+	PTR_L	a6, PT_R10(sp)
+	PTR_L	a7, PT_R11(sp)
+#endif
+	PTR_ADDIU	sp, PT_SIZE
+	.endm
+
+	.macro RETURN_BACK
+	jr ra
+	 move ra, AT
+	.endm
+
+/*
+ * The -mmcount-ra-address option of gcc 4.5 uses register $12 to pass
+ * the location of the parent's return address.
+ */
+#define MCOUNT_RA_ADDRESS_REG	$12
+
+#ifdef CONFIG_DYNAMIC_FTRACE
+
+NESTED(ftrace_caller, PT_SIZE, ra)
+	.globl _mcount
+_mcount:
+EXPORT_SYMBOL(_mcount)
+	b	ftrace_stub
+#ifdef CONFIG_32BIT
+	 addiu sp,sp,8
+#else
+	 nop
+#endif
+
+	/* When tracing is activated, it calls ftrace_caller+8 (aka here) */
+	MCOUNT_SAVE_REGS
+#ifdef KBUILD_MCOUNT_RA_ADDRESS
+	PTR_S	MCOUNT_RA_ADDRESS_REG, PT_R12(sp)
+#endif
+
+	PTR_SUBU a0, ra, 8	/* arg1: self address */
+	PTR_LA   t1, _stext
+	sltu     t2, a0, t1	/* t2 = (a0 < _stext) */
+	PTR_LA   t1, _etext
+	sltu     t3, t1, a0	/* t3 = (a0 > _etext) */
+	or       t1, t2, t3
+	beqz     t1, ftrace_call
+	 nop
+#if defined(KBUILD_MCOUNT_RA_ADDRESS) && defined(CONFIG_32BIT)
+	PTR_SUBU a0, a0, 16	/* arg1: adjust to module's recorded callsite */
+#else
+	PTR_SUBU a0, a0, 12
+#endif
+
+	.globl ftrace_call
+ftrace_call:
+	nop	/* a placeholder for the call to a real tracing function */
+	 move	a1, AT		/* arg2: parent's return address */
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+	.globl ftrace_graph_call
+ftrace_graph_call:
+	nop
+	 nop
+#endif
+
+	MCOUNT_RESTORE_REGS
+	.globl ftrace_stub
+ftrace_stub:
+	RETURN_BACK
+	END(ftrace_caller)
+
+#else	/* ! CONFIG_DYNAMIC_FTRACE */
+
+NESTED(_mcount, PT_SIZE, ra)
+EXPORT_SYMBOL(_mcount)
+	PTR_LA	t1, ftrace_stub
+	PTR_L	t2, ftrace_trace_function /* Prepare t2 for (1) */
+	beq	t1, t2, fgraph_trace
+	 nop
+
+	MCOUNT_SAVE_REGS
+
+	move	a0, ra		/* arg1: self return address */
+	jalr	t2		/* (1) call *ftrace_trace_function */
+	 move	a1, AT		/* arg2: parent's return address */
+
+	MCOUNT_RESTORE_REGS
+
+fgraph_trace:
+#ifdef	CONFIG_FUNCTION_GRAPH_TRACER
+	PTR_LA	t1, ftrace_stub
+	PTR_L	t3, ftrace_graph_return
+	bne	t1, t3, ftrace_graph_caller
+	 nop
+	PTR_LA	t1, ftrace_graph_entry_stub
+	PTR_L	t3, ftrace_graph_entry
+	bne	t1, t3, ftrace_graph_caller
+	 nop
+#endif
+
+#ifdef CONFIG_32BIT
+	addiu sp, sp, 8
+#endif
+
+	.globl ftrace_stub
+ftrace_stub:
+	RETURN_BACK
+	END(_mcount)
+
+#endif	/* ! CONFIG_DYNAMIC_FTRACE */
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+
+NESTED(ftrace_graph_caller, PT_SIZE, ra)
+#ifndef CONFIG_DYNAMIC_FTRACE
+	MCOUNT_SAVE_REGS
+#endif
+
+	/* arg1: Get the location of the parent's return address */
+#ifdef KBUILD_MCOUNT_RA_ADDRESS
+#ifdef CONFIG_DYNAMIC_FTRACE
+	PTR_L	a0, PT_R12(sp)
+#else
+	move	a0, MCOUNT_RA_ADDRESS_REG
+#endif
+	bnez	a0, 1f	/* non-leaf func: stored in MCOUNT_RA_ADDRESS_REG */
+	 nop
+#endif
+	PTR_LA	a0, PT_R1(sp)	/* leaf func: the location in current stack */
+1:
+
+	/* arg2: Get self return address */
+#ifdef CONFIG_DYNAMIC_FTRACE
+	PTR_L	a1, PT_R31(sp)
+#else
+	move	a1, ra
+#endif
+
+	/* arg3: Get frame pointer of current stack */
+#ifdef CONFIG_64BIT
+	PTR_LA	a2, PT_SIZE(sp)
+#else
+	PTR_LA	a2, (PT_SIZE+8)(sp)
+#endif
+
+	jal	prepare_ftrace_return
+	 nop
+	MCOUNT_RESTORE_REGS
+#ifndef CONFIG_DYNAMIC_FTRACE
+#ifdef CONFIG_32BIT
+	addiu sp, sp, 8
+#endif
+#endif
+	RETURN_BACK
+	END(ftrace_graph_caller)
+
+	.align	2
+	.globl	return_to_handler
+return_to_handler:
+	PTR_SUBU	sp, PT_SIZE
+	PTR_S	v0, PT_R2(sp)
+
+	jal	ftrace_return_to_handler
+	 PTR_S	v1, PT_R3(sp)
+
+	/* restore the real parent address: v0 -> ra */
+	move	ra, v0
+
+	PTR_L	v0, PT_R2(sp)
+	PTR_L	v1, PT_R3(sp)
+	jr	ra
+	 PTR_ADDIU	sp, PT_SIZE
+#endif /* CONFIG_FUNCTION_GRAPH_TRACER */
+
+	.set at
+	.set reorder
diff --git a/arch/mips/kernel/mips-cm.c b/arch/mips/kernel/mips-cm.c
new file mode 100644
index 000000000..72c8374a3
--- /dev/null
+++ b/arch/mips/kernel/mips-cm.c
@@ -0,0 +1,514 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2013 Imagination Technologies
+ * Author: Paul Burton <paul.burton@mips.com>
+ */
+
+#include <linux/errno.h>
+#include <linux/percpu.h>
+#include <linux/spinlock.h>
+
+#include <asm/mips-cps.h>
+#include <asm/mipsregs.h>
+
+void __iomem *mips_gcr_base;
+void __iomem *mips_cm_l2sync_base;
+int mips_cm_is64;
+
+static char *cm2_tr[8] = {
+	"mem",	"gcr",	"gic",	"mmio",
+	"0x04", "cpc", "0x06", "0x07"
+};
+
+/* CM3 Tag ECC transaction type */
+static char *cm3_tr[16] = {
+	[0x0] = "ReqNoData",
+	[0x1] = "0x1",
+	[0x2] = "ReqWData",
+	[0x3] = "0x3",
+	[0x4] = "IReqNoResp",
+	[0x5] = "IReqWResp",
+	[0x6] = "IReqNoRespDat",
+	[0x7] = "IReqWRespDat",
+	[0x8] = "RespNoData",
+	[0x9] = "RespDataFol",
+	[0xa] = "RespWData",
+	[0xb] = "RespDataOnly",
+	[0xc] = "IRespNoData",
+	[0xd] = "IRespDataFol",
+	[0xe] = "IRespWData",
+	[0xf] = "IRespDataOnly"
+};
+
+static char *cm2_cmd[32] = {
+	[0x00] = "0x00",
+	[0x01] = "Legacy Write",
+	[0x02] = "Legacy Read",
+	[0x03] = "0x03",
+	[0x04] = "0x04",
+	[0x05] = "0x05",
+	[0x06] = "0x06",
+	[0x07] = "0x07",
+	[0x08] = "Coherent Read Own",
+	[0x09] = "Coherent Read Share",
+	[0x0a] = "Coherent Read Discard",
+	[0x0b] = "Coherent Ready Share Always",
+	[0x0c] = "Coherent Upgrade",
+	[0x0d] = "Coherent Writeback",
+	[0x0e] = "0x0e",
+	[0x0f] = "0x0f",
+	[0x10] = "Coherent Copyback",
+	[0x11] = "Coherent Copyback Invalidate",
+	[0x12] = "Coherent Invalidate",
+	[0x13] = "Coherent Write Invalidate",
+	[0x14] = "Coherent Completion Sync",
+	[0x15] = "0x15",
+	[0x16] = "0x16",
+	[0x17] = "0x17",
+	[0x18] = "0x18",
+	[0x19] = "0x19",
+	[0x1a] = "0x1a",
+	[0x1b] = "0x1b",
+	[0x1c] = "0x1c",
+	[0x1d] = "0x1d",
+	[0x1e] = "0x1e",
+	[0x1f] = "0x1f"
+};
+
+/* CM3 Tag ECC command type */
+static char *cm3_cmd[16] = {
+	[0x0] = "Legacy Read",
+	[0x1] = "Legacy Write",
+	[0x2] = "Coherent Read Own",
+	[0x3] = "Coherent Read Share",
+	[0x4] = "Coherent Read Discard",
+	[0x5] = "Coherent Evicted",
+	[0x6] = "Coherent Upgrade",
+	[0x7] = "Coherent Upgrade for Store Conditional",
+	[0x8] = "Coherent Writeback",
+	[0x9] = "Coherent Write Invalidate",
+	[0xa] = "0xa",
+	[0xb] = "0xb",
+	[0xc] = "0xc",
+	[0xd] = "0xd",
+	[0xe] = "0xe",
+	[0xf] = "0xf"
+};
+
+/* CM3 Tag ECC command group */
+static char *cm3_cmd_group[8] = {
+	[0x0] = "Normal",
+	[0x1] = "Registers",
+	[0x2] = "TLB",
+	[0x3] = "0x3",
+	[0x4] = "L1I",
+	[0x5] = "L1D",
+	[0x6] = "L3",
+	[0x7] = "L2"
+};
+
+static char *cm2_core[8] = {
+	"Invalid/OK",	"Invalid/Data",
+	"Shared/OK",	"Shared/Data",
+	"Modified/OK",	"Modified/Data",
+	"Exclusive/OK", "Exclusive/Data"
+};
+
+static char *cm2_l2_type[4] = {
+	[0x0] = "None",
+	[0x1] = "Tag RAM single/double ECC error",
+	[0x2] = "Data RAM single/double ECC error",
+	[0x3] = "WS RAM uncorrectable dirty parity"
+};
+
+static char *cm2_l2_instr[32] = {
+	[0x00] = "L2_NOP",
+	[0x01] = "L2_ERR_CORR",
+	[0x02] = "L2_TAG_INV",
+	[0x03] = "L2_WS_CLEAN",
+	[0x04] = "L2_RD_MDYFY_WR",
+	[0x05] = "L2_WS_MRU",
+	[0x06] = "L2_EVICT_LN2",
+	[0x07] = "0x07",
+	[0x08] = "L2_EVICT",
+	[0x09] = "L2_REFL",
+	[0x0a] = "L2_RD",
+	[0x0b] = "L2_WR",
+	[0x0c] = "L2_EVICT_MRU",
+	[0x0d] = "L2_SYNC",
+	[0x0e] = "L2_REFL_ERR",
+	[0x0f] = "0x0f",
+	[0x10] = "L2_INDX_WB_INV",
+	[0x11] = "L2_INDX_LD_TAG",
+	[0x12] = "L2_INDX_ST_TAG",
+	[0x13] = "L2_INDX_ST_DATA",
+	[0x14] = "L2_INDX_ST_ECC",
+	[0x15] = "0x15",
+	[0x16] = "0x16",
+	[0x17] = "0x17",
+	[0x18] = "L2_FTCH_AND_LCK",
+	[0x19] = "L2_HIT_INV",
+	[0x1a] = "L2_HIT_WB_INV",
+	[0x1b] = "L2_HIT_WB",
+	[0x1c] = "0x1c",
+	[0x1d] = "0x1d",
+	[0x1e] = "0x1e",
+	[0x1f] = "0x1f"
+};
+
+static char *cm2_causes[32] = {
+	"None", "GC_WR_ERR", "GC_RD_ERR", "COH_WR_ERR",
+	"COH_RD_ERR", "MMIO_WR_ERR", "MMIO_RD_ERR", "0x07",
+	"0x08", "0x09", "0x0a", "0x0b",
+	"0x0c", "0x0d", "0x0e", "0x0f",
+	"0x10", "INTVN_WR_ERR", "INTVN_RD_ERR", "0x13",
+	"0x14", "0x15", "0x16", "0x17",
+	"L2_RD_UNCORR", "L2_WR_UNCORR", "L2_CORR", "0x1b",
+	"0x1c", "0x1d", "0x1e", "0x1f"
+};
+
+static char *cm3_causes[32] = {
+	"0x0", "MP_CORRECTABLE_ECC_ERR", "MP_REQUEST_DECODE_ERR",
+	"MP_UNCORRECTABLE_ECC_ERR", "MP_PARITY_ERR", "MP_COHERENCE_ERR",
+	"CMBIU_REQUEST_DECODE_ERR", "CMBIU_PARITY_ERR", "CMBIU_AXI_RESP_ERR",
+	"0x9", "RBI_BUS_ERR", "0xb", "0xc", "0xd", "0xe", "0xf", "0x10",
+	"0x11", "0x12", "0x13", "0x14", "0x15", "0x16", "0x17", "0x18",
+	"0x19", "0x1a", "0x1b", "0x1c", "0x1d", "0x1e", "0x1f"
+};
+
+static DEFINE_PER_CPU_ALIGNED(spinlock_t, cm_core_lock);
+static DEFINE_PER_CPU_ALIGNED(unsigned long, cm_core_lock_flags);
+
+phys_addr_t __mips_cm_phys_base(void)
+{
+	u32 config3 = read_c0_config3();
+	unsigned long cmgcr;
+
+	/* Check the CMGCRBase register is implemented */
+	if (!(config3 & MIPS_CONF3_CMGCR))
+		return 0;
+
+	/* Read the address from CMGCRBase */
+	cmgcr = read_c0_cmgcrbase();
+	return (cmgcr & MIPS_CMGCRF_BASE) << (36 - 32);
+}
+
+phys_addr_t mips_cm_phys_base(void)
+	__attribute__((weak, alias("__mips_cm_phys_base")));
+
+phys_addr_t __mips_cm_l2sync_phys_base(void)
+{
+	u32 base_reg;
+
+	/*
+	 * If the L2-only sync region is already enabled then leave it at it's
+	 * current location.
+	 */
+	base_reg = read_gcr_l2_only_sync_base();
+	if (base_reg & CM_GCR_L2_ONLY_SYNC_BASE_SYNCEN)
+		return base_reg & CM_GCR_L2_ONLY_SYNC_BASE_SYNCBASE;
+
+	/* Default to following the CM */
+	return mips_cm_phys_base() + MIPS_CM_GCR_SIZE;
+}
+
+phys_addr_t mips_cm_l2sync_phys_base(void)
+	__attribute__((weak, alias("__mips_cm_l2sync_phys_base")));
+
+static void mips_cm_probe_l2sync(void)
+{
+	unsigned major_rev;
+	phys_addr_t addr;
+
+	/* L2-only sync was introduced with CM major revision 6 */
+	major_rev = FIELD_GET(CM_GCR_REV_MAJOR, read_gcr_rev());
+	if (major_rev < 6)
+		return;
+
+	/* Find a location for the L2 sync region */
+	addr = mips_cm_l2sync_phys_base();
+	BUG_ON((addr & CM_GCR_L2_ONLY_SYNC_BASE_SYNCBASE) != addr);
+	if (!addr)
+		return;
+
+	/* Set the region base address & enable it */
+	write_gcr_l2_only_sync_base(addr | CM_GCR_L2_ONLY_SYNC_BASE_SYNCEN);
+
+	/* Map the region */
+	mips_cm_l2sync_base = ioremap(addr, MIPS_CM_L2SYNC_SIZE);
+}
+
+int mips_cm_probe(void)
+{
+	phys_addr_t addr;
+	u32 base_reg;
+	unsigned cpu;
+
+	/*
+	 * No need to probe again if we have already been
+	 * here before.
+	 */
+	if (mips_gcr_base)
+		return 0;
+
+	addr = mips_cm_phys_base();
+	BUG_ON((addr & CM_GCR_BASE_GCRBASE) != addr);
+	if (!addr)
+		return -ENODEV;
+
+	mips_gcr_base = ioremap(addr, MIPS_CM_GCR_SIZE);
+	if (!mips_gcr_base)
+		return -ENXIO;
+
+	/* sanity check that we're looking at a CM */
+	base_reg = read_gcr_base();
+	if ((base_reg & CM_GCR_BASE_GCRBASE) != addr) {
+		pr_err("GCRs appear to have been moved (expected them at 0x%08lx)!\n",
+		       (unsigned long)addr);
+		mips_gcr_base = NULL;
+		return -ENODEV;
+	}
+
+	/* set default target to memory */
+	change_gcr_base(CM_GCR_BASE_CMDEFTGT, CM_GCR_BASE_CMDEFTGT_MEM);
+
+	/* disable CM regions */
+	write_gcr_reg0_base(CM_GCR_REGn_BASE_BASEADDR);
+	write_gcr_reg0_mask(CM_GCR_REGn_MASK_ADDRMASK);
+	write_gcr_reg1_base(CM_GCR_REGn_BASE_BASEADDR);
+	write_gcr_reg1_mask(CM_GCR_REGn_MASK_ADDRMASK);
+	write_gcr_reg2_base(CM_GCR_REGn_BASE_BASEADDR);
+	write_gcr_reg2_mask(CM_GCR_REGn_MASK_ADDRMASK);
+	write_gcr_reg3_base(CM_GCR_REGn_BASE_BASEADDR);
+	write_gcr_reg3_mask(CM_GCR_REGn_MASK_ADDRMASK);
+
+	/* probe for an L2-only sync region */
+	mips_cm_probe_l2sync();
+
+	/* determine register width for this CM */
+	mips_cm_is64 = IS_ENABLED(CONFIG_64BIT) && (mips_cm_revision() >= CM_REV_CM3);
+
+	for_each_possible_cpu(cpu)
+		spin_lock_init(&per_cpu(cm_core_lock, cpu));
+
+	return 0;
+}
+
+void mips_cm_lock_other(unsigned int cluster, unsigned int core,
+			unsigned int vp, unsigned int block)
+{
+	unsigned int curr_core, cm_rev;
+	u32 val;
+
+	cm_rev = mips_cm_revision();
+	preempt_disable();
+
+	if (cm_rev >= CM_REV_CM3) {
+		val = FIELD_PREP(CM3_GCR_Cx_OTHER_CORE, core) |
+		      FIELD_PREP(CM3_GCR_Cx_OTHER_VP, vp);
+
+		if (cm_rev >= CM_REV_CM3_5) {
+			val |= CM_GCR_Cx_OTHER_CLUSTER_EN;
+			val |= FIELD_PREP(CM_GCR_Cx_OTHER_CLUSTER, cluster);
+			val |= FIELD_PREP(CM_GCR_Cx_OTHER_BLOCK, block);
+		} else {
+			WARN_ON(cluster != 0);
+			WARN_ON(block != CM_GCR_Cx_OTHER_BLOCK_LOCAL);
+		}
+
+		/*
+		 * We need to disable interrupts in SMP systems in order to
+		 * ensure that we don't interrupt the caller with code which
+		 * may modify the redirect register. We do so here in a
+		 * slightly obscure way by using a spin lock, since this has
+		 * the neat property of also catching any nested uses of
+		 * mips_cm_lock_other() leading to a deadlock or a nice warning
+		 * with lockdep enabled.
+		 */
+		spin_lock_irqsave(this_cpu_ptr(&cm_core_lock),
+				  *this_cpu_ptr(&cm_core_lock_flags));
+	} else {
+		WARN_ON(cluster != 0);
+		WARN_ON(block != CM_GCR_Cx_OTHER_BLOCK_LOCAL);
+
+		/*
+		 * We only have a GCR_CL_OTHER per core in systems with
+		 * CM 2.5 & older, so have to ensure other VP(E)s don't
+		 * race with us.
+		 */
+		curr_core = cpu_core(&current_cpu_data);
+		spin_lock_irqsave(&per_cpu(cm_core_lock, curr_core),
+				  per_cpu(cm_core_lock_flags, curr_core));
+
+		val = FIELD_PREP(CM_GCR_Cx_OTHER_CORENUM, core);
+	}
+
+	write_gcr_cl_other(val);
+
+	/*
+	 * Ensure the core-other region reflects the appropriate core &
+	 * VP before any accesses to it occur.
+	 */
+	mb();
+}
+
+void mips_cm_unlock_other(void)
+{
+	unsigned int curr_core;
+
+	if (mips_cm_revision() < CM_REV_CM3) {
+		curr_core = cpu_core(&current_cpu_data);
+		spin_unlock_irqrestore(&per_cpu(cm_core_lock, curr_core),
+				       per_cpu(cm_core_lock_flags, curr_core));
+	} else {
+		spin_unlock_irqrestore(this_cpu_ptr(&cm_core_lock),
+				       *this_cpu_ptr(&cm_core_lock_flags));
+	}
+
+	preempt_enable();
+}
+
+void mips_cm_error_report(void)
+{
+	u64 cm_error, cm_addr, cm_other;
+	unsigned long revision;
+	int ocause, cause;
+	char buf[256];
+
+	if (!mips_cm_present())
+		return;
+
+	revision = mips_cm_revision();
+	cm_error = read_gcr_error_cause();
+	cm_addr = read_gcr_error_addr();
+	cm_other = read_gcr_error_mult();
+
+	if (revision < CM_REV_CM3) { /* CM2 */
+		cause = FIELD_GET(CM_GCR_ERROR_CAUSE_ERRTYPE, cm_error);
+		ocause = FIELD_GET(CM_GCR_ERROR_MULT_ERR2ND, cm_other);
+
+		if (!cause)
+			return;
+
+		if (cause < 16) {
+			unsigned long cca_bits = (cm_error >> 15) & 7;
+			unsigned long tr_bits = (cm_error >> 12) & 7;
+			unsigned long cmd_bits = (cm_error >> 7) & 0x1f;
+			unsigned long stag_bits = (cm_error >> 3) & 15;
+			unsigned long sport_bits = (cm_error >> 0) & 7;
+
+			snprintf(buf, sizeof(buf),
+				 "CCA=%lu TR=%s MCmd=%s STag=%lu "
+				 "SPort=%lu\n", cca_bits, cm2_tr[tr_bits],
+				 cm2_cmd[cmd_bits], stag_bits, sport_bits);
+		} else if (cause < 24) {
+			/* glob state & sresp together */
+			unsigned long c3_bits = (cm_error >> 18) & 7;
+			unsigned long c2_bits = (cm_error >> 15) & 7;
+			unsigned long c1_bits = (cm_error >> 12) & 7;
+			unsigned long c0_bits = (cm_error >> 9) & 7;
+			unsigned long sc_bit = (cm_error >> 8) & 1;
+			unsigned long cmd_bits = (cm_error >> 3) & 0x1f;
+			unsigned long sport_bits = (cm_error >> 0) & 7;
+
+			snprintf(buf, sizeof(buf),
+				 "C3=%s C2=%s C1=%s C0=%s SC=%s "
+				 "MCmd=%s SPort=%lu\n",
+				 cm2_core[c3_bits], cm2_core[c2_bits],
+				 cm2_core[c1_bits], cm2_core[c0_bits],
+				 sc_bit ? "True" : "False",
+				 cm2_cmd[cmd_bits], sport_bits);
+		} else {
+			unsigned long muc_bit = (cm_error >> 23) & 1;
+			unsigned long ins_bits = (cm_error >> 18) & 0x1f;
+			unsigned long arr_bits = (cm_error >> 16) & 3;
+			unsigned long dw_bits = (cm_error >> 12) & 15;
+			unsigned long way_bits = (cm_error >> 9) & 7;
+			unsigned long mway_bit = (cm_error >> 8) & 1;
+			unsigned long syn_bits = (cm_error >> 0) & 0xFF;
+
+			snprintf(buf, sizeof(buf),
+				 "Type=%s%s Instr=%s DW=%lu Way=%lu "
+				 "MWay=%s Syndrome=0x%02lx",
+				 muc_bit ? "Multi-UC " : "",
+				 cm2_l2_type[arr_bits],
+				 cm2_l2_instr[ins_bits], dw_bits, way_bits,
+				 mway_bit ? "True" : "False", syn_bits);
+		}
+		pr_err("CM_ERROR=%08llx %s <%s>\n", cm_error,
+		       cm2_causes[cause], buf);
+		pr_err("CM_ADDR =%08llx\n", cm_addr);
+		pr_err("CM_OTHER=%08llx %s\n", cm_other, cm2_causes[ocause]);
+	} else { /* CM3 */
+		ulong core_id_bits, vp_id_bits, cmd_bits, cmd_group_bits;
+		ulong cm3_cca_bits, mcp_bits, cm3_tr_bits, sched_bit;
+
+		cause = FIELD_GET(CM3_GCR_ERROR_CAUSE_ERRTYPE, cm_error);
+		ocause = FIELD_GET(CM_GCR_ERROR_MULT_ERR2ND, cm_other);
+
+		if (!cause)
+			return;
+
+		/* Used by cause == {1,2,3} */
+		core_id_bits = (cm_error >> 22) & 0xf;
+		vp_id_bits = (cm_error >> 18) & 0xf;
+		cmd_bits = (cm_error >> 14) & 0xf;
+		cmd_group_bits = (cm_error >> 11) & 0xf;
+		cm3_cca_bits = (cm_error >> 8) & 7;
+		mcp_bits = (cm_error >> 5) & 0xf;
+		cm3_tr_bits = (cm_error >> 1) & 0xf;
+		sched_bit = cm_error & 0x1;
+
+		if (cause == 1 || cause == 3) { /* Tag ECC */
+			unsigned long tag_ecc = (cm_error >> 57) & 0x1;
+			unsigned long tag_way_bits = (cm_error >> 29) & 0xffff;
+			unsigned long dword_bits = (cm_error >> 49) & 0xff;
+			unsigned long data_way_bits = (cm_error >> 45) & 0xf;
+			unsigned long data_sets_bits = (cm_error >> 29) & 0xfff;
+			unsigned long bank_bit = (cm_error >> 28) & 0x1;
+			snprintf(buf, sizeof(buf),
+				 "%s ECC Error: Way=%lu (DWORD=%lu, Sets=%lu)"
+				 "Bank=%lu CoreID=%lu VPID=%lu Command=%s"
+				 "Command Group=%s CCA=%lu MCP=%d"
+				 "Transaction type=%s Scheduler=%lu\n",
+				 tag_ecc ? "TAG" : "DATA",
+				 tag_ecc ? (unsigned long)ffs(tag_way_bits) - 1 :
+				 data_way_bits, bank_bit, dword_bits,
+				 data_sets_bits,
+				 core_id_bits, vp_id_bits,
+				 cm3_cmd[cmd_bits],
+				 cm3_cmd_group[cmd_group_bits],
+				 cm3_cca_bits, 1 << mcp_bits,
+				 cm3_tr[cm3_tr_bits], sched_bit);
+		} else if (cause == 2) {
+			unsigned long data_error_type = (cm_error >> 41) & 0xfff;
+			unsigned long data_decode_cmd = (cm_error >> 37) & 0xf;
+			unsigned long data_decode_group = (cm_error >> 34) & 0x7;
+			unsigned long data_decode_destination_id = (cm_error >> 28) & 0x3f;
+
+			snprintf(buf, sizeof(buf),
+				 "Decode Request Error: Type=%lu, Command=%lu"
+				 "Command Group=%lu Destination ID=%lu"
+				 "CoreID=%lu VPID=%lu Command=%s"
+				 "Command Group=%s CCA=%lu MCP=%d"
+				 "Transaction type=%s Scheduler=%lu\n",
+				 data_error_type, data_decode_cmd,
+				 data_decode_group, data_decode_destination_id,
+				 core_id_bits, vp_id_bits,
+				 cm3_cmd[cmd_bits],
+				 cm3_cmd_group[cmd_group_bits],
+				 cm3_cca_bits, 1 << mcp_bits,
+				 cm3_tr[cm3_tr_bits], sched_bit);
+		} else {
+			buf[0] = 0;
+		}
+
+		pr_err("CM_ERROR=%llx %s <%s>\n", cm_error,
+		       cm3_causes[cause], buf);
+		pr_err("CM_ADDR =%llx\n", cm_addr);
+		pr_err("CM_OTHER=%llx %s\n", cm_other, cm3_causes[ocause]);
+	}
+
+	/* reprime cause register */
+	write_gcr_error_cause(cm_error);
+}
diff --git a/arch/mips/kernel/mips-cpc.c b/arch/mips/kernel/mips-cpc.c
new file mode 100644
index 000000000..d005be84c
--- /dev/null
+++ b/arch/mips/kernel/mips-cpc.c
@@ -0,0 +1,122 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2013 Imagination Technologies
+ * Author: Paul Burton <paul.burton@mips.com>
+ */
+
+#include <linux/errno.h>
+#include <linux/percpu.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/spinlock.h>
+
+#include <asm/mips-cps.h>
+
+void __iomem *mips_cpc_base;
+
+static DEFINE_PER_CPU_ALIGNED(spinlock_t, cpc_core_lock);
+
+static DEFINE_PER_CPU_ALIGNED(unsigned long, cpc_core_lock_flags);
+
+phys_addr_t __weak mips_cpc_default_phys_base(void)
+{
+	struct device_node *cpc_node;
+	struct resource res;
+	int err;
+
+	cpc_node = of_find_compatible_node(of_root, NULL, "mti,mips-cpc");
+	if (cpc_node) {
+		err = of_address_to_resource(cpc_node, 0, &res);
+		of_node_put(cpc_node);
+		if (!err)
+			return res.start;
+	}
+
+	return 0;
+}
+
+/**
+ * mips_cpc_phys_base - retrieve the physical base address of the CPC
+ *
+ * This function returns the physical base address of the Cluster Power
+ * Controller memory mapped registers, or 0 if no Cluster Power Controller
+ * is present.
+ */
+static phys_addr_t mips_cpc_phys_base(void)
+{
+	unsigned long cpc_base;
+
+	if (!mips_cm_present())
+		return 0;
+
+	if (!(read_gcr_cpc_status() & CM_GCR_CPC_STATUS_EX))
+		return 0;
+
+	/* If the CPC is already enabled, leave it so */
+	cpc_base = read_gcr_cpc_base();
+	if (cpc_base & CM_GCR_CPC_BASE_CPCEN)
+		return cpc_base & CM_GCR_CPC_BASE_CPCBASE;
+
+	/* Otherwise, use the default address */
+	cpc_base = mips_cpc_default_phys_base();
+	if (!cpc_base)
+		return cpc_base;
+
+	/* Enable the CPC, mapped at the default address */
+	write_gcr_cpc_base(cpc_base | CM_GCR_CPC_BASE_CPCEN);
+	return cpc_base;
+}
+
+int mips_cpc_probe(void)
+{
+	phys_addr_t addr;
+	unsigned int cpu;
+
+	for_each_possible_cpu(cpu)
+		spin_lock_init(&per_cpu(cpc_core_lock, cpu));
+
+	addr = mips_cpc_phys_base();
+	if (!addr)
+		return -ENODEV;
+
+	mips_cpc_base = ioremap(addr, 0x8000);
+	if (!mips_cpc_base)
+		return -ENXIO;
+
+	return 0;
+}
+
+void mips_cpc_lock_other(unsigned int core)
+{
+	unsigned int curr_core;
+
+	if (mips_cm_revision() >= CM_REV_CM3)
+		/* Systems with CM >= 3 lock the CPC via mips_cm_lock_other */
+		return;
+
+	preempt_disable();
+	curr_core = cpu_core(&current_cpu_data);
+	spin_lock_irqsave(&per_cpu(cpc_core_lock, curr_core),
+			  per_cpu(cpc_core_lock_flags, curr_core));
+	write_cpc_cl_other(core << __ffs(CPC_Cx_OTHER_CORENUM));
+
+	/*
+	 * Ensure the core-other region reflects the appropriate core &
+	 * VP before any accesses to it occur.
+	 */
+	mb();
+}
+
+void mips_cpc_unlock_other(void)
+{
+	unsigned int curr_core;
+
+	if (mips_cm_revision() >= CM_REV_CM3)
+		/* Systems with CM >= 3 lock the CPC via mips_cm_lock_other */
+		return;
+
+	curr_core = cpu_core(&current_cpu_data);
+	spin_unlock_irqrestore(&per_cpu(cpc_core_lock, curr_core),
+			       per_cpu(cpc_core_lock_flags, curr_core));
+	preempt_enable();
+}
diff --git a/arch/mips/kernel/mips-mt-fpaff.c b/arch/mips/kernel/mips-mt-fpaff.c
new file mode 100644
index 000000000..6c590ef27
--- /dev/null
+++ b/arch/mips/kernel/mips-mt-fpaff.c
@@ -0,0 +1,219 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * General MIPS MT support routines, usable in AP/SP and SMVP.
+ * Copyright (C) 2005 Mips Technologies, Inc
+ */
+#include <linux/cpu.h>
+#include <linux/cpuset.h>
+#include <linux/cpumask.h>
+#include <linux/delay.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/sched/task.h>
+#include <linux/cred.h>
+#include <linux/security.h>
+#include <linux/types.h>
+#include <linux/uaccess.h>
+
+/*
+ * CPU mask used to set process affinity for MT VPEs/TCs with FPUs
+ */
+cpumask_t mt_fpu_cpumask;
+
+static int fpaff_threshold = -1;
+unsigned long mt_fpemul_threshold;
+
+/*
+ * Replacement functions for the sys_sched_setaffinity() and
+ * sys_sched_getaffinity() system calls, so that we can integrate
+ * FPU affinity with the user's requested processor affinity.
+ * This code is 98% identical with the sys_sched_setaffinity()
+ * and sys_sched_getaffinity() system calls, and should be
+ * updated when kernel/sched/core.c changes.
+ */
+
+/*
+ * find_process_by_pid - find a process with a matching PID value.
+ * used in sys_sched_set/getaffinity() in kernel/sched/core.c, so
+ * cloned here.
+ */
+static inline struct task_struct *find_process_by_pid(pid_t pid)
+{
+	return pid ? find_task_by_vpid(pid) : current;
+}
+
+/*
+ * check the target process has a UID that matches the current process's
+ */
+static bool check_same_owner(struct task_struct *p)
+{
+	const struct cred *cred = current_cred(), *pcred;
+	bool match;
+
+	rcu_read_lock();
+	pcred = __task_cred(p);
+	match = (uid_eq(cred->euid, pcred->euid) ||
+		 uid_eq(cred->euid, pcred->uid));
+	rcu_read_unlock();
+	return match;
+}
+
+/*
+ * mipsmt_sys_sched_setaffinity - set the cpu affinity of a process
+ */
+asmlinkage long mipsmt_sys_sched_setaffinity(pid_t pid, unsigned int len,
+				      unsigned long __user *user_mask_ptr)
+{
+	cpumask_var_t cpus_allowed, new_mask, effective_mask;
+	struct thread_info *ti;
+	struct task_struct *p;
+	int retval;
+
+	if (len < sizeof(new_mask))
+		return -EINVAL;
+
+	if (copy_from_user(&new_mask, user_mask_ptr, sizeof(new_mask)))
+		return -EFAULT;
+
+	get_online_cpus();
+	rcu_read_lock();
+
+	p = find_process_by_pid(pid);
+	if (!p) {
+		rcu_read_unlock();
+		put_online_cpus();
+		return -ESRCH;
+	}
+
+	/* Prevent p going away */
+	get_task_struct(p);
+	rcu_read_unlock();
+
+	if (!alloc_cpumask_var(&cpus_allowed, GFP_KERNEL)) {
+		retval = -ENOMEM;
+		goto out_put_task;
+	}
+	if (!alloc_cpumask_var(&new_mask, GFP_KERNEL)) {
+		retval = -ENOMEM;
+		goto out_free_cpus_allowed;
+	}
+	if (!alloc_cpumask_var(&effective_mask, GFP_KERNEL)) {
+		retval = -ENOMEM;
+		goto out_free_new_mask;
+	}
+	if (!check_same_owner(p) && !capable(CAP_SYS_NICE)) {
+		retval = -EPERM;
+		goto out_unlock;
+	}
+
+	retval = security_task_setscheduler(p);
+	if (retval)
+		goto out_unlock;
+
+	/* Record new user-specified CPU set for future reference */
+	cpumask_copy(&p->thread.user_cpus_allowed, new_mask);
+
+ again:
+	/* Compute new global allowed CPU set if necessary */
+	ti = task_thread_info(p);
+	if (test_ti_thread_flag(ti, TIF_FPUBOUND) &&
+	    cpumask_intersects(new_mask, &mt_fpu_cpumask)) {
+		cpumask_and(effective_mask, new_mask, &mt_fpu_cpumask);
+		retval = set_cpus_allowed_ptr(p, effective_mask);
+	} else {
+		cpumask_copy(effective_mask, new_mask);
+		clear_ti_thread_flag(ti, TIF_FPUBOUND);
+		retval = set_cpus_allowed_ptr(p, new_mask);
+	}
+
+	if (!retval) {
+		cpuset_cpus_allowed(p, cpus_allowed);
+		if (!cpumask_subset(effective_mask, cpus_allowed)) {
+			/*
+			 * We must have raced with a concurrent cpuset
+			 * update. Just reset the cpus_allowed to the
+			 * cpuset's cpus_allowed
+			 */
+			cpumask_copy(new_mask, cpus_allowed);
+			goto again;
+		}
+	}
+out_unlock:
+	free_cpumask_var(effective_mask);
+out_free_new_mask:
+	free_cpumask_var(new_mask);
+out_free_cpus_allowed:
+	free_cpumask_var(cpus_allowed);
+out_put_task:
+	put_task_struct(p);
+	put_online_cpus();
+	return retval;
+}
+
+/*
+ * mipsmt_sys_sched_getaffinity - get the cpu affinity of a process
+ */
+asmlinkage long mipsmt_sys_sched_getaffinity(pid_t pid, unsigned int len,
+				      unsigned long __user *user_mask_ptr)
+{
+	unsigned int real_len;
+	cpumask_t allowed, mask;
+	int retval;
+	struct task_struct *p;
+
+	real_len = sizeof(mask);
+	if (len < real_len)
+		return -EINVAL;
+
+	get_online_cpus();
+	rcu_read_lock();
+
+	retval = -ESRCH;
+	p = find_process_by_pid(pid);
+	if (!p)
+		goto out_unlock;
+	retval = security_task_getscheduler(p);
+	if (retval)
+		goto out_unlock;
+
+	cpumask_or(&allowed, &p->thread.user_cpus_allowed, p->cpus_ptr);
+	cpumask_and(&mask, &allowed, cpu_active_mask);
+
+out_unlock:
+	rcu_read_unlock();
+	put_online_cpus();
+	if (retval)
+		return retval;
+	if (copy_to_user(user_mask_ptr, &mask, real_len))
+		return -EFAULT;
+	return real_len;
+}
+
+
+static int __init fpaff_thresh(char *str)
+{
+	get_option(&str, &fpaff_threshold);
+	return 1;
+}
+__setup("fpaff=", fpaff_thresh);
+
+/*
+ * FPU Use Factor empirically derived from experiments on 34K
+ */
+#define FPUSEFACTOR 2000
+
+static __init int mt_fp_affinity_init(void)
+{
+	if (fpaff_threshold >= 0) {
+		mt_fpemul_threshold = fpaff_threshold;
+	} else {
+		mt_fpemul_threshold =
+			(FPUSEFACTOR * (loops_per_jiffy/(500000/HZ))) / HZ;
+	}
+	printk(KERN_DEBUG "FPU Affinity set after %ld emulations\n",
+	       mt_fpemul_threshold);
+
+	return 0;
+}
+arch_initcall(mt_fp_affinity_init);
diff --git a/arch/mips/kernel/mips-mt.c b/arch/mips/kernel/mips-mt.c
new file mode 100644
index 000000000..d5f7362e8
--- /dev/null
+++ b/arch/mips/kernel/mips-mt.c
@@ -0,0 +1,246 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * General MIPS MT support routines, usable in AP/SP and SMVP.
+ * Copyright (C) 2005 Mips Technologies, Inc
+ */
+
+#include <linux/device.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/export.h>
+#include <linux/interrupt.h>
+#include <linux/security.h>
+
+#include <asm/cpu.h>
+#include <asm/processor.h>
+#include <linux/atomic.h>
+#include <asm/hardirq.h>
+#include <asm/mmu_context.h>
+#include <asm/mipsmtregs.h>
+#include <asm/r4kcache.h>
+#include <asm/cacheflush.h>
+
+int vpelimit;
+
+static int __init maxvpes(char *str)
+{
+	get_option(&str, &vpelimit);
+
+	return 1;
+}
+
+__setup("maxvpes=", maxvpes);
+
+int tclimit;
+
+static int __init maxtcs(char *str)
+{
+	get_option(&str, &tclimit);
+
+	return 1;
+}
+
+__setup("maxtcs=", maxtcs);
+
+/*
+ * Dump new MIPS MT state for the core. Does not leave TCs halted.
+ * Takes an argument which taken to be a pre-call MVPControl value.
+ */
+
+void mips_mt_regdump(unsigned long mvpctl)
+{
+	unsigned long flags;
+	unsigned long vpflags;
+	unsigned long mvpconf0;
+	int nvpe;
+	int ntc;
+	int i;
+	int tc;
+	unsigned long haltval;
+	unsigned long tcstatval;
+
+	local_irq_save(flags);
+	vpflags = dvpe();
+	printk("=== MIPS MT State Dump ===\n");
+	printk("-- Global State --\n");
+	printk("   MVPControl Passed: %08lx\n", mvpctl);
+	printk("   MVPControl Read: %08lx\n", vpflags);
+	printk("   MVPConf0 : %08lx\n", (mvpconf0 = read_c0_mvpconf0()));
+	nvpe = ((mvpconf0 & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT) + 1;
+	ntc = ((mvpconf0 & MVPCONF0_PTC) >> MVPCONF0_PTC_SHIFT) + 1;
+	printk("-- per-VPE State --\n");
+	for (i = 0; i < nvpe; i++) {
+		for (tc = 0; tc < ntc; tc++) {
+			settc(tc);
+			if ((read_tc_c0_tcbind() & TCBIND_CURVPE) == i) {
+				printk("  VPE %d\n", i);
+				printk("   VPEControl : %08lx\n",
+				       read_vpe_c0_vpecontrol());
+				printk("   VPEConf0 : %08lx\n",
+				       read_vpe_c0_vpeconf0());
+				printk("   VPE%d.Status : %08lx\n",
+				       i, read_vpe_c0_status());
+				printk("   VPE%d.EPC : %08lx %pS\n",
+				       i, read_vpe_c0_epc(),
+				       (void *) read_vpe_c0_epc());
+				printk("   VPE%d.Cause : %08lx\n",
+				       i, read_vpe_c0_cause());
+				printk("   VPE%d.Config7 : %08lx\n",
+				       i, read_vpe_c0_config7());
+				break; /* Next VPE */
+			}
+		}
+	}
+	printk("-- per-TC State --\n");
+	for (tc = 0; tc < ntc; tc++) {
+		settc(tc);
+		if (read_tc_c0_tcbind() == read_c0_tcbind()) {
+			/* Are we dumping ourself?  */
+			haltval = 0; /* Then we're not halted, and mustn't be */
+			tcstatval = flags; /* And pre-dump TCStatus is flags */
+			printk("  TC %d (current TC with VPE EPC above)\n", tc);
+		} else {
+			haltval = read_tc_c0_tchalt();
+			write_tc_c0_tchalt(1);
+			tcstatval = read_tc_c0_tcstatus();
+			printk("  TC %d\n", tc);
+		}
+		printk("   TCStatus : %08lx\n", tcstatval);
+		printk("   TCBind : %08lx\n", read_tc_c0_tcbind());
+		printk("   TCRestart : %08lx %pS\n",
+		       read_tc_c0_tcrestart(), (void *) read_tc_c0_tcrestart());
+		printk("   TCHalt : %08lx\n", haltval);
+		printk("   TCContext : %08lx\n", read_tc_c0_tccontext());
+		if (!haltval)
+			write_tc_c0_tchalt(0);
+	}
+	printk("===========================\n");
+	evpe(vpflags);
+	local_irq_restore(flags);
+}
+
+static int mt_opt_rpsctl = -1;
+static int mt_opt_nblsu = -1;
+static int mt_opt_forceconfig7;
+static int mt_opt_config7 = -1;
+
+static int __init rpsctl_set(char *str)
+{
+	get_option(&str, &mt_opt_rpsctl);
+	return 1;
+}
+__setup("rpsctl=", rpsctl_set);
+
+static int __init nblsu_set(char *str)
+{
+	get_option(&str, &mt_opt_nblsu);
+	return 1;
+}
+__setup("nblsu=", nblsu_set);
+
+static int __init config7_set(char *str)
+{
+	get_option(&str, &mt_opt_config7);
+	mt_opt_forceconfig7 = 1;
+	return 1;
+}
+__setup("config7=", config7_set);
+
+static unsigned int itc_base;
+
+static int __init set_itc_base(char *str)
+{
+	get_option(&str, &itc_base);
+	return 1;
+}
+
+__setup("itcbase=", set_itc_base);
+
+void mips_mt_set_cpuoptions(void)
+{
+	unsigned int oconfig7 = read_c0_config7();
+	unsigned int nconfig7 = oconfig7;
+
+	if (mt_opt_rpsctl >= 0) {
+		printk("34K return prediction stack override set to %d.\n",
+			mt_opt_rpsctl);
+		if (mt_opt_rpsctl)
+			nconfig7 |= (1 << 2);
+		else
+			nconfig7 &= ~(1 << 2);
+	}
+	if (mt_opt_nblsu >= 0) {
+		printk("34K ALU/LSU sync override set to %d.\n", mt_opt_nblsu);
+		if (mt_opt_nblsu)
+			nconfig7 |= (1 << 5);
+		else
+			nconfig7 &= ~(1 << 5);
+	}
+	if (mt_opt_forceconfig7) {
+		printk("CP0.Config7 forced to 0x%08x.\n", mt_opt_config7);
+		nconfig7 = mt_opt_config7;
+	}
+	if (oconfig7 != nconfig7) {
+		__asm__ __volatile("sync");
+		write_c0_config7(nconfig7);
+		ehb();
+		printk("Config7: 0x%08x\n", read_c0_config7());
+	}
+
+	if (itc_base != 0) {
+		/*
+		 * Configure ITC mapping.  This code is very
+		 * specific to the 34K core family, which uses
+		 * a special mode bit ("ITC") in the ErrCtl
+		 * register to enable access to ITC control
+		 * registers via cache "tag" operations.
+		 */
+		unsigned long ectlval;
+		unsigned long itcblkgrn;
+
+		/* ErrCtl register is known as "ecc" to Linux */
+		ectlval = read_c0_ecc();
+		write_c0_ecc(ectlval | (0x1 << 26));
+		ehb();
+#define INDEX_0 (0x80000000)
+#define INDEX_8 (0x80000008)
+		/* Read "cache tag" for Dcache pseudo-index 8 */
+		cache_op(Index_Load_Tag_D, INDEX_8);
+		ehb();
+		itcblkgrn = read_c0_dtaglo();
+		itcblkgrn &= 0xfffe0000;
+		/* Set for 128 byte pitch of ITC cells */
+		itcblkgrn |= 0x00000c00;
+		/* Stage in Tag register */
+		write_c0_dtaglo(itcblkgrn);
+		ehb();
+		/* Write out to ITU with CACHE op */
+		cache_op(Index_Store_Tag_D, INDEX_8);
+		/* Now set base address, and turn ITC on with 0x1 bit */
+		write_c0_dtaglo((itc_base & 0xfffffc00) | 0x1 );
+		ehb();
+		/* Write out to ITU with CACHE op */
+		cache_op(Index_Store_Tag_D, INDEX_0);
+		write_c0_ecc(ectlval);
+		ehb();
+		printk("Mapped %ld ITC cells starting at 0x%08x\n",
+			((itcblkgrn & 0x7fe00000) >> 20), itc_base);
+	}
+}
+
+struct class *mt_class;
+
+static int __init mt_init(void)
+{
+	struct class *mtc;
+
+	mtc = class_create(THIS_MODULE, "mt");
+	if (IS_ERR(mtc))
+		return PTR_ERR(mtc);
+
+	mt_class = mtc;
+
+	return 0;
+}
+
+subsys_initcall(mt_init);
diff --git a/arch/mips/kernel/mips-r2-to-r6-emul.c b/arch/mips/kernel/mips-r2-to-r6-emul.c
new file mode 100644
index 000000000..a39ec755e
--- /dev/null
+++ b/arch/mips/kernel/mips-r2-to-r6-emul.c
@@ -0,0 +1,2363 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (c) 2014 Imagination Technologies Ltd.
+ * Author: Leonid Yegoshin <Leonid.Yegoshin@imgtec.com>
+ * Author: Markos Chandras <markos.chandras@imgtec.com>
+ *
+ *      MIPS R2 user space instruction emulator for MIPS R6
+ *
+ */
+#include <linux/bug.h>
+#include <linux/compiler.h>
+#include <linux/debugfs.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/ptrace.h>
+#include <linux/seq_file.h>
+
+#include <asm/asm.h>
+#include <asm/branch.h>
+#include <asm/break.h>
+#include <asm/debug.h>
+#include <asm/fpu.h>
+#include <asm/fpu_emulator.h>
+#include <asm/inst.h>
+#include <asm/mips-r2-to-r6-emul.h>
+#include <asm/local.h>
+#include <asm/mipsregs.h>
+#include <asm/ptrace.h>
+#include <linux/uaccess.h>
+
+#ifdef CONFIG_64BIT
+#define ADDIU	"daddiu "
+#define INS	"dins "
+#define EXT	"dext "
+#else
+#define ADDIU	"addiu "
+#define INS	"ins "
+#define EXT	"ext "
+#endif /* CONFIG_64BIT */
+
+#define SB	"sb "
+#define LB	"lb "
+#define LL	"ll "
+#define SC	"sc "
+
+#ifdef CONFIG_DEBUG_FS
+static DEFINE_PER_CPU(struct mips_r2_emulator_stats, mipsr2emustats);
+static DEFINE_PER_CPU(struct mips_r2_emulator_stats, mipsr2bdemustats);
+static DEFINE_PER_CPU(struct mips_r2br_emulator_stats, mipsr2bremustats);
+#endif
+
+extern const unsigned int fpucondbit[8];
+
+#define MIPS_R2_EMUL_TOTAL_PASS	10
+
+int mipsr2_emulation = 0;
+
+static int __init mipsr2emu_enable(char *s)
+{
+	mipsr2_emulation = 1;
+
+	pr_info("MIPS R2-to-R6 Emulator Enabled!");
+
+	return 1;
+}
+__setup("mipsr2emu", mipsr2emu_enable);
+
+/**
+ * mipsr6_emul - Emulate some frequent R2/R5/R6 instructions in delay slot
+ * for performance instead of the traditional way of using a stack trampoline
+ * which is rather slow.
+ * @regs: Process register set
+ * @ir: Instruction
+ */
+static inline int mipsr6_emul(struct pt_regs *regs, u32 ir)
+{
+	switch (MIPSInst_OPCODE(ir)) {
+	case addiu_op:
+		if (MIPSInst_RT(ir))
+			regs->regs[MIPSInst_RT(ir)] =
+				(s32)regs->regs[MIPSInst_RS(ir)] +
+				(s32)MIPSInst_SIMM(ir);
+		return 0;
+	case daddiu_op:
+		if (IS_ENABLED(CONFIG_32BIT))
+			break;
+
+		if (MIPSInst_RT(ir))
+			regs->regs[MIPSInst_RT(ir)] =
+				(s64)regs->regs[MIPSInst_RS(ir)] +
+				(s64)MIPSInst_SIMM(ir);
+		return 0;
+	case lwc1_op:
+	case swc1_op:
+	case cop1_op:
+	case cop1x_op:
+		/* FPU instructions in delay slot */
+		return -SIGFPE;
+	case spec_op:
+		switch (MIPSInst_FUNC(ir)) {
+		case or_op:
+			if (MIPSInst_RD(ir))
+				regs->regs[MIPSInst_RD(ir)] =
+					regs->regs[MIPSInst_RS(ir)] |
+					regs->regs[MIPSInst_RT(ir)];
+			return 0;
+		case sll_op:
+			if (MIPSInst_RS(ir))
+				break;
+
+			if (MIPSInst_RD(ir))
+				regs->regs[MIPSInst_RD(ir)] =
+					(s32)(((u32)regs->regs[MIPSInst_RT(ir)]) <<
+						MIPSInst_FD(ir));
+			return 0;
+		case srl_op:
+			if (MIPSInst_RS(ir))
+				break;
+
+			if (MIPSInst_RD(ir))
+				regs->regs[MIPSInst_RD(ir)] =
+					(s32)(((u32)regs->regs[MIPSInst_RT(ir)]) >>
+						MIPSInst_FD(ir));
+			return 0;
+		case addu_op:
+			if (MIPSInst_FD(ir))
+				break;
+
+			if (MIPSInst_RD(ir))
+				regs->regs[MIPSInst_RD(ir)] =
+					(s32)((u32)regs->regs[MIPSInst_RS(ir)] +
+					      (u32)regs->regs[MIPSInst_RT(ir)]);
+			return 0;
+		case subu_op:
+			if (MIPSInst_FD(ir))
+				break;
+
+			if (MIPSInst_RD(ir))
+				regs->regs[MIPSInst_RD(ir)] =
+					(s32)((u32)regs->regs[MIPSInst_RS(ir)] -
+					      (u32)regs->regs[MIPSInst_RT(ir)]);
+			return 0;
+		case dsll_op:
+			if (IS_ENABLED(CONFIG_32BIT) || MIPSInst_RS(ir))
+				break;
+
+			if (MIPSInst_RD(ir))
+				regs->regs[MIPSInst_RD(ir)] =
+					(s64)(((u64)regs->regs[MIPSInst_RT(ir)]) <<
+						MIPSInst_FD(ir));
+			return 0;
+		case dsrl_op:
+			if (IS_ENABLED(CONFIG_32BIT) || MIPSInst_RS(ir))
+				break;
+
+			if (MIPSInst_RD(ir))
+				regs->regs[MIPSInst_RD(ir)] =
+					(s64)(((u64)regs->regs[MIPSInst_RT(ir)]) >>
+						MIPSInst_FD(ir));
+			return 0;
+		case daddu_op:
+			if (IS_ENABLED(CONFIG_32BIT) || MIPSInst_FD(ir))
+				break;
+
+			if (MIPSInst_RD(ir))
+				regs->regs[MIPSInst_RD(ir)] =
+					(u64)regs->regs[MIPSInst_RS(ir)] +
+					(u64)regs->regs[MIPSInst_RT(ir)];
+			return 0;
+		case dsubu_op:
+			if (IS_ENABLED(CONFIG_32BIT) || MIPSInst_FD(ir))
+				break;
+
+			if (MIPSInst_RD(ir))
+				regs->regs[MIPSInst_RD(ir)] =
+					(s64)((u64)regs->regs[MIPSInst_RS(ir)] -
+					      (u64)regs->regs[MIPSInst_RT(ir)]);
+			return 0;
+		}
+		break;
+	default:
+		pr_debug("No fastpath BD emulation for instruction 0x%08x (op: %02x)\n",
+			 ir, MIPSInst_OPCODE(ir));
+	}
+
+	return SIGILL;
+}
+
+/**
+ * movf_func - Emulate a MOVF instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int movf_func(struct pt_regs *regs, u32 ir)
+{
+	u32 csr;
+	u32 cond;
+
+	csr = current->thread.fpu.fcr31;
+	cond = fpucondbit[MIPSInst_RT(ir) >> 2];
+
+	if (((csr & cond) == 0) && MIPSInst_RD(ir))
+		regs->regs[MIPSInst_RD(ir)] = regs->regs[MIPSInst_RS(ir)];
+
+	MIPS_R2_STATS(movs);
+
+	return 0;
+}
+
+/**
+ * movt_func - Emulate a MOVT instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int movt_func(struct pt_regs *regs, u32 ir)
+{
+	u32 csr;
+	u32 cond;
+
+	csr = current->thread.fpu.fcr31;
+	cond = fpucondbit[MIPSInst_RT(ir) >> 2];
+
+	if (((csr & cond) != 0) && MIPSInst_RD(ir))
+		regs->regs[MIPSInst_RD(ir)] = regs->regs[MIPSInst_RS(ir)];
+
+	MIPS_R2_STATS(movs);
+
+	return 0;
+}
+
+/**
+ * jr_func - Emulate a JR instruction.
+ * @pt_regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns SIGILL if JR was in delay slot, SIGEMT if we
+ * can't compute the EPC, SIGSEGV if we can't access the
+ * userland instruction or 0 on success.
+ */
+static int jr_func(struct pt_regs *regs, u32 ir)
+{
+	int err;
+	unsigned long cepc, epc, nepc;
+	u32 nir;
+
+	if (delay_slot(regs))
+		return SIGILL;
+
+	/* EPC after the RI/JR instruction */
+	nepc = regs->cp0_epc;
+	/* Roll back to the reserved R2 JR instruction */
+	regs->cp0_epc -= 4;
+	epc = regs->cp0_epc;
+	err = __compute_return_epc(regs);
+
+	if (err < 0)
+		return SIGEMT;
+
+
+	/* Computed EPC */
+	cepc = regs->cp0_epc;
+
+	/* Get DS instruction */
+	err = __get_user(nir, (u32 __user *)nepc);
+	if (err)
+		return SIGSEGV;
+
+	MIPS_R2BR_STATS(jrs);
+
+	/* If nir == 0(NOP), then nothing else to do */
+	if (nir) {
+		/*
+		 * Negative err means FPU instruction in BD-slot,
+		 * Zero err means 'BD-slot emulation done'
+		 * For anything else we go back to trampoline emulation.
+		 */
+		err = mipsr6_emul(regs, nir);
+		if (err > 0) {
+			regs->cp0_epc = nepc;
+			err = mips_dsemul(regs, nir, epc, cepc);
+			if (err == SIGILL)
+				err = SIGEMT;
+			MIPS_R2_STATS(dsemul);
+		}
+	}
+
+	return err;
+}
+
+/**
+ * movz_func - Emulate a MOVZ instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int movz_func(struct pt_regs *regs, u32 ir)
+{
+	if (((regs->regs[MIPSInst_RT(ir)]) == 0) && MIPSInst_RD(ir))
+		regs->regs[MIPSInst_RD(ir)] = regs->regs[MIPSInst_RS(ir)];
+	MIPS_R2_STATS(movs);
+
+	return 0;
+}
+
+/**
+ * movn_func - Emulate a MOVZ instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int movn_func(struct pt_regs *regs, u32 ir)
+{
+	if (((regs->regs[MIPSInst_RT(ir)]) != 0) && MIPSInst_RD(ir))
+		regs->regs[MIPSInst_RD(ir)] = regs->regs[MIPSInst_RS(ir)];
+	MIPS_R2_STATS(movs);
+
+	return 0;
+}
+
+/**
+ * mfhi_func - Emulate a MFHI instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int mfhi_func(struct pt_regs *regs, u32 ir)
+{
+	if (MIPSInst_RD(ir))
+		regs->regs[MIPSInst_RD(ir)] = regs->hi;
+
+	MIPS_R2_STATS(hilo);
+
+	return 0;
+}
+
+/**
+ * mthi_func - Emulate a MTHI instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int mthi_func(struct pt_regs *regs, u32 ir)
+{
+	regs->hi = regs->regs[MIPSInst_RS(ir)];
+
+	MIPS_R2_STATS(hilo);
+
+	return 0;
+}
+
+/**
+ * mflo_func - Emulate a MFLO instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int mflo_func(struct pt_regs *regs, u32 ir)
+{
+	if (MIPSInst_RD(ir))
+		regs->regs[MIPSInst_RD(ir)] = regs->lo;
+
+	MIPS_R2_STATS(hilo);
+
+	return 0;
+}
+
+/**
+ * mtlo_func - Emulate a MTLO instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int mtlo_func(struct pt_regs *regs, u32 ir)
+{
+	regs->lo = regs->regs[MIPSInst_RS(ir)];
+
+	MIPS_R2_STATS(hilo);
+
+	return 0;
+}
+
+/**
+ * mult_func - Emulate a MULT instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int mult_func(struct pt_regs *regs, u32 ir)
+{
+	s64 res;
+	s32 rt, rs;
+
+	rt = regs->regs[MIPSInst_RT(ir)];
+	rs = regs->regs[MIPSInst_RS(ir)];
+	res = (s64)rt * (s64)rs;
+
+	rs = res;
+	regs->lo = (s64)rs;
+	rt = res >> 32;
+	res = (s64)rt;
+	regs->hi = res;
+
+	MIPS_R2_STATS(muls);
+
+	return 0;
+}
+
+/**
+ * multu_func - Emulate a MULTU instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int multu_func(struct pt_regs *regs, u32 ir)
+{
+	u64 res;
+	u32 rt, rs;
+
+	rt = regs->regs[MIPSInst_RT(ir)];
+	rs = regs->regs[MIPSInst_RS(ir)];
+	res = (u64)rt * (u64)rs;
+	rt = res;
+	regs->lo = (s64)(s32)rt;
+	regs->hi = (s64)(s32)(res >> 32);
+
+	MIPS_R2_STATS(muls);
+
+	return 0;
+}
+
+/**
+ * div_func - Emulate a DIV instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int div_func(struct pt_regs *regs, u32 ir)
+{
+	s32 rt, rs;
+
+	rt = regs->regs[MIPSInst_RT(ir)];
+	rs = regs->regs[MIPSInst_RS(ir)];
+
+	regs->lo = (s64)(rs / rt);
+	regs->hi = (s64)(rs % rt);
+
+	MIPS_R2_STATS(divs);
+
+	return 0;
+}
+
+/**
+ * divu_func - Emulate a DIVU instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int divu_func(struct pt_regs *regs, u32 ir)
+{
+	u32 rt, rs;
+
+	rt = regs->regs[MIPSInst_RT(ir)];
+	rs = regs->regs[MIPSInst_RS(ir)];
+
+	regs->lo = (s64)(rs / rt);
+	regs->hi = (s64)(rs % rt);
+
+	MIPS_R2_STATS(divs);
+
+	return 0;
+}
+
+/**
+ * dmult_func - Emulate a DMULT instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 on success or SIGILL for 32-bit kernels.
+ */
+static int dmult_func(struct pt_regs *regs, u32 ir)
+{
+	s64 res;
+	s64 rt, rs;
+
+	if (IS_ENABLED(CONFIG_32BIT))
+		return SIGILL;
+
+	rt = regs->regs[MIPSInst_RT(ir)];
+	rs = regs->regs[MIPSInst_RS(ir)];
+	res = rt * rs;
+
+	regs->lo = res;
+	__asm__ __volatile__(
+		"dmuh %0, %1, %2\t\n"
+		: "=r"(res)
+		: "r"(rt), "r"(rs));
+
+	regs->hi = res;
+
+	MIPS_R2_STATS(muls);
+
+	return 0;
+}
+
+/**
+ * dmultu_func - Emulate a DMULTU instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 on success or SIGILL for 32-bit kernels.
+ */
+static int dmultu_func(struct pt_regs *regs, u32 ir)
+{
+	u64 res;
+	u64 rt, rs;
+
+	if (IS_ENABLED(CONFIG_32BIT))
+		return SIGILL;
+
+	rt = regs->regs[MIPSInst_RT(ir)];
+	rs = regs->regs[MIPSInst_RS(ir)];
+	res = rt * rs;
+
+	regs->lo = res;
+	__asm__ __volatile__(
+		"dmuhu %0, %1, %2\t\n"
+		: "=r"(res)
+		: "r"(rt), "r"(rs));
+
+	regs->hi = res;
+
+	MIPS_R2_STATS(muls);
+
+	return 0;
+}
+
+/**
+ * ddiv_func - Emulate a DDIV instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 on success or SIGILL for 32-bit kernels.
+ */
+static int ddiv_func(struct pt_regs *regs, u32 ir)
+{
+	s64 rt, rs;
+
+	if (IS_ENABLED(CONFIG_32BIT))
+		return SIGILL;
+
+	rt = regs->regs[MIPSInst_RT(ir)];
+	rs = regs->regs[MIPSInst_RS(ir)];
+
+	regs->lo = rs / rt;
+	regs->hi = rs % rt;
+
+	MIPS_R2_STATS(divs);
+
+	return 0;
+}
+
+/**
+ * ddivu_func - Emulate a DDIVU instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 on success or SIGILL for 32-bit kernels.
+ */
+static int ddivu_func(struct pt_regs *regs, u32 ir)
+{
+	u64 rt, rs;
+
+	if (IS_ENABLED(CONFIG_32BIT))
+		return SIGILL;
+
+	rt = regs->regs[MIPSInst_RT(ir)];
+	rs = regs->regs[MIPSInst_RS(ir)];
+
+	regs->lo = rs / rt;
+	regs->hi = rs % rt;
+
+	MIPS_R2_STATS(divs);
+
+	return 0;
+}
+
+/* R6 removed instructions for the SPECIAL opcode */
+static const struct r2_decoder_table spec_op_table[] = {
+	{ 0xfc1ff83f, 0x00000008, jr_func },
+	{ 0xfc00ffff, 0x00000018, mult_func },
+	{ 0xfc00ffff, 0x00000019, multu_func },
+	{ 0xfc00ffff, 0x0000001c, dmult_func },
+	{ 0xfc00ffff, 0x0000001d, dmultu_func },
+	{ 0xffff07ff, 0x00000010, mfhi_func },
+	{ 0xfc1fffff, 0x00000011, mthi_func },
+	{ 0xffff07ff, 0x00000012, mflo_func },
+	{ 0xfc1fffff, 0x00000013, mtlo_func },
+	{ 0xfc0307ff, 0x00000001, movf_func },
+	{ 0xfc0307ff, 0x00010001, movt_func },
+	{ 0xfc0007ff, 0x0000000a, movz_func },
+	{ 0xfc0007ff, 0x0000000b, movn_func },
+	{ 0xfc00ffff, 0x0000001a, div_func },
+	{ 0xfc00ffff, 0x0000001b, divu_func },
+	{ 0xfc00ffff, 0x0000001e, ddiv_func },
+	{ 0xfc00ffff, 0x0000001f, ddivu_func },
+	{}
+};
+
+/**
+ * madd_func - Emulate a MADD instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int madd_func(struct pt_regs *regs, u32 ir)
+{
+	s64 res;
+	s32 rt, rs;
+
+	rt = regs->regs[MIPSInst_RT(ir)];
+	rs = regs->regs[MIPSInst_RS(ir)];
+	res = (s64)rt * (s64)rs;
+	rt = regs->hi;
+	rs = regs->lo;
+	res += ((((s64)rt) << 32) | (u32)rs);
+
+	rt = res;
+	regs->lo = (s64)rt;
+	rs = res >> 32;
+	regs->hi = (s64)rs;
+
+	MIPS_R2_STATS(dsps);
+
+	return 0;
+}
+
+/**
+ * maddu_func - Emulate a MADDU instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int maddu_func(struct pt_regs *regs, u32 ir)
+{
+	u64 res;
+	u32 rt, rs;
+
+	rt = regs->regs[MIPSInst_RT(ir)];
+	rs = regs->regs[MIPSInst_RS(ir)];
+	res = (u64)rt * (u64)rs;
+	rt = regs->hi;
+	rs = regs->lo;
+	res += ((((s64)rt) << 32) | (u32)rs);
+
+	rt = res;
+	regs->lo = (s64)(s32)rt;
+	rs = res >> 32;
+	regs->hi = (s64)(s32)rs;
+
+	MIPS_R2_STATS(dsps);
+
+	return 0;
+}
+
+/**
+ * msub_func - Emulate a MSUB instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int msub_func(struct pt_regs *regs, u32 ir)
+{
+	s64 res;
+	s32 rt, rs;
+
+	rt = regs->regs[MIPSInst_RT(ir)];
+	rs = regs->regs[MIPSInst_RS(ir)];
+	res = (s64)rt * (s64)rs;
+	rt = regs->hi;
+	rs = regs->lo;
+	res = ((((s64)rt) << 32) | (u32)rs) - res;
+
+	rt = res;
+	regs->lo = (s64)rt;
+	rs = res >> 32;
+	regs->hi = (s64)rs;
+
+	MIPS_R2_STATS(dsps);
+
+	return 0;
+}
+
+/**
+ * msubu_func - Emulate a MSUBU instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int msubu_func(struct pt_regs *regs, u32 ir)
+{
+	u64 res;
+	u32 rt, rs;
+
+	rt = regs->regs[MIPSInst_RT(ir)];
+	rs = regs->regs[MIPSInst_RS(ir)];
+	res = (u64)rt * (u64)rs;
+	rt = regs->hi;
+	rs = regs->lo;
+	res = ((((s64)rt) << 32) | (u32)rs) - res;
+
+	rt = res;
+	regs->lo = (s64)(s32)rt;
+	rs = res >> 32;
+	regs->hi = (s64)(s32)rs;
+
+	MIPS_R2_STATS(dsps);
+
+	return 0;
+}
+
+/**
+ * mul_func - Emulate a MUL instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int mul_func(struct pt_regs *regs, u32 ir)
+{
+	s64 res;
+	s32 rt, rs;
+
+	if (!MIPSInst_RD(ir))
+		return 0;
+	rt = regs->regs[MIPSInst_RT(ir)];
+	rs = regs->regs[MIPSInst_RS(ir)];
+	res = (s64)rt * (s64)rs;
+
+	rs = res;
+	regs->regs[MIPSInst_RD(ir)] = (s64)rs;
+
+	MIPS_R2_STATS(muls);
+
+	return 0;
+}
+
+/**
+ * clz_func - Emulate a CLZ instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int clz_func(struct pt_regs *regs, u32 ir)
+{
+	u32 res;
+	u32 rs;
+
+	if (!MIPSInst_RD(ir))
+		return 0;
+
+	rs = regs->regs[MIPSInst_RS(ir)];
+	__asm__ __volatile__("clz %0, %1" : "=r"(res) : "r"(rs));
+	regs->regs[MIPSInst_RD(ir)] = res;
+
+	MIPS_R2_STATS(bops);
+
+	return 0;
+}
+
+/**
+ * clo_func - Emulate a CLO instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+
+static int clo_func(struct pt_regs *regs, u32 ir)
+{
+	u32 res;
+	u32 rs;
+
+	if (!MIPSInst_RD(ir))
+		return 0;
+
+	rs = regs->regs[MIPSInst_RS(ir)];
+	__asm__ __volatile__("clo %0, %1" : "=r"(res) : "r"(rs));
+	regs->regs[MIPSInst_RD(ir)] = res;
+
+	MIPS_R2_STATS(bops);
+
+	return 0;
+}
+
+/**
+ * dclz_func - Emulate a DCLZ instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int dclz_func(struct pt_regs *regs, u32 ir)
+{
+	u64 res;
+	u64 rs;
+
+	if (IS_ENABLED(CONFIG_32BIT))
+		return SIGILL;
+
+	if (!MIPSInst_RD(ir))
+		return 0;
+
+	rs = regs->regs[MIPSInst_RS(ir)];
+	__asm__ __volatile__("dclz %0, %1" : "=r"(res) : "r"(rs));
+	regs->regs[MIPSInst_RD(ir)] = res;
+
+	MIPS_R2_STATS(bops);
+
+	return 0;
+}
+
+/**
+ * dclo_func - Emulate a DCLO instruction
+ * @regs: Process register set
+ * @ir: Instruction
+ *
+ * Returns 0 since it always succeeds.
+ */
+static int dclo_func(struct pt_regs *regs, u32 ir)
+{
+	u64 res;
+	u64 rs;
+
+	if (IS_ENABLED(CONFIG_32BIT))
+		return SIGILL;
+
+	if (!MIPSInst_RD(ir))
+		return 0;
+
+	rs = regs->regs[MIPSInst_RS(ir)];
+	__asm__ __volatile__("dclo %0, %1" : "=r"(res) : "r"(rs));
+	regs->regs[MIPSInst_RD(ir)] = res;
+
+	MIPS_R2_STATS(bops);
+
+	return 0;
+}
+
+/* R6 removed instructions for the SPECIAL2 opcode */
+static const struct r2_decoder_table spec2_op_table[] = {
+	{ 0xfc00ffff, 0x70000000, madd_func },
+	{ 0xfc00ffff, 0x70000001, maddu_func },
+	{ 0xfc0007ff, 0x70000002, mul_func },
+	{ 0xfc00ffff, 0x70000004, msub_func },
+	{ 0xfc00ffff, 0x70000005, msubu_func },
+	{ 0xfc0007ff, 0x70000020, clz_func },
+	{ 0xfc0007ff, 0x70000021, clo_func },
+	{ 0xfc0007ff, 0x70000024, dclz_func },
+	{ 0xfc0007ff, 0x70000025, dclo_func },
+	{ }
+};
+
+static inline int mipsr2_find_op_func(struct pt_regs *regs, u32 inst,
+				      const struct r2_decoder_table *table)
+{
+	const struct r2_decoder_table *p;
+	int err;
+
+	for (p = table; p->func; p++) {
+		if ((inst & p->mask) == p->code) {
+			err = (p->func)(regs, inst);
+			return err;
+		}
+	}
+	return SIGILL;
+}
+
+/**
+ * mipsr2_decoder: Decode and emulate a MIPS R2 instruction
+ * @regs: Process register set
+ * @inst: Instruction to decode and emulate
+ * @fcr31: Floating Point Control and Status Register Cause bits returned
+ */
+int mipsr2_decoder(struct pt_regs *regs, u32 inst, unsigned long *fcr31)
+{
+	int err = 0;
+	unsigned long vaddr;
+	u32 nir;
+	unsigned long cpc, epc, nepc, r31, res, rs, rt;
+
+	void __user *fault_addr = NULL;
+	int pass = 0;
+
+repeat:
+	r31 = regs->regs[31];
+	epc = regs->cp0_epc;
+	err = compute_return_epc(regs);
+	if (err < 0) {
+		BUG();
+		return SIGEMT;
+	}
+	pr_debug("Emulating the 0x%08x R2 instruction @ 0x%08lx (pass=%d))\n",
+		 inst, epc, pass);
+
+	switch (MIPSInst_OPCODE(inst)) {
+	case spec_op:
+		err = mipsr2_find_op_func(regs, inst, spec_op_table);
+		if (err < 0) {
+			/* FPU instruction under JR */
+			regs->cp0_cause |= CAUSEF_BD;
+			goto fpu_emul;
+		}
+		break;
+	case spec2_op:
+		err = mipsr2_find_op_func(regs, inst, spec2_op_table);
+		break;
+	case bcond_op:
+		rt = MIPSInst_RT(inst);
+		rs = MIPSInst_RS(inst);
+		switch (rt) {
+		case tgei_op:
+			if ((long)regs->regs[rs] >= MIPSInst_SIMM(inst))
+				do_trap_or_bp(regs, 0, 0, "TGEI");
+
+			MIPS_R2_STATS(traps);
+
+			break;
+		case tgeiu_op:
+			if (regs->regs[rs] >= MIPSInst_UIMM(inst))
+				do_trap_or_bp(regs, 0, 0, "TGEIU");
+
+			MIPS_R2_STATS(traps);
+
+			break;
+		case tlti_op:
+			if ((long)regs->regs[rs] < MIPSInst_SIMM(inst))
+				do_trap_or_bp(regs, 0, 0, "TLTI");
+
+			MIPS_R2_STATS(traps);
+
+			break;
+		case tltiu_op:
+			if (regs->regs[rs] < MIPSInst_UIMM(inst))
+				do_trap_or_bp(regs, 0, 0, "TLTIU");
+
+			MIPS_R2_STATS(traps);
+
+			break;
+		case teqi_op:
+			if (regs->regs[rs] == MIPSInst_SIMM(inst))
+				do_trap_or_bp(regs, 0, 0, "TEQI");
+
+			MIPS_R2_STATS(traps);
+
+			break;
+		case tnei_op:
+			if (regs->regs[rs] != MIPSInst_SIMM(inst))
+				do_trap_or_bp(regs, 0, 0, "TNEI");
+
+			MIPS_R2_STATS(traps);
+
+			break;
+		case bltzl_op:
+		case bgezl_op:
+		case bltzall_op:
+		case bgezall_op:
+			if (delay_slot(regs)) {
+				err = SIGILL;
+				break;
+			}
+			regs->regs[31] = r31;
+			regs->cp0_epc = epc;
+			err = __compute_return_epc(regs);
+			if (err < 0)
+				return SIGEMT;
+			if (err != BRANCH_LIKELY_TAKEN)
+				break;
+			cpc = regs->cp0_epc;
+			nepc = epc + 4;
+			err = __get_user(nir, (u32 __user *)nepc);
+			if (err) {
+				err = SIGSEGV;
+				break;
+			}
+			/*
+			 * This will probably be optimized away when
+			 * CONFIG_DEBUG_FS is not enabled
+			 */
+			switch (rt) {
+			case bltzl_op:
+				MIPS_R2BR_STATS(bltzl);
+				break;
+			case bgezl_op:
+				MIPS_R2BR_STATS(bgezl);
+				break;
+			case bltzall_op:
+				MIPS_R2BR_STATS(bltzall);
+				break;
+			case bgezall_op:
+				MIPS_R2BR_STATS(bgezall);
+				break;
+			}
+
+			switch (MIPSInst_OPCODE(nir)) {
+			case cop1_op:
+			case cop1x_op:
+			case lwc1_op:
+			case swc1_op:
+				regs->cp0_cause |= CAUSEF_BD;
+				goto fpu_emul;
+			}
+			if (nir) {
+				err = mipsr6_emul(regs, nir);
+				if (err > 0) {
+					err = mips_dsemul(regs, nir, epc, cpc);
+					if (err == SIGILL)
+						err = SIGEMT;
+					MIPS_R2_STATS(dsemul);
+				}
+			}
+			break;
+		case bltzal_op:
+		case bgezal_op:
+			if (delay_slot(regs)) {
+				err = SIGILL;
+				break;
+			}
+			regs->regs[31] = r31;
+			regs->cp0_epc = epc;
+			err = __compute_return_epc(regs);
+			if (err < 0)
+				return SIGEMT;
+			cpc = regs->cp0_epc;
+			nepc = epc + 4;
+			err = __get_user(nir, (u32 __user *)nepc);
+			if (err) {
+				err = SIGSEGV;
+				break;
+			}
+			/*
+			 * This will probably be optimized away when
+			 * CONFIG_DEBUG_FS is not enabled
+			 */
+			switch (rt) {
+			case bltzal_op:
+				MIPS_R2BR_STATS(bltzal);
+				break;
+			case bgezal_op:
+				MIPS_R2BR_STATS(bgezal);
+				break;
+			}
+
+			switch (MIPSInst_OPCODE(nir)) {
+			case cop1_op:
+			case cop1x_op:
+			case lwc1_op:
+			case swc1_op:
+				regs->cp0_cause |= CAUSEF_BD;
+				goto fpu_emul;
+			}
+			if (nir) {
+				err = mipsr6_emul(regs, nir);
+				if (err > 0) {
+					err = mips_dsemul(regs, nir, epc, cpc);
+					if (err == SIGILL)
+						err = SIGEMT;
+					MIPS_R2_STATS(dsemul);
+				}
+			}
+			break;
+		default:
+			regs->regs[31] = r31;
+			regs->cp0_epc = epc;
+			err = SIGILL;
+			break;
+		}
+		break;
+
+	case blezl_op:
+	case bgtzl_op:
+		/*
+		 * For BLEZL and BGTZL, rt field must be set to 0. If this
+		 * is not the case, this may be an encoding of a MIPS R6
+		 * instruction, so return to CPU execution if this occurs
+		 */
+		if (MIPSInst_RT(inst)) {
+			err = SIGILL;
+			break;
+		}
+		fallthrough;
+	case beql_op:
+	case bnel_op:
+		if (delay_slot(regs)) {
+			err = SIGILL;
+			break;
+		}
+		regs->regs[31] = r31;
+		regs->cp0_epc = epc;
+		err = __compute_return_epc(regs);
+		if (err < 0)
+			return SIGEMT;
+		if (err != BRANCH_LIKELY_TAKEN)
+			break;
+		cpc = regs->cp0_epc;
+		nepc = epc + 4;
+		err = __get_user(nir, (u32 __user *)nepc);
+		if (err) {
+			err = SIGSEGV;
+			break;
+		}
+		/*
+		 * This will probably be optimized away when
+		 * CONFIG_DEBUG_FS is not enabled
+		 */
+		switch (MIPSInst_OPCODE(inst)) {
+		case beql_op:
+			MIPS_R2BR_STATS(beql);
+			break;
+		case bnel_op:
+			MIPS_R2BR_STATS(bnel);
+			break;
+		case blezl_op:
+			MIPS_R2BR_STATS(blezl);
+			break;
+		case bgtzl_op:
+			MIPS_R2BR_STATS(bgtzl);
+			break;
+		}
+
+		switch (MIPSInst_OPCODE(nir)) {
+		case cop1_op:
+		case cop1x_op:
+		case lwc1_op:
+		case swc1_op:
+			regs->cp0_cause |= CAUSEF_BD;
+			goto fpu_emul;
+		}
+		if (nir) {
+			err = mipsr6_emul(regs, nir);
+			if (err > 0) {
+				err = mips_dsemul(regs, nir, epc, cpc);
+				if (err == SIGILL)
+					err = SIGEMT;
+				MIPS_R2_STATS(dsemul);
+			}
+		}
+		break;
+	case lwc1_op:
+	case swc1_op:
+	case cop1_op:
+	case cop1x_op:
+fpu_emul:
+		regs->regs[31] = r31;
+		regs->cp0_epc = epc;
+
+		err = fpu_emulator_cop1Handler(regs, &current->thread.fpu, 0,
+					       &fault_addr);
+
+		/*
+		 * We can't allow the emulated instruction to leave any
+		 * enabled Cause bits set in $fcr31.
+		 */
+		*fcr31 = res = mask_fcr31_x(current->thread.fpu.fcr31);
+		current->thread.fpu.fcr31 &= ~res;
+
+		/*
+		 * this is a tricky issue - lose_fpu() uses LL/SC atomics
+		 * if FPU is owned and effectively cancels user level LL/SC.
+		 * So, it could be logical to don't restore FPU ownership here.
+		 * But the sequence of multiple FPU instructions is much much
+		 * more often than LL-FPU-SC and I prefer loop here until
+		 * next scheduler cycle cancels FPU ownership
+		 */
+		own_fpu(1);	/* Restore FPU state. */
+
+		if (err)
+			current->thread.cp0_baduaddr = (unsigned long)fault_addr;
+
+		MIPS_R2_STATS(fpus);
+
+		break;
+
+	case lwl_op:
+		rt = regs->regs[MIPSInst_RT(inst)];
+		vaddr = regs->regs[MIPSInst_RS(inst)] + MIPSInst_SIMM(inst);
+		if (!access_ok((void __user *)vaddr, 4)) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGSEGV;
+			break;
+		}
+		__asm__ __volatile__(
+			"	.set	push\n"
+			"	.set	reorder\n"
+#ifdef CONFIG_CPU_LITTLE_ENDIAN
+			"1:"	LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 24, 8\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+				ADDIU	"%2, %2, -1\n"
+			"2:"	LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 16, 8\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+				ADDIU	"%2, %2, -1\n"
+			"3:"	LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 8, 8\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+				ADDIU	"%2, %2, -1\n"
+			"4:"	LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 0, 8\n"
+#else /* !CONFIG_CPU_LITTLE_ENDIAN */
+			"1:"	LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 24, 8\n"
+				ADDIU	"%2, %2, 1\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+			"2:"	LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 16, 8\n"
+				ADDIU	"%2, %2, 1\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+			"3:"	LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 8, 8\n"
+				ADDIU	"%2, %2, 1\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+			"4:"	LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 0, 8\n"
+#endif /* CONFIG_CPU_LITTLE_ENDIAN */
+			"9:	sll	%0, %0, 0\n"
+			"10:\n"
+			"	.insn\n"
+			"	.section	.fixup,\"ax\"\n"
+			"8:	li	%3,%4\n"
+			"	j	10b\n"
+			"	.previous\n"
+			"	.section	__ex_table,\"a\"\n"
+			STR(PTR) " 1b,8b\n"
+			STR(PTR) " 2b,8b\n"
+			STR(PTR) " 3b,8b\n"
+			STR(PTR) " 4b,8b\n"
+			"	.previous\n"
+			"	.set	pop\n"
+			: "+&r"(rt), "=&r"(rs),
+			  "+&r"(vaddr), "+&r"(err)
+			: "i"(SIGSEGV));
+
+		if (MIPSInst_RT(inst) && !err)
+			regs->regs[MIPSInst_RT(inst)] = rt;
+
+		MIPS_R2_STATS(loads);
+
+		break;
+
+	case lwr_op:
+		rt = regs->regs[MIPSInst_RT(inst)];
+		vaddr = regs->regs[MIPSInst_RS(inst)] + MIPSInst_SIMM(inst);
+		if (!access_ok((void __user *)vaddr, 4)) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGSEGV;
+			break;
+		}
+		__asm__ __volatile__(
+			"       .set	push\n"
+			"       .set	reorder\n"
+#ifdef CONFIG_CPU_LITTLE_ENDIAN
+			"1:"    LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 0, 8\n"
+				ADDIU	"%2, %2, 1\n"
+			"       andi	%1, %2, 0x3\n"
+			"       beq	$0, %1, 9f\n"
+			"2:"    LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 8, 8\n"
+				ADDIU	"%2, %2, 1\n"
+			"       andi	%1, %2, 0x3\n"
+			"       beq	$0, %1, 9f\n"
+			"3:"    LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 16, 8\n"
+				ADDIU	"%2, %2, 1\n"
+			"       andi	%1, %2, 0x3\n"
+			"       beq	$0, %1, 9f\n"
+			"4:"    LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 24, 8\n"
+			"       sll	%0, %0, 0\n"
+#else /* !CONFIG_CPU_LITTLE_ENDIAN */
+			"1:"    LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 0, 8\n"
+			"       andi	%1, %2, 0x3\n"
+			"       beq	$0, %1, 9f\n"
+				ADDIU	"%2, %2, -1\n"
+			"2:"    LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 8, 8\n"
+			"       andi	%1, %2, 0x3\n"
+			"       beq	$0, %1, 9f\n"
+				ADDIU	"%2, %2, -1\n"
+			"3:"    LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 16, 8\n"
+			"       andi	%1, %2, 0x3\n"
+			"       beq	$0, %1, 9f\n"
+				ADDIU	"%2, %2, -1\n"
+			"4:"    LB	"%1, 0(%2)\n"
+				INS	"%0, %1, 24, 8\n"
+			"       sll	%0, %0, 0\n"
+#endif /* CONFIG_CPU_LITTLE_ENDIAN */
+			"9:\n"
+			"10:\n"
+			"	.insn\n"
+			"	.section	.fixup,\"ax\"\n"
+			"8:	li	%3,%4\n"
+			"	j	10b\n"
+			"       .previous\n"
+			"	.section	__ex_table,\"a\"\n"
+			STR(PTR) " 1b,8b\n"
+			STR(PTR) " 2b,8b\n"
+			STR(PTR) " 3b,8b\n"
+			STR(PTR) " 4b,8b\n"
+			"	.previous\n"
+			"	.set	pop\n"
+			: "+&r"(rt), "=&r"(rs),
+			  "+&r"(vaddr), "+&r"(err)
+			: "i"(SIGSEGV));
+		if (MIPSInst_RT(inst) && !err)
+			regs->regs[MIPSInst_RT(inst)] = rt;
+
+		MIPS_R2_STATS(loads);
+
+		break;
+
+	case swl_op:
+		rt = regs->regs[MIPSInst_RT(inst)];
+		vaddr = regs->regs[MIPSInst_RS(inst)] + MIPSInst_SIMM(inst);
+		if (!access_ok((void __user *)vaddr, 4)) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGSEGV;
+			break;
+		}
+		__asm__ __volatile__(
+			"	.set	push\n"
+			"	.set	reorder\n"
+#ifdef CONFIG_CPU_LITTLE_ENDIAN
+				EXT	"%1, %0, 24, 8\n"
+			"1:"	SB	"%1, 0(%2)\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+				ADDIU	"%2, %2, -1\n"
+				EXT	"%1, %0, 16, 8\n"
+			"2:"	SB	"%1, 0(%2)\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+				ADDIU	"%2, %2, -1\n"
+				EXT	"%1, %0, 8, 8\n"
+			"3:"	SB	"%1, 0(%2)\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+				ADDIU	"%2, %2, -1\n"
+				EXT	"%1, %0, 0, 8\n"
+			"4:"	SB	"%1, 0(%2)\n"
+#else /* !CONFIG_CPU_LITTLE_ENDIAN */
+				EXT	"%1, %0, 24, 8\n"
+			"1:"	SB	"%1, 0(%2)\n"
+				ADDIU	"%2, %2, 1\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+				EXT	"%1, %0, 16, 8\n"
+			"2:"	SB	"%1, 0(%2)\n"
+				ADDIU	"%2, %2, 1\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+				EXT	"%1, %0, 8, 8\n"
+			"3:"	SB	"%1, 0(%2)\n"
+				ADDIU	"%2, %2, 1\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+				EXT	"%1, %0, 0, 8\n"
+			"4:"	SB	"%1, 0(%2)\n"
+#endif /* CONFIG_CPU_LITTLE_ENDIAN */
+			"9:\n"
+			"	.insn\n"
+			"       .section        .fixup,\"ax\"\n"
+			"8:	li	%3,%4\n"
+			"	j	9b\n"
+			"	.previous\n"
+			"	.section        __ex_table,\"a\"\n"
+			STR(PTR) " 1b,8b\n"
+			STR(PTR) " 2b,8b\n"
+			STR(PTR) " 3b,8b\n"
+			STR(PTR) " 4b,8b\n"
+			"	.previous\n"
+			"	.set	pop\n"
+			: "+&r"(rt), "=&r"(rs),
+			  "+&r"(vaddr), "+&r"(err)
+			: "i"(SIGSEGV)
+			: "memory");
+
+		MIPS_R2_STATS(stores);
+
+		break;
+
+	case swr_op:
+		rt = regs->regs[MIPSInst_RT(inst)];
+		vaddr = regs->regs[MIPSInst_RS(inst)] + MIPSInst_SIMM(inst);
+		if (!access_ok((void __user *)vaddr, 4)) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGSEGV;
+			break;
+		}
+		__asm__ __volatile__(
+			"	.set	push\n"
+			"	.set	reorder\n"
+#ifdef CONFIG_CPU_LITTLE_ENDIAN
+				EXT	"%1, %0, 0, 8\n"
+			"1:"	SB	"%1, 0(%2)\n"
+				ADDIU	"%2, %2, 1\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+				EXT	"%1, %0, 8, 8\n"
+			"2:"	SB	"%1, 0(%2)\n"
+				ADDIU	"%2, %2, 1\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+				EXT	"%1, %0, 16, 8\n"
+			"3:"	SB	"%1, 0(%2)\n"
+				ADDIU	"%2, %2, 1\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+				EXT	"%1, %0, 24, 8\n"
+			"4:"	SB	"%1, 0(%2)\n"
+#else /* !CONFIG_CPU_LITTLE_ENDIAN */
+				EXT	"%1, %0, 0, 8\n"
+			"1:"	SB	"%1, 0(%2)\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+				ADDIU	"%2, %2, -1\n"
+				EXT	"%1, %0, 8, 8\n"
+			"2:"	SB	"%1, 0(%2)\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+				ADDIU	"%2, %2, -1\n"
+				EXT	"%1, %0, 16, 8\n"
+			"3:"	SB	"%1, 0(%2)\n"
+			"	andi	%1, %2, 0x3\n"
+			"	beq	$0, %1, 9f\n"
+				ADDIU	"%2, %2, -1\n"
+				EXT	"%1, %0, 24, 8\n"
+			"4:"	SB	"%1, 0(%2)\n"
+#endif /* CONFIG_CPU_LITTLE_ENDIAN */
+			"9:\n"
+			"	.insn\n"
+			"	.section        .fixup,\"ax\"\n"
+			"8:	li	%3,%4\n"
+			"	j	9b\n"
+			"	.previous\n"
+			"	.section        __ex_table,\"a\"\n"
+			STR(PTR) " 1b,8b\n"
+			STR(PTR) " 2b,8b\n"
+			STR(PTR) " 3b,8b\n"
+			STR(PTR) " 4b,8b\n"
+			"	.previous\n"
+			"	.set	pop\n"
+			: "+&r"(rt), "=&r"(rs),
+			  "+&r"(vaddr), "+&r"(err)
+			: "i"(SIGSEGV)
+			: "memory");
+
+		MIPS_R2_STATS(stores);
+
+		break;
+
+	case ldl_op:
+		if (IS_ENABLED(CONFIG_32BIT)) {
+		    err = SIGILL;
+		    break;
+		}
+
+		rt = regs->regs[MIPSInst_RT(inst)];
+		vaddr = regs->regs[MIPSInst_RS(inst)] + MIPSInst_SIMM(inst);
+		if (!access_ok((void __user *)vaddr, 8)) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGSEGV;
+			break;
+		}
+		__asm__ __volatile__(
+			"	.set    push\n"
+			"	.set    reorder\n"
+#ifdef CONFIG_CPU_LITTLE_ENDIAN
+			"1:	lb	%1, 0(%2)\n"
+			"	dinsu	%0, %1, 56, 8\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	daddiu	%2, %2, -1\n"
+			"2:	lb	%1, 0(%2)\n"
+			"	dinsu	%0, %1, 48, 8\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	daddiu	%2, %2, -1\n"
+			"3:	lb	%1, 0(%2)\n"
+			"	dinsu	%0, %1, 40, 8\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	daddiu	%2, %2, -1\n"
+			"4:	lb	%1, 0(%2)\n"
+			"	dinsu	%0, %1, 32, 8\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	daddiu	%2, %2, -1\n"
+			"5:	lb	%1, 0(%2)\n"
+			"	dins	%0, %1, 24, 8\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	daddiu	%2, %2, -1\n"
+			"6:	lb	%1, 0(%2)\n"
+			"	dins	%0, %1, 16, 8\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	daddiu	%2, %2, -1\n"
+			"7:	lb	%1, 0(%2)\n"
+			"	dins	%0, %1, 8, 8\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	daddiu	%2, %2, -1\n"
+			"0:	lb	%1, 0(%2)\n"
+			"	dins	%0, %1, 0, 8\n"
+#else /* !CONFIG_CPU_LITTLE_ENDIAN */
+			"1:	lb	%1, 0(%2)\n"
+			"	dinsu	%0, %1, 56, 8\n"
+			"	daddiu	%2, %2, 1\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"2:	lb	%1, 0(%2)\n"
+			"	dinsu	%0, %1, 48, 8\n"
+			"	daddiu	%2, %2, 1\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"3:	lb	%1, 0(%2)\n"
+			"	dinsu	%0, %1, 40, 8\n"
+			"	daddiu  %2, %2, 1\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"4:	lb	%1, 0(%2)\n"
+			"	dinsu	%0, %1, 32, 8\n"
+			"	daddiu	%2, %2, 1\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"5:	lb	%1, 0(%2)\n"
+			"	dins	%0, %1, 24, 8\n"
+			"	daddiu	%2, %2, 1\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"6:	lb	%1, 0(%2)\n"
+			"	dins	%0, %1, 16, 8\n"
+			"	daddiu	%2, %2, 1\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"7:	lb	%1, 0(%2)\n"
+			"	dins	%0, %1, 8, 8\n"
+			"	daddiu	%2, %2, 1\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"0:	lb	%1, 0(%2)\n"
+			"	dins	%0, %1, 0, 8\n"
+#endif /* CONFIG_CPU_LITTLE_ENDIAN */
+			"9:\n"
+			"	.insn\n"
+			"	.section        .fixup,\"ax\"\n"
+			"8:	li	%3,%4\n"
+			"	j	9b\n"
+			"	.previous\n"
+			"	.section        __ex_table,\"a\"\n"
+			STR(PTR) " 1b,8b\n"
+			STR(PTR) " 2b,8b\n"
+			STR(PTR) " 3b,8b\n"
+			STR(PTR) " 4b,8b\n"
+			STR(PTR) " 5b,8b\n"
+			STR(PTR) " 6b,8b\n"
+			STR(PTR) " 7b,8b\n"
+			STR(PTR) " 0b,8b\n"
+			"	.previous\n"
+			"	.set	pop\n"
+			: "+&r"(rt), "=&r"(rs),
+			  "+&r"(vaddr), "+&r"(err)
+			: "i"(SIGSEGV));
+		if (MIPSInst_RT(inst) && !err)
+			regs->regs[MIPSInst_RT(inst)] = rt;
+
+		MIPS_R2_STATS(loads);
+		break;
+
+	case ldr_op:
+		if (IS_ENABLED(CONFIG_32BIT)) {
+		    err = SIGILL;
+		    break;
+		}
+
+		rt = regs->regs[MIPSInst_RT(inst)];
+		vaddr = regs->regs[MIPSInst_RS(inst)] + MIPSInst_SIMM(inst);
+		if (!access_ok((void __user *)vaddr, 8)) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGSEGV;
+			break;
+		}
+		__asm__ __volatile__(
+			"	.set    push\n"
+			"	.set    reorder\n"
+#ifdef CONFIG_CPU_LITTLE_ENDIAN
+			"1:	lb      %1, 0(%2)\n"
+			"	dins   %0, %1, 0, 8\n"
+			"	daddiu  %2, %2, 1\n"
+			"	andi    %1, %2, 0x7\n"
+			"	beq     $0, %1, 9f\n"
+			"2:	lb      %1, 0(%2)\n"
+			"	dins   %0, %1, 8, 8\n"
+			"	daddiu  %2, %2, 1\n"
+			"	andi    %1, %2, 0x7\n"
+			"	beq     $0, %1, 9f\n"
+			"3:	lb      %1, 0(%2)\n"
+			"	dins   %0, %1, 16, 8\n"
+			"	daddiu  %2, %2, 1\n"
+			"	andi    %1, %2, 0x7\n"
+			"	beq     $0, %1, 9f\n"
+			"4:	lb      %1, 0(%2)\n"
+			"	dins   %0, %1, 24, 8\n"
+			"	daddiu  %2, %2, 1\n"
+			"	andi    %1, %2, 0x7\n"
+			"	beq     $0, %1, 9f\n"
+			"5:	lb      %1, 0(%2)\n"
+			"	dinsu    %0, %1, 32, 8\n"
+			"	daddiu  %2, %2, 1\n"
+			"	andi    %1, %2, 0x7\n"
+			"	beq     $0, %1, 9f\n"
+			"6:	lb      %1, 0(%2)\n"
+			"	dinsu    %0, %1, 40, 8\n"
+			"	daddiu  %2, %2, 1\n"
+			"	andi    %1, %2, 0x7\n"
+			"	beq     $0, %1, 9f\n"
+			"7:	lb      %1, 0(%2)\n"
+			"	dinsu    %0, %1, 48, 8\n"
+			"	daddiu  %2, %2, 1\n"
+			"	andi    %1, %2, 0x7\n"
+			"	beq     $0, %1, 9f\n"
+			"0:	lb      %1, 0(%2)\n"
+			"	dinsu    %0, %1, 56, 8\n"
+#else /* !CONFIG_CPU_LITTLE_ENDIAN */
+			"1:	lb      %1, 0(%2)\n"
+			"	dins   %0, %1, 0, 8\n"
+			"	andi    %1, %2, 0x7\n"
+			"	beq     $0, %1, 9f\n"
+			"	daddiu  %2, %2, -1\n"
+			"2:	lb      %1, 0(%2)\n"
+			"	dins   %0, %1, 8, 8\n"
+			"	andi    %1, %2, 0x7\n"
+			"	beq     $0, %1, 9f\n"
+			"	daddiu  %2, %2, -1\n"
+			"3:	lb      %1, 0(%2)\n"
+			"	dins   %0, %1, 16, 8\n"
+			"	andi    %1, %2, 0x7\n"
+			"	beq     $0, %1, 9f\n"
+			"	daddiu  %2, %2, -1\n"
+			"4:	lb      %1, 0(%2)\n"
+			"	dins   %0, %1, 24, 8\n"
+			"	andi    %1, %2, 0x7\n"
+			"	beq     $0, %1, 9f\n"
+			"	daddiu  %2, %2, -1\n"
+			"5:	lb      %1, 0(%2)\n"
+			"	dinsu    %0, %1, 32, 8\n"
+			"	andi    %1, %2, 0x7\n"
+			"	beq     $0, %1, 9f\n"
+			"	daddiu  %2, %2, -1\n"
+			"6:	lb      %1, 0(%2)\n"
+			"	dinsu    %0, %1, 40, 8\n"
+			"	andi    %1, %2, 0x7\n"
+			"	beq     $0, %1, 9f\n"
+			"	daddiu  %2, %2, -1\n"
+			"7:	lb      %1, 0(%2)\n"
+			"	dinsu    %0, %1, 48, 8\n"
+			"	andi    %1, %2, 0x7\n"
+			"	beq     $0, %1, 9f\n"
+			"	daddiu  %2, %2, -1\n"
+			"0:	lb      %1, 0(%2)\n"
+			"	dinsu    %0, %1, 56, 8\n"
+#endif /* CONFIG_CPU_LITTLE_ENDIAN */
+			"9:\n"
+			"	.insn\n"
+			"	.section        .fixup,\"ax\"\n"
+			"8:	li     %3,%4\n"
+			"	j      9b\n"
+			"	.previous\n"
+			"	.section        __ex_table,\"a\"\n"
+			STR(PTR) " 1b,8b\n"
+			STR(PTR) " 2b,8b\n"
+			STR(PTR) " 3b,8b\n"
+			STR(PTR) " 4b,8b\n"
+			STR(PTR) " 5b,8b\n"
+			STR(PTR) " 6b,8b\n"
+			STR(PTR) " 7b,8b\n"
+			STR(PTR) " 0b,8b\n"
+			"	.previous\n"
+			"	.set    pop\n"
+			: "+&r"(rt), "=&r"(rs),
+			  "+&r"(vaddr), "+&r"(err)
+			: "i"(SIGSEGV));
+		if (MIPSInst_RT(inst) && !err)
+			regs->regs[MIPSInst_RT(inst)] = rt;
+
+		MIPS_R2_STATS(loads);
+		break;
+
+	case sdl_op:
+		if (IS_ENABLED(CONFIG_32BIT)) {
+		    err = SIGILL;
+		    break;
+		}
+
+		rt = regs->regs[MIPSInst_RT(inst)];
+		vaddr = regs->regs[MIPSInst_RS(inst)] + MIPSInst_SIMM(inst);
+		if (!access_ok((void __user *)vaddr, 8)) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGSEGV;
+			break;
+		}
+		__asm__ __volatile__(
+			"	.set	push\n"
+			"	.set	reorder\n"
+#ifdef CONFIG_CPU_LITTLE_ENDIAN
+			"	dextu	%1, %0, 56, 8\n"
+			"1:	sb	%1, 0(%2)\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	daddiu	%2, %2, -1\n"
+			"	dextu	%1, %0, 48, 8\n"
+			"2:	sb	%1, 0(%2)\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	daddiu	%2, %2, -1\n"
+			"	dextu	%1, %0, 40, 8\n"
+			"3:	sb	%1, 0(%2)\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	daddiu	%2, %2, -1\n"
+			"	dextu	%1, %0, 32, 8\n"
+			"4:	sb	%1, 0(%2)\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	daddiu	%2, %2, -1\n"
+			"	dext	%1, %0, 24, 8\n"
+			"5:	sb	%1, 0(%2)\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	daddiu	%2, %2, -1\n"
+			"	dext	%1, %0, 16, 8\n"
+			"6:	sb	%1, 0(%2)\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	daddiu	%2, %2, -1\n"
+			"	dext	%1, %0, 8, 8\n"
+			"7:	sb	%1, 0(%2)\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	daddiu	%2, %2, -1\n"
+			"	dext	%1, %0, 0, 8\n"
+			"0:	sb	%1, 0(%2)\n"
+#else /* !CONFIG_CPU_LITTLE_ENDIAN */
+			"	dextu	%1, %0, 56, 8\n"
+			"1:	sb	%1, 0(%2)\n"
+			"	daddiu	%2, %2, 1\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	dextu	%1, %0, 48, 8\n"
+			"2:	sb	%1, 0(%2)\n"
+			"	daddiu	%2, %2, 1\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	dextu	%1, %0, 40, 8\n"
+			"3:	sb	%1, 0(%2)\n"
+			"	daddiu	%2, %2, 1\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	dextu	%1, %0, 32, 8\n"
+			"4:	sb	%1, 0(%2)\n"
+			"	daddiu	%2, %2, 1\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	dext	%1, %0, 24, 8\n"
+			"5:	sb	%1, 0(%2)\n"
+			"	daddiu	%2, %2, 1\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	dext	%1, %0, 16, 8\n"
+			"6:	sb	%1, 0(%2)\n"
+			"	daddiu	%2, %2, 1\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	dext	%1, %0, 8, 8\n"
+			"7:	sb	%1, 0(%2)\n"
+			"	daddiu	%2, %2, 1\n"
+			"	andi	%1, %2, 0x7\n"
+			"	beq	$0, %1, 9f\n"
+			"	dext	%1, %0, 0, 8\n"
+			"0:	sb	%1, 0(%2)\n"
+#endif /* CONFIG_CPU_LITTLE_ENDIAN */
+			"9:\n"
+			"	.insn\n"
+			"	.section        .fixup,\"ax\"\n"
+			"8:	li	%3,%4\n"
+			"	j	9b\n"
+			"	.previous\n"
+			"	.section        __ex_table,\"a\"\n"
+			STR(PTR) " 1b,8b\n"
+			STR(PTR) " 2b,8b\n"
+			STR(PTR) " 3b,8b\n"
+			STR(PTR) " 4b,8b\n"
+			STR(PTR) " 5b,8b\n"
+			STR(PTR) " 6b,8b\n"
+			STR(PTR) " 7b,8b\n"
+			STR(PTR) " 0b,8b\n"
+			"	.previous\n"
+			"	.set	pop\n"
+			: "+&r"(rt), "=&r"(rs),
+			  "+&r"(vaddr), "+&r"(err)
+			: "i"(SIGSEGV)
+			: "memory");
+
+		MIPS_R2_STATS(stores);
+		break;
+
+	case sdr_op:
+		if (IS_ENABLED(CONFIG_32BIT)) {
+		    err = SIGILL;
+		    break;
+		}
+
+		rt = regs->regs[MIPSInst_RT(inst)];
+		vaddr = regs->regs[MIPSInst_RS(inst)] + MIPSInst_SIMM(inst);
+		if (!access_ok((void __user *)vaddr, 8)) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGSEGV;
+			break;
+		}
+		__asm__ __volatile__(
+			"       .set	push\n"
+			"       .set	reorder\n"
+#ifdef CONFIG_CPU_LITTLE_ENDIAN
+			"       dext	%1, %0, 0, 8\n"
+			"1:     sb	%1, 0(%2)\n"
+			"       daddiu	%2, %2, 1\n"
+			"       andi	%1, %2, 0x7\n"
+			"       beq	$0, %1, 9f\n"
+			"       dext	%1, %0, 8, 8\n"
+			"2:     sb	%1, 0(%2)\n"
+			"       daddiu	%2, %2, 1\n"
+			"       andi	%1, %2, 0x7\n"
+			"       beq	$0, %1, 9f\n"
+			"       dext	%1, %0, 16, 8\n"
+			"3:     sb	%1, 0(%2)\n"
+			"       daddiu	%2, %2, 1\n"
+			"       andi	%1, %2, 0x7\n"
+			"       beq	$0, %1, 9f\n"
+			"       dext	%1, %0, 24, 8\n"
+			"4:     sb	%1, 0(%2)\n"
+			"       daddiu	%2, %2, 1\n"
+			"       andi	%1, %2, 0x7\n"
+			"       beq	$0, %1, 9f\n"
+			"       dextu	%1, %0, 32, 8\n"
+			"5:     sb	%1, 0(%2)\n"
+			"       daddiu	%2, %2, 1\n"
+			"       andi	%1, %2, 0x7\n"
+			"       beq	$0, %1, 9f\n"
+			"       dextu	%1, %0, 40, 8\n"
+			"6:     sb	%1, 0(%2)\n"
+			"       daddiu	%2, %2, 1\n"
+			"       andi	%1, %2, 0x7\n"
+			"       beq	$0, %1, 9f\n"
+			"       dextu	%1, %0, 48, 8\n"
+			"7:     sb	%1, 0(%2)\n"
+			"       daddiu	%2, %2, 1\n"
+			"       andi	%1, %2, 0x7\n"
+			"       beq	$0, %1, 9f\n"
+			"       dextu	%1, %0, 56, 8\n"
+			"0:     sb	%1, 0(%2)\n"
+#else /* !CONFIG_CPU_LITTLE_ENDIAN */
+			"       dext	%1, %0, 0, 8\n"
+			"1:     sb	%1, 0(%2)\n"
+			"       andi	%1, %2, 0x7\n"
+			"       beq	$0, %1, 9f\n"
+			"       daddiu	%2, %2, -1\n"
+			"       dext	%1, %0, 8, 8\n"
+			"2:     sb	%1, 0(%2)\n"
+			"       andi	%1, %2, 0x7\n"
+			"       beq	$0, %1, 9f\n"
+			"       daddiu	%2, %2, -1\n"
+			"       dext	%1, %0, 16, 8\n"
+			"3:     sb	%1, 0(%2)\n"
+			"       andi	%1, %2, 0x7\n"
+			"       beq	$0, %1, 9f\n"
+			"       daddiu	%2, %2, -1\n"
+			"       dext	%1, %0, 24, 8\n"
+			"4:     sb	%1, 0(%2)\n"
+			"       andi	%1, %2, 0x7\n"
+			"       beq	$0, %1, 9f\n"
+			"       daddiu	%2, %2, -1\n"
+			"       dextu	%1, %0, 32, 8\n"
+			"5:     sb	%1, 0(%2)\n"
+			"       andi	%1, %2, 0x7\n"
+			"       beq	$0, %1, 9f\n"
+			"       daddiu	%2, %2, -1\n"
+			"       dextu	%1, %0, 40, 8\n"
+			"6:     sb	%1, 0(%2)\n"
+			"       andi	%1, %2, 0x7\n"
+			"       beq	$0, %1, 9f\n"
+			"       daddiu	%2, %2, -1\n"
+			"       dextu	%1, %0, 48, 8\n"
+			"7:     sb	%1, 0(%2)\n"
+			"       andi	%1, %2, 0x7\n"
+			"       beq	$0, %1, 9f\n"
+			"       daddiu	%2, %2, -1\n"
+			"       dextu	%1, %0, 56, 8\n"
+			"0:     sb	%1, 0(%2)\n"
+#endif /* CONFIG_CPU_LITTLE_ENDIAN */
+			"9:\n"
+			"       .insn\n"
+			"       .section        .fixup,\"ax\"\n"
+			"8:     li	%3,%4\n"
+			"       j	9b\n"
+			"       .previous\n"
+			"       .section        __ex_table,\"a\"\n"
+			STR(PTR) " 1b,8b\n"
+			STR(PTR) " 2b,8b\n"
+			STR(PTR) " 3b,8b\n"
+			STR(PTR) " 4b,8b\n"
+			STR(PTR) " 5b,8b\n"
+			STR(PTR) " 6b,8b\n"
+			STR(PTR) " 7b,8b\n"
+			STR(PTR) " 0b,8b\n"
+			"       .previous\n"
+			"       .set	pop\n"
+			: "+&r"(rt), "=&r"(rs),
+			  "+&r"(vaddr), "+&r"(err)
+			: "i"(SIGSEGV)
+			: "memory");
+
+		MIPS_R2_STATS(stores);
+
+		break;
+	case ll_op:
+		vaddr = regs->regs[MIPSInst_RS(inst)] + MIPSInst_SIMM(inst);
+		if (vaddr & 0x3) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGBUS;
+			break;
+		}
+		if (!access_ok((void __user *)vaddr, 4)) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGBUS;
+			break;
+		}
+
+		if (!cpu_has_rw_llb) {
+			/*
+			 * An LL/SC block can't be safely emulated without
+			 * a Config5/LLB availability. So it's probably time to
+			 * kill our process before things get any worse. This is
+			 * because Config5/LLB allows us to use ERETNC so that
+			 * the LLAddr/LLB bit is not cleared when we return from
+			 * an exception. MIPS R2 LL/SC instructions trap with an
+			 * RI exception so once we emulate them here, we return
+			 * back to userland with ERETNC. That preserves the
+			 * LLAddr/LLB so the subsequent SC instruction will
+			 * succeed preserving the atomic semantics of the LL/SC
+			 * block. Without that, there is no safe way to emulate
+			 * an LL/SC block in MIPSR2 userland.
+			 */
+			pr_err("Can't emulate MIPSR2 LL/SC without Config5/LLB\n");
+			err = SIGKILL;
+			break;
+		}
+
+		__asm__ __volatile__(
+			"1:\n"
+			"ll	%0, 0(%2)\n"
+			"2:\n"
+			".insn\n"
+			".section        .fixup,\"ax\"\n"
+			"3:\n"
+			"li	%1, %3\n"
+			"j	2b\n"
+			".previous\n"
+			".section        __ex_table,\"a\"\n"
+			STR(PTR) " 1b,3b\n"
+			".previous\n"
+			: "=&r"(res), "+&r"(err)
+			: "r"(vaddr), "i"(SIGSEGV)
+			: "memory");
+
+		if (MIPSInst_RT(inst) && !err)
+			regs->regs[MIPSInst_RT(inst)] = res;
+		MIPS_R2_STATS(llsc);
+
+		break;
+
+	case sc_op:
+		vaddr = regs->regs[MIPSInst_RS(inst)] + MIPSInst_SIMM(inst);
+		if (vaddr & 0x3) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGBUS;
+			break;
+		}
+		if (!access_ok((void __user *)vaddr, 4)) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGBUS;
+			break;
+		}
+
+		if (!cpu_has_rw_llb) {
+			/*
+			 * An LL/SC block can't be safely emulated without
+			 * a Config5/LLB availability. So it's probably time to
+			 * kill our process before things get any worse. This is
+			 * because Config5/LLB allows us to use ERETNC so that
+			 * the LLAddr/LLB bit is not cleared when we return from
+			 * an exception. MIPS R2 LL/SC instructions trap with an
+			 * RI exception so once we emulate them here, we return
+			 * back to userland with ERETNC. That preserves the
+			 * LLAddr/LLB so the subsequent SC instruction will
+			 * succeed preserving the atomic semantics of the LL/SC
+			 * block. Without that, there is no safe way to emulate
+			 * an LL/SC block in MIPSR2 userland.
+			 */
+			pr_err("Can't emulate MIPSR2 LL/SC without Config5/LLB\n");
+			err = SIGKILL;
+			break;
+		}
+
+		res = regs->regs[MIPSInst_RT(inst)];
+
+		__asm__ __volatile__(
+			"1:\n"
+			"sc	%0, 0(%2)\n"
+			"2:\n"
+			".insn\n"
+			".section        .fixup,\"ax\"\n"
+			"3:\n"
+			"li	%1, %3\n"
+			"j	2b\n"
+			".previous\n"
+			".section        __ex_table,\"a\"\n"
+			STR(PTR) " 1b,3b\n"
+			".previous\n"
+			: "+&r"(res), "+&r"(err)
+			: "r"(vaddr), "i"(SIGSEGV));
+
+		if (MIPSInst_RT(inst) && !err)
+			regs->regs[MIPSInst_RT(inst)] = res;
+
+		MIPS_R2_STATS(llsc);
+
+		break;
+
+	case lld_op:
+		if (IS_ENABLED(CONFIG_32BIT)) {
+		    err = SIGILL;
+		    break;
+		}
+
+		vaddr = regs->regs[MIPSInst_RS(inst)] + MIPSInst_SIMM(inst);
+		if (vaddr & 0x7) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGBUS;
+			break;
+		}
+		if (!access_ok((void __user *)vaddr, 8)) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGBUS;
+			break;
+		}
+
+		if (!cpu_has_rw_llb) {
+			/*
+			 * An LL/SC block can't be safely emulated without
+			 * a Config5/LLB availability. So it's probably time to
+			 * kill our process before things get any worse. This is
+			 * because Config5/LLB allows us to use ERETNC so that
+			 * the LLAddr/LLB bit is not cleared when we return from
+			 * an exception. MIPS R2 LL/SC instructions trap with an
+			 * RI exception so once we emulate them here, we return
+			 * back to userland with ERETNC. That preserves the
+			 * LLAddr/LLB so the subsequent SC instruction will
+			 * succeed preserving the atomic semantics of the LL/SC
+			 * block. Without that, there is no safe way to emulate
+			 * an LL/SC block in MIPSR2 userland.
+			 */
+			pr_err("Can't emulate MIPSR2 LL/SC without Config5/LLB\n");
+			err = SIGKILL;
+			break;
+		}
+
+		__asm__ __volatile__(
+			"1:\n"
+			"lld	%0, 0(%2)\n"
+			"2:\n"
+			".insn\n"
+			".section        .fixup,\"ax\"\n"
+			"3:\n"
+			"li	%1, %3\n"
+			"j	2b\n"
+			".previous\n"
+			".section        __ex_table,\"a\"\n"
+			STR(PTR) " 1b,3b\n"
+			".previous\n"
+			: "=&r"(res), "+&r"(err)
+			: "r"(vaddr), "i"(SIGSEGV)
+			: "memory");
+		if (MIPSInst_RT(inst) && !err)
+			regs->regs[MIPSInst_RT(inst)] = res;
+
+		MIPS_R2_STATS(llsc);
+
+		break;
+
+	case scd_op:
+		if (IS_ENABLED(CONFIG_32BIT)) {
+		    err = SIGILL;
+		    break;
+		}
+
+		vaddr = regs->regs[MIPSInst_RS(inst)] + MIPSInst_SIMM(inst);
+		if (vaddr & 0x7) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGBUS;
+			break;
+		}
+		if (!access_ok((void __user *)vaddr, 8)) {
+			current->thread.cp0_baduaddr = vaddr;
+			err = SIGBUS;
+			break;
+		}
+
+		if (!cpu_has_rw_llb) {
+			/*
+			 * An LL/SC block can't be safely emulated without
+			 * a Config5/LLB availability. So it's probably time to
+			 * kill our process before things get any worse. This is
+			 * because Config5/LLB allows us to use ERETNC so that
+			 * the LLAddr/LLB bit is not cleared when we return from
+			 * an exception. MIPS R2 LL/SC instructions trap with an
+			 * RI exception so once we emulate them here, we return
+			 * back to userland with ERETNC. That preserves the
+			 * LLAddr/LLB so the subsequent SC instruction will
+			 * succeed preserving the atomic semantics of the LL/SC
+			 * block. Without that, there is no safe way to emulate
+			 * an LL/SC block in MIPSR2 userland.
+			 */
+			pr_err("Can't emulate MIPSR2 LL/SC without Config5/LLB\n");
+			err = SIGKILL;
+			break;
+		}
+
+		res = regs->regs[MIPSInst_RT(inst)];
+
+		__asm__ __volatile__(
+			"1:\n"
+			"scd	%0, 0(%2)\n"
+			"2:\n"
+			".insn\n"
+			".section        .fixup,\"ax\"\n"
+			"3:\n"
+			"li	%1, %3\n"
+			"j	2b\n"
+			".previous\n"
+			".section        __ex_table,\"a\"\n"
+			STR(PTR) " 1b,3b\n"
+			".previous\n"
+			: "+&r"(res), "+&r"(err)
+			: "r"(vaddr), "i"(SIGSEGV));
+
+		if (MIPSInst_RT(inst) && !err)
+			regs->regs[MIPSInst_RT(inst)] = res;
+
+		MIPS_R2_STATS(llsc);
+
+		break;
+	case pref_op:
+		/* skip it */
+		break;
+	default:
+		err = SIGILL;
+	}
+
+	/*
+	 * Let's not return to userland just yet. It's costly and
+	 * it's likely we have more R2 instructions to emulate
+	 */
+	if (!err && (pass++ < MIPS_R2_EMUL_TOTAL_PASS)) {
+		regs->cp0_cause &= ~CAUSEF_BD;
+		err = get_user(inst, (u32 __user *)regs->cp0_epc);
+		if (!err)
+			goto repeat;
+
+		if (err < 0)
+			err = SIGSEGV;
+	}
+
+	if (err && (err != SIGEMT)) {
+		regs->regs[31] = r31;
+		regs->cp0_epc = epc;
+	}
+
+	/* Likely a MIPS R6 compatible instruction */
+	if (pass && (err == SIGILL))
+		err = 0;
+
+	return err;
+}
+
+#ifdef CONFIG_DEBUG_FS
+
+static int mipsr2_emul_show(struct seq_file *s, void *unused)
+{
+
+	seq_printf(s, "Instruction\tTotal\tBDslot\n------------------------------\n");
+	seq_printf(s, "movs\t\t%ld\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2emustats.movs),
+		   (unsigned long)__this_cpu_read(mipsr2bdemustats.movs));
+	seq_printf(s, "hilo\t\t%ld\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2emustats.hilo),
+		   (unsigned long)__this_cpu_read(mipsr2bdemustats.hilo));
+	seq_printf(s, "muls\t\t%ld\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2emustats.muls),
+		   (unsigned long)__this_cpu_read(mipsr2bdemustats.muls));
+	seq_printf(s, "divs\t\t%ld\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2emustats.divs),
+		   (unsigned long)__this_cpu_read(mipsr2bdemustats.divs));
+	seq_printf(s, "dsps\t\t%ld\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2emustats.dsps),
+		   (unsigned long)__this_cpu_read(mipsr2bdemustats.dsps));
+	seq_printf(s, "bops\t\t%ld\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2emustats.bops),
+		   (unsigned long)__this_cpu_read(mipsr2bdemustats.bops));
+	seq_printf(s, "traps\t\t%ld\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2emustats.traps),
+		   (unsigned long)__this_cpu_read(mipsr2bdemustats.traps));
+	seq_printf(s, "fpus\t\t%ld\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2emustats.fpus),
+		   (unsigned long)__this_cpu_read(mipsr2bdemustats.fpus));
+	seq_printf(s, "loads\t\t%ld\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2emustats.loads),
+		   (unsigned long)__this_cpu_read(mipsr2bdemustats.loads));
+	seq_printf(s, "stores\t\t%ld\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2emustats.stores),
+		   (unsigned long)__this_cpu_read(mipsr2bdemustats.stores));
+	seq_printf(s, "llsc\t\t%ld\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2emustats.llsc),
+		   (unsigned long)__this_cpu_read(mipsr2bdemustats.llsc));
+	seq_printf(s, "dsemul\t\t%ld\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2emustats.dsemul),
+		   (unsigned long)__this_cpu_read(mipsr2bdemustats.dsemul));
+	seq_printf(s, "jr\t\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2bremustats.jrs));
+	seq_printf(s, "bltzl\t\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2bremustats.bltzl));
+	seq_printf(s, "bgezl\t\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2bremustats.bgezl));
+	seq_printf(s, "bltzll\t\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2bremustats.bltzll));
+	seq_printf(s, "bgezll\t\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2bremustats.bgezll));
+	seq_printf(s, "bltzal\t\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2bremustats.bltzal));
+	seq_printf(s, "bgezal\t\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2bremustats.bgezal));
+	seq_printf(s, "beql\t\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2bremustats.beql));
+	seq_printf(s, "bnel\t\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2bremustats.bnel));
+	seq_printf(s, "blezl\t\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2bremustats.blezl));
+	seq_printf(s, "bgtzl\t\t%ld\n",
+		   (unsigned long)__this_cpu_read(mipsr2bremustats.bgtzl));
+
+	return 0;
+}
+
+static int mipsr2_clear_show(struct seq_file *s, void *unused)
+{
+	mipsr2_emul_show(s, unused);
+
+	__this_cpu_write((mipsr2emustats).movs, 0);
+	__this_cpu_write((mipsr2bdemustats).movs, 0);
+	__this_cpu_write((mipsr2emustats).hilo, 0);
+	__this_cpu_write((mipsr2bdemustats).hilo, 0);
+	__this_cpu_write((mipsr2emustats).muls, 0);
+	__this_cpu_write((mipsr2bdemustats).muls, 0);
+	__this_cpu_write((mipsr2emustats).divs, 0);
+	__this_cpu_write((mipsr2bdemustats).divs, 0);
+	__this_cpu_write((mipsr2emustats).dsps, 0);
+	__this_cpu_write((mipsr2bdemustats).dsps, 0);
+	__this_cpu_write((mipsr2emustats).bops, 0);
+	__this_cpu_write((mipsr2bdemustats).bops, 0);
+	__this_cpu_write((mipsr2emustats).traps, 0);
+	__this_cpu_write((mipsr2bdemustats).traps, 0);
+	__this_cpu_write((mipsr2emustats).fpus, 0);
+	__this_cpu_write((mipsr2bdemustats).fpus, 0);
+	__this_cpu_write((mipsr2emustats).loads, 0);
+	__this_cpu_write((mipsr2bdemustats).loads, 0);
+	__this_cpu_write((mipsr2emustats).stores, 0);
+	__this_cpu_write((mipsr2bdemustats).stores, 0);
+	__this_cpu_write((mipsr2emustats).llsc, 0);
+	__this_cpu_write((mipsr2bdemustats).llsc, 0);
+	__this_cpu_write((mipsr2emustats).dsemul, 0);
+	__this_cpu_write((mipsr2bdemustats).dsemul, 0);
+	__this_cpu_write((mipsr2bremustats).jrs, 0);
+	__this_cpu_write((mipsr2bremustats).bltzl, 0);
+	__this_cpu_write((mipsr2bremustats).bgezl, 0);
+	__this_cpu_write((mipsr2bremustats).bltzll, 0);
+	__this_cpu_write((mipsr2bremustats).bgezll, 0);
+	__this_cpu_write((mipsr2bremustats).bltzall, 0);
+	__this_cpu_write((mipsr2bremustats).bgezall, 0);
+	__this_cpu_write((mipsr2bremustats).bltzal, 0);
+	__this_cpu_write((mipsr2bremustats).bgezal, 0);
+	__this_cpu_write((mipsr2bremustats).beql, 0);
+	__this_cpu_write((mipsr2bremustats).bnel, 0);
+	__this_cpu_write((mipsr2bremustats).blezl, 0);
+	__this_cpu_write((mipsr2bremustats).bgtzl, 0);
+
+	return 0;
+}
+
+DEFINE_SHOW_ATTRIBUTE(mipsr2_emul);
+DEFINE_SHOW_ATTRIBUTE(mipsr2_clear);
+
+static int __init mipsr2_init_debugfs(void)
+{
+	debugfs_create_file("r2_emul_stats", S_IRUGO, mips_debugfs_dir, NULL,
+			    &mipsr2_emul_fops);
+	debugfs_create_file("r2_emul_stats_clear", S_IRUGO, mips_debugfs_dir,
+			    NULL, &mipsr2_clear_fops);
+	return 0;
+}
+
+device_initcall(mipsr2_init_debugfs);
+
+#endif /* CONFIG_DEBUG_FS */
diff --git a/arch/mips/kernel/module.c b/arch/mips/kernel/module.c
new file mode 100644
index 000000000..3c0c3d126
--- /dev/null
+++ b/arch/mips/kernel/module.c
@@ -0,0 +1,457 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ *
+ *  Copyright (C) 2001 Rusty Russell.
+ *  Copyright (C) 2003, 2004 Ralf Baechle (ralf@linux-mips.org)
+ *  Copyright (C) 2005 Thiemo Seufer
+ */
+
+#undef DEBUG
+
+#include <linux/extable.h>
+#include <linux/moduleloader.h>
+#include <linux/elf.h>
+#include <linux/mm.h>
+#include <linux/numa.h>
+#include <linux/vmalloc.h>
+#include <linux/slab.h>
+#include <linux/fs.h>
+#include <linux/string.h>
+#include <linux/kernel.h>
+#include <linux/spinlock.h>
+#include <linux/jump_label.h>
+
+
+struct mips_hi16 {
+	struct mips_hi16 *next;
+	Elf_Addr *addr;
+	Elf_Addr value;
+};
+
+static LIST_HEAD(dbe_list);
+static DEFINE_SPINLOCK(dbe_lock);
+
+#ifdef MODULE_START
+void *module_alloc(unsigned long size)
+{
+	return __vmalloc_node_range(size, 1, MODULE_START, MODULE_END,
+				GFP_KERNEL, PAGE_KERNEL, 0, NUMA_NO_NODE,
+				__builtin_return_address(0));
+}
+#endif
+
+static int apply_r_mips_none(struct module *me, u32 *location,
+			     u32 base, Elf_Addr v, bool rela)
+{
+	return 0;
+}
+
+static int apply_r_mips_32(struct module *me, u32 *location,
+			   u32 base, Elf_Addr v, bool rela)
+{
+	*location = base + v;
+
+	return 0;
+}
+
+static int apply_r_mips_26(struct module *me, u32 *location,
+			   u32 base, Elf_Addr v, bool rela)
+{
+	if (v % 4) {
+		pr_err("module %s: dangerous R_MIPS_26 relocation\n",
+		       me->name);
+		return -ENOEXEC;
+	}
+
+	if ((v & 0xf0000000) != (((unsigned long)location + 4) & 0xf0000000)) {
+		pr_err("module %s: relocation overflow\n",
+		       me->name);
+		return -ENOEXEC;
+	}
+
+	*location = (*location & ~0x03ffffff) |
+		    ((base + (v >> 2)) & 0x03ffffff);
+
+	return 0;
+}
+
+static int apply_r_mips_hi16(struct module *me, u32 *location,
+			     u32 base, Elf_Addr v, bool rela)
+{
+	struct mips_hi16 *n;
+
+	if (rela) {
+		*location = (*location & 0xffff0000) |
+			    ((((long long) v + 0x8000LL) >> 16) & 0xffff);
+		return 0;
+	}
+
+	/*
+	 * We cannot relocate this one now because we don't know the value of
+	 * the carry we need to add.  Save the information, and let LO16 do the
+	 * actual relocation.
+	 */
+	n = kmalloc(sizeof *n, GFP_KERNEL);
+	if (!n)
+		return -ENOMEM;
+
+	n->addr = (Elf_Addr *)location;
+	n->value = v;
+	n->next = me->arch.r_mips_hi16_list;
+	me->arch.r_mips_hi16_list = n;
+
+	return 0;
+}
+
+static void free_relocation_chain(struct mips_hi16 *l)
+{
+	struct mips_hi16 *next;
+
+	while (l) {
+		next = l->next;
+		kfree(l);
+		l = next;
+	}
+}
+
+static int apply_r_mips_lo16(struct module *me, u32 *location,
+			     u32 base, Elf_Addr v, bool rela)
+{
+	unsigned long insnlo = base;
+	struct mips_hi16 *l;
+	Elf_Addr val, vallo;
+
+	if (rela) {
+		*location = (*location & 0xffff0000) | (v & 0xffff);
+		return 0;
+	}
+
+	/* Sign extend the addend we extract from the lo insn.	*/
+	vallo = ((insnlo & 0xffff) ^ 0x8000) - 0x8000;
+
+	if (me->arch.r_mips_hi16_list != NULL) {
+		l = me->arch.r_mips_hi16_list;
+		while (l != NULL) {
+			struct mips_hi16 *next;
+			unsigned long insn;
+
+			/*
+			 * The value for the HI16 had best be the same.
+			 */
+			if (v != l->value)
+				goto out_danger;
+
+			/*
+			 * Do the HI16 relocation.  Note that we actually don't
+			 * need to know anything about the LO16 itself, except
+			 * where to find the low 16 bits of the addend needed
+			 * by the LO16.
+			 */
+			insn = *l->addr;
+			val = ((insn & 0xffff) << 16) + vallo;
+			val += v;
+
+			/*
+			 * Account for the sign extension that will happen in
+			 * the low bits.
+			 */
+			val = ((val >> 16) + ((val & 0x8000) != 0)) & 0xffff;
+
+			insn = (insn & ~0xffff) | val;
+			*l->addr = insn;
+
+			next = l->next;
+			kfree(l);
+			l = next;
+		}
+
+		me->arch.r_mips_hi16_list = NULL;
+	}
+
+	/*
+	 * Ok, we're done with the HI16 relocs.	 Now deal with the LO16.
+	 */
+	val = v + vallo;
+	insnlo = (insnlo & ~0xffff) | (val & 0xffff);
+	*location = insnlo;
+
+	return 0;
+
+out_danger:
+	free_relocation_chain(l);
+	me->arch.r_mips_hi16_list = NULL;
+
+	pr_err("module %s: dangerous R_MIPS_LO16 relocation\n", me->name);
+
+	return -ENOEXEC;
+}
+
+static int apply_r_mips_pc(struct module *me, u32 *location, u32 base,
+			   Elf_Addr v, unsigned int bits)
+{
+	unsigned long mask = GENMASK(bits - 1, 0);
+	unsigned long se_bits;
+	long offset;
+
+	if (v % 4) {
+		pr_err("module %s: dangerous R_MIPS_PC%u relocation\n",
+		       me->name, bits);
+		return -ENOEXEC;
+	}
+
+	/* retrieve & sign extend implicit addend if any */
+	offset = base & mask;
+	offset |= (offset & BIT(bits - 1)) ? ~mask : 0;
+
+	offset += ((long)v - (long)location) >> 2;
+
+	/* check the sign bit onwards are identical - ie. we didn't overflow */
+	se_bits = (offset & BIT(bits - 1)) ? ~0ul : 0;
+	if ((offset & ~mask) != (se_bits & ~mask)) {
+		pr_err("module %s: relocation overflow\n", me->name);
+		return -ENOEXEC;
+	}
+
+	*location = (*location & ~mask) | (offset & mask);
+
+	return 0;
+}
+
+static int apply_r_mips_pc16(struct module *me, u32 *location,
+			     u32 base, Elf_Addr v, bool rela)
+{
+	return apply_r_mips_pc(me, location, base, v, 16);
+}
+
+static int apply_r_mips_pc21(struct module *me, u32 *location,
+			     u32 base, Elf_Addr v, bool rela)
+{
+	return apply_r_mips_pc(me, location, base, v, 21);
+}
+
+static int apply_r_mips_pc26(struct module *me, u32 *location,
+			     u32 base, Elf_Addr v, bool rela)
+{
+	return apply_r_mips_pc(me, location, base, v, 26);
+}
+
+static int apply_r_mips_64(struct module *me, u32 *location,
+			   u32 base, Elf_Addr v, bool rela)
+{
+	if (WARN_ON(!rela))
+		return -EINVAL;
+
+	*(Elf_Addr *)location = v;
+
+	return 0;
+}
+
+static int apply_r_mips_higher(struct module *me, u32 *location,
+			       u32 base, Elf_Addr v, bool rela)
+{
+	if (WARN_ON(!rela))
+		return -EINVAL;
+
+	*location = (*location & 0xffff0000) |
+		    ((((long long)v + 0x80008000LL) >> 32) & 0xffff);
+
+	return 0;
+}
+
+static int apply_r_mips_highest(struct module *me, u32 *location,
+				u32 base, Elf_Addr v, bool rela)
+{
+	if (WARN_ON(!rela))
+		return -EINVAL;
+
+	*location = (*location & 0xffff0000) |
+		    ((((long long)v + 0x800080008000LL) >> 48) & 0xffff);
+
+	return 0;
+}
+
+/**
+ * reloc_handler() - Apply a particular relocation to a module
+ * @me: the module to apply the reloc to
+ * @location: the address at which the reloc is to be applied
+ * @base: the existing value at location for REL-style; 0 for RELA-style
+ * @v: the value of the reloc, with addend for RELA-style
+ *
+ * Each implemented reloc_handler function applies a particular type of
+ * relocation to the module @me. Relocs that may be found in either REL or RELA
+ * variants can be handled by making use of the @base & @v parameters which are
+ * set to values which abstract the difference away from the particular reloc
+ * implementations.
+ *
+ * Return: 0 upon success, else -ERRNO
+ */
+typedef int (*reloc_handler)(struct module *me, u32 *location,
+			     u32 base, Elf_Addr v, bool rela);
+
+/* The handlers for known reloc types */
+static reloc_handler reloc_handlers[] = {
+	[R_MIPS_NONE]		= apply_r_mips_none,
+	[R_MIPS_32]		= apply_r_mips_32,
+	[R_MIPS_26]		= apply_r_mips_26,
+	[R_MIPS_HI16]		= apply_r_mips_hi16,
+	[R_MIPS_LO16]		= apply_r_mips_lo16,
+	[R_MIPS_PC16]		= apply_r_mips_pc16,
+	[R_MIPS_64]		= apply_r_mips_64,
+	[R_MIPS_HIGHER]		= apply_r_mips_higher,
+	[R_MIPS_HIGHEST]	= apply_r_mips_highest,
+	[R_MIPS_PC21_S2]	= apply_r_mips_pc21,
+	[R_MIPS_PC26_S2]	= apply_r_mips_pc26,
+};
+
+static int __apply_relocate(Elf_Shdr *sechdrs, const char *strtab,
+			    unsigned int symindex, unsigned int relsec,
+			    struct module *me, bool rela)
+{
+	union {
+		Elf_Mips_Rel *rel;
+		Elf_Mips_Rela *rela;
+	} r;
+	reloc_handler handler;
+	Elf_Sym *sym;
+	u32 *location, base;
+	unsigned int i, type;
+	Elf_Addr v;
+	int err = 0;
+	size_t reloc_sz;
+
+	pr_debug("Applying relocate section %u to %u\n", relsec,
+	       sechdrs[relsec].sh_info);
+
+	r.rel = (void *)sechdrs[relsec].sh_addr;
+	reloc_sz = rela ? sizeof(*r.rela) : sizeof(*r.rel);
+	me->arch.r_mips_hi16_list = NULL;
+	for (i = 0; i < sechdrs[relsec].sh_size / reloc_sz; i++) {
+		/* This is where to make the change */
+		location = (void *)sechdrs[sechdrs[relsec].sh_info].sh_addr
+			+ r.rel->r_offset;
+		/* This is the symbol it is referring to */
+		sym = (Elf_Sym *)sechdrs[symindex].sh_addr
+			+ ELF_MIPS_R_SYM(*r.rel);
+		if (sym->st_value >= -MAX_ERRNO) {
+			/* Ignore unresolved weak symbol */
+			if (ELF_ST_BIND(sym->st_info) == STB_WEAK)
+				continue;
+			pr_warn("%s: Unknown symbol %s\n",
+				me->name, strtab + sym->st_name);
+			err = -ENOENT;
+			goto out;
+		}
+
+		type = ELF_MIPS_R_TYPE(*r.rel);
+		if (type < ARRAY_SIZE(reloc_handlers))
+			handler = reloc_handlers[type];
+		else
+			handler = NULL;
+
+		if (!handler) {
+			pr_err("%s: Unknown relocation type %u\n",
+			       me->name, type);
+			err = -EINVAL;
+			goto out;
+		}
+
+		if (rela) {
+			v = sym->st_value + r.rela->r_addend;
+			base = 0;
+			r.rela = &r.rela[1];
+		} else {
+			v = sym->st_value;
+			base = *location;
+			r.rel = &r.rel[1];
+		}
+
+		err = handler(me, location, base, v, rela);
+		if (err)
+			goto out;
+	}
+
+out:
+	/*
+	 * Normally the hi16 list should be deallocated at this point. A
+	 * malformed binary however could contain a series of R_MIPS_HI16
+	 * relocations not followed by a R_MIPS_LO16 relocation, or if we hit
+	 * an error processing a reloc we might have gotten here before
+	 * reaching the R_MIPS_LO16. In either case, free up the list and
+	 * return an error.
+	 */
+	if (me->arch.r_mips_hi16_list) {
+		free_relocation_chain(me->arch.r_mips_hi16_list);
+		me->arch.r_mips_hi16_list = NULL;
+		err = err ?: -ENOEXEC;
+	}
+
+	return err;
+}
+
+int apply_relocate(Elf_Shdr *sechdrs, const char *strtab,
+		   unsigned int symindex, unsigned int relsec,
+		   struct module *me)
+{
+	return __apply_relocate(sechdrs, strtab, symindex, relsec, me, false);
+}
+
+#ifdef CONFIG_MODULES_USE_ELF_RELA
+int apply_relocate_add(Elf_Shdr *sechdrs, const char *strtab,
+		       unsigned int symindex, unsigned int relsec,
+		       struct module *me)
+{
+	return __apply_relocate(sechdrs, strtab, symindex, relsec, me, true);
+}
+#endif /* CONFIG_MODULES_USE_ELF_RELA */
+
+/* Given an address, look for it in the module exception tables. */
+const struct exception_table_entry *search_module_dbetables(unsigned long addr)
+{
+	unsigned long flags;
+	const struct exception_table_entry *e = NULL;
+	struct mod_arch_specific *dbe;
+
+	spin_lock_irqsave(&dbe_lock, flags);
+	list_for_each_entry(dbe, &dbe_list, dbe_list) {
+		e = search_extable(dbe->dbe_start,
+				   dbe->dbe_end - dbe->dbe_start, addr);
+		if (e)
+			break;
+	}
+	spin_unlock_irqrestore(&dbe_lock, flags);
+
+	/* Now, if we found one, we are running inside it now, hence
+	   we cannot unload the module, hence no refcnt needed. */
+	return e;
+}
+
+/* Put in dbe list if necessary. */
+int module_finalize(const Elf_Ehdr *hdr,
+		    const Elf_Shdr *sechdrs,
+		    struct module *me)
+{
+	const Elf_Shdr *s;
+	char *secstrings = (void *)hdr + sechdrs[hdr->e_shstrndx].sh_offset;
+
+	/* Make jump label nops. */
+	jump_label_apply_nops(me);
+
+	INIT_LIST_HEAD(&me->arch.dbe_list);
+	for (s = sechdrs; s < sechdrs + hdr->e_shnum; s++) {
+		if (strcmp("__dbe_table", secstrings + s->sh_name) != 0)
+			continue;
+		me->arch.dbe_start = (void *)s->sh_addr;
+		me->arch.dbe_end = (void *)s->sh_addr + s->sh_size;
+		spin_lock_irq(&dbe_lock);
+		list_add(&me->arch.dbe_list, &dbe_list);
+		spin_unlock_irq(&dbe_lock);
+	}
+	return 0;
+}
+
+void module_arch_cleanup(struct module *mod)
+{
+	spin_lock_irq(&dbe_lock);
+	list_del(&mod->arch.dbe_list);
+	spin_unlock_irq(&dbe_lock);
+}
diff --git a/arch/mips/kernel/octeon_switch.S b/arch/mips/kernel/octeon_switch.S
new file mode 100644
index 000000000..896080b44
--- /dev/null
+++ b/arch/mips/kernel/octeon_switch.S
@@ -0,0 +1,554 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1994, 1995, 1996, 1998, 1999, 2002, 2003 Ralf Baechle
+ * Copyright (C) 1996 David S. Miller (davem@davemloft.net)
+ * Copyright (C) 1994, 1995, 1996, by Andreas Busse
+ * Copyright (C) 1999 Silicon Graphics, Inc.
+ * Copyright (C) 2000 MIPS Technologies, Inc.
+ *    written by Carsten Langgaard, carstenl@mips.com
+ */
+#include <asm/asm.h>
+#include <asm/export.h>
+#include <asm/asm-offsets.h>
+#include <asm/mipsregs.h>
+#include <asm/regdef.h>
+#include <asm/stackframe.h>
+
+/*
+ * task_struct *resume(task_struct *prev, task_struct *next,
+ *		       struct thread_info *next_ti)
+ */
+	.align	7
+	LEAF(resume)
+	.set arch=octeon
+	mfc0	t1, CP0_STATUS
+	LONG_S	t1, THREAD_STATUS(a0)
+	cpu_save_nonscratch a0
+	LONG_S	ra, THREAD_REG31(a0)
+
+#if CONFIG_CAVIUM_OCTEON_CVMSEG_SIZE > 0
+	/* Check if we need to store CVMSEG state */
+	dmfc0	t0, $11,7	/* CvmMemCtl */
+	bbit0	t0, 6, 3f	/* Is user access enabled? */
+
+	/* Store the CVMSEG state */
+	/* Extract the size of CVMSEG */
+	andi	t0, 0x3f
+	/* Multiply * (cache line size/sizeof(long)/2) */
+	sll	t0, 7-LONGLOG-1
+	li	t1, -32768	/* Base address of CVMSEG */
+	LONG_ADDI t2, a0, THREAD_CVMSEG /* Where to store CVMSEG to */
+	synciobdma
+2:
+	.set noreorder
+	LONG_L	t8, 0(t1)	/* Load from CVMSEG */
+	subu	t0, 1		/* Decrement loop var */
+	LONG_L	t9, LONGSIZE(t1)/* Load from CVMSEG */
+	LONG_ADDU t1, LONGSIZE*2 /* Increment loc in CVMSEG */
+	LONG_S	t8, 0(t2)	/* Store CVMSEG to thread storage */
+	LONG_ADDU t2, LONGSIZE*2 /* Increment loc in thread storage */
+	bnez	t0, 2b		/* Loop until we've copied it all */
+	 LONG_S t9, -LONGSIZE(t2)/* Store CVMSEG to thread storage */
+	.set reorder
+
+	/* Disable access to CVMSEG */
+	dmfc0	t0, $11,7	/* CvmMemCtl */
+	xori	t0, t0, 0x40	/* Bit 6 is CVMSEG user enable */
+	dmtc0	t0, $11,7	/* CvmMemCtl */
+#endif
+3:
+
+#if defined(CONFIG_STACKPROTECTOR) && !defined(CONFIG_SMP)
+	PTR_LA	t8, __stack_chk_guard
+	LONG_L	t9, TASK_STACK_CANARY(a1)
+	LONG_S	t9, 0(t8)
+#endif
+
+	/*
+	 * The order of restoring the registers takes care of the race
+	 * updating $28, $29 and kernelsp without disabling ints.
+	 */
+	move	$28, a2
+	cpu_restore_nonscratch a1
+
+	PTR_ADDU	t0, $28, _THREAD_SIZE - 32
+	set_saved_sp	t0, t1, t2
+
+	mfc0	t1, CP0_STATUS		/* Do we really need this? */
+	li	a3, 0xff01
+	and	t1, a3
+	LONG_L	a2, THREAD_STATUS(a1)
+	nor	a3, $0, a3
+	and	a2, a3
+	or	a2, t1
+	mtc0	a2, CP0_STATUS
+	move	v0, a0
+	jr	ra
+	END(resume)
+
+/*
+ * void octeon_cop2_save(struct octeon_cop2_state *a0)
+ */
+	.align	7
+	.set push
+	.set noreorder
+	LEAF(octeon_cop2_save)
+
+	dmfc0	t9, $9,7	/* CvmCtl register. */
+
+	/* Save the COP2 CRC state */
+	dmfc2	t0, 0x0201
+	dmfc2	t1, 0x0202
+	dmfc2	t2, 0x0200
+	sd	t0, OCTEON_CP2_CRC_IV(a0)
+	sd	t1, OCTEON_CP2_CRC_LENGTH(a0)
+	/* Skip next instructions if CvmCtl[NODFA_CP2] set */
+	bbit1	t9, 28, 1f
+	 sd	t2, OCTEON_CP2_CRC_POLY(a0)
+
+	/* Save the LLM state */
+	dmfc2	t0, 0x0402
+	dmfc2	t1, 0x040A
+	sd	t0, OCTEON_CP2_LLM_DAT(a0)
+
+1:	bbit1	t9, 26, 3f	/* done if CvmCtl[NOCRYPTO] set */
+	 sd	t1, OCTEON_CP2_LLM_DAT+8(a0)
+
+	/* Save the COP2 crypto state */
+	/* this part is mostly common to both pass 1 and later revisions */
+	dmfc2	t0, 0x0084
+	dmfc2	t1, 0x0080
+	dmfc2	t2, 0x0081
+	dmfc2	t3, 0x0082
+	sd	t0, OCTEON_CP2_3DES_IV(a0)
+	dmfc2	t0, 0x0088
+	sd	t1, OCTEON_CP2_3DES_KEY(a0)
+	dmfc2	t1, 0x0111			/* only necessary for pass 1 */
+	sd	t2, OCTEON_CP2_3DES_KEY+8(a0)
+	dmfc2	t2, 0x0102
+	sd	t3, OCTEON_CP2_3DES_KEY+16(a0)
+	dmfc2	t3, 0x0103
+	sd	t0, OCTEON_CP2_3DES_RESULT(a0)
+	dmfc2	t0, 0x0104
+	sd	t1, OCTEON_CP2_AES_INP0(a0)	/* only necessary for pass 1 */
+	dmfc2	t1, 0x0105
+	sd	t2, OCTEON_CP2_AES_IV(a0)
+	dmfc2	t2, 0x0106
+	sd	t3, OCTEON_CP2_AES_IV+8(a0)
+	dmfc2	t3, 0x0107
+	sd	t0, OCTEON_CP2_AES_KEY(a0)
+	dmfc2	t0, 0x0110
+	sd	t1, OCTEON_CP2_AES_KEY+8(a0)
+	dmfc2	t1, 0x0100
+	sd	t2, OCTEON_CP2_AES_KEY+16(a0)
+	dmfc2	t2, 0x0101
+	sd	t3, OCTEON_CP2_AES_KEY+24(a0)
+	mfc0	v0, $15,0	/* Get the processor ID register */
+	sd	t0, OCTEON_CP2_AES_KEYLEN(a0)
+	li	v1, 0x000d0000	/* This is the processor ID of Octeon Pass1 */
+	sd	t1, OCTEON_CP2_AES_RESULT(a0)
+	/* Skip to the Pass1 version of the remainder of the COP2 state */
+	beq	v0, v1, 2f
+	 sd	t2, OCTEON_CP2_AES_RESULT+8(a0)
+
+	/* the non-pass1 state when !CvmCtl[NOCRYPTO] */
+	dmfc2	t1, 0x0240
+	dmfc2	t2, 0x0241
+	ori	v1, v1, 0x9500 /* lowest OCTEON III PrId*/
+	dmfc2	t3, 0x0242
+	subu	v1, v0, v1 /* prid - lowest OCTEON III PrId */
+	dmfc2	t0, 0x0243
+	sd	t1, OCTEON_CP2_HSH_DATW(a0)
+	dmfc2	t1, 0x0244
+	sd	t2, OCTEON_CP2_HSH_DATW+8(a0)
+	dmfc2	t2, 0x0245
+	sd	t3, OCTEON_CP2_HSH_DATW+16(a0)
+	dmfc2	t3, 0x0246
+	sd	t0, OCTEON_CP2_HSH_DATW+24(a0)
+	dmfc2	t0, 0x0247
+	sd	t1, OCTEON_CP2_HSH_DATW+32(a0)
+	dmfc2	t1, 0x0248
+	sd	t2, OCTEON_CP2_HSH_DATW+40(a0)
+	dmfc2	t2, 0x0249
+	sd	t3, OCTEON_CP2_HSH_DATW+48(a0)
+	dmfc2	t3, 0x024A
+	sd	t0, OCTEON_CP2_HSH_DATW+56(a0)
+	dmfc2	t0, 0x024B
+	sd	t1, OCTEON_CP2_HSH_DATW+64(a0)
+	dmfc2	t1, 0x024C
+	sd	t2, OCTEON_CP2_HSH_DATW+72(a0)
+	dmfc2	t2, 0x024D
+	sd	t3, OCTEON_CP2_HSH_DATW+80(a0)
+	dmfc2	t3, 0x024E
+	sd	t0, OCTEON_CP2_HSH_DATW+88(a0)
+	dmfc2	t0, 0x0250
+	sd	t1, OCTEON_CP2_HSH_DATW+96(a0)
+	dmfc2	t1, 0x0251
+	sd	t2, OCTEON_CP2_HSH_DATW+104(a0)
+	dmfc2	t2, 0x0252
+	sd	t3, OCTEON_CP2_HSH_DATW+112(a0)
+	dmfc2	t3, 0x0253
+	sd	t0, OCTEON_CP2_HSH_IVW(a0)
+	dmfc2	t0, 0x0254
+	sd	t1, OCTEON_CP2_HSH_IVW+8(a0)
+	dmfc2	t1, 0x0255
+	sd	t2, OCTEON_CP2_HSH_IVW+16(a0)
+	dmfc2	t2, 0x0256
+	sd	t3, OCTEON_CP2_HSH_IVW+24(a0)
+	dmfc2	t3, 0x0257
+	sd	t0, OCTEON_CP2_HSH_IVW+32(a0)
+	dmfc2	t0, 0x0258
+	sd	t1, OCTEON_CP2_HSH_IVW+40(a0)
+	dmfc2	t1, 0x0259
+	sd	t2, OCTEON_CP2_HSH_IVW+48(a0)
+	dmfc2	t2, 0x025E
+	sd	t3, OCTEON_CP2_HSH_IVW+56(a0)
+	dmfc2	t3, 0x025A
+	sd	t0, OCTEON_CP2_GFM_MULT(a0)
+	dmfc2	t0, 0x025B
+	sd	t1, OCTEON_CP2_GFM_MULT+8(a0)
+	sd	t2, OCTEON_CP2_GFM_POLY(a0)
+	sd	t3, OCTEON_CP2_GFM_RESULT(a0)
+	bltz	v1, 4f
+	 sd	t0, OCTEON_CP2_GFM_RESULT+8(a0)
+	/* OCTEON III things*/
+	dmfc2	t0, 0x024F
+	dmfc2	t1, 0x0050
+	sd	t0, OCTEON_CP2_SHA3(a0)
+	sd	t1, OCTEON_CP2_SHA3+8(a0)
+4:
+	jr	ra
+	 nop
+
+2:	/* pass 1 special stuff when !CvmCtl[NOCRYPTO] */
+	dmfc2	t3, 0x0040
+	dmfc2	t0, 0x0041
+	dmfc2	t1, 0x0042
+	dmfc2	t2, 0x0043
+	sd	t3, OCTEON_CP2_HSH_DATW(a0)
+	dmfc2	t3, 0x0044
+	sd	t0, OCTEON_CP2_HSH_DATW+8(a0)
+	dmfc2	t0, 0x0045
+	sd	t1, OCTEON_CP2_HSH_DATW+16(a0)
+	dmfc2	t1, 0x0046
+	sd	t2, OCTEON_CP2_HSH_DATW+24(a0)
+	dmfc2	t2, 0x0048
+	sd	t3, OCTEON_CP2_HSH_DATW+32(a0)
+	dmfc2	t3, 0x0049
+	sd	t0, OCTEON_CP2_HSH_DATW+40(a0)
+	dmfc2	t0, 0x004A
+	sd	t1, OCTEON_CP2_HSH_DATW+48(a0)
+	sd	t2, OCTEON_CP2_HSH_IVW(a0)
+	sd	t3, OCTEON_CP2_HSH_IVW+8(a0)
+	sd	t0, OCTEON_CP2_HSH_IVW+16(a0)
+
+3:	/* pass 1 or CvmCtl[NOCRYPTO] set */
+	jr	ra
+	 nop
+	END(octeon_cop2_save)
+	.set pop
+
+/*
+ * void octeon_cop2_restore(struct octeon_cop2_state *a0)
+ */
+	.align	7
+	.set push
+	.set noreorder
+	LEAF(octeon_cop2_restore)
+	/* First cache line was prefetched before the call */
+	pref	4,  128(a0)
+	dmfc0	t9, $9,7	/* CvmCtl register. */
+
+	pref	4,  256(a0)
+	ld	t0, OCTEON_CP2_CRC_IV(a0)
+	pref	4,  384(a0)
+	ld	t1, OCTEON_CP2_CRC_LENGTH(a0)
+	ld	t2, OCTEON_CP2_CRC_POLY(a0)
+
+	/* Restore the COP2 CRC state */
+	dmtc2	t0, 0x0201
+	dmtc2	t1, 0x1202
+	bbit1	t9, 28, 2f	/* Skip LLM if CvmCtl[NODFA_CP2] is set */
+	 dmtc2	t2, 0x4200
+
+	/* Restore the LLM state */
+	ld	t0, OCTEON_CP2_LLM_DAT(a0)
+	ld	t1, OCTEON_CP2_LLM_DAT+8(a0)
+	dmtc2	t0, 0x0402
+	dmtc2	t1, 0x040A
+
+2:
+	bbit1	t9, 26, done_restore	/* done if CvmCtl[NOCRYPTO] set */
+	 nop
+
+	/* Restore the COP2 crypto state common to pass 1 and pass 2 */
+	ld	t0, OCTEON_CP2_3DES_IV(a0)
+	ld	t1, OCTEON_CP2_3DES_KEY(a0)
+	ld	t2, OCTEON_CP2_3DES_KEY+8(a0)
+	dmtc2	t0, 0x0084
+	ld	t0, OCTEON_CP2_3DES_KEY+16(a0)
+	dmtc2	t1, 0x0080
+	ld	t1, OCTEON_CP2_3DES_RESULT(a0)
+	dmtc2	t2, 0x0081
+	ld	t2, OCTEON_CP2_AES_INP0(a0) /* only really needed for pass 1 */
+	dmtc2	t0, 0x0082
+	ld	t0, OCTEON_CP2_AES_IV(a0)
+	dmtc2	t1, 0x0098
+	ld	t1, OCTEON_CP2_AES_IV+8(a0)
+	dmtc2	t2, 0x010A		    /* only really needed for pass 1 */
+	ld	t2, OCTEON_CP2_AES_KEY(a0)
+	dmtc2	t0, 0x0102
+	ld	t0, OCTEON_CP2_AES_KEY+8(a0)
+	dmtc2	t1, 0x0103
+	ld	t1, OCTEON_CP2_AES_KEY+16(a0)
+	dmtc2	t2, 0x0104
+	ld	t2, OCTEON_CP2_AES_KEY+24(a0)
+	dmtc2	t0, 0x0105
+	ld	t0, OCTEON_CP2_AES_KEYLEN(a0)
+	dmtc2	t1, 0x0106
+	ld	t1, OCTEON_CP2_AES_RESULT(a0)
+	dmtc2	t2, 0x0107
+	ld	t2, OCTEON_CP2_AES_RESULT+8(a0)
+	mfc0	t3, $15,0	/* Get the processor ID register */
+	dmtc2	t0, 0x0110
+	li	v0, 0x000d0000	/* This is the processor ID of Octeon Pass1 */
+	dmtc2	t1, 0x0100
+	bne	v0, t3, 3f	/* Skip the next stuff for non-pass1 */
+	 dmtc2	t2, 0x0101
+
+	/* this code is specific for pass 1 */
+	ld	t0, OCTEON_CP2_HSH_DATW(a0)
+	ld	t1, OCTEON_CP2_HSH_DATW+8(a0)
+	ld	t2, OCTEON_CP2_HSH_DATW+16(a0)
+	dmtc2	t0, 0x0040
+	ld	t0, OCTEON_CP2_HSH_DATW+24(a0)
+	dmtc2	t1, 0x0041
+	ld	t1, OCTEON_CP2_HSH_DATW+32(a0)
+	dmtc2	t2, 0x0042
+	ld	t2, OCTEON_CP2_HSH_DATW+40(a0)
+	dmtc2	t0, 0x0043
+	ld	t0, OCTEON_CP2_HSH_DATW+48(a0)
+	dmtc2	t1, 0x0044
+	ld	t1, OCTEON_CP2_HSH_IVW(a0)
+	dmtc2	t2, 0x0045
+	ld	t2, OCTEON_CP2_HSH_IVW+8(a0)
+	dmtc2	t0, 0x0046
+	ld	t0, OCTEON_CP2_HSH_IVW+16(a0)
+	dmtc2	t1, 0x0048
+	dmtc2	t2, 0x0049
+	b done_restore	 /* unconditional branch */
+	 dmtc2	t0, 0x004A
+
+3:	/* this is post-pass1 code */
+	ld	t2, OCTEON_CP2_HSH_DATW(a0)
+	ori	v0, v0, 0x9500 /* lowest OCTEON III PrId*/
+	ld	t0, OCTEON_CP2_HSH_DATW+8(a0)
+	ld	t1, OCTEON_CP2_HSH_DATW+16(a0)
+	dmtc2	t2, 0x0240
+	ld	t2, OCTEON_CP2_HSH_DATW+24(a0)
+	dmtc2	t0, 0x0241
+	ld	t0, OCTEON_CP2_HSH_DATW+32(a0)
+	dmtc2	t1, 0x0242
+	ld	t1, OCTEON_CP2_HSH_DATW+40(a0)
+	dmtc2	t2, 0x0243
+	ld	t2, OCTEON_CP2_HSH_DATW+48(a0)
+	dmtc2	t0, 0x0244
+	ld	t0, OCTEON_CP2_HSH_DATW+56(a0)
+	dmtc2	t1, 0x0245
+	ld	t1, OCTEON_CP2_HSH_DATW+64(a0)
+	dmtc2	t2, 0x0246
+	ld	t2, OCTEON_CP2_HSH_DATW+72(a0)
+	dmtc2	t0, 0x0247
+	ld	t0, OCTEON_CP2_HSH_DATW+80(a0)
+	dmtc2	t1, 0x0248
+	ld	t1, OCTEON_CP2_HSH_DATW+88(a0)
+	dmtc2	t2, 0x0249
+	ld	t2, OCTEON_CP2_HSH_DATW+96(a0)
+	dmtc2	t0, 0x024A
+	ld	t0, OCTEON_CP2_HSH_DATW+104(a0)
+	dmtc2	t1, 0x024B
+	ld	t1, OCTEON_CP2_HSH_DATW+112(a0)
+	dmtc2	t2, 0x024C
+	ld	t2, OCTEON_CP2_HSH_IVW(a0)
+	dmtc2	t0, 0x024D
+	ld	t0, OCTEON_CP2_HSH_IVW+8(a0)
+	dmtc2	t1, 0x024E
+	ld	t1, OCTEON_CP2_HSH_IVW+16(a0)
+	dmtc2	t2, 0x0250
+	ld	t2, OCTEON_CP2_HSH_IVW+24(a0)
+	dmtc2	t0, 0x0251
+	ld	t0, OCTEON_CP2_HSH_IVW+32(a0)
+	dmtc2	t1, 0x0252
+	ld	t1, OCTEON_CP2_HSH_IVW+40(a0)
+	dmtc2	t2, 0x0253
+	ld	t2, OCTEON_CP2_HSH_IVW+48(a0)
+	dmtc2	t0, 0x0254
+	ld	t0, OCTEON_CP2_HSH_IVW+56(a0)
+	dmtc2	t1, 0x0255
+	ld	t1, OCTEON_CP2_GFM_MULT(a0)
+	dmtc2	t2, 0x0256
+	ld	t2, OCTEON_CP2_GFM_MULT+8(a0)
+	dmtc2	t0, 0x0257
+	ld	t0, OCTEON_CP2_GFM_POLY(a0)
+	dmtc2	t1, 0x0258
+	ld	t1, OCTEON_CP2_GFM_RESULT(a0)
+	dmtc2	t2, 0x0259
+	ld	t2, OCTEON_CP2_GFM_RESULT+8(a0)
+	dmtc2	t0, 0x025E
+	subu	v0, t3, v0 /* prid - lowest OCTEON III PrId */
+	dmtc2	t1, 0x025A
+	bltz	v0, done_restore
+	 dmtc2	t2, 0x025B
+	/* OCTEON III things*/
+	ld	t0, OCTEON_CP2_SHA3(a0)
+	ld	t1, OCTEON_CP2_SHA3+8(a0)
+	dmtc2	t0, 0x0051
+	dmtc2	t1, 0x0050
+done_restore:
+	jr	ra
+	 nop
+	END(octeon_cop2_restore)
+	.set pop
+
+/*
+ * void octeon_mult_save()
+ * sp is assumed to point to a struct pt_regs
+ *
+ * NOTE: This is called in SAVE_TEMP in stackframe.h. It can
+ *       safely modify v1,k0, k1,$10-$15, and $24.  It will
+ *	 be overwritten with a processor specific version of the code.
+ */
+	.p2align 7
+	.set push
+	.set noreorder
+	LEAF(octeon_mult_save)
+	jr	ra
+	 nop
+	.space 30 * 4, 0
+octeon_mult_save_end:
+	EXPORT(octeon_mult_save_end)
+	END(octeon_mult_save)
+
+	LEAF(octeon_mult_save2)
+	/* Save the multiplier state OCTEON II and earlier*/
+	v3mulu	k0, $0, $0
+	v3mulu	k1, $0, $0
+	sd	k0, PT_MTP(sp)	      /* PT_MTP	   has P0 */
+	v3mulu	k0, $0, $0
+	sd	k1, PT_MTP+8(sp)      /* PT_MTP+8  has P1 */
+	ori	k1, $0, 1
+	v3mulu	k1, k1, $0
+	sd	k0, PT_MTP+16(sp)     /* PT_MTP+16 has P2 */
+	v3mulu	k0, $0, $0
+	sd	k1, PT_MPL(sp)	      /* PT_MPL	   has MPL0 */
+	v3mulu	k1, $0, $0
+	sd	k0, PT_MPL+8(sp)      /* PT_MPL+8  has MPL1 */
+	jr	ra
+	 sd	k1, PT_MPL+16(sp)     /* PT_MPL+16 has MPL2 */
+octeon_mult_save2_end:
+	EXPORT(octeon_mult_save2_end)
+	END(octeon_mult_save2)
+
+	LEAF(octeon_mult_save3)
+	/* Save the multiplier state OCTEON III */
+	v3mulu	$10, $0, $0		/* read P0 */
+	v3mulu	$11, $0, $0		/* read P1 */
+	v3mulu	$12, $0, $0		/* read P2 */
+	sd	$10, PT_MTP+(0*8)(sp)	/* store P0 */
+	v3mulu	$10, $0, $0		/* read P3 */
+	sd	$11, PT_MTP+(1*8)(sp)	/*  store P1 */
+	v3mulu	$11, $0, $0		/* read P4 */
+	sd	$12, PT_MTP+(2*8)(sp)	/* store P2 */
+	ori	$13, $0, 1
+	v3mulu	$12, $0, $0		/* read P5 */
+	sd	$10, PT_MTP+(3*8)(sp)	/* store P3 */
+	v3mulu	$13, $13, $0		/* P4-P0 = MPL5-MPL1, $13 = MPL0 */
+	sd	$11, PT_MTP+(4*8)(sp)	/* store P4 */
+	v3mulu	$10, $0, $0		/* read MPL1 */
+	sd	$12, PT_MTP+(5*8)(sp)	/* store P5 */
+	v3mulu	$11, $0, $0		/* read MPL2 */
+	sd	$13, PT_MPL+(0*8)(sp)	/* store MPL0 */
+	v3mulu	$12, $0, $0		/* read MPL3 */
+	sd	$10, PT_MPL+(1*8)(sp)	/* store MPL1 */
+	v3mulu	$10, $0, $0		/* read MPL4 */
+	sd	$11, PT_MPL+(2*8)(sp)	/* store MPL2 */
+	v3mulu	$11, $0, $0		/* read MPL5 */
+	sd	$12, PT_MPL+(3*8)(sp)	/* store MPL3 */
+	sd	$10, PT_MPL+(4*8)(sp)	/* store MPL4 */
+	jr	ra
+	 sd	$11, PT_MPL+(5*8)(sp)	/* store MPL5 */
+octeon_mult_save3_end:
+	EXPORT(octeon_mult_save3_end)
+	END(octeon_mult_save3)
+	.set pop
+
+/*
+ * void octeon_mult_restore()
+ * sp is assumed to point to a struct pt_regs
+ *
+ * NOTE: This is called in RESTORE_TEMP in stackframe.h.
+ */
+	.p2align 7
+	.set push
+	.set noreorder
+	LEAF(octeon_mult_restore)
+	jr	ra
+	 nop
+	.space 30 * 4, 0
+octeon_mult_restore_end:
+	EXPORT(octeon_mult_restore_end)
+	END(octeon_mult_restore)
+
+	LEAF(octeon_mult_restore2)
+	ld	v0, PT_MPL(sp)        	/* MPL0 */
+	ld	v1, PT_MPL+8(sp)      	/* MPL1 */
+	ld	k0, PT_MPL+16(sp)     	/* MPL2 */
+	/* Restore the multiplier state */
+	ld	k1, PT_MTP+16(sp)     	/* P2 */
+	mtm0	v0			/* MPL0 */
+	ld	v0, PT_MTP+8(sp)	/* P1 */
+	mtm1	v1			/* MPL1 */
+	ld	v1, PT_MTP(sp)   	/* P0 */
+	mtm2	k0			/* MPL2 */
+	mtp2	k1			/* P2 */
+	mtp1	v0			/* P1 */
+	jr	ra
+	 mtp0	v1			/* P0 */
+octeon_mult_restore2_end:
+	EXPORT(octeon_mult_restore2_end)
+	END(octeon_mult_restore2)
+
+	LEAF(octeon_mult_restore3)
+	ld	$12, PT_MPL+(0*8)(sp)	/* read MPL0 */
+	ld	$13, PT_MPL+(3*8)(sp)	/* read MPL3 */
+	ld	$10, PT_MPL+(1*8)(sp)	/* read MPL1 */
+	ld	$11, PT_MPL+(4*8)(sp)	/* read MPL4 */
+	.word	0x718d0008
+	/* mtm0	$12, $13		   restore MPL0 and MPL3 */
+	ld	$12, PT_MPL+(2*8)(sp)	/* read MPL2 */
+	.word	0x714b000c
+	/* mtm1	$10, $11		   restore MPL1 and MPL4 */
+	ld	$13, PT_MPL+(5*8)(sp)	/* read MPL5 */
+	ld	$10, PT_MTP+(0*8)(sp)	/* read P0 */
+	ld	$11, PT_MTP+(3*8)(sp)	/* read P3 */
+	.word	0x718d000d
+	/* mtm2	$12, $13		   restore MPL2 and MPL5 */
+	ld	$12, PT_MTP+(1*8)(sp)	/* read P1 */
+	.word	0x714b0009
+	/* mtp0	$10, $11		   restore P0 and P3 */
+	ld	$13, PT_MTP+(4*8)(sp)	/* read P4 */
+	ld	$10, PT_MTP+(2*8)(sp)	/* read P2 */
+	ld	$11, PT_MTP+(5*8)(sp)	/* read P5 */
+	.word	0x718d000a
+	/* mtp1	$12, $13		   restore P1 and P4 */
+	jr	ra
+	.word	0x714b000b
+	/* mtp2	$10, $11		   restore P2 and P5 */
+
+octeon_mult_restore3_end:
+	EXPORT(octeon_mult_restore3_end)
+	END(octeon_mult_restore3)
+	.set pop
diff --git a/arch/mips/kernel/perf_event.c b/arch/mips/kernel/perf_event.c
new file mode 100644
index 000000000..5d7a9c039
--- /dev/null
+++ b/arch/mips/kernel/perf_event.c
@@ -0,0 +1,67 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Linux performance counter support for MIPS.
+ *
+ * Copyright (C) 2010 MIPS Technologies, Inc.
+ * Author: Deng-Cheng Zhu
+ *
+ * This code is based on the implementation for ARM, which is in turn
+ * based on the sparc64 perf event code and the x86 code. Performance
+ * counter access is based on the MIPS Oprofile code. And the callchain
+ * support references the code of MIPS stacktrace.c.
+ */
+
+#include <linux/perf_event.h>
+#include <linux/sched/task_stack.h>
+
+#include <asm/stacktrace.h>
+
+/* Callchain handling code. */
+
+/*
+ * Leave userspace callchain empty for now. When we find a way to trace
+ * the user stack callchains, we will add it here.
+ */
+
+static void save_raw_perf_callchain(struct perf_callchain_entry_ctx *entry,
+				    unsigned long reg29)
+{
+	unsigned long *sp = (unsigned long *)reg29;
+	unsigned long addr;
+
+	while (!kstack_end(sp)) {
+		addr = *sp++;
+		if (__kernel_text_address(addr)) {
+			perf_callchain_store(entry, addr);
+			if (entry->nr >= entry->max_stack)
+				break;
+		}
+	}
+}
+
+void perf_callchain_kernel(struct perf_callchain_entry_ctx *entry,
+			   struct pt_regs *regs)
+{
+	unsigned long sp = regs->regs[29];
+#ifdef CONFIG_KALLSYMS
+	unsigned long ra = regs->regs[31];
+	unsigned long pc = regs->cp0_epc;
+
+	if (raw_show_trace || !__kernel_text_address(pc)) {
+		unsigned long stack_page =
+			(unsigned long)task_stack_page(current);
+		if (stack_page && sp >= stack_page &&
+		    sp <= stack_page + THREAD_SIZE - 32)
+			save_raw_perf_callchain(entry, sp);
+		return;
+	}
+	do {
+		perf_callchain_store(entry, pc);
+		if (entry->nr >= entry->max_stack)
+			break;
+		pc = unwind_stack(current, &sp, pc, &ra);
+	} while (pc);
+#else
+	save_raw_perf_callchain(entry, sp);
+#endif
+}
diff --git a/arch/mips/kernel/perf_event_mipsxx.c b/arch/mips/kernel/perf_event_mipsxx.c
new file mode 100644
index 000000000..011eb6bbf
--- /dev/null
+++ b/arch/mips/kernel/perf_event_mipsxx.c
@@ -0,0 +1,2138 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Linux performance counter support for MIPS.
+ *
+ * Copyright (C) 2010 MIPS Technologies, Inc.
+ * Copyright (C) 2011 Cavium Networks, Inc.
+ * Author: Deng-Cheng Zhu
+ *
+ * This code is based on the implementation for ARM, which is in turn
+ * based on the sparc64 perf event code and the x86 code. Performance
+ * counter access is based on the MIPS Oprofile code. And the callchain
+ * support references the code of MIPS stacktrace.c.
+ */
+
+#include <linux/cpumask.h>
+#include <linux/interrupt.h>
+#include <linux/smp.h>
+#include <linux/kernel.h>
+#include <linux/perf_event.h>
+#include <linux/uaccess.h>
+
+#include <asm/irq.h>
+#include <asm/irq_regs.h>
+#include <asm/stacktrace.h>
+#include <asm/time.h> /* For perf_irq */
+
+#define MIPS_MAX_HWEVENTS 4
+#define MIPS_TCS_PER_COUNTER 2
+#define MIPS_CPUID_TO_COUNTER_MASK (MIPS_TCS_PER_COUNTER - 1)
+
+struct cpu_hw_events {
+	/* Array of events on this cpu. */
+	struct perf_event	*events[MIPS_MAX_HWEVENTS];
+
+	/*
+	 * Set the bit (indexed by the counter number) when the counter
+	 * is used for an event.
+	 */
+	unsigned long		used_mask[BITS_TO_LONGS(MIPS_MAX_HWEVENTS)];
+
+	/*
+	 * Software copy of the control register for each performance counter.
+	 * MIPS CPUs vary in performance counters. They use this differently,
+	 * and even may not use it.
+	 */
+	unsigned int		saved_ctrl[MIPS_MAX_HWEVENTS];
+};
+DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
+	.saved_ctrl = {0},
+};
+
+/* The description of MIPS performance events. */
+struct mips_perf_event {
+	unsigned int event_id;
+	/*
+	 * MIPS performance counters are indexed starting from 0.
+	 * CNTR_EVEN indicates the indexes of the counters to be used are
+	 * even numbers.
+	 */
+	unsigned int cntr_mask;
+	#define CNTR_EVEN	0x55555555
+	#define CNTR_ODD	0xaaaaaaaa
+	#define CNTR_ALL	0xffffffff
+	enum {
+		T  = 0,
+		V  = 1,
+		P  = 2,
+	} range;
+};
+
+static struct mips_perf_event raw_event;
+static DEFINE_MUTEX(raw_event_mutex);
+
+#define C(x) PERF_COUNT_HW_CACHE_##x
+
+struct mips_pmu {
+	u64		max_period;
+	u64		valid_count;
+	u64		overflow;
+	const char	*name;
+	int		irq;
+	u64		(*read_counter)(unsigned int idx);
+	void		(*write_counter)(unsigned int idx, u64 val);
+	const struct mips_perf_event *(*map_raw_event)(u64 config);
+	const struct mips_perf_event (*general_event_map)[PERF_COUNT_HW_MAX];
+	const struct mips_perf_event (*cache_event_map)
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX];
+	unsigned int	num_counters;
+};
+
+static int counter_bits;
+static struct mips_pmu mipspmu;
+
+#define M_PERFCTL_EVENT(event)		(((event) << MIPS_PERFCTRL_EVENT_S) & \
+					 MIPS_PERFCTRL_EVENT)
+#define M_PERFCTL_VPEID(vpe)		((vpe)	  << MIPS_PERFCTRL_VPEID_S)
+
+#ifdef CONFIG_CPU_BMIPS5000
+#define M_PERFCTL_MT_EN(filter)		0
+#else /* !CONFIG_CPU_BMIPS5000 */
+#define M_PERFCTL_MT_EN(filter)		(filter)
+#endif /* CONFIG_CPU_BMIPS5000 */
+
+#define	   M_TC_EN_ALL			M_PERFCTL_MT_EN(MIPS_PERFCTRL_MT_EN_ALL)
+#define	   M_TC_EN_VPE			M_PERFCTL_MT_EN(MIPS_PERFCTRL_MT_EN_VPE)
+#define	   M_TC_EN_TC			M_PERFCTL_MT_EN(MIPS_PERFCTRL_MT_EN_TC)
+
+#define M_PERFCTL_COUNT_EVENT_WHENEVER	(MIPS_PERFCTRL_EXL |		\
+					 MIPS_PERFCTRL_K |		\
+					 MIPS_PERFCTRL_U |		\
+					 MIPS_PERFCTRL_S |		\
+					 MIPS_PERFCTRL_IE)
+
+#ifdef CONFIG_MIPS_MT_SMP
+#define M_PERFCTL_CONFIG_MASK		0x3fff801f
+#else
+#define M_PERFCTL_CONFIG_MASK		0x1f
+#endif
+
+#define CNTR_BIT_MASK(n)	(((n) == 64) ? ~0ULL : ((1ULL<<(n))-1))
+
+#ifdef CONFIG_MIPS_PERF_SHARED_TC_COUNTERS
+static DEFINE_RWLOCK(pmuint_rwlock);
+
+#if defined(CONFIG_CPU_BMIPS5000)
+#define vpe_id()	(cpu_has_mipsmt_pertccounters ? \
+			 0 : (smp_processor_id() & MIPS_CPUID_TO_COUNTER_MASK))
+#else
+#define vpe_id()	(cpu_has_mipsmt_pertccounters ? \
+			 0 : cpu_vpe_id(&current_cpu_data))
+#endif
+
+/* Copied from op_model_mipsxx.c */
+static unsigned int vpe_shift(void)
+{
+	if (num_possible_cpus() > 1)
+		return 1;
+
+	return 0;
+}
+
+static unsigned int counters_total_to_per_cpu(unsigned int counters)
+{
+	return counters >> vpe_shift();
+}
+
+#else /* !CONFIG_MIPS_PERF_SHARED_TC_COUNTERS */
+#define vpe_id()	0
+
+#endif /* CONFIG_MIPS_PERF_SHARED_TC_COUNTERS */
+
+static void resume_local_counters(void);
+static void pause_local_counters(void);
+static irqreturn_t mipsxx_pmu_handle_irq(int, void *);
+static int mipsxx_pmu_handle_shared_irq(void);
+
+/* 0: Not Loongson-3
+ * 1: Loongson-3A1000/3B1000/3B1500
+ * 2: Loongson-3A2000/3A3000
+ * 3: Loongson-3A4000+
+ */
+
+#define LOONGSON_PMU_TYPE0 0
+#define LOONGSON_PMU_TYPE1 1
+#define LOONGSON_PMU_TYPE2 2
+#define LOONGSON_PMU_TYPE3 3
+
+static inline int get_loongson3_pmu_type(void)
+{
+	if (boot_cpu_type() != CPU_LOONGSON64)
+		return LOONGSON_PMU_TYPE0;
+	if ((boot_cpu_data.processor_id & PRID_COMP_MASK) == PRID_COMP_LEGACY)
+		return LOONGSON_PMU_TYPE1;
+	if ((boot_cpu_data.processor_id & PRID_IMP_MASK) == PRID_IMP_LOONGSON_64C)
+		return LOONGSON_PMU_TYPE2;
+	if ((boot_cpu_data.processor_id & PRID_IMP_MASK) == PRID_IMP_LOONGSON_64G)
+		return LOONGSON_PMU_TYPE3;
+
+	return LOONGSON_PMU_TYPE0;
+}
+
+static unsigned int mipsxx_pmu_swizzle_perf_idx(unsigned int idx)
+{
+	if (vpe_id() == 1)
+		idx = (idx + 2) & 3;
+	return idx;
+}
+
+static u64 mipsxx_pmu_read_counter(unsigned int idx)
+{
+	idx = mipsxx_pmu_swizzle_perf_idx(idx);
+
+	switch (idx) {
+	case 0:
+		/*
+		 * The counters are unsigned, we must cast to truncate
+		 * off the high bits.
+		 */
+		return (u32)read_c0_perfcntr0();
+	case 1:
+		return (u32)read_c0_perfcntr1();
+	case 2:
+		return (u32)read_c0_perfcntr2();
+	case 3:
+		return (u32)read_c0_perfcntr3();
+	default:
+		WARN_ONCE(1, "Invalid performance counter number (%d)\n", idx);
+		return 0;
+	}
+}
+
+static u64 mipsxx_pmu_read_counter_64(unsigned int idx)
+{
+	u64 mask = CNTR_BIT_MASK(counter_bits);
+	idx = mipsxx_pmu_swizzle_perf_idx(idx);
+
+	switch (idx) {
+	case 0:
+		return read_c0_perfcntr0_64() & mask;
+	case 1:
+		return read_c0_perfcntr1_64() & mask;
+	case 2:
+		return read_c0_perfcntr2_64() & mask;
+	case 3:
+		return read_c0_perfcntr3_64() & mask;
+	default:
+		WARN_ONCE(1, "Invalid performance counter number (%d)\n", idx);
+		return 0;
+	}
+}
+
+static void mipsxx_pmu_write_counter(unsigned int idx, u64 val)
+{
+	idx = mipsxx_pmu_swizzle_perf_idx(idx);
+
+	switch (idx) {
+	case 0:
+		write_c0_perfcntr0(val);
+		return;
+	case 1:
+		write_c0_perfcntr1(val);
+		return;
+	case 2:
+		write_c0_perfcntr2(val);
+		return;
+	case 3:
+		write_c0_perfcntr3(val);
+		return;
+	}
+}
+
+static void mipsxx_pmu_write_counter_64(unsigned int idx, u64 val)
+{
+	val &= CNTR_BIT_MASK(counter_bits);
+	idx = mipsxx_pmu_swizzle_perf_idx(idx);
+
+	switch (idx) {
+	case 0:
+		write_c0_perfcntr0_64(val);
+		return;
+	case 1:
+		write_c0_perfcntr1_64(val);
+		return;
+	case 2:
+		write_c0_perfcntr2_64(val);
+		return;
+	case 3:
+		write_c0_perfcntr3_64(val);
+		return;
+	}
+}
+
+static unsigned int mipsxx_pmu_read_control(unsigned int idx)
+{
+	idx = mipsxx_pmu_swizzle_perf_idx(idx);
+
+	switch (idx) {
+	case 0:
+		return read_c0_perfctrl0();
+	case 1:
+		return read_c0_perfctrl1();
+	case 2:
+		return read_c0_perfctrl2();
+	case 3:
+		return read_c0_perfctrl3();
+	default:
+		WARN_ONCE(1, "Invalid performance counter number (%d)\n", idx);
+		return 0;
+	}
+}
+
+static void mipsxx_pmu_write_control(unsigned int idx, unsigned int val)
+{
+	idx = mipsxx_pmu_swizzle_perf_idx(idx);
+
+	switch (idx) {
+	case 0:
+		write_c0_perfctrl0(val);
+		return;
+	case 1:
+		write_c0_perfctrl1(val);
+		return;
+	case 2:
+		write_c0_perfctrl2(val);
+		return;
+	case 3:
+		write_c0_perfctrl3(val);
+		return;
+	}
+}
+
+static int mipsxx_pmu_alloc_counter(struct cpu_hw_events *cpuc,
+				    struct hw_perf_event *hwc)
+{
+	int i;
+	unsigned long cntr_mask;
+
+	/*
+	 * We only need to care the counter mask. The range has been
+	 * checked definitely.
+	 */
+	if (get_loongson3_pmu_type() == LOONGSON_PMU_TYPE2)
+		cntr_mask = (hwc->event_base >> 10) & 0xffff;
+	else
+		cntr_mask = (hwc->event_base >> 8) & 0xffff;
+
+	for (i = mipspmu.num_counters - 1; i >= 0; i--) {
+		/*
+		 * Note that some MIPS perf events can be counted by both
+		 * even and odd counters, wheresas many other are only by
+		 * even _or_ odd counters. This introduces an issue that
+		 * when the former kind of event takes the counter the
+		 * latter kind of event wants to use, then the "counter
+		 * allocation" for the latter event will fail. In fact if
+		 * they can be dynamically swapped, they both feel happy.
+		 * But here we leave this issue alone for now.
+		 */
+		if (test_bit(i, &cntr_mask) &&
+			!test_and_set_bit(i, cpuc->used_mask))
+			return i;
+	}
+
+	return -EAGAIN;
+}
+
+static void mipsxx_pmu_enable_event(struct hw_perf_event *evt, int idx)
+{
+	struct perf_event *event = container_of(evt, struct perf_event, hw);
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	unsigned int range = evt->event_base >> 24;
+
+	WARN_ON(idx < 0 || idx >= mipspmu.num_counters);
+
+	if (get_loongson3_pmu_type() == LOONGSON_PMU_TYPE2)
+		cpuc->saved_ctrl[idx] = M_PERFCTL_EVENT(evt->event_base & 0x3ff) |
+			(evt->config_base & M_PERFCTL_CONFIG_MASK) |
+			/* Make sure interrupt enabled. */
+			MIPS_PERFCTRL_IE;
+	else
+		cpuc->saved_ctrl[idx] = M_PERFCTL_EVENT(evt->event_base & 0xff) |
+			(evt->config_base & M_PERFCTL_CONFIG_MASK) |
+			/* Make sure interrupt enabled. */
+			MIPS_PERFCTRL_IE;
+
+	if (IS_ENABLED(CONFIG_CPU_BMIPS5000)) {
+		/* enable the counter for the calling thread */
+		cpuc->saved_ctrl[idx] |=
+			(1 << (12 + vpe_id())) | BRCM_PERFCTRL_TC;
+	} else if (IS_ENABLED(CONFIG_MIPS_MT_SMP) && range > V) {
+		/* The counter is processor wide. Set it up to count all TCs. */
+		pr_debug("Enabling perf counter for all TCs\n");
+		cpuc->saved_ctrl[idx] |= M_TC_EN_ALL;
+	} else {
+		unsigned int cpu, ctrl;
+
+		/*
+		 * Set up the counter for a particular CPU when event->cpu is
+		 * a valid CPU number. Otherwise set up the counter for the CPU
+		 * scheduling this thread.
+		 */
+		cpu = (event->cpu >= 0) ? event->cpu : smp_processor_id();
+
+		ctrl = M_PERFCTL_VPEID(cpu_vpe_id(&cpu_data[cpu]));
+		ctrl |= M_TC_EN_VPE;
+		cpuc->saved_ctrl[idx] |= ctrl;
+		pr_debug("Enabling perf counter for CPU%d\n", cpu);
+	}
+	/*
+	 * We do not actually let the counter run. Leave it until start().
+	 */
+}
+
+static void mipsxx_pmu_disable_event(int idx)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	unsigned long flags;
+
+	WARN_ON(idx < 0 || idx >= mipspmu.num_counters);
+
+	local_irq_save(flags);
+	cpuc->saved_ctrl[idx] = mipsxx_pmu_read_control(idx) &
+		~M_PERFCTL_COUNT_EVENT_WHENEVER;
+	mipsxx_pmu_write_control(idx, cpuc->saved_ctrl[idx]);
+	local_irq_restore(flags);
+}
+
+static int mipspmu_event_set_period(struct perf_event *event,
+				    struct hw_perf_event *hwc,
+				    int idx)
+{
+	u64 left = local64_read(&hwc->period_left);
+	u64 period = hwc->sample_period;
+	int ret = 0;
+
+	if (unlikely((left + period) & (1ULL << 63))) {
+		/* left underflowed by more than period. */
+		left = period;
+		local64_set(&hwc->period_left, left);
+		hwc->last_period = period;
+		ret = 1;
+	} else	if (unlikely((left + period) <= period)) {
+		/* left underflowed by less than period. */
+		left += period;
+		local64_set(&hwc->period_left, left);
+		hwc->last_period = period;
+		ret = 1;
+	}
+
+	if (left > mipspmu.max_period) {
+		left = mipspmu.max_period;
+		local64_set(&hwc->period_left, left);
+	}
+
+	local64_set(&hwc->prev_count, mipspmu.overflow - left);
+
+	if (get_loongson3_pmu_type() == LOONGSON_PMU_TYPE2)
+		mipsxx_pmu_write_control(idx,
+				M_PERFCTL_EVENT(hwc->event_base & 0x3ff));
+
+	mipspmu.write_counter(idx, mipspmu.overflow - left);
+
+	perf_event_update_userpage(event);
+
+	return ret;
+}
+
+static void mipspmu_event_update(struct perf_event *event,
+				 struct hw_perf_event *hwc,
+				 int idx)
+{
+	u64 prev_raw_count, new_raw_count;
+	u64 delta;
+
+again:
+	prev_raw_count = local64_read(&hwc->prev_count);
+	new_raw_count = mipspmu.read_counter(idx);
+
+	if (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
+				new_raw_count) != prev_raw_count)
+		goto again;
+
+	delta = new_raw_count - prev_raw_count;
+
+	local64_add(delta, &event->count);
+	local64_sub(delta, &hwc->period_left);
+}
+
+static void mipspmu_start(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE));
+
+	hwc->state = 0;
+
+	/* Set the period for the event. */
+	mipspmu_event_set_period(event, hwc, hwc->idx);
+
+	/* Enable the event. */
+	mipsxx_pmu_enable_event(hwc, hwc->idx);
+}
+
+static void mipspmu_stop(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (!(hwc->state & PERF_HES_STOPPED)) {
+		/* We are working on a local event. */
+		mipsxx_pmu_disable_event(hwc->idx);
+		barrier();
+		mipspmu_event_update(event, hwc, hwc->idx);
+		hwc->state |= PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	}
+}
+
+static int mipspmu_add(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	int idx;
+	int err = 0;
+
+	perf_pmu_disable(event->pmu);
+
+	/* To look for a free counter for this event. */
+	idx = mipsxx_pmu_alloc_counter(cpuc, hwc);
+	if (idx < 0) {
+		err = idx;
+		goto out;
+	}
+
+	/*
+	 * If there is an event in the counter we are going to use then
+	 * make sure it is disabled.
+	 */
+	event->hw.idx = idx;
+	mipsxx_pmu_disable_event(idx);
+	cpuc->events[idx] = event;
+
+	hwc->state = PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	if (flags & PERF_EF_START)
+		mipspmu_start(event, PERF_EF_RELOAD);
+
+	/* Propagate our changes to the userspace mapping. */
+	perf_event_update_userpage(event);
+
+out:
+	perf_pmu_enable(event->pmu);
+	return err;
+}
+
+static void mipspmu_del(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	int idx = hwc->idx;
+
+	WARN_ON(idx < 0 || idx >= mipspmu.num_counters);
+
+	mipspmu_stop(event, PERF_EF_UPDATE);
+	cpuc->events[idx] = NULL;
+	clear_bit(idx, cpuc->used_mask);
+
+	perf_event_update_userpage(event);
+}
+
+static void mipspmu_read(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	/* Don't read disabled counters! */
+	if (hwc->idx < 0)
+		return;
+
+	mipspmu_event_update(event, hwc, hwc->idx);
+}
+
+static void mipspmu_enable(struct pmu *pmu)
+{
+#ifdef CONFIG_MIPS_PERF_SHARED_TC_COUNTERS
+	write_unlock(&pmuint_rwlock);
+#endif
+	resume_local_counters();
+}
+
+/*
+ * MIPS performance counters can be per-TC. The control registers can
+ * not be directly accessed across CPUs. Hence if we want to do global
+ * control, we need cross CPU calls. on_each_cpu() can help us, but we
+ * can not make sure this function is called with interrupts enabled. So
+ * here we pause local counters and then grab a rwlock and leave the
+ * counters on other CPUs alone. If any counter interrupt raises while
+ * we own the write lock, simply pause local counters on that CPU and
+ * spin in the handler. Also we know we won't be switched to another
+ * CPU after pausing local counters and before grabbing the lock.
+ */
+static void mipspmu_disable(struct pmu *pmu)
+{
+	pause_local_counters();
+#ifdef CONFIG_MIPS_PERF_SHARED_TC_COUNTERS
+	write_lock(&pmuint_rwlock);
+#endif
+}
+
+static atomic_t active_events = ATOMIC_INIT(0);
+static DEFINE_MUTEX(pmu_reserve_mutex);
+static int (*save_perf_irq)(void);
+
+static int mipspmu_get_irq(void)
+{
+	int err;
+
+	if (mipspmu.irq >= 0) {
+		/* Request my own irq handler. */
+		err = request_irq(mipspmu.irq, mipsxx_pmu_handle_irq,
+				  IRQF_PERCPU | IRQF_NOBALANCING |
+				  IRQF_NO_THREAD | IRQF_NO_SUSPEND |
+				  IRQF_SHARED,
+				  "mips_perf_pmu", &mipspmu);
+		if (err) {
+			pr_warn("Unable to request IRQ%d for MIPS performance counters!\n",
+				mipspmu.irq);
+		}
+	} else if (cp0_perfcount_irq < 0) {
+		/*
+		 * We are sharing the irq number with the timer interrupt.
+		 */
+		save_perf_irq = perf_irq;
+		perf_irq = mipsxx_pmu_handle_shared_irq;
+		err = 0;
+	} else {
+		pr_warn("The platform hasn't properly defined its interrupt controller\n");
+		err = -ENOENT;
+	}
+
+	return err;
+}
+
+static void mipspmu_free_irq(void)
+{
+	if (mipspmu.irq >= 0)
+		free_irq(mipspmu.irq, &mipspmu);
+	else if (cp0_perfcount_irq < 0)
+		perf_irq = save_perf_irq;
+}
+
+/*
+ * mipsxx/rm9000/loongson2 have different performance counters, they have
+ * specific low-level init routines.
+ */
+static void reset_counters(void *arg);
+static int __hw_perf_event_init(struct perf_event *event);
+
+static void hw_perf_event_destroy(struct perf_event *event)
+{
+	if (atomic_dec_and_mutex_lock(&active_events,
+				&pmu_reserve_mutex)) {
+		/*
+		 * We must not call the destroy function with interrupts
+		 * disabled.
+		 */
+		on_each_cpu(reset_counters,
+			(void *)(long)mipspmu.num_counters, 1);
+		mipspmu_free_irq();
+		mutex_unlock(&pmu_reserve_mutex);
+	}
+}
+
+static int mipspmu_event_init(struct perf_event *event)
+{
+	int err = 0;
+
+	/* does not support taken branch sampling */
+	if (has_branch_stack(event))
+		return -EOPNOTSUPP;
+
+	switch (event->attr.type) {
+	case PERF_TYPE_RAW:
+	case PERF_TYPE_HARDWARE:
+	case PERF_TYPE_HW_CACHE:
+		break;
+
+	default:
+		return -ENOENT;
+	}
+
+	if (event->cpu >= 0 && !cpu_online(event->cpu))
+		return -ENODEV;
+
+	if (!atomic_inc_not_zero(&active_events)) {
+		mutex_lock(&pmu_reserve_mutex);
+		if (atomic_read(&active_events) == 0)
+			err = mipspmu_get_irq();
+
+		if (!err)
+			atomic_inc(&active_events);
+		mutex_unlock(&pmu_reserve_mutex);
+	}
+
+	if (err)
+		return err;
+
+	return __hw_perf_event_init(event);
+}
+
+static struct pmu pmu = {
+	.pmu_enable	= mipspmu_enable,
+	.pmu_disable	= mipspmu_disable,
+	.event_init	= mipspmu_event_init,
+	.add		= mipspmu_add,
+	.del		= mipspmu_del,
+	.start		= mipspmu_start,
+	.stop		= mipspmu_stop,
+	.read		= mipspmu_read,
+};
+
+static unsigned int mipspmu_perf_event_encode(const struct mips_perf_event *pev)
+{
+/*
+ * Top 8 bits for range, next 16 bits for cntr_mask, lowest 8 bits for
+ * event_id.
+ */
+#ifdef CONFIG_MIPS_MT_SMP
+	if (num_possible_cpus() > 1)
+		return ((unsigned int)pev->range << 24) |
+			(pev->cntr_mask & 0xffff00) |
+			(pev->event_id & 0xff);
+	else
+#endif /* CONFIG_MIPS_MT_SMP */
+	{
+		if (get_loongson3_pmu_type() == LOONGSON_PMU_TYPE2)
+			return (pev->cntr_mask & 0xfffc00) |
+				(pev->event_id & 0x3ff);
+		else
+			return (pev->cntr_mask & 0xffff00) |
+				(pev->event_id & 0xff);
+	}
+}
+
+static const struct mips_perf_event *mipspmu_map_general_event(int idx)
+{
+
+	if ((*mipspmu.general_event_map)[idx].cntr_mask == 0)
+		return ERR_PTR(-EOPNOTSUPP);
+	return &(*mipspmu.general_event_map)[idx];
+}
+
+static const struct mips_perf_event *mipspmu_map_cache_event(u64 config)
+{
+	unsigned int cache_type, cache_op, cache_result;
+	const struct mips_perf_event *pev;
+
+	cache_type = (config >> 0) & 0xff;
+	if (cache_type >= PERF_COUNT_HW_CACHE_MAX)
+		return ERR_PTR(-EINVAL);
+
+	cache_op = (config >> 8) & 0xff;
+	if (cache_op >= PERF_COUNT_HW_CACHE_OP_MAX)
+		return ERR_PTR(-EINVAL);
+
+	cache_result = (config >> 16) & 0xff;
+	if (cache_result >= PERF_COUNT_HW_CACHE_RESULT_MAX)
+		return ERR_PTR(-EINVAL);
+
+	pev = &((*mipspmu.cache_event_map)
+					[cache_type]
+					[cache_op]
+					[cache_result]);
+
+	if (pev->cntr_mask == 0)
+		return ERR_PTR(-EOPNOTSUPP);
+
+	return pev;
+
+}
+
+static int validate_group(struct perf_event *event)
+{
+	struct perf_event *sibling, *leader = event->group_leader;
+	struct cpu_hw_events fake_cpuc;
+
+	memset(&fake_cpuc, 0, sizeof(fake_cpuc));
+
+	if (mipsxx_pmu_alloc_counter(&fake_cpuc, &leader->hw) < 0)
+		return -EINVAL;
+
+	for_each_sibling_event(sibling, leader) {
+		if (mipsxx_pmu_alloc_counter(&fake_cpuc, &sibling->hw) < 0)
+			return -EINVAL;
+	}
+
+	if (mipsxx_pmu_alloc_counter(&fake_cpuc, &event->hw) < 0)
+		return -EINVAL;
+
+	return 0;
+}
+
+/* This is needed by specific irq handlers in perf_event_*.c */
+static void handle_associated_event(struct cpu_hw_events *cpuc,
+				    int idx, struct perf_sample_data *data,
+				    struct pt_regs *regs)
+{
+	struct perf_event *event = cpuc->events[idx];
+	struct hw_perf_event *hwc = &event->hw;
+
+	mipspmu_event_update(event, hwc, idx);
+	data->period = event->hw.last_period;
+	if (!mipspmu_event_set_period(event, hwc, idx))
+		return;
+
+	if (perf_event_overflow(event, data, regs))
+		mipsxx_pmu_disable_event(idx);
+}
+
+
+static int __n_counters(void)
+{
+	if (!cpu_has_perf)
+		return 0;
+	if (!(read_c0_perfctrl0() & MIPS_PERFCTRL_M))
+		return 1;
+	if (!(read_c0_perfctrl1() & MIPS_PERFCTRL_M))
+		return 2;
+	if (!(read_c0_perfctrl2() & MIPS_PERFCTRL_M))
+		return 3;
+
+	return 4;
+}
+
+static int n_counters(void)
+{
+	int counters;
+
+	switch (current_cpu_type()) {
+	case CPU_R10000:
+		counters = 2;
+		break;
+
+	case CPU_R12000:
+	case CPU_R14000:
+	case CPU_R16000:
+		counters = 4;
+		break;
+
+	default:
+		counters = __n_counters();
+	}
+
+	return counters;
+}
+
+static void loongson3_reset_counters(void *arg)
+{
+	int counters = (int)(long)arg;
+
+	switch (counters) {
+	case 4:
+		mipsxx_pmu_write_control(3, 0);
+		mipspmu.write_counter(3, 0);
+		mipsxx_pmu_write_control(3, 127<<5);
+		mipspmu.write_counter(3, 0);
+		mipsxx_pmu_write_control(3, 191<<5);
+		mipspmu.write_counter(3, 0);
+		mipsxx_pmu_write_control(3, 255<<5);
+		mipspmu.write_counter(3, 0);
+		mipsxx_pmu_write_control(3, 319<<5);
+		mipspmu.write_counter(3, 0);
+		mipsxx_pmu_write_control(3, 383<<5);
+		mipspmu.write_counter(3, 0);
+		mipsxx_pmu_write_control(3, 575<<5);
+		mipspmu.write_counter(3, 0);
+		fallthrough;
+	case 3:
+		mipsxx_pmu_write_control(2, 0);
+		mipspmu.write_counter(2, 0);
+		mipsxx_pmu_write_control(2, 127<<5);
+		mipspmu.write_counter(2, 0);
+		mipsxx_pmu_write_control(2, 191<<5);
+		mipspmu.write_counter(2, 0);
+		mipsxx_pmu_write_control(2, 255<<5);
+		mipspmu.write_counter(2, 0);
+		mipsxx_pmu_write_control(2, 319<<5);
+		mipspmu.write_counter(2, 0);
+		mipsxx_pmu_write_control(2, 383<<5);
+		mipspmu.write_counter(2, 0);
+		mipsxx_pmu_write_control(2, 575<<5);
+		mipspmu.write_counter(2, 0);
+		fallthrough;
+	case 2:
+		mipsxx_pmu_write_control(1, 0);
+		mipspmu.write_counter(1, 0);
+		mipsxx_pmu_write_control(1, 127<<5);
+		mipspmu.write_counter(1, 0);
+		mipsxx_pmu_write_control(1, 191<<5);
+		mipspmu.write_counter(1, 0);
+		mipsxx_pmu_write_control(1, 255<<5);
+		mipspmu.write_counter(1, 0);
+		mipsxx_pmu_write_control(1, 319<<5);
+		mipspmu.write_counter(1, 0);
+		mipsxx_pmu_write_control(1, 383<<5);
+		mipspmu.write_counter(1, 0);
+		mipsxx_pmu_write_control(1, 575<<5);
+		mipspmu.write_counter(1, 0);
+		fallthrough;
+	case 1:
+		mipsxx_pmu_write_control(0, 0);
+		mipspmu.write_counter(0, 0);
+		mipsxx_pmu_write_control(0, 127<<5);
+		mipspmu.write_counter(0, 0);
+		mipsxx_pmu_write_control(0, 191<<5);
+		mipspmu.write_counter(0, 0);
+		mipsxx_pmu_write_control(0, 255<<5);
+		mipspmu.write_counter(0, 0);
+		mipsxx_pmu_write_control(0, 319<<5);
+		mipspmu.write_counter(0, 0);
+		mipsxx_pmu_write_control(0, 383<<5);
+		mipspmu.write_counter(0, 0);
+		mipsxx_pmu_write_control(0, 575<<5);
+		mipspmu.write_counter(0, 0);
+		break;
+	}
+}
+
+static void reset_counters(void *arg)
+{
+	int counters = (int)(long)arg;
+
+	if (get_loongson3_pmu_type() == LOONGSON_PMU_TYPE2) {
+		loongson3_reset_counters(arg);
+		return;
+	}
+
+	switch (counters) {
+	case 4:
+		mipsxx_pmu_write_control(3, 0);
+		mipspmu.write_counter(3, 0);
+		fallthrough;
+	case 3:
+		mipsxx_pmu_write_control(2, 0);
+		mipspmu.write_counter(2, 0);
+		fallthrough;
+	case 2:
+		mipsxx_pmu_write_control(1, 0);
+		mipspmu.write_counter(1, 0);
+		fallthrough;
+	case 1:
+		mipsxx_pmu_write_control(0, 0);
+		mipspmu.write_counter(0, 0);
+		break;
+	}
+}
+
+/* 24K/34K/1004K/interAptiv/loongson1 cores share the same event map. */
+static const struct mips_perf_event mipsxxcore_event_map
+				[PERF_COUNT_HW_MAX] = {
+	[PERF_COUNT_HW_CPU_CYCLES] = { 0x00, CNTR_EVEN | CNTR_ODD, P },
+	[PERF_COUNT_HW_INSTRUCTIONS] = { 0x01, CNTR_EVEN | CNTR_ODD, T },
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = { 0x02, CNTR_EVEN, T },
+	[PERF_COUNT_HW_BRANCH_MISSES] = { 0x02, CNTR_ODD, T },
+};
+
+/* 74K/proAptiv core has different branch event code. */
+static const struct mips_perf_event mipsxxcore_event_map2
+				[PERF_COUNT_HW_MAX] = {
+	[PERF_COUNT_HW_CPU_CYCLES] = { 0x00, CNTR_EVEN | CNTR_ODD, P },
+	[PERF_COUNT_HW_INSTRUCTIONS] = { 0x01, CNTR_EVEN | CNTR_ODD, T },
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = { 0x27, CNTR_EVEN, T },
+	[PERF_COUNT_HW_BRANCH_MISSES] = { 0x27, CNTR_ODD, T },
+};
+
+static const struct mips_perf_event i6x00_event_map[PERF_COUNT_HW_MAX] = {
+	[PERF_COUNT_HW_CPU_CYCLES]          = { 0x00, CNTR_EVEN | CNTR_ODD },
+	[PERF_COUNT_HW_INSTRUCTIONS]        = { 0x01, CNTR_EVEN | CNTR_ODD },
+	/* These only count dcache, not icache */
+	[PERF_COUNT_HW_CACHE_REFERENCES]    = { 0x45, CNTR_EVEN | CNTR_ODD },
+	[PERF_COUNT_HW_CACHE_MISSES]        = { 0x48, CNTR_EVEN | CNTR_ODD },
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = { 0x15, CNTR_EVEN | CNTR_ODD },
+	[PERF_COUNT_HW_BRANCH_MISSES]       = { 0x16, CNTR_EVEN | CNTR_ODD },
+};
+
+static const struct mips_perf_event loongson3_event_map1[PERF_COUNT_HW_MAX] = {
+	[PERF_COUNT_HW_CPU_CYCLES] = { 0x00, CNTR_EVEN },
+	[PERF_COUNT_HW_INSTRUCTIONS] = { 0x00, CNTR_ODD },
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = { 0x01, CNTR_EVEN },
+	[PERF_COUNT_HW_BRANCH_MISSES] = { 0x01, CNTR_ODD },
+};
+
+static const struct mips_perf_event loongson3_event_map2[PERF_COUNT_HW_MAX] = {
+	[PERF_COUNT_HW_CPU_CYCLES] = { 0x80, CNTR_ALL },
+	[PERF_COUNT_HW_INSTRUCTIONS] = { 0x81, CNTR_ALL },
+	[PERF_COUNT_HW_CACHE_MISSES] = { 0x18, CNTR_ALL },
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = { 0x94, CNTR_ALL },
+	[PERF_COUNT_HW_BRANCH_MISSES] = { 0x9c, CNTR_ALL },
+};
+
+static const struct mips_perf_event loongson3_event_map3[PERF_COUNT_HW_MAX] = {
+	[PERF_COUNT_HW_CPU_CYCLES] = { 0x00, CNTR_ALL },
+	[PERF_COUNT_HW_INSTRUCTIONS] = { 0x01, CNTR_ALL },
+	[PERF_COUNT_HW_CACHE_REFERENCES] = { 0x1c, CNTR_ALL },
+	[PERF_COUNT_HW_CACHE_MISSES] = { 0x1d, CNTR_ALL },
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = { 0x02, CNTR_ALL },
+	[PERF_COUNT_HW_BRANCH_MISSES] = { 0x08, CNTR_ALL },
+};
+
+static const struct mips_perf_event octeon_event_map[PERF_COUNT_HW_MAX] = {
+	[PERF_COUNT_HW_CPU_CYCLES] = { 0x01, CNTR_ALL },
+	[PERF_COUNT_HW_INSTRUCTIONS] = { 0x03, CNTR_ALL },
+	[PERF_COUNT_HW_CACHE_REFERENCES] = { 0x2b, CNTR_ALL },
+	[PERF_COUNT_HW_CACHE_MISSES] = { 0x2e, CNTR_ALL	 },
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = { 0x08, CNTR_ALL },
+	[PERF_COUNT_HW_BRANCH_MISSES] = { 0x09, CNTR_ALL },
+	[PERF_COUNT_HW_BUS_CYCLES] = { 0x25, CNTR_ALL },
+};
+
+static const struct mips_perf_event bmips5000_event_map
+				[PERF_COUNT_HW_MAX] = {
+	[PERF_COUNT_HW_CPU_CYCLES] = { 0x00, CNTR_EVEN | CNTR_ODD, T },
+	[PERF_COUNT_HW_INSTRUCTIONS] = { 0x01, CNTR_EVEN | CNTR_ODD, T },
+	[PERF_COUNT_HW_BRANCH_MISSES] = { 0x02, CNTR_ODD, T },
+};
+
+static const struct mips_perf_event xlp_event_map[PERF_COUNT_HW_MAX] = {
+	[PERF_COUNT_HW_CPU_CYCLES] = { 0x01, CNTR_ALL },
+	[PERF_COUNT_HW_INSTRUCTIONS] = { 0x18, CNTR_ALL }, /* PAPI_TOT_INS */
+	[PERF_COUNT_HW_CACHE_REFERENCES] = { 0x04, CNTR_ALL }, /* PAPI_L1_ICA */
+	[PERF_COUNT_HW_CACHE_MISSES] = { 0x07, CNTR_ALL }, /* PAPI_L1_ICM */
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = { 0x1b, CNTR_ALL }, /* PAPI_BR_CN */
+	[PERF_COUNT_HW_BRANCH_MISSES] = { 0x1c, CNTR_ALL }, /* PAPI_BR_MSP */
+};
+
+/* 24K/34K/1004K/interAptiv/loongson1 cores share the same cache event map. */
+static const struct mips_perf_event mipsxxcore_cache_map
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
+[C(L1D)] = {
+	/*
+	 * Like some other architectures (e.g. ARM), the performance
+	 * counters don't differentiate between read and write
+	 * accesses/misses, so this isn't strictly correct, but it's the
+	 * best we can do. Writes and reads get combined.
+	 */
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x0a, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x0b, CNTR_EVEN | CNTR_ODD, T },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x0a, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x0b, CNTR_EVEN | CNTR_ODD, T },
+	},
+},
+[C(L1I)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x09, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x09, CNTR_ODD, T },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x09, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x09, CNTR_ODD, T },
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)]	= { 0x14, CNTR_EVEN, T },
+		/*
+		 * Note that MIPS has only "hit" events countable for
+		 * the prefetch operation.
+		 */
+	},
+},
+[C(LL)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x15, CNTR_ODD, P },
+		[C(RESULT_MISS)]	= { 0x16, CNTR_EVEN, P },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x15, CNTR_ODD, P },
+		[C(RESULT_MISS)]	= { 0x16, CNTR_EVEN, P },
+	},
+},
+[C(DTLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x06, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x06, CNTR_ODD, T },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x06, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x06, CNTR_ODD, T },
+	},
+},
+[C(ITLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x05, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x05, CNTR_ODD, T },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x05, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x05, CNTR_ODD, T },
+	},
+},
+[C(BPU)] = {
+	/* Using the same code for *HW_BRANCH* */
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x02, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x02, CNTR_ODD, T },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x02, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x02, CNTR_ODD, T },
+	},
+},
+};
+
+/* 74K/proAptiv core has completely different cache event map. */
+static const struct mips_perf_event mipsxxcore_cache_map2
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
+[C(L1D)] = {
+	/*
+	 * Like some other architectures (e.g. ARM), the performance
+	 * counters don't differentiate between read and write
+	 * accesses/misses, so this isn't strictly correct, but it's the
+	 * best we can do. Writes and reads get combined.
+	 */
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x17, CNTR_ODD, T },
+		[C(RESULT_MISS)]	= { 0x18, CNTR_ODD, T },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x17, CNTR_ODD, T },
+		[C(RESULT_MISS)]	= { 0x18, CNTR_ODD, T },
+	},
+},
+[C(L1I)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x06, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x06, CNTR_ODD, T },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x06, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x06, CNTR_ODD, T },
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)]	= { 0x34, CNTR_EVEN, T },
+		/*
+		 * Note that MIPS has only "hit" events countable for
+		 * the prefetch operation.
+		 */
+	},
+},
+[C(LL)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x1c, CNTR_ODD, P },
+		[C(RESULT_MISS)]	= { 0x1d, CNTR_EVEN, P },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x1c, CNTR_ODD, P },
+		[C(RESULT_MISS)]	= { 0x1d, CNTR_EVEN, P },
+	},
+},
+/*
+ * 74K core does not have specific DTLB events. proAptiv core has
+ * "speculative" DTLB events which are numbered 0x63 (even/odd) and
+ * not included here. One can use raw events if really needed.
+ */
+[C(ITLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x04, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x04, CNTR_ODD, T },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x04, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x04, CNTR_ODD, T },
+	},
+},
+[C(BPU)] = {
+	/* Using the same code for *HW_BRANCH* */
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x27, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x27, CNTR_ODD, T },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x27, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 0x27, CNTR_ODD, T },
+	},
+},
+};
+
+static const struct mips_perf_event i6x00_cache_map
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
+[C(L1D)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x46, CNTR_EVEN | CNTR_ODD },
+		[C(RESULT_MISS)]	= { 0x49, CNTR_EVEN | CNTR_ODD },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x47, CNTR_EVEN | CNTR_ODD },
+		[C(RESULT_MISS)]	= { 0x4a, CNTR_EVEN | CNTR_ODD },
+	},
+},
+[C(L1I)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x84, CNTR_EVEN | CNTR_ODD },
+		[C(RESULT_MISS)]	= { 0x85, CNTR_EVEN | CNTR_ODD },
+	},
+},
+[C(DTLB)] = {
+	/* Can't distinguish read & write */
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x40, CNTR_EVEN | CNTR_ODD },
+		[C(RESULT_MISS)]	= { 0x41, CNTR_EVEN | CNTR_ODD },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x40, CNTR_EVEN | CNTR_ODD },
+		[C(RESULT_MISS)]	= { 0x41, CNTR_EVEN | CNTR_ODD },
+	},
+},
+[C(BPU)] = {
+	/* Conditional branches / mispredicted */
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x15, CNTR_EVEN | CNTR_ODD },
+		[C(RESULT_MISS)]	= { 0x16, CNTR_EVEN | CNTR_ODD },
+	},
+},
+};
+
+static const struct mips_perf_event loongson3_cache_map1
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
+[C(L1D)] = {
+	/*
+	 * Like some other architectures (e.g. ARM), the performance
+	 * counters don't differentiate between read and write
+	 * accesses/misses, so this isn't strictly correct, but it's the
+	 * best we can do. Writes and reads get combined.
+	 */
+	[C(OP_READ)] = {
+		[C(RESULT_MISS)]        = { 0x04, CNTR_ODD },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_MISS)]        = { 0x04, CNTR_ODD },
+	},
+},
+[C(L1I)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_MISS)]        = { 0x04, CNTR_EVEN },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_MISS)]        = { 0x04, CNTR_EVEN },
+	},
+},
+[C(DTLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_MISS)]        = { 0x09, CNTR_ODD },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_MISS)]        = { 0x09, CNTR_ODD },
+	},
+},
+[C(ITLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_MISS)]        = { 0x0c, CNTR_ODD },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_MISS)]        = { 0x0c, CNTR_ODD },
+	},
+},
+[C(BPU)] = {
+	/* Using the same code for *HW_BRANCH* */
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]      = { 0x01, CNTR_EVEN },
+		[C(RESULT_MISS)]        = { 0x01, CNTR_ODD },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]      = { 0x01, CNTR_EVEN },
+		[C(RESULT_MISS)]        = { 0x01, CNTR_ODD },
+	},
+},
+};
+
+static const struct mips_perf_event loongson3_cache_map2
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
+[C(L1D)] = {
+	/*
+	 * Like some other architectures (e.g. ARM), the performance
+	 * counters don't differentiate between read and write
+	 * accesses/misses, so this isn't strictly correct, but it's the
+	 * best we can do. Writes and reads get combined.
+	 */
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x156, CNTR_ALL },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x155, CNTR_ALL },
+		[C(RESULT_MISS)]        = { 0x153, CNTR_ALL },
+	},
+},
+[C(L1I)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_MISS)]	= { 0x18, CNTR_ALL },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_MISS)]        = { 0x18, CNTR_ALL },
+	},
+},
+[C(LL)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x1b6, CNTR_ALL },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x1b7, CNTR_ALL },
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)]	= { 0x1bf, CNTR_ALL },
+	},
+},
+[C(DTLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_MISS)]        = { 0x92, CNTR_ALL },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_MISS)]        = { 0x92, CNTR_ALL },
+	},
+},
+[C(ITLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_MISS)]	= { 0x1a, CNTR_ALL },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_MISS)]	= { 0x1a, CNTR_ALL },
+	},
+},
+[C(BPU)] = {
+	/* Using the same code for *HW_BRANCH* */
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]      = { 0x94, CNTR_ALL },
+		[C(RESULT_MISS)]        = { 0x9c, CNTR_ALL },
+	},
+},
+};
+
+static const struct mips_perf_event loongson3_cache_map3
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
+[C(L1D)] = {
+	/*
+	 * Like some other architectures (e.g. ARM), the performance
+	 * counters don't differentiate between read and write
+	 * accesses/misses, so this isn't strictly correct, but it's the
+	 * best we can do. Writes and reads get combined.
+	 */
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]      = { 0x1e, CNTR_ALL },
+		[C(RESULT_MISS)]        = { 0x1f, CNTR_ALL },
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)]	= { 0xaa, CNTR_ALL },
+		[C(RESULT_MISS)]	= { 0xa9, CNTR_ALL },
+	},
+},
+[C(L1I)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x1c, CNTR_ALL },
+		[C(RESULT_MISS)]	= { 0x1d, CNTR_ALL },
+	},
+},
+[C(LL)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x2e, CNTR_ALL },
+		[C(RESULT_MISS)]	= { 0x2f, CNTR_ALL },
+	},
+},
+[C(DTLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]      = { 0x14, CNTR_ALL },
+		[C(RESULT_MISS)]	= { 0x1b, CNTR_ALL },
+	},
+},
+[C(ITLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_MISS)]	= { 0x1a, CNTR_ALL },
+	},
+},
+[C(BPU)] = {
+	/* Using the same code for *HW_BRANCH* */
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]      = { 0x02, CNTR_ALL },
+		[C(RESULT_MISS)]        = { 0x08, CNTR_ALL },
+	},
+},
+};
+
+/* BMIPS5000 */
+static const struct mips_perf_event bmips5000_cache_map
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
+[C(L1D)] = {
+	/*
+	 * Like some other architectures (e.g. ARM), the performance
+	 * counters don't differentiate between read and write
+	 * accesses/misses, so this isn't strictly correct, but it's the
+	 * best we can do. Writes and reads get combined.
+	 */
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 12, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 12, CNTR_ODD, T },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 12, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 12, CNTR_ODD, T },
+	},
+},
+[C(L1I)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 10, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 10, CNTR_ODD, T },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 10, CNTR_EVEN, T },
+		[C(RESULT_MISS)]	= { 10, CNTR_ODD, T },
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)]	= { 23, CNTR_EVEN, T },
+		/*
+		 * Note that MIPS has only "hit" events countable for
+		 * the prefetch operation.
+		 */
+	},
+},
+[C(LL)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 28, CNTR_EVEN, P },
+		[C(RESULT_MISS)]	= { 28, CNTR_ODD, P },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 28, CNTR_EVEN, P },
+		[C(RESULT_MISS)]	= { 28, CNTR_ODD, P },
+	},
+},
+[C(BPU)] = {
+	/* Using the same code for *HW_BRANCH* */
+	[C(OP_READ)] = {
+		[C(RESULT_MISS)]	= { 0x02, CNTR_ODD, T },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_MISS)]	= { 0x02, CNTR_ODD, T },
+	},
+},
+};
+
+static const struct mips_perf_event octeon_cache_map
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
+[C(L1D)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x2b, CNTR_ALL },
+		[C(RESULT_MISS)]	= { 0x2e, CNTR_ALL },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x30, CNTR_ALL },
+	},
+},
+[C(L1I)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x18, CNTR_ALL },
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)]	= { 0x19, CNTR_ALL },
+	},
+},
+[C(DTLB)] = {
+	/*
+	 * Only general DTLB misses are counted use the same event for
+	 * read and write.
+	 */
+	[C(OP_READ)] = {
+		[C(RESULT_MISS)]	= { 0x35, CNTR_ALL },
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_MISS)]	= { 0x35, CNTR_ALL },
+	},
+},
+[C(ITLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_MISS)]	= { 0x37, CNTR_ALL },
+	},
+},
+};
+
+static const struct mips_perf_event xlp_cache_map
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
+[C(L1D)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x31, CNTR_ALL }, /* PAPI_L1_DCR */
+		[C(RESULT_MISS)]	= { 0x30, CNTR_ALL }, /* PAPI_L1_LDM */
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x2f, CNTR_ALL }, /* PAPI_L1_DCW */
+		[C(RESULT_MISS)]	= { 0x2e, CNTR_ALL }, /* PAPI_L1_STM */
+	},
+},
+[C(L1I)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x04, CNTR_ALL }, /* PAPI_L1_ICA */
+		[C(RESULT_MISS)]	= { 0x07, CNTR_ALL }, /* PAPI_L1_ICM */
+	},
+},
+[C(LL)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)]	= { 0x35, CNTR_ALL }, /* PAPI_L2_DCR */
+		[C(RESULT_MISS)]	= { 0x37, CNTR_ALL }, /* PAPI_L2_LDM */
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)]	= { 0x34, CNTR_ALL }, /* PAPI_L2_DCA */
+		[C(RESULT_MISS)]	= { 0x36, CNTR_ALL }, /* PAPI_L2_DCM */
+	},
+},
+[C(DTLB)] = {
+	/*
+	 * Only general DTLB misses are counted use the same event for
+	 * read and write.
+	 */
+	[C(OP_READ)] = {
+		[C(RESULT_MISS)]	= { 0x2d, CNTR_ALL }, /* PAPI_TLB_DM */
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_MISS)]	= { 0x2d, CNTR_ALL }, /* PAPI_TLB_DM */
+	},
+},
+[C(ITLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_MISS)]	= { 0x08, CNTR_ALL }, /* PAPI_TLB_IM */
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_MISS)]	= { 0x08, CNTR_ALL }, /* PAPI_TLB_IM */
+	},
+},
+[C(BPU)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_MISS)]	= { 0x25, CNTR_ALL },
+	},
+},
+};
+
+static int __hw_perf_event_init(struct perf_event *event)
+{
+	struct perf_event_attr *attr = &event->attr;
+	struct hw_perf_event *hwc = &event->hw;
+	const struct mips_perf_event *pev;
+	int err;
+
+	/* Returning MIPS event descriptor for generic perf event. */
+	if (PERF_TYPE_HARDWARE == event->attr.type) {
+		if (event->attr.config >= PERF_COUNT_HW_MAX)
+			return -EINVAL;
+		pev = mipspmu_map_general_event(event->attr.config);
+	} else if (PERF_TYPE_HW_CACHE == event->attr.type) {
+		pev = mipspmu_map_cache_event(event->attr.config);
+	} else if (PERF_TYPE_RAW == event->attr.type) {
+		/* We are working on the global raw event. */
+		mutex_lock(&raw_event_mutex);
+		pev = mipspmu.map_raw_event(event->attr.config);
+	} else {
+		/* The event type is not (yet) supported. */
+		return -EOPNOTSUPP;
+	}
+
+	if (IS_ERR(pev)) {
+		if (PERF_TYPE_RAW == event->attr.type)
+			mutex_unlock(&raw_event_mutex);
+		return PTR_ERR(pev);
+	}
+
+	/*
+	 * We allow max flexibility on how each individual counter shared
+	 * by the single CPU operates (the mode exclusion and the range).
+	 */
+	hwc->config_base = MIPS_PERFCTRL_IE;
+
+	hwc->event_base = mipspmu_perf_event_encode(pev);
+	if (PERF_TYPE_RAW == event->attr.type)
+		mutex_unlock(&raw_event_mutex);
+
+	if (!attr->exclude_user)
+		hwc->config_base |= MIPS_PERFCTRL_U;
+	if (!attr->exclude_kernel) {
+		hwc->config_base |= MIPS_PERFCTRL_K;
+		/* MIPS kernel mode: KSU == 00b || EXL == 1 || ERL == 1 */
+		hwc->config_base |= MIPS_PERFCTRL_EXL;
+	}
+	if (!attr->exclude_hv)
+		hwc->config_base |= MIPS_PERFCTRL_S;
+
+	hwc->config_base &= M_PERFCTL_CONFIG_MASK;
+	/*
+	 * The event can belong to another cpu. We do not assign a local
+	 * counter for it for now.
+	 */
+	hwc->idx = -1;
+	hwc->config = 0;
+
+	if (!hwc->sample_period) {
+		hwc->sample_period  = mipspmu.max_period;
+		hwc->last_period    = hwc->sample_period;
+		local64_set(&hwc->period_left, hwc->sample_period);
+	}
+
+	err = 0;
+	if (event->group_leader != event)
+		err = validate_group(event);
+
+	event->destroy = hw_perf_event_destroy;
+
+	if (err)
+		event->destroy(event);
+
+	return err;
+}
+
+static void pause_local_counters(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	int ctr = mipspmu.num_counters;
+	unsigned long flags;
+
+	local_irq_save(flags);
+	do {
+		ctr--;
+		cpuc->saved_ctrl[ctr] = mipsxx_pmu_read_control(ctr);
+		mipsxx_pmu_write_control(ctr, cpuc->saved_ctrl[ctr] &
+					 ~M_PERFCTL_COUNT_EVENT_WHENEVER);
+	} while (ctr > 0);
+	local_irq_restore(flags);
+}
+
+static void resume_local_counters(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	int ctr = mipspmu.num_counters;
+
+	do {
+		ctr--;
+		mipsxx_pmu_write_control(ctr, cpuc->saved_ctrl[ctr]);
+	} while (ctr > 0);
+}
+
+static int mipsxx_pmu_handle_shared_irq(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct perf_sample_data data;
+	unsigned int counters = mipspmu.num_counters;
+	u64 counter;
+	int n, handled = IRQ_NONE;
+	struct pt_regs *regs;
+
+	if (cpu_has_perf_cntr_intr_bit && !(read_c0_cause() & CAUSEF_PCI))
+		return handled;
+	/*
+	 * First we pause the local counters, so that when we are locked
+	 * here, the counters are all paused. When it gets locked due to
+	 * perf_disable(), the timer interrupt handler will be delayed.
+	 *
+	 * See also mipsxx_pmu_start().
+	 */
+	pause_local_counters();
+#ifdef CONFIG_MIPS_PERF_SHARED_TC_COUNTERS
+	read_lock(&pmuint_rwlock);
+#endif
+
+	regs = get_irq_regs();
+
+	perf_sample_data_init(&data, 0, 0);
+
+	for (n = counters - 1; n >= 0; n--) {
+		if (!test_bit(n, cpuc->used_mask))
+			continue;
+
+		counter = mipspmu.read_counter(n);
+		if (!(counter & mipspmu.overflow))
+			continue;
+
+		handle_associated_event(cpuc, n, &data, regs);
+		handled = IRQ_HANDLED;
+	}
+
+#ifdef CONFIG_MIPS_PERF_SHARED_TC_COUNTERS
+	read_unlock(&pmuint_rwlock);
+#endif
+	resume_local_counters();
+
+	/*
+	 * Do all the work for the pending perf events. We can do this
+	 * in here because the performance counter interrupt is a regular
+	 * interrupt, not NMI.
+	 */
+	if (handled == IRQ_HANDLED)
+		irq_work_run();
+
+	return handled;
+}
+
+static irqreturn_t mipsxx_pmu_handle_irq(int irq, void *dev)
+{
+	return mipsxx_pmu_handle_shared_irq();
+}
+
+/* 24K */
+#define IS_BOTH_COUNTERS_24K_EVENT(b)					\
+	((b) == 0 || (b) == 1 || (b) == 11)
+
+/* 34K */
+#define IS_BOTH_COUNTERS_34K_EVENT(b)					\
+	((b) == 0 || (b) == 1 || (b) == 11)
+#ifdef CONFIG_MIPS_MT_SMP
+#define IS_RANGE_P_34K_EVENT(r, b)					\
+	((b) == 0 || (r) == 18 || (b) == 21 || (b) == 22 ||		\
+	 (b) == 25 || (b) == 39 || (r) == 44 || (r) == 174 ||		\
+	 (r) == 176 || ((b) >= 50 && (b) <= 55) ||			\
+	 ((b) >= 64 && (b) <= 67))
+#define IS_RANGE_V_34K_EVENT(r) ((r) == 47)
+#endif
+
+/* 74K */
+#define IS_BOTH_COUNTERS_74K_EVENT(b)					\
+	((b) == 0 || (b) == 1)
+
+/* proAptiv */
+#define IS_BOTH_COUNTERS_PROAPTIV_EVENT(b)				\
+	((b) == 0 || (b) == 1)
+/* P5600 */
+#define IS_BOTH_COUNTERS_P5600_EVENT(b)					\
+	((b) == 0 || (b) == 1)
+
+/* 1004K */
+#define IS_BOTH_COUNTERS_1004K_EVENT(b)					\
+	((b) == 0 || (b) == 1 || (b) == 11)
+#ifdef CONFIG_MIPS_MT_SMP
+#define IS_RANGE_P_1004K_EVENT(r, b)					\
+	((b) == 0 || (r) == 18 || (b) == 21 || (b) == 22 ||		\
+	 (b) == 25 || (b) == 36 || (b) == 39 || (r) == 44 ||		\
+	 (r) == 174 || (r) == 176 || ((b) >= 50 && (b) <= 59) ||	\
+	 (r) == 188 || (b) == 61 || (b) == 62 ||			\
+	 ((b) >= 64 && (b) <= 67))
+#define IS_RANGE_V_1004K_EVENT(r)	((r) == 47)
+#endif
+
+/* interAptiv */
+#define IS_BOTH_COUNTERS_INTERAPTIV_EVENT(b)				\
+	((b) == 0 || (b) == 1 || (b) == 11)
+#ifdef CONFIG_MIPS_MT_SMP
+/* The P/V/T info is not provided for "(b) == 38" in SUM, assume P. */
+#define IS_RANGE_P_INTERAPTIV_EVENT(r, b)				\
+	((b) == 0 || (r) == 18 || (b) == 21 || (b) == 22 ||		\
+	 (b) == 25 || (b) == 36 || (b) == 38 || (b) == 39 ||		\
+	 (r) == 44 || (r) == 174 || (r) == 176 || ((b) >= 50 &&		\
+	 (b) <= 59) || (r) == 188 || (b) == 61 || (b) == 62 ||		\
+	 ((b) >= 64 && (b) <= 67))
+#define IS_RANGE_V_INTERAPTIV_EVENT(r)	((r) == 47 || (r) == 175)
+#endif
+
+/* BMIPS5000 */
+#define IS_BOTH_COUNTERS_BMIPS5000_EVENT(b)				\
+	((b) == 0 || (b) == 1)
+
+
+/*
+ * For most cores the user can use 0-255 raw events, where 0-127 for the events
+ * of even counters, and 128-255 for odd counters. Note that bit 7 is used to
+ * indicate the even/odd bank selector. So, for example, when user wants to take
+ * the Event Num of 15 for odd counters (by referring to the user manual), then
+ * 128 needs to be added to 15 as the input for the event config, i.e., 143 (0x8F)
+ * to be used.
+ *
+ * Some newer cores have even more events, in which case the user can use raw
+ * events 0-511, where 0-255 are for the events of even counters, and 256-511
+ * are for odd counters, so bit 8 is used to indicate the even/odd bank selector.
+ */
+static const struct mips_perf_event *mipsxx_pmu_map_raw_event(u64 config)
+{
+	/* currently most cores have 7-bit event numbers */
+	int pmu_type;
+	unsigned int raw_id = config & 0xff;
+	unsigned int base_id = raw_id & 0x7f;
+
+	switch (current_cpu_type()) {
+	case CPU_24K:
+		if (IS_BOTH_COUNTERS_24K_EVENT(base_id))
+			raw_event.cntr_mask = CNTR_EVEN | CNTR_ODD;
+		else
+			raw_event.cntr_mask =
+				raw_id > 127 ? CNTR_ODD : CNTR_EVEN;
+#ifdef CONFIG_MIPS_MT_SMP
+		/*
+		 * This is actually doing nothing. Non-multithreading
+		 * CPUs will not check and calculate the range.
+		 */
+		raw_event.range = P;
+#endif
+		break;
+	case CPU_34K:
+		if (IS_BOTH_COUNTERS_34K_EVENT(base_id))
+			raw_event.cntr_mask = CNTR_EVEN | CNTR_ODD;
+		else
+			raw_event.cntr_mask =
+				raw_id > 127 ? CNTR_ODD : CNTR_EVEN;
+#ifdef CONFIG_MIPS_MT_SMP
+		if (IS_RANGE_P_34K_EVENT(raw_id, base_id))
+			raw_event.range = P;
+		else if (unlikely(IS_RANGE_V_34K_EVENT(raw_id)))
+			raw_event.range = V;
+		else
+			raw_event.range = T;
+#endif
+		break;
+	case CPU_74K:
+	case CPU_1074K:
+		if (IS_BOTH_COUNTERS_74K_EVENT(base_id))
+			raw_event.cntr_mask = CNTR_EVEN | CNTR_ODD;
+		else
+			raw_event.cntr_mask =
+				raw_id > 127 ? CNTR_ODD : CNTR_EVEN;
+#ifdef CONFIG_MIPS_MT_SMP
+		raw_event.range = P;
+#endif
+		break;
+	case CPU_PROAPTIV:
+		if (IS_BOTH_COUNTERS_PROAPTIV_EVENT(base_id))
+			raw_event.cntr_mask = CNTR_EVEN | CNTR_ODD;
+		else
+			raw_event.cntr_mask =
+				raw_id > 127 ? CNTR_ODD : CNTR_EVEN;
+#ifdef CONFIG_MIPS_MT_SMP
+		raw_event.range = P;
+#endif
+		break;
+	case CPU_P5600:
+	case CPU_P6600:
+		/* 8-bit event numbers */
+		raw_id = config & 0x1ff;
+		base_id = raw_id & 0xff;
+		if (IS_BOTH_COUNTERS_P5600_EVENT(base_id))
+			raw_event.cntr_mask = CNTR_EVEN | CNTR_ODD;
+		else
+			raw_event.cntr_mask =
+				raw_id > 255 ? CNTR_ODD : CNTR_EVEN;
+#ifdef CONFIG_MIPS_MT_SMP
+		raw_event.range = P;
+#endif
+		break;
+	case CPU_I6400:
+	case CPU_I6500:
+		/* 8-bit event numbers */
+		base_id = config & 0xff;
+		raw_event.cntr_mask = CNTR_EVEN | CNTR_ODD;
+		break;
+	case CPU_1004K:
+		if (IS_BOTH_COUNTERS_1004K_EVENT(base_id))
+			raw_event.cntr_mask = CNTR_EVEN | CNTR_ODD;
+		else
+			raw_event.cntr_mask =
+				raw_id > 127 ? CNTR_ODD : CNTR_EVEN;
+#ifdef CONFIG_MIPS_MT_SMP
+		if (IS_RANGE_P_1004K_EVENT(raw_id, base_id))
+			raw_event.range = P;
+		else if (unlikely(IS_RANGE_V_1004K_EVENT(raw_id)))
+			raw_event.range = V;
+		else
+			raw_event.range = T;
+#endif
+		break;
+	case CPU_INTERAPTIV:
+		if (IS_BOTH_COUNTERS_INTERAPTIV_EVENT(base_id))
+			raw_event.cntr_mask = CNTR_EVEN | CNTR_ODD;
+		else
+			raw_event.cntr_mask =
+				raw_id > 127 ? CNTR_ODD : CNTR_EVEN;
+#ifdef CONFIG_MIPS_MT_SMP
+		if (IS_RANGE_P_INTERAPTIV_EVENT(raw_id, base_id))
+			raw_event.range = P;
+		else if (unlikely(IS_RANGE_V_INTERAPTIV_EVENT(raw_id)))
+			raw_event.range = V;
+		else
+			raw_event.range = T;
+#endif
+		break;
+	case CPU_BMIPS5000:
+		if (IS_BOTH_COUNTERS_BMIPS5000_EVENT(base_id))
+			raw_event.cntr_mask = CNTR_EVEN | CNTR_ODD;
+		else
+			raw_event.cntr_mask =
+				raw_id > 127 ? CNTR_ODD : CNTR_EVEN;
+		break;
+	case CPU_LOONGSON64:
+		pmu_type = get_loongson3_pmu_type();
+
+		switch (pmu_type) {
+		case LOONGSON_PMU_TYPE1:
+			raw_event.cntr_mask =
+				raw_id > 127 ? CNTR_ODD : CNTR_EVEN;
+			break;
+		case LOONGSON_PMU_TYPE2:
+			base_id = config & 0x3ff;
+			raw_event.cntr_mask = CNTR_ALL;
+
+			if ((base_id >= 1 && base_id < 28) ||
+				(base_id >= 64 && base_id < 90) ||
+				(base_id >= 128 && base_id < 164) ||
+				(base_id >= 192 && base_id < 200) ||
+				(base_id >= 256 && base_id < 275) ||
+				(base_id >= 320 && base_id < 361) ||
+				(base_id >= 384 && base_id < 574))
+				break;
+
+			return ERR_PTR(-EOPNOTSUPP);
+		case LOONGSON_PMU_TYPE3:
+			base_id = raw_id;
+			raw_event.cntr_mask = CNTR_ALL;
+			break;
+		}
+		break;
+	}
+
+	raw_event.event_id = base_id;
+
+	return &raw_event;
+}
+
+static const struct mips_perf_event *octeon_pmu_map_raw_event(u64 config)
+{
+	unsigned int raw_id = config & 0xff;
+	unsigned int base_id = raw_id & 0x7f;
+
+
+	raw_event.cntr_mask = CNTR_ALL;
+	raw_event.event_id = base_id;
+
+	if (current_cpu_type() == CPU_CAVIUM_OCTEON2) {
+		if (base_id > 0x42)
+			return ERR_PTR(-EOPNOTSUPP);
+	} else {
+		if (base_id > 0x3a)
+			return ERR_PTR(-EOPNOTSUPP);
+	}
+
+	switch (base_id) {
+	case 0x00:
+	case 0x0f:
+	case 0x1e:
+	case 0x1f:
+	case 0x2f:
+	case 0x34:
+	case 0x3b ... 0x3f:
+		return ERR_PTR(-EOPNOTSUPP);
+	default:
+		break;
+	}
+
+	return &raw_event;
+}
+
+static const struct mips_perf_event *xlp_pmu_map_raw_event(u64 config)
+{
+	unsigned int raw_id = config & 0xff;
+
+	/* Only 1-63 are defined */
+	if ((raw_id < 0x01) || (raw_id > 0x3f))
+		return ERR_PTR(-EOPNOTSUPP);
+
+	raw_event.cntr_mask = CNTR_ALL;
+	raw_event.event_id = raw_id;
+
+	return &raw_event;
+}
+
+static int __init
+init_hw_perf_events(void)
+{
+	int counters, irq, pmu_type;
+
+	pr_info("Performance counters: ");
+
+	counters = n_counters();
+	if (counters == 0) {
+		pr_cont("No available PMU.\n");
+		return -ENODEV;
+	}
+
+#ifdef CONFIG_MIPS_PERF_SHARED_TC_COUNTERS
+	if (!cpu_has_mipsmt_pertccounters)
+		counters = counters_total_to_per_cpu(counters);
+#endif
+
+	if (get_c0_perfcount_int)
+		irq = get_c0_perfcount_int();
+	else if (cp0_perfcount_irq >= 0)
+		irq = MIPS_CPU_IRQ_BASE + cp0_perfcount_irq;
+	else
+		irq = -1;
+
+	mipspmu.map_raw_event = mipsxx_pmu_map_raw_event;
+
+	switch (current_cpu_type()) {
+	case CPU_24K:
+		mipspmu.name = "mips/24K";
+		mipspmu.general_event_map = &mipsxxcore_event_map;
+		mipspmu.cache_event_map = &mipsxxcore_cache_map;
+		break;
+	case CPU_34K:
+		mipspmu.name = "mips/34K";
+		mipspmu.general_event_map = &mipsxxcore_event_map;
+		mipspmu.cache_event_map = &mipsxxcore_cache_map;
+		break;
+	case CPU_74K:
+		mipspmu.name = "mips/74K";
+		mipspmu.general_event_map = &mipsxxcore_event_map2;
+		mipspmu.cache_event_map = &mipsxxcore_cache_map2;
+		break;
+	case CPU_PROAPTIV:
+		mipspmu.name = "mips/proAptiv";
+		mipspmu.general_event_map = &mipsxxcore_event_map2;
+		mipspmu.cache_event_map = &mipsxxcore_cache_map2;
+		break;
+	case CPU_P5600:
+		mipspmu.name = "mips/P5600";
+		mipspmu.general_event_map = &mipsxxcore_event_map2;
+		mipspmu.cache_event_map = &mipsxxcore_cache_map2;
+		break;
+	case CPU_P6600:
+		mipspmu.name = "mips/P6600";
+		mipspmu.general_event_map = &mipsxxcore_event_map2;
+		mipspmu.cache_event_map = &mipsxxcore_cache_map2;
+		break;
+	case CPU_I6400:
+		mipspmu.name = "mips/I6400";
+		mipspmu.general_event_map = &i6x00_event_map;
+		mipspmu.cache_event_map = &i6x00_cache_map;
+		break;
+	case CPU_I6500:
+		mipspmu.name = "mips/I6500";
+		mipspmu.general_event_map = &i6x00_event_map;
+		mipspmu.cache_event_map = &i6x00_cache_map;
+		break;
+	case CPU_1004K:
+		mipspmu.name = "mips/1004K";
+		mipspmu.general_event_map = &mipsxxcore_event_map;
+		mipspmu.cache_event_map = &mipsxxcore_cache_map;
+		break;
+	case CPU_1074K:
+		mipspmu.name = "mips/1074K";
+		mipspmu.general_event_map = &mipsxxcore_event_map;
+		mipspmu.cache_event_map = &mipsxxcore_cache_map;
+		break;
+	case CPU_INTERAPTIV:
+		mipspmu.name = "mips/interAptiv";
+		mipspmu.general_event_map = &mipsxxcore_event_map;
+		mipspmu.cache_event_map = &mipsxxcore_cache_map;
+		break;
+	case CPU_LOONGSON32:
+		mipspmu.name = "mips/loongson1";
+		mipspmu.general_event_map = &mipsxxcore_event_map;
+		mipspmu.cache_event_map = &mipsxxcore_cache_map;
+		break;
+	case CPU_LOONGSON64:
+		mipspmu.name = "mips/loongson3";
+		pmu_type = get_loongson3_pmu_type();
+
+		switch (pmu_type) {
+		case LOONGSON_PMU_TYPE1:
+			counters = 2;
+			mipspmu.general_event_map = &loongson3_event_map1;
+			mipspmu.cache_event_map = &loongson3_cache_map1;
+			break;
+		case LOONGSON_PMU_TYPE2:
+			counters = 4;
+			mipspmu.general_event_map = &loongson3_event_map2;
+			mipspmu.cache_event_map = &loongson3_cache_map2;
+			break;
+		case LOONGSON_PMU_TYPE3:
+			counters = 4;
+			mipspmu.general_event_map = &loongson3_event_map3;
+			mipspmu.cache_event_map = &loongson3_cache_map3;
+			break;
+		}
+		break;
+	case CPU_CAVIUM_OCTEON:
+	case CPU_CAVIUM_OCTEON_PLUS:
+	case CPU_CAVIUM_OCTEON2:
+		mipspmu.name = "octeon";
+		mipspmu.general_event_map = &octeon_event_map;
+		mipspmu.cache_event_map = &octeon_cache_map;
+		mipspmu.map_raw_event = octeon_pmu_map_raw_event;
+		break;
+	case CPU_BMIPS5000:
+		mipspmu.name = "BMIPS5000";
+		mipspmu.general_event_map = &bmips5000_event_map;
+		mipspmu.cache_event_map = &bmips5000_cache_map;
+		break;
+	case CPU_XLP:
+		mipspmu.name = "xlp";
+		mipspmu.general_event_map = &xlp_event_map;
+		mipspmu.cache_event_map = &xlp_cache_map;
+		mipspmu.map_raw_event = xlp_pmu_map_raw_event;
+		break;
+	default:
+		pr_cont("Either hardware does not support performance "
+			"counters, or not yet implemented.\n");
+		return -ENODEV;
+	}
+
+	mipspmu.num_counters = counters;
+	mipspmu.irq = irq;
+
+	if (read_c0_perfctrl0() & MIPS_PERFCTRL_W) {
+		if (get_loongson3_pmu_type() == LOONGSON_PMU_TYPE2) {
+			counter_bits = 48;
+			mipspmu.max_period = (1ULL << 47) - 1;
+			mipspmu.valid_count = (1ULL << 47) - 1;
+			mipspmu.overflow = 1ULL << 47;
+		} else {
+			counter_bits = 64;
+			mipspmu.max_period = (1ULL << 63) - 1;
+			mipspmu.valid_count = (1ULL << 63) - 1;
+			mipspmu.overflow = 1ULL << 63;
+		}
+		mipspmu.read_counter = mipsxx_pmu_read_counter_64;
+		mipspmu.write_counter = mipsxx_pmu_write_counter_64;
+	} else {
+		counter_bits = 32;
+		mipspmu.max_period = (1ULL << 31) - 1;
+		mipspmu.valid_count = (1ULL << 31) - 1;
+		mipspmu.overflow = 1ULL << 31;
+		mipspmu.read_counter = mipsxx_pmu_read_counter;
+		mipspmu.write_counter = mipsxx_pmu_write_counter;
+	}
+
+	on_each_cpu(reset_counters, (void *)(long)counters, 1);
+
+	pr_cont("%s PMU enabled, %d %d-bit counters available to each "
+		"CPU, irq %d%s\n", mipspmu.name, counters, counter_bits, irq,
+		irq < 0 ? " (share with timer interrupt)" : "");
+
+	perf_pmu_register(&pmu, "cpu", PERF_TYPE_RAW);
+
+	return 0;
+}
+early_initcall(init_hw_perf_events);
diff --git a/arch/mips/kernel/pm-cps.c b/arch/mips/kernel/pm-cps.c
new file mode 100644
index 000000000..9bf60d7d4
--- /dev/null
+++ b/arch/mips/kernel/pm-cps.c
@@ -0,0 +1,738 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2014 Imagination Technologies
+ * Author: Paul Burton <paul.burton@mips.com>
+ */
+
+#include <linux/cpuhotplug.h>
+#include <linux/init.h>
+#include <linux/percpu.h>
+#include <linux/slab.h>
+#include <linux/suspend.h>
+
+#include <asm/asm-offsets.h>
+#include <asm/cacheflush.h>
+#include <asm/cacheops.h>
+#include <asm/idle.h>
+#include <asm/mips-cps.h>
+#include <asm/mipsmtregs.h>
+#include <asm/pm.h>
+#include <asm/pm-cps.h>
+#include <asm/smp-cps.h>
+#include <asm/uasm.h>
+
+/*
+ * cps_nc_entry_fn - type of a generated non-coherent state entry function
+ * @online: the count of online coupled VPEs
+ * @nc_ready_count: pointer to a non-coherent mapping of the core ready_count
+ *
+ * The code entering & exiting non-coherent states is generated at runtime
+ * using uasm, in order to ensure that the compiler cannot insert a stray
+ * memory access at an unfortunate time and to allow the generation of optimal
+ * core-specific code particularly for cache routines. If coupled_coherence
+ * is non-zero and this is the entry function for the CPS_PM_NC_WAIT state,
+ * returns the number of VPEs that were in the wait state at the point this
+ * VPE left it. Returns garbage if coupled_coherence is zero or this is not
+ * the entry function for CPS_PM_NC_WAIT.
+ */
+typedef unsigned (*cps_nc_entry_fn)(unsigned online, u32 *nc_ready_count);
+
+/*
+ * The entry point of the generated non-coherent idle state entry/exit
+ * functions. Actually per-core rather than per-CPU.
+ */
+static DEFINE_PER_CPU_READ_MOSTLY(cps_nc_entry_fn[CPS_PM_STATE_COUNT],
+				  nc_asm_enter);
+
+/* Bitmap indicating which states are supported by the system */
+static DECLARE_BITMAP(state_support, CPS_PM_STATE_COUNT);
+
+/*
+ * Indicates the number of coupled VPEs ready to operate in a non-coherent
+ * state. Actually per-core rather than per-CPU.
+ */
+static DEFINE_PER_CPU_ALIGNED(u32*, ready_count);
+
+/* Indicates online CPUs coupled with the current CPU */
+static DEFINE_PER_CPU_ALIGNED(cpumask_t, online_coupled);
+
+/*
+ * Used to synchronize entry to deep idle states. Actually per-core rather
+ * than per-CPU.
+ */
+static DEFINE_PER_CPU_ALIGNED(atomic_t, pm_barrier);
+
+/* Saved CPU state across the CPS_PM_POWER_GATED state */
+DEFINE_PER_CPU_ALIGNED(struct mips_static_suspend_state, cps_cpu_state);
+
+/* A somewhat arbitrary number of labels & relocs for uasm */
+static struct uasm_label labels[32];
+static struct uasm_reloc relocs[32];
+
+enum mips_reg {
+	zero, at, v0, v1, a0, a1, a2, a3,
+	t0, t1, t2, t3, t4, t5, t6, t7,
+	s0, s1, s2, s3, s4, s5, s6, s7,
+	t8, t9, k0, k1, gp, sp, fp, ra,
+};
+
+bool cps_pm_support_state(enum cps_pm_state state)
+{
+	return test_bit(state, state_support);
+}
+
+static void coupled_barrier(atomic_t *a, unsigned online)
+{
+	/*
+	 * This function is effectively the same as
+	 * cpuidle_coupled_parallel_barrier, which can't be used here since
+	 * there's no cpuidle device.
+	 */
+
+	if (!coupled_coherence)
+		return;
+
+	smp_mb__before_atomic();
+	atomic_inc(a);
+
+	while (atomic_read(a) < online)
+		cpu_relax();
+
+	if (atomic_inc_return(a) == online * 2) {
+		atomic_set(a, 0);
+		return;
+	}
+
+	while (atomic_read(a) > online)
+		cpu_relax();
+}
+
+int cps_pm_enter_state(enum cps_pm_state state)
+{
+	unsigned cpu = smp_processor_id();
+	unsigned core = cpu_core(&current_cpu_data);
+	unsigned online, left;
+	cpumask_t *coupled_mask = this_cpu_ptr(&online_coupled);
+	u32 *core_ready_count, *nc_core_ready_count;
+	void *nc_addr;
+	cps_nc_entry_fn entry;
+	struct core_boot_config *core_cfg;
+	struct vpe_boot_config *vpe_cfg;
+
+	/* Check that there is an entry function for this state */
+	entry = per_cpu(nc_asm_enter, core)[state];
+	if (!entry)
+		return -EINVAL;
+
+	/* Calculate which coupled CPUs (VPEs) are online */
+#if defined(CONFIG_MIPS_MT) || defined(CONFIG_CPU_MIPSR6)
+	if (cpu_online(cpu)) {
+		cpumask_and(coupled_mask, cpu_online_mask,
+			    &cpu_sibling_map[cpu]);
+		online = cpumask_weight(coupled_mask);
+		cpumask_clear_cpu(cpu, coupled_mask);
+	} else
+#endif
+	{
+		cpumask_clear(coupled_mask);
+		online = 1;
+	}
+
+	/* Setup the VPE to run mips_cps_pm_restore when started again */
+	if (IS_ENABLED(CONFIG_CPU_PM) && state == CPS_PM_POWER_GATED) {
+		/* Power gating relies upon CPS SMP */
+		if (!mips_cps_smp_in_use())
+			return -EINVAL;
+
+		core_cfg = &mips_cps_core_bootcfg[core];
+		vpe_cfg = &core_cfg->vpe_config[cpu_vpe_id(&current_cpu_data)];
+		vpe_cfg->pc = (unsigned long)mips_cps_pm_restore;
+		vpe_cfg->gp = (unsigned long)current_thread_info();
+		vpe_cfg->sp = 0;
+	}
+
+	/* Indicate that this CPU might not be coherent */
+	cpumask_clear_cpu(cpu, &cpu_coherent_mask);
+	smp_mb__after_atomic();
+
+	/* Create a non-coherent mapping of the core ready_count */
+	core_ready_count = per_cpu(ready_count, core);
+	nc_addr = kmap_noncoherent(virt_to_page(core_ready_count),
+				   (unsigned long)core_ready_count);
+	nc_addr += ((unsigned long)core_ready_count & ~PAGE_MASK);
+	nc_core_ready_count = nc_addr;
+
+	/* Ensure ready_count is zero-initialised before the assembly runs */
+	WRITE_ONCE(*nc_core_ready_count, 0);
+	coupled_barrier(&per_cpu(pm_barrier, core), online);
+
+	/* Run the generated entry code */
+	left = entry(online, nc_core_ready_count);
+
+	/* Remove the non-coherent mapping of ready_count */
+	kunmap_noncoherent();
+
+	/* Indicate that this CPU is definitely coherent */
+	cpumask_set_cpu(cpu, &cpu_coherent_mask);
+
+	/*
+	 * If this VPE is the first to leave the non-coherent wait state then
+	 * it needs to wake up any coupled VPEs still running their wait
+	 * instruction so that they return to cpuidle, which can then complete
+	 * coordination between the coupled VPEs & provide the governor with
+	 * a chance to reflect on the length of time the VPEs were in the
+	 * idle state.
+	 */
+	if (coupled_coherence && (state == CPS_PM_NC_WAIT) && (left == online))
+		arch_send_call_function_ipi_mask(coupled_mask);
+
+	return 0;
+}
+
+static void cps_gen_cache_routine(u32 **pp, struct uasm_label **pl,
+				  struct uasm_reloc **pr,
+				  const struct cache_desc *cache,
+				  unsigned op, int lbl)
+{
+	unsigned cache_size = cache->ways << cache->waybit;
+	unsigned i;
+	const unsigned unroll_lines = 32;
+
+	/* If the cache isn't present this function has it easy */
+	if (cache->flags & MIPS_CACHE_NOT_PRESENT)
+		return;
+
+	/* Load base address */
+	UASM_i_LA(pp, t0, (long)CKSEG0);
+
+	/* Calculate end address */
+	if (cache_size < 0x8000)
+		uasm_i_addiu(pp, t1, t0, cache_size);
+	else
+		UASM_i_LA(pp, t1, (long)(CKSEG0 + cache_size));
+
+	/* Start of cache op loop */
+	uasm_build_label(pl, *pp, lbl);
+
+	/* Generate the cache ops */
+	for (i = 0; i < unroll_lines; i++) {
+		if (cpu_has_mips_r6) {
+			uasm_i_cache(pp, op, 0, t0);
+			uasm_i_addiu(pp, t0, t0, cache->linesz);
+		} else {
+			uasm_i_cache(pp, op, i * cache->linesz, t0);
+		}
+	}
+
+	if (!cpu_has_mips_r6)
+		/* Update the base address */
+		uasm_i_addiu(pp, t0, t0, unroll_lines * cache->linesz);
+
+	/* Loop if we haven't reached the end address yet */
+	uasm_il_bne(pp, pr, t0, t1, lbl);
+	uasm_i_nop(pp);
+}
+
+static int cps_gen_flush_fsb(u32 **pp, struct uasm_label **pl,
+			     struct uasm_reloc **pr,
+			     const struct cpuinfo_mips *cpu_info,
+			     int lbl)
+{
+	unsigned i, fsb_size = 8;
+	unsigned num_loads = (fsb_size * 3) / 2;
+	unsigned line_stride = 2;
+	unsigned line_size = cpu_info->dcache.linesz;
+	unsigned perf_counter, perf_event;
+	unsigned revision = cpu_info->processor_id & PRID_REV_MASK;
+
+	/*
+	 * Determine whether this CPU requires an FSB flush, and if so which
+	 * performance counter/event reflect stalls due to a full FSB.
+	 */
+	switch (__get_cpu_type(cpu_info->cputype)) {
+	case CPU_INTERAPTIV:
+		perf_counter = 1;
+		perf_event = 51;
+		break;
+
+	case CPU_PROAPTIV:
+		/* Newer proAptiv cores don't require this workaround */
+		if (revision >= PRID_REV_ENCODE_332(1, 1, 0))
+			return 0;
+
+		/* On older ones it's unavailable */
+		return -1;
+
+	default:
+		/* Assume that the CPU does not need this workaround */
+		return 0;
+	}
+
+	/*
+	 * Ensure that the fill/store buffer (FSB) is not holding the results
+	 * of a prefetch, since if it is then the CPC sequencer may become
+	 * stuck in the D3 (ClrBus) state whilst entering a low power state.
+	 */
+
+	/* Preserve perf counter setup */
+	uasm_i_mfc0(pp, t2, 25, (perf_counter * 2) + 0); /* PerfCtlN */
+	uasm_i_mfc0(pp, t3, 25, (perf_counter * 2) + 1); /* PerfCntN */
+
+	/* Setup perf counter to count FSB full pipeline stalls */
+	uasm_i_addiu(pp, t0, zero, (perf_event << 5) | 0xf);
+	uasm_i_mtc0(pp, t0, 25, (perf_counter * 2) + 0); /* PerfCtlN */
+	uasm_i_ehb(pp);
+	uasm_i_mtc0(pp, zero, 25, (perf_counter * 2) + 1); /* PerfCntN */
+	uasm_i_ehb(pp);
+
+	/* Base address for loads */
+	UASM_i_LA(pp, t0, (long)CKSEG0);
+
+	/* Start of clear loop */
+	uasm_build_label(pl, *pp, lbl);
+
+	/* Perform some loads to fill the FSB */
+	for (i = 0; i < num_loads; i++)
+		uasm_i_lw(pp, zero, i * line_size * line_stride, t0);
+
+	/*
+	 * Invalidate the new D-cache entries so that the cache will need
+	 * refilling (via the FSB) if the loop is executed again.
+	 */
+	for (i = 0; i < num_loads; i++) {
+		uasm_i_cache(pp, Hit_Invalidate_D,
+			     i * line_size * line_stride, t0);
+		uasm_i_cache(pp, Hit_Writeback_Inv_SD,
+			     i * line_size * line_stride, t0);
+	}
+
+	/* Barrier ensuring previous cache invalidates are complete */
+	uasm_i_sync(pp, __SYNC_full);
+	uasm_i_ehb(pp);
+
+	/* Check whether the pipeline stalled due to the FSB being full */
+	uasm_i_mfc0(pp, t1, 25, (perf_counter * 2) + 1); /* PerfCntN */
+
+	/* Loop if it didn't */
+	uasm_il_beqz(pp, pr, t1, lbl);
+	uasm_i_nop(pp);
+
+	/* Restore perf counter 1. The count may well now be wrong... */
+	uasm_i_mtc0(pp, t2, 25, (perf_counter * 2) + 0); /* PerfCtlN */
+	uasm_i_ehb(pp);
+	uasm_i_mtc0(pp, t3, 25, (perf_counter * 2) + 1); /* PerfCntN */
+	uasm_i_ehb(pp);
+
+	return 0;
+}
+
+static void cps_gen_set_top_bit(u32 **pp, struct uasm_label **pl,
+				struct uasm_reloc **pr,
+				unsigned r_addr, int lbl)
+{
+	uasm_i_lui(pp, t0, uasm_rel_hi(0x80000000));
+	uasm_build_label(pl, *pp, lbl);
+	uasm_i_ll(pp, t1, 0, r_addr);
+	uasm_i_or(pp, t1, t1, t0);
+	uasm_i_sc(pp, t1, 0, r_addr);
+	uasm_il_beqz(pp, pr, t1, lbl);
+	uasm_i_nop(pp);
+}
+
+static void *cps_gen_entry_code(unsigned cpu, enum cps_pm_state state)
+{
+	struct uasm_label *l = labels;
+	struct uasm_reloc *r = relocs;
+	u32 *buf, *p;
+	const unsigned r_online = a0;
+	const unsigned r_nc_count = a1;
+	const unsigned r_pcohctl = t7;
+	const unsigned max_instrs = 256;
+	unsigned cpc_cmd;
+	int err;
+	enum {
+		lbl_incready = 1,
+		lbl_poll_cont,
+		lbl_secondary_hang,
+		lbl_disable_coherence,
+		lbl_flush_fsb,
+		lbl_invicache,
+		lbl_flushdcache,
+		lbl_hang,
+		lbl_set_cont,
+		lbl_secondary_cont,
+		lbl_decready,
+	};
+
+	/* Allocate a buffer to hold the generated code */
+	p = buf = kcalloc(max_instrs, sizeof(u32), GFP_KERNEL);
+	if (!buf)
+		return NULL;
+
+	/* Clear labels & relocs ready for (re)use */
+	memset(labels, 0, sizeof(labels));
+	memset(relocs, 0, sizeof(relocs));
+
+	if (IS_ENABLED(CONFIG_CPU_PM) && state == CPS_PM_POWER_GATED) {
+		/* Power gating relies upon CPS SMP */
+		if (!mips_cps_smp_in_use())
+			goto out_err;
+
+		/*
+		 * Save CPU state. Note the non-standard calling convention
+		 * with the return address placed in v0 to avoid clobbering
+		 * the ra register before it is saved.
+		 */
+		UASM_i_LA(&p, t0, (long)mips_cps_pm_save);
+		uasm_i_jalr(&p, v0, t0);
+		uasm_i_nop(&p);
+	}
+
+	/*
+	 * Load addresses of required CM & CPC registers. This is done early
+	 * because they're needed in both the enable & disable coherence steps
+	 * but in the coupled case the enable step will only run on one VPE.
+	 */
+	UASM_i_LA(&p, r_pcohctl, (long)addr_gcr_cl_coherence());
+
+	if (coupled_coherence) {
+		/* Increment ready_count */
+		uasm_i_sync(&p, __SYNC_mb);
+		uasm_build_label(&l, p, lbl_incready);
+		uasm_i_ll(&p, t1, 0, r_nc_count);
+		uasm_i_addiu(&p, t2, t1, 1);
+		uasm_i_sc(&p, t2, 0, r_nc_count);
+		uasm_il_beqz(&p, &r, t2, lbl_incready);
+		uasm_i_addiu(&p, t1, t1, 1);
+
+		/* Barrier ensuring all CPUs see the updated r_nc_count value */
+		uasm_i_sync(&p, __SYNC_mb);
+
+		/*
+		 * If this is the last VPE to become ready for non-coherence
+		 * then it should branch below.
+		 */
+		uasm_il_beq(&p, &r, t1, r_online, lbl_disable_coherence);
+		uasm_i_nop(&p);
+
+		if (state < CPS_PM_POWER_GATED) {
+			/*
+			 * Otherwise this is not the last VPE to become ready
+			 * for non-coherence. It needs to wait until coherence
+			 * has been disabled before proceeding, which it will do
+			 * by polling for the top bit of ready_count being set.
+			 */
+			uasm_i_addiu(&p, t1, zero, -1);
+			uasm_build_label(&l, p, lbl_poll_cont);
+			uasm_i_lw(&p, t0, 0, r_nc_count);
+			uasm_il_bltz(&p, &r, t0, lbl_secondary_cont);
+			uasm_i_ehb(&p);
+			if (cpu_has_mipsmt)
+				uasm_i_yield(&p, zero, t1);
+			uasm_il_b(&p, &r, lbl_poll_cont);
+			uasm_i_nop(&p);
+		} else {
+			/*
+			 * The core will lose power & this VPE will not continue
+			 * so it can simply halt here.
+			 */
+			if (cpu_has_mipsmt) {
+				/* Halt the VPE via C0 tchalt register */
+				uasm_i_addiu(&p, t0, zero, TCHALT_H);
+				uasm_i_mtc0(&p, t0, 2, 4);
+			} else if (cpu_has_vp) {
+				/* Halt the VP via the CPC VP_STOP register */
+				unsigned int vpe_id;
+
+				vpe_id = cpu_vpe_id(&cpu_data[cpu]);
+				uasm_i_addiu(&p, t0, zero, 1 << vpe_id);
+				UASM_i_LA(&p, t1, (long)addr_cpc_cl_vp_stop());
+				uasm_i_sw(&p, t0, 0, t1);
+			} else {
+				BUG();
+			}
+			uasm_build_label(&l, p, lbl_secondary_hang);
+			uasm_il_b(&p, &r, lbl_secondary_hang);
+			uasm_i_nop(&p);
+		}
+	}
+
+	/*
+	 * This is the point of no return - this VPE will now proceed to
+	 * disable coherence. At this point we *must* be sure that no other
+	 * VPE within the core will interfere with the L1 dcache.
+	 */
+	uasm_build_label(&l, p, lbl_disable_coherence);
+
+	/* Invalidate the L1 icache */
+	cps_gen_cache_routine(&p, &l, &r, &cpu_data[cpu].icache,
+			      Index_Invalidate_I, lbl_invicache);
+
+	/* Writeback & invalidate the L1 dcache */
+	cps_gen_cache_routine(&p, &l, &r, &cpu_data[cpu].dcache,
+			      Index_Writeback_Inv_D, lbl_flushdcache);
+
+	/* Barrier ensuring previous cache invalidates are complete */
+	uasm_i_sync(&p, __SYNC_full);
+	uasm_i_ehb(&p);
+
+	if (mips_cm_revision() < CM_REV_CM3) {
+		/*
+		* Disable all but self interventions. The load from COHCTL is
+		* defined by the interAptiv & proAptiv SUMs as ensuring that the
+		*  operation resulting from the preceding store is complete.
+		*/
+		uasm_i_addiu(&p, t0, zero, 1 << cpu_core(&cpu_data[cpu]));
+		uasm_i_sw(&p, t0, 0, r_pcohctl);
+		uasm_i_lw(&p, t0, 0, r_pcohctl);
+
+		/* Barrier to ensure write to coherence control is complete */
+		uasm_i_sync(&p, __SYNC_full);
+		uasm_i_ehb(&p);
+	}
+
+	/* Disable coherence */
+	uasm_i_sw(&p, zero, 0, r_pcohctl);
+	uasm_i_lw(&p, t0, 0, r_pcohctl);
+
+	if (state >= CPS_PM_CLOCK_GATED) {
+		err = cps_gen_flush_fsb(&p, &l, &r, &cpu_data[cpu],
+					lbl_flush_fsb);
+		if (err)
+			goto out_err;
+
+		/* Determine the CPC command to issue */
+		switch (state) {
+		case CPS_PM_CLOCK_GATED:
+			cpc_cmd = CPC_Cx_CMD_CLOCKOFF;
+			break;
+		case CPS_PM_POWER_GATED:
+			cpc_cmd = CPC_Cx_CMD_PWRDOWN;
+			break;
+		default:
+			BUG();
+			goto out_err;
+		}
+
+		/* Issue the CPC command */
+		UASM_i_LA(&p, t0, (long)addr_cpc_cl_cmd());
+		uasm_i_addiu(&p, t1, zero, cpc_cmd);
+		uasm_i_sw(&p, t1, 0, t0);
+
+		if (state == CPS_PM_POWER_GATED) {
+			/* If anything goes wrong just hang */
+			uasm_build_label(&l, p, lbl_hang);
+			uasm_il_b(&p, &r, lbl_hang);
+			uasm_i_nop(&p);
+
+			/*
+			 * There's no point generating more code, the core is
+			 * powered down & if powered back up will run from the
+			 * reset vector not from here.
+			 */
+			goto gen_done;
+		}
+
+		/* Barrier to ensure write to CPC command is complete */
+		uasm_i_sync(&p, __SYNC_full);
+		uasm_i_ehb(&p);
+	}
+
+	if (state == CPS_PM_NC_WAIT) {
+		/*
+		 * At this point it is safe for all VPEs to proceed with
+		 * execution. This VPE will set the top bit of ready_count
+		 * to indicate to the other VPEs that they may continue.
+		 */
+		if (coupled_coherence)
+			cps_gen_set_top_bit(&p, &l, &r, r_nc_count,
+					    lbl_set_cont);
+
+		/*
+		 * VPEs which did not disable coherence will continue
+		 * executing, after coherence has been disabled, from this
+		 * point.
+		 */
+		uasm_build_label(&l, p, lbl_secondary_cont);
+
+		/* Now perform our wait */
+		uasm_i_wait(&p, 0);
+	}
+
+	/*
+	 * Re-enable coherence. Note that for CPS_PM_NC_WAIT all coupled VPEs
+	 * will run this. The first will actually re-enable coherence & the
+	 * rest will just be performing a rather unusual nop.
+	 */
+	uasm_i_addiu(&p, t0, zero, mips_cm_revision() < CM_REV_CM3
+				? CM_GCR_Cx_COHERENCE_COHDOMAINEN
+				: CM3_GCR_Cx_COHERENCE_COHEN);
+
+	uasm_i_sw(&p, t0, 0, r_pcohctl);
+	uasm_i_lw(&p, t0, 0, r_pcohctl);
+
+	/* Barrier to ensure write to coherence control is complete */
+	uasm_i_sync(&p, __SYNC_full);
+	uasm_i_ehb(&p);
+
+	if (coupled_coherence && (state == CPS_PM_NC_WAIT)) {
+		/* Decrement ready_count */
+		uasm_build_label(&l, p, lbl_decready);
+		uasm_i_sync(&p, __SYNC_mb);
+		uasm_i_ll(&p, t1, 0, r_nc_count);
+		uasm_i_addiu(&p, t2, t1, -1);
+		uasm_i_sc(&p, t2, 0, r_nc_count);
+		uasm_il_beqz(&p, &r, t2, lbl_decready);
+		uasm_i_andi(&p, v0, t1, (1 << fls(smp_num_siblings)) - 1);
+
+		/* Barrier ensuring all CPUs see the updated r_nc_count value */
+		uasm_i_sync(&p, __SYNC_mb);
+	}
+
+	if (coupled_coherence && (state == CPS_PM_CLOCK_GATED)) {
+		/*
+		 * At this point it is safe for all VPEs to proceed with
+		 * execution. This VPE will set the top bit of ready_count
+		 * to indicate to the other VPEs that they may continue.
+		 */
+		cps_gen_set_top_bit(&p, &l, &r, r_nc_count, lbl_set_cont);
+
+		/*
+		 * This core will be reliant upon another core sending a
+		 * power-up command to the CPC in order to resume operation.
+		 * Thus an arbitrary VPE can't trigger the core leaving the
+		 * idle state and the one that disables coherence might as well
+		 * be the one to re-enable it. The rest will continue from here
+		 * after that has been done.
+		 */
+		uasm_build_label(&l, p, lbl_secondary_cont);
+
+		/* Barrier ensuring all CPUs see the updated r_nc_count value */
+		uasm_i_sync(&p, __SYNC_mb);
+	}
+
+	/* The core is coherent, time to return to C code */
+	uasm_i_jr(&p, ra);
+	uasm_i_nop(&p);
+
+gen_done:
+	/* Ensure the code didn't exceed the resources allocated for it */
+	BUG_ON((p - buf) > max_instrs);
+	BUG_ON((l - labels) > ARRAY_SIZE(labels));
+	BUG_ON((r - relocs) > ARRAY_SIZE(relocs));
+
+	/* Patch branch offsets */
+	uasm_resolve_relocs(relocs, labels);
+
+	/* Flush the icache */
+	local_flush_icache_range((unsigned long)buf, (unsigned long)p);
+
+	return buf;
+out_err:
+	kfree(buf);
+	return NULL;
+}
+
+static int cps_pm_online_cpu(unsigned int cpu)
+{
+	enum cps_pm_state state;
+	unsigned core = cpu_core(&cpu_data[cpu]);
+	void *entry_fn, *core_rc;
+
+	for (state = CPS_PM_NC_WAIT; state < CPS_PM_STATE_COUNT; state++) {
+		if (per_cpu(nc_asm_enter, core)[state])
+			continue;
+		if (!test_bit(state, state_support))
+			continue;
+
+		entry_fn = cps_gen_entry_code(cpu, state);
+		if (!entry_fn) {
+			pr_err("Failed to generate core %u state %u entry\n",
+			       core, state);
+			clear_bit(state, state_support);
+		}
+
+		per_cpu(nc_asm_enter, core)[state] = entry_fn;
+	}
+
+	if (!per_cpu(ready_count, core)) {
+		core_rc = kmalloc(sizeof(u32), GFP_KERNEL);
+		if (!core_rc) {
+			pr_err("Failed allocate core %u ready_count\n", core);
+			return -ENOMEM;
+		}
+		per_cpu(ready_count, core) = core_rc;
+	}
+
+	return 0;
+}
+
+static int cps_pm_power_notifier(struct notifier_block *this,
+				 unsigned long event, void *ptr)
+{
+	unsigned int stat;
+
+	switch (event) {
+	case PM_SUSPEND_PREPARE:
+		stat = read_cpc_cl_stat_conf();
+		/*
+		 * If we're attempting to suspend the system and power down all
+		 * of the cores, the JTAG detect bit indicates that the CPC will
+		 * instead put the cores into clock-off state. In this state
+		 * a connected debugger can cause the CPU to attempt
+		 * interactions with the powered down system. At best this will
+		 * fail. At worst, it can hang the NoC, requiring a hard reset.
+		 * To avoid this, just block system suspend if a JTAG probe
+		 * is detected.
+		 */
+		if (stat & CPC_Cx_STAT_CONF_EJTAG_PROBE) {
+			pr_warn("JTAG probe is connected - abort suspend\n");
+			return NOTIFY_BAD;
+		}
+		return NOTIFY_DONE;
+	default:
+		return NOTIFY_DONE;
+	}
+}
+
+static int __init cps_pm_init(void)
+{
+	/* A CM is required for all non-coherent states */
+	if (!mips_cm_present()) {
+		pr_warn("pm-cps: no CM, non-coherent states unavailable\n");
+		return 0;
+	}
+
+	/*
+	 * If interrupts were enabled whilst running a wait instruction on a
+	 * non-coherent core then the VPE may end up processing interrupts
+	 * whilst non-coherent. That would be bad.
+	 */
+	if (cpu_wait == r4k_wait_irqoff)
+		set_bit(CPS_PM_NC_WAIT, state_support);
+	else
+		pr_warn("pm-cps: non-coherent wait unavailable\n");
+
+	/* Detect whether a CPC is present */
+	if (mips_cpc_present()) {
+		/* Detect whether clock gating is implemented */
+		if (read_cpc_cl_stat_conf() & CPC_Cx_STAT_CONF_CLKGAT_IMPL)
+			set_bit(CPS_PM_CLOCK_GATED, state_support);
+		else
+			pr_warn("pm-cps: CPC does not support clock gating\n");
+
+		/* Power gating is available with CPS SMP & any CPC */
+		if (mips_cps_smp_in_use())
+			set_bit(CPS_PM_POWER_GATED, state_support);
+		else
+			pr_warn("pm-cps: CPS SMP not in use, power gating unavailable\n");
+	} else {
+		pr_warn("pm-cps: no CPC, clock & power gating unavailable\n");
+	}
+
+	pm_notifier(cps_pm_power_notifier, 0);
+
+	return cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "mips/cps_pm:online",
+				 cps_pm_online_cpu, NULL);
+}
+arch_initcall(cps_pm_init);
diff --git a/arch/mips/kernel/pm.c b/arch/mips/kernel/pm.c
new file mode 100644
index 000000000..486ed2bf2
--- /dev/null
+++ b/arch/mips/kernel/pm.c
@@ -0,0 +1,95 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2014 Imagination Technologies Ltd.
+ *
+ * CPU PM notifiers for saving/restoring general CPU state.
+ */
+
+#include <linux/cpu_pm.h>
+#include <linux/init.h>
+
+#include <asm/dsp.h>
+#include <asm/fpu.h>
+#include <asm/mmu_context.h>
+#include <asm/pm.h>
+#include <asm/watch.h>
+
+/* Used by PM helper macros in asm/pm.h */
+struct mips_static_suspend_state mips_static_suspend_state;
+
+/**
+ * mips_cpu_save() - Save general CPU state.
+ * Ensures that general CPU context is saved, notably FPU and DSP.
+ */
+static int mips_cpu_save(void)
+{
+	/* Save FPU state */
+	lose_fpu(1);
+
+	/* Save DSP state */
+	save_dsp(current);
+
+	return 0;
+}
+
+/**
+ * mips_cpu_restore() - Restore general CPU state.
+ * Restores important CPU context.
+ */
+static void mips_cpu_restore(void)
+{
+	unsigned int cpu = smp_processor_id();
+
+	/* Restore ASID */
+	if (current->mm)
+		write_c0_entryhi(cpu_asid(cpu, current->mm));
+
+	/* Restore DSP state */
+	restore_dsp(current);
+
+	/* Restore UserLocal */
+	if (cpu_has_userlocal)
+		write_c0_userlocal(current_thread_info()->tp_value);
+
+	/* Restore watch registers */
+	__restore_watch(current);
+}
+
+/**
+ * mips_pm_notifier() - Notifier for preserving general CPU context.
+ * @self:	Notifier block.
+ * @cmd:	CPU PM event.
+ * @v:		Private data (unused).
+ *
+ * This is called when a CPU power management event occurs, and is used to
+ * ensure that important CPU context is preserved across a CPU power down.
+ */
+static int mips_pm_notifier(struct notifier_block *self, unsigned long cmd,
+			    void *v)
+{
+	int ret;
+
+	switch (cmd) {
+	case CPU_PM_ENTER:
+		ret = mips_cpu_save();
+		if (ret)
+			return NOTIFY_STOP;
+		break;
+	case CPU_PM_ENTER_FAILED:
+	case CPU_PM_EXIT:
+		mips_cpu_restore();
+		break;
+	}
+
+	return NOTIFY_OK;
+}
+
+static struct notifier_block mips_pm_notifier_block = {
+	.notifier_call = mips_pm_notifier,
+};
+
+static int __init mips_pm_init(void)
+{
+	return cpu_pm_register_notifier(&mips_pm_notifier_block);
+}
+arch_initcall(mips_pm_init);
diff --git a/arch/mips/kernel/probes-common.h b/arch/mips/kernel/probes-common.h
new file mode 100644
index 000000000..73e1d5e95
--- /dev/null
+++ b/arch/mips/kernel/probes-common.h
@@ -0,0 +1,79 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Copyright (C) 2016 Imagination Technologies
+ * Author: Marcin Nowakowski <marcin.nowakowski@mips.com>
+ */
+
+#ifndef __PROBES_COMMON_H
+#define __PROBES_COMMON_H
+
+#include <asm/inst.h>
+
+int __insn_is_compact_branch(union mips_instruction insn);
+
+static inline int __insn_has_delay_slot(const union mips_instruction insn)
+{
+	switch (insn.i_format.opcode) {
+	/*
+	 * jr and jalr are in r_format format.
+	 */
+	case spec_op:
+		switch (insn.r_format.func) {
+		case jalr_op:
+		case jr_op:
+			return 1;
+		}
+		break;
+
+	/*
+	 * This group contains:
+	 * bltz_op, bgez_op, bltzl_op, bgezl_op,
+	 * bltzal_op, bgezal_op, bltzall_op, bgezall_op.
+	 */
+	case bcond_op:
+		switch (insn.i_format.rt) {
+		case bltz_op:
+		case bltzl_op:
+		case bgez_op:
+		case bgezl_op:
+		case bltzal_op:
+		case bltzall_op:
+		case bgezal_op:
+		case bgezall_op:
+		case bposge32_op:
+			return 1;
+		}
+		break;
+
+	/*
+	 * These are unconditional and in j_format.
+	 */
+	case jal_op:
+	case j_op:
+	case beq_op:
+	case beql_op:
+	case bne_op:
+	case bnel_op:
+	case blez_op: /* not really i_format */
+	case blezl_op:
+	case bgtz_op:
+	case bgtzl_op:
+		return 1;
+
+	/*
+	 * And now the FPA/cp1 branch instructions.
+	 */
+	case cop1_op:
+#ifdef CONFIG_CPU_CAVIUM_OCTEON
+	case lwc2_op: /* This is bbit0 on Octeon */
+	case ldc2_op: /* This is bbit032 on Octeon */
+	case swc2_op: /* This is bbit1 on Octeon */
+	case sdc2_op: /* This is bbit132 on Octeon */
+#endif
+		return 1;
+	}
+
+	return 0;
+}
+
+#endif  /* __PROBES_COMMON_H */
diff --git a/arch/mips/kernel/proc.c b/arch/mips/kernel/proc.c
new file mode 100644
index 000000000..33a02f381
--- /dev/null
+++ b/arch/mips/kernel/proc.c
@@ -0,0 +1,193 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ *  Copyright (C) 1995, 1996, 2001  Ralf Baechle
+ *  Copyright (C) 2001, 2004  MIPS Technologies, Inc.
+ *  Copyright (C) 2004	Maciej W. Rozycki
+ */
+#include <linux/delay.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/seq_file.h>
+#include <asm/bootinfo.h>
+#include <asm/cpu.h>
+#include <asm/cpu-features.h>
+#include <asm/idle.h>
+#include <asm/mipsregs.h>
+#include <asm/processor.h>
+#include <asm/prom.h>
+
+unsigned int vced_count, vcei_count;
+
+/*
+ *  * No lock; only written during early bootup by CPU 0.
+ *   */
+static RAW_NOTIFIER_HEAD(proc_cpuinfo_chain);
+
+int __ref register_proc_cpuinfo_notifier(struct notifier_block *nb)
+{
+	return raw_notifier_chain_register(&proc_cpuinfo_chain, nb);
+}
+
+int proc_cpuinfo_notifier_call_chain(unsigned long val, void *v)
+{
+	return raw_notifier_call_chain(&proc_cpuinfo_chain, val, v);
+}
+
+static int show_cpuinfo(struct seq_file *m, void *v)
+{
+	struct proc_cpuinfo_notifier_args proc_cpuinfo_notifier_args;
+	unsigned long n = (unsigned long) v - 1;
+	unsigned int version = cpu_data[n].processor_id;
+	unsigned int fp_vers = cpu_data[n].fpu_id;
+	char fmt [64];
+	int i;
+
+#ifdef CONFIG_SMP
+	if (!cpu_online(n))
+		return 0;
+#endif
+
+	/*
+	 * For the first processor also print the system type
+	 */
+	if (n == 0) {
+		seq_printf(m, "system type\t\t: %s\n", get_system_type());
+		if (mips_get_machine_name())
+			seq_printf(m, "machine\t\t\t: %s\n",
+				   mips_get_machine_name());
+	}
+
+	seq_printf(m, "processor\t\t: %ld\n", n);
+	sprintf(fmt, "cpu model\t\t: %%s V%%d.%%d%s\n",
+		      cpu_data[n].options & MIPS_CPU_FPU ? "  FPU V%d.%d" : "");
+	seq_printf(m, fmt, __cpu_name[n],
+		      (version >> 4) & 0x0f, version & 0x0f,
+		      (fp_vers >> 4) & 0x0f, fp_vers & 0x0f);
+	seq_printf(m, "BogoMIPS\t\t: %u.%02u\n",
+		      cpu_data[n].udelay_val / (500000/HZ),
+		      (cpu_data[n].udelay_val / (5000/HZ)) % 100);
+	seq_printf(m, "wait instruction\t: %s\n", cpu_wait ? "yes" : "no");
+	seq_printf(m, "microsecond timers\t: %s\n",
+		      cpu_has_counter ? "yes" : "no");
+	seq_printf(m, "tlb_entries\t\t: %d\n", cpu_data[n].tlbsize);
+	seq_printf(m, "extra interrupt vector\t: %s\n",
+		      cpu_has_divec ? "yes" : "no");
+	seq_printf(m, "hardware watchpoint\t: %s",
+		      cpu_has_watch ? "yes, " : "no\n");
+	if (cpu_has_watch) {
+		seq_printf(m, "count: %d, address/irw mask: [",
+		      cpu_data[n].watch_reg_count);
+		for (i = 0; i < cpu_data[n].watch_reg_count; i++)
+			seq_printf(m, "%s0x%04x", i ? ", " : "" ,
+				cpu_data[n].watch_reg_masks[i]);
+		seq_printf(m, "]\n");
+	}
+
+	seq_printf(m, "isa\t\t\t:"); 
+	if (cpu_has_mips_1)
+		seq_printf(m, " mips1");
+	if (cpu_has_mips_2)
+		seq_printf(m, "%s", " mips2");
+	if (cpu_has_mips_3)
+		seq_printf(m, "%s", " mips3");
+	if (cpu_has_mips_4)
+		seq_printf(m, "%s", " mips4");
+	if (cpu_has_mips_5)
+		seq_printf(m, "%s", " mips5");
+	if (cpu_has_mips32r1)
+		seq_printf(m, "%s", " mips32r1");
+	if (cpu_has_mips32r2)
+		seq_printf(m, "%s", " mips32r2");
+	if (cpu_has_mips32r5)
+		seq_printf(m, "%s", " mips32r5");
+	if (cpu_has_mips32r6)
+		seq_printf(m, "%s", " mips32r6");
+	if (cpu_has_mips64r1)
+		seq_printf(m, "%s", " mips64r1");
+	if (cpu_has_mips64r2)
+		seq_printf(m, "%s", " mips64r2");
+	if (cpu_has_mips64r5)
+		seq_printf(m, "%s", " mips64r5");
+	if (cpu_has_mips64r6)
+		seq_printf(m, "%s", " mips64r6");
+	seq_printf(m, "\n");
+
+	seq_printf(m, "ASEs implemented\t:");
+	if (cpu_has_mips16)	seq_printf(m, "%s", " mips16");
+	if (cpu_has_mips16e2)	seq_printf(m, "%s", " mips16e2");
+	if (cpu_has_mdmx)	seq_printf(m, "%s", " mdmx");
+	if (cpu_has_mips3d)	seq_printf(m, "%s", " mips3d");
+	if (cpu_has_smartmips)	seq_printf(m, "%s", " smartmips");
+	if (cpu_has_dsp)	seq_printf(m, "%s", " dsp");
+	if (cpu_has_dsp2)	seq_printf(m, "%s", " dsp2");
+	if (cpu_has_dsp3)	seq_printf(m, "%s", " dsp3");
+	if (cpu_has_mipsmt)	seq_printf(m, "%s", " mt");
+	if (cpu_has_mmips)	seq_printf(m, "%s", " micromips");
+	if (cpu_has_vz)		seq_printf(m, "%s", " vz");
+	if (cpu_has_msa)	seq_printf(m, "%s", " msa");
+	if (cpu_has_eva)	seq_printf(m, "%s", " eva");
+	if (cpu_has_htw)	seq_printf(m, "%s", " htw");
+	if (cpu_has_xpa)	seq_printf(m, "%s", " xpa");
+	if (cpu_has_loongson_mmi)	seq_printf(m, "%s", " loongson-mmi");
+	if (cpu_has_loongson_cam)	seq_printf(m, "%s", " loongson-cam");
+	if (cpu_has_loongson_ext)	seq_printf(m, "%s", " loongson-ext");
+	if (cpu_has_loongson_ext2)	seq_printf(m, "%s", " loongson-ext2");
+	seq_printf(m, "\n");
+
+	if (cpu_has_mmips) {
+		seq_printf(m, "micromips kernel\t: %s\n",
+		      (read_c0_config3() & MIPS_CONF3_ISA_OE) ?  "yes" : "no");
+	}
+	seq_printf(m, "shadow register sets\t: %d\n",
+		      cpu_data[n].srsets);
+	seq_printf(m, "kscratch registers\t: %d\n",
+		      hweight8(cpu_data[n].kscratch_mask));
+	seq_printf(m, "package\t\t\t: %d\n", cpu_data[n].package);
+	seq_printf(m, "core\t\t\t: %d\n", cpu_core(&cpu_data[n]));
+
+#if defined(CONFIG_MIPS_MT_SMP) || defined(CONFIG_CPU_MIPSR6)
+	if (cpu_has_mipsmt)
+		seq_printf(m, "VPE\t\t\t: %d\n", cpu_vpe_id(&cpu_data[n]));
+	else if (cpu_has_vp)
+		seq_printf(m, "VP\t\t\t: %d\n", cpu_vpe_id(&cpu_data[n]));
+#endif
+
+	sprintf(fmt, "VCE%%c exceptions\t\t: %s\n",
+		      cpu_has_vce ? "%u" : "not available");
+	seq_printf(m, fmt, 'D', vced_count);
+	seq_printf(m, fmt, 'I', vcei_count);
+
+	proc_cpuinfo_notifier_args.m = m;
+	proc_cpuinfo_notifier_args.n = n;
+
+	raw_notifier_call_chain(&proc_cpuinfo_chain, 0,
+				&proc_cpuinfo_notifier_args);
+
+	seq_printf(m, "\n");
+
+	return 0;
+}
+
+static void *c_start(struct seq_file *m, loff_t *pos)
+{
+	unsigned long i = *pos;
+
+	return i < nr_cpu_ids ? (void *) (i + 1) : NULL;
+}
+
+static void *c_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	++*pos;
+	return c_start(m, pos);
+}
+
+static void c_stop(struct seq_file *m, void *v)
+{
+}
+
+const struct seq_operations cpuinfo_op = {
+	.start	= c_start,
+	.next	= c_next,
+	.stop	= c_stop,
+	.show	= show_cpuinfo,
+};
diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
new file mode 100644
index 000000000..98ecaf6f3
--- /dev/null
+++ b/arch/mips/kernel/process.c
@@ -0,0 +1,896 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1994 - 1999, 2000 by Ralf Baechle and others.
+ * Copyright (C) 2005, 2006 by Ralf Baechle (ralf@linux-mips.org)
+ * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ * Copyright (C) 2004 Thiemo Seufer
+ * Copyright (C) 2013  Imagination Technologies Ltd.
+ */
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/sched/debug.h>
+#include <linux/sched/task.h>
+#include <linux/sched/task_stack.h>
+#include <linux/tick.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/stddef.h>
+#include <linux/unistd.h>
+#include <linux/export.h>
+#include <linux/ptrace.h>
+#include <linux/mman.h>
+#include <linux/personality.h>
+#include <linux/sys.h>
+#include <linux/init.h>
+#include <linux/completion.h>
+#include <linux/kallsyms.h>
+#include <linux/random.h>
+#include <linux/prctl.h>
+#include <linux/nmi.h>
+#include <linux/cpu.h>
+
+#include <asm/abi.h>
+#include <asm/asm.h>
+#include <asm/bootinfo.h>
+#include <asm/cpu.h>
+#include <asm/dsemul.h>
+#include <asm/dsp.h>
+#include <asm/fpu.h>
+#include <asm/irq.h>
+#include <asm/mips-cps.h>
+#include <asm/msa.h>
+#include <asm/mipsregs.h>
+#include <asm/processor.h>
+#include <asm/reg.h>
+#include <linux/uaccess.h>
+#include <asm/io.h>
+#include <asm/elf.h>
+#include <asm/isadep.h>
+#include <asm/inst.h>
+#include <asm/stacktrace.h>
+#include <asm/irq_regs.h>
+#include <asm/exec.h>
+
+#ifdef CONFIG_HOTPLUG_CPU
+void arch_cpu_idle_dead(void)
+{
+	play_dead();
+}
+#endif
+
+asmlinkage void ret_from_fork(void);
+asmlinkage void ret_from_kernel_thread(void);
+
+void start_thread(struct pt_regs * regs, unsigned long pc, unsigned long sp)
+{
+	unsigned long status;
+
+	/* New thread loses kernel privileges. */
+	status = regs->cp0_status & ~(ST0_CU0|ST0_CU1|ST0_CU2|ST0_FR|KU_MASK);
+	status |= KU_USER;
+	regs->cp0_status = status;
+	lose_fpu(0);
+	clear_thread_flag(TIF_MSA_CTX_LIVE);
+	clear_used_math();
+#ifdef CONFIG_MIPS_FP_SUPPORT
+	atomic_set(&current->thread.bd_emu_frame, BD_EMUFRAME_NONE);
+#endif
+	init_dsp();
+	regs->cp0_epc = pc;
+	regs->regs[29] = sp;
+}
+
+void exit_thread(struct task_struct *tsk)
+{
+	/*
+	 * User threads may have allocated a delay slot emulation frame.
+	 * If so, clean up that allocation.
+	 */
+	if (!(current->flags & PF_KTHREAD))
+		dsemul_thread_cleanup(tsk);
+}
+
+int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
+{
+	/*
+	 * Save any process state which is live in hardware registers to the
+	 * parent context prior to duplication. This prevents the new child
+	 * state becoming stale if the parent is preempted before copy_thread()
+	 * gets a chance to save the parent's live hardware registers to the
+	 * child context.
+	 */
+	preempt_disable();
+
+	if (is_msa_enabled())
+		save_msa(current);
+	else if (is_fpu_owner())
+		_save_fp(current);
+
+	save_dsp(current);
+
+	preempt_enable();
+
+	*dst = *src;
+	return 0;
+}
+
+/*
+ * Copy architecture-specific thread state
+ */
+int copy_thread(unsigned long clone_flags, unsigned long usp,
+		unsigned long kthread_arg, struct task_struct *p,
+		unsigned long tls)
+{
+	struct thread_info *ti = task_thread_info(p);
+	struct pt_regs *childregs, *regs = current_pt_regs();
+	unsigned long childksp;
+
+	childksp = (unsigned long)task_stack_page(p) + THREAD_SIZE - 32;
+
+	/* set up new TSS. */
+	childregs = (struct pt_regs *) childksp - 1;
+	/*  Put the stack after the struct pt_regs.  */
+	childksp = (unsigned long) childregs;
+	p->thread.cp0_status = (read_c0_status() & ~(ST0_CU2|ST0_CU1)) | ST0_KERNEL_CUMASK;
+	if (unlikely(p->flags & (PF_KTHREAD | PF_IO_WORKER))) {
+		/* kernel thread */
+		unsigned long status = p->thread.cp0_status;
+		memset(childregs, 0, sizeof(struct pt_regs));
+		ti->addr_limit = KERNEL_DS;
+		p->thread.reg16 = usp; /* fn */
+		p->thread.reg17 = kthread_arg;
+		p->thread.reg29 = childksp;
+		p->thread.reg31 = (unsigned long) ret_from_kernel_thread;
+#if defined(CONFIG_CPU_R3000) || defined(CONFIG_CPU_TX39XX)
+		status = (status & ~(ST0_KUP | ST0_IEP | ST0_IEC)) |
+			 ((status & (ST0_KUC | ST0_IEC)) << 2);
+#else
+		status |= ST0_EXL;
+#endif
+		childregs->cp0_status = status;
+		return 0;
+	}
+
+	/* user thread */
+	*childregs = *regs;
+	childregs->regs[7] = 0; /* Clear error flag */
+	childregs->regs[2] = 0; /* Child gets zero as return value */
+	if (usp)
+		childregs->regs[29] = usp;
+	ti->addr_limit = USER_DS;
+
+	p->thread.reg29 = (unsigned long) childregs;
+	p->thread.reg31 = (unsigned long) ret_from_fork;
+
+	/*
+	 * New tasks lose permission to use the fpu. This accelerates context
+	 * switching for most programs since they don't use the fpu.
+	 */
+	childregs->cp0_status &= ~(ST0_CU2|ST0_CU1);
+
+	clear_tsk_thread_flag(p, TIF_USEDFPU);
+	clear_tsk_thread_flag(p, TIF_USEDMSA);
+	clear_tsk_thread_flag(p, TIF_MSA_CTX_LIVE);
+
+#ifdef CONFIG_MIPS_MT_FPAFF
+	clear_tsk_thread_flag(p, TIF_FPUBOUND);
+#endif /* CONFIG_MIPS_MT_FPAFF */
+
+#ifdef CONFIG_MIPS_FP_SUPPORT
+	atomic_set(&p->thread.bd_emu_frame, BD_EMUFRAME_NONE);
+#endif
+
+	if (clone_flags & CLONE_SETTLS)
+		ti->tp_value = tls;
+
+	return 0;
+}
+
+#ifdef CONFIG_STACKPROTECTOR
+#include <linux/stackprotector.h>
+unsigned long __stack_chk_guard __read_mostly;
+EXPORT_SYMBOL(__stack_chk_guard);
+#endif
+
+struct mips_frame_info {
+	void		*func;
+	unsigned long	func_size;
+	int		frame_size;
+	int		pc_offset;
+};
+
+#define J_TARGET(pc,target)	\
+		(((unsigned long)(pc) & 0xf0000000) | ((target) << 2))
+
+static inline int is_ra_save_ins(union mips_instruction *ip, int *poff)
+{
+#ifdef CONFIG_CPU_MICROMIPS
+	/*
+	 * swsp ra,offset
+	 * swm16 reglist,offset(sp)
+	 * swm32 reglist,offset(sp)
+	 * sw32 ra,offset(sp)
+	 * jradiussp - NOT SUPPORTED
+	 *
+	 * microMIPS is way more fun...
+	 */
+	if (mm_insn_16bit(ip->word >> 16)) {
+		switch (ip->mm16_r5_format.opcode) {
+		case mm_swsp16_op:
+			if (ip->mm16_r5_format.rt != 31)
+				return 0;
+
+			*poff = ip->mm16_r5_format.imm;
+			*poff = (*poff << 2) / sizeof(ulong);
+			return 1;
+
+		case mm_pool16c_op:
+			switch (ip->mm16_m_format.func) {
+			case mm_swm16_op:
+				*poff = ip->mm16_m_format.imm;
+				*poff += 1 + ip->mm16_m_format.rlist;
+				*poff = (*poff << 2) / sizeof(ulong);
+				return 1;
+
+			default:
+				return 0;
+			}
+
+		default:
+			return 0;
+		}
+	}
+
+	switch (ip->i_format.opcode) {
+	case mm_sw32_op:
+		if (ip->i_format.rs != 29)
+			return 0;
+		if (ip->i_format.rt != 31)
+			return 0;
+
+		*poff = ip->i_format.simmediate / sizeof(ulong);
+		return 1;
+
+	case mm_pool32b_op:
+		switch (ip->mm_m_format.func) {
+		case mm_swm32_func:
+			if (ip->mm_m_format.rd < 0x10)
+				return 0;
+			if (ip->mm_m_format.base != 29)
+				return 0;
+
+			*poff = ip->mm_m_format.simmediate;
+			*poff += (ip->mm_m_format.rd & 0xf) * sizeof(u32);
+			*poff /= sizeof(ulong);
+			return 1;
+		default:
+			return 0;
+		}
+
+	default:
+		return 0;
+	}
+#else
+	/* sw / sd $ra, offset($sp) */
+	if ((ip->i_format.opcode == sw_op || ip->i_format.opcode == sd_op) &&
+		ip->i_format.rs == 29 && ip->i_format.rt == 31) {
+		*poff = ip->i_format.simmediate / sizeof(ulong);
+		return 1;
+	}
+#ifdef CONFIG_CPU_LOONGSON64
+	if ((ip->loongson3_lswc2_format.opcode == swc2_op) &&
+		      (ip->loongson3_lswc2_format.ls == 1) &&
+		      (ip->loongson3_lswc2_format.fr == 0) &&
+		      (ip->loongson3_lswc2_format.base == 29)) {
+		if (ip->loongson3_lswc2_format.rt == 31) {
+			*poff = ip->loongson3_lswc2_format.offset << 1;
+			return 1;
+		}
+		if (ip->loongson3_lswc2_format.rq == 31) {
+			*poff = (ip->loongson3_lswc2_format.offset << 1) + 1;
+			return 1;
+		}
+	}
+#endif
+	return 0;
+#endif
+}
+
+static inline int is_jump_ins(union mips_instruction *ip)
+{
+#ifdef CONFIG_CPU_MICROMIPS
+	/*
+	 * jr16,jrc,jalr16,jalr16
+	 * jal
+	 * jalr/jr,jalr.hb/jr.hb,jalrs,jalrs.hb
+	 * jraddiusp - NOT SUPPORTED
+	 *
+	 * microMIPS is kind of more fun...
+	 */
+	if (mm_insn_16bit(ip->word >> 16)) {
+		if ((ip->mm16_r5_format.opcode == mm_pool16c_op &&
+		    (ip->mm16_r5_format.rt & mm_jr16_op) == mm_jr16_op))
+			return 1;
+		return 0;
+	}
+
+	if (ip->j_format.opcode == mm_j32_op)
+		return 1;
+	if (ip->j_format.opcode == mm_jal32_op)
+		return 1;
+	if (ip->r_format.opcode != mm_pool32a_op ||
+			ip->r_format.func != mm_pool32axf_op)
+		return 0;
+	return ((ip->u_format.uimmediate >> 6) & mm_jalr_op) == mm_jalr_op;
+#else
+	if (ip->j_format.opcode == j_op)
+		return 1;
+	if (ip->j_format.opcode == jal_op)
+		return 1;
+	if (ip->r_format.opcode != spec_op)
+		return 0;
+	return ip->r_format.func == jalr_op || ip->r_format.func == jr_op;
+#endif
+}
+
+static inline int is_sp_move_ins(union mips_instruction *ip, int *frame_size)
+{
+#ifdef CONFIG_CPU_MICROMIPS
+	unsigned short tmp;
+
+	/*
+	 * addiusp -imm
+	 * addius5 sp,-imm
+	 * addiu32 sp,sp,-imm
+	 * jradiussp - NOT SUPPORTED
+	 *
+	 * microMIPS is not more fun...
+	 */
+	if (mm_insn_16bit(ip->word >> 16)) {
+		if (ip->mm16_r3_format.opcode == mm_pool16d_op &&
+		    ip->mm16_r3_format.simmediate & mm_addiusp_func) {
+			tmp = ip->mm_b0_format.simmediate >> 1;
+			tmp = ((tmp & 0x1ff) ^ 0x100) - 0x100;
+			if ((tmp + 2) < 4) /* 0x0,0x1,0x1fe,0x1ff are special */
+				tmp ^= 0x100;
+			*frame_size = -(signed short)(tmp << 2);
+			return 1;
+		}
+		if (ip->mm16_r5_format.opcode == mm_pool16d_op &&
+		    ip->mm16_r5_format.rt == 29) {
+			tmp = ip->mm16_r5_format.imm >> 1;
+			*frame_size = -(signed short)(tmp & 0xf);
+			return 1;
+		}
+		return 0;
+	}
+
+	if (ip->mm_i_format.opcode == mm_addiu32_op &&
+	    ip->mm_i_format.rt == 29 && ip->mm_i_format.rs == 29) {
+		*frame_size = -ip->i_format.simmediate;
+		return 1;
+	}
+#else
+	/* addiu/daddiu sp,sp,-imm */
+	if (ip->i_format.rs != 29 || ip->i_format.rt != 29)
+		return 0;
+
+	if (ip->i_format.opcode == addiu_op ||
+	    ip->i_format.opcode == daddiu_op) {
+		*frame_size = -ip->i_format.simmediate;
+		return 1;
+	}
+#endif
+	return 0;
+}
+
+static int get_frame_info(struct mips_frame_info *info)
+{
+	bool is_mmips = IS_ENABLED(CONFIG_CPU_MICROMIPS);
+	union mips_instruction insn, *ip;
+	const unsigned int max_insns = 128;
+	unsigned int last_insn_size = 0;
+	unsigned int i;
+	bool saw_jump = false;
+
+	info->pc_offset = -1;
+	info->frame_size = 0;
+
+	ip = (void *)msk_isa16_mode((ulong)info->func);
+	if (!ip)
+		goto err;
+
+	for (i = 0; i < max_insns; i++) {
+		ip = (void *)ip + last_insn_size;
+
+		if (is_mmips && mm_insn_16bit(ip->halfword[0])) {
+			insn.word = ip->halfword[0] << 16;
+			last_insn_size = 2;
+		} else if (is_mmips) {
+			insn.word = ip->halfword[0] << 16 | ip->halfword[1];
+			last_insn_size = 4;
+		} else {
+			insn.word = ip->word;
+			last_insn_size = 4;
+		}
+
+		if (!info->frame_size) {
+			is_sp_move_ins(&insn, &info->frame_size);
+			continue;
+		} else if (!saw_jump && is_jump_ins(ip)) {
+			/*
+			 * If we see a jump instruction, we are finished
+			 * with the frame save.
+			 *
+			 * Some functions can have a shortcut return at
+			 * the beginning of the function, so don't start
+			 * looking for jump instruction until we see the
+			 * frame setup.
+			 *
+			 * The RA save instruction can get put into the
+			 * delay slot of the jump instruction, so look
+			 * at the next instruction, too.
+			 */
+			saw_jump = true;
+			continue;
+		}
+		if (info->pc_offset == -1 &&
+		    is_ra_save_ins(&insn, &info->pc_offset))
+			break;
+		if (saw_jump)
+			break;
+	}
+	if (info->frame_size && info->pc_offset >= 0) /* nested */
+		return 0;
+	if (info->pc_offset < 0) /* leaf */
+		return 1;
+	/* prologue seems bogus... */
+err:
+	return -1;
+}
+
+static struct mips_frame_info schedule_mfi __read_mostly;
+
+#ifdef CONFIG_KALLSYMS
+static unsigned long get___schedule_addr(void)
+{
+	return kallsyms_lookup_name("__schedule");
+}
+#else
+static unsigned long get___schedule_addr(void)
+{
+	union mips_instruction *ip = (void *)schedule;
+	int max_insns = 8;
+	int i;
+
+	for (i = 0; i < max_insns; i++, ip++) {
+		if (ip->j_format.opcode == j_op)
+			return J_TARGET(ip, ip->j_format.target);
+	}
+	return 0;
+}
+#endif
+
+static int __init frame_info_init(void)
+{
+	unsigned long size = 0;
+#ifdef CONFIG_KALLSYMS
+	unsigned long ofs;
+#endif
+	unsigned long addr;
+
+	addr = get___schedule_addr();
+	if (!addr)
+		addr = (unsigned long)schedule;
+
+#ifdef CONFIG_KALLSYMS
+	kallsyms_lookup_size_offset(addr, &size, &ofs);
+#endif
+	schedule_mfi.func = (void *)addr;
+	schedule_mfi.func_size = size;
+
+	get_frame_info(&schedule_mfi);
+
+	/*
+	 * Without schedule() frame info, result given by
+	 * thread_saved_pc() and get_wchan() are not reliable.
+	 */
+	if (schedule_mfi.pc_offset < 0)
+		printk("Can't analyze schedule() prologue at %p\n", schedule);
+
+	return 0;
+}
+
+arch_initcall(frame_info_init);
+
+/*
+ * Return saved PC of a blocked thread.
+ */
+static unsigned long thread_saved_pc(struct task_struct *tsk)
+{
+	struct thread_struct *t = &tsk->thread;
+
+	/* New born processes are a special case */
+	if (t->reg31 == (unsigned long) ret_from_fork)
+		return t->reg31;
+	if (schedule_mfi.pc_offset < 0)
+		return 0;
+	return ((unsigned long *)t->reg29)[schedule_mfi.pc_offset];
+}
+
+
+#ifdef CONFIG_KALLSYMS
+/* generic stack unwinding function */
+unsigned long notrace unwind_stack_by_address(unsigned long stack_page,
+					      unsigned long *sp,
+					      unsigned long pc,
+					      unsigned long *ra)
+{
+	unsigned long low, high, irq_stack_high;
+	struct mips_frame_info info;
+	unsigned long size, ofs;
+	struct pt_regs *regs;
+	int leaf;
+
+	if (!stack_page)
+		return 0;
+
+	/*
+	 * IRQ stacks start at IRQ_STACK_START
+	 * task stacks at THREAD_SIZE - 32
+	 */
+	low = stack_page;
+	if (!preemptible() && on_irq_stack(raw_smp_processor_id(), *sp)) {
+		high = stack_page + IRQ_STACK_START;
+		irq_stack_high = high;
+	} else {
+		high = stack_page + THREAD_SIZE - 32;
+		irq_stack_high = 0;
+	}
+
+	/*
+	 * If we reached the top of the interrupt stack, start unwinding
+	 * the interrupted task stack.
+	 */
+	if (unlikely(*sp == irq_stack_high)) {
+		unsigned long task_sp = *(unsigned long *)*sp;
+
+		/*
+		 * Check that the pointer saved in the IRQ stack head points to
+		 * something within the stack of the current task
+		 */
+		if (!object_is_on_stack((void *)task_sp))
+			return 0;
+
+		/*
+		 * Follow pointer to tasks kernel stack frame where interrupted
+		 * state was saved.
+		 */
+		regs = (struct pt_regs *)task_sp;
+		pc = regs->cp0_epc;
+		if (!user_mode(regs) && __kernel_text_address(pc)) {
+			*sp = regs->regs[29];
+			*ra = regs->regs[31];
+			return pc;
+		}
+		return 0;
+	}
+	if (!kallsyms_lookup_size_offset(pc, &size, &ofs))
+		return 0;
+	/*
+	 * Return ra if an exception occurred at the first instruction
+	 */
+	if (unlikely(ofs == 0)) {
+		pc = *ra;
+		*ra = 0;
+		return pc;
+	}
+
+	info.func = (void *)(pc - ofs);
+	info.func_size = ofs;	/* analyze from start to ofs */
+	leaf = get_frame_info(&info);
+	if (leaf < 0)
+		return 0;
+
+	if (*sp < low || *sp + info.frame_size > high)
+		return 0;
+
+	if (leaf)
+		/*
+		 * For some extreme cases, get_frame_info() can
+		 * consider wrongly a nested function as a leaf
+		 * one. In that cases avoid to return always the
+		 * same value.
+		 */
+		pc = pc != *ra ? *ra : 0;
+	else
+		pc = ((unsigned long *)(*sp))[info.pc_offset];
+
+	*sp += info.frame_size;
+	*ra = 0;
+	return __kernel_text_address(pc) ? pc : 0;
+}
+EXPORT_SYMBOL(unwind_stack_by_address);
+
+/* used by show_backtrace() */
+unsigned long unwind_stack(struct task_struct *task, unsigned long *sp,
+			   unsigned long pc, unsigned long *ra)
+{
+	unsigned long stack_page = 0;
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		if (on_irq_stack(cpu, *sp)) {
+			stack_page = (unsigned long)irq_stack[cpu];
+			break;
+		}
+	}
+
+	if (!stack_page)
+		stack_page = (unsigned long)task_stack_page(task);
+
+	return unwind_stack_by_address(stack_page, sp, pc, ra);
+}
+#endif
+
+/*
+ * get_wchan - a maintenance nightmare^W^Wpain in the ass ...
+ */
+unsigned long get_wchan(struct task_struct *task)
+{
+	unsigned long pc = 0;
+#ifdef CONFIG_KALLSYMS
+	unsigned long sp;
+	unsigned long ra = 0;
+#endif
+
+	if (!task || task == current || task->state == TASK_RUNNING)
+		goto out;
+	if (!task_stack_page(task))
+		goto out;
+
+	pc = thread_saved_pc(task);
+
+#ifdef CONFIG_KALLSYMS
+	sp = task->thread.reg29 + schedule_mfi.frame_size;
+
+	while (in_sched_functions(pc))
+		pc = unwind_stack(task, &sp, pc, &ra);
+#endif
+
+out:
+	return pc;
+}
+
+unsigned long mips_stack_top(void)
+{
+	unsigned long top = TASK_SIZE & PAGE_MASK;
+
+	if (IS_ENABLED(CONFIG_MIPS_FP_SUPPORT)) {
+		/* One page for branch delay slot "emulation" */
+		top -= PAGE_SIZE;
+	}
+
+	/* Space for the VDSO, data page & GIC user page */
+	top -= PAGE_ALIGN(current->thread.abi->vdso->size);
+	top -= PAGE_SIZE;
+	top -= mips_gic_present() ? PAGE_SIZE : 0;
+
+	/* Space for cache colour alignment */
+	if (cpu_has_dc_aliases)
+		top -= shm_align_mask + 1;
+
+	/* Space to randomize the VDSO base */
+	if (current->flags & PF_RANDOMIZE)
+		top -= VDSO_RANDOMIZE_SIZE;
+
+	return top;
+}
+
+/*
+ * Don't forget that the stack pointer must be aligned on a 8 bytes
+ * boundary for 32-bits ABI and 16 bytes for 64-bits ABI.
+ */
+unsigned long arch_align_stack(unsigned long sp)
+{
+	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
+		sp -= get_random_int() & ~PAGE_MASK;
+
+	return sp & ALMASK;
+}
+
+static DEFINE_PER_CPU(call_single_data_t, backtrace_csd);
+static struct cpumask backtrace_csd_busy;
+
+static void handle_backtrace(void *info)
+{
+	nmi_cpu_backtrace(get_irq_regs());
+	cpumask_clear_cpu(smp_processor_id(), &backtrace_csd_busy);
+}
+
+static void raise_backtrace(cpumask_t *mask)
+{
+	call_single_data_t *csd;
+	int cpu;
+
+	for_each_cpu(cpu, mask) {
+		/*
+		 * If we previously sent an IPI to the target CPU & it hasn't
+		 * cleared its bit in the busy cpumask then it didn't handle
+		 * our previous IPI & it's not safe for us to reuse the
+		 * call_single_data_t.
+		 */
+		if (cpumask_test_and_set_cpu(cpu, &backtrace_csd_busy)) {
+			pr_warn("Unable to send backtrace IPI to CPU%u - perhaps it hung?\n",
+				cpu);
+			continue;
+		}
+
+		csd = &per_cpu(backtrace_csd, cpu);
+		csd->func = handle_backtrace;
+		smp_call_function_single_async(cpu, csd);
+	}
+}
+
+void arch_trigger_cpumask_backtrace(const cpumask_t *mask, bool exclude_self)
+{
+	nmi_trigger_cpumask_backtrace(mask, exclude_self, raise_backtrace);
+}
+
+int mips_get_process_fp_mode(struct task_struct *task)
+{
+	int value = 0;
+
+	if (!test_tsk_thread_flag(task, TIF_32BIT_FPREGS))
+		value |= PR_FP_MODE_FR;
+	if (test_tsk_thread_flag(task, TIF_HYBRID_FPREGS))
+		value |= PR_FP_MODE_FRE;
+
+	return value;
+}
+
+static long prepare_for_fp_mode_switch(void *unused)
+{
+	/*
+	 * This is icky, but we use this to simply ensure that all CPUs have
+	 * context switched, regardless of whether they were previously running
+	 * kernel or user code. This ensures that no CPU that a mode-switching
+	 * program may execute on keeps its FPU enabled (& in the old mode)
+	 * throughout the mode switch.
+	 */
+	return 0;
+}
+
+int mips_set_process_fp_mode(struct task_struct *task, unsigned int value)
+{
+	const unsigned int known_bits = PR_FP_MODE_FR | PR_FP_MODE_FRE;
+	struct task_struct *t;
+	struct cpumask process_cpus;
+	int cpu;
+
+	/* If nothing to change, return right away, successfully.  */
+	if (value == mips_get_process_fp_mode(task))
+		return 0;
+
+	/* Only accept a mode change if 64-bit FP enabled for o32.  */
+	if (!IS_ENABLED(CONFIG_MIPS_O32_FP64_SUPPORT))
+		return -EOPNOTSUPP;
+
+	/* And only for o32 tasks.  */
+	if (IS_ENABLED(CONFIG_64BIT) && !test_thread_flag(TIF_32BIT_REGS))
+		return -EOPNOTSUPP;
+
+	/* Check the value is valid */
+	if (value & ~known_bits)
+		return -EOPNOTSUPP;
+
+	/* Setting FRE without FR is not supported.  */
+	if ((value & (PR_FP_MODE_FR | PR_FP_MODE_FRE)) == PR_FP_MODE_FRE)
+		return -EOPNOTSUPP;
+
+	/* Avoid inadvertently triggering emulation */
+	if ((value & PR_FP_MODE_FR) && raw_cpu_has_fpu &&
+	    !(raw_current_cpu_data.fpu_id & MIPS_FPIR_F64))
+		return -EOPNOTSUPP;
+	if ((value & PR_FP_MODE_FRE) && raw_cpu_has_fpu && !cpu_has_fre)
+		return -EOPNOTSUPP;
+
+	/* FR = 0 not supported in MIPS R6 */
+	if (!(value & PR_FP_MODE_FR) && raw_cpu_has_fpu && cpu_has_mips_r6)
+		return -EOPNOTSUPP;
+
+	/* Indicate the new FP mode in each thread */
+	for_each_thread(task, t) {
+		/* Update desired FP register width */
+		if (value & PR_FP_MODE_FR) {
+			clear_tsk_thread_flag(t, TIF_32BIT_FPREGS);
+		} else {
+			set_tsk_thread_flag(t, TIF_32BIT_FPREGS);
+			clear_tsk_thread_flag(t, TIF_MSA_CTX_LIVE);
+		}
+
+		/* Update desired FP single layout */
+		if (value & PR_FP_MODE_FRE)
+			set_tsk_thread_flag(t, TIF_HYBRID_FPREGS);
+		else
+			clear_tsk_thread_flag(t, TIF_HYBRID_FPREGS);
+	}
+
+	/*
+	 * We need to ensure that all threads in the process have switched mode
+	 * before returning, in order to allow userland to not worry about
+	 * races. We can do this by forcing all CPUs that any thread in the
+	 * process may be running on to schedule something else - in this case
+	 * prepare_for_fp_mode_switch().
+	 *
+	 * We begin by generating a mask of all CPUs that any thread in the
+	 * process may be running on.
+	 */
+	cpumask_clear(&process_cpus);
+	for_each_thread(task, t)
+		cpumask_set_cpu(task_cpu(t), &process_cpus);
+
+	/*
+	 * Now we schedule prepare_for_fp_mode_switch() on each of those CPUs.
+	 *
+	 * The CPUs may have rescheduled already since we switched mode or
+	 * generated the cpumask, but that doesn't matter. If the task in this
+	 * process is scheduled out then our scheduling
+	 * prepare_for_fp_mode_switch() will simply be redundant. If it's
+	 * scheduled in then it will already have picked up the new FP mode
+	 * whilst doing so.
+	 */
+	get_online_cpus();
+	for_each_cpu_and(cpu, &process_cpus, cpu_online_mask)
+		work_on_cpu(cpu, prepare_for_fp_mode_switch, NULL);
+	put_online_cpus();
+
+	return 0;
+}
+
+#if defined(CONFIG_32BIT) || defined(CONFIG_MIPS32_O32)
+void mips_dump_regs32(u32 *uregs, const struct pt_regs *regs)
+{
+	unsigned int i;
+
+	for (i = MIPS32_EF_R1; i <= MIPS32_EF_R31; i++) {
+		/* k0/k1 are copied as zero. */
+		if (i == MIPS32_EF_R26 || i == MIPS32_EF_R27)
+			uregs[i] = 0;
+		else
+			uregs[i] = regs->regs[i - MIPS32_EF_R0];
+	}
+
+	uregs[MIPS32_EF_LO] = regs->lo;
+	uregs[MIPS32_EF_HI] = regs->hi;
+	uregs[MIPS32_EF_CP0_EPC] = regs->cp0_epc;
+	uregs[MIPS32_EF_CP0_BADVADDR] = regs->cp0_badvaddr;
+	uregs[MIPS32_EF_CP0_STATUS] = regs->cp0_status;
+	uregs[MIPS32_EF_CP0_CAUSE] = regs->cp0_cause;
+}
+#endif /* CONFIG_32BIT || CONFIG_MIPS32_O32 */
+
+#ifdef CONFIG_64BIT
+void mips_dump_regs64(u64 *uregs, const struct pt_regs *regs)
+{
+	unsigned int i;
+
+	for (i = MIPS64_EF_R1; i <= MIPS64_EF_R31; i++) {
+		/* k0/k1 are copied as zero. */
+		if (i == MIPS64_EF_R26 || i == MIPS64_EF_R27)
+			uregs[i] = 0;
+		else
+			uregs[i] = regs->regs[i - MIPS64_EF_R0];
+	}
+
+	uregs[MIPS64_EF_LO] = regs->lo;
+	uregs[MIPS64_EF_HI] = regs->hi;
+	uregs[MIPS64_EF_CP0_EPC] = regs->cp0_epc;
+	uregs[MIPS64_EF_CP0_BADVADDR] = regs->cp0_badvaddr;
+	uregs[MIPS64_EF_CP0_STATUS] = regs->cp0_status;
+	uregs[MIPS64_EF_CP0_CAUSE] = regs->cp0_cause;
+}
+#endif /* CONFIG_64BIT */
diff --git a/arch/mips/kernel/prom.c b/arch/mips/kernel/prom.c
new file mode 100644
index 000000000..6abebd57b
--- /dev/null
+++ b/arch/mips/kernel/prom.c
@@ -0,0 +1,67 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * MIPS support for CONFIG_OF device tree support
+ *
+ * Copyright (C) 2010 Cisco Systems Inc. <dediao@cisco.com>
+ */
+
+#include <linux/init.h>
+#include <linux/export.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/memblock.h>
+#include <linux/debugfs.h>
+#include <linux/of.h>
+#include <linux/of_fdt.h>
+#include <linux/of_platform.h>
+
+#include <asm/bootinfo.h>
+#include <asm/page.h>
+#include <asm/prom.h>
+
+static char mips_machine_name[64] = "Unknown";
+
+__init void mips_set_machine_name(const char *name)
+{
+	if (name == NULL)
+		return;
+
+	strlcpy(mips_machine_name, name, sizeof(mips_machine_name));
+	pr_info("MIPS: machine is %s\n", mips_get_machine_name());
+}
+
+char *mips_get_machine_name(void)
+{
+	return mips_machine_name;
+}
+
+#ifdef CONFIG_USE_OF
+
+void __init __dt_setup_arch(void *bph)
+{
+	if (!early_init_dt_scan(bph))
+		return;
+
+	mips_set_machine_name(of_flat_dt_get_machine_name());
+}
+
+int __init __dt_register_buses(const char *bus0, const char *bus1)
+{
+	static struct of_device_id of_ids[3];
+
+	if (!of_have_populated_dt())
+		panic("device tree not present");
+
+	strlcpy(of_ids[0].compatible, bus0, sizeof(of_ids[0].compatible));
+	if (bus1) {
+		strlcpy(of_ids[1].compatible, bus1,
+			sizeof(of_ids[1].compatible));
+	}
+
+	if (of_platform_populate(NULL, of_ids, NULL, NULL))
+		panic("failed to populate DT");
+
+	return 0;
+}
+
+#endif
diff --git a/arch/mips/kernel/ptrace.c b/arch/mips/kernel/ptrace.c
new file mode 100644
index 000000000..db7c5be1d
--- /dev/null
+++ b/arch/mips/kernel/ptrace.c
@@ -0,0 +1,1382 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1992 Ross Biro
+ * Copyright (C) Linus Torvalds
+ * Copyright (C) 1994, 95, 96, 97, 98, 2000 Ralf Baechle
+ * Copyright (C) 1996 David S. Miller
+ * Kevin D. Kissell, kevink@mips.com and Carsten Langgaard, carstenl@mips.com
+ * Copyright (C) 1999 MIPS Technologies, Inc.
+ * Copyright (C) 2000 Ulf Carlsson
+ *
+ * At this time Linux/MIPS64 only supports syscall tracing, even for 32-bit
+ * binaries.
+ */
+#include <linux/compiler.h>
+#include <linux/context_tracking.h>
+#include <linux/elf.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/sched/task_stack.h>
+#include <linux/mm.h>
+#include <linux/errno.h>
+#include <linux/ptrace.h>
+#include <linux/regset.h>
+#include <linux/smp.h>
+#include <linux/security.h>
+#include <linux/stddef.h>
+#include <linux/tracehook.h>
+#include <linux/audit.h>
+#include <linux/seccomp.h>
+#include <linux/ftrace.h>
+
+#include <asm/byteorder.h>
+#include <asm/cpu.h>
+#include <asm/cpu-info.h>
+#include <asm/dsp.h>
+#include <asm/fpu.h>
+#include <asm/mipsregs.h>
+#include <asm/mipsmtregs.h>
+#include <asm/page.h>
+#include <asm/processor.h>
+#include <asm/syscall.h>
+#include <linux/uaccess.h>
+#include <asm/bootinfo.h>
+#include <asm/reg.h>
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/syscalls.h>
+
+/*
+ * Called by kernel/ptrace.c when detaching..
+ *
+ * Make sure single step bits etc are not set.
+ */
+void ptrace_disable(struct task_struct *child)
+{
+	/* Don't load the watchpoint registers for the ex-child. */
+	clear_tsk_thread_flag(child, TIF_LOAD_WATCH);
+}
+
+/*
+ * Read a general register set.	 We always use the 64-bit format, even
+ * for 32-bit kernels and for 32-bit processes on a 64-bit kernel.
+ * Registers are sign extended to fill the available space.
+ */
+int ptrace_getregs(struct task_struct *child, struct user_pt_regs __user *data)
+{
+	struct pt_regs *regs;
+	int i;
+
+	if (!access_ok(data, 38 * 8))
+		return -EIO;
+
+	regs = task_pt_regs(child);
+
+	for (i = 0; i < 32; i++)
+		__put_user((long)regs->regs[i], (__s64 __user *)&data->regs[i]);
+	__put_user((long)regs->lo, (__s64 __user *)&data->lo);
+	__put_user((long)regs->hi, (__s64 __user *)&data->hi);
+	__put_user((long)regs->cp0_epc, (__s64 __user *)&data->cp0_epc);
+	__put_user((long)regs->cp0_badvaddr, (__s64 __user *)&data->cp0_badvaddr);
+	__put_user((long)regs->cp0_status, (__s64 __user *)&data->cp0_status);
+	__put_user((long)regs->cp0_cause, (__s64 __user *)&data->cp0_cause);
+
+	return 0;
+}
+
+/*
+ * Write a general register set.  As for PTRACE_GETREGS, we always use
+ * the 64-bit format.  On a 32-bit kernel only the lower order half
+ * (according to endianness) will be used.
+ */
+int ptrace_setregs(struct task_struct *child, struct user_pt_regs __user *data)
+{
+	struct pt_regs *regs;
+	int i;
+
+	if (!access_ok(data, 38 * 8))
+		return -EIO;
+
+	regs = task_pt_regs(child);
+
+	for (i = 0; i < 32; i++)
+		__get_user(regs->regs[i], (__s64 __user *)&data->regs[i]);
+	__get_user(regs->lo, (__s64 __user *)&data->lo);
+	__get_user(regs->hi, (__s64 __user *)&data->hi);
+	__get_user(regs->cp0_epc, (__s64 __user *)&data->cp0_epc);
+
+	/* badvaddr, status, and cause may not be written.  */
+
+	/* System call number may have been changed */
+	mips_syscall_update_nr(child, regs);
+
+	return 0;
+}
+
+int ptrace_get_watch_regs(struct task_struct *child,
+			  struct pt_watch_regs __user *addr)
+{
+	enum pt_watch_style style;
+	int i;
+
+	if (!cpu_has_watch || boot_cpu_data.watch_reg_use_cnt == 0)
+		return -EIO;
+	if (!access_ok(addr, sizeof(struct pt_watch_regs)))
+		return -EIO;
+
+#ifdef CONFIG_32BIT
+	style = pt_watch_style_mips32;
+#define WATCH_STYLE mips32
+#else
+	style = pt_watch_style_mips64;
+#define WATCH_STYLE mips64
+#endif
+
+	__put_user(style, &addr->style);
+	__put_user(boot_cpu_data.watch_reg_use_cnt,
+		   &addr->WATCH_STYLE.num_valid);
+	for (i = 0; i < boot_cpu_data.watch_reg_use_cnt; i++) {
+		__put_user(child->thread.watch.mips3264.watchlo[i],
+			   &addr->WATCH_STYLE.watchlo[i]);
+		__put_user(child->thread.watch.mips3264.watchhi[i] &
+				(MIPS_WATCHHI_MASK | MIPS_WATCHHI_IRW),
+			   &addr->WATCH_STYLE.watchhi[i]);
+		__put_user(boot_cpu_data.watch_reg_masks[i],
+			   &addr->WATCH_STYLE.watch_masks[i]);
+	}
+	for (; i < 8; i++) {
+		__put_user(0, &addr->WATCH_STYLE.watchlo[i]);
+		__put_user(0, &addr->WATCH_STYLE.watchhi[i]);
+		__put_user(0, &addr->WATCH_STYLE.watch_masks[i]);
+	}
+
+	return 0;
+}
+
+int ptrace_set_watch_regs(struct task_struct *child,
+			  struct pt_watch_regs __user *addr)
+{
+	int i;
+	int watch_active = 0;
+	unsigned long lt[NUM_WATCH_REGS];
+	u16 ht[NUM_WATCH_REGS];
+
+	if (!cpu_has_watch || boot_cpu_data.watch_reg_use_cnt == 0)
+		return -EIO;
+	if (!access_ok(addr, sizeof(struct pt_watch_regs)))
+		return -EIO;
+	/* Check the values. */
+	for (i = 0; i < boot_cpu_data.watch_reg_use_cnt; i++) {
+		__get_user(lt[i], &addr->WATCH_STYLE.watchlo[i]);
+#ifdef CONFIG_32BIT
+		if (lt[i] & __UA_LIMIT)
+			return -EINVAL;
+#else
+		if (test_tsk_thread_flag(child, TIF_32BIT_ADDR)) {
+			if (lt[i] & 0xffffffff80000000UL)
+				return -EINVAL;
+		} else {
+			if (lt[i] & __UA_LIMIT)
+				return -EINVAL;
+		}
+#endif
+		__get_user(ht[i], &addr->WATCH_STYLE.watchhi[i]);
+		if (ht[i] & ~MIPS_WATCHHI_MASK)
+			return -EINVAL;
+	}
+	/* Install them. */
+	for (i = 0; i < boot_cpu_data.watch_reg_use_cnt; i++) {
+		if (lt[i] & MIPS_WATCHLO_IRW)
+			watch_active = 1;
+		child->thread.watch.mips3264.watchlo[i] = lt[i];
+		/* Set the G bit. */
+		child->thread.watch.mips3264.watchhi[i] = ht[i];
+	}
+
+	if (watch_active)
+		set_tsk_thread_flag(child, TIF_LOAD_WATCH);
+	else
+		clear_tsk_thread_flag(child, TIF_LOAD_WATCH);
+
+	return 0;
+}
+
+/* regset get/set implementations */
+
+#if defined(CONFIG_32BIT) || defined(CONFIG_MIPS32_O32)
+
+static int gpr32_get(struct task_struct *target,
+		     const struct user_regset *regset,
+		     struct membuf to)
+{
+	struct pt_regs *regs = task_pt_regs(target);
+	u32 uregs[ELF_NGREG] = {};
+
+	mips_dump_regs32(uregs, regs);
+	return membuf_write(&to, uregs, sizeof(uregs));
+}
+
+static int gpr32_set(struct task_struct *target,
+		     const struct user_regset *regset,
+		     unsigned int pos, unsigned int count,
+		     const void *kbuf, const void __user *ubuf)
+{
+	struct pt_regs *regs = task_pt_regs(target);
+	u32 uregs[ELF_NGREG];
+	unsigned start, num_regs, i;
+	int err;
+
+	start = pos / sizeof(u32);
+	num_regs = count / sizeof(u32);
+
+	if (start + num_regs > ELF_NGREG)
+		return -EIO;
+
+	err = user_regset_copyin(&pos, &count, &kbuf, &ubuf, uregs, 0,
+				 sizeof(uregs));
+	if (err)
+		return err;
+
+	for (i = start; i < num_regs; i++) {
+		/*
+		 * Cast all values to signed here so that if this is a 64-bit
+		 * kernel, the supplied 32-bit values will be sign extended.
+		 */
+		switch (i) {
+		case MIPS32_EF_R1 ... MIPS32_EF_R25:
+			/* k0/k1 are ignored. */
+		case MIPS32_EF_R28 ... MIPS32_EF_R31:
+			regs->regs[i - MIPS32_EF_R0] = (s32)uregs[i];
+			break;
+		case MIPS32_EF_LO:
+			regs->lo = (s32)uregs[i];
+			break;
+		case MIPS32_EF_HI:
+			regs->hi = (s32)uregs[i];
+			break;
+		case MIPS32_EF_CP0_EPC:
+			regs->cp0_epc = (s32)uregs[i];
+			break;
+		}
+	}
+
+	/* System call number may have been changed */
+	mips_syscall_update_nr(target, regs);
+
+	return 0;
+}
+
+#endif /* CONFIG_32BIT || CONFIG_MIPS32_O32 */
+
+#ifdef CONFIG_64BIT
+
+static int gpr64_get(struct task_struct *target,
+		     const struct user_regset *regset,
+		     struct membuf to)
+{
+	struct pt_regs *regs = task_pt_regs(target);
+	u64 uregs[ELF_NGREG] = {};
+
+	mips_dump_regs64(uregs, regs);
+	return membuf_write(&to, uregs, sizeof(uregs));
+}
+
+static int gpr64_set(struct task_struct *target,
+		     const struct user_regset *regset,
+		     unsigned int pos, unsigned int count,
+		     const void *kbuf, const void __user *ubuf)
+{
+	struct pt_regs *regs = task_pt_regs(target);
+	u64 uregs[ELF_NGREG];
+	unsigned start, num_regs, i;
+	int err;
+
+	start = pos / sizeof(u64);
+	num_regs = count / sizeof(u64);
+
+	if (start + num_regs > ELF_NGREG)
+		return -EIO;
+
+	err = user_regset_copyin(&pos, &count, &kbuf, &ubuf, uregs, 0,
+				 sizeof(uregs));
+	if (err)
+		return err;
+
+	for (i = start; i < num_regs; i++) {
+		switch (i) {
+		case MIPS64_EF_R1 ... MIPS64_EF_R25:
+			/* k0/k1 are ignored. */
+		case MIPS64_EF_R28 ... MIPS64_EF_R31:
+			regs->regs[i - MIPS64_EF_R0] = uregs[i];
+			break;
+		case MIPS64_EF_LO:
+			regs->lo = uregs[i];
+			break;
+		case MIPS64_EF_HI:
+			regs->hi = uregs[i];
+			break;
+		case MIPS64_EF_CP0_EPC:
+			regs->cp0_epc = uregs[i];
+			break;
+		}
+	}
+
+	/* System call number may have been changed */
+	mips_syscall_update_nr(target, regs);
+
+	return 0;
+}
+
+#endif /* CONFIG_64BIT */
+
+
+#ifdef CONFIG_MIPS_FP_SUPPORT
+
+/*
+ * Poke at FCSR according to its mask.  Set the Cause bits even
+ * if a corresponding Enable bit is set.  This will be noticed at
+ * the time the thread is switched to and SIGFPE thrown accordingly.
+ */
+static void ptrace_setfcr31(struct task_struct *child, u32 value)
+{
+	u32 fcr31;
+	u32 mask;
+
+	fcr31 = child->thread.fpu.fcr31;
+	mask = boot_cpu_data.fpu_msk31;
+	child->thread.fpu.fcr31 = (value & ~mask) | (fcr31 & mask);
+}
+
+int ptrace_getfpregs(struct task_struct *child, __u32 __user *data)
+{
+	int i;
+
+	if (!access_ok(data, 33 * 8))
+		return -EIO;
+
+	if (tsk_used_math(child)) {
+		union fpureg *fregs = get_fpu_regs(child);
+		for (i = 0; i < 32; i++)
+			__put_user(get_fpr64(&fregs[i], 0),
+				   i + (__u64 __user *)data);
+	} else {
+		for (i = 0; i < 32; i++)
+			__put_user((__u64) -1, i + (__u64 __user *) data);
+	}
+
+	__put_user(child->thread.fpu.fcr31, data + 64);
+	__put_user(boot_cpu_data.fpu_id, data + 65);
+
+	return 0;
+}
+
+int ptrace_setfpregs(struct task_struct *child, __u32 __user *data)
+{
+	union fpureg *fregs;
+	u64 fpr_val;
+	u32 value;
+	int i;
+
+	if (!access_ok(data, 33 * 8))
+		return -EIO;
+
+	init_fp_ctx(child);
+	fregs = get_fpu_regs(child);
+
+	for (i = 0; i < 32; i++) {
+		__get_user(fpr_val, i + (__u64 __user *)data);
+		set_fpr64(&fregs[i], 0, fpr_val);
+	}
+
+	__get_user(value, data + 64);
+	ptrace_setfcr31(child, value);
+
+	/* FIR may not be written.  */
+
+	return 0;
+}
+
+/*
+ * Copy the floating-point context to the supplied NT_PRFPREG buffer,
+ * !CONFIG_CPU_HAS_MSA variant.  FP context's general register slots
+ * correspond 1:1 to buffer slots.  Only general registers are copied.
+ */
+static void fpr_get_fpa(struct task_struct *target,
+		       struct membuf *to)
+{
+	membuf_write(to, &target->thread.fpu,
+			NUM_FPU_REGS * sizeof(elf_fpreg_t));
+}
+
+/*
+ * Copy the floating-point context to the supplied NT_PRFPREG buffer,
+ * CONFIG_CPU_HAS_MSA variant.  Only lower 64 bits of FP context's
+ * general register slots are copied to buffer slots.  Only general
+ * registers are copied.
+ */
+static void fpr_get_msa(struct task_struct *target, struct membuf *to)
+{
+	unsigned int i;
+
+	BUILD_BUG_ON(sizeof(u64) != sizeof(elf_fpreg_t));
+	for (i = 0; i < NUM_FPU_REGS; i++)
+		membuf_store(to, get_fpr64(&target->thread.fpu.fpr[i], 0));
+}
+
+/*
+ * Copy the floating-point context to the supplied NT_PRFPREG buffer.
+ * Choose the appropriate helper for general registers, and then copy
+ * the FCSR and FIR registers separately.
+ */
+static int fpr_get(struct task_struct *target,
+		   const struct user_regset *regset,
+		   struct membuf to)
+{
+	if (sizeof(target->thread.fpu.fpr[0]) == sizeof(elf_fpreg_t))
+		fpr_get_fpa(target, &to);
+	else
+		fpr_get_msa(target, &to);
+
+	membuf_write(&to, &target->thread.fpu.fcr31, sizeof(u32));
+	membuf_write(&to, &boot_cpu_data.fpu_id, sizeof(u32));
+	return 0;
+}
+
+/*
+ * Copy the supplied NT_PRFPREG buffer to the floating-point context,
+ * !CONFIG_CPU_HAS_MSA variant.   Buffer slots correspond 1:1 to FP
+ * context's general register slots.  Only general registers are copied.
+ */
+static int fpr_set_fpa(struct task_struct *target,
+		       unsigned int *pos, unsigned int *count,
+		       const void **kbuf, const void __user **ubuf)
+{
+	return user_regset_copyin(pos, count, kbuf, ubuf,
+				  &target->thread.fpu,
+				  0, NUM_FPU_REGS * sizeof(elf_fpreg_t));
+}
+
+/*
+ * Copy the supplied NT_PRFPREG buffer to the floating-point context,
+ * CONFIG_CPU_HAS_MSA variant.  Buffer slots are copied to lower 64
+ * bits only of FP context's general register slots.  Only general
+ * registers are copied.
+ */
+static int fpr_set_msa(struct task_struct *target,
+		       unsigned int *pos, unsigned int *count,
+		       const void **kbuf, const void __user **ubuf)
+{
+	unsigned int i;
+	u64 fpr_val;
+	int err;
+
+	BUILD_BUG_ON(sizeof(fpr_val) != sizeof(elf_fpreg_t));
+	for (i = 0; i < NUM_FPU_REGS && *count > 0; i++) {
+		err = user_regset_copyin(pos, count, kbuf, ubuf,
+					 &fpr_val, i * sizeof(elf_fpreg_t),
+					 (i + 1) * sizeof(elf_fpreg_t));
+		if (err)
+			return err;
+		set_fpr64(&target->thread.fpu.fpr[i], 0, fpr_val);
+	}
+
+	return 0;
+}
+
+/*
+ * Copy the supplied NT_PRFPREG buffer to the floating-point context.
+ * Choose the appropriate helper for general registers, and then copy
+ * the FCSR register separately.  Ignore the incoming FIR register
+ * contents though, as the register is read-only.
+ *
+ * We optimize for the case where `count % sizeof(elf_fpreg_t) == 0',
+ * which is supposed to have been guaranteed by the kernel before
+ * calling us, e.g. in `ptrace_regset'.  We enforce that requirement,
+ * so that we can safely avoid preinitializing temporaries for
+ * partial register writes.
+ */
+static int fpr_set(struct task_struct *target,
+		   const struct user_regset *regset,
+		   unsigned int pos, unsigned int count,
+		   const void *kbuf, const void __user *ubuf)
+{
+	const int fcr31_pos = NUM_FPU_REGS * sizeof(elf_fpreg_t);
+	const int fir_pos = fcr31_pos + sizeof(u32);
+	u32 fcr31;
+	int err;
+
+	BUG_ON(count % sizeof(elf_fpreg_t));
+
+	if (pos + count > sizeof(elf_fpregset_t))
+		return -EIO;
+
+	init_fp_ctx(target);
+
+	if (sizeof(target->thread.fpu.fpr[0]) == sizeof(elf_fpreg_t))
+		err = fpr_set_fpa(target, &pos, &count, &kbuf, &ubuf);
+	else
+		err = fpr_set_msa(target, &pos, &count, &kbuf, &ubuf);
+	if (err)
+		return err;
+
+	if (count > 0) {
+		err = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
+					 &fcr31,
+					 fcr31_pos, fcr31_pos + sizeof(u32));
+		if (err)
+			return err;
+
+		ptrace_setfcr31(target, fcr31);
+	}
+
+	if (count > 0)
+		err = user_regset_copyin_ignore(&pos, &count, &kbuf, &ubuf,
+						fir_pos,
+						fir_pos + sizeof(u32));
+
+	return err;
+}
+
+/* Copy the FP mode setting to the supplied NT_MIPS_FP_MODE buffer.  */
+static int fp_mode_get(struct task_struct *target,
+		       const struct user_regset *regset,
+		       struct membuf to)
+{
+	return membuf_store(&to, (int)mips_get_process_fp_mode(target));
+}
+
+/*
+ * Copy the supplied NT_MIPS_FP_MODE buffer to the FP mode setting.
+ *
+ * We optimize for the case where `count % sizeof(int) == 0', which
+ * is supposed to have been guaranteed by the kernel before calling
+ * us, e.g. in `ptrace_regset'.  We enforce that requirement, so
+ * that we can safely avoid preinitializing temporaries for partial
+ * mode writes.
+ */
+static int fp_mode_set(struct task_struct *target,
+		       const struct user_regset *regset,
+		       unsigned int pos, unsigned int count,
+		       const void *kbuf, const void __user *ubuf)
+{
+	int fp_mode;
+	int err;
+
+	BUG_ON(count % sizeof(int));
+
+	if (pos + count > sizeof(fp_mode))
+		return -EIO;
+
+	err = user_regset_copyin(&pos, &count, &kbuf, &ubuf, &fp_mode, 0,
+				 sizeof(fp_mode));
+	if (err)
+		return err;
+
+	if (count > 0)
+		err = mips_set_process_fp_mode(target, fp_mode);
+
+	return err;
+}
+
+#endif /* CONFIG_MIPS_FP_SUPPORT */
+
+#ifdef CONFIG_CPU_HAS_MSA
+
+struct msa_control_regs {
+	unsigned int fir;
+	unsigned int fcsr;
+	unsigned int msair;
+	unsigned int msacsr;
+};
+
+static void copy_pad_fprs(struct task_struct *target,
+			 const struct user_regset *regset,
+			 struct membuf *to,
+			 unsigned int live_sz)
+{
+	int i, j;
+	unsigned long long fill = ~0ull;
+	unsigned int cp_sz, pad_sz;
+
+	cp_sz = min(regset->size, live_sz);
+	pad_sz = regset->size - cp_sz;
+	WARN_ON(pad_sz % sizeof(fill));
+
+	for (i = 0; i < NUM_FPU_REGS; i++) {
+		membuf_write(to, &target->thread.fpu.fpr[i], cp_sz);
+		for (j = 0; j < (pad_sz / sizeof(fill)); j++)
+			membuf_store(to, fill);
+	}
+}
+
+static int msa_get(struct task_struct *target,
+		   const struct user_regset *regset,
+		   struct membuf to)
+{
+	const unsigned int wr_size = NUM_FPU_REGS * regset->size;
+	const struct msa_control_regs ctrl_regs = {
+		.fir = boot_cpu_data.fpu_id,
+		.fcsr = target->thread.fpu.fcr31,
+		.msair = boot_cpu_data.msa_id,
+		.msacsr = target->thread.fpu.msacsr,
+	};
+
+	if (!tsk_used_math(target)) {
+		/* The task hasn't used FP or MSA, fill with 0xff */
+		copy_pad_fprs(target, regset, &to, 0);
+	} else if (!test_tsk_thread_flag(target, TIF_MSA_CTX_LIVE)) {
+		/* Copy scalar FP context, fill the rest with 0xff */
+		copy_pad_fprs(target, regset, &to, 8);
+	} else if (sizeof(target->thread.fpu.fpr[0]) == regset->size) {
+		/* Trivially copy the vector registers */
+		membuf_write(&to, &target->thread.fpu.fpr, wr_size);
+	} else {
+		/* Copy as much context as possible, fill the rest with 0xff */
+		copy_pad_fprs(target, regset, &to,
+				sizeof(target->thread.fpu.fpr[0]));
+	}
+
+	return membuf_write(&to, &ctrl_regs, sizeof(ctrl_regs));
+}
+
+static int msa_set(struct task_struct *target,
+		   const struct user_regset *regset,
+		   unsigned int pos, unsigned int count,
+		   const void *kbuf, const void __user *ubuf)
+{
+	const unsigned int wr_size = NUM_FPU_REGS * regset->size;
+	struct msa_control_regs ctrl_regs;
+	unsigned int cp_sz;
+	int i, err, start;
+
+	init_fp_ctx(target);
+
+	if (sizeof(target->thread.fpu.fpr[0]) == regset->size) {
+		/* Trivially copy the vector registers */
+		err = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
+					 &target->thread.fpu.fpr,
+					 0, wr_size);
+	} else {
+		/* Copy as much context as possible */
+		cp_sz = min_t(unsigned int, regset->size,
+			      sizeof(target->thread.fpu.fpr[0]));
+
+		i = start = err = 0;
+		for (; i < NUM_FPU_REGS; i++, start += regset->size) {
+			err |= user_regset_copyin(&pos, &count, &kbuf, &ubuf,
+						  &target->thread.fpu.fpr[i],
+						  start, start + cp_sz);
+		}
+	}
+
+	if (!err)
+		err = user_regset_copyin(&pos, &count, &kbuf, &ubuf, &ctrl_regs,
+					 wr_size, wr_size + sizeof(ctrl_regs));
+	if (!err) {
+		target->thread.fpu.fcr31 = ctrl_regs.fcsr & ~FPU_CSR_ALL_X;
+		target->thread.fpu.msacsr = ctrl_regs.msacsr & ~MSA_CSR_CAUSEF;
+	}
+
+	return err;
+}
+
+#endif /* CONFIG_CPU_HAS_MSA */
+
+#if defined(CONFIG_32BIT) || defined(CONFIG_MIPS32_O32)
+
+/*
+ * Copy the DSP context to the supplied 32-bit NT_MIPS_DSP buffer.
+ */
+static int dsp32_get(struct task_struct *target,
+		     const struct user_regset *regset,
+		     struct membuf to)
+{
+	u32 dspregs[NUM_DSP_REGS + 1];
+	unsigned int i;
+
+	BUG_ON(to.left % sizeof(u32));
+
+	if (!cpu_has_dsp)
+		return -EIO;
+
+	for (i = 0; i < NUM_DSP_REGS; i++)
+		dspregs[i] = target->thread.dsp.dspr[i];
+	dspregs[NUM_DSP_REGS] = target->thread.dsp.dspcontrol;
+	return membuf_write(&to, dspregs, sizeof(dspregs));
+}
+
+/*
+ * Copy the supplied 32-bit NT_MIPS_DSP buffer to the DSP context.
+ */
+static int dsp32_set(struct task_struct *target,
+		     const struct user_regset *regset,
+		     unsigned int pos, unsigned int count,
+		     const void *kbuf, const void __user *ubuf)
+{
+	unsigned int start, num_regs, i;
+	u32 dspregs[NUM_DSP_REGS + 1];
+	int err;
+
+	BUG_ON(count % sizeof(u32));
+
+	if (!cpu_has_dsp)
+		return -EIO;
+
+	start = pos / sizeof(u32);
+	num_regs = count / sizeof(u32);
+
+	if (start + num_regs > NUM_DSP_REGS + 1)
+		return -EIO;
+
+	err = user_regset_copyin(&pos, &count, &kbuf, &ubuf, dspregs, 0,
+				 sizeof(dspregs));
+	if (err)
+		return err;
+
+	for (i = start; i < num_regs; i++)
+		switch (i) {
+		case 0 ... NUM_DSP_REGS - 1:
+			target->thread.dsp.dspr[i] = (s32)dspregs[i];
+			break;
+		case NUM_DSP_REGS:
+			target->thread.dsp.dspcontrol = (s32)dspregs[i];
+			break;
+		}
+
+	return 0;
+}
+
+#endif /* CONFIG_32BIT || CONFIG_MIPS32_O32 */
+
+#ifdef CONFIG_64BIT
+
+/*
+ * Copy the DSP context to the supplied 64-bit NT_MIPS_DSP buffer.
+ */
+static int dsp64_get(struct task_struct *target,
+		     const struct user_regset *regset,
+		     struct membuf to)
+{
+	u64 dspregs[NUM_DSP_REGS + 1];
+	unsigned int i;
+
+	BUG_ON(to.left % sizeof(u64));
+
+	if (!cpu_has_dsp)
+		return -EIO;
+
+	for (i = 0; i < NUM_DSP_REGS; i++)
+		dspregs[i] = target->thread.dsp.dspr[i];
+	dspregs[NUM_DSP_REGS] = target->thread.dsp.dspcontrol;
+	return membuf_write(&to, dspregs, sizeof(dspregs));
+}
+
+/*
+ * Copy the supplied 64-bit NT_MIPS_DSP buffer to the DSP context.
+ */
+static int dsp64_set(struct task_struct *target,
+		     const struct user_regset *regset,
+		     unsigned int pos, unsigned int count,
+		     const void *kbuf, const void __user *ubuf)
+{
+	unsigned int start, num_regs, i;
+	u64 dspregs[NUM_DSP_REGS + 1];
+	int err;
+
+	BUG_ON(count % sizeof(u64));
+
+	if (!cpu_has_dsp)
+		return -EIO;
+
+	start = pos / sizeof(u64);
+	num_regs = count / sizeof(u64);
+
+	if (start + num_regs > NUM_DSP_REGS + 1)
+		return -EIO;
+
+	err = user_regset_copyin(&pos, &count, &kbuf, &ubuf, dspregs, 0,
+				 sizeof(dspregs));
+	if (err)
+		return err;
+
+	for (i = start; i < num_regs; i++)
+		switch (i) {
+		case 0 ... NUM_DSP_REGS - 1:
+			target->thread.dsp.dspr[i] = dspregs[i];
+			break;
+		case NUM_DSP_REGS:
+			target->thread.dsp.dspcontrol = dspregs[i];
+			break;
+		}
+
+	return 0;
+}
+
+#endif /* CONFIG_64BIT */
+
+/*
+ * Determine whether the DSP context is present.
+ */
+static int dsp_active(struct task_struct *target,
+		      const struct user_regset *regset)
+{
+	return cpu_has_dsp ? NUM_DSP_REGS + 1 : -ENODEV;
+}
+
+enum mips_regset {
+	REGSET_GPR,
+	REGSET_DSP,
+#ifdef CONFIG_MIPS_FP_SUPPORT
+	REGSET_FPR,
+	REGSET_FP_MODE,
+#endif
+#ifdef CONFIG_CPU_HAS_MSA
+	REGSET_MSA,
+#endif
+};
+
+struct pt_regs_offset {
+	const char *name;
+	int offset;
+};
+
+#define REG_OFFSET_NAME(reg, r) {					\
+	.name = #reg,							\
+	.offset = offsetof(struct pt_regs, r)				\
+}
+
+#define REG_OFFSET_END {						\
+	.name = NULL,							\
+	.offset = 0							\
+}
+
+static const struct pt_regs_offset regoffset_table[] = {
+	REG_OFFSET_NAME(r0, regs[0]),
+	REG_OFFSET_NAME(r1, regs[1]),
+	REG_OFFSET_NAME(r2, regs[2]),
+	REG_OFFSET_NAME(r3, regs[3]),
+	REG_OFFSET_NAME(r4, regs[4]),
+	REG_OFFSET_NAME(r5, regs[5]),
+	REG_OFFSET_NAME(r6, regs[6]),
+	REG_OFFSET_NAME(r7, regs[7]),
+	REG_OFFSET_NAME(r8, regs[8]),
+	REG_OFFSET_NAME(r9, regs[9]),
+	REG_OFFSET_NAME(r10, regs[10]),
+	REG_OFFSET_NAME(r11, regs[11]),
+	REG_OFFSET_NAME(r12, regs[12]),
+	REG_OFFSET_NAME(r13, regs[13]),
+	REG_OFFSET_NAME(r14, regs[14]),
+	REG_OFFSET_NAME(r15, regs[15]),
+	REG_OFFSET_NAME(r16, regs[16]),
+	REG_OFFSET_NAME(r17, regs[17]),
+	REG_OFFSET_NAME(r18, regs[18]),
+	REG_OFFSET_NAME(r19, regs[19]),
+	REG_OFFSET_NAME(r20, regs[20]),
+	REG_OFFSET_NAME(r21, regs[21]),
+	REG_OFFSET_NAME(r22, regs[22]),
+	REG_OFFSET_NAME(r23, regs[23]),
+	REG_OFFSET_NAME(r24, regs[24]),
+	REG_OFFSET_NAME(r25, regs[25]),
+	REG_OFFSET_NAME(r26, regs[26]),
+	REG_OFFSET_NAME(r27, regs[27]),
+	REG_OFFSET_NAME(r28, regs[28]),
+	REG_OFFSET_NAME(r29, regs[29]),
+	REG_OFFSET_NAME(r30, regs[30]),
+	REG_OFFSET_NAME(r31, regs[31]),
+	REG_OFFSET_NAME(c0_status, cp0_status),
+	REG_OFFSET_NAME(hi, hi),
+	REG_OFFSET_NAME(lo, lo),
+#ifdef CONFIG_CPU_HAS_SMARTMIPS
+	REG_OFFSET_NAME(acx, acx),
+#endif
+	REG_OFFSET_NAME(c0_badvaddr, cp0_badvaddr),
+	REG_OFFSET_NAME(c0_cause, cp0_cause),
+	REG_OFFSET_NAME(c0_epc, cp0_epc),
+#ifdef CONFIG_CPU_CAVIUM_OCTEON
+	REG_OFFSET_NAME(mpl0, mpl[0]),
+	REG_OFFSET_NAME(mpl1, mpl[1]),
+	REG_OFFSET_NAME(mpl2, mpl[2]),
+	REG_OFFSET_NAME(mtp0, mtp[0]),
+	REG_OFFSET_NAME(mtp1, mtp[1]),
+	REG_OFFSET_NAME(mtp2, mtp[2]),
+#endif
+	REG_OFFSET_END,
+};
+
+/**
+ * regs_query_register_offset() - query register offset from its name
+ * @name:       the name of a register
+ *
+ * regs_query_register_offset() returns the offset of a register in struct
+ * pt_regs from its name. If the name is invalid, this returns -EINVAL;
+ */
+int regs_query_register_offset(const char *name)
+{
+        const struct pt_regs_offset *roff;
+        for (roff = regoffset_table; roff->name != NULL; roff++)
+                if (!strcmp(roff->name, name))
+                        return roff->offset;
+        return -EINVAL;
+}
+
+#if defined(CONFIG_32BIT) || defined(CONFIG_MIPS32_O32)
+
+static const struct user_regset mips_regsets[] = {
+	[REGSET_GPR] = {
+		.core_note_type	= NT_PRSTATUS,
+		.n		= ELF_NGREG,
+		.size		= sizeof(unsigned int),
+		.align		= sizeof(unsigned int),
+		.regset_get		= gpr32_get,
+		.set		= gpr32_set,
+	},
+	[REGSET_DSP] = {
+		.core_note_type	= NT_MIPS_DSP,
+		.n		= NUM_DSP_REGS + 1,
+		.size		= sizeof(u32),
+		.align		= sizeof(u32),
+		.regset_get		= dsp32_get,
+		.set		= dsp32_set,
+		.active		= dsp_active,
+	},
+#ifdef CONFIG_MIPS_FP_SUPPORT
+	[REGSET_FPR] = {
+		.core_note_type	= NT_PRFPREG,
+		.n		= ELF_NFPREG,
+		.size		= sizeof(elf_fpreg_t),
+		.align		= sizeof(elf_fpreg_t),
+		.regset_get		= fpr_get,
+		.set		= fpr_set,
+	},
+	[REGSET_FP_MODE] = {
+		.core_note_type	= NT_MIPS_FP_MODE,
+		.n		= 1,
+		.size		= sizeof(int),
+		.align		= sizeof(int),
+		.regset_get		= fp_mode_get,
+		.set		= fp_mode_set,
+	},
+#endif
+#ifdef CONFIG_CPU_HAS_MSA
+	[REGSET_MSA] = {
+		.core_note_type	= NT_MIPS_MSA,
+		.n		= NUM_FPU_REGS + 1,
+		.size		= 16,
+		.align		= 16,
+		.regset_get		= msa_get,
+		.set		= msa_set,
+	},
+#endif
+};
+
+static const struct user_regset_view user_mips_view = {
+	.name		= "mips",
+	.e_machine	= ELF_ARCH,
+	.ei_osabi	= ELF_OSABI,
+	.regsets	= mips_regsets,
+	.n		= ARRAY_SIZE(mips_regsets),
+};
+
+#endif /* CONFIG_32BIT || CONFIG_MIPS32_O32 */
+
+#ifdef CONFIG_64BIT
+
+static const struct user_regset mips64_regsets[] = {
+	[REGSET_GPR] = {
+		.core_note_type	= NT_PRSTATUS,
+		.n		= ELF_NGREG,
+		.size		= sizeof(unsigned long),
+		.align		= sizeof(unsigned long),
+		.regset_get		= gpr64_get,
+		.set		= gpr64_set,
+	},
+	[REGSET_DSP] = {
+		.core_note_type	= NT_MIPS_DSP,
+		.n		= NUM_DSP_REGS + 1,
+		.size		= sizeof(u64),
+		.align		= sizeof(u64),
+		.regset_get		= dsp64_get,
+		.set		= dsp64_set,
+		.active		= dsp_active,
+	},
+#ifdef CONFIG_MIPS_FP_SUPPORT
+	[REGSET_FP_MODE] = {
+		.core_note_type	= NT_MIPS_FP_MODE,
+		.n		= 1,
+		.size		= sizeof(int),
+		.align		= sizeof(int),
+		.regset_get		= fp_mode_get,
+		.set		= fp_mode_set,
+	},
+	[REGSET_FPR] = {
+		.core_note_type	= NT_PRFPREG,
+		.n		= ELF_NFPREG,
+		.size		= sizeof(elf_fpreg_t),
+		.align		= sizeof(elf_fpreg_t),
+		.regset_get		= fpr_get,
+		.set		= fpr_set,
+	},
+#endif
+#ifdef CONFIG_CPU_HAS_MSA
+	[REGSET_MSA] = {
+		.core_note_type	= NT_MIPS_MSA,
+		.n		= NUM_FPU_REGS + 1,
+		.size		= 16,
+		.align		= 16,
+		.regset_get		= msa_get,
+		.set		= msa_set,
+	},
+#endif
+};
+
+static const struct user_regset_view user_mips64_view = {
+	.name		= "mips64",
+	.e_machine	= ELF_ARCH,
+	.ei_osabi	= ELF_OSABI,
+	.regsets	= mips64_regsets,
+	.n		= ARRAY_SIZE(mips64_regsets),
+};
+
+#ifdef CONFIG_MIPS32_N32
+
+static const struct user_regset_view user_mipsn32_view = {
+	.name		= "mipsn32",
+	.e_flags	= EF_MIPS_ABI2,
+	.e_machine	= ELF_ARCH,
+	.ei_osabi	= ELF_OSABI,
+	.regsets	= mips64_regsets,
+	.n		= ARRAY_SIZE(mips64_regsets),
+};
+
+#endif /* CONFIG_MIPS32_N32 */
+
+#endif /* CONFIG_64BIT */
+
+const struct user_regset_view *task_user_regset_view(struct task_struct *task)
+{
+#ifdef CONFIG_32BIT
+	return &user_mips_view;
+#else
+#ifdef CONFIG_MIPS32_O32
+	if (test_tsk_thread_flag(task, TIF_32BIT_REGS))
+		return &user_mips_view;
+#endif
+#ifdef CONFIG_MIPS32_N32
+	if (test_tsk_thread_flag(task, TIF_32BIT_ADDR))
+		return &user_mipsn32_view;
+#endif
+	return &user_mips64_view;
+#endif
+}
+
+long arch_ptrace(struct task_struct *child, long request,
+		 unsigned long addr, unsigned long data)
+{
+	int ret;
+	void __user *addrp = (void __user *) addr;
+	void __user *datavp = (void __user *) data;
+	unsigned long __user *datalp = (void __user *) data;
+
+	switch (request) {
+	/* when I and D space are separate, these will need to be fixed. */
+	case PTRACE_PEEKTEXT: /* read word at location addr. */
+	case PTRACE_PEEKDATA:
+		ret = generic_ptrace_peekdata(child, addr, data);
+		break;
+
+	/* Read the word at location addr in the USER area. */
+	case PTRACE_PEEKUSR: {
+		struct pt_regs *regs;
+		unsigned long tmp = 0;
+
+		regs = task_pt_regs(child);
+		ret = 0;  /* Default return value. */
+
+		switch (addr) {
+		case 0 ... 31:
+			tmp = regs->regs[addr];
+			break;
+#ifdef CONFIG_MIPS_FP_SUPPORT
+		case FPR_BASE ... FPR_BASE + 31: {
+			union fpureg *fregs;
+
+			if (!tsk_used_math(child)) {
+				/* FP not yet used */
+				tmp = -1;
+				break;
+			}
+			fregs = get_fpu_regs(child);
+
+#ifdef CONFIG_32BIT
+			if (test_tsk_thread_flag(child, TIF_32BIT_FPREGS)) {
+				/*
+				 * The odd registers are actually the high
+				 * order bits of the values stored in the even
+				 * registers.
+				 */
+				tmp = get_fpr32(&fregs[(addr & ~1) - FPR_BASE],
+						addr & 1);
+				break;
+			}
+#endif
+			tmp = get_fpr64(&fregs[addr - FPR_BASE], 0);
+			break;
+		}
+		case FPC_CSR:
+			tmp = child->thread.fpu.fcr31;
+			break;
+		case FPC_EIR:
+			/* implementation / version register */
+			tmp = boot_cpu_data.fpu_id;
+			break;
+#endif
+		case PC:
+			tmp = regs->cp0_epc;
+			break;
+		case CAUSE:
+			tmp = regs->cp0_cause;
+			break;
+		case BADVADDR:
+			tmp = regs->cp0_badvaddr;
+			break;
+		case MMHI:
+			tmp = regs->hi;
+			break;
+		case MMLO:
+			tmp = regs->lo;
+			break;
+#ifdef CONFIG_CPU_HAS_SMARTMIPS
+		case ACX:
+			tmp = regs->acx;
+			break;
+#endif
+		case DSP_BASE ... DSP_BASE + 5: {
+			dspreg_t *dregs;
+
+			if (!cpu_has_dsp) {
+				tmp = 0;
+				ret = -EIO;
+				goto out;
+			}
+			dregs = __get_dsp_regs(child);
+			tmp = dregs[addr - DSP_BASE];
+			break;
+		}
+		case DSP_CONTROL:
+			if (!cpu_has_dsp) {
+				tmp = 0;
+				ret = -EIO;
+				goto out;
+			}
+			tmp = child->thread.dsp.dspcontrol;
+			break;
+		default:
+			tmp = 0;
+			ret = -EIO;
+			goto out;
+		}
+		ret = put_user(tmp, datalp);
+		break;
+	}
+
+	/* when I and D space are separate, this will have to be fixed. */
+	case PTRACE_POKETEXT: /* write the word at location addr. */
+	case PTRACE_POKEDATA:
+		ret = generic_ptrace_pokedata(child, addr, data);
+		break;
+
+	case PTRACE_POKEUSR: {
+		struct pt_regs *regs;
+		ret = 0;
+		regs = task_pt_regs(child);
+
+		switch (addr) {
+		case 0 ... 31:
+			regs->regs[addr] = data;
+			/* System call number may have been changed */
+			if (addr == 2)
+				mips_syscall_update_nr(child, regs);
+			else if (addr == 4 &&
+				 mips_syscall_is_indirect(child, regs))
+				mips_syscall_update_nr(child, regs);
+			break;
+#ifdef CONFIG_MIPS_FP_SUPPORT
+		case FPR_BASE ... FPR_BASE + 31: {
+			union fpureg *fregs = get_fpu_regs(child);
+
+			init_fp_ctx(child);
+#ifdef CONFIG_32BIT
+			if (test_tsk_thread_flag(child, TIF_32BIT_FPREGS)) {
+				/*
+				 * The odd registers are actually the high
+				 * order bits of the values stored in the even
+				 * registers.
+				 */
+				set_fpr32(&fregs[(addr & ~1) - FPR_BASE],
+					  addr & 1, data);
+				break;
+			}
+#endif
+			set_fpr64(&fregs[addr - FPR_BASE], 0, data);
+			break;
+		}
+		case FPC_CSR:
+			init_fp_ctx(child);
+			ptrace_setfcr31(child, data);
+			break;
+#endif
+		case PC:
+			regs->cp0_epc = data;
+			break;
+		case MMHI:
+			regs->hi = data;
+			break;
+		case MMLO:
+			regs->lo = data;
+			break;
+#ifdef CONFIG_CPU_HAS_SMARTMIPS
+		case ACX:
+			regs->acx = data;
+			break;
+#endif
+		case DSP_BASE ... DSP_BASE + 5: {
+			dspreg_t *dregs;
+
+			if (!cpu_has_dsp) {
+				ret = -EIO;
+				break;
+			}
+
+			dregs = __get_dsp_regs(child);
+			dregs[addr - DSP_BASE] = data;
+			break;
+		}
+		case DSP_CONTROL:
+			if (!cpu_has_dsp) {
+				ret = -EIO;
+				break;
+			}
+			child->thread.dsp.dspcontrol = data;
+			break;
+		default:
+			/* The rest are not allowed. */
+			ret = -EIO;
+			break;
+		}
+		break;
+		}
+
+	case PTRACE_GETREGS:
+		ret = ptrace_getregs(child, datavp);
+		break;
+
+	case PTRACE_SETREGS:
+		ret = ptrace_setregs(child, datavp);
+		break;
+
+#ifdef CONFIG_MIPS_FP_SUPPORT
+	case PTRACE_GETFPREGS:
+		ret = ptrace_getfpregs(child, datavp);
+		break;
+
+	case PTRACE_SETFPREGS:
+		ret = ptrace_setfpregs(child, datavp);
+		break;
+#endif
+	case PTRACE_GET_THREAD_AREA:
+		ret = put_user(task_thread_info(child)->tp_value, datalp);
+		break;
+
+	case PTRACE_GET_WATCH_REGS:
+		ret = ptrace_get_watch_regs(child, addrp);
+		break;
+
+	case PTRACE_SET_WATCH_REGS:
+		ret = ptrace_set_watch_regs(child, addrp);
+		break;
+
+	default:
+		ret = ptrace_request(child, request, addr, data);
+		break;
+	}
+ out:
+	return ret;
+}
+
+/*
+ * Notification of system call entry/exit
+ * - triggered by current->work.syscall_trace
+ */
+asmlinkage long syscall_trace_enter(struct pt_regs *regs, long syscall)
+{
+	user_exit();
+
+	current_thread_info()->syscall = syscall;
+
+	if (test_thread_flag(TIF_SYSCALL_TRACE)) {
+		if (tracehook_report_syscall_entry(regs))
+			return -1;
+		syscall = current_thread_info()->syscall;
+	}
+
+#ifdef CONFIG_SECCOMP
+	if (unlikely(test_thread_flag(TIF_SECCOMP))) {
+		int ret, i;
+		struct seccomp_data sd;
+		unsigned long args[6];
+
+		sd.nr = syscall;
+		sd.arch = syscall_get_arch(current);
+		syscall_get_arguments(current, regs, args);
+		for (i = 0; i < 6; i++)
+			sd.args[i] = args[i];
+		sd.instruction_pointer = KSTK_EIP(current);
+
+		ret = __secure_computing(&sd);
+		if (ret == -1)
+			return ret;
+		syscall = current_thread_info()->syscall;
+	}
+#endif
+
+	if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
+		trace_sys_enter(regs, regs->regs[2]);
+
+	audit_syscall_entry(syscall, regs->regs[4], regs->regs[5],
+			    regs->regs[6], regs->regs[7]);
+
+	/*
+	 * Negative syscall numbers are mistaken for rejected syscalls, but
+	 * won't have had the return value set appropriately, so we do so now.
+	 */
+	if (syscall < 0)
+		syscall_set_return_value(current, regs, -ENOSYS, 0);
+	return syscall;
+}
+
+/*
+ * Notification of system call entry/exit
+ * - triggered by current->work.syscall_trace
+ */
+asmlinkage void syscall_trace_leave(struct pt_regs *regs)
+{
+        /*
+	 * We may come here right after calling schedule_user()
+	 * or do_notify_resume(), in which case we can be in RCU
+	 * user mode.
+	 */
+	user_exit();
+
+	audit_syscall_exit(regs);
+
+	if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
+		trace_sys_exit(regs, regs_return_value(regs));
+
+	if (test_thread_flag(TIF_SYSCALL_TRACE))
+		tracehook_report_syscall_exit(regs, 0);
+
+	user_enter();
+}
diff --git a/arch/mips/kernel/ptrace32.c b/arch/mips/kernel/ptrace32.c
new file mode 100644
index 000000000..afcf27a87
--- /dev/null
+++ b/arch/mips/kernel/ptrace32.c
@@ -0,0 +1,317 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1992 Ross Biro
+ * Copyright (C) Linus Torvalds
+ * Copyright (C) 1994, 95, 96, 97, 98, 2000 Ralf Baechle
+ * Copyright (C) 1996 David S. Miller
+ * Kevin D. Kissell, kevink@mips.com and Carsten Langgaard, carstenl@mips.com
+ * Copyright (C) 1999 MIPS Technologies, Inc.
+ * Copyright (C) 2000 Ulf Carlsson
+ *
+ * At this time Linux/MIPS64 only supports syscall tracing, even for 32-bit
+ * binaries.
+ */
+#include <linux/compiler.h>
+#include <linux/compat.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/sched/task_stack.h>
+#include <linux/mm.h>
+#include <linux/errno.h>
+#include <linux/ptrace.h>
+#include <linux/smp.h>
+#include <linux/security.h>
+
+#include <asm/cpu.h>
+#include <asm/dsp.h>
+#include <asm/fpu.h>
+#include <asm/mipsregs.h>
+#include <asm/mipsmtregs.h>
+#include <asm/page.h>
+#include <asm/reg.h>
+#include <asm/syscall.h>
+#include <linux/uaccess.h>
+#include <asm/bootinfo.h>
+
+/*
+ * Tracing a 32-bit process with a 64-bit strace and vice versa will not
+ * work.  I don't know how to fix this.
+ */
+long compat_arch_ptrace(struct task_struct *child, compat_long_t request,
+			compat_ulong_t caddr, compat_ulong_t cdata)
+{
+	int addr = caddr;
+	int data = cdata;
+	int ret;
+
+	switch (request) {
+
+	/*
+	 * Read 4 bytes of the other process' storage
+	 *  data is a pointer specifying where the user wants the
+	 *	4 bytes copied into
+	 *  addr is a pointer in the user's storage that contains an 8 byte
+	 *	address in the other process of the 4 bytes that is to be read
+	 * (this is run in a 32-bit process looking at a 64-bit process)
+	 * when I and D space are separate, these will need to be fixed.
+	 */
+	case PTRACE_PEEKTEXT_3264:
+	case PTRACE_PEEKDATA_3264: {
+		u32 tmp;
+		int copied;
+		u32 __user * addrOthers;
+
+		ret = -EIO;
+
+		/* Get the addr in the other process that we want to read */
+		if (get_user(addrOthers, (u32 __user * __user *) (unsigned long) addr) != 0)
+			break;
+
+		copied = ptrace_access_vm(child, (u64)addrOthers, &tmp,
+				sizeof(tmp), FOLL_FORCE);
+		if (copied != sizeof(tmp))
+			break;
+		ret = put_user(tmp, (u32 __user *) (unsigned long) data);
+		break;
+	}
+
+	/* Read the word at location addr in the USER area. */
+	case PTRACE_PEEKUSR: {
+		struct pt_regs *regs;
+		unsigned int tmp;
+
+		regs = task_pt_regs(child);
+		ret = 0;  /* Default return value. */
+
+		switch (addr) {
+		case 0 ... 31:
+			tmp = regs->regs[addr];
+			break;
+#ifdef CONFIG_MIPS_FP_SUPPORT
+		case FPR_BASE ... FPR_BASE + 31: {
+			union fpureg *fregs;
+
+			if (!tsk_used_math(child)) {
+				/* FP not yet used */
+				tmp = -1;
+				break;
+			}
+			fregs = get_fpu_regs(child);
+			if (test_tsk_thread_flag(child, TIF_32BIT_FPREGS)) {
+				/*
+				 * The odd registers are actually the high
+				 * order bits of the values stored in the even
+				 * registers.
+				 */
+				tmp = get_fpr32(&fregs[(addr & ~1) - FPR_BASE],
+						addr & 1);
+				break;
+			}
+			tmp = get_fpr64(&fregs[addr - FPR_BASE], 0);
+			break;
+		}
+		case FPC_CSR:
+			tmp = child->thread.fpu.fcr31;
+			break;
+		case FPC_EIR:
+			/* implementation / version register */
+			tmp = boot_cpu_data.fpu_id;
+			break;
+#endif /* CONFIG_MIPS_FP_SUPPORT */
+		case PC:
+			tmp = regs->cp0_epc;
+			break;
+		case CAUSE:
+			tmp = regs->cp0_cause;
+			break;
+		case BADVADDR:
+			tmp = regs->cp0_badvaddr;
+			break;
+		case MMHI:
+			tmp = regs->hi;
+			break;
+		case MMLO:
+			tmp = regs->lo;
+			break;
+		case DSP_BASE ... DSP_BASE + 5: {
+			dspreg_t *dregs;
+
+			if (!cpu_has_dsp) {
+				tmp = 0;
+				ret = -EIO;
+				goto out;
+			}
+			dregs = __get_dsp_regs(child);
+			tmp = dregs[addr - DSP_BASE];
+			break;
+		}
+		case DSP_CONTROL:
+			if (!cpu_has_dsp) {
+				tmp = 0;
+				ret = -EIO;
+				goto out;
+			}
+			tmp = child->thread.dsp.dspcontrol;
+			break;
+		default:
+			tmp = 0;
+			ret = -EIO;
+			goto out;
+		}
+		ret = put_user(tmp, (unsigned __user *) (unsigned long) data);
+		break;
+	}
+
+	/*
+	 * Write 4 bytes into the other process' storage
+	 *  data is the 4 bytes that the user wants written
+	 *  addr is a pointer in the user's storage that contains an
+	 *	8 byte address in the other process where the 4 bytes
+	 *	that is to be written
+	 * (this is run in a 32-bit process looking at a 64-bit process)
+	 * when I and D space are separate, these will need to be fixed.
+	 */
+	case PTRACE_POKETEXT_3264:
+	case PTRACE_POKEDATA_3264: {
+		u32 __user * addrOthers;
+
+		/* Get the addr in the other process that we want to write into */
+		ret = -EIO;
+		if (get_user(addrOthers, (u32 __user * __user *) (unsigned long) addr) != 0)
+			break;
+		ret = 0;
+		if (ptrace_access_vm(child, (u64)addrOthers, &data,
+					sizeof(data),
+					FOLL_FORCE | FOLL_WRITE) == sizeof(data))
+			break;
+		ret = -EIO;
+		break;
+	}
+
+	case PTRACE_POKEUSR: {
+		struct pt_regs *regs;
+		ret = 0;
+		regs = task_pt_regs(child);
+
+		switch (addr) {
+		case 0 ... 31:
+			regs->regs[addr] = data;
+			/* System call number may have been changed */
+			if (addr == 2)
+				mips_syscall_update_nr(child, regs);
+			else if (addr == 4 &&
+				 mips_syscall_is_indirect(child, regs))
+				mips_syscall_update_nr(child, regs);
+			break;
+#ifdef CONFIG_MIPS_FP_SUPPORT
+		case FPR_BASE ... FPR_BASE + 31: {
+			union fpureg *fregs = get_fpu_regs(child);
+
+			if (!tsk_used_math(child)) {
+				/* FP not yet used  */
+				memset(&child->thread.fpu, ~0,
+				       sizeof(child->thread.fpu));
+				child->thread.fpu.fcr31 = 0;
+			}
+			if (test_tsk_thread_flag(child, TIF_32BIT_FPREGS)) {
+				/*
+				 * The odd registers are actually the high
+				 * order bits of the values stored in the even
+				 * registers.
+				 */
+				set_fpr32(&fregs[(addr & ~1) - FPR_BASE],
+					  addr & 1, data);
+				break;
+			}
+			set_fpr64(&fregs[addr - FPR_BASE], 0, data);
+			break;
+		}
+		case FPC_CSR:
+			child->thread.fpu.fcr31 = data;
+			break;
+#endif /* CONFIG_MIPS_FP_SUPPORT */
+		case PC:
+			regs->cp0_epc = data;
+			break;
+		case MMHI:
+			regs->hi = data;
+			break;
+		case MMLO:
+			regs->lo = data;
+			break;
+		case DSP_BASE ... DSP_BASE + 5: {
+			dspreg_t *dregs;
+
+			if (!cpu_has_dsp) {
+				ret = -EIO;
+				break;
+			}
+
+			dregs = __get_dsp_regs(child);
+			dregs[addr - DSP_BASE] = data;
+			break;
+		}
+		case DSP_CONTROL:
+			if (!cpu_has_dsp) {
+				ret = -EIO;
+				break;
+			}
+			child->thread.dsp.dspcontrol = data;
+			break;
+		default:
+			/* The rest are not allowed. */
+			ret = -EIO;
+			break;
+		}
+		break;
+		}
+
+	case PTRACE_GETREGS:
+		ret = ptrace_getregs(child,
+				(struct user_pt_regs __user *) (__u64) data);
+		break;
+
+	case PTRACE_SETREGS:
+		ret = ptrace_setregs(child,
+				(struct user_pt_regs __user *) (__u64) data);
+		break;
+
+#ifdef CONFIG_MIPS_FP_SUPPORT
+	case PTRACE_GETFPREGS:
+		ret = ptrace_getfpregs(child, (__u32 __user *) (__u64) data);
+		break;
+
+	case PTRACE_SETFPREGS:
+		ret = ptrace_setfpregs(child, (__u32 __user *) (__u64) data);
+		break;
+#endif
+	case PTRACE_GET_THREAD_AREA:
+		ret = put_user(task_thread_info(child)->tp_value,
+				(unsigned int __user *) (unsigned long) data);
+		break;
+
+	case PTRACE_GET_THREAD_AREA_3264:
+		ret = put_user(task_thread_info(child)->tp_value,
+				(unsigned long __user *) (unsigned long) data);
+		break;
+
+	case PTRACE_GET_WATCH_REGS:
+		ret = ptrace_get_watch_regs(child,
+			(struct pt_watch_regs __user *) (unsigned long) addr);
+		break;
+
+	case PTRACE_SET_WATCH_REGS:
+		ret = ptrace_set_watch_regs(child,
+			(struct pt_watch_regs __user *) (unsigned long) addr);
+		break;
+
+	default:
+		ret = compat_ptrace_request(child, request, addr, data);
+		break;
+	}
+out:
+	return ret;
+}
diff --git a/arch/mips/kernel/r2300_fpu.S b/arch/mips/kernel/r2300_fpu.S
new file mode 100644
index 000000000..cbf6db98c
--- /dev/null
+++ b/arch/mips/kernel/r2300_fpu.S
@@ -0,0 +1,130 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1996, 1998 by Ralf Baechle
+ *
+ * Multi-arch abstraction and asm macros for easier reading:
+ * Copyright (C) 1996 David S. Miller (davem@davemloft.net)
+ *
+ * Further modifications to make this work:
+ * Copyright (c) 1998 Harald Koerfgen
+ */
+#include <asm/asm.h>
+#include <asm/asmmacro.h>
+#include <asm/errno.h>
+#include <asm/export.h>
+#include <asm/fpregdef.h>
+#include <asm/mipsregs.h>
+#include <asm/asm-offsets.h>
+#include <asm/regdef.h>
+
+#define EX(a,b)							\
+9:	a,##b;							\
+	.section __ex_table,"a";				\
+	PTR	9b,fault;					\
+	.previous
+
+#define EX2(a,b)						\
+9:	a,##b;							\
+	.section __ex_table,"a";				\
+	PTR	9b,fault;					\
+	PTR	9b+4,fault;					\
+	.previous
+
+	.set	mips1
+
+/*
+ * Save a thread's fp context.
+ */
+LEAF(_save_fp)
+EXPORT_SYMBOL(_save_fp)
+	fpu_save_single a0, t1			# clobbers t1
+	jr	ra
+	END(_save_fp)
+
+/*
+ * Restore a thread's fp context.
+ */
+LEAF(_restore_fp)
+	fpu_restore_single a0, t1		# clobbers t1
+	jr	ra
+	END(_restore_fp)
+
+	.set	noreorder
+
+/**
+ * _save_fp_context() - save FP context from the FPU
+ * @a0 - pointer to fpregs field of sigcontext
+ * @a1 - pointer to fpc_csr field of sigcontext
+ *
+ * Save FP context, including the 32 FP data registers and the FP
+ * control & status register, from the FPU to signal context.
+ */
+LEAF(_save_fp_context)
+	.set	push
+	SET_HARDFLOAT
+	li	v0, 0					# assume success
+	cfc1	t1, fcr31
+	EX2(s.d $f0, 0(a0))
+	EX2(s.d $f2, 16(a0))
+	EX2(s.d $f4, 32(a0))
+	EX2(s.d $f6, 48(a0))
+	EX2(s.d $f8, 64(a0))
+	EX2(s.d $f10, 80(a0))
+	EX2(s.d $f12, 96(a0))
+	EX2(s.d $f14, 112(a0))
+	EX2(s.d $f16, 128(a0))
+	EX2(s.d $f18, 144(a0))
+	EX2(s.d $f20, 160(a0))
+	EX2(s.d $f22, 176(a0))
+	EX2(s.d $f24, 192(a0))
+	EX2(s.d $f26, 208(a0))
+	EX2(s.d $f28, 224(a0))
+	EX2(s.d $f30, 240(a0))
+	jr	ra
+	 EX(sw	t1, (a1))
+	.set	pop
+	END(_save_fp_context)
+
+/**
+ * _restore_fp_context() - restore FP context to the FPU
+ * @a0 - pointer to fpregs field of sigcontext
+ * @a1 - pointer to fpc_csr field of sigcontext
+ *
+ * Restore FP context, including the 32 FP data registers and the FP
+ * control & status register, from signal context to the FPU.
+ */
+LEAF(_restore_fp_context)
+	.set	push
+	SET_HARDFLOAT
+	li	v0, 0					# assume success
+	EX(lw t0, (a1))
+	EX2(l.d $f0, 0(a0))
+	EX2(l.d $f2, 16(a0))
+	EX2(l.d $f4, 32(a0))
+	EX2(l.d $f6, 48(a0))
+	EX2(l.d $f8, 64(a0))
+	EX2(l.d $f10, 80(a0))
+	EX2(l.d $f12, 96(a0))
+	EX2(l.d $f14, 112(a0))
+	EX2(l.d $f16, 128(a0))
+	EX2(l.d $f18, 144(a0))
+	EX2(l.d $f20, 160(a0))
+	EX2(l.d $f22, 176(a0))
+	EX2(l.d $f24, 192(a0))
+	EX2(l.d $f26, 208(a0))
+	EX2(l.d $f28, 224(a0))
+	EX2(l.d $f30, 240(a0))
+	jr	ra
+	 ctc1	t0, fcr31
+	.set	pop
+	END(_restore_fp_context)
+	.set	reorder
+
+	.type	fault, @function
+	.ent	fault
+fault:	li	v0, -EFAULT
+	jr	ra
+	.end	fault
diff --git a/arch/mips/kernel/r2300_switch.S b/arch/mips/kernel/r2300_switch.S
new file mode 100644
index 000000000..71b1aafae
--- /dev/null
+++ b/arch/mips/kernel/r2300_switch.S
@@ -0,0 +1,65 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * r2300_switch.S: R2300 specific task switching code.
+ *
+ * Copyright (C) 1994, 1995, 1996, 1999 by Ralf Baechle
+ * Copyright (C) 1994, 1995, 1996 by Andreas Busse
+ *
+ * Multi-cpu abstraction and macros for easier reading:
+ * Copyright (C) 1996 David S. Miller (davem@davemloft.net)
+ *
+ * Further modifications to make this work:
+ * Copyright (c) 1998-2000 Harald Koerfgen
+ */
+#include <asm/asm.h>
+#include <asm/cachectl.h>
+#include <asm/export.h>
+#include <asm/fpregdef.h>
+#include <asm/mipsregs.h>
+#include <asm/asm-offsets.h>
+#include <asm/regdef.h>
+#include <asm/stackframe.h>
+#include <asm/thread_info.h>
+
+#include <asm/asmmacro.h>
+
+	.set	mips1
+	.align	5
+
+/*
+ * task_struct *resume(task_struct *prev, task_struct *next,
+ *		       struct thread_info *next_ti)
+ */
+LEAF(resume)
+	mfc0	t1, CP0_STATUS
+	sw	t1, THREAD_STATUS(a0)
+	cpu_save_nonscratch a0
+	sw	ra, THREAD_REG31(a0)
+
+#if defined(CONFIG_STACKPROTECTOR) && !defined(CONFIG_SMP)
+	PTR_LA	t8, __stack_chk_guard
+	LONG_L	t9, TASK_STACK_CANARY(a1)
+	LONG_S	t9, 0(t8)
+#endif
+
+	/*
+	 * The order of restoring the registers takes care of the race
+	 * updating $28, $29 and kernelsp without disabling ints.
+	 */
+	move	$28, a2
+	cpu_restore_nonscratch a1
+
+	addiu	t1, $28, _THREAD_SIZE - 32
+	sw	t1, kernelsp
+
+	mfc0	t1, CP0_STATUS		/* Do we really need this? */
+	li	a3, 0xff01
+	and	t1, a3
+	lw	a2, THREAD_STATUS(a1)
+	nor	a3, $0, a3
+	and	a2, a3
+	or	a2, t1
+	mtc0	a2, CP0_STATUS
+	move	v0, a0
+	jr	ra
+	END(resume)
diff --git a/arch/mips/kernel/r4k-bugs64.c b/arch/mips/kernel/r4k-bugs64.c
new file mode 100644
index 000000000..1ff19f1ea
--- /dev/null
+++ b/arch/mips/kernel/r4k-bugs64.c
@@ -0,0 +1,322 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2003, 2004, 2007  Maciej W. Rozycki
+ */
+#include <linux/context_tracking.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/ptrace.h>
+#include <linux/stddef.h>
+
+#include <asm/bugs.h>
+#include <asm/compiler.h>
+#include <asm/cpu.h>
+#include <asm/fpu.h>
+#include <asm/mipsregs.h>
+#include <asm/setup.h>
+
+static char bug64hit[] __initdata =
+	"reliable operation impossible!\n%s";
+static char nowar[] __initdata =
+	"Please report to <linux-mips@linux-mips.org>.";
+static char r4kwar[] __initdata =
+	"Enable CPU_R4000_WORKAROUNDS to rectify.";
+static char daddiwar[] __initdata =
+	"Enable CPU_DADDI_WORKAROUNDS to rectify.";
+
+static __always_inline __init
+void align_mod(const int align, const int mod)
+{
+	asm volatile(
+		".set	push\n\t"
+		".set	noreorder\n\t"
+		".balign %0\n\t"
+		".rept	%1\n\t"
+		"nop\n\t"
+		".endr\n\t"
+		".set	pop"
+		:
+		: "n"(align), "n"(mod));
+}
+
+static __always_inline __init
+void mult_sh_align_mod(long *v1, long *v2, long *w,
+		       const int align, const int mod)
+{
+	unsigned long flags;
+	int m1, m2;
+	long p, s, lv1, lv2, lw;
+
+	/*
+	 * We want the multiply and the shift to be isolated from the
+	 * rest of the code to disable gcc optimizations.  Hence the
+	 * asm statements that execute nothing, but make gcc not know
+	 * what the values of m1, m2 and s are and what lv2 and p are
+	 * used for.
+	 */
+
+	local_irq_save(flags);
+	/*
+	 * The following code leads to a wrong result of the first
+	 * dsll32 when executed on R4000 rev. 2.2 or 3.0 (PRId
+	 * 00000422 or 00000430, respectively).
+	 *
+	 * See "MIPS R4000PC/SC Errata, Processor Revision 2.2 and
+	 * 3.0" by MIPS Technologies, Inc., errata #16 and #28 for
+	 * details.  I got no permission to duplicate them here,
+	 * sigh... --macro
+	 */
+	asm volatile(
+		""
+		: "=r" (m1), "=r" (m2), "=r" (s)
+		: "0" (5), "1" (8), "2" (5));
+	align_mod(align, mod);
+	/*
+	 * The trailing nop is needed to fulfill the two-instruction
+	 * requirement between reading hi/lo and staring a mult/div.
+	 * Leaving it out may cause gas insert a nop itself breaking
+	 * the desired alignment of the next chunk.
+	 */
+	asm volatile(
+		".set	push\n\t"
+		".set	noat\n\t"
+		".set	noreorder\n\t"
+		".set	nomacro\n\t"
+		"mult	%2, %3\n\t"
+		"dsll32 %0, %4, %5\n\t"
+		"mflo	$0\n\t"
+		"dsll32 %1, %4, %5\n\t"
+		"nop\n\t"
+		".set	pop"
+		: "=&r" (lv1), "=r" (lw)
+		: "r" (m1), "r" (m2), "r" (s), "I" (0)
+		: "hi", "lo", "$0");
+	/* We have to use single integers for m1 and m2 and a double
+	 * one for p to be sure the mulsidi3 gcc's RTL multiplication
+	 * instruction has the workaround applied.  Older versions of
+	 * gcc have correct umulsi3 and mulsi3, but other
+	 * multiplication variants lack the workaround.
+	 */
+	asm volatile(
+		""
+		: "=r" (m1), "=r" (m2), "=r" (s)
+		: "0" (m1), "1" (m2), "2" (s));
+	align_mod(align, mod);
+	p = m1 * m2;
+	lv2 = s << 32;
+	asm volatile(
+		""
+		: "=r" (lv2)
+		: "0" (lv2), "r" (p));
+	local_irq_restore(flags);
+
+	*v1 = lv1;
+	*v2 = lv2;
+	*w = lw;
+}
+
+static __always_inline __init void check_mult_sh(void)
+{
+	long v1[8], v2[8], w[8];
+	int bug, fix, i;
+
+	printk("Checking for the multiply/shift bug... ");
+
+	/*
+	 * Testing discovered false negatives for certain code offsets
+	 * into cache lines.  Hence we test all possible offsets for
+	 * the worst assumption of an R4000 I-cache line width of 32
+	 * bytes.
+	 *
+	 * We can't use a loop as alignment directives need to be
+	 * immediates.
+	 */
+	mult_sh_align_mod(&v1[0], &v2[0], &w[0], 32, 0);
+	mult_sh_align_mod(&v1[1], &v2[1], &w[1], 32, 1);
+	mult_sh_align_mod(&v1[2], &v2[2], &w[2], 32, 2);
+	mult_sh_align_mod(&v1[3], &v2[3], &w[3], 32, 3);
+	mult_sh_align_mod(&v1[4], &v2[4], &w[4], 32, 4);
+	mult_sh_align_mod(&v1[5], &v2[5], &w[5], 32, 5);
+	mult_sh_align_mod(&v1[6], &v2[6], &w[6], 32, 6);
+	mult_sh_align_mod(&v1[7], &v2[7], &w[7], 32, 7);
+
+	bug = 0;
+	for (i = 0; i < 8; i++)
+		if (v1[i] != w[i])
+			bug = 1;
+
+	if (bug == 0) {
+		pr_cont("no.\n");
+		return;
+	}
+
+	pr_cont("yes, workaround... ");
+
+	fix = 1;
+	for (i = 0; i < 8; i++)
+		if (v2[i] != w[i])
+			fix = 0;
+
+	if (fix == 1) {
+		pr_cont("yes.\n");
+		return;
+	}
+
+	pr_cont("no.\n");
+	panic(bug64hit, !R4000_WAR ? r4kwar : nowar);
+}
+
+static volatile int daddi_ov;
+
+asmlinkage void __init do_daddi_ov(struct pt_regs *regs)
+{
+	enum ctx_state prev_state;
+
+	prev_state = exception_enter();
+	daddi_ov = 1;
+	regs->cp0_epc += 4;
+	exception_exit(prev_state);
+}
+
+static __init void check_daddi(void)
+{
+	extern asmlinkage void handle_daddi_ov(void);
+	unsigned long flags;
+	void *handler;
+	long v, tmp;
+
+	printk("Checking for the daddi bug... ");
+
+	local_irq_save(flags);
+	handler = set_except_vector(EXCCODE_OV, handle_daddi_ov);
+	/*
+	 * The following code fails to trigger an overflow exception
+	 * when executed on R4000 rev. 2.2 or 3.0 (PRId 00000422 or
+	 * 00000430, respectively).
+	 *
+	 * See "MIPS R4000PC/SC Errata, Processor Revision 2.2 and
+	 * 3.0" by MIPS Technologies, Inc., erratum #23 for details.
+	 * I got no permission to duplicate it here, sigh... --macro
+	 */
+	asm volatile(
+		".set	push\n\t"
+		".set	noat\n\t"
+		".set	noreorder\n\t"
+		".set	nomacro\n\t"
+		"addiu	%1, $0, %2\n\t"
+		"dsrl	%1, %1, 1\n\t"
+#ifdef HAVE_AS_SET_DADDI
+		".set	daddi\n\t"
+#endif
+		"daddi	%0, %1, %3\n\t"
+		".set	pop"
+		: "=r" (v), "=&r" (tmp)
+		: "I" (0xffffffffffffdb9aUL), "I" (0x1234));
+	set_except_vector(EXCCODE_OV, handler);
+	local_irq_restore(flags);
+
+	if (daddi_ov) {
+		pr_cont("no.\n");
+		return;
+	}
+
+	pr_cont("yes, workaround... ");
+
+	local_irq_save(flags);
+	handler = set_except_vector(EXCCODE_OV, handle_daddi_ov);
+	asm volatile(
+		"addiu	%1, $0, %2\n\t"
+		"dsrl	%1, %1, 1\n\t"
+		"daddi	%0, %1, %3"
+		: "=r" (v), "=&r" (tmp)
+		: "I" (0xffffffffffffdb9aUL), "I" (0x1234));
+	set_except_vector(EXCCODE_OV, handler);
+	local_irq_restore(flags);
+
+	if (daddi_ov) {
+		pr_cont("yes.\n");
+		return;
+	}
+
+	pr_cont("no.\n");
+	panic(bug64hit, !DADDI_WAR ? daddiwar : nowar);
+}
+
+int daddiu_bug	= -1;
+
+static __init void check_daddiu(void)
+{
+	long v, w, tmp;
+
+	printk("Checking for the daddiu bug... ");
+
+	/*
+	 * The following code leads to a wrong result of daddiu when
+	 * executed on R4400 rev. 1.0 (PRId 00000440).
+	 *
+	 * See "MIPS R4400PC/SC Errata, Processor Revision 1.0" by
+	 * MIPS Technologies, Inc., erratum #7 for details.
+	 *
+	 * According to "MIPS R4000PC/SC Errata, Processor Revision
+	 * 2.2 and 3.0" by MIPS Technologies, Inc., erratum #41 this
+	 * problem affects R4000 rev. 2.2 and 3.0 (PRId 00000422 and
+	 * 00000430, respectively), too.  Testing failed to trigger it
+	 * so far.
+	 *
+	 * I got no permission to duplicate the errata here, sigh...
+	 * --macro
+	 */
+	asm volatile(
+		".set	push\n\t"
+		".set	noat\n\t"
+		".set	noreorder\n\t"
+		".set	nomacro\n\t"
+		"addiu	%2, $0, %3\n\t"
+		"dsrl	%2, %2, 1\n\t"
+#ifdef HAVE_AS_SET_DADDI
+		".set	daddi\n\t"
+#endif
+		"daddiu %0, %2, %4\n\t"
+		"addiu	%1, $0, %4\n\t"
+		"daddu	%1, %2\n\t"
+		".set	pop"
+		: "=&r" (v), "=&r" (w), "=&r" (tmp)
+		: "I" (0xffffffffffffdb9aUL), "I" (0x1234));
+
+	daddiu_bug = v != w;
+
+	if (!daddiu_bug) {
+		pr_cont("no.\n");
+		return;
+	}
+
+	pr_cont("yes, workaround... ");
+
+	asm volatile(
+		"addiu	%2, $0, %3\n\t"
+		"dsrl	%2, %2, 1\n\t"
+		"daddiu %0, %2, %4\n\t"
+		"addiu	%1, $0, %4\n\t"
+		"daddu	%1, %2"
+		: "=&r" (v), "=&r" (w), "=&r" (tmp)
+		: "I" (0xffffffffffffdb9aUL), "I" (0x1234));
+
+	if (v == w) {
+		pr_cont("yes.\n");
+		return;
+	}
+
+	pr_cont("no.\n");
+	panic(bug64hit, !DADDI_WAR ? daddiwar : nowar);
+}
+
+void __init check_bugs64_early(void)
+{
+	check_mult_sh();
+	check_daddiu();
+}
+
+void __init check_bugs64(void)
+{
+	check_daddi();
+}
diff --git a/arch/mips/kernel/r4k_fpu.S b/arch/mips/kernel/r4k_fpu.S
new file mode 100644
index 000000000..b91e91106
--- /dev/null
+++ b/arch/mips/kernel/r4k_fpu.S
@@ -0,0 +1,417 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1996, 98, 99, 2000, 01 Ralf Baechle
+ *
+ * Multi-arch abstraction and asm macros for easier reading:
+ * Copyright (C) 1996 David S. Miller (davem@davemloft.net)
+ *
+ * Carsten Langgaard, carstenl@mips.com
+ * Copyright (C) 2000 MIPS Technologies, Inc.
+ * Copyright (C) 1999, 2001 Silicon Graphics, Inc.
+ */
+#include <asm/asm.h>
+#include <asm/asmmacro.h>
+#include <asm/errno.h>
+#include <asm/export.h>
+#include <asm/fpregdef.h>
+#include <asm/mipsregs.h>
+#include <asm/asm-offsets.h>
+#include <asm/regdef.h>
+
+/* preprocessor replaces the fp in ".set fp=64" with $30 otherwise */
+#undef fp
+
+	.macro	EX insn, reg, src
+	.set	push
+	SET_HARDFLOAT
+	.set	nomacro
+.ex\@:	\insn	\reg, \src
+	.set	pop
+	.section __ex_table,"a"
+	PTR	.ex\@, fault
+	.previous
+	.endm
+
+/*
+ * Save a thread's fp context.
+ */
+LEAF(_save_fp)
+EXPORT_SYMBOL(_save_fp)
+#if defined(CONFIG_64BIT) || defined(CONFIG_CPU_MIPSR2) || \
+    defined(CONFIG_CPU_MIPSR5) || defined(CONFIG_CPU_MIPSR6)
+	mfc0	t0, CP0_STATUS
+#endif
+	fpu_save_double a0 t0 t1		# clobbers t1
+	jr	ra
+	END(_save_fp)
+
+/*
+ * Restore a thread's fp context.
+ */
+LEAF(_restore_fp)
+#if defined(CONFIG_64BIT) || defined(CONFIG_CPU_MIPSR2) || \
+    defined(CONFIG_CPU_MIPSR5) || defined(CONFIG_CPU_MIPSR6)
+	mfc0	t0, CP0_STATUS
+#endif
+	fpu_restore_double a0 t0 t1		# clobbers t1
+	jr	ra
+	END(_restore_fp)
+
+#ifdef CONFIG_CPU_HAS_MSA
+
+/*
+ * Save a thread's MSA vector context.
+ */
+LEAF(_save_msa)
+EXPORT_SYMBOL(_save_msa)
+	msa_save_all	a0
+	jr	ra
+	END(_save_msa)
+
+/*
+ * Restore a thread's MSA vector context.
+ */
+LEAF(_restore_msa)
+	msa_restore_all	a0
+	jr	ra
+	END(_restore_msa)
+
+LEAF(_init_msa_upper)
+	msa_init_all_upper
+	jr	ra
+	END(_init_msa_upper)
+
+#endif
+
+	.set	noreorder
+
+/**
+ * _save_fp_context() - save FP context from the FPU
+ * @a0 - pointer to fpregs field of sigcontext
+ * @a1 - pointer to fpc_csr field of sigcontext
+ *
+ * Save FP context, including the 32 FP data registers and the FP
+ * control & status register, from the FPU to signal context.
+ */
+LEAF(_save_fp_context)
+	.set	push
+	SET_HARDFLOAT
+	cfc1	t1, fcr31
+	.set	pop
+
+#if defined(CONFIG_64BIT) || defined(CONFIG_CPU_MIPSR2) || \
+    defined(CONFIG_CPU_MIPSR5) || defined(CONFIG_CPU_MIPSR6)
+	.set	push
+	SET_HARDFLOAT
+#if defined(CONFIG_CPU_MIPSR2) || defined(CONFIG_CPU_MIPSR5)
+	.set	mips32r2
+	.set	fp=64
+	mfc0	t0, CP0_STATUS
+	sll	t0, t0, 5
+	bgez	t0, 1f			# skip storing odd if FR=0
+	 nop
+#endif
+	/* Store the 16 odd double precision registers */
+	EX	sdc1 $f1, 8(a0)
+	EX	sdc1 $f3, 24(a0)
+	EX	sdc1 $f5, 40(a0)
+	EX	sdc1 $f7, 56(a0)
+	EX	sdc1 $f9, 72(a0)
+	EX	sdc1 $f11, 88(a0)
+	EX	sdc1 $f13, 104(a0)
+	EX	sdc1 $f15, 120(a0)
+	EX	sdc1 $f17, 136(a0)
+	EX	sdc1 $f19, 152(a0)
+	EX	sdc1 $f21, 168(a0)
+	EX	sdc1 $f23, 184(a0)
+	EX	sdc1 $f25, 200(a0)
+	EX	sdc1 $f27, 216(a0)
+	EX	sdc1 $f29, 232(a0)
+	EX	sdc1 $f31, 248(a0)
+1:	.set	pop
+#endif
+
+	.set push
+	SET_HARDFLOAT
+	/* Store the 16 even double precision registers */
+	EX	sdc1 $f0, 0(a0)
+	EX	sdc1 $f2, 16(a0)
+	EX	sdc1 $f4, 32(a0)
+	EX	sdc1 $f6, 48(a0)
+	EX	sdc1 $f8, 64(a0)
+	EX	sdc1 $f10, 80(a0)
+	EX	sdc1 $f12, 96(a0)
+	EX	sdc1 $f14, 112(a0)
+	EX	sdc1 $f16, 128(a0)
+	EX	sdc1 $f18, 144(a0)
+	EX	sdc1 $f20, 160(a0)
+	EX	sdc1 $f22, 176(a0)
+	EX	sdc1 $f24, 192(a0)
+	EX	sdc1 $f26, 208(a0)
+	EX	sdc1 $f28, 224(a0)
+	EX	sdc1 $f30, 240(a0)
+	EX	sw t1, 0(a1)
+	jr	ra
+	 li	v0, 0					# success
+	.set pop
+	END(_save_fp_context)
+
+/**
+ * _restore_fp_context() - restore FP context to the FPU
+ * @a0 - pointer to fpregs field of sigcontext
+ * @a1 - pointer to fpc_csr field of sigcontext
+ *
+ * Restore FP context, including the 32 FP data registers and the FP
+ * control & status register, from signal context to the FPU.
+ */
+LEAF(_restore_fp_context)
+	EX	lw t1, 0(a1)
+
+#if defined(CONFIG_64BIT) || defined(CONFIG_CPU_MIPSR2) || \
+    defined(CONFIG_CPU_MIPSR5) || defined(CONFIG_CPU_MIPSR6)
+	.set	push
+	SET_HARDFLOAT
+#if defined(CONFIG_CPU_MIPSR2) || defined(CONFIG_CPU_MIPSR5)
+	.set	mips32r2
+	.set	fp=64
+	mfc0	t0, CP0_STATUS
+	sll	t0, t0, 5
+	bgez	t0, 1f			# skip loading odd if FR=0
+	 nop
+#endif
+	EX	ldc1 $f1, 8(a0)
+	EX	ldc1 $f3, 24(a0)
+	EX	ldc1 $f5, 40(a0)
+	EX	ldc1 $f7, 56(a0)
+	EX	ldc1 $f9, 72(a0)
+	EX	ldc1 $f11, 88(a0)
+	EX	ldc1 $f13, 104(a0)
+	EX	ldc1 $f15, 120(a0)
+	EX	ldc1 $f17, 136(a0)
+	EX	ldc1 $f19, 152(a0)
+	EX	ldc1 $f21, 168(a0)
+	EX	ldc1 $f23, 184(a0)
+	EX	ldc1 $f25, 200(a0)
+	EX	ldc1 $f27, 216(a0)
+	EX	ldc1 $f29, 232(a0)
+	EX	ldc1 $f31, 248(a0)
+1:	.set pop
+#endif
+	.set push
+	SET_HARDFLOAT
+	EX	ldc1 $f0, 0(a0)
+	EX	ldc1 $f2, 16(a0)
+	EX	ldc1 $f4, 32(a0)
+	EX	ldc1 $f6, 48(a0)
+	EX	ldc1 $f8, 64(a0)
+	EX	ldc1 $f10, 80(a0)
+	EX	ldc1 $f12, 96(a0)
+	EX	ldc1 $f14, 112(a0)
+	EX	ldc1 $f16, 128(a0)
+	EX	ldc1 $f18, 144(a0)
+	EX	ldc1 $f20, 160(a0)
+	EX	ldc1 $f22, 176(a0)
+	EX	ldc1 $f24, 192(a0)
+	EX	ldc1 $f26, 208(a0)
+	EX	ldc1 $f28, 224(a0)
+	EX	ldc1 $f30, 240(a0)
+	ctc1	t1, fcr31
+	.set pop
+	jr	ra
+	 li	v0, 0					# success
+	END(_restore_fp_context)
+
+#ifdef CONFIG_CPU_HAS_MSA
+
+	.macro	op_one_wr	op, idx, base
+	.align	4
+\idx:	\op	\idx, 0, \base
+	jr	ra
+	 nop
+	.endm
+
+	.macro	op_msa_wr	name, op
+LEAF(\name)
+	.set		push
+	.set		noreorder
+	sll		t0, a0, 4
+	PTR_LA		t1, 0f
+	PTR_ADDU	t0, t0, t1
+	jr		t0
+	  nop
+	op_one_wr	\op, 0, a1
+	op_one_wr	\op, 1, a1
+	op_one_wr	\op, 2, a1
+	op_one_wr	\op, 3, a1
+	op_one_wr	\op, 4, a1
+	op_one_wr	\op, 5, a1
+	op_one_wr	\op, 6, a1
+	op_one_wr	\op, 7, a1
+	op_one_wr	\op, 8, a1
+	op_one_wr	\op, 9, a1
+	op_one_wr	\op, 10, a1
+	op_one_wr	\op, 11, a1
+	op_one_wr	\op, 12, a1
+	op_one_wr	\op, 13, a1
+	op_one_wr	\op, 14, a1
+	op_one_wr	\op, 15, a1
+	op_one_wr	\op, 16, a1
+	op_one_wr	\op, 17, a1
+	op_one_wr	\op, 18, a1
+	op_one_wr	\op, 19, a1
+	op_one_wr	\op, 20, a1
+	op_one_wr	\op, 21, a1
+	op_one_wr	\op, 22, a1
+	op_one_wr	\op, 23, a1
+	op_one_wr	\op, 24, a1
+	op_one_wr	\op, 25, a1
+	op_one_wr	\op, 26, a1
+	op_one_wr	\op, 27, a1
+	op_one_wr	\op, 28, a1
+	op_one_wr	\op, 29, a1
+	op_one_wr	\op, 30, a1
+	op_one_wr	\op, 31, a1
+	.set		pop
+	END(\name)
+	.endm
+
+	op_msa_wr	read_msa_wr_b, st_b
+	op_msa_wr	read_msa_wr_h, st_h
+	op_msa_wr	read_msa_wr_w, st_w
+	op_msa_wr	read_msa_wr_d, st_d
+
+	op_msa_wr	write_msa_wr_b, ld_b
+	op_msa_wr	write_msa_wr_h, ld_h
+	op_msa_wr	write_msa_wr_w, ld_w
+	op_msa_wr	write_msa_wr_d, ld_d
+
+#endif /* CONFIG_CPU_HAS_MSA */
+
+#ifdef CONFIG_CPU_HAS_MSA
+
+	.macro	save_msa_upper	wr, off, base
+	.set	push
+	.set	noat
+#ifdef CONFIG_64BIT
+	copy_s_d \wr, 1
+	EX sd	$1, \off(\base)
+#elif defined(CONFIG_CPU_LITTLE_ENDIAN)
+	copy_s_w \wr, 2
+	EX sw	$1, \off(\base)
+	copy_s_w \wr, 3
+	EX sw	$1, (\off+4)(\base)
+#else /* CONFIG_CPU_BIG_ENDIAN */
+	copy_s_w \wr, 2
+	EX sw	$1, (\off+4)(\base)
+	copy_s_w \wr, 3
+	EX sw	$1, \off(\base)
+#endif
+	.set	pop
+	.endm
+
+LEAF(_save_msa_all_upper)
+	save_msa_upper	0, 0x00, a0
+	save_msa_upper	1, 0x08, a0
+	save_msa_upper	2, 0x10, a0
+	save_msa_upper	3, 0x18, a0
+	save_msa_upper	4, 0x20, a0
+	save_msa_upper	5, 0x28, a0
+	save_msa_upper	6, 0x30, a0
+	save_msa_upper	7, 0x38, a0
+	save_msa_upper	8, 0x40, a0
+	save_msa_upper	9, 0x48, a0
+	save_msa_upper	10, 0x50, a0
+	save_msa_upper	11, 0x58, a0
+	save_msa_upper	12, 0x60, a0
+	save_msa_upper	13, 0x68, a0
+	save_msa_upper	14, 0x70, a0
+	save_msa_upper	15, 0x78, a0
+	save_msa_upper	16, 0x80, a0
+	save_msa_upper	17, 0x88, a0
+	save_msa_upper	18, 0x90, a0
+	save_msa_upper	19, 0x98, a0
+	save_msa_upper	20, 0xa0, a0
+	save_msa_upper	21, 0xa8, a0
+	save_msa_upper	22, 0xb0, a0
+	save_msa_upper	23, 0xb8, a0
+	save_msa_upper	24, 0xc0, a0
+	save_msa_upper	25, 0xc8, a0
+	save_msa_upper	26, 0xd0, a0
+	save_msa_upper	27, 0xd8, a0
+	save_msa_upper	28, 0xe0, a0
+	save_msa_upper	29, 0xe8, a0
+	save_msa_upper	30, 0xf0, a0
+	save_msa_upper	31, 0xf8, a0
+	jr	ra
+	 li	v0, 0
+	END(_save_msa_all_upper)
+
+	.macro	restore_msa_upper	wr, off, base
+	.set	push
+	.set	noat
+#ifdef CONFIG_64BIT
+	EX ld	$1, \off(\base)
+	insert_d \wr, 1
+#elif defined(CONFIG_CPU_LITTLE_ENDIAN)
+	EX lw	$1, \off(\base)
+	insert_w \wr, 2
+	EX lw	$1, (\off+4)(\base)
+	insert_w \wr, 3
+#else /* CONFIG_CPU_BIG_ENDIAN */
+	EX lw	$1, (\off+4)(\base)
+	insert_w \wr, 2
+	EX lw	$1, \off(\base)
+	insert_w \wr, 3
+#endif
+	.set	pop
+	.endm
+
+LEAF(_restore_msa_all_upper)
+	restore_msa_upper	0, 0x00, a0
+	restore_msa_upper	1, 0x08, a0
+	restore_msa_upper	2, 0x10, a0
+	restore_msa_upper	3, 0x18, a0
+	restore_msa_upper	4, 0x20, a0
+	restore_msa_upper	5, 0x28, a0
+	restore_msa_upper	6, 0x30, a0
+	restore_msa_upper	7, 0x38, a0
+	restore_msa_upper	8, 0x40, a0
+	restore_msa_upper	9, 0x48, a0
+	restore_msa_upper	10, 0x50, a0
+	restore_msa_upper	11, 0x58, a0
+	restore_msa_upper	12, 0x60, a0
+	restore_msa_upper	13, 0x68, a0
+	restore_msa_upper	14, 0x70, a0
+	restore_msa_upper	15, 0x78, a0
+	restore_msa_upper	16, 0x80, a0
+	restore_msa_upper	17, 0x88, a0
+	restore_msa_upper	18, 0x90, a0
+	restore_msa_upper	19, 0x98, a0
+	restore_msa_upper	20, 0xa0, a0
+	restore_msa_upper	21, 0xa8, a0
+	restore_msa_upper	22, 0xb0, a0
+	restore_msa_upper	23, 0xb8, a0
+	restore_msa_upper	24, 0xc0, a0
+	restore_msa_upper	25, 0xc8, a0
+	restore_msa_upper	26, 0xd0, a0
+	restore_msa_upper	27, 0xd8, a0
+	restore_msa_upper	28, 0xe0, a0
+	restore_msa_upper	29, 0xe8, a0
+	restore_msa_upper	30, 0xf0, a0
+	restore_msa_upper	31, 0xf8, a0
+	jr	ra
+	 li	v0, 0
+	END(_restore_msa_all_upper)
+
+#endif /* CONFIG_CPU_HAS_MSA */
+
+	.set	reorder
+
+	.type	fault, @function
+	.ent	fault
+fault:	li	v0, -EFAULT				# failure
+	jr	ra
+	.end	fault
diff --git a/arch/mips/kernel/r4k_switch.S b/arch/mips/kernel/r4k_switch.S
new file mode 100644
index 000000000..58232ae6c
--- /dev/null
+++ b/arch/mips/kernel/r4k_switch.S
@@ -0,0 +1,59 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1994, 1995, 1996, 1998, 1999, 2002, 2003 Ralf Baechle
+ * Copyright (C) 1996 David S. Miller (davem@davemloft.net)
+ * Copyright (C) 1994, 1995, 1996, by Andreas Busse
+ * Copyright (C) 1999 Silicon Graphics, Inc.
+ * Copyright (C) 2000 MIPS Technologies, Inc.
+ *    written by Carsten Langgaard, carstenl@mips.com
+ */
+#include <asm/asm.h>
+#include <asm/cachectl.h>
+#include <asm/mipsregs.h>
+#include <asm/asm-offsets.h>
+#include <asm/regdef.h>
+#include <asm/stackframe.h>
+#include <asm/thread_info.h>
+
+#include <asm/asmmacro.h>
+
+/*
+ * task_struct *resume(task_struct *prev, task_struct *next,
+ *		       struct thread_info *next_ti)
+ */
+	.align	5
+	LEAF(resume)
+	mfc0	t1, CP0_STATUS
+	LONG_S	t1, THREAD_STATUS(a0)
+	cpu_save_nonscratch a0
+	LONG_S	ra, THREAD_REG31(a0)
+
+#if defined(CONFIG_STACKPROTECTOR) && !defined(CONFIG_SMP)
+	PTR_LA	t8, __stack_chk_guard
+	LONG_L	t9, TASK_STACK_CANARY(a1)
+	LONG_S	t9, 0(t8)
+#endif
+
+	/*
+	 * The order of restoring the registers takes care of the race
+	 * updating $28, $29 and kernelsp without disabling ints.
+	 */
+	move	$28, a2
+	cpu_restore_nonscratch a1
+
+	PTR_ADDU	t0, $28, _THREAD_SIZE - 32
+	set_saved_sp	t0, t1, t2
+	mfc0	t1, CP0_STATUS		/* Do we really need this? */
+	li	a3, 0xff01
+	and	t1, a3
+	LONG_L	a2, THREAD_STATUS(a1)
+	nor	a3, $0, a3
+	and	a2, a3
+	or	a2, t1
+	mtc0	a2, CP0_STATUS
+	move	v0, a0
+	jr	ra
+	END(resume)
diff --git a/arch/mips/kernel/relocate.c b/arch/mips/kernel/relocate.c
new file mode 100644
index 000000000..dab8febb5
--- /dev/null
+++ b/arch/mips/kernel/relocate.c
@@ -0,0 +1,446 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Support for Kernel relocation at boot time
+ *
+ * Copyright (C) 2015, Imagination Technologies Ltd.
+ * Authors: Matt Redfearn (matt.redfearn@mips.com)
+ */
+#include <asm/bootinfo.h>
+#include <asm/cacheflush.h>
+#include <asm/fw/fw.h>
+#include <asm/sections.h>
+#include <asm/setup.h>
+#include <asm/timex.h>
+#include <linux/elf.h>
+#include <linux/kernel.h>
+#include <linux/libfdt.h>
+#include <linux/of_fdt.h>
+#include <linux/sched/task.h>
+#include <linux/start_kernel.h>
+#include <linux/string.h>
+#include <linux/printk.h>
+
+#define RELOCATED(x) ((void *)((long)x + offset))
+
+extern u32 _relocation_start[];	/* End kernel image / start relocation table */
+extern u32 _relocation_end[];	/* End relocation table */
+
+extern long __start___ex_table;	/* Start exception table */
+extern long __stop___ex_table;	/* End exception table */
+
+extern void __weak plat_fdt_relocated(void *new_location);
+
+/*
+ * This function may be defined for a platform to perform any post-relocation
+ * fixup necessary.
+ * Return non-zero to abort relocation
+ */
+int __weak plat_post_relocation(long offset)
+{
+	return 0;
+}
+
+static inline u32 __init get_synci_step(void)
+{
+	u32 res;
+
+	__asm__("rdhwr  %0, $1" : "=r" (res));
+
+	return res;
+}
+
+static void __init sync_icache(void *kbase, unsigned long kernel_length)
+{
+	void *kend = kbase + kernel_length;
+	u32 step = get_synci_step();
+
+	do {
+		__asm__ __volatile__(
+			"synci  0(%0)"
+			: /* no output */
+			: "r" (kbase));
+
+		kbase += step;
+	} while (kbase < kend);
+
+	/* Completion barrier */
+	__sync();
+}
+
+static int __init apply_r_mips_64_rel(u32 *loc_orig, u32 *loc_new, long offset)
+{
+	*(u64 *)loc_new += offset;
+
+	return 0;
+}
+
+static int __init apply_r_mips_32_rel(u32 *loc_orig, u32 *loc_new, long offset)
+{
+	*loc_new += offset;
+
+	return 0;
+}
+
+static int __init apply_r_mips_26_rel(u32 *loc_orig, u32 *loc_new, long offset)
+{
+	unsigned long target_addr = (*loc_orig) & 0x03ffffff;
+
+	if (offset % 4) {
+		pr_err("Dangerous R_MIPS_26 REL relocation\n");
+		return -ENOEXEC;
+	}
+
+	/* Original target address */
+	target_addr <<= 2;
+	target_addr += (unsigned long)loc_orig & ~0x03ffffff;
+
+	/* Get the new target address */
+	target_addr += offset;
+
+	if ((target_addr & 0xf0000000) != ((unsigned long)loc_new & 0xf0000000)) {
+		pr_err("R_MIPS_26 REL relocation overflow\n");
+		return -ENOEXEC;
+	}
+
+	target_addr -= (unsigned long)loc_new & ~0x03ffffff;
+	target_addr >>= 2;
+
+	*loc_new = (*loc_new & ~0x03ffffff) | (target_addr & 0x03ffffff);
+
+	return 0;
+}
+
+
+static int __init apply_r_mips_hi16_rel(u32 *loc_orig, u32 *loc_new, long offset)
+{
+	unsigned long insn = *loc_orig;
+	unsigned long target = (insn & 0xffff) << 16; /* high 16bits of target */
+
+	target += offset;
+
+	*loc_new = (insn & ~0xffff) | ((target >> 16) & 0xffff);
+	return 0;
+}
+
+static int (*reloc_handlers_rel[]) (u32 *, u32 *, long) __initdata = {
+	[R_MIPS_64]		= apply_r_mips_64_rel,
+	[R_MIPS_32]		= apply_r_mips_32_rel,
+	[R_MIPS_26]		= apply_r_mips_26_rel,
+	[R_MIPS_HI16]		= apply_r_mips_hi16_rel,
+};
+
+int __init do_relocations(void *kbase_old, void *kbase_new, long offset)
+{
+	u32 *r;
+	u32 *loc_orig;
+	u32 *loc_new;
+	int type;
+	int res;
+
+	for (r = _relocation_start; r < _relocation_end; r++) {
+		/* Sentinel for last relocation */
+		if (*r == 0)
+			break;
+
+		type = (*r >> 24) & 0xff;
+		loc_orig = kbase_old + ((*r & 0x00ffffff) << 2);
+		loc_new = RELOCATED(loc_orig);
+
+		if (reloc_handlers_rel[type] == NULL) {
+			/* Unsupported relocation */
+			pr_err("Unhandled relocation type %d at 0x%pK\n",
+			       type, loc_orig);
+			return -ENOEXEC;
+		}
+
+		res = reloc_handlers_rel[type](loc_orig, loc_new, offset);
+		if (res)
+			return res;
+	}
+
+	return 0;
+}
+
+/*
+ * The exception table is filled in by the relocs tool after vmlinux is linked.
+ * It must be relocated separately since there will not be any relocation
+ * information for it filled in by the linker.
+ */
+static int __init relocate_exception_table(long offset)
+{
+	unsigned long *etable_start, *etable_end, *e;
+
+	etable_start = RELOCATED(&__start___ex_table);
+	etable_end = RELOCATED(&__stop___ex_table);
+
+	for (e = etable_start; e < etable_end; e++)
+		*e += offset;
+
+	return 0;
+}
+
+#ifdef CONFIG_RANDOMIZE_BASE
+
+static inline __init unsigned long rotate_xor(unsigned long hash,
+					      const void *area, size_t size)
+{
+	const typeof(hash) *ptr = PTR_ALIGN(area, sizeof(hash));
+	size_t diff, i;
+
+	diff = (void *)ptr - area;
+	if (unlikely(size < diff + sizeof(hash)))
+		return hash;
+
+	size = ALIGN_DOWN(size - diff, sizeof(hash));
+
+	for (i = 0; i < size / sizeof(hash); i++) {
+		/* Rotate by odd number of bits and XOR. */
+		hash = (hash << ((sizeof(hash) * 8) - 7)) | (hash >> 7);
+		hash ^= ptr[i];
+	}
+
+	return hash;
+}
+
+static inline __init unsigned long get_random_boot(void)
+{
+	unsigned long entropy = random_get_entropy();
+	unsigned long hash = 0;
+
+	/* Attempt to create a simple but unpredictable starting entropy. */
+	hash = rotate_xor(hash, linux_banner, strlen(linux_banner));
+
+	/* Add in any runtime entropy we can get */
+	hash = rotate_xor(hash, &entropy, sizeof(entropy));
+
+#if defined(CONFIG_USE_OF)
+	/* Get any additional entropy passed in device tree */
+	if (initial_boot_params) {
+		int node, len;
+		u64 *prop;
+
+		node = fdt_path_offset(initial_boot_params, "/chosen");
+		if (node >= 0) {
+			prop = fdt_getprop_w(initial_boot_params, node,
+					     "kaslr-seed", &len);
+			if (prop && (len == sizeof(u64)))
+				hash = rotate_xor(hash, prop, sizeof(*prop));
+		}
+	}
+#endif /* CONFIG_USE_OF */
+
+	return hash;
+}
+
+static inline __init bool kaslr_disabled(void)
+{
+	char *str;
+
+#if defined(CONFIG_CMDLINE_BOOL)
+	const char *builtin_cmdline = CONFIG_CMDLINE;
+
+	str = strstr(builtin_cmdline, "nokaslr");
+	if (str == builtin_cmdline ||
+	    (str > builtin_cmdline && *(str - 1) == ' '))
+		return true;
+#endif
+	str = strstr(arcs_cmdline, "nokaslr");
+	if (str == arcs_cmdline || (str > arcs_cmdline && *(str - 1) == ' '))
+		return true;
+
+	return false;
+}
+
+static inline void __init *determine_relocation_address(void)
+{
+	/* Choose a new address for the kernel */
+	unsigned long kernel_length;
+	void *dest = &_text;
+	unsigned long offset;
+
+	if (kaslr_disabled())
+		return dest;
+
+	kernel_length = (long)_end - (long)(&_text);
+
+	offset = get_random_boot() << 16;
+	offset &= (CONFIG_RANDOMIZE_BASE_MAX_OFFSET - 1);
+	if (offset < kernel_length)
+		offset += ALIGN(kernel_length, 0xffff);
+
+	return RELOCATED(dest);
+}
+
+#else
+
+static inline void __init *determine_relocation_address(void)
+{
+	/*
+	 * Choose a new address for the kernel
+	 * For now we'll hard code the destination
+	 */
+	return (void *)0xffffffff81000000;
+}
+
+#endif
+
+static inline int __init relocation_addr_valid(void *loc_new)
+{
+	if ((unsigned long)loc_new & 0x0000ffff) {
+		/* Inappropriately aligned new location */
+		return 0;
+	}
+	if ((unsigned long)loc_new < (unsigned long)&_end) {
+		/* New location overlaps original kernel */
+		return 0;
+	}
+	return 1;
+}
+
+void *__init relocate_kernel(void)
+{
+	void *loc_new;
+	unsigned long kernel_length;
+	unsigned long bss_length;
+	long offset = 0;
+	int res = 1;
+	/* Default to original kernel entry point */
+	void *kernel_entry = start_kernel;
+	void *fdt = NULL;
+
+	/* Get the command line */
+	fw_init_cmdline();
+#if defined(CONFIG_USE_OF)
+	/* Deal with the device tree */
+	fdt = plat_get_fdt();
+	early_init_dt_scan(fdt);
+	if (boot_command_line[0]) {
+		/* Boot command line was passed in device tree */
+		strlcpy(arcs_cmdline, boot_command_line, COMMAND_LINE_SIZE);
+	}
+#endif /* CONFIG_USE_OF */
+
+	kernel_length = (long)(&_relocation_start) - (long)(&_text);
+	bss_length = (long)&__bss_stop - (long)&__bss_start;
+
+	loc_new = determine_relocation_address();
+
+	/* Sanity check relocation address */
+	if (relocation_addr_valid(loc_new))
+		offset = (unsigned long)loc_new - (unsigned long)(&_text);
+
+	/* Reset the command line now so we don't end up with a duplicate */
+	arcs_cmdline[0] = '\0';
+
+	if (offset) {
+		void (*fdt_relocated_)(void *) = NULL;
+#if defined(CONFIG_USE_OF)
+		unsigned long fdt_phys = virt_to_phys(fdt);
+
+		/*
+		 * If built-in dtb is used then it will have been relocated
+		 * during kernel _text relocation. If appended DTB is used
+		 * then it will not be relocated, but it should remain
+		 * intact in the original location. If dtb is loaded by
+		 * the bootloader then it may need to be moved if it crosses
+		 * the target memory area
+		 */
+
+		if (fdt_phys >= virt_to_phys(RELOCATED(&_text)) &&
+			fdt_phys <= virt_to_phys(RELOCATED(&_end))) {
+			void *fdt_relocated =
+				RELOCATED(ALIGN((long)&_end, PAGE_SIZE));
+			memcpy(fdt_relocated, fdt, fdt_totalsize(fdt));
+			fdt = fdt_relocated;
+			fdt_relocated_ = RELOCATED(&plat_fdt_relocated);
+		}
+#endif /* CONFIG_USE_OF */
+
+		/* Copy the kernel to it's new location */
+		memcpy(loc_new, &_text, kernel_length);
+
+		/* Perform relocations on the new kernel */
+		res = do_relocations(&_text, loc_new, offset);
+		if (res < 0)
+			goto out;
+
+		/* Sync the caches ready for execution of new kernel */
+		sync_icache(loc_new, kernel_length);
+
+		res = relocate_exception_table(offset);
+		if (res < 0)
+			goto out;
+
+		/*
+		 * The original .bss has already been cleared, and
+		 * some variables such as command line parameters
+		 * stored to it so make a copy in the new location.
+		 */
+		memcpy(RELOCATED(&__bss_start), &__bss_start, bss_length);
+
+		/*
+		 * If fdt was stored outside of the kernel image and
+		 * had to be moved then update platform's state data
+		 * with the new fdt location
+		 */
+		if (fdt_relocated_)
+			fdt_relocated_(fdt);
+
+		/*
+		 * Last chance for the platform to abort relocation.
+		 * This may also be used by the platform to perform any
+		 * initialisation required now that the new kernel is
+		 * resident in memory and ready to be executed.
+		 */
+		if (plat_post_relocation(offset))
+			goto out;
+
+		/* The current thread is now within the relocated image */
+		__current_thread_info = RELOCATED(&init_thread_union);
+
+		/* Return the new kernel's entry point */
+		kernel_entry = RELOCATED(start_kernel);
+	}
+out:
+	return kernel_entry;
+}
+
+/*
+ * Show relocation information on panic.
+ */
+void show_kernel_relocation(const char *level)
+{
+	unsigned long offset;
+
+	offset = __pa_symbol(_text) - __pa_symbol(VMLINUX_LOAD_ADDRESS);
+
+	if (IS_ENABLED(CONFIG_RELOCATABLE) && offset > 0) {
+		printk(level);
+		pr_cont("Kernel relocated by 0x%pK\n", (void *)offset);
+		pr_cont(" .text @ 0x%pK\n", _text);
+		pr_cont(" .data @ 0x%pK\n", _sdata);
+		pr_cont(" .bss  @ 0x%pK\n", __bss_start);
+	}
+}
+
+static int kernel_location_notifier_fn(struct notifier_block *self,
+				       unsigned long v, void *p)
+{
+	show_kernel_relocation(KERN_EMERG);
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block kernel_location_notifier = {
+	.notifier_call = kernel_location_notifier_fn
+};
+
+static int __init register_kernel_offset_dumper(void)
+{
+	atomic_notifier_chain_register(&panic_notifier_list,
+				       &kernel_location_notifier);
+	return 0;
+}
+__initcall(register_kernel_offset_dumper);
diff --git a/arch/mips/kernel/relocate_kernel.S b/arch/mips/kernel/relocate_kernel.S
new file mode 100644
index 000000000..ac870893b
--- /dev/null
+++ b/arch/mips/kernel/relocate_kernel.S
@@ -0,0 +1,190 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * relocate_kernel.S for kexec
+ * Created by <nschichan@corp.free.fr> on Thu Oct 12 17:49:57 2006
+ */
+
+#include <asm/asm.h>
+#include <asm/asmmacro.h>
+#include <asm/regdef.h>
+#include <asm/mipsregs.h>
+#include <asm/stackframe.h>
+#include <asm/addrspace.h>
+
+LEAF(relocate_new_kernel)
+	PTR_L a0,	arg0
+	PTR_L a1,	arg1
+	PTR_L a2,	arg2
+	PTR_L a3,	arg3
+
+	PTR_L		s0, kexec_indirection_page
+	PTR_L		s1, kexec_start_address
+
+process_entry:
+	PTR_L		s2, (s0)
+	PTR_ADDIU	s0, s0, SZREG
+
+	/*
+	 * In case of a kdump/crash kernel, the indirection page is not
+	 * populated as the kernel is directly copied to a reserved location
+	 */
+	beqz		s2, done
+
+	/* destination page */
+	and		s3, s2, 0x1
+	beq		s3, zero, 1f
+	and		s4, s2, ~0x1	/* store destination addr in s4 */
+	b		process_entry
+
+1:
+	/* indirection page, update s0	*/
+	and		s3, s2, 0x2
+	beq		s3, zero, 1f
+	and		s0, s2, ~0x2
+	b		process_entry
+
+1:
+	/* done page */
+	and		s3, s2, 0x4
+	beq		s3, zero, 1f
+	b		done
+1:
+	/* source page */
+	and		s3, s2, 0x8
+	beq		s3, zero, process_entry
+	and		s2, s2, ~0x8
+	li		s6, (1 << _PAGE_SHIFT) / SZREG
+
+copy_word:
+	/* copy page word by word */
+	REG_L		s5, (s2)
+	REG_S		s5, (s4)
+	PTR_ADDIU	s4, s4, SZREG
+	PTR_ADDIU	s2, s2, SZREG
+	LONG_ADDIU	s6, s6, -1
+	beq		s6, zero, process_entry
+	b		copy_word
+	b		process_entry
+
+done:
+#ifdef CONFIG_SMP
+	/* kexec_flag reset is signal to other CPUs what kernel
+	   was moved to it's location. Note - we need relocated address
+	   of kexec_flag.  */
+
+	bal		1f
+ 1:	move		t1,ra;
+	PTR_LA		t2,1b
+	PTR_LA		t0,kexec_flag
+	PTR_SUB		t0,t0,t2;
+	PTR_ADD		t0,t1,t0;
+	LONG_S		zero,(t0)
+#endif
+
+#ifdef CONFIG_CPU_CAVIUM_OCTEON
+	/* We need to flush I-cache before jumping to new kernel.
+	 * Unfortunately, this code is cpu-specific.
+	 */
+	.set push
+	.set noreorder
+	syncw
+	syncw
+	synci		0($0)
+	.set pop
+#else
+	sync
+#endif
+	/* jump to kexec_start_address */
+	j		s1
+	END(relocate_new_kernel)
+
+#ifdef CONFIG_SMP
+/*
+ * Other CPUs should wait until code is relocated and
+ * then start at entry (?) point.
+ */
+LEAF(kexec_smp_wait)
+	PTR_L		a0, s_arg0
+	PTR_L		a1, s_arg1
+	PTR_L		a2, s_arg2
+	PTR_L		a3, s_arg3
+	PTR_L		s1, kexec_start_address
+
+	/* Non-relocated address works for args and kexec_start_address ( old
+	 * kernel is not overwritten). But we need relocated address of
+	 * kexec_flag.
+	 */
+
+	bal		1f
+1:	move		t1,ra;
+	PTR_LA		t2,1b
+	PTR_LA		t0,kexec_flag
+	PTR_SUB		t0,t0,t2;
+	PTR_ADD		t0,t1,t0;
+
+1:	LONG_L		s0, (t0)
+	bne		s0, zero,1b
+
+#ifdef CONFIG_CPU_CAVIUM_OCTEON
+	.set push
+	.set noreorder
+	synci		0($0)
+	.set pop
+#else
+	sync
+#endif
+	j		s1
+	END(kexec_smp_wait)
+#endif
+
+#ifdef __mips64
+       /* all PTR's must be aligned to 8 byte in 64-bit mode */
+       .align  3
+#endif
+
+/* All parameters to new kernel are passed in registers a0-a3.
+ * kexec_args[0..3] are used to prepare register values.
+ */
+
+kexec_args:
+	EXPORT(kexec_args)
+arg0:	PTR		0x0
+arg1:	PTR		0x0
+arg2:	PTR		0x0
+arg3:	PTR		0x0
+	.size	kexec_args,PTRSIZE*4
+
+#ifdef CONFIG_SMP
+/*
+ * Secondary CPUs may have different kernel parameters in
+ * their registers a0-a3. secondary_kexec_args[0..3] are used
+ * to prepare register values.
+ */
+secondary_kexec_args:
+	EXPORT(secondary_kexec_args)
+s_arg0: PTR		0x0
+s_arg1: PTR		0x0
+s_arg2: PTR		0x0
+s_arg3: PTR		0x0
+	.size	secondary_kexec_args,PTRSIZE*4
+kexec_flag:
+	LONG		0x1
+
+#endif
+
+kexec_start_address:
+	EXPORT(kexec_start_address)
+	PTR		0x0
+	.size		kexec_start_address, PTRSIZE
+
+kexec_indirection_page:
+	EXPORT(kexec_indirection_page)
+	PTR		0
+	.size		kexec_indirection_page, PTRSIZE
+
+relocate_new_kernel_end:
+
+relocate_new_kernel_size:
+	EXPORT(relocate_new_kernel_size)
+	PTR		relocate_new_kernel_end - relocate_new_kernel
+	.size		relocate_new_kernel_size, PTRSIZE
diff --git a/arch/mips/kernel/reset.c b/arch/mips/kernel/reset.c
new file mode 100644
index 000000000..6288780b7
--- /dev/null
+++ b/arch/mips/kernel/reset.c
@@ -0,0 +1,125 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2001, 06 by Ralf Baechle (ralf@linux-mips.org)
+ * Copyright (C) 2001 MIPS Technologies, Inc.
+ */
+#include <linux/kernel.h>
+#include <linux/export.h>
+#include <linux/pm.h>
+#include <linux/types.h>
+#include <linux/reboot.h>
+#include <linux/delay.h>
+
+#include <asm/compiler.h>
+#include <asm/idle.h>
+#include <asm/mipsregs.h>
+#include <asm/reboot.h>
+
+/*
+ * Urgs ...  Too many MIPS machines to handle this in a generic way.
+ * So handle all using function pointers to machine specific
+ * functions.
+ */
+void (*_machine_restart)(char *command);
+void (*_machine_halt)(void);
+void (*pm_power_off)(void);
+
+EXPORT_SYMBOL(pm_power_off);
+
+static void machine_hang(void)
+{
+	/*
+	 * We're hanging the system so we don't want to be interrupted anymore.
+	 * Any interrupt handlers that ran would at best be useless & at worst
+	 * go awry because the system isn't in a functional state.
+	 */
+	local_irq_disable();
+
+	/*
+	 * Mask all interrupts, giving us a better chance of remaining in the
+	 * low power wait state.
+	 */
+	clear_c0_status(ST0_IM);
+
+	while (true) {
+		if (cpu_has_mips_r) {
+			/*
+			 * We know that the wait instruction is supported so
+			 * make use of it directly, leaving interrupts
+			 * disabled.
+			 */
+			asm volatile(
+				".set	push\n\t"
+				".set	" MIPS_ISA_ARCH_LEVEL "\n\t"
+				"wait\n\t"
+				".set	pop");
+		} else if (cpu_wait) {
+			/*
+			 * Try the cpu_wait() callback. This isn't ideal since
+			 * it'll re-enable interrupts, but that ought to be
+			 * harmless given that they're all masked.
+			 */
+			cpu_wait();
+			local_irq_disable();
+		} else {
+			/*
+			 * We're going to burn some power running round the
+			 * loop, but we don't really have a choice. This isn't
+			 * a path we should expect to run for long during
+			 * typical use anyway.
+			 */
+		}
+
+		/*
+		 * In most modern MIPS CPUs interrupts will cause the wait
+		 * instruction to graduate even when disabled, and in some
+		 * cases even when masked. In order to prevent a timer
+		 * interrupt from continuously taking us out of the low power
+		 * wait state, we clear any pending timer interrupt here.
+		 */
+		if (cpu_has_counter)
+			write_c0_compare(0);
+	}
+}
+
+void machine_restart(char *command)
+{
+	if (_machine_restart)
+		_machine_restart(command);
+
+#ifdef CONFIG_SMP
+	preempt_disable();
+	smp_send_stop();
+#endif
+	do_kernel_restart(command);
+	mdelay(1000);
+	pr_emerg("Reboot failed -- System halted\n");
+	machine_hang();
+}
+
+void machine_halt(void)
+{
+	if (_machine_halt)
+		_machine_halt();
+
+#ifdef CONFIG_SMP
+	preempt_disable();
+	smp_send_stop();
+#endif
+	machine_hang();
+}
+
+void machine_power_off(void)
+{
+	if (pm_power_off)
+		pm_power_off();
+
+#ifdef CONFIG_SMP
+	preempt_disable();
+	smp_send_stop();
+#endif
+	machine_hang();
+}
diff --git a/arch/mips/kernel/rtlx-cmp.c b/arch/mips/kernel/rtlx-cmp.c
new file mode 100644
index 000000000..d26dcc4b4
--- /dev/null
+++ b/arch/mips/kernel/rtlx-cmp.c
@@ -0,0 +1,122 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2005 MIPS Technologies, Inc.  All rights reserved.
+ * Copyright (C) 2013 Imagination Technologies Ltd.
+ */
+#include <linux/device.h>
+#include <linux/fs.h>
+#include <linux/err.h>
+#include <linux/wait.h>
+#include <linux/sched.h>
+#include <linux/smp.h>
+
+#include <asm/mips_mt.h>
+#include <asm/vpe.h>
+#include <asm/rtlx.h>
+
+static int major;
+
+static void rtlx_interrupt(void)
+{
+	int i;
+	struct rtlx_info *info;
+	struct rtlx_info **p = vpe_get_shared(aprp_cpu_index());
+
+	if (p == NULL || *p == NULL)
+		return;
+
+	info = *p;
+
+	if (info->ap_int_pending == 1 && smp_processor_id() == 0) {
+		for (i = 0; i < RTLX_CHANNELS; i++) {
+			wake_up(&channel_wqs[i].lx_queue);
+			wake_up(&channel_wqs[i].rt_queue);
+		}
+		info->ap_int_pending = 0;
+	}
+}
+
+void _interrupt_sp(void)
+{
+	smp_send_reschedule(aprp_cpu_index());
+}
+
+int __init rtlx_module_init(void)
+{
+	struct device *dev;
+	int i, err;
+
+	if (!cpu_has_mipsmt) {
+		pr_warn("VPE loader: not a MIPS MT capable processor\n");
+		return -ENODEV;
+	}
+
+	if (num_possible_cpus() - aprp_cpu_index() < 1) {
+		pr_warn("No TCs reserved for AP/SP, not initializing RTLX.\n"
+			"Pass maxcpus=<n> argument as kernel argument\n");
+
+		return -ENODEV;
+	}
+
+	major = register_chrdev(0, RTLX_MODULE_NAME, &rtlx_fops);
+	if (major < 0) {
+		pr_err("rtlx_module_init: unable to register device\n");
+		return major;
+	}
+
+	/* initialise the wait queues */
+	for (i = 0; i < RTLX_CHANNELS; i++) {
+		init_waitqueue_head(&channel_wqs[i].rt_queue);
+		init_waitqueue_head(&channel_wqs[i].lx_queue);
+		atomic_set(&channel_wqs[i].in_open, 0);
+		mutex_init(&channel_wqs[i].mutex);
+
+		dev = device_create(mt_class, NULL, MKDEV(major, i), NULL,
+				    "%s%d", RTLX_MODULE_NAME, i);
+		if (IS_ERR(dev)) {
+			while (i--)
+				device_destroy(mt_class, MKDEV(major, i));
+
+			err = PTR_ERR(dev);
+			goto out_chrdev;
+		}
+	}
+
+	/* set up notifiers */
+	rtlx_notify.start = rtlx_starting;
+	rtlx_notify.stop = rtlx_stopping;
+	vpe_notify(aprp_cpu_index(), &rtlx_notify);
+
+	if (cpu_has_vint) {
+		aprp_hook = rtlx_interrupt;
+	} else {
+		pr_err("APRP RTLX init on non-vectored-interrupt processor\n");
+		err = -ENODEV;
+		goto out_class;
+	}
+
+	return 0;
+
+out_class:
+	for (i = 0; i < RTLX_CHANNELS; i++)
+		device_destroy(mt_class, MKDEV(major, i));
+out_chrdev:
+	unregister_chrdev(major, RTLX_MODULE_NAME);
+
+	return err;
+}
+
+void __exit rtlx_module_exit(void)
+{
+	int i;
+
+	for (i = 0; i < RTLX_CHANNELS; i++)
+		device_destroy(mt_class, MKDEV(major, i));
+
+	unregister_chrdev(major, RTLX_MODULE_NAME);
+
+	aprp_hook = NULL;
+}
diff --git a/arch/mips/kernel/rtlx-mt.c b/arch/mips/kernel/rtlx-mt.c
new file mode 100644
index 000000000..38c6925a1
--- /dev/null
+++ b/arch/mips/kernel/rtlx-mt.c
@@ -0,0 +1,147 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2005 MIPS Technologies, Inc.  All rights reserved.
+ * Copyright (C) 2013 Imagination Technologies Ltd.
+ */
+#include <linux/device.h>
+#include <linux/fs.h>
+#include <linux/err.h>
+#include <linux/wait.h>
+#include <linux/sched.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+
+#include <asm/mips_mt.h>
+#include <asm/vpe.h>
+#include <asm/rtlx.h>
+
+static int major;
+
+static void rtlx_dispatch(void)
+{
+	if (read_c0_cause() & read_c0_status() & C_SW0)
+		do_IRQ(MIPS_CPU_IRQ_BASE + MIPS_CPU_RTLX_IRQ);
+}
+
+/*
+ * Interrupt handler may be called before rtlx_init has otherwise had
+ * a chance to run.
+ */
+static irqreturn_t rtlx_interrupt(int irq, void *dev_id)
+{
+	unsigned int vpeflags;
+	unsigned long flags;
+	int i;
+
+	local_irq_save(flags);
+	vpeflags = dvpe();
+	set_c0_status(0x100 << MIPS_CPU_RTLX_IRQ);
+	irq_enable_hazard();
+	evpe(vpeflags);
+	local_irq_restore(flags);
+
+	for (i = 0; i < RTLX_CHANNELS; i++) {
+		wake_up(&channel_wqs[i].lx_queue);
+		wake_up(&channel_wqs[i].rt_queue);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static int rtlx_irq_num = MIPS_CPU_IRQ_BASE + MIPS_CPU_RTLX_IRQ;
+
+void _interrupt_sp(void)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	dvpe();
+	settc(1);
+	write_vpe_c0_cause(read_vpe_c0_cause() | C_SW0);
+	evpe(EVPE_ENABLE);
+	local_irq_restore(flags);
+}
+
+int __init rtlx_module_init(void)
+{
+	struct device *dev;
+	int i, err;
+
+	if (!cpu_has_mipsmt) {
+		pr_warn("VPE loader: not a MIPS MT capable processor\n");
+		return -ENODEV;
+	}
+
+	if (aprp_cpu_index() == 0) {
+		pr_warn("No TCs reserved for AP/SP, not initializing RTLX.\n"
+			"Pass maxtcs=<n> argument as kernel argument\n");
+
+		return -ENODEV;
+	}
+
+	major = register_chrdev(0, RTLX_MODULE_NAME, &rtlx_fops);
+	if (major < 0) {
+		pr_err("rtlx_module_init: unable to register device\n");
+		return major;
+	}
+
+	/* initialise the wait queues */
+	for (i = 0; i < RTLX_CHANNELS; i++) {
+		init_waitqueue_head(&channel_wqs[i].rt_queue);
+		init_waitqueue_head(&channel_wqs[i].lx_queue);
+		atomic_set(&channel_wqs[i].in_open, 0);
+		mutex_init(&channel_wqs[i].mutex);
+
+		dev = device_create(mt_class, NULL, MKDEV(major, i), NULL,
+				    "%s%d", RTLX_MODULE_NAME, i);
+		if (IS_ERR(dev)) {
+			while (i--)
+				device_destroy(mt_class, MKDEV(major, i));
+
+			err = PTR_ERR(dev);
+			goto out_chrdev;
+		}
+	}
+
+	/* set up notifiers */
+	rtlx_notify.start = rtlx_starting;
+	rtlx_notify.stop = rtlx_stopping;
+	vpe_notify(aprp_cpu_index(), &rtlx_notify);
+
+	if (cpu_has_vint) {
+		aprp_hook = rtlx_dispatch;
+	} else {
+		pr_err("APRP RTLX init on non-vectored-interrupt processor\n");
+		err = -ENODEV;
+		goto out_class;
+	}
+
+	err = request_irq(rtlx_irq_num, rtlx_interrupt, 0, "RTLX", rtlx);
+	if (err)
+		goto out_class;
+
+	return 0;
+
+out_class:
+	for (i = 0; i < RTLX_CHANNELS; i++)
+		device_destroy(mt_class, MKDEV(major, i));
+out_chrdev:
+	unregister_chrdev(major, RTLX_MODULE_NAME);
+
+	return err;
+}
+
+void __exit rtlx_module_exit(void)
+{
+	int i;
+
+	for (i = 0; i < RTLX_CHANNELS; i++)
+		device_destroy(mt_class, MKDEV(major, i));
+
+	unregister_chrdev(major, RTLX_MODULE_NAME);
+
+	aprp_hook = NULL;
+}
diff --git a/arch/mips/kernel/rtlx.c b/arch/mips/kernel/rtlx.c
new file mode 100644
index 000000000..18c509c59
--- /dev/null
+++ b/arch/mips/kernel/rtlx.c
@@ -0,0 +1,409 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2005 MIPS Technologies, Inc.  All rights reserved.
+ * Copyright (C) 2005, 06 Ralf Baechle (ralf@linux-mips.org)
+ * Copyright (C) 2013 Imagination Technologies Ltd.
+ */
+#include <linux/kernel.h>
+#include <linux/fs.h>
+#include <linux/syscalls.h>
+#include <linux/moduleloader.h>
+#include <linux/atomic.h>
+#include <linux/sched/signal.h>
+
+#include <asm/mipsmtregs.h>
+#include <asm/mips_mt.h>
+#include <asm/processor.h>
+#include <asm/rtlx.h>
+#include <asm/setup.h>
+#include <asm/vpe.h>
+
+static int sp_stopping;
+struct rtlx_info *rtlx;
+struct chan_waitqueues channel_wqs[RTLX_CHANNELS];
+struct vpe_notifications rtlx_notify;
+void (*aprp_hook)(void) = NULL;
+EXPORT_SYMBOL(aprp_hook);
+
+static void __used dump_rtlx(void)
+{
+	int i;
+
+	pr_info("id 0x%lx state %d\n", rtlx->id, rtlx->state);
+
+	for (i = 0; i < RTLX_CHANNELS; i++) {
+		struct rtlx_channel *chan = &rtlx->channel[i];
+
+		pr_info(" rt_state %d lx_state %d buffer_size %d\n",
+			chan->rt_state, chan->lx_state, chan->buffer_size);
+
+		pr_info(" rt_read %d rt_write %d\n",
+			chan->rt_read, chan->rt_write);
+
+		pr_info(" lx_read %d lx_write %d\n",
+			chan->lx_read, chan->lx_write);
+
+		pr_info(" rt_buffer <%s>\n", chan->rt_buffer);
+		pr_info(" lx_buffer <%s>\n", chan->lx_buffer);
+	}
+}
+
+/* call when we have the address of the shared structure from the SP side. */
+static int rtlx_init(struct rtlx_info *rtlxi)
+{
+	if (rtlxi->id != RTLX_ID) {
+		pr_err("no valid RTLX id at 0x%p 0x%lx\n", rtlxi, rtlxi->id);
+		return -ENOEXEC;
+	}
+
+	rtlx = rtlxi;
+
+	return 0;
+}
+
+/* notifications */
+void rtlx_starting(int vpe)
+{
+	int i;
+	sp_stopping = 0;
+
+	/* force a reload of rtlx */
+	rtlx = NULL;
+
+	/* wake up any sleeping rtlx_open's */
+	for (i = 0; i < RTLX_CHANNELS; i++)
+		wake_up_interruptible(&channel_wqs[i].lx_queue);
+}
+
+void rtlx_stopping(int vpe)
+{
+	int i;
+
+	sp_stopping = 1;
+	for (i = 0; i < RTLX_CHANNELS; i++)
+		wake_up_interruptible(&channel_wqs[i].lx_queue);
+}
+
+
+int rtlx_open(int index, int can_sleep)
+{
+	struct rtlx_info **p;
+	struct rtlx_channel *chan;
+	enum rtlx_state state;
+	int ret = 0;
+
+	if (index >= RTLX_CHANNELS) {
+		pr_debug("rtlx_open index out of range\n");
+		return -ENOSYS;
+	}
+
+	if (atomic_inc_return(&channel_wqs[index].in_open) > 1) {
+		pr_debug("rtlx_open channel %d already opened\n", index);
+		ret = -EBUSY;
+		goto out_fail;
+	}
+
+	if (rtlx == NULL) {
+		p = vpe_get_shared(aprp_cpu_index());
+		if (p == NULL) {
+			if (can_sleep) {
+				ret = __wait_event_interruptible(
+					channel_wqs[index].lx_queue,
+					(p = vpe_get_shared(aprp_cpu_index())));
+				if (ret)
+					goto out_fail;
+			} else {
+				pr_debug("No SP program loaded, and device opened with O_NONBLOCK\n");
+				ret = -ENOSYS;
+				goto out_fail;
+			}
+		}
+
+		smp_rmb();
+		if (*p == NULL) {
+			if (can_sleep) {
+				DEFINE_WAIT(wait);
+
+				for (;;) {
+					prepare_to_wait(
+						&channel_wqs[index].lx_queue,
+						&wait, TASK_INTERRUPTIBLE);
+					smp_rmb();
+					if (*p != NULL)
+						break;
+					if (!signal_pending(current)) {
+						schedule();
+						continue;
+					}
+					ret = -ERESTARTSYS;
+					goto out_fail;
+				}
+				finish_wait(&channel_wqs[index].lx_queue,
+					    &wait);
+			} else {
+				pr_err(" *vpe_get_shared is NULL. Has an SP program been loaded?\n");
+				ret = -ENOSYS;
+				goto out_fail;
+			}
+		}
+
+		if ((unsigned int)*p < KSEG0) {
+			pr_warn("vpe_get_shared returned an invalid pointer maybe an error code %d\n",
+				(int)*p);
+			ret = -ENOSYS;
+			goto out_fail;
+		}
+
+		ret = rtlx_init(*p);
+		if (ret < 0)
+			goto out_ret;
+	}
+
+	chan = &rtlx->channel[index];
+
+	state = xchg(&chan->lx_state, RTLX_STATE_OPENED);
+	if (state == RTLX_STATE_OPENED) {
+		ret = -EBUSY;
+		goto out_fail;
+	}
+
+out_fail:
+	smp_mb();
+	atomic_dec(&channel_wqs[index].in_open);
+	smp_mb();
+
+out_ret:
+	return ret;
+}
+
+int rtlx_release(int index)
+{
+	if (rtlx == NULL) {
+		pr_err("rtlx_release() with null rtlx\n");
+		return 0;
+	}
+	rtlx->channel[index].lx_state = RTLX_STATE_UNUSED;
+	return 0;
+}
+
+unsigned int rtlx_read_poll(int index, int can_sleep)
+{
+	struct rtlx_channel *chan;
+
+	if (rtlx == NULL)
+		return 0;
+
+	chan = &rtlx->channel[index];
+
+	/* data available to read? */
+	if (chan->lx_read == chan->lx_write) {
+		if (can_sleep) {
+			int ret = __wait_event_interruptible(
+				channel_wqs[index].lx_queue,
+				(chan->lx_read != chan->lx_write) ||
+				sp_stopping);
+			if (ret)
+				return ret;
+
+			if (sp_stopping)
+				return 0;
+		} else
+			return 0;
+	}
+
+	return (chan->lx_write + chan->buffer_size - chan->lx_read)
+	       % chan->buffer_size;
+}
+
+static inline int write_spacefree(int read, int write, int size)
+{
+	if (read == write) {
+		/*
+		 * Never fill the buffer completely, so indexes are always
+		 * equal if empty and only empty, or !equal if data available
+		 */
+		return size - 1;
+	}
+
+	return ((read + size - write) % size) - 1;
+}
+
+unsigned int rtlx_write_poll(int index)
+{
+	struct rtlx_channel *chan = &rtlx->channel[index];
+
+	return write_spacefree(chan->rt_read, chan->rt_write,
+				chan->buffer_size);
+}
+
+ssize_t rtlx_read(int index, void __user *buff, size_t count)
+{
+	size_t lx_write, fl = 0L;
+	struct rtlx_channel *lx;
+	unsigned long failed;
+
+	if (rtlx == NULL)
+		return -ENOSYS;
+
+	lx = &rtlx->channel[index];
+
+	mutex_lock(&channel_wqs[index].mutex);
+	smp_rmb();
+	lx_write = lx->lx_write;
+
+	/* find out how much in total */
+	count = min(count,
+		     (size_t)(lx_write + lx->buffer_size - lx->lx_read)
+		     % lx->buffer_size);
+
+	/* then how much from the read pointer onwards */
+	fl = min(count, (size_t)lx->buffer_size - lx->lx_read);
+
+	failed = copy_to_user(buff, lx->lx_buffer + lx->lx_read, fl);
+	if (failed)
+		goto out;
+
+	/* and if there is anything left at the beginning of the buffer */
+	if (count - fl)
+		failed = copy_to_user(buff + fl, lx->lx_buffer, count - fl);
+
+out:
+	count -= failed;
+
+	smp_wmb();
+	lx->lx_read = (lx->lx_read + count) % lx->buffer_size;
+	smp_wmb();
+	mutex_unlock(&channel_wqs[index].mutex);
+
+	return count;
+}
+
+ssize_t rtlx_write(int index, const void __user *buffer, size_t count)
+{
+	struct rtlx_channel *rt;
+	unsigned long failed;
+	size_t rt_read;
+	size_t fl;
+
+	if (rtlx == NULL)
+		return -ENOSYS;
+
+	rt = &rtlx->channel[index];
+
+	mutex_lock(&channel_wqs[index].mutex);
+	smp_rmb();
+	rt_read = rt->rt_read;
+
+	/* total number of bytes to copy */
+	count = min_t(size_t, count, write_spacefree(rt_read, rt->rt_write,
+						     rt->buffer_size));
+
+	/* first bit from write pointer to the end of the buffer, or count */
+	fl = min(count, (size_t) rt->buffer_size - rt->rt_write);
+
+	failed = copy_from_user(rt->rt_buffer + rt->rt_write, buffer, fl);
+	if (failed)
+		goto out;
+
+	/* if there's any left copy to the beginning of the buffer */
+	if (count - fl)
+		failed = copy_from_user(rt->rt_buffer, buffer + fl, count - fl);
+
+out:
+	count -= failed;
+
+	smp_wmb();
+	rt->rt_write = (rt->rt_write + count) % rt->buffer_size;
+	smp_wmb();
+	mutex_unlock(&channel_wqs[index].mutex);
+
+	_interrupt_sp();
+
+	return count;
+}
+
+
+static int file_open(struct inode *inode, struct file *filp)
+{
+	return rtlx_open(iminor(inode), (filp->f_flags & O_NONBLOCK) ? 0 : 1);
+}
+
+static int file_release(struct inode *inode, struct file *filp)
+{
+	return rtlx_release(iminor(inode));
+}
+
+static __poll_t file_poll(struct file *file, poll_table *wait)
+{
+	int minor = iminor(file_inode(file));
+	__poll_t mask = 0;
+
+	poll_wait(file, &channel_wqs[minor].rt_queue, wait);
+	poll_wait(file, &channel_wqs[minor].lx_queue, wait);
+
+	if (rtlx == NULL)
+		return 0;
+
+	/* data available to read? */
+	if (rtlx_read_poll(minor, 0))
+		mask |= EPOLLIN | EPOLLRDNORM;
+
+	/* space to write */
+	if (rtlx_write_poll(minor))
+		mask |= EPOLLOUT | EPOLLWRNORM;
+
+	return mask;
+}
+
+static ssize_t file_read(struct file *file, char __user *buffer, size_t count,
+			 loff_t *ppos)
+{
+	int minor = iminor(file_inode(file));
+
+	/* data available? */
+	if (!rtlx_read_poll(minor, (file->f_flags & O_NONBLOCK) ? 0 : 1))
+		return 0;	/* -EAGAIN makes 'cat' whine */
+
+	return rtlx_read(minor, buffer, count);
+}
+
+static ssize_t file_write(struct file *file, const char __user *buffer,
+			  size_t count, loff_t *ppos)
+{
+	int minor = iminor(file_inode(file));
+
+	/* any space left... */
+	if (!rtlx_write_poll(minor)) {
+		int ret;
+
+		if (file->f_flags & O_NONBLOCK)
+			return -EAGAIN;
+
+		ret = __wait_event_interruptible(channel_wqs[minor].rt_queue,
+					   rtlx_write_poll(minor));
+		if (ret)
+			return ret;
+	}
+
+	return rtlx_write(minor, buffer, count);
+}
+
+const struct file_operations rtlx_fops = {
+	.owner =   THIS_MODULE,
+	.open =    file_open,
+	.release = file_release,
+	.write =   file_write,
+	.read =    file_read,
+	.poll =    file_poll,
+	.llseek =  noop_llseek,
+};
+
+module_init(rtlx_module_init);
+module_exit(rtlx_module_exit);
+
+MODULE_DESCRIPTION("MIPS RTLX");
+MODULE_AUTHOR("Elizabeth Oldham, MIPS Technologies, Inc.");
+MODULE_LICENSE("GPL");
diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S
new file mode 100644
index 000000000..b449b6866
--- /dev/null
+++ b/arch/mips/kernel/scall32-o32.S
@@ -0,0 +1,225 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1995-99, 2000- 02, 06 Ralf Baechle <ralf@linux-mips.org>
+ * Copyright (C) 2001 MIPS Technologies, Inc.
+ * Copyright (C) 2004 Thiemo Seufer
+ * Copyright (C) 2014 Imagination Technologies Ltd.
+ */
+#include <linux/errno.h>
+#include <asm/asm.h>
+#include <asm/asmmacro.h>
+#include <asm/irqflags.h>
+#include <asm/mipsregs.h>
+#include <asm/regdef.h>
+#include <asm/stackframe.h>
+#include <asm/isadep.h>
+#include <asm/sysmips.h>
+#include <asm/thread_info.h>
+#include <asm/unistd.h>
+#include <asm/war.h>
+#include <asm/asm-offsets.h>
+
+	.align	5
+NESTED(handle_sys, PT_SIZE, sp)
+	.set	noat
+	SAVE_SOME
+	TRACE_IRQS_ON_RELOAD
+	STI
+	.set	at
+
+	lw	t1, PT_EPC(sp)		# skip syscall on return
+
+	addiu	t1, 4			# skip to next instruction
+	sw	t1, PT_EPC(sp)
+
+	sw	a3, PT_R26(sp)		# save a3 for syscall restarting
+
+	/*
+	 * More than four arguments.  Try to deal with it by copying the
+	 * stack arguments from the user stack to the kernel stack.
+	 * This Sucks (TM).
+	 */
+	lw	t0, PT_R29(sp)		# get old user stack pointer
+
+	/*
+	 * We intentionally keep the kernel stack a little below the top of
+	 * userspace so we don't have to do a slower byte accurate check here.
+	 */
+	lw	t5, TI_ADDR_LIMIT($28)
+	addu	t4, t0, 32
+	and	t5, t4
+	bltz	t5, bad_stack		# -> sp is bad
+
+	/*
+	 * Ok, copy the args from the luser stack to the kernel stack.
+	 */
+
+	.set    push
+	.set    noreorder
+	.set	nomacro
+
+load_a4: user_lw(t5, 16(t0))		# argument #5 from usp
+load_a5: user_lw(t6, 20(t0))		# argument #6 from usp
+load_a6: user_lw(t7, 24(t0))		# argument #7 from usp
+load_a7: user_lw(t8, 28(t0))		# argument #8 from usp
+loads_done:
+
+	sw	t5, 16(sp)		# argument #5 to ksp
+	sw	t6, 20(sp)		# argument #6 to ksp
+	sw	t7, 24(sp)		# argument #7 to ksp
+	sw	t8, 28(sp)		# argument #8 to ksp
+	.set	pop
+
+	.section __ex_table,"a"
+	PTR	load_a4, bad_stack_a4
+	PTR	load_a5, bad_stack_a5
+	PTR	load_a6, bad_stack_a6
+	PTR	load_a7, bad_stack_a7
+	.previous
+
+	lw	t0, TI_FLAGS($28)	# syscall tracing enabled?
+	li	t1, _TIF_WORK_SYSCALL_ENTRY
+	and	t0, t1
+	bnez	t0, syscall_trace_entry # -> yes
+syscall_common:
+	subu	v0, v0, __NR_O32_Linux	# check syscall number
+	sltiu	t0, v0, __NR_O32_Linux_syscalls
+	beqz	t0, illegal_syscall
+
+	sll	t0, v0, 2
+	la	t1, sys_call_table
+	addu	t1, t0
+	lw	t2, (t1)		# syscall routine
+
+	beqz	t2, illegal_syscall
+
+	jalr	t2			# Do The Real Thing (TM)
+
+	li	t0, -EMAXERRNO - 1	# error?
+	sltu	t0, t0, v0
+	sw	t0, PT_R7(sp)		# set error flag
+	beqz	t0, 1f
+
+	lw	t1, PT_R2(sp)		# syscall number
+	negu	v0			# error
+	sw	t1, PT_R0(sp)		# save it for syscall restarting
+1:	sw	v0, PT_R2(sp)		# result
+
+o32_syscall_exit:
+	j	syscall_exit_partial
+
+/* ------------------------------------------------------------------------ */
+
+syscall_trace_entry:
+	SAVE_STATIC
+	move	a0, sp
+
+	/*
+	 * syscall number is in v0 unless we called syscall(__NR_###)
+	 * where the real syscall number is in a0
+	 */
+	move	a1, v0
+	subu	t2, v0,  __NR_O32_Linux
+	bnez	t2, 1f /* __NR_syscall at offset 0 */
+	lw	a1, PT_R4(sp)
+
+1:	jal	syscall_trace_enter
+
+	bltz	v0, 1f			# seccomp failed? Skip syscall
+
+	RESTORE_STATIC
+	lw	v0, PT_R2(sp)		# Restore syscall (maybe modified)
+	lw	a0, PT_R4(sp)		# Restore argument registers
+	lw	a1, PT_R5(sp)
+	lw	a2, PT_R6(sp)
+	lw	a3, PT_R7(sp)
+	j	syscall_common
+
+1:	j	syscall_exit
+
+/* ------------------------------------------------------------------------ */
+
+	/*
+	 * Our open-coded access area sanity test for the stack pointer
+	 * failed. We probably should handle this case a bit more drastic.
+	 */
+bad_stack:
+	li	v0, EFAULT
+	sw	v0, PT_R2(sp)
+	li	t0, 1				# set error flag
+	sw	t0, PT_R7(sp)
+	j	o32_syscall_exit
+
+bad_stack_a4:
+	li	t5, 0
+	b	load_a5
+
+bad_stack_a5:
+	li	t6, 0
+	b	load_a6
+
+bad_stack_a6:
+	li	t7, 0
+	b	load_a7
+
+bad_stack_a7:
+	li	t8, 0
+	b	loads_done
+
+	/*
+	 * The system call does not exist in this kernel
+	 */
+illegal_syscall:
+	li	v0, ENOSYS			# error
+	sw	v0, PT_R2(sp)
+	li	t0, 1				# set error flag
+	sw	t0, PT_R7(sp)
+	j	o32_syscall_exit
+	END(handle_sys)
+
+	LEAF(sys_syscall)
+	subu	t0, a0, __NR_O32_Linux	# check syscall number
+	sltiu	v0, t0, __NR_O32_Linux_syscalls
+	beqz	t0, einval		# do not recurse
+	sll	t1, t0, 2
+	beqz	v0, einval
+	lw	t2, sys_call_table(t1)		# syscall routine
+
+	move	a0, a1				# shift argument registers
+	move	a1, a2
+	move	a2, a3
+	lw	a3, 16(sp)
+	lw	t4, 20(sp)
+	lw	t5, 24(sp)
+	lw	t6, 28(sp)
+	sw	t4, 16(sp)
+	sw	t5, 20(sp)
+	sw	t6, 24(sp)
+	jr	t2
+	/* Unreached */
+
+einval: li	v0, -ENOSYS
+	jr	ra
+	END(sys_syscall)
+
+#ifdef CONFIG_MIPS_MT_FPAFF
+	/*
+	 * For FPU affinity scheduling on MIPS MT processors, we need to
+	 * intercept sys_sched_xxxaffinity() calls until we get a proper hook
+	 * in kernel/sched/core.c.  Considered only temporary we only support
+	 * these hooks for the 32-bit kernel - there is no MIPS64 MT processor
+	 * atm.
+	 */
+#define sys_sched_setaffinity	mipsmt_sys_sched_setaffinity
+#define sys_sched_getaffinity	mipsmt_sys_sched_getaffinity
+#endif /* CONFIG_MIPS_MT_FPAFF */
+
+#define __SYSCALL(nr, entry) 	PTR entry
+	.align	2
+	.type	sys_call_table, @object
+EXPORT(sys_call_table)
+#include <asm/syscall_table_32_o32.h>
+#undef __SYSCALL
diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S
new file mode 100644
index 000000000..35d8c86b1
--- /dev/null
+++ b/arch/mips/kernel/scall64-n32.S
@@ -0,0 +1,108 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1995, 96, 97, 98, 99, 2000, 01 by Ralf Baechle
+ * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ * Copyright (C) 2001 MIPS Technologies, Inc.
+ */
+#include <linux/errno.h>
+#include <asm/asm.h>
+#include <asm/asmmacro.h>
+#include <asm/irqflags.h>
+#include <asm/mipsregs.h>
+#include <asm/regdef.h>
+#include <asm/stackframe.h>
+#include <asm/thread_info.h>
+#include <asm/unistd.h>
+
+#ifndef CONFIG_MIPS32_O32
+/* No O32, so define handle_sys here */
+#define handle_sysn32 handle_sys
+#endif
+
+	.align	5
+NESTED(handle_sysn32, PT_SIZE, sp)
+#ifndef CONFIG_MIPS32_O32
+	.set	noat
+	SAVE_SOME
+	TRACE_IRQS_ON_RELOAD
+	STI
+	.set	at
+#endif
+
+	dsubu	t0, v0, __NR_N32_Linux	# check syscall number
+	sltiu	t0, t0, __NR_N32_Linux_syscalls
+
+#ifndef CONFIG_MIPS32_O32
+	ld	t1, PT_EPC(sp)		# skip syscall on return
+	daddiu	t1, 4			# skip to next instruction
+	sd	t1, PT_EPC(sp)
+#endif
+	beqz	t0, not_n32_scall
+
+	sd	a3, PT_R26(sp)		# save a3 for syscall restarting
+
+	li	t1, _TIF_WORK_SYSCALL_ENTRY
+	LONG_L	t0, TI_FLAGS($28)	# syscall tracing enabled?
+	and	t0, t1, t0
+	bnez	t0, n32_syscall_trace_entry
+
+syscall_common:
+	dsll	t0, v0, 3		# offset into table
+	ld	t2, (sysn32_call_table - (__NR_N32_Linux * 8))(t0)
+
+	jalr	t2			# Do The Real Thing (TM)
+
+	li	t0, -EMAXERRNO - 1	# error?
+	sltu	t0, t0, v0
+	sd	t0, PT_R7(sp)		# set error flag
+	beqz	t0, 1f
+
+	ld	t1, PT_R2(sp)		# syscall number
+	dnegu	v0			# error
+	sd	t1, PT_R0(sp)		# save it for syscall restarting
+1:	sd	v0, PT_R2(sp)		# result
+
+	j	syscall_exit_partial
+
+/* ------------------------------------------------------------------------ */
+
+n32_syscall_trace_entry:
+	SAVE_STATIC
+	move	a0, sp
+	move	a1, v0
+	jal	syscall_trace_enter
+
+	bltz	v0, 1f			# seccomp failed? Skip syscall
+
+	RESTORE_STATIC
+	ld	v0, PT_R2(sp)		# Restore syscall (maybe modified)
+	ld	a0, PT_R4(sp)		# Restore argument registers
+	ld	a1, PT_R5(sp)
+	ld	a2, PT_R6(sp)
+	ld	a3, PT_R7(sp)
+	ld	a4, PT_R8(sp)
+	ld	a5, PT_R9(sp)
+
+	dsubu	t2, v0, __NR_N32_Linux	# check (new) syscall number
+	sltiu   t0, t2, __NR_N32_Linux_syscalls
+	beqz	t0, not_n32_scall
+
+	j	syscall_common
+
+1:	j	syscall_exit
+
+not_n32_scall:
+	/* This is not an n32 compatibility syscall, pass it on to
+	   the n64 syscall handlers.  */
+	j	handle_sys64
+
+	END(handle_sysn32)
+
+#define __SYSCALL(nr, entry)	PTR entry
+	.type	sysn32_call_table, @object
+EXPORT(sysn32_call_table)
+#include <asm/syscall_table_64_n32.h>
+#undef __SYSCALL
diff --git a/arch/mips/kernel/scall64-n64.S b/arch/mips/kernel/scall64-n64.S
new file mode 100644
index 000000000..23b2e2b16
--- /dev/null
+++ b/arch/mips/kernel/scall64-n64.S
@@ -0,0 +1,117 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1995, 96, 97, 98, 99, 2000, 01, 02 by Ralf Baechle
+ * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ * Copyright (C) 2001 MIPS Technologies, Inc.
+ */
+#include <linux/errno.h>
+#include <asm/asm.h>
+#include <asm/asmmacro.h>
+#include <asm/irqflags.h>
+#include <asm/mipsregs.h>
+#include <asm/regdef.h>
+#include <asm/stackframe.h>
+#include <asm/asm-offsets.h>
+#include <asm/sysmips.h>
+#include <asm/thread_info.h>
+#include <asm/unistd.h>
+#include <asm/war.h>
+
+#ifndef CONFIG_BINFMT_ELF32
+/* Neither O32 nor N32, so define handle_sys here */
+#define handle_sys64 handle_sys
+#endif
+
+	.align	5
+NESTED(handle_sys64, PT_SIZE, sp)
+#if !defined(CONFIG_MIPS32_O32) && !defined(CONFIG_MIPS32_N32)
+	/*
+	 * When 32-bit compatibility is configured scall_o32.S
+	 * already did this.
+	 */
+	.set	noat
+	SAVE_SOME
+	TRACE_IRQS_ON_RELOAD
+	STI
+	.set	at
+#endif
+
+#if !defined(CONFIG_MIPS32_O32) && !defined(CONFIG_MIPS32_N32)
+	ld	t1, PT_EPC(sp)		# skip syscall on return
+	daddiu	t1, 4			# skip to next instruction
+	sd	t1, PT_EPC(sp)
+#endif
+
+	sd	a3, PT_R26(sp)		# save a3 for syscall restarting
+
+	li	t1, _TIF_WORK_SYSCALL_ENTRY
+	LONG_L	t0, TI_FLAGS($28)	# syscall tracing enabled?
+	and	t0, t1, t0
+	bnez	t0, syscall_trace_entry
+
+syscall_common:
+	dsubu	t2, v0, __NR_64_Linux
+	sltiu   t0, t2, __NR_64_Linux_syscalls
+	beqz	t0, illegal_syscall
+
+	dsll	t0, t2, 3		# offset into table
+	dla	t2, sys_call_table
+	daddu	t0, t2, t0
+	ld	t2, (t0)		# syscall routine
+	beqz	t2, illegal_syscall
+
+	jalr	t2			# Do The Real Thing (TM)
+
+	li	t0, -EMAXERRNO - 1	# error?
+	sltu	t0, t0, v0
+	sd	t0, PT_R7(sp)		# set error flag
+	beqz	t0, 1f
+
+	ld	t1, PT_R2(sp)		# syscall number
+	dnegu	v0			# error
+	sd	t1, PT_R0(sp)		# save it for syscall restarting
+1:	sd	v0, PT_R2(sp)		# result
+
+n64_syscall_exit:
+	j	syscall_exit_partial
+
+/* ------------------------------------------------------------------------ */
+
+syscall_trace_entry:
+	SAVE_STATIC
+	move	a0, sp
+	move	a1, v0
+	jal	syscall_trace_enter
+
+	bltz	v0, 1f			# seccomp failed? Skip syscall
+
+	RESTORE_STATIC
+	ld	v0, PT_R2(sp)		# Restore syscall (maybe modified)
+	ld	a0, PT_R4(sp)		# Restore argument registers
+	ld	a1, PT_R5(sp)
+	ld	a2, PT_R6(sp)
+	ld	a3, PT_R7(sp)
+	ld	a4, PT_R8(sp)
+	ld	a5, PT_R9(sp)
+	j	syscall_common
+
+1:	j	syscall_exit
+
+illegal_syscall:
+	/* This also isn't a 64-bit syscall, throw an error.  */
+	li	v0, ENOSYS			# error
+	sd	v0, PT_R2(sp)
+	li	t0, 1				# set error flag
+	sd	t0, PT_R7(sp)
+	j	n64_syscall_exit
+	END(handle_sys64)
+
+#define __SYSCALL(nr, entry)	PTR entry
+	.align	3
+	.type	sys_call_table, @object
+EXPORT(sys_call_table)
+#include <asm/syscall_table_64_n64.h>
+#undef __SYSCALL
diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S
new file mode 100644
index 000000000..50c9a57e0
--- /dev/null
+++ b/arch/mips/kernel/scall64-o32.S
@@ -0,0 +1,221 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1995 - 2000, 2001 by Ralf Baechle
+ * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ * Copyright (C) 2001 MIPS Technologies, Inc.
+ * Copyright (C) 2004 Thiemo Seufer
+ *
+ * Hairy, the userspace application uses a different argument passing
+ * convention than the kernel, so we have to translate things from o32
+ * to ABI64 calling convention.	 64-bit syscalls are also processed
+ * here for now.
+ */
+#include <linux/errno.h>
+#include <asm/asm.h>
+#include <asm/asmmacro.h>
+#include <asm/irqflags.h>
+#include <asm/mipsregs.h>
+#include <asm/regdef.h>
+#include <asm/stackframe.h>
+#include <asm/thread_info.h>
+#include <asm/unistd.h>
+#include <asm/sysmips.h>
+
+	.align	5
+NESTED(handle_sys, PT_SIZE, sp)
+	.set	noat
+	SAVE_SOME
+	TRACE_IRQS_ON_RELOAD
+	STI
+	.set	at
+	ld	t1, PT_EPC(sp)		# skip syscall on return
+
+	dsubu	t0, v0, __NR_O32_Linux	# check syscall number
+	sltiu	t0, t0, __NR_O32_Linux_syscalls
+	daddiu	t1, 4			# skip to next instruction
+	sd	t1, PT_EPC(sp)
+	beqz	t0, not_o32_scall
+#if 0
+ SAVE_ALL
+ move a1, v0
+ ASM_PRINT("Scall %ld\n")
+ RESTORE_ALL
+#endif
+
+	/* We don't want to stumble over broken sign extensions from
+	   userland. O32 does never use the upper half. */
+	sll	a0, a0, 0
+	sll	a1, a1, 0
+	sll	a2, a2, 0
+	sll	a3, a3, 0
+
+	sd	a3, PT_R26(sp)		# save a3 for syscall restarting
+
+	/*
+	 * More than four arguments.  Try to deal with it by copying the
+	 * stack arguments from the user stack to the kernel stack.
+	 * This Sucks (TM).
+	 *
+	 * We intentionally keep the kernel stack a little below the top of
+	 * userspace so we don't have to do a slower byte accurate check here.
+	 */
+	ld	t0, PT_R29(sp)		# get old user stack pointer
+	daddu	t1, t0, 32
+	bltz	t1, bad_stack
+
+load_a4: lw	a4, 16(t0)		# argument #5 from usp
+load_a5: lw	a5, 20(t0)		# argument #6 from usp
+load_a6: lw	a6, 24(t0)		# argument #7 from usp
+load_a7: lw	a7, 28(t0)		# argument #8 from usp
+loads_done:
+
+	.section __ex_table,"a"
+	PTR	load_a4, bad_stack_a4
+	PTR	load_a5, bad_stack_a5
+	PTR	load_a6, bad_stack_a6
+	PTR	load_a7, bad_stack_a7
+	.previous
+
+	li	t1, _TIF_WORK_SYSCALL_ENTRY
+	LONG_L	t0, TI_FLAGS($28)	# syscall tracing enabled?
+	and	t0, t1, t0
+	bnez	t0, trace_a_syscall
+
+syscall_common:
+	dsll	t0, v0, 3		# offset into table
+	ld	t2, (sys32_call_table - (__NR_O32_Linux * 8))(t0)
+
+	jalr	t2			# Do The Real Thing (TM)
+
+	li	t0, -EMAXERRNO - 1	# error?
+	sltu	t0, t0, v0
+	sd	t0, PT_R7(sp)		# set error flag
+	beqz	t0, 1f
+
+	ld	t1, PT_R2(sp)		# syscall number
+	dnegu	v0			# error
+	sd	t1, PT_R0(sp)		# save it for syscall restarting
+1:	sd	v0, PT_R2(sp)		# result
+
+o32_syscall_exit:
+	j	syscall_exit_partial
+
+/* ------------------------------------------------------------------------ */
+
+trace_a_syscall:
+	SAVE_STATIC
+	sd	a4, PT_R8(sp)		# Save argument registers
+	sd	a5, PT_R9(sp)
+	sd	a6, PT_R10(sp)
+	sd	a7, PT_R11(sp)		# For indirect syscalls
+
+	move	a0, sp
+	/*
+	 * absolute syscall number is in v0 unless we called syscall(__NR_###)
+	 * where the real syscall number is in a0
+	 * note: NR_syscall is the first O32 syscall but the macro is
+	 * only defined when compiling with -mabi=32 (CONFIG_32BIT)
+	 * therefore __NR_O32_Linux is used (4000)
+	 */
+	.set	push
+	.set	reorder
+	subu	t1, v0,  __NR_O32_Linux
+	move	a1, v0
+	bnez	t1, 1f /* __NR_syscall at offset 0 */
+	ld	a1, PT_R4(sp) /* Arg1 for __NR_syscall case */
+	.set	pop
+
+1:	jal	syscall_trace_enter
+
+	bltz	v0, 1f			# seccomp failed? Skip syscall
+
+	RESTORE_STATIC
+	ld	v0, PT_R2(sp)		# Restore syscall (maybe modified)
+	ld	a0, PT_R4(sp)		# Restore argument registers
+	ld	a1, PT_R5(sp)
+	ld	a2, PT_R6(sp)
+	ld	a3, PT_R7(sp)
+	ld	a4, PT_R8(sp)
+	ld	a5, PT_R9(sp)
+	ld	a6, PT_R10(sp)
+	ld	a7, PT_R11(sp)		# For indirect syscalls
+
+	dsubu	t0, v0, __NR_O32_Linux	# check (new) syscall number
+	sltiu	t0, t0, __NR_O32_Linux_syscalls
+	beqz	t0, not_o32_scall
+
+	j	syscall_common
+
+1:	j	syscall_exit
+
+/* ------------------------------------------------------------------------ */
+
+	/*
+	 * The stackpointer for a call with more than 4 arguments is bad.
+	 */
+bad_stack:
+	li	v0, EFAULT
+	sd	v0, PT_R2(sp)
+	li	t0, 1			# set error flag
+	sd	t0, PT_R7(sp)
+	j	o32_syscall_exit
+
+bad_stack_a4:
+	li	a4, 0
+	b	load_a5
+
+bad_stack_a5:
+	li	a5, 0
+	b	load_a6
+
+bad_stack_a6:
+	li	a6, 0
+	b	load_a7
+
+bad_stack_a7:
+	li	a7, 0
+	b	loads_done
+
+not_o32_scall:
+	/*
+	 * This is not an o32 compatibility syscall, pass it on
+	 * to the 64-bit syscall handlers.
+	 */
+#ifdef CONFIG_MIPS32_N32
+	j	handle_sysn32
+#else
+	j	handle_sys64
+#endif
+	END(handle_sys)
+
+LEAF(sys32_syscall)
+	subu	t0, a0, __NR_O32_Linux	# check syscall number
+	sltiu	v0, t0, __NR_O32_Linux_syscalls
+	beqz	t0, einval		# do not recurse
+	dsll	t1, t0, 3
+	beqz	v0, einval
+	ld	t2, sys32_call_table(t1)		# syscall routine
+
+	move	a0, a1			# shift argument registers
+	move	a1, a2
+	move	a2, a3
+	move	a3, a4
+	move	a4, a5
+	move	a5, a6
+	move	a6, a7
+	jr	t2
+	/* Unreached */
+
+einval: li	v0, -ENOSYS
+	jr	ra
+	END(sys32_syscall)
+
+#define __SYSCALL(nr, entry)	PTR entry
+	.align	3
+	.type	sys32_call_table,@object
+EXPORT(sys32_call_table)
+#include <asm/syscall_table_64_o32.h>
+#undef __SYSCALL
diff --git a/arch/mips/kernel/segment.c b/arch/mips/kernel/segment.c
new file mode 100644
index 000000000..0a9bd7b09
--- /dev/null
+++ b/arch/mips/kernel/segment.c
@@ -0,0 +1,104 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2013 Imagination Technologies Ltd.
+ */
+
+#include <linux/kernel.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+#include <asm/cpu.h>
+#include <asm/debug.h>
+#include <asm/mipsregs.h>
+
+static void build_segment_config(char *str, unsigned int cfg)
+{
+	unsigned int am;
+	static const char * const am_str[] = {
+		"UK", "MK", "MSK", "MUSK", "MUSUK", "USK",
+		"RSRVD", "UUSK"};
+
+	/* Segment access mode. */
+	am = (cfg & MIPS_SEGCFG_AM) >> MIPS_SEGCFG_AM_SHIFT;
+	str += sprintf(str, "%-5s", am_str[am]);
+
+	/*
+	 * Access modes MK, MSK and MUSK are mapped segments. Therefore
+	 * there is no direct physical address mapping unless it becomes
+	 * unmapped uncached at error level due to EU.
+	 */
+	if ((am == 0) || (am > 3) || (cfg & MIPS_SEGCFG_EU))
+		str += sprintf(str, "         %03lx",
+			((cfg & MIPS_SEGCFG_PA) >> MIPS_SEGCFG_PA_SHIFT));
+	else
+		str += sprintf(str, "         UND");
+
+	if ((am == 0) || (am > 3))
+		str += sprintf(str, "         %01ld",
+			((cfg & MIPS_SEGCFG_C) >> MIPS_SEGCFG_C_SHIFT));
+	else
+		str += sprintf(str, "         U");
+
+	/* Exception configuration. */
+	str += sprintf(str, "       %01ld\n",
+		((cfg & MIPS_SEGCFG_EU) >> MIPS_SEGCFG_EU_SHIFT));
+}
+
+static int show_segments(struct seq_file *m, void *v)
+{
+	unsigned int segcfg;
+	char str[42];
+
+	seq_puts(m, "Segment   Virtual    Size   Access Mode   Physical   Caching   EU\n");
+	seq_puts(m, "-------   -------    ----   -----------   --------   -------   --\n");
+
+	segcfg = read_c0_segctl0();
+	build_segment_config(str, segcfg);
+	seq_printf(m, "   0      e0000000   512M      %s", str);
+
+	segcfg >>= 16;
+	build_segment_config(str, segcfg);
+	seq_printf(m, "   1      c0000000   512M      %s", str);
+
+	segcfg = read_c0_segctl1();
+	build_segment_config(str, segcfg);
+	seq_printf(m, "   2      a0000000   512M      %s", str);
+
+	segcfg >>= 16;
+	build_segment_config(str, segcfg);
+	seq_printf(m, "   3      80000000   512M      %s", str);
+
+	segcfg = read_c0_segctl2();
+	build_segment_config(str, segcfg);
+	seq_printf(m, "   4      40000000    1G       %s", str);
+
+	segcfg >>= 16;
+	build_segment_config(str, segcfg);
+	seq_printf(m, "   5      00000000    1G       %s\n", str);
+
+	return 0;
+}
+
+static int segments_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, show_segments, NULL);
+}
+
+static const struct file_operations segments_fops = {
+	.open		= segments_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+static int __init segments_info(void)
+{
+	if (cpu_has_segments)
+		debugfs_create_file("segments", S_IRUGO, mips_debugfs_dir, NULL,
+				    &segments_fops);
+	return 0;
+}
+
+device_initcall(segments_info);
diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c
new file mode 100644
index 000000000..66643cb65
--- /dev/null
+++ b/arch/mips/kernel/setup.c
@@ -0,0 +1,844 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1995 Linus Torvalds
+ * Copyright (C) 1995 Waldorf Electronics
+ * Copyright (C) 1994, 95, 96, 97, 98, 99, 2000, 01, 02, 03  Ralf Baechle
+ * Copyright (C) 1996 Stoned Elipot
+ * Copyright (C) 1999 Silicon Graphics, Inc.
+ * Copyright (C) 2000, 2001, 2002, 2007	 Maciej W. Rozycki
+ */
+#include <linux/init.h>
+#include <linux/cpu.h>
+#include <linux/delay.h>
+#include <linux/ioport.h>
+#include <linux/export.h>
+#include <linux/screen_info.h>
+#include <linux/memblock.h>
+#include <linux/initrd.h>
+#include <linux/root_dev.h>
+#include <linux/highmem.h>
+#include <linux/console.h>
+#include <linux/pfn.h>
+#include <linux/debugfs.h>
+#include <linux/kexec.h>
+#include <linux/sizes.h>
+#include <linux/device.h>
+#include <linux/dma-map-ops.h>
+#include <linux/decompress/generic.h>
+#include <linux/of_fdt.h>
+#include <linux/of_reserved_mem.h>
+#include <linux/dmi.h>
+
+#include <asm/addrspace.h>
+#include <asm/bootinfo.h>
+#include <asm/bugs.h>
+#include <asm/cache.h>
+#include <asm/cdmm.h>
+#include <asm/cpu.h>
+#include <asm/debug.h>
+#include <asm/dma-coherence.h>
+#include <asm/sections.h>
+#include <asm/setup.h>
+#include <asm/smp-ops.h>
+#include <asm/prom.h>
+
+#ifdef CONFIG_MIPS_ELF_APPENDED_DTB
+const char __section(".appended_dtb") __appended_dtb[0x100000];
+#endif /* CONFIG_MIPS_ELF_APPENDED_DTB */
+
+struct cpuinfo_mips cpu_data[NR_CPUS] __read_mostly;
+
+EXPORT_SYMBOL(cpu_data);
+
+#ifdef CONFIG_VT
+struct screen_info screen_info;
+#endif
+
+/*
+ * Setup information
+ *
+ * These are initialized so they are in the .data section
+ */
+unsigned long mips_machtype __read_mostly = MACH_UNKNOWN;
+
+EXPORT_SYMBOL(mips_machtype);
+
+static char __initdata command_line[COMMAND_LINE_SIZE];
+char __initdata arcs_cmdline[COMMAND_LINE_SIZE];
+
+#ifdef CONFIG_CMDLINE_BOOL
+static const char builtin_cmdline[] __initconst = CONFIG_CMDLINE;
+#else
+static const char builtin_cmdline[] __initconst = "";
+#endif
+
+/*
+ * mips_io_port_base is the begin of the address space to which x86 style
+ * I/O ports are mapped.
+ */
+unsigned long mips_io_port_base = -1;
+EXPORT_SYMBOL(mips_io_port_base);
+
+static struct resource code_resource = { .name = "Kernel code", };
+static struct resource data_resource = { .name = "Kernel data", };
+static struct resource bss_resource = { .name = "Kernel bss", };
+
+static void *detect_magic __initdata = detect_memory_region;
+
+#ifdef CONFIG_MIPS_AUTO_PFN_OFFSET
+unsigned long ARCH_PFN_OFFSET;
+EXPORT_SYMBOL(ARCH_PFN_OFFSET);
+#endif
+
+void __init detect_memory_region(phys_addr_t start, phys_addr_t sz_min, phys_addr_t sz_max)
+{
+	void *dm = &detect_magic;
+	phys_addr_t size;
+
+	for (size = sz_min; size < sz_max; size <<= 1) {
+		if (!memcmp(dm, dm + size, sizeof(detect_magic)))
+			break;
+	}
+
+	pr_debug("Memory: %lluMB of RAM detected at 0x%llx (min: %lluMB, max: %lluMB)\n",
+		((unsigned long long) size) / SZ_1M,
+		(unsigned long long) start,
+		((unsigned long long) sz_min) / SZ_1M,
+		((unsigned long long) sz_max) / SZ_1M);
+
+	memblock_add(start, size);
+}
+
+/*
+ * Manage initrd
+ */
+#ifdef CONFIG_BLK_DEV_INITRD
+
+static int __init rd_start_early(char *p)
+{
+	unsigned long start = memparse(p, &p);
+
+#ifdef CONFIG_64BIT
+	/* Guess if the sign extension was forgotten by bootloader */
+	if (start < XKPHYS)
+		start = (int)start;
+#endif
+	initrd_start = start;
+	initrd_end += start;
+	return 0;
+}
+early_param("rd_start", rd_start_early);
+
+static int __init rd_size_early(char *p)
+{
+	initrd_end += memparse(p, &p);
+	return 0;
+}
+early_param("rd_size", rd_size_early);
+
+/* it returns the next free pfn after initrd */
+static unsigned long __init init_initrd(void)
+{
+	unsigned long end;
+
+	/*
+	 * Board specific code or command line parser should have
+	 * already set up initrd_start and initrd_end. In these cases
+	 * perfom sanity checks and use them if all looks good.
+	 */
+	if (!initrd_start || initrd_end <= initrd_start)
+		goto disable;
+
+	if (initrd_start & ~PAGE_MASK) {
+		pr_err("initrd start must be page aligned\n");
+		goto disable;
+	}
+
+	/*
+	 * Sanitize initrd addresses. For example firmware
+	 * can't guess if they need to pass them through
+	 * 64-bits values if the kernel has been built in pure
+	 * 32-bit. We need also to switch from KSEG0 to XKPHYS
+	 * addresses now, so the code can now safely use __pa().
+	 */
+	end = __pa(initrd_end);
+	initrd_end = (unsigned long)__va(end);
+	initrd_start = (unsigned long)__va(__pa(initrd_start));
+
+	if (initrd_start < PAGE_OFFSET) {
+		pr_err("initrd start < PAGE_OFFSET\n");
+		goto disable;
+	}
+
+	ROOT_DEV = Root_RAM0;
+	return PFN_UP(end);
+disable:
+	initrd_start = 0;
+	initrd_end = 0;
+	return 0;
+}
+
+/* In some conditions (e.g. big endian bootloader with a little endian
+   kernel), the initrd might appear byte swapped.  Try to detect this and
+   byte swap it if needed.  */
+static void __init maybe_bswap_initrd(void)
+{
+#if defined(CONFIG_CPU_CAVIUM_OCTEON)
+	u64 buf;
+
+	/* Check for CPIO signature */
+	if (!memcmp((void *)initrd_start, "070701", 6))
+		return;
+
+	/* Check for compressed initrd */
+	if (decompress_method((unsigned char *)initrd_start, 8, NULL))
+		return;
+
+	/* Try again with a byte swapped header */
+	buf = swab64p((u64 *)initrd_start);
+	if (!memcmp(&buf, "070701", 6) ||
+	    decompress_method((unsigned char *)(&buf), 8, NULL)) {
+		unsigned long i;
+
+		pr_info("Byteswapped initrd detected\n");
+		for (i = initrd_start; i < ALIGN(initrd_end, 8); i += 8)
+			swab64s((u64 *)i);
+	}
+#endif
+}
+
+static void __init finalize_initrd(void)
+{
+	unsigned long size = initrd_end - initrd_start;
+
+	if (size == 0) {
+		printk(KERN_INFO "Initrd not found or empty");
+		goto disable;
+	}
+	if (__pa(initrd_end) > PFN_PHYS(max_low_pfn)) {
+		printk(KERN_ERR "Initrd extends beyond end of memory");
+		goto disable;
+	}
+
+	maybe_bswap_initrd();
+
+	memblock_reserve(__pa(initrd_start), size);
+	initrd_below_start_ok = 1;
+
+	pr_info("Initial ramdisk at: 0x%lx (%lu bytes)\n",
+		initrd_start, size);
+	return;
+disable:
+	printk(KERN_CONT " - disabling initrd\n");
+	initrd_start = 0;
+	initrd_end = 0;
+}
+
+#else  /* !CONFIG_BLK_DEV_INITRD */
+
+static unsigned long __init init_initrd(void)
+{
+	return 0;
+}
+
+#define finalize_initrd()	do {} while (0)
+
+#endif
+
+/*
+ * Initialize the bootmem allocator. It also setup initrd related data
+ * if needed.
+ */
+#if defined(CONFIG_SGI_IP27) || (defined(CONFIG_CPU_LOONGSON64) && defined(CONFIG_NUMA))
+
+static void __init bootmem_init(void)
+{
+	init_initrd();
+	finalize_initrd();
+}
+
+#else  /* !CONFIG_SGI_IP27 */
+
+static void __init bootmem_init(void)
+{
+	phys_addr_t ramstart, ramend;
+	unsigned long start, end;
+	int i;
+
+	ramstart = memblock_start_of_DRAM();
+	ramend = memblock_end_of_DRAM();
+
+	/*
+	 * Sanity check any INITRD first. We don't take it into account
+	 * for bootmem setup initially, rely on the end-of-kernel-code
+	 * as our memory range starting point. Once bootmem is inited we
+	 * will reserve the area used for the initrd.
+	 */
+	init_initrd();
+
+	/* Reserve memory occupied by kernel. */
+	memblock_reserve(__pa_symbol(&_text),
+			__pa_symbol(&_end) - __pa_symbol(&_text));
+
+	/* max_low_pfn is not a number of pages but the end pfn of low mem */
+
+#ifdef CONFIG_MIPS_AUTO_PFN_OFFSET
+	ARCH_PFN_OFFSET = PFN_UP(ramstart);
+#else
+	/*
+	 * Reserve any memory between the start of RAM and PHYS_OFFSET
+	 */
+	if (ramstart > PHYS_OFFSET)
+		memblock_reserve(PHYS_OFFSET, ramstart - PHYS_OFFSET);
+
+	if (PFN_UP(ramstart) > ARCH_PFN_OFFSET) {
+		pr_info("Wasting %lu bytes for tracking %lu unused pages\n",
+			(unsigned long)((PFN_UP(ramstart) - ARCH_PFN_OFFSET) * sizeof(struct page)),
+			(unsigned long)(PFN_UP(ramstart) - ARCH_PFN_OFFSET));
+	}
+#endif
+
+	min_low_pfn = ARCH_PFN_OFFSET;
+	max_pfn = PFN_DOWN(ramend);
+	for_each_mem_pfn_range(i, MAX_NUMNODES, &start, &end, NULL) {
+		/*
+		 * Skip highmem here so we get an accurate max_low_pfn if low
+		 * memory stops short of high memory.
+		 * If the region overlaps HIGHMEM_START, end is clipped so
+		 * max_pfn excludes the highmem portion.
+		 */
+		if (start >= PFN_DOWN(HIGHMEM_START))
+			continue;
+		if (end > PFN_DOWN(HIGHMEM_START))
+			end = PFN_DOWN(HIGHMEM_START);
+		if (end > max_low_pfn)
+			max_low_pfn = end;
+	}
+
+	if (min_low_pfn >= max_low_pfn)
+		panic("Incorrect memory mapping !!!");
+
+	if (max_pfn > PFN_DOWN(HIGHMEM_START)) {
+		max_low_pfn = PFN_DOWN(HIGHMEM_START);
+#ifdef CONFIG_HIGHMEM
+		highstart_pfn = max_low_pfn;
+		highend_pfn = max_pfn;
+#else
+		max_pfn = max_low_pfn;
+#endif
+	}
+
+	/*
+	 * Reserve initrd memory if needed.
+	 */
+	finalize_initrd();
+}
+
+#endif	/* CONFIG_SGI_IP27 */
+
+static int usermem __initdata;
+
+static int __init early_parse_mem(char *p)
+{
+	phys_addr_t start, size;
+
+	/*
+	 * If a user specifies memory size, we
+	 * blow away any automatically generated
+	 * size.
+	 */
+	if (usermem == 0) {
+		usermem = 1;
+		memblock_remove(memblock_start_of_DRAM(),
+			memblock_end_of_DRAM() - memblock_start_of_DRAM());
+	}
+	start = 0;
+	size = memparse(p, &p);
+	if (*p == '@')
+		start = memparse(p + 1, &p);
+
+	memblock_add(start, size);
+
+	return 0;
+}
+early_param("mem", early_parse_mem);
+
+static int __init early_parse_memmap(char *p)
+{
+	char *oldp;
+	u64 start_at, mem_size;
+
+	if (!p)
+		return -EINVAL;
+
+	if (!strncmp(p, "exactmap", 8)) {
+		pr_err("\"memmap=exactmap\" invalid on MIPS\n");
+		return 0;
+	}
+
+	oldp = p;
+	mem_size = memparse(p, &p);
+	if (p == oldp)
+		return -EINVAL;
+
+	if (*p == '@') {
+		start_at = memparse(p+1, &p);
+		memblock_add(start_at, mem_size);
+	} else if (*p == '#') {
+		pr_err("\"memmap=nn#ss\" (force ACPI data) invalid on MIPS\n");
+		return -EINVAL;
+	} else if (*p == '$') {
+		start_at = memparse(p+1, &p);
+		memblock_add(start_at, mem_size);
+		memblock_reserve(start_at, mem_size);
+	} else {
+		pr_err("\"memmap\" invalid format!\n");
+		return -EINVAL;
+	}
+
+	if (*p == '\0') {
+		usermem = 1;
+		return 0;
+	} else
+		return -EINVAL;
+}
+early_param("memmap", early_parse_memmap);
+
+#ifdef CONFIG_PROC_VMCORE
+static unsigned long setup_elfcorehdr, setup_elfcorehdr_size;
+static int __init early_parse_elfcorehdr(char *p)
+{
+	phys_addr_t start, end;
+	u64 i;
+
+	setup_elfcorehdr = memparse(p, &p);
+
+	for_each_mem_range(i, &start, &end) {
+		if (setup_elfcorehdr >= start && setup_elfcorehdr < end) {
+			/*
+			 * Reserve from the elf core header to the end of
+			 * the memory segment, that should all be kdump
+			 * reserved memory.
+			 */
+			setup_elfcorehdr_size = end - setup_elfcorehdr;
+			break;
+		}
+	}
+	/*
+	 * If we don't find it in the memory map, then we shouldn't
+	 * have to worry about it, as the new kernel won't use it.
+	 */
+	return 0;
+}
+early_param("elfcorehdr", early_parse_elfcorehdr);
+#endif
+
+#ifdef CONFIG_KEXEC
+
+/* 64M alignment for crash kernel regions */
+#define CRASH_ALIGN	SZ_64M
+#define CRASH_ADDR_MAX	SZ_512M
+
+static void __init mips_parse_crashkernel(void)
+{
+	unsigned long long total_mem;
+	unsigned long long crash_size, crash_base;
+	int ret;
+
+	total_mem = memblock_phys_mem_size();
+	ret = parse_crashkernel(boot_command_line, total_mem,
+				&crash_size, &crash_base);
+	if (ret != 0 || crash_size <= 0)
+		return;
+
+	if (crash_base <= 0) {
+		crash_base = memblock_find_in_range(CRASH_ALIGN, CRASH_ADDR_MAX,
+							crash_size, CRASH_ALIGN);
+		if (!crash_base) {
+			pr_warn("crashkernel reservation failed - No suitable area found.\n");
+			return;
+		}
+	} else {
+		unsigned long long start;
+
+		start = memblock_find_in_range(crash_base, crash_base + crash_size,
+						crash_size, 1);
+		if (start != crash_base) {
+			pr_warn("Invalid memory region reserved for crash kernel\n");
+			return;
+		}
+	}
+
+	crashk_res.start = crash_base;
+	crashk_res.end	 = crash_base + crash_size - 1;
+}
+
+static void __init request_crashkernel(struct resource *res)
+{
+	int ret;
+
+	if (crashk_res.start == crashk_res.end)
+		return;
+
+	ret = request_resource(res, &crashk_res);
+	if (!ret)
+		pr_info("Reserving %ldMB of memory at %ldMB for crashkernel\n",
+			(unsigned long)(resource_size(&crashk_res) >> 20),
+			(unsigned long)(crashk_res.start  >> 20));
+}
+#else /* !defined(CONFIG_KEXEC)		*/
+static void __init mips_parse_crashkernel(void)
+{
+}
+
+static void __init request_crashkernel(struct resource *res)
+{
+}
+#endif /* !defined(CONFIG_KEXEC)  */
+
+static void __init check_kernel_sections_mem(void)
+{
+	phys_addr_t start = __pa_symbol(&_text);
+	phys_addr_t size = __pa_symbol(&_end) - start;
+
+	if (!memblock_is_region_memory(start, size)) {
+		pr_info("Kernel sections are not in the memory maps\n");
+		memblock_add(start, size);
+	}
+}
+
+static void __init bootcmdline_append(const char *s, size_t max)
+{
+	if (!s[0] || !max)
+		return;
+
+	if (boot_command_line[0])
+		strlcat(boot_command_line, " ", COMMAND_LINE_SIZE);
+
+	strlcat(boot_command_line, s, max);
+}
+
+#ifdef CONFIG_OF_EARLY_FLATTREE
+
+static int __init bootcmdline_scan_chosen(unsigned long node, const char *uname,
+					  int depth, void *data)
+{
+	bool *dt_bootargs = data;
+	const char *p;
+	int l;
+
+	if (depth != 1 || !data ||
+	    (strcmp(uname, "chosen") != 0 && strcmp(uname, "chosen@0") != 0))
+		return 0;
+
+	p = of_get_flat_dt_prop(node, "bootargs", &l);
+	if (p != NULL && l > 0) {
+		bootcmdline_append(p, min(l, COMMAND_LINE_SIZE));
+		*dt_bootargs = true;
+	}
+
+	return 1;
+}
+
+#endif /* CONFIG_OF_EARLY_FLATTREE */
+
+static void __init bootcmdline_init(void)
+{
+	bool dt_bootargs = false;
+
+	/*
+	 * If CMDLINE_OVERRIDE is enabled then initializing the command line is
+	 * trivial - we simply use the built-in command line unconditionally &
+	 * unmodified.
+	 */
+	if (IS_ENABLED(CONFIG_CMDLINE_OVERRIDE)) {
+		strlcpy(boot_command_line, builtin_cmdline, COMMAND_LINE_SIZE);
+		return;
+	}
+
+	/*
+	 * If the user specified a built-in command line &
+	 * MIPS_CMDLINE_BUILTIN_EXTEND, then the built-in command line is
+	 * prepended to arguments from the bootloader or DT so we'll copy them
+	 * to the start of boot_command_line here. Otherwise, empty
+	 * boot_command_line to undo anything early_init_dt_scan_chosen() did.
+	 */
+	if (IS_ENABLED(CONFIG_MIPS_CMDLINE_BUILTIN_EXTEND))
+		strlcpy(boot_command_line, builtin_cmdline, COMMAND_LINE_SIZE);
+	else
+		boot_command_line[0] = 0;
+
+#ifdef CONFIG_OF_EARLY_FLATTREE
+	/*
+	 * If we're configured to take boot arguments from DT, look for those
+	 * now.
+	 */
+	if (IS_ENABLED(CONFIG_MIPS_CMDLINE_FROM_DTB) ||
+	    IS_ENABLED(CONFIG_MIPS_CMDLINE_DTB_EXTEND))
+		of_scan_flat_dt(bootcmdline_scan_chosen, &dt_bootargs);
+#endif
+
+	/*
+	 * If we didn't get any arguments from DT (regardless of whether that's
+	 * because we weren't configured to look for them, or because we looked
+	 * & found none) then we'll take arguments from the bootloader.
+	 * plat_mem_setup() should have filled arcs_cmdline with arguments from
+	 * the bootloader.
+	 */
+	if (IS_ENABLED(CONFIG_MIPS_CMDLINE_DTB_EXTEND) || !dt_bootargs)
+		bootcmdline_append(arcs_cmdline, COMMAND_LINE_SIZE);
+
+	/*
+	 * If the user specified a built-in command line & we didn't already
+	 * prepend it, we append it to boot_command_line here.
+	 */
+	if (IS_ENABLED(CONFIG_CMDLINE_BOOL) &&
+	    !IS_ENABLED(CONFIG_MIPS_CMDLINE_BUILTIN_EXTEND))
+		bootcmdline_append(builtin_cmdline, COMMAND_LINE_SIZE);
+}
+
+/*
+ * arch_mem_init - initialize memory management subsystem
+ *
+ *  o plat_mem_setup() detects the memory configuration and will record detected
+ *    memory areas using memblock_add.
+ *
+ * At this stage the memory configuration of the system is known to the
+ * kernel but generic memory management system is still entirely uninitialized.
+ *
+ *  o bootmem_init()
+ *  o sparse_init()
+ *  o paging_init()
+ *  o dma_contiguous_reserve()
+ *
+ * At this stage the bootmem allocator is ready to use.
+ *
+ * NOTE: historically plat_mem_setup did the entire platform initialization.
+ *	 This was rather impractical because it meant plat_mem_setup had to
+ * get away without any kind of memory allocator.  To keep old code from
+ * breaking plat_setup was just renamed to plat_mem_setup and a second platform
+ * initialization hook for anything else was introduced.
+ */
+static void __init arch_mem_init(char **cmdline_p)
+{
+	/* call board setup routine */
+	plat_mem_setup();
+	memblock_set_bottom_up(true);
+
+	bootcmdline_init();
+	strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE);
+	*cmdline_p = command_line;
+
+	parse_early_param();
+
+	if (usermem)
+		pr_info("User-defined physical RAM map overwrite\n");
+
+	check_kernel_sections_mem();
+
+	early_init_fdt_reserve_self();
+	early_init_fdt_scan_reserved_mem();
+
+#ifndef CONFIG_NUMA
+	memblock_set_node(0, PHYS_ADDR_MAX, &memblock.memory, 0);
+#endif
+	bootmem_init();
+
+	/*
+	 * Prevent memblock from allocating high memory.
+	 * This cannot be done before max_low_pfn is detected, so up
+	 * to this point is possible to only reserve physical memory
+	 * with memblock_reserve; memblock_alloc* can be used
+	 * only after this point
+	 */
+	memblock_set_current_limit(PFN_PHYS(max_low_pfn));
+
+#ifdef CONFIG_PROC_VMCORE
+	if (setup_elfcorehdr && setup_elfcorehdr_size) {
+		printk(KERN_INFO "kdump reserved memory at %lx-%lx\n",
+		       setup_elfcorehdr, setup_elfcorehdr_size);
+		memblock_reserve(setup_elfcorehdr, setup_elfcorehdr_size);
+	}
+#endif
+
+	mips_parse_crashkernel();
+#ifdef CONFIG_KEXEC
+	if (crashk_res.start != crashk_res.end)
+		memblock_reserve(crashk_res.start, resource_size(&crashk_res));
+#endif
+	device_tree_init();
+
+	/*
+	 * In order to reduce the possibility of kernel panic when failed to
+	 * get IO TLB memory under CONFIG_SWIOTLB, it is better to allocate
+	 * low memory as small as possible before plat_swiotlb_setup(), so
+	 * make sparse_init() using top-down allocation.
+	 */
+	memblock_set_bottom_up(false);
+	sparse_init();
+	memblock_set_bottom_up(true);
+
+	plat_swiotlb_setup();
+
+	dma_contiguous_reserve(PFN_PHYS(max_low_pfn));
+
+	/* Reserve for hibernation. */
+	memblock_reserve(__pa_symbol(&__nosave_begin),
+		__pa_symbol(&__nosave_end) - __pa_symbol(&__nosave_begin));
+
+	fdt_init_reserved_mem();
+
+	memblock_dump_all();
+
+	early_memtest(PFN_PHYS(ARCH_PFN_OFFSET), PFN_PHYS(max_low_pfn));
+}
+
+static void __init resource_init(void)
+{
+	phys_addr_t start, end;
+	u64 i;
+
+	if (UNCAC_BASE != IO_BASE)
+		return;
+
+	code_resource.start = __pa_symbol(&_text);
+	code_resource.end = __pa_symbol(&_etext) - 1;
+	data_resource.start = __pa_symbol(&_etext);
+	data_resource.end = __pa_symbol(&_edata) - 1;
+	bss_resource.start = __pa_symbol(&__bss_start);
+	bss_resource.end = __pa_symbol(&__bss_stop) - 1;
+
+	for_each_mem_range(i, &start, &end) {
+		struct resource *res;
+
+		res = memblock_alloc(sizeof(struct resource), SMP_CACHE_BYTES);
+		if (!res)
+			panic("%s: Failed to allocate %zu bytes\n", __func__,
+			      sizeof(struct resource));
+
+		res->start = start;
+		/*
+		 * In memblock, end points to the first byte after the
+		 * range while in resourses, end points to the last byte in
+		 * the range.
+		 */
+		res->end = end - 1;
+		res->flags = IORESOURCE_SYSTEM_RAM | IORESOURCE_BUSY;
+		res->name = "System RAM";
+
+		request_resource(&iomem_resource, res);
+
+		/*
+		 *  We don't know which RAM region contains kernel data,
+		 *  so we try it repeatedly and let the resource manager
+		 *  test it.
+		 */
+		request_resource(res, &code_resource);
+		request_resource(res, &data_resource);
+		request_resource(res, &bss_resource);
+		request_crashkernel(res);
+	}
+}
+
+#ifdef CONFIG_SMP
+static void __init prefill_possible_map(void)
+{
+	int i, possible = num_possible_cpus();
+
+	if (possible > nr_cpu_ids)
+		possible = nr_cpu_ids;
+
+	for (i = 0; i < possible; i++)
+		set_cpu_possible(i, true);
+	for (; i < NR_CPUS; i++)
+		set_cpu_possible(i, false);
+
+	nr_cpu_ids = possible;
+}
+#else
+static inline void prefill_possible_map(void) {}
+#endif
+
+void __init setup_arch(char **cmdline_p)
+{
+	cpu_probe();
+	mips_cm_probe();
+	prom_init();
+
+	setup_early_fdc_console();
+#ifdef CONFIG_EARLY_PRINTK
+	setup_early_printk();
+#endif
+	cpu_report();
+	check_bugs_early();
+
+#if defined(CONFIG_VT)
+#if defined(CONFIG_VGA_CONSOLE)
+	conswitchp = &vga_con;
+#endif
+#endif
+
+	arch_mem_init(cmdline_p);
+	dmi_setup();
+
+	resource_init();
+	plat_smp_setup();
+	prefill_possible_map();
+
+	cpu_cache_init();
+	paging_init();
+}
+
+unsigned long kernelsp[NR_CPUS];
+unsigned long fw_arg0, fw_arg1, fw_arg2, fw_arg3;
+
+#ifdef CONFIG_USE_OF
+unsigned long fw_passed_dtb;
+#endif
+
+#ifdef CONFIG_DEBUG_FS
+struct dentry *mips_debugfs_dir;
+static int __init debugfs_mips(void)
+{
+	mips_debugfs_dir = debugfs_create_dir("mips", NULL);
+	return 0;
+}
+arch_initcall(debugfs_mips);
+#endif
+
+#ifdef CONFIG_DMA_MAYBE_COHERENT
+/* User defined DMA coherency from command line. */
+enum coherent_io_user_state coherentio = IO_COHERENCE_DEFAULT;
+EXPORT_SYMBOL_GPL(coherentio);
+int hw_coherentio;	/* Actual hardware supported DMA coherency setting. */
+
+static int __init setcoherentio(char *str)
+{
+	coherentio = IO_COHERENCE_ENABLED;
+	pr_info("Hardware DMA cache coherency (command line)\n");
+	return 0;
+}
+early_param("coherentio", setcoherentio);
+
+static int __init setnocoherentio(char *str)
+{
+	coherentio = IO_COHERENCE_DISABLED;
+	pr_info("Software DMA cache coherency (command line)\n");
+	return 0;
+}
+early_param("nocoherentio", setnocoherentio);
+#endif
+
+void __init arch_cpu_finalize_init(void)
+{
+	unsigned int cpu = smp_processor_id();
+
+	cpu_data[cpu].udelay_val = loops_per_jiffy;
+	check_bugs32();
+
+	if (IS_ENABLED(CONFIG_CPU_R4X00_BUGS64))
+		check_bugs64();
+}
diff --git a/arch/mips/kernel/signal-common.h b/arch/mips/kernel/signal-common.h
new file mode 100644
index 000000000..f50d48435
--- /dev/null
+++ b/arch/mips/kernel/signal-common.h
@@ -0,0 +1,43 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1991, 1992  Linus Torvalds
+ * Copyright (C) 1994 - 2000  Ralf Baechle
+ * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ */
+
+#ifndef __SIGNAL_COMMON_H
+#define __SIGNAL_COMMON_H
+
+/* #define DEBUG_SIG */
+
+#ifdef DEBUG_SIG
+#  define DEBUGP(fmt, args...) printk("%s: " fmt, __func__, ##args)
+#else
+#  define DEBUGP(fmt, args...)
+#endif
+
+/*
+ * Determine which stack to use..
+ */
+extern void __user *get_sigframe(struct ksignal *ksig, struct pt_regs *regs,
+				 size_t frame_size);
+/* Check and clear pending FPU exceptions in saved CSR */
+extern int fpcsr_pending(unsigned int __user *fpcsr);
+
+/* Make sure we will not lose FPU ownership */
+#define lock_fpu_owner()	({ preempt_disable(); pagefault_disable(); })
+#define unlock_fpu_owner()	({ pagefault_enable(); preempt_enable(); })
+
+/* Assembly functions to move context to/from the FPU */
+extern asmlinkage int
+_save_fp_context(void __user *fpregs, void __user *csr);
+extern asmlinkage int
+_restore_fp_context(void __user *fpregs, void __user *csr);
+
+extern asmlinkage int _save_msa_all_upper(void __user *buf);
+extern asmlinkage int _restore_msa_all_upper(void __user *buf);
+
+#endif	/* __SIGNAL_COMMON_H */
diff --git a/arch/mips/kernel/signal.c b/arch/mips/kernel/signal.c
new file mode 100644
index 000000000..f1e985109
--- /dev/null
+++ b/arch/mips/kernel/signal.c
@@ -0,0 +1,962 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1991, 1992  Linus Torvalds
+ * Copyright (C) 1994 - 2000  Ralf Baechle
+ * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ * Copyright (C) 2014, Imagination Technologies Ltd.
+ */
+#include <linux/cache.h>
+#include <linux/context_tracking.h>
+#include <linux/irqflags.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/personality.h>
+#include <linux/smp.h>
+#include <linux/kernel.h>
+#include <linux/signal.h>
+#include <linux/errno.h>
+#include <linux/wait.h>
+#include <linux/ptrace.h>
+#include <linux/unistd.h>
+#include <linux/uprobes.h>
+#include <linux/compiler.h>
+#include <linux/syscalls.h>
+#include <linux/uaccess.h>
+#include <linux/tracehook.h>
+
+#include <asm/abi.h>
+#include <asm/asm.h>
+#include <linux/bitops.h>
+#include <asm/cacheflush.h>
+#include <asm/fpu.h>
+#include <asm/sim.h>
+#include <asm/ucontext.h>
+#include <asm/cpu-features.h>
+#include <asm/war.h>
+#include <asm/dsp.h>
+#include <asm/inst.h>
+#include <asm/msa.h>
+
+#include "signal-common.h"
+
+static int (*save_fp_context)(void __user *sc);
+static int (*restore_fp_context)(void __user *sc);
+
+struct sigframe {
+	u32 sf_ass[4];		/* argument save space for o32 */
+	u32 sf_pad[2];		/* Was: signal trampoline */
+
+	/* Matches struct ucontext from its uc_mcontext field onwards */
+	struct sigcontext sf_sc;
+	sigset_t sf_mask;
+	unsigned long long sf_extcontext[];
+};
+
+struct rt_sigframe {
+	u32 rs_ass[4];		/* argument save space for o32 */
+	u32 rs_pad[2];		/* Was: signal trampoline */
+	struct siginfo rs_info;
+	struct ucontext rs_uc;
+};
+
+#ifdef CONFIG_MIPS_FP_SUPPORT
+
+/*
+ * Thread saved context copy to/from a signal context presumed to be on the
+ * user stack, and therefore accessed with appropriate macros from uaccess.h.
+ */
+static int copy_fp_to_sigcontext(void __user *sc)
+{
+	struct mips_abi *abi = current->thread.abi;
+	uint64_t __user *fpregs = sc + abi->off_sc_fpregs;
+	uint32_t __user *csr = sc + abi->off_sc_fpc_csr;
+	int i;
+	int err = 0;
+	int inc = test_thread_flag(TIF_32BIT_FPREGS) ? 2 : 1;
+
+	for (i = 0; i < NUM_FPU_REGS; i += inc) {
+		err |=
+		    __put_user(get_fpr64(&current->thread.fpu.fpr[i], 0),
+			       &fpregs[i]);
+	}
+	err |= __put_user(current->thread.fpu.fcr31, csr);
+
+	return err;
+}
+
+static int copy_fp_from_sigcontext(void __user *sc)
+{
+	struct mips_abi *abi = current->thread.abi;
+	uint64_t __user *fpregs = sc + abi->off_sc_fpregs;
+	uint32_t __user *csr = sc + abi->off_sc_fpc_csr;
+	int i;
+	int err = 0;
+	int inc = test_thread_flag(TIF_32BIT_FPREGS) ? 2 : 1;
+	u64 fpr_val;
+
+	for (i = 0; i < NUM_FPU_REGS; i += inc) {
+		err |= __get_user(fpr_val, &fpregs[i]);
+		set_fpr64(&current->thread.fpu.fpr[i], 0, fpr_val);
+	}
+	err |= __get_user(current->thread.fpu.fcr31, csr);
+
+	return err;
+}
+
+#else /* !CONFIG_MIPS_FP_SUPPORT */
+
+static int copy_fp_to_sigcontext(void __user *sc)
+{
+	return 0;
+}
+
+static int copy_fp_from_sigcontext(void __user *sc)
+{
+	return 0;
+}
+
+#endif /* !CONFIG_MIPS_FP_SUPPORT */
+
+/*
+ * Wrappers for the assembly _{save,restore}_fp_context functions.
+ */
+static int save_hw_fp_context(void __user *sc)
+{
+	struct mips_abi *abi = current->thread.abi;
+	uint64_t __user *fpregs = sc + abi->off_sc_fpregs;
+	uint32_t __user *csr = sc + abi->off_sc_fpc_csr;
+
+	return _save_fp_context(fpregs, csr);
+}
+
+static int restore_hw_fp_context(void __user *sc)
+{
+	struct mips_abi *abi = current->thread.abi;
+	uint64_t __user *fpregs = sc + abi->off_sc_fpregs;
+	uint32_t __user *csr = sc + abi->off_sc_fpc_csr;
+
+	return _restore_fp_context(fpregs, csr);
+}
+
+/*
+ * Extended context handling.
+ */
+
+static inline void __user *sc_to_extcontext(void __user *sc)
+{
+	struct ucontext __user *uc;
+
+	/*
+	 * We can just pretend the sigcontext is always embedded in a struct
+	 * ucontext here, because the offset from sigcontext to extended
+	 * context is the same in the struct sigframe case.
+	 */
+	uc = container_of(sc, struct ucontext, uc_mcontext);
+	return &uc->uc_extcontext;
+}
+
+#ifdef CONFIG_CPU_HAS_MSA
+
+static int save_msa_extcontext(void __user *buf)
+{
+	struct msa_extcontext __user *msa = buf;
+	uint64_t val;
+	int i, err;
+
+	if (!thread_msa_context_live())
+		return 0;
+
+	/*
+	 * Ensure that we can't lose the live MSA context between checking
+	 * for it & writing it to memory.
+	 */
+	preempt_disable();
+
+	if (is_msa_enabled()) {
+		/*
+		 * There are no EVA versions of the vector register load/store
+		 * instructions, so MSA context has to be saved to kernel memory
+		 * and then copied to user memory. The save to kernel memory
+		 * should already have been done when handling scalar FP
+		 * context.
+		 */
+		BUG_ON(IS_ENABLED(CONFIG_EVA));
+
+		err = __put_user(read_msa_csr(), &msa->csr);
+		err |= _save_msa_all_upper(&msa->wr);
+
+		preempt_enable();
+	} else {
+		preempt_enable();
+
+		err = __put_user(current->thread.fpu.msacsr, &msa->csr);
+
+		for (i = 0; i < NUM_FPU_REGS; i++) {
+			val = get_fpr64(&current->thread.fpu.fpr[i], 1);
+			err |= __put_user(val, &msa->wr[i]);
+		}
+	}
+
+	err |= __put_user(MSA_EXTCONTEXT_MAGIC, &msa->ext.magic);
+	err |= __put_user(sizeof(*msa), &msa->ext.size);
+
+	return err ? -EFAULT : sizeof(*msa);
+}
+
+static int restore_msa_extcontext(void __user *buf, unsigned int size)
+{
+	struct msa_extcontext __user *msa = buf;
+	unsigned long long val;
+	unsigned int csr;
+	int i, err;
+
+	if (size != sizeof(*msa))
+		return -EINVAL;
+
+	err = get_user(csr, &msa->csr);
+	if (err)
+		return err;
+
+	preempt_disable();
+
+	if (is_msa_enabled()) {
+		/*
+		 * There are no EVA versions of the vector register load/store
+		 * instructions, so MSA context has to be copied to kernel
+		 * memory and later loaded to registers. The same is true of
+		 * scalar FP context, so FPU & MSA should have already been
+		 * disabled whilst handling scalar FP context.
+		 */
+		BUG_ON(IS_ENABLED(CONFIG_EVA));
+
+		write_msa_csr(csr);
+		err |= _restore_msa_all_upper(&msa->wr);
+		preempt_enable();
+	} else {
+		preempt_enable();
+
+		current->thread.fpu.msacsr = csr;
+
+		for (i = 0; i < NUM_FPU_REGS; i++) {
+			err |= __get_user(val, &msa->wr[i]);
+			set_fpr64(&current->thread.fpu.fpr[i], 1, val);
+		}
+	}
+
+	return err;
+}
+
+#else /* !CONFIG_CPU_HAS_MSA */
+
+static int save_msa_extcontext(void __user *buf)
+{
+	return 0;
+}
+
+static int restore_msa_extcontext(void __user *buf, unsigned int size)
+{
+	return SIGSYS;
+}
+
+#endif /* !CONFIG_CPU_HAS_MSA */
+
+static int save_extcontext(void __user *buf)
+{
+	int sz;
+
+	sz = save_msa_extcontext(buf);
+	if (sz < 0)
+		return sz;
+	buf += sz;
+
+	/* If no context was saved then trivially return */
+	if (!sz)
+		return 0;
+
+	/* Write the end marker */
+	if (__put_user(END_EXTCONTEXT_MAGIC, (u32 *)buf))
+		return -EFAULT;
+
+	sz += sizeof(((struct extcontext *)NULL)->magic);
+	return sz;
+}
+
+static int restore_extcontext(void __user *buf)
+{
+	struct extcontext ext;
+	int err;
+
+	while (1) {
+		err = __get_user(ext.magic, (unsigned int *)buf);
+		if (err)
+			return err;
+
+		if (ext.magic == END_EXTCONTEXT_MAGIC)
+			return 0;
+
+		err = __get_user(ext.size, (unsigned int *)(buf
+			+ offsetof(struct extcontext, size)));
+		if (err)
+			return err;
+
+		switch (ext.magic) {
+		case MSA_EXTCONTEXT_MAGIC:
+			err = restore_msa_extcontext(buf, ext.size);
+			break;
+
+		default:
+			err = -EINVAL;
+			break;
+		}
+
+		if (err)
+			return err;
+
+		buf += ext.size;
+	}
+}
+
+/*
+ * Helper routines
+ */
+int protected_save_fp_context(void __user *sc)
+{
+	struct mips_abi *abi = current->thread.abi;
+	uint64_t __user *fpregs = sc + abi->off_sc_fpregs;
+	uint32_t __user *csr = sc + abi->off_sc_fpc_csr;
+	uint32_t __user *used_math = sc + abi->off_sc_used_math;
+	unsigned int used, ext_sz;
+	int err;
+
+	used = used_math() ? USED_FP : 0;
+	if (!used)
+		goto fp_done;
+
+	if (!test_thread_flag(TIF_32BIT_FPREGS))
+		used |= USED_FR1;
+	if (test_thread_flag(TIF_HYBRID_FPREGS))
+		used |= USED_HYBRID_FPRS;
+
+	/*
+	 * EVA does not have userland equivalents of ldc1 or sdc1, so
+	 * save to the kernel FP context & copy that to userland below.
+	 */
+	if (IS_ENABLED(CONFIG_EVA))
+		lose_fpu(1);
+
+	while (1) {
+		lock_fpu_owner();
+		if (is_fpu_owner()) {
+			err = save_fp_context(sc);
+			unlock_fpu_owner();
+		} else {
+			unlock_fpu_owner();
+			err = copy_fp_to_sigcontext(sc);
+		}
+		if (likely(!err))
+			break;
+		/* touch the sigcontext and try again */
+		err = __put_user(0, &fpregs[0]) |
+			__put_user(0, &fpregs[31]) |
+			__put_user(0, csr);
+		if (err)
+			return err;	/* really bad sigcontext */
+	}
+
+fp_done:
+	ext_sz = err = save_extcontext(sc_to_extcontext(sc));
+	if (err < 0)
+		return err;
+	used |= ext_sz ? USED_EXTCONTEXT : 0;
+
+	return __put_user(used, used_math);
+}
+
+int protected_restore_fp_context(void __user *sc)
+{
+	struct mips_abi *abi = current->thread.abi;
+	uint64_t __user *fpregs = sc + abi->off_sc_fpregs;
+	uint32_t __user *csr = sc + abi->off_sc_fpc_csr;
+	uint32_t __user *used_math = sc + abi->off_sc_used_math;
+	unsigned int used;
+	int err, sig = 0, tmp __maybe_unused;
+
+	err = __get_user(used, used_math);
+	conditional_used_math(used & USED_FP);
+
+	/*
+	 * The signal handler may have used FPU; give it up if the program
+	 * doesn't want it following sigreturn.
+	 */
+	if (err || !(used & USED_FP))
+		lose_fpu(0);
+	if (err)
+		return err;
+	if (!(used & USED_FP))
+		goto fp_done;
+
+	err = sig = fpcsr_pending(csr);
+	if (err < 0)
+		return err;
+
+	/*
+	 * EVA does not have userland equivalents of ldc1 or sdc1, so we
+	 * disable the FPU here such that the code below simply copies to
+	 * the kernel FP context.
+	 */
+	if (IS_ENABLED(CONFIG_EVA))
+		lose_fpu(0);
+
+	while (1) {
+		lock_fpu_owner();
+		if (is_fpu_owner()) {
+			err = restore_fp_context(sc);
+			unlock_fpu_owner();
+		} else {
+			unlock_fpu_owner();
+			err = copy_fp_from_sigcontext(sc);
+		}
+		if (likely(!err))
+			break;
+		/* touch the sigcontext and try again */
+		err = __get_user(tmp, &fpregs[0]) |
+			__get_user(tmp, &fpregs[31]) |
+			__get_user(tmp, csr);
+		if (err)
+			break;	/* really bad sigcontext */
+	}
+
+fp_done:
+	if (!err && (used & USED_EXTCONTEXT))
+		err = restore_extcontext(sc_to_extcontext(sc));
+
+	return err ?: sig;
+}
+
+int setup_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc)
+{
+	int err = 0;
+	int i;
+
+	err |= __put_user(regs->cp0_epc, &sc->sc_pc);
+
+	err |= __put_user(0, &sc->sc_regs[0]);
+	for (i = 1; i < 32; i++)
+		err |= __put_user(regs->regs[i], &sc->sc_regs[i]);
+
+#ifdef CONFIG_CPU_HAS_SMARTMIPS
+	err |= __put_user(regs->acx, &sc->sc_acx);
+#endif
+	err |= __put_user(regs->hi, &sc->sc_mdhi);
+	err |= __put_user(regs->lo, &sc->sc_mdlo);
+	if (cpu_has_dsp) {
+		err |= __put_user(mfhi1(), &sc->sc_hi1);
+		err |= __put_user(mflo1(), &sc->sc_lo1);
+		err |= __put_user(mfhi2(), &sc->sc_hi2);
+		err |= __put_user(mflo2(), &sc->sc_lo2);
+		err |= __put_user(mfhi3(), &sc->sc_hi3);
+		err |= __put_user(mflo3(), &sc->sc_lo3);
+		err |= __put_user(rddsp(DSP_MASK), &sc->sc_dsp);
+	}
+
+
+	/*
+	 * Save FPU state to signal context. Signal handler
+	 * will "inherit" current FPU state.
+	 */
+	err |= protected_save_fp_context(sc);
+
+	return err;
+}
+
+static size_t extcontext_max_size(void)
+{
+	size_t sz = 0;
+
+	/*
+	 * The assumption here is that between this point & the point at which
+	 * the extended context is saved the size of the context should only
+	 * ever be able to shrink (if the task is preempted), but never grow.
+	 * That is, what this function returns is an upper bound on the size of
+	 * the extended context for the current task at the current time.
+	 */
+
+	if (thread_msa_context_live())
+		sz += sizeof(struct msa_extcontext);
+
+	/* If any context is saved then we'll append the end marker */
+	if (sz)
+		sz += sizeof(((struct extcontext *)NULL)->magic);
+
+	return sz;
+}
+
+int fpcsr_pending(unsigned int __user *fpcsr)
+{
+	int err, sig = 0;
+	unsigned int csr, enabled;
+
+	err = __get_user(csr, fpcsr);
+	enabled = FPU_CSR_UNI_X | ((csr & FPU_CSR_ALL_E) << 5);
+	/*
+	 * If the signal handler set some FPU exceptions, clear it and
+	 * send SIGFPE.
+	 */
+	if (csr & enabled) {
+		csr &= ~enabled;
+		err |= __put_user(csr, fpcsr);
+		sig = SIGFPE;
+	}
+	return err ?: sig;
+}
+
+int restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc)
+{
+	unsigned long treg;
+	int err = 0;
+	int i;
+
+	/* Always make any pending restarted system calls return -EINTR */
+	current->restart_block.fn = do_no_restart_syscall;
+
+	err |= __get_user(regs->cp0_epc, &sc->sc_pc);
+
+#ifdef CONFIG_CPU_HAS_SMARTMIPS
+	err |= __get_user(regs->acx, &sc->sc_acx);
+#endif
+	err |= __get_user(regs->hi, &sc->sc_mdhi);
+	err |= __get_user(regs->lo, &sc->sc_mdlo);
+	if (cpu_has_dsp) {
+		err |= __get_user(treg, &sc->sc_hi1); mthi1(treg);
+		err |= __get_user(treg, &sc->sc_lo1); mtlo1(treg);
+		err |= __get_user(treg, &sc->sc_hi2); mthi2(treg);
+		err |= __get_user(treg, &sc->sc_lo2); mtlo2(treg);
+		err |= __get_user(treg, &sc->sc_hi3); mthi3(treg);
+		err |= __get_user(treg, &sc->sc_lo3); mtlo3(treg);
+		err |= __get_user(treg, &sc->sc_dsp); wrdsp(treg, DSP_MASK);
+	}
+
+	for (i = 1; i < 32; i++)
+		err |= __get_user(regs->regs[i], &sc->sc_regs[i]);
+
+	return err ?: protected_restore_fp_context(sc);
+}
+
+#ifdef CONFIG_WAR_ICACHE_REFILLS
+#define SIGMASK		~(cpu_icache_line_size()-1)
+#else
+#define SIGMASK		ALMASK
+#endif
+
+void __user *get_sigframe(struct ksignal *ksig, struct pt_regs *regs,
+			  size_t frame_size)
+{
+	unsigned long sp;
+
+	/* Leave space for potential extended context */
+	frame_size += extcontext_max_size();
+
+	/* Default to using normal stack */
+	sp = regs->regs[29];
+
+	/*
+	 * FPU emulator may have it's own trampoline active just
+	 * above the user stack, 16-bytes before the next lowest
+	 * 16 byte boundary.  Try to avoid trashing it.
+	 */
+	sp -= 32;
+
+	sp = sigsp(sp, ksig);
+
+	return (void __user *)((sp - frame_size) & SIGMASK);
+}
+
+/*
+ * Atomically swap in the new signal mask, and wait for a signal.
+ */
+
+#ifdef CONFIG_TRAD_SIGNALS
+SYSCALL_DEFINE1(sigsuspend, sigset_t __user *, uset)
+{
+	return sys_rt_sigsuspend(uset, sizeof(sigset_t));
+}
+#endif
+
+#ifdef CONFIG_TRAD_SIGNALS
+SYSCALL_DEFINE3(sigaction, int, sig, const struct sigaction __user *, act,
+	struct sigaction __user *, oact)
+{
+	struct k_sigaction new_ka, old_ka;
+	int ret;
+	int err = 0;
+
+	if (act) {
+		old_sigset_t mask;
+
+		if (!access_ok(act, sizeof(*act)))
+			return -EFAULT;
+		err |= __get_user(new_ka.sa.sa_handler, &act->sa_handler);
+		err |= __get_user(new_ka.sa.sa_flags, &act->sa_flags);
+		err |= __get_user(mask, &act->sa_mask.sig[0]);
+		if (err)
+			return -EFAULT;
+
+		siginitset(&new_ka.sa.sa_mask, mask);
+	}
+
+	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
+
+	if (!ret && oact) {
+		if (!access_ok(oact, sizeof(*oact)))
+			return -EFAULT;
+		err |= __put_user(old_ka.sa.sa_flags, &oact->sa_flags);
+		err |= __put_user(old_ka.sa.sa_handler, &oact->sa_handler);
+		err |= __put_user(old_ka.sa.sa_mask.sig[0], oact->sa_mask.sig);
+		err |= __put_user(0, &oact->sa_mask.sig[1]);
+		err |= __put_user(0, &oact->sa_mask.sig[2]);
+		err |= __put_user(0, &oact->sa_mask.sig[3]);
+		if (err)
+			return -EFAULT;
+	}
+
+	return ret;
+}
+#endif
+
+#ifdef CONFIG_TRAD_SIGNALS
+asmlinkage void sys_sigreturn(void)
+{
+	struct sigframe __user *frame;
+	struct pt_regs *regs;
+	sigset_t blocked;
+	int sig;
+
+	regs = current_pt_regs();
+	frame = (struct sigframe __user *)regs->regs[29];
+	if (!access_ok(frame, sizeof(*frame)))
+		goto badframe;
+	if (__copy_from_user(&blocked, &frame->sf_mask, sizeof(blocked)))
+		goto badframe;
+
+	set_current_blocked(&blocked);
+
+	sig = restore_sigcontext(regs, &frame->sf_sc);
+	if (sig < 0)
+		goto badframe;
+	else if (sig)
+		force_sig(sig);
+
+	/*
+	 * Don't let your children do this ...
+	 */
+	__asm__ __volatile__(
+		"move\t$29, %0\n\t"
+		"j\tsyscall_exit"
+		: /* no outputs */
+		: "r" (regs));
+	/* Unreached */
+
+badframe:
+	force_sig(SIGSEGV);
+}
+#endif /* CONFIG_TRAD_SIGNALS */
+
+asmlinkage void sys_rt_sigreturn(void)
+{
+	struct rt_sigframe __user *frame;
+	struct pt_regs *regs;
+	sigset_t set;
+	int sig;
+
+	regs = current_pt_regs();
+	frame = (struct rt_sigframe __user *)regs->regs[29];
+	if (!access_ok(frame, sizeof(*frame)))
+		goto badframe;
+	if (__copy_from_user(&set, &frame->rs_uc.uc_sigmask, sizeof(set)))
+		goto badframe;
+
+	set_current_blocked(&set);
+
+	sig = restore_sigcontext(regs, &frame->rs_uc.uc_mcontext);
+	if (sig < 0)
+		goto badframe;
+	else if (sig)
+		force_sig(sig);
+
+	if (restore_altstack(&frame->rs_uc.uc_stack))
+		goto badframe;
+
+	/*
+	 * Don't let your children do this ...
+	 */
+	__asm__ __volatile__(
+		"move\t$29, %0\n\t"
+		"j\tsyscall_exit"
+		: /* no outputs */
+		: "r" (regs));
+	/* Unreached */
+
+badframe:
+	force_sig(SIGSEGV);
+}
+
+#ifdef CONFIG_TRAD_SIGNALS
+static int setup_frame(void *sig_return, struct ksignal *ksig,
+		       struct pt_regs *regs, sigset_t *set)
+{
+	struct sigframe __user *frame;
+	int err = 0;
+
+	frame = get_sigframe(ksig, regs, sizeof(*frame));
+	if (!access_ok(frame, sizeof (*frame)))
+		return -EFAULT;
+
+	err |= setup_sigcontext(regs, &frame->sf_sc);
+	err |= __copy_to_user(&frame->sf_mask, set, sizeof(*set));
+	if (err)
+		return -EFAULT;
+
+	/*
+	 * Arguments to signal handler:
+	 *
+	 *   a0 = signal number
+	 *   a1 = 0 (should be cause)
+	 *   a2 = pointer to struct sigcontext
+	 *
+	 * $25 and c0_epc point to the signal handler, $29 points to the
+	 * struct sigframe.
+	 */
+	regs->regs[ 4] = ksig->sig;
+	regs->regs[ 5] = 0;
+	regs->regs[ 6] = (unsigned long) &frame->sf_sc;
+	regs->regs[29] = (unsigned long) frame;
+	regs->regs[31] = (unsigned long) sig_return;
+	regs->cp0_epc = regs->regs[25] = (unsigned long) ksig->ka.sa.sa_handler;
+
+	DEBUGP("SIG deliver (%s:%d): sp=0x%p pc=0x%lx ra=0x%lx\n",
+	       current->comm, current->pid,
+	       frame, regs->cp0_epc, regs->regs[31]);
+	return 0;
+}
+#endif
+
+static int setup_rt_frame(void *sig_return, struct ksignal *ksig,
+			  struct pt_regs *regs, sigset_t *set)
+{
+	struct rt_sigframe __user *frame;
+	int err = 0;
+
+	frame = get_sigframe(ksig, regs, sizeof(*frame));
+	if (!access_ok(frame, sizeof (*frame)))
+		return -EFAULT;
+
+	/* Create siginfo.  */
+	err |= copy_siginfo_to_user(&frame->rs_info, &ksig->info);
+
+	/* Create the ucontext.	 */
+	err |= __put_user(0, &frame->rs_uc.uc_flags);
+	err |= __put_user(NULL, &frame->rs_uc.uc_link);
+	err |= __save_altstack(&frame->rs_uc.uc_stack, regs->regs[29]);
+	err |= setup_sigcontext(regs, &frame->rs_uc.uc_mcontext);
+	err |= __copy_to_user(&frame->rs_uc.uc_sigmask, set, sizeof(*set));
+
+	if (err)
+		return -EFAULT;
+
+	/*
+	 * Arguments to signal handler:
+	 *
+	 *   a0 = signal number
+	 *   a1 = 0 (should be cause)
+	 *   a2 = pointer to ucontext
+	 *
+	 * $25 and c0_epc point to the signal handler, $29 points to
+	 * the struct rt_sigframe.
+	 */
+	regs->regs[ 4] = ksig->sig;
+	regs->regs[ 5] = (unsigned long) &frame->rs_info;
+	regs->regs[ 6] = (unsigned long) &frame->rs_uc;
+	regs->regs[29] = (unsigned long) frame;
+	regs->regs[31] = (unsigned long) sig_return;
+	regs->cp0_epc = regs->regs[25] = (unsigned long) ksig->ka.sa.sa_handler;
+
+	DEBUGP("SIG deliver (%s:%d): sp=0x%p pc=0x%lx ra=0x%lx\n",
+	       current->comm, current->pid,
+	       frame, regs->cp0_epc, regs->regs[31]);
+
+	return 0;
+}
+
+struct mips_abi mips_abi = {
+#ifdef CONFIG_TRAD_SIGNALS
+	.setup_frame	= setup_frame,
+#endif
+	.setup_rt_frame = setup_rt_frame,
+	.restart	= __NR_restart_syscall,
+
+	.off_sc_fpregs = offsetof(struct sigcontext, sc_fpregs),
+	.off_sc_fpc_csr = offsetof(struct sigcontext, sc_fpc_csr),
+	.off_sc_used_math = offsetof(struct sigcontext, sc_used_math),
+
+	.vdso		= &vdso_image,
+};
+
+static void handle_signal(struct ksignal *ksig, struct pt_regs *regs)
+{
+	sigset_t *oldset = sigmask_to_save();
+	int ret;
+	struct mips_abi *abi = current->thread.abi;
+	void *vdso = current->mm->context.vdso;
+
+	/*
+	 * If we were emulating a delay slot instruction, exit that frame such
+	 * that addresses in the sigframe are as expected for userland and we
+	 * don't have a problem if we reuse the thread's frame for an
+	 * instruction within the signal handler.
+	 */
+	dsemul_thread_rollback(regs);
+
+	if (regs->regs[0]) {
+		switch(regs->regs[2]) {
+		case ERESTART_RESTARTBLOCK:
+		case ERESTARTNOHAND:
+			regs->regs[2] = EINTR;
+			break;
+		case ERESTARTSYS:
+			if (!(ksig->ka.sa.sa_flags & SA_RESTART)) {
+				regs->regs[2] = EINTR;
+				break;
+			}
+			fallthrough;
+		case ERESTARTNOINTR:
+			regs->regs[7] = regs->regs[26];
+			regs->regs[2] = regs->regs[0];
+			regs->cp0_epc -= 4;
+		}
+
+		regs->regs[0] = 0;		/* Don't deal with this again.	*/
+	}
+
+	rseq_signal_deliver(ksig, regs);
+
+	if (sig_uses_siginfo(&ksig->ka, abi))
+		ret = abi->setup_rt_frame(vdso + abi->vdso->off_rt_sigreturn,
+					  ksig, regs, oldset);
+	else
+		ret = abi->setup_frame(vdso + abi->vdso->off_sigreturn,
+				       ksig, regs, oldset);
+
+	signal_setup_done(ret, ksig, 0);
+}
+
+static void do_signal(struct pt_regs *regs)
+{
+	struct ksignal ksig;
+
+	if (get_signal(&ksig)) {
+		/* Whee!  Actually deliver the signal.	*/
+		handle_signal(&ksig, regs);
+		return;
+	}
+
+	if (regs->regs[0]) {
+		switch (regs->regs[2]) {
+		case ERESTARTNOHAND:
+		case ERESTARTSYS:
+		case ERESTARTNOINTR:
+			regs->regs[2] = regs->regs[0];
+			regs->regs[7] = regs->regs[26];
+			regs->cp0_epc -= 4;
+			break;
+
+		case ERESTART_RESTARTBLOCK:
+			regs->regs[2] = current->thread.abi->restart;
+			regs->regs[7] = regs->regs[26];
+			regs->cp0_epc -= 4;
+			break;
+		}
+		regs->regs[0] = 0;	/* Don't deal with this again.	*/
+	}
+
+	/*
+	 * If there's no signal to deliver, we just put the saved sigmask
+	 * back
+	 */
+	restore_saved_sigmask();
+}
+
+/*
+ * notification of userspace execution resumption
+ * - triggered by the TIF_WORK_MASK flags
+ */
+asmlinkage void do_notify_resume(struct pt_regs *regs, void *unused,
+	__u32 thread_info_flags)
+{
+	local_irq_enable();
+
+	user_exit();
+
+	if (thread_info_flags & _TIF_UPROBE)
+		uprobe_notify_resume(regs);
+
+	/* deal with pending signal delivery */
+	if (thread_info_flags & (_TIF_SIGPENDING | _TIF_NOTIFY_SIGNAL))
+		do_signal(regs);
+
+	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
+		tracehook_notify_resume(regs);
+		rseq_handle_notify_resume(NULL, regs);
+	}
+
+	user_enter();
+}
+
+#if defined(CONFIG_SMP) && defined(CONFIG_MIPS_FP_SUPPORT)
+static int smp_save_fp_context(void __user *sc)
+{
+	return raw_cpu_has_fpu
+	       ? save_hw_fp_context(sc)
+	       : copy_fp_to_sigcontext(sc);
+}
+
+static int smp_restore_fp_context(void __user *sc)
+{
+	return raw_cpu_has_fpu
+	       ? restore_hw_fp_context(sc)
+	       : copy_fp_from_sigcontext(sc);
+}
+#endif
+
+static int signal_setup(void)
+{
+	/*
+	 * The offset from sigcontext to extended context should be the same
+	 * regardless of the type of signal, such that userland can always know
+	 * where to look if it wishes to find the extended context structures.
+	 */
+	BUILD_BUG_ON((offsetof(struct sigframe, sf_extcontext) -
+		      offsetof(struct sigframe, sf_sc)) !=
+		     (offsetof(struct rt_sigframe, rs_uc.uc_extcontext) -
+		      offsetof(struct rt_sigframe, rs_uc.uc_mcontext)));
+
+#if defined(CONFIG_SMP) && defined(CONFIG_MIPS_FP_SUPPORT)
+	/* For now just do the cpu_has_fpu check when the functions are invoked */
+	save_fp_context = smp_save_fp_context;
+	restore_fp_context = smp_restore_fp_context;
+#else
+	if (cpu_has_fpu) {
+		save_fp_context = save_hw_fp_context;
+		restore_fp_context = restore_hw_fp_context;
+	} else {
+		save_fp_context = copy_fp_to_sigcontext;
+		restore_fp_context = copy_fp_from_sigcontext;
+	}
+#endif /* CONFIG_SMP */
+
+	return 0;
+}
+
+arch_initcall(signal_setup);
diff --git a/arch/mips/kernel/signal32.c b/arch/mips/kernel/signal32.c
new file mode 100644
index 000000000..59b896543
--- /dev/null
+++ b/arch/mips/kernel/signal32.c
@@ -0,0 +1,78 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1991, 1992  Linus Torvalds
+ * Copyright (C) 1994 - 2000, 2006  Ralf Baechle
+ * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ * Copyright (C) 2016, Imagination Technologies Ltd.
+ */
+#include <linux/compat.h>
+#include <linux/compiler.h>
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/signal.h>
+#include <linux/syscalls.h>
+
+#include <asm/compat-signal.h>
+#include <linux/uaccess.h>
+#include <asm/unistd.h>
+
+#include "signal-common.h"
+
+/* 32-bit compatibility types */
+
+typedef unsigned int __sighandler32_t;
+typedef void (*vfptr_t)(void);
+
+/*
+ * Atomically swap in the new signal mask, and wait for a signal.
+ */
+
+asmlinkage int sys32_sigsuspend(compat_sigset_t __user *uset)
+{
+	return compat_sys_rt_sigsuspend(uset, sizeof(compat_sigset_t));
+}
+
+SYSCALL_DEFINE3(32_sigaction, long, sig, const struct compat_sigaction __user *, act,
+	struct compat_sigaction __user *, oact)
+{
+	struct k_sigaction new_ka, old_ka;
+	int ret;
+	int err = 0;
+
+	if (act) {
+		old_sigset_t mask;
+		s32 handler;
+
+		if (!access_ok(act, sizeof(*act)))
+			return -EFAULT;
+		err |= __get_user(handler, &act->sa_handler);
+		new_ka.sa.sa_handler = (void __user *)(s64)handler;
+		err |= __get_user(new_ka.sa.sa_flags, &act->sa_flags);
+		err |= __get_user(mask, &act->sa_mask.sig[0]);
+		if (err)
+			return -EFAULT;
+
+		siginitset(&new_ka.sa.sa_mask, mask);
+	}
+
+	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
+
+	if (!ret && oact) {
+		if (!access_ok(oact, sizeof(*oact)))
+			return -EFAULT;
+		err |= __put_user(old_ka.sa.sa_flags, &oact->sa_flags);
+		err |= __put_user((u32)(u64)old_ka.sa.sa_handler,
+				  &oact->sa_handler);
+		err |= __put_user(old_ka.sa.sa_mask.sig[0], oact->sa_mask.sig);
+		err |= __put_user(0, &oact->sa_mask.sig[1]);
+		err |= __put_user(0, &oact->sa_mask.sig[2]);
+		err |= __put_user(0, &oact->sa_mask.sig[3]);
+		if (err)
+			return -EFAULT;
+	}
+
+	return ret;
+}
diff --git a/arch/mips/kernel/signal_n32.c b/arch/mips/kernel/signal_n32.c
new file mode 100644
index 000000000..7bd00fad6
--- /dev/null
+++ b/arch/mips/kernel/signal_n32.c
@@ -0,0 +1,149 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2003 Broadcom Corporation
+ */
+#include <linux/cache.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/kernel.h>
+#include <linux/signal.h>
+#include <linux/errno.h>
+#include <linux/wait.h>
+#include <linux/ptrace.h>
+#include <linux/unistd.h>
+#include <linux/compat.h>
+#include <linux/bitops.h>
+
+#include <asm/abi.h>
+#include <asm/asm.h>
+#include <asm/cacheflush.h>
+#include <asm/compat-signal.h>
+#include <asm/sim.h>
+#include <linux/uaccess.h>
+#include <asm/ucontext.h>
+#include <asm/fpu.h>
+#include <asm/cpu-features.h>
+#include <asm/war.h>
+
+#include "signal-common.h"
+
+/*
+ * Including <asm/unistd.h> would give use the 64-bit syscall numbers ...
+ */
+#define __NR_N32_restart_syscall	6214
+
+extern int setup_sigcontext(struct pt_regs *, struct sigcontext __user *);
+extern int restore_sigcontext(struct pt_regs *, struct sigcontext __user *);
+
+struct ucontextn32 {
+	u32		    uc_flags;
+	s32		    uc_link;
+	compat_stack_t      uc_stack;
+	struct sigcontext   uc_mcontext;
+	compat_sigset_t	    uc_sigmask;	  /* mask last for extensibility */
+};
+
+struct rt_sigframe_n32 {
+	u32 rs_ass[4];			/* argument save space for o32 */
+	u32 rs_pad[2];			/* Was: signal trampoline */
+	struct compat_siginfo rs_info;
+	struct ucontextn32 rs_uc;
+};
+
+asmlinkage void sysn32_rt_sigreturn(void)
+{
+	struct rt_sigframe_n32 __user *frame;
+	struct pt_regs *regs;
+	sigset_t set;
+	int sig;
+
+	regs = current_pt_regs();
+	frame = (struct rt_sigframe_n32 __user *)regs->regs[29];
+	if (!access_ok(frame, sizeof(*frame)))
+		goto badframe;
+	if (__copy_conv_sigset_from_user(&set, &frame->rs_uc.uc_sigmask))
+		goto badframe;
+
+	set_current_blocked(&set);
+
+	sig = restore_sigcontext(regs, &frame->rs_uc.uc_mcontext);
+	if (sig < 0)
+		goto badframe;
+	else if (sig)
+		force_sig(sig);
+
+	if (compat_restore_altstack(&frame->rs_uc.uc_stack))
+		goto badframe;
+
+	/*
+	 * Don't let your children do this ...
+	 */
+	__asm__ __volatile__(
+		"move\t$29, %0\n\t"
+		"j\tsyscall_exit"
+		: /* no outputs */
+		: "r" (regs));
+	/* Unreached */
+
+badframe:
+	force_sig(SIGSEGV);
+}
+
+static int setup_rt_frame_n32(void *sig_return, struct ksignal *ksig,
+			      struct pt_regs *regs, sigset_t *set)
+{
+	struct rt_sigframe_n32 __user *frame;
+	int err = 0;
+
+	frame = get_sigframe(ksig, regs, sizeof(*frame));
+	if (!access_ok(frame, sizeof (*frame)))
+		return -EFAULT;
+
+	/* Create siginfo.  */
+	err |= copy_siginfo_to_user32(&frame->rs_info, &ksig->info);
+
+	/* Create the ucontext.	 */
+	err |= __put_user(0, &frame->rs_uc.uc_flags);
+	err |= __put_user(0, &frame->rs_uc.uc_link);
+	err |= __compat_save_altstack(&frame->rs_uc.uc_stack, regs->regs[29]);
+	err |= setup_sigcontext(regs, &frame->rs_uc.uc_mcontext);
+	err |= __copy_conv_sigset_to_user(&frame->rs_uc.uc_sigmask, set);
+
+	if (err)
+		return -EFAULT;
+
+	/*
+	 * Arguments to signal handler:
+	 *
+	 *   a0 = signal number
+	 *   a1 = 0 (should be cause)
+	 *   a2 = pointer to ucontext
+	 *
+	 * $25 and c0_epc point to the signal handler, $29 points to
+	 * the struct rt_sigframe.
+	 */
+	regs->regs[ 4] = ksig->sig;
+	regs->regs[ 5] = (unsigned long) &frame->rs_info;
+	regs->regs[ 6] = (unsigned long) &frame->rs_uc;
+	regs->regs[29] = (unsigned long) frame;
+	regs->regs[31] = (unsigned long) sig_return;
+	regs->cp0_epc = regs->regs[25] = (unsigned long) ksig->ka.sa.sa_handler;
+
+	DEBUGP("SIG deliver (%s:%d): sp=0x%p pc=0x%lx ra=0x%lx\n",
+	       current->comm, current->pid,
+	       frame, regs->cp0_epc, regs->regs[31]);
+
+	return 0;
+}
+
+struct mips_abi mips_abi_n32 = {
+	.setup_rt_frame = setup_rt_frame_n32,
+	.restart	= __NR_N32_restart_syscall,
+
+	.off_sc_fpregs = offsetof(struct sigcontext, sc_fpregs),
+	.off_sc_fpc_csr = offsetof(struct sigcontext, sc_fpc_csr),
+	.off_sc_used_math = offsetof(struct sigcontext, sc_used_math),
+
+	.vdso		= &vdso_image_n32,
+};
diff --git a/arch/mips/kernel/signal_o32.c b/arch/mips/kernel/signal_o32.c
new file mode 100644
index 000000000..299a7a28c
--- /dev/null
+++ b/arch/mips/kernel/signal_o32.c
@@ -0,0 +1,290 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1991, 1992  Linus Torvalds
+ * Copyright (C) 1994 - 2000, 2006  Ralf Baechle
+ * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ * Copyright (C) 2016, Imagination Technologies Ltd.
+ */
+#include <linux/compiler.h>
+#include <linux/errno.h>
+#include <linux/signal.h>
+#include <linux/sched/signal.h>
+#include <linux/uaccess.h>
+
+#include <asm/abi.h>
+#include <asm/compat-signal.h>
+#include <asm/dsp.h>
+#include <asm/sim.h>
+#include <asm/unistd.h>
+
+#include "signal-common.h"
+
+/*
+ * Including <asm/unistd.h> would give use the 64-bit syscall numbers ...
+ */
+#define __NR_O32_restart_syscall	4253
+
+struct sigframe32 {
+	u32 sf_ass[4];		/* argument save space for o32 */
+	u32 sf_pad[2];		/* Was: signal trampoline */
+	struct sigcontext32 sf_sc;
+	compat_sigset_t sf_mask;
+};
+
+struct ucontext32 {
+	u32		    uc_flags;
+	s32		    uc_link;
+	compat_stack_t      uc_stack;
+	struct sigcontext32 uc_mcontext;
+	compat_sigset_t	    uc_sigmask;	  /* mask last for extensibility */
+};
+
+struct rt_sigframe32 {
+	u32 rs_ass[4];			/* argument save space for o32 */
+	u32 rs_pad[2];			/* Was: signal trampoline */
+	compat_siginfo_t rs_info;
+	struct ucontext32 rs_uc;
+};
+
+static int setup_sigcontext32(struct pt_regs *regs,
+			      struct sigcontext32 __user *sc)
+{
+	int err = 0;
+	int i;
+
+	err |= __put_user(regs->cp0_epc, &sc->sc_pc);
+
+	err |= __put_user(0, &sc->sc_regs[0]);
+	for (i = 1; i < 32; i++)
+		err |= __put_user(regs->regs[i], &sc->sc_regs[i]);
+
+	err |= __put_user(regs->hi, &sc->sc_mdhi);
+	err |= __put_user(regs->lo, &sc->sc_mdlo);
+	if (cpu_has_dsp) {
+		err |= __put_user(rddsp(DSP_MASK), &sc->sc_dsp);
+		err |= __put_user(mfhi1(), &sc->sc_hi1);
+		err |= __put_user(mflo1(), &sc->sc_lo1);
+		err |= __put_user(mfhi2(), &sc->sc_hi2);
+		err |= __put_user(mflo2(), &sc->sc_lo2);
+		err |= __put_user(mfhi3(), &sc->sc_hi3);
+		err |= __put_user(mflo3(), &sc->sc_lo3);
+	}
+
+	/*
+	 * Save FPU state to signal context.  Signal handler
+	 * will "inherit" current FPU state.
+	 */
+	err |= protected_save_fp_context(sc);
+
+	return err;
+}
+
+static int restore_sigcontext32(struct pt_regs *regs,
+				struct sigcontext32 __user *sc)
+{
+	int err = 0;
+	s32 treg;
+	int i;
+
+	/* Always make any pending restarted system calls return -EINTR */
+	current->restart_block.fn = do_no_restart_syscall;
+
+	err |= __get_user(regs->cp0_epc, &sc->sc_pc);
+	err |= __get_user(regs->hi, &sc->sc_mdhi);
+	err |= __get_user(regs->lo, &sc->sc_mdlo);
+	if (cpu_has_dsp) {
+		err |= __get_user(treg, &sc->sc_hi1); mthi1(treg);
+		err |= __get_user(treg, &sc->sc_lo1); mtlo1(treg);
+		err |= __get_user(treg, &sc->sc_hi2); mthi2(treg);
+		err |= __get_user(treg, &sc->sc_lo2); mtlo2(treg);
+		err |= __get_user(treg, &sc->sc_hi3); mthi3(treg);
+		err |= __get_user(treg, &sc->sc_lo3); mtlo3(treg);
+		err |= __get_user(treg, &sc->sc_dsp); wrdsp(treg, DSP_MASK);
+	}
+
+	for (i = 1; i < 32; i++)
+		err |= __get_user(regs->regs[i], &sc->sc_regs[i]);
+
+	return err ?: protected_restore_fp_context(sc);
+}
+
+static int setup_frame_32(void *sig_return, struct ksignal *ksig,
+			  struct pt_regs *regs, sigset_t *set)
+{
+	struct sigframe32 __user *frame;
+	int err = 0;
+
+	frame = get_sigframe(ksig, regs, sizeof(*frame));
+	if (!access_ok(frame, sizeof (*frame)))
+		return -EFAULT;
+
+	err |= setup_sigcontext32(regs, &frame->sf_sc);
+	err |= __copy_conv_sigset_to_user(&frame->sf_mask, set);
+
+	if (err)
+		return -EFAULT;
+
+	/*
+	 * Arguments to signal handler:
+	 *
+	 *   a0 = signal number
+	 *   a1 = 0 (should be cause)
+	 *   a2 = pointer to struct sigcontext
+	 *
+	 * $25 and c0_epc point to the signal handler, $29 points to the
+	 * struct sigframe.
+	 */
+	regs->regs[ 4] = ksig->sig;
+	regs->regs[ 5] = 0;
+	regs->regs[ 6] = (unsigned long) &frame->sf_sc;
+	regs->regs[29] = (unsigned long) frame;
+	regs->regs[31] = (unsigned long) sig_return;
+	regs->cp0_epc = regs->regs[25] = (unsigned long) ksig->ka.sa.sa_handler;
+
+	DEBUGP("SIG deliver (%s:%d): sp=0x%p pc=0x%lx ra=0x%lx\n",
+	       current->comm, current->pid,
+	       frame, regs->cp0_epc, regs->regs[31]);
+
+	return 0;
+}
+
+asmlinkage void sys32_rt_sigreturn(void)
+{
+	struct rt_sigframe32 __user *frame;
+	struct pt_regs *regs;
+	sigset_t set;
+	int sig;
+
+	regs = current_pt_regs();
+	frame = (struct rt_sigframe32 __user *)regs->regs[29];
+	if (!access_ok(frame, sizeof(*frame)))
+		goto badframe;
+	if (__copy_conv_sigset_from_user(&set, &frame->rs_uc.uc_sigmask))
+		goto badframe;
+
+	set_current_blocked(&set);
+
+	sig = restore_sigcontext32(regs, &frame->rs_uc.uc_mcontext);
+	if (sig < 0)
+		goto badframe;
+	else if (sig)
+		force_sig(sig);
+
+	if (compat_restore_altstack(&frame->rs_uc.uc_stack))
+		goto badframe;
+
+	/*
+	 * Don't let your children do this ...
+	 */
+	__asm__ __volatile__(
+		"move\t$29, %0\n\t"
+		"j\tsyscall_exit"
+		: /* no outputs */
+		: "r" (regs));
+	/* Unreached */
+
+badframe:
+	force_sig(SIGSEGV);
+}
+
+static int setup_rt_frame_32(void *sig_return, struct ksignal *ksig,
+			     struct pt_regs *regs, sigset_t *set)
+{
+	struct rt_sigframe32 __user *frame;
+	int err = 0;
+
+	frame = get_sigframe(ksig, regs, sizeof(*frame));
+	if (!access_ok(frame, sizeof (*frame)))
+		return -EFAULT;
+
+	/* Convert (siginfo_t -> compat_siginfo_t) and copy to user. */
+	err |= copy_siginfo_to_user32(&frame->rs_info, &ksig->info);
+
+	/* Create the ucontext.	 */
+	err |= __put_user(0, &frame->rs_uc.uc_flags);
+	err |= __put_user(0, &frame->rs_uc.uc_link);
+	err |= __compat_save_altstack(&frame->rs_uc.uc_stack, regs->regs[29]);
+	err |= setup_sigcontext32(regs, &frame->rs_uc.uc_mcontext);
+	err |= __copy_conv_sigset_to_user(&frame->rs_uc.uc_sigmask, set);
+
+	if (err)
+		return -EFAULT;
+
+	/*
+	 * Arguments to signal handler:
+	 *
+	 *   a0 = signal number
+	 *   a1 = 0 (should be cause)
+	 *   a2 = pointer to ucontext
+	 *
+	 * $25 and c0_epc point to the signal handler, $29 points to
+	 * the struct rt_sigframe32.
+	 */
+	regs->regs[ 4] = ksig->sig;
+	regs->regs[ 5] = (unsigned long) &frame->rs_info;
+	regs->regs[ 6] = (unsigned long) &frame->rs_uc;
+	regs->regs[29] = (unsigned long) frame;
+	regs->regs[31] = (unsigned long) sig_return;
+	regs->cp0_epc = regs->regs[25] = (unsigned long) ksig->ka.sa.sa_handler;
+
+	DEBUGP("SIG deliver (%s:%d): sp=0x%p pc=0x%lx ra=0x%lx\n",
+	       current->comm, current->pid,
+	       frame, regs->cp0_epc, regs->regs[31]);
+
+	return 0;
+}
+
+/*
+ * o32 compatibility on 64-bit kernels, without DSP ASE
+ */
+struct mips_abi mips_abi_32 = {
+	.setup_frame	= setup_frame_32,
+	.setup_rt_frame = setup_rt_frame_32,
+	.restart	= __NR_O32_restart_syscall,
+
+	.off_sc_fpregs = offsetof(struct sigcontext32, sc_fpregs),
+	.off_sc_fpc_csr = offsetof(struct sigcontext32, sc_fpc_csr),
+	.off_sc_used_math = offsetof(struct sigcontext32, sc_used_math),
+
+	.vdso		= &vdso_image_o32,
+};
+
+
+asmlinkage void sys32_sigreturn(void)
+{
+	struct sigframe32 __user *frame;
+	struct pt_regs *regs;
+	sigset_t blocked;
+	int sig;
+
+	regs = current_pt_regs();
+	frame = (struct sigframe32 __user *)regs->regs[29];
+	if (!access_ok(frame, sizeof(*frame)))
+		goto badframe;
+	if (__copy_conv_sigset_from_user(&blocked, &frame->sf_mask))
+		goto badframe;
+
+	set_current_blocked(&blocked);
+
+	sig = restore_sigcontext32(regs, &frame->sf_sc);
+	if (sig < 0)
+		goto badframe;
+	else if (sig)
+		force_sig(sig);
+
+	/*
+	 * Don't let your children do this ...
+	 */
+	__asm__ __volatile__(
+		"move\t$29, %0\n\t"
+		"j\tsyscall_exit"
+		: /* no outputs */
+		: "r" (regs));
+	/* Unreached */
+
+badframe:
+	force_sig(SIGSEGV);
+}
diff --git a/arch/mips/kernel/smp-bmips.c b/arch/mips/kernel/smp-bmips.c
new file mode 100644
index 000000000..1dbfb5aad
--- /dev/null
+++ b/arch/mips/kernel/smp-bmips.c
@@ -0,0 +1,667 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2011 by Kevin Cernekee (cernekee@gmail.com)
+ *
+ * SMP support for BMIPS
+ */
+
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/sched/hotplug.h>
+#include <linux/sched/task_stack.h>
+#include <linux/mm.h>
+#include <linux/delay.h>
+#include <linux/smp.h>
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/cpu.h>
+#include <linux/cpumask.h>
+#include <linux/reboot.h>
+#include <linux/io.h>
+#include <linux/compiler.h>
+#include <linux/linkage.h>
+#include <linux/bug.h>
+#include <linux/kernel.h>
+#include <linux/kexec.h>
+
+#include <asm/time.h>
+#include <asm/processor.h>
+#include <asm/bootinfo.h>
+#include <asm/cacheflush.h>
+#include <asm/tlbflush.h>
+#include <asm/mipsregs.h>
+#include <asm/bmips.h>
+#include <asm/traps.h>
+#include <asm/barrier.h>
+#include <asm/cpu-features.h>
+
+static int __maybe_unused max_cpus = 1;
+
+/* these may be configured by the platform code */
+int bmips_smp_enabled = 1;
+int bmips_cpu_offset;
+cpumask_t bmips_booted_mask;
+unsigned long bmips_tp1_irqs = IE_IRQ1;
+
+#define RESET_FROM_KSEG0		0x80080800
+#define RESET_FROM_KSEG1		0xa0080800
+
+static void bmips_set_reset_vec(int cpu, u32 val);
+
+#ifdef CONFIG_SMP
+
+/* initial $sp, $gp - used by arch/mips/kernel/bmips_vec.S */
+unsigned long bmips_smp_boot_sp;
+unsigned long bmips_smp_boot_gp;
+
+static void bmips43xx_send_ipi_single(int cpu, unsigned int action);
+static void bmips5000_send_ipi_single(int cpu, unsigned int action);
+static irqreturn_t bmips43xx_ipi_interrupt(int irq, void *dev_id);
+static irqreturn_t bmips5000_ipi_interrupt(int irq, void *dev_id);
+
+/* SW interrupts 0,1 are used for interprocessor signaling */
+#define IPI0_IRQ			(MIPS_CPU_IRQ_BASE + 0)
+#define IPI1_IRQ			(MIPS_CPU_IRQ_BASE + 1)
+
+#define CPUNUM(cpu, shift)		(((cpu) + bmips_cpu_offset) << (shift))
+#define ACTION_CLR_IPI(cpu, ipi)	(0x2000 | CPUNUM(cpu, 9) | ((ipi) << 8))
+#define ACTION_SET_IPI(cpu, ipi)	(0x3000 | CPUNUM(cpu, 9) | ((ipi) << 8))
+#define ACTION_BOOT_THREAD(cpu)		(0x08 | CPUNUM(cpu, 0))
+
+static void __init bmips_smp_setup(void)
+{
+	int i, cpu = 1, boot_cpu = 0;
+	int cpu_hw_intr;
+
+	switch (current_cpu_type()) {
+	case CPU_BMIPS4350:
+	case CPU_BMIPS4380:
+		/* arbitration priority */
+		clear_c0_brcm_cmt_ctrl(0x30);
+
+		/* NBK and weak order flags */
+		set_c0_brcm_config_0(0x30000);
+
+		/* Find out if we are running on TP0 or TP1 */
+		boot_cpu = !!(read_c0_brcm_cmt_local() & (1 << 31));
+
+		/*
+		 * MIPS interrupts 0,1 (SW INT 0,1) cross over to the other
+		 * thread
+		 * MIPS interrupt 2 (HW INT 0) is the CPU0 L1 controller output
+		 * MIPS interrupt 3 (HW INT 1) is the CPU1 L1 controller output
+		 */
+		if (boot_cpu == 0)
+			cpu_hw_intr = 0x02;
+		else
+			cpu_hw_intr = 0x1d;
+
+		change_c0_brcm_cmt_intr(0xf8018000,
+					(cpu_hw_intr << 27) | (0x03 << 15));
+
+		/* single core, 2 threads (2 pipelines) */
+		max_cpus = 2;
+
+		break;
+	case CPU_BMIPS5000:
+		/* enable raceless SW interrupts */
+		set_c0_brcm_config(0x03 << 22);
+
+		/* route HW interrupt 0 to CPU0, HW interrupt 1 to CPU1 */
+		change_c0_brcm_mode(0x1f << 27, 0x02 << 27);
+
+		/* N cores, 2 threads per core */
+		max_cpus = (((read_c0_brcm_config() >> 6) & 0x03) + 1) << 1;
+
+		/* clear any pending SW interrupts */
+		for (i = 0; i < max_cpus; i++) {
+			write_c0_brcm_action(ACTION_CLR_IPI(i, 0));
+			write_c0_brcm_action(ACTION_CLR_IPI(i, 1));
+		}
+
+		break;
+	default:
+		max_cpus = 1;
+	}
+
+	if (!bmips_smp_enabled)
+		max_cpus = 1;
+
+	/* this can be overridden by the BSP */
+	if (!board_ebase_setup)
+		board_ebase_setup = &bmips_ebase_setup;
+
+	__cpu_number_map[boot_cpu] = 0;
+	__cpu_logical_map[0] = boot_cpu;
+
+	for (i = 0; i < max_cpus; i++) {
+		if (i != boot_cpu) {
+			__cpu_number_map[i] = cpu;
+			__cpu_logical_map[cpu] = i;
+			cpu++;
+		}
+		set_cpu_possible(i, 1);
+		set_cpu_present(i, 1);
+	}
+}
+
+/*
+ * IPI IRQ setup - runs on CPU0
+ */
+static void bmips_prepare_cpus(unsigned int max_cpus)
+{
+	irqreturn_t (*bmips_ipi_interrupt)(int irq, void *dev_id);
+
+	switch (current_cpu_type()) {
+	case CPU_BMIPS4350:
+	case CPU_BMIPS4380:
+		bmips_ipi_interrupt = bmips43xx_ipi_interrupt;
+		break;
+	case CPU_BMIPS5000:
+		bmips_ipi_interrupt = bmips5000_ipi_interrupt;
+		break;
+	default:
+		return;
+	}
+
+	if (request_irq(IPI0_IRQ, bmips_ipi_interrupt,
+			IRQF_PERCPU | IRQF_NO_SUSPEND, "smp_ipi0", NULL))
+		panic("Can't request IPI0 interrupt");
+	if (request_irq(IPI1_IRQ, bmips_ipi_interrupt,
+			IRQF_PERCPU | IRQF_NO_SUSPEND, "smp_ipi1", NULL))
+		panic("Can't request IPI1 interrupt");
+}
+
+/*
+ * Tell the hardware to boot CPUx - runs on CPU0
+ */
+static int bmips_boot_secondary(int cpu, struct task_struct *idle)
+{
+	bmips_smp_boot_sp = __KSTK_TOS(idle);
+	bmips_smp_boot_gp = (unsigned long)task_thread_info(idle);
+	mb();
+
+	/*
+	 * Initial boot sequence for secondary CPU:
+	 *   bmips_reset_nmi_vec @ a000_0000 ->
+	 *   bmips_smp_entry ->
+	 *   plat_wired_tlb_setup (cached function call; optional) ->
+	 *   start_secondary (cached jump)
+	 *
+	 * Warm restart sequence:
+	 *   play_dead WAIT loop ->
+	 *   bmips_smp_int_vec @ BMIPS_WARM_RESTART_VEC ->
+	 *   eret to play_dead ->
+	 *   bmips_secondary_reentry ->
+	 *   start_secondary
+	 */
+
+	pr_info("SMP: Booting CPU%d...\n", cpu);
+
+	if (cpumask_test_cpu(cpu, &bmips_booted_mask)) {
+		/* kseg1 might not exist if this CPU enabled XKS01 */
+		bmips_set_reset_vec(cpu, RESET_FROM_KSEG0);
+
+		switch (current_cpu_type()) {
+		case CPU_BMIPS4350:
+		case CPU_BMIPS4380:
+			bmips43xx_send_ipi_single(cpu, 0);
+			break;
+		case CPU_BMIPS5000:
+			bmips5000_send_ipi_single(cpu, 0);
+			break;
+		}
+	} else {
+		bmips_set_reset_vec(cpu, RESET_FROM_KSEG1);
+
+		switch (current_cpu_type()) {
+		case CPU_BMIPS4350:
+		case CPU_BMIPS4380:
+			/* Reset slave TP1 if booting from TP0 */
+			if (cpu_logical_map(cpu) == 1)
+				set_c0_brcm_cmt_ctrl(0x01);
+			break;
+		case CPU_BMIPS5000:
+			write_c0_brcm_action(ACTION_BOOT_THREAD(cpu));
+			break;
+		}
+		cpumask_set_cpu(cpu, &bmips_booted_mask);
+	}
+
+	return 0;
+}
+
+/*
+ * Early setup - runs on secondary CPU after cache probe
+ */
+static void bmips_init_secondary(void)
+{
+	bmips_cpu_setup();
+
+	switch (current_cpu_type()) {
+	case CPU_BMIPS4350:
+	case CPU_BMIPS4380:
+		clear_c0_cause(smp_processor_id() ? C_SW1 : C_SW0);
+		break;
+	case CPU_BMIPS5000:
+		write_c0_brcm_action(ACTION_CLR_IPI(smp_processor_id(), 0));
+		cpu_set_core(&current_cpu_data, (read_c0_brcm_config() >> 25) & 3);
+		break;
+	}
+}
+
+/*
+ * Late setup - runs on secondary CPU before entering the idle loop
+ */
+static void bmips_smp_finish(void)
+{
+	pr_info("SMP: CPU%d is running\n", smp_processor_id());
+
+	/* make sure there won't be a timer interrupt for a little while */
+	write_c0_compare(read_c0_count() + mips_hpt_frequency / HZ);
+
+	irq_enable_hazard();
+	set_c0_status(IE_SW0 | IE_SW1 | bmips_tp1_irqs | IE_IRQ5 | ST0_IE);
+	irq_enable_hazard();
+}
+
+/*
+ * BMIPS5000 raceless IPIs
+ *
+ * Each CPU has two inbound SW IRQs which are independent of all other CPUs.
+ * IPI0 is used for SMP_RESCHEDULE_YOURSELF
+ * IPI1 is used for SMP_CALL_FUNCTION
+ */
+
+static void bmips5000_send_ipi_single(int cpu, unsigned int action)
+{
+	write_c0_brcm_action(ACTION_SET_IPI(cpu, action == SMP_CALL_FUNCTION));
+}
+
+static irqreturn_t bmips5000_ipi_interrupt(int irq, void *dev_id)
+{
+	int action = irq - IPI0_IRQ;
+
+	write_c0_brcm_action(ACTION_CLR_IPI(smp_processor_id(), action));
+
+	if (action == 0)
+		scheduler_ipi();
+	else
+		generic_smp_call_function_interrupt();
+
+	return IRQ_HANDLED;
+}
+
+static void bmips5000_send_ipi_mask(const struct cpumask *mask,
+	unsigned int action)
+{
+	unsigned int i;
+
+	for_each_cpu(i, mask)
+		bmips5000_send_ipi_single(i, action);
+}
+
+/*
+ * BMIPS43xx racey IPIs
+ *
+ * We use one inbound SW IRQ for each CPU.
+ *
+ * A spinlock must be held in order to keep CPUx from accidentally clearing
+ * an incoming IPI when it writes CP0 CAUSE to raise an IPI on CPUy.  The
+ * same spinlock is used to protect the action masks.
+ */
+
+static DEFINE_SPINLOCK(ipi_lock);
+static DEFINE_PER_CPU(int, ipi_action_mask);
+
+static void bmips43xx_send_ipi_single(int cpu, unsigned int action)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&ipi_lock, flags);
+	set_c0_cause(cpu ? C_SW1 : C_SW0);
+	per_cpu(ipi_action_mask, cpu) |= action;
+	irq_enable_hazard();
+	spin_unlock_irqrestore(&ipi_lock, flags);
+}
+
+static irqreturn_t bmips43xx_ipi_interrupt(int irq, void *dev_id)
+{
+	unsigned long flags;
+	int action, cpu = irq - IPI0_IRQ;
+
+	spin_lock_irqsave(&ipi_lock, flags);
+	action = __this_cpu_read(ipi_action_mask);
+	per_cpu(ipi_action_mask, cpu) = 0;
+	clear_c0_cause(cpu ? C_SW1 : C_SW0);
+	spin_unlock_irqrestore(&ipi_lock, flags);
+
+	if (action & SMP_RESCHEDULE_YOURSELF)
+		scheduler_ipi();
+	if (action & SMP_CALL_FUNCTION)
+		generic_smp_call_function_interrupt();
+
+	return IRQ_HANDLED;
+}
+
+static void bmips43xx_send_ipi_mask(const struct cpumask *mask,
+	unsigned int action)
+{
+	unsigned int i;
+
+	for_each_cpu(i, mask)
+		bmips43xx_send_ipi_single(i, action);
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+
+static int bmips_cpu_disable(void)
+{
+	unsigned int cpu = smp_processor_id();
+
+	if (cpu == 0)
+		return -EBUSY;
+
+	pr_info("SMP: CPU%d is offline\n", cpu);
+
+	set_cpu_online(cpu, false);
+	calculate_cpu_foreign_map();
+	irq_cpu_offline();
+	clear_c0_status(IE_IRQ5);
+
+	local_flush_tlb_all();
+	local_flush_icache_range(0, ~0);
+
+	return 0;
+}
+
+static void bmips_cpu_die(unsigned int cpu)
+{
+}
+
+void __ref play_dead(void)
+{
+	idle_task_exit();
+
+	/* flush data cache */
+	_dma_cache_wback_inv(0, ~0);
+
+	/*
+	 * Wakeup is on SW0 or SW1; disable everything else
+	 * Use BEV !IV (BMIPS_WARM_RESTART_VEC) to avoid the regular Linux
+	 * IRQ handlers; this clears ST0_IE and returns immediately.
+	 */
+	clear_c0_cause(CAUSEF_IV | C_SW0 | C_SW1);
+	change_c0_status(
+		IE_IRQ5 | bmips_tp1_irqs | IE_SW0 | IE_SW1 | ST0_IE | ST0_BEV,
+		IE_SW0 | IE_SW1 | ST0_IE | ST0_BEV);
+	irq_disable_hazard();
+
+	/*
+	 * wait for SW interrupt from bmips_boot_secondary(), then jump
+	 * back to start_secondary()
+	 */
+	__asm__ __volatile__(
+	"	wait\n"
+	"	j	bmips_secondary_reentry\n"
+	: : : "memory");
+}
+
+#endif /* CONFIG_HOTPLUG_CPU */
+
+const struct plat_smp_ops bmips43xx_smp_ops = {
+	.smp_setup		= bmips_smp_setup,
+	.prepare_cpus		= bmips_prepare_cpus,
+	.boot_secondary		= bmips_boot_secondary,
+	.smp_finish		= bmips_smp_finish,
+	.init_secondary		= bmips_init_secondary,
+	.send_ipi_single	= bmips43xx_send_ipi_single,
+	.send_ipi_mask		= bmips43xx_send_ipi_mask,
+#ifdef CONFIG_HOTPLUG_CPU
+	.cpu_disable		= bmips_cpu_disable,
+	.cpu_die		= bmips_cpu_die,
+#endif
+#ifdef CONFIG_KEXEC
+	.kexec_nonboot_cpu	= kexec_nonboot_cpu_jump,
+#endif
+};
+
+const struct plat_smp_ops bmips5000_smp_ops = {
+	.smp_setup		= bmips_smp_setup,
+	.prepare_cpus		= bmips_prepare_cpus,
+	.boot_secondary		= bmips_boot_secondary,
+	.smp_finish		= bmips_smp_finish,
+	.init_secondary		= bmips_init_secondary,
+	.send_ipi_single	= bmips5000_send_ipi_single,
+	.send_ipi_mask		= bmips5000_send_ipi_mask,
+#ifdef CONFIG_HOTPLUG_CPU
+	.cpu_disable		= bmips_cpu_disable,
+	.cpu_die		= bmips_cpu_die,
+#endif
+#ifdef CONFIG_KEXEC
+	.kexec_nonboot_cpu	= kexec_nonboot_cpu_jump,
+#endif
+};
+
+#endif /* CONFIG_SMP */
+
+/***********************************************************************
+ * BMIPS vector relocation
+ * This is primarily used for SMP boot, but it is applicable to some
+ * UP BMIPS systems as well.
+ ***********************************************************************/
+
+static void bmips_wr_vec(unsigned long dst, char *start, char *end)
+{
+	memcpy((void *)dst, start, end - start);
+	dma_cache_wback(dst, end - start);
+	local_flush_icache_range(dst, dst + (end - start));
+	instruction_hazard();
+}
+
+static inline void bmips_nmi_handler_setup(void)
+{
+	bmips_wr_vec(BMIPS_NMI_RESET_VEC, bmips_reset_nmi_vec,
+		bmips_reset_nmi_vec_end);
+	bmips_wr_vec(BMIPS_WARM_RESTART_VEC, bmips_smp_int_vec,
+		bmips_smp_int_vec_end);
+}
+
+struct reset_vec_info {
+	int cpu;
+	u32 val;
+};
+
+static void bmips_set_reset_vec_remote(void *vinfo)
+{
+	struct reset_vec_info *info = vinfo;
+	int shift = info->cpu & 0x01 ? 16 : 0;
+	u32 mask = ~(0xffff << shift), val = info->val >> 16;
+
+	preempt_disable();
+	if (smp_processor_id() > 0) {
+		smp_call_function_single(0, &bmips_set_reset_vec_remote,
+					 info, 1);
+	} else {
+		if (info->cpu & 0x02) {
+			/* BMIPS5200 "should" use mask/shift, but it's buggy */
+			bmips_write_zscm_reg(0xa0, (val << 16) | val);
+			bmips_read_zscm_reg(0xa0);
+		} else {
+			write_c0_brcm_bootvec((read_c0_brcm_bootvec() & mask) |
+					      (val << shift));
+		}
+	}
+	preempt_enable();
+}
+
+static void bmips_set_reset_vec(int cpu, u32 val)
+{
+	struct reset_vec_info info;
+
+	if (current_cpu_type() == CPU_BMIPS5000) {
+		/* this needs to run from CPU0 (which is always online) */
+		info.cpu = cpu;
+		info.val = val;
+		bmips_set_reset_vec_remote(&info);
+	} else {
+		void __iomem *cbr = BMIPS_GET_CBR();
+
+		if (cpu == 0)
+			__raw_writel(val, cbr + BMIPS_RELO_VECTOR_CONTROL_0);
+		else {
+			if (current_cpu_type() != CPU_BMIPS4380)
+				return;
+			__raw_writel(val, cbr + BMIPS_RELO_VECTOR_CONTROL_1);
+		}
+	}
+	__sync();
+	back_to_back_c0_hazard();
+}
+
+void bmips_ebase_setup(void)
+{
+	unsigned long new_ebase = ebase;
+
+	BUG_ON(ebase != CKSEG0);
+
+	switch (current_cpu_type()) {
+	case CPU_BMIPS4350:
+		/*
+		 * BMIPS4350 cannot relocate the normal vectors, but it
+		 * can relocate the BEV=1 vectors.  So CPU1 starts up at
+		 * the relocated BEV=1, IV=0 general exception vector @
+		 * 0xa000_0380.
+		 *
+		 * set_uncached_handler() is used here because:
+		 *  - CPU1 will run this from uncached space
+		 *  - None of the cacheflush functions are set up yet
+		 */
+		set_uncached_handler(BMIPS_WARM_RESTART_VEC - CKSEG0,
+			&bmips_smp_int_vec, 0x80);
+		__sync();
+		return;
+	case CPU_BMIPS3300:
+	case CPU_BMIPS4380:
+		/*
+		 * 0x8000_0000: reset/NMI (initially in kseg1)
+		 * 0x8000_0400: normal vectors
+		 */
+		new_ebase = 0x80000400;
+		bmips_set_reset_vec(0, RESET_FROM_KSEG0);
+		break;
+	case CPU_BMIPS5000:
+		/*
+		 * 0x8000_0000: reset/NMI (initially in kseg1)
+		 * 0x8000_1000: normal vectors
+		 */
+		new_ebase = 0x80001000;
+		bmips_set_reset_vec(0, RESET_FROM_KSEG0);
+		write_c0_ebase(new_ebase);
+		break;
+	default:
+		return;
+	}
+
+	board_nmi_handler_setup = &bmips_nmi_handler_setup;
+	ebase = new_ebase;
+}
+
+asmlinkage void __weak plat_wired_tlb_setup(void)
+{
+	/*
+	 * Called when starting/restarting a secondary CPU.
+	 * Kernel stacks and other important data might only be accessible
+	 * once the wired entries are present.
+	 */
+}
+
+void bmips_cpu_setup(void)
+{
+	void __iomem __maybe_unused *cbr = BMIPS_GET_CBR();
+	u32 __maybe_unused cfg;
+
+	switch (current_cpu_type()) {
+	case CPU_BMIPS3300:
+		/* Set BIU to async mode */
+		set_c0_brcm_bus_pll(BIT(22));
+		__sync();
+
+		/* put the BIU back in sync mode */
+		clear_c0_brcm_bus_pll(BIT(22));
+
+		/* clear BHTD to enable branch history table */
+		clear_c0_brcm_reset(BIT(16));
+
+		/* Flush and enable RAC */
+		cfg = __raw_readl(cbr + BMIPS_RAC_CONFIG);
+		__raw_writel(cfg | 0x100, cbr + BMIPS_RAC_CONFIG);
+		__raw_readl(cbr + BMIPS_RAC_CONFIG);
+
+		cfg = __raw_readl(cbr + BMIPS_RAC_CONFIG);
+		__raw_writel(cfg | 0xf, cbr + BMIPS_RAC_CONFIG);
+		__raw_readl(cbr + BMIPS_RAC_CONFIG);
+
+		cfg = __raw_readl(cbr + BMIPS_RAC_ADDRESS_RANGE);
+		__raw_writel(cfg | 0x0fff0000, cbr + BMIPS_RAC_ADDRESS_RANGE);
+		__raw_readl(cbr + BMIPS_RAC_ADDRESS_RANGE);
+		break;
+
+	case CPU_BMIPS4380:
+		/* CBG workaround for early BMIPS4380 CPUs */
+		switch (read_c0_prid()) {
+		case 0x2a040:
+		case 0x2a042:
+		case 0x2a044:
+		case 0x2a060:
+			cfg = __raw_readl(cbr + BMIPS_L2_CONFIG);
+			__raw_writel(cfg & ~0x07000000, cbr + BMIPS_L2_CONFIG);
+			__raw_readl(cbr + BMIPS_L2_CONFIG);
+		}
+
+		/* clear BHTD to enable branch history table */
+		clear_c0_brcm_config_0(BIT(21));
+
+		/* XI/ROTR enable */
+		set_c0_brcm_config_0(BIT(23));
+		set_c0_brcm_cmt_ctrl(BIT(15));
+		break;
+
+	case CPU_BMIPS5000:
+		/* enable RDHWR, BRDHWR */
+		set_c0_brcm_config(BIT(17) | BIT(21));
+
+		/* Disable JTB */
+		__asm__ __volatile__(
+		"	.set	noreorder\n"
+		"	li	$8, 0x5a455048\n"
+		"	.word	0x4088b00f\n"	/* mtc0	t0, $22, 15 */
+		"	.word	0x4008b008\n"	/* mfc0	t0, $22, 8 */
+		"	li	$9, 0x00008000\n"
+		"	or	$8, $8, $9\n"
+		"	.word	0x4088b008\n"	/* mtc0	t0, $22, 8 */
+		"	sync\n"
+		"	li	$8, 0x0\n"
+		"	.word	0x4088b00f\n"	/* mtc0	t0, $22, 15 */
+		"	.set	reorder\n"
+		: : : "$8", "$9");
+
+		/* XI enable */
+		set_c0_brcm_config(BIT(27));
+
+		/* enable MIPS32R2 ROR instruction for XI TLB handlers */
+		__asm__ __volatile__(
+		"	li	$8, 0x5a455048\n"
+		"	.word	0x4088b00f\n"	/* mtc0 $8, $22, 15 */
+		"	nop; nop; nop\n"
+		"	.word	0x4008b008\n"	/* mfc0 $8, $22, 8 */
+		"	lui	$9, 0x0100\n"
+		"	or	$8, $9\n"
+		"	.word	0x4088b008\n"	/* mtc0 $8, $22, 8 */
+		: : : "$8", "$9");
+		break;
+	}
+}
diff --git a/arch/mips/kernel/smp-cmp.c b/arch/mips/kernel/smp-cmp.c
new file mode 100644
index 000000000..76f5824cd
--- /dev/null
+++ b/arch/mips/kernel/smp-cmp.c
@@ -0,0 +1,148 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *
+ * Copyright (C) 2007 MIPS Technologies, Inc.
+ *    Chris Dearman (chris@mips.com)
+ */
+
+#undef DEBUG
+
+#include <linux/kernel.h>
+#include <linux/sched/task_stack.h>
+#include <linux/smp.h>
+#include <linux/cpumask.h>
+#include <linux/interrupt.h>
+#include <linux/compiler.h>
+
+#include <linux/atomic.h>
+#include <asm/cacheflush.h>
+#include <asm/cpu.h>
+#include <asm/processor.h>
+#include <asm/hardirq.h>
+#include <asm/mmu_context.h>
+#include <asm/smp.h>
+#include <asm/time.h>
+#include <asm/mipsregs.h>
+#include <asm/mipsmtregs.h>
+#include <asm/mips_mt.h>
+#include <asm/amon.h>
+
+static void cmp_init_secondary(void)
+{
+	struct cpuinfo_mips *c __maybe_unused = &current_cpu_data;
+
+	/* Assume GIC is present */
+	change_c0_status(ST0_IM, STATUSF_IP2 | STATUSF_IP3 | STATUSF_IP4 |
+				 STATUSF_IP5 | STATUSF_IP6 | STATUSF_IP7);
+
+	/* Enable per-cpu interrupts: platform specific */
+
+#ifdef CONFIG_MIPS_MT_SMP
+	if (cpu_has_mipsmt)
+		cpu_set_vpe_id(c, (read_c0_tcbind() >> TCBIND_CURVPE_SHIFT) &
+				  TCBIND_CURVPE);
+#endif
+}
+
+static void cmp_smp_finish(void)
+{
+	pr_debug("SMPCMP: CPU%d: %s\n", smp_processor_id(), __func__);
+
+	/* CDFIXME: remove this? */
+	write_c0_compare(read_c0_count() + (8 * mips_hpt_frequency / HZ));
+
+#ifdef CONFIG_MIPS_MT_FPAFF
+	/* If we have an FPU, enroll ourselves in the FPU-full mask */
+	if (cpu_has_fpu)
+		cpumask_set_cpu(smp_processor_id(), &mt_fpu_cpumask);
+#endif /* CONFIG_MIPS_MT_FPAFF */
+
+	local_irq_enable();
+}
+
+/*
+ * Setup the PC, SP, and GP of a secondary processor and start it running
+ * smp_bootstrap is the place to resume from
+ * __KSTK_TOS(idle) is apparently the stack pointer
+ * (unsigned long)idle->thread_info the gp
+ */
+static int cmp_boot_secondary(int cpu, struct task_struct *idle)
+{
+	struct thread_info *gp = task_thread_info(idle);
+	unsigned long sp = __KSTK_TOS(idle);
+	unsigned long pc = (unsigned long)&smp_bootstrap;
+	unsigned long a0 = 0;
+
+	pr_debug("SMPCMP: CPU%d: %s cpu %d\n", smp_processor_id(),
+		__func__, cpu);
+
+#if 0
+	/* Needed? */
+	flush_icache_range((unsigned long)gp,
+			   (unsigned long)(gp + sizeof(struct thread_info)));
+#endif
+
+	amon_cpu_start(cpu, pc, sp, (unsigned long)gp, a0);
+	return 0;
+}
+
+/*
+ * Common setup before any secondaries are started
+ */
+void __init cmp_smp_setup(void)
+{
+	int i;
+	int ncpu = 0;
+
+	pr_debug("SMPCMP: CPU%d: %s\n", smp_processor_id(), __func__);
+
+#ifdef CONFIG_MIPS_MT_FPAFF
+	/* If we have an FPU, enroll ourselves in the FPU-full mask */
+	if (cpu_has_fpu)
+		cpumask_set_cpu(0, &mt_fpu_cpumask);
+#endif /* CONFIG_MIPS_MT_FPAFF */
+
+	for (i = 1; i < NR_CPUS; i++) {
+		if (amon_cpu_avail(i)) {
+			set_cpu_possible(i, true);
+			__cpu_number_map[i]	= ++ncpu;
+			__cpu_logical_map[ncpu] = i;
+		}
+	}
+
+	if (cpu_has_mipsmt) {
+		unsigned int nvpe = 1;
+#ifdef CONFIG_MIPS_MT_SMP
+		unsigned int mvpconf0 = read_c0_mvpconf0();
+
+		nvpe = ((mvpconf0 & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT) + 1;
+#endif
+		smp_num_siblings = nvpe;
+	}
+	pr_info("Detected %i available secondary CPU(s)\n", ncpu);
+}
+
+void __init cmp_prepare_cpus(unsigned int max_cpus)
+{
+	pr_debug("SMPCMP: CPU%d: %s max_cpus=%d\n",
+		 smp_processor_id(), __func__, max_cpus);
+
+#ifdef CONFIG_MIPS_MT
+	/*
+	 * FIXME: some of these options are per-system, some per-core and
+	 * some per-cpu
+	 */
+	mips_mt_set_cpuoptions();
+#endif
+
+}
+
+const struct plat_smp_ops cmp_smp_ops = {
+	.send_ipi_single	= mips_smp_send_ipi_single,
+	.send_ipi_mask		= mips_smp_send_ipi_mask,
+	.init_secondary		= cmp_init_secondary,
+	.smp_finish		= cmp_smp_finish,
+	.boot_secondary		= cmp_boot_secondary,
+	.smp_setup		= cmp_smp_setup,
+	.prepare_cpus		= cmp_prepare_cpus,
+};
diff --git a/arch/mips/kernel/smp-cps.c b/arch/mips/kernel/smp-cps.c
new file mode 100644
index 000000000..f659adb68
--- /dev/null
+++ b/arch/mips/kernel/smp-cps.c
@@ -0,0 +1,647 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2013 Imagination Technologies
+ * Author: Paul Burton <paul.burton@mips.com>
+ */
+
+#include <linux/cpu.h>
+#include <linux/delay.h>
+#include <linux/io.h>
+#include <linux/sched/task_stack.h>
+#include <linux/sched/hotplug.h>
+#include <linux/slab.h>
+#include <linux/smp.h>
+#include <linux/types.h>
+
+#include <asm/bcache.h>
+#include <asm/mips-cps.h>
+#include <asm/mips_mt.h>
+#include <asm/mipsregs.h>
+#include <asm/pm-cps.h>
+#include <asm/r4kcache.h>
+#include <asm/smp-cps.h>
+#include <asm/time.h>
+#include <asm/uasm.h>
+
+static bool threads_disabled;
+static DECLARE_BITMAP(core_power, NR_CPUS);
+
+struct core_boot_config *mips_cps_core_bootcfg;
+
+static int __init setup_nothreads(char *s)
+{
+	threads_disabled = true;
+	return 0;
+}
+early_param("nothreads", setup_nothreads);
+
+static unsigned core_vpe_count(unsigned int cluster, unsigned core)
+{
+	if (threads_disabled)
+		return 1;
+
+	return mips_cps_numvps(cluster, core);
+}
+
+static void __init cps_smp_setup(void)
+{
+	unsigned int nclusters, ncores, nvpes, core_vpes;
+	unsigned long core_entry;
+	int cl, c, v;
+
+	/* Detect & record VPE topology */
+	nvpes = 0;
+	nclusters = mips_cps_numclusters();
+	pr_info("%s topology ", cpu_has_mips_r6 ? "VP" : "VPE");
+	for (cl = 0; cl < nclusters; cl++) {
+		if (cl > 0)
+			pr_cont(",");
+		pr_cont("{");
+
+		ncores = mips_cps_numcores(cl);
+		for (c = 0; c < ncores; c++) {
+			core_vpes = core_vpe_count(cl, c);
+
+			if (c > 0)
+				pr_cont(",");
+			pr_cont("%u", core_vpes);
+
+			/* Use the number of VPEs in cluster 0 core 0 for smp_num_siblings */
+			if (!cl && !c)
+				smp_num_siblings = core_vpes;
+
+			for (v = 0; v < min_t(int, core_vpes, NR_CPUS - nvpes); v++) {
+				cpu_set_cluster(&cpu_data[nvpes + v], cl);
+				cpu_set_core(&cpu_data[nvpes + v], c);
+				cpu_set_vpe_id(&cpu_data[nvpes + v], v);
+			}
+
+			nvpes += core_vpes;
+		}
+
+		pr_cont("}");
+	}
+	pr_cont(" total %u\n", nvpes);
+
+	/* Indicate present CPUs (CPU being synonymous with VPE) */
+	for (v = 0; v < min_t(unsigned, nvpes, NR_CPUS); v++) {
+		set_cpu_possible(v, cpu_cluster(&cpu_data[v]) == 0);
+		set_cpu_present(v, cpu_cluster(&cpu_data[v]) == 0);
+		__cpu_number_map[v] = v;
+		__cpu_logical_map[v] = v;
+	}
+
+	/* Set a coherent default CCA (CWB) */
+	change_c0_config(CONF_CM_CMASK, 0x5);
+
+	/* Core 0 is powered up (we're running on it) */
+	bitmap_set(core_power, 0, 1);
+
+	/* Initialise core 0 */
+	mips_cps_core_init();
+
+	/* Make core 0 coherent with everything */
+	write_gcr_cl_coherence(0xff);
+
+	if (mips_cm_revision() >= CM_REV_CM3) {
+		core_entry = CKSEG1ADDR((unsigned long)mips_cps_core_entry);
+		write_gcr_bev_base(core_entry);
+	}
+
+#ifdef CONFIG_MIPS_MT_FPAFF
+	/* If we have an FPU, enroll ourselves in the FPU-full mask */
+	if (cpu_has_fpu)
+		cpumask_set_cpu(0, &mt_fpu_cpumask);
+#endif /* CONFIG_MIPS_MT_FPAFF */
+}
+
+static void __init cps_prepare_cpus(unsigned int max_cpus)
+{
+	unsigned ncores, core_vpes, c, cca;
+	bool cca_unsuitable, cores_limited;
+	u32 *entry_code;
+
+	mips_mt_set_cpuoptions();
+
+	/* Detect whether the CCA is unsuited to multi-core SMP */
+	cca = read_c0_config() & CONF_CM_CMASK;
+	switch (cca) {
+	case 0x4: /* CWBE */
+	case 0x5: /* CWB */
+		/* The CCA is coherent, multi-core is fine */
+		cca_unsuitable = false;
+		break;
+
+	default:
+		/* CCA is not coherent, multi-core is not usable */
+		cca_unsuitable = true;
+	}
+
+	/* Warn the user if the CCA prevents multi-core */
+	cores_limited = false;
+	if (cca_unsuitable || cpu_has_dc_aliases) {
+		for_each_present_cpu(c) {
+			if (cpus_are_siblings(smp_processor_id(), c))
+				continue;
+
+			set_cpu_present(c, false);
+			cores_limited = true;
+		}
+	}
+	if (cores_limited)
+		pr_warn("Using only one core due to %s%s%s\n",
+			cca_unsuitable ? "unsuitable CCA" : "",
+			(cca_unsuitable && cpu_has_dc_aliases) ? " & " : "",
+			cpu_has_dc_aliases ? "dcache aliasing" : "");
+
+	/*
+	 * Patch the start of mips_cps_core_entry to provide:
+	 *
+	 * s0 = kseg0 CCA
+	 */
+	entry_code = (u32 *)&mips_cps_core_entry;
+	uasm_i_addiu(&entry_code, 16, 0, cca);
+	blast_dcache_range((unsigned long)&mips_cps_core_entry,
+			   (unsigned long)entry_code);
+	bc_wback_inv((unsigned long)&mips_cps_core_entry,
+		     (void *)entry_code - (void *)&mips_cps_core_entry);
+	__sync();
+
+	/* Allocate core boot configuration structs */
+	ncores = mips_cps_numcores(0);
+	mips_cps_core_bootcfg = kcalloc(ncores, sizeof(*mips_cps_core_bootcfg),
+					GFP_KERNEL);
+	if (!mips_cps_core_bootcfg) {
+		pr_err("Failed to allocate boot config for %u cores\n", ncores);
+		goto err_out;
+	}
+
+	/* Allocate VPE boot configuration structs */
+	for (c = 0; c < ncores; c++) {
+		core_vpes = core_vpe_count(0, c);
+		mips_cps_core_bootcfg[c].vpe_config = kcalloc(core_vpes,
+				sizeof(*mips_cps_core_bootcfg[c].vpe_config),
+				GFP_KERNEL);
+		if (!mips_cps_core_bootcfg[c].vpe_config) {
+			pr_err("Failed to allocate %u VPE boot configs\n",
+			       core_vpes);
+			goto err_out;
+		}
+	}
+
+	/* Mark this CPU as booted */
+	atomic_set(&mips_cps_core_bootcfg[cpu_core(&current_cpu_data)].vpe_mask,
+		   1 << cpu_vpe_id(&current_cpu_data));
+
+	return;
+err_out:
+	/* Clean up allocations */
+	if (mips_cps_core_bootcfg) {
+		for (c = 0; c < ncores; c++)
+			kfree(mips_cps_core_bootcfg[c].vpe_config);
+		kfree(mips_cps_core_bootcfg);
+		mips_cps_core_bootcfg = NULL;
+	}
+
+	/* Effectively disable SMP by declaring CPUs not present */
+	for_each_possible_cpu(c) {
+		if (c == 0)
+			continue;
+		set_cpu_present(c, false);
+	}
+}
+
+static void boot_core(unsigned int core, unsigned int vpe_id)
+{
+	u32 stat, seq_state;
+	unsigned timeout;
+
+	/* Select the appropriate core */
+	mips_cm_lock_other(0, core, 0, CM_GCR_Cx_OTHER_BLOCK_LOCAL);
+
+	/* Set its reset vector */
+	write_gcr_co_reset_base(CKSEG1ADDR((unsigned long)mips_cps_core_entry));
+
+	/* Ensure its coherency is disabled */
+	write_gcr_co_coherence(0);
+
+	/* Start it with the legacy memory map and exception base */
+	write_gcr_co_reset_ext_base(CM_GCR_Cx_RESET_EXT_BASE_UEB);
+
+	/* Ensure the core can access the GCRs */
+	set_gcr_access(1 << core);
+
+	if (mips_cpc_present()) {
+		/* Reset the core */
+		mips_cpc_lock_other(core);
+
+		if (mips_cm_revision() >= CM_REV_CM3) {
+			/* Run only the requested VP following the reset */
+			write_cpc_co_vp_stop(0xf);
+			write_cpc_co_vp_run(1 << vpe_id);
+
+			/*
+			 * Ensure that the VP_RUN register is written before the
+			 * core leaves reset.
+			 */
+			wmb();
+		}
+
+		write_cpc_co_cmd(CPC_Cx_CMD_RESET);
+
+		timeout = 100;
+		while (true) {
+			stat = read_cpc_co_stat_conf();
+			seq_state = stat & CPC_Cx_STAT_CONF_SEQSTATE;
+			seq_state >>= __ffs(CPC_Cx_STAT_CONF_SEQSTATE);
+
+			/* U6 == coherent execution, ie. the core is up */
+			if (seq_state == CPC_Cx_STAT_CONF_SEQSTATE_U6)
+				break;
+
+			/* Delay a little while before we start warning */
+			if (timeout) {
+				timeout--;
+				mdelay(10);
+				continue;
+			}
+
+			pr_warn("Waiting for core %u to start... STAT_CONF=0x%x\n",
+				core, stat);
+			mdelay(1000);
+		}
+
+		mips_cpc_unlock_other();
+	} else {
+		/* Take the core out of reset */
+		write_gcr_co_reset_release(0);
+	}
+
+	mips_cm_unlock_other();
+
+	/* The core is now powered up */
+	bitmap_set(core_power, core, 1);
+}
+
+static void remote_vpe_boot(void *dummy)
+{
+	unsigned core = cpu_core(&current_cpu_data);
+	struct core_boot_config *core_cfg = &mips_cps_core_bootcfg[core];
+
+	mips_cps_boot_vpes(core_cfg, cpu_vpe_id(&current_cpu_data));
+}
+
+static int cps_boot_secondary(int cpu, struct task_struct *idle)
+{
+	unsigned core = cpu_core(&cpu_data[cpu]);
+	unsigned vpe_id = cpu_vpe_id(&cpu_data[cpu]);
+	struct core_boot_config *core_cfg = &mips_cps_core_bootcfg[core];
+	struct vpe_boot_config *vpe_cfg = &core_cfg->vpe_config[vpe_id];
+	unsigned long core_entry;
+	unsigned int remote;
+	int err;
+
+	/* We don't yet support booting CPUs in other clusters */
+	if (cpu_cluster(&cpu_data[cpu]) != cpu_cluster(&raw_current_cpu_data))
+		return -ENOSYS;
+
+	vpe_cfg->pc = (unsigned long)&smp_bootstrap;
+	vpe_cfg->sp = __KSTK_TOS(idle);
+	vpe_cfg->gp = (unsigned long)task_thread_info(idle);
+
+	atomic_or(1 << cpu_vpe_id(&cpu_data[cpu]), &core_cfg->vpe_mask);
+
+	preempt_disable();
+
+	if (!test_bit(core, core_power)) {
+		/* Boot a VPE on a powered down core */
+		boot_core(core, vpe_id);
+		goto out;
+	}
+
+	if (cpu_has_vp) {
+		mips_cm_lock_other(0, core, vpe_id, CM_GCR_Cx_OTHER_BLOCK_LOCAL);
+		core_entry = CKSEG1ADDR((unsigned long)mips_cps_core_entry);
+		write_gcr_co_reset_base(core_entry);
+		mips_cm_unlock_other();
+	}
+
+	if (!cpus_are_siblings(cpu, smp_processor_id())) {
+		/* Boot a VPE on another powered up core */
+		for (remote = 0; remote < NR_CPUS; remote++) {
+			if (!cpus_are_siblings(cpu, remote))
+				continue;
+			if (cpu_online(remote))
+				break;
+		}
+		if (remote >= NR_CPUS) {
+			pr_crit("No online CPU in core %u to start CPU%d\n",
+				core, cpu);
+			goto out;
+		}
+
+		err = smp_call_function_single(remote, remote_vpe_boot,
+					       NULL, 1);
+		if (err)
+			panic("Failed to call remote CPU\n");
+		goto out;
+	}
+
+	BUG_ON(!cpu_has_mipsmt && !cpu_has_vp);
+
+	/* Boot a VPE on this core */
+	mips_cps_boot_vpes(core_cfg, vpe_id);
+out:
+	preempt_enable();
+	return 0;
+}
+
+static void cps_init_secondary(void)
+{
+	/* Disable MT - we only want to run 1 TC per VPE */
+	if (cpu_has_mipsmt)
+		dmt();
+
+	if (mips_cm_revision() >= CM_REV_CM3) {
+		unsigned int ident = read_gic_vl_ident();
+
+		/*
+		 * Ensure that our calculation of the VP ID matches up with
+		 * what the GIC reports, otherwise we'll have configured
+		 * interrupts incorrectly.
+		 */
+		BUG_ON(ident != mips_cm_vp_id(smp_processor_id()));
+	}
+
+	if (cpu_has_veic)
+		clear_c0_status(ST0_IM);
+	else
+		change_c0_status(ST0_IM, STATUSF_IP2 | STATUSF_IP3 |
+					 STATUSF_IP4 | STATUSF_IP5 |
+					 STATUSF_IP6 | STATUSF_IP7);
+}
+
+static void cps_smp_finish(void)
+{
+	write_c0_compare(read_c0_count() + (8 * mips_hpt_frequency / HZ));
+
+#ifdef CONFIG_MIPS_MT_FPAFF
+	/* If we have an FPU, enroll ourselves in the FPU-full mask */
+	if (cpu_has_fpu)
+		cpumask_set_cpu(smp_processor_id(), &mt_fpu_cpumask);
+#endif /* CONFIG_MIPS_MT_FPAFF */
+
+	local_irq_enable();
+}
+
+#if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_KEXEC)
+
+enum cpu_death {
+	CPU_DEATH_HALT,
+	CPU_DEATH_POWER,
+};
+
+static void cps_shutdown_this_cpu(enum cpu_death death)
+{
+	unsigned int cpu, core, vpe_id;
+
+	cpu = smp_processor_id();
+	core = cpu_core(&cpu_data[cpu]);
+
+	if (death == CPU_DEATH_HALT) {
+		vpe_id = cpu_vpe_id(&cpu_data[cpu]);
+
+		pr_debug("Halting core %d VP%d\n", core, vpe_id);
+		if (cpu_has_mipsmt) {
+			/* Halt this TC */
+			write_c0_tchalt(TCHALT_H);
+			instruction_hazard();
+		} else if (cpu_has_vp) {
+			write_cpc_cl_vp_stop(1 << vpe_id);
+
+			/* Ensure that the VP_STOP register is written */
+			wmb();
+		}
+	} else {
+		if (IS_ENABLED(CONFIG_HOTPLUG_CPU)) {
+			pr_debug("Gating power to core %d\n", core);
+			/* Power down the core */
+			cps_pm_enter_state(CPS_PM_POWER_GATED);
+		}
+	}
+}
+
+#ifdef CONFIG_KEXEC
+
+static void cps_kexec_nonboot_cpu(void)
+{
+	if (cpu_has_mipsmt || cpu_has_vp)
+		cps_shutdown_this_cpu(CPU_DEATH_HALT);
+	else
+		cps_shutdown_this_cpu(CPU_DEATH_POWER);
+}
+
+#endif /* CONFIG_KEXEC */
+
+#endif /* CONFIG_HOTPLUG_CPU || CONFIG_KEXEC */
+
+#ifdef CONFIG_HOTPLUG_CPU
+
+static int cps_cpu_disable(void)
+{
+	unsigned cpu = smp_processor_id();
+	struct core_boot_config *core_cfg;
+
+	if (!cpu)
+		return -EBUSY;
+
+	if (!cps_pm_support_state(CPS_PM_POWER_GATED))
+		return -EINVAL;
+
+	core_cfg = &mips_cps_core_bootcfg[cpu_core(&current_cpu_data)];
+	atomic_sub(1 << cpu_vpe_id(&current_cpu_data), &core_cfg->vpe_mask);
+	smp_mb__after_atomic();
+	set_cpu_online(cpu, false);
+	calculate_cpu_foreign_map();
+
+	return 0;
+}
+
+static unsigned cpu_death_sibling;
+static enum cpu_death cpu_death;
+
+void play_dead(void)
+{
+	unsigned int cpu;
+
+	local_irq_disable();
+	idle_task_exit();
+	cpu = smp_processor_id();
+	cpu_death = CPU_DEATH_POWER;
+
+	pr_debug("CPU%d going offline\n", cpu);
+
+	if (cpu_has_mipsmt || cpu_has_vp) {
+		/* Look for another online VPE within the core */
+		for_each_online_cpu(cpu_death_sibling) {
+			if (!cpus_are_siblings(cpu, cpu_death_sibling))
+				continue;
+
+			/*
+			 * There is an online VPE within the core. Just halt
+			 * this TC and leave the core alone.
+			 */
+			cpu_death = CPU_DEATH_HALT;
+			break;
+		}
+	}
+
+	/* This CPU has chosen its way out */
+	(void)cpu_report_death();
+
+	cps_shutdown_this_cpu(cpu_death);
+
+	/* This should never be reached */
+	panic("Failed to offline CPU %u", cpu);
+}
+
+static void wait_for_sibling_halt(void *ptr_cpu)
+{
+	unsigned cpu = (unsigned long)ptr_cpu;
+	unsigned vpe_id = cpu_vpe_id(&cpu_data[cpu]);
+	unsigned halted;
+	unsigned long flags;
+
+	do {
+		local_irq_save(flags);
+		settc(vpe_id);
+		halted = read_tc_c0_tchalt();
+		local_irq_restore(flags);
+	} while (!(halted & TCHALT_H));
+}
+
+static void cps_cpu_die(unsigned int cpu)
+{
+	unsigned core = cpu_core(&cpu_data[cpu]);
+	unsigned int vpe_id = cpu_vpe_id(&cpu_data[cpu]);
+	ktime_t fail_time;
+	unsigned stat;
+	int err;
+
+	/* Wait for the cpu to choose its way out */
+	if (!cpu_wait_death(cpu, 5)) {
+		pr_err("CPU%u: didn't offline\n", cpu);
+		return;
+	}
+
+	/*
+	 * Now wait for the CPU to actually offline. Without doing this that
+	 * offlining may race with one or more of:
+	 *
+	 *   - Onlining the CPU again.
+	 *   - Powering down the core if another VPE within it is offlined.
+	 *   - A sibling VPE entering a non-coherent state.
+	 *
+	 * In the non-MT halt case (ie. infinite loop) the CPU is doing nothing
+	 * with which we could race, so do nothing.
+	 */
+	if (cpu_death == CPU_DEATH_POWER) {
+		/*
+		 * Wait for the core to enter a powered down or clock gated
+		 * state, the latter happening when a JTAG probe is connected
+		 * in which case the CPC will refuse to power down the core.
+		 */
+		fail_time = ktime_add_ms(ktime_get(), 2000);
+		do {
+			mips_cm_lock_other(0, core, 0, CM_GCR_Cx_OTHER_BLOCK_LOCAL);
+			mips_cpc_lock_other(core);
+			stat = read_cpc_co_stat_conf();
+			stat &= CPC_Cx_STAT_CONF_SEQSTATE;
+			stat >>= __ffs(CPC_Cx_STAT_CONF_SEQSTATE);
+			mips_cpc_unlock_other();
+			mips_cm_unlock_other();
+
+			if (stat == CPC_Cx_STAT_CONF_SEQSTATE_D0 ||
+			    stat == CPC_Cx_STAT_CONF_SEQSTATE_D2 ||
+			    stat == CPC_Cx_STAT_CONF_SEQSTATE_U2)
+				break;
+
+			/*
+			 * The core ought to have powered down, but didn't &
+			 * now we don't really know what state it's in. It's
+			 * likely that its _pwr_up pin has been wired to logic
+			 * 1 & it powered back up as soon as we powered it
+			 * down...
+			 *
+			 * The best we can do is warn the user & continue in
+			 * the hope that the core is doing nothing harmful &
+			 * might behave properly if we online it later.
+			 */
+			if (WARN(ktime_after(ktime_get(), fail_time),
+				 "CPU%u hasn't powered down, seq. state %u\n",
+				 cpu, stat))
+				break;
+		} while (1);
+
+		/* Indicate the core is powered off */
+		bitmap_clear(core_power, core, 1);
+	} else if (cpu_has_mipsmt) {
+		/*
+		 * Have a CPU with access to the offlined CPUs registers wait
+		 * for its TC to halt.
+		 */
+		err = smp_call_function_single(cpu_death_sibling,
+					       wait_for_sibling_halt,
+					       (void *)(unsigned long)cpu, 1);
+		if (err)
+			panic("Failed to call remote sibling CPU\n");
+	} else if (cpu_has_vp) {
+		do {
+			mips_cm_lock_other(0, core, vpe_id, CM_GCR_Cx_OTHER_BLOCK_LOCAL);
+			stat = read_cpc_co_vp_running();
+			mips_cm_unlock_other();
+		} while (stat & (1 << vpe_id));
+	}
+}
+
+#endif /* CONFIG_HOTPLUG_CPU */
+
+static const struct plat_smp_ops cps_smp_ops = {
+	.smp_setup		= cps_smp_setup,
+	.prepare_cpus		= cps_prepare_cpus,
+	.boot_secondary		= cps_boot_secondary,
+	.init_secondary		= cps_init_secondary,
+	.smp_finish		= cps_smp_finish,
+	.send_ipi_single	= mips_smp_send_ipi_single,
+	.send_ipi_mask		= mips_smp_send_ipi_mask,
+#ifdef CONFIG_HOTPLUG_CPU
+	.cpu_disable		= cps_cpu_disable,
+	.cpu_die		= cps_cpu_die,
+#endif
+#ifdef CONFIG_KEXEC
+	.kexec_nonboot_cpu	= cps_kexec_nonboot_cpu,
+#endif
+};
+
+bool mips_cps_smp_in_use(void)
+{
+	extern const struct plat_smp_ops *mp_ops;
+	return mp_ops == &cps_smp_ops;
+}
+
+int register_cps_smp_ops(void)
+{
+	if (!mips_cm_present()) {
+		pr_warn("MIPS CPS SMP unable to proceed without a CM\n");
+		return -ENODEV;
+	}
+
+	/* check we have a GIC - we need one for IPIs */
+	if (!(read_gcr_gic_status() & CM_GCR_GIC_STATUS_EX)) {
+		pr_warn("MIPS CPS SMP unable to proceed without a GIC\n");
+		return -ENODEV;
+	}
+
+	register_smp_ops(&cps_smp_ops);
+	return 0;
+}
diff --git a/arch/mips/kernel/smp-mt.c b/arch/mips/kernel/smp-mt.c
new file mode 100644
index 000000000..5f04a0141
--- /dev/null
+++ b/arch/mips/kernel/smp-mt.c
@@ -0,0 +1,240 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *
+ * Copyright (C) 2004, 05, 06 MIPS Technologies, Inc.
+ *    Elizabeth Clarke (beth@mips.com)
+ *    Ralf Baechle (ralf@linux-mips.org)
+ * Copyright (C) 2006 Ralf Baechle (ralf@linux-mips.org)
+ */
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/cpumask.h>
+#include <linux/interrupt.h>
+#include <linux/compiler.h>
+#include <linux/sched/task_stack.h>
+#include <linux/smp.h>
+
+#include <linux/atomic.h>
+#include <asm/cacheflush.h>
+#include <asm/cpu.h>
+#include <asm/processor.h>
+#include <asm/hardirq.h>
+#include <asm/mmu_context.h>
+#include <asm/time.h>
+#include <asm/mipsregs.h>
+#include <asm/mipsmtregs.h>
+#include <asm/mips_mt.h>
+#include <asm/mips-cps.h>
+
+static void __init smvp_copy_vpe_config(void)
+{
+	write_vpe_c0_status(
+		(read_c0_status() & ~(ST0_IM | ST0_IE | ST0_KSU)) | ST0_CU0);
+
+	/* set config to be the same as vpe0, particularly kseg0 coherency alg */
+	write_vpe_c0_config( read_c0_config());
+
+	/* make sure there are no software interrupts pending */
+	write_vpe_c0_cause(0);
+
+	/* Propagate Config7 */
+	write_vpe_c0_config7(read_c0_config7());
+
+	write_vpe_c0_count(read_c0_count());
+}
+
+static unsigned int __init smvp_vpe_init(unsigned int tc, unsigned int mvpconf0,
+	unsigned int ncpu)
+{
+	if (tc > ((mvpconf0 & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT))
+		return ncpu;
+
+	/* Deactivate all but VPE 0 */
+	if (tc != 0) {
+		unsigned long tmp = read_vpe_c0_vpeconf0();
+
+		tmp &= ~VPECONF0_VPA;
+
+		/* master VPE */
+		tmp |= VPECONF0_MVP;
+		write_vpe_c0_vpeconf0(tmp);
+
+		/* Record this as available CPU */
+		set_cpu_possible(tc, true);
+		set_cpu_present(tc, true);
+		__cpu_number_map[tc]	= ++ncpu;
+		__cpu_logical_map[ncpu] = tc;
+	}
+
+	/* Disable multi-threading with TC's */
+	write_vpe_c0_vpecontrol(read_vpe_c0_vpecontrol() & ~VPECONTROL_TE);
+
+	if (tc != 0)
+		smvp_copy_vpe_config();
+
+	cpu_set_vpe_id(&cpu_data[ncpu], tc);
+
+	return ncpu;
+}
+
+static void __init smvp_tc_init(unsigned int tc, unsigned int mvpconf0)
+{
+	unsigned long tmp;
+
+	if (!tc)
+		return;
+
+	/* bind a TC to each VPE, May as well put all excess TC's
+	   on the last VPE */
+	if (tc >= (((mvpconf0 & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT)+1))
+		write_tc_c0_tcbind(read_tc_c0_tcbind() | ((mvpconf0 & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT));
+	else {
+		write_tc_c0_tcbind(read_tc_c0_tcbind() | tc);
+
+		/* and set XTC */
+		write_vpe_c0_vpeconf0(read_vpe_c0_vpeconf0() | (tc << VPECONF0_XTC_SHIFT));
+	}
+
+	tmp = read_tc_c0_tcstatus();
+
+	/* mark not allocated and not dynamically allocatable */
+	tmp &= ~(TCSTATUS_A | TCSTATUS_DA);
+	tmp |= TCSTATUS_IXMT;		/* interrupt exempt */
+	write_tc_c0_tcstatus(tmp);
+
+	write_tc_c0_tchalt(TCHALT_H);
+}
+
+static void vsmp_init_secondary(void)
+{
+	/* This is Malta specific: IPI,performance and timer interrupts */
+	if (mips_gic_present())
+		change_c0_status(ST0_IM, STATUSF_IP2 | STATUSF_IP3 |
+					 STATUSF_IP4 | STATUSF_IP5 |
+					 STATUSF_IP6 | STATUSF_IP7);
+	else
+		change_c0_status(ST0_IM, STATUSF_IP0 | STATUSF_IP1 |
+					 STATUSF_IP6 | STATUSF_IP7);
+}
+
+static void vsmp_smp_finish(void)
+{
+	/* CDFIXME: remove this? */
+	write_c0_compare(read_c0_count() + (8* mips_hpt_frequency/HZ));
+
+#ifdef CONFIG_MIPS_MT_FPAFF
+	/* If we have an FPU, enroll ourselves in the FPU-full mask */
+	if (cpu_has_fpu)
+		cpumask_set_cpu(smp_processor_id(), &mt_fpu_cpumask);
+#endif /* CONFIG_MIPS_MT_FPAFF */
+
+	local_irq_enable();
+}
+
+/*
+ * Setup the PC, SP, and GP of a secondary processor and start it
+ * running!
+ * smp_bootstrap is the place to resume from
+ * __KSTK_TOS(idle) is apparently the stack pointer
+ * (unsigned long)idle->thread_info the gp
+ * assumes a 1:1 mapping of TC => VPE
+ */
+static int vsmp_boot_secondary(int cpu, struct task_struct *idle)
+{
+	struct thread_info *gp = task_thread_info(idle);
+	dvpe();
+	set_c0_mvpcontrol(MVPCONTROL_VPC);
+
+	settc(cpu);
+
+	/* restart */
+	write_tc_c0_tcrestart((unsigned long)&smp_bootstrap);
+
+	/* enable the tc this vpe/cpu will be running */
+	write_tc_c0_tcstatus((read_tc_c0_tcstatus() & ~TCSTATUS_IXMT) | TCSTATUS_A);
+
+	write_tc_c0_tchalt(0);
+
+	/* enable the VPE */
+	write_vpe_c0_vpeconf0(read_vpe_c0_vpeconf0() | VPECONF0_VPA);
+
+	/* stack pointer */
+	write_tc_gpr_sp( __KSTK_TOS(idle));
+
+	/* global pointer */
+	write_tc_gpr_gp((unsigned long)gp);
+
+	flush_icache_range((unsigned long)gp,
+			   (unsigned long)(gp + sizeof(struct thread_info)));
+
+	/* finally out of configuration and into chaos */
+	clear_c0_mvpcontrol(MVPCONTROL_VPC);
+
+	evpe(EVPE_ENABLE);
+
+	return 0;
+}
+
+/*
+ * Common setup before any secondaries are started
+ * Make sure all CPU's are in a sensible state before we boot any of the
+ * secondaries
+ */
+static void __init vsmp_smp_setup(void)
+{
+	unsigned int mvpconf0, ntc, tc, ncpu = 0;
+	unsigned int nvpe;
+
+#ifdef CONFIG_MIPS_MT_FPAFF
+	/* If we have an FPU, enroll ourselves in the FPU-full mask */
+	if (cpu_has_fpu)
+		cpumask_set_cpu(0, &mt_fpu_cpumask);
+#endif /* CONFIG_MIPS_MT_FPAFF */
+	if (!cpu_has_mipsmt)
+		return;
+
+	/* disable MT so we can configure */
+	dvpe();
+	dmt();
+
+	/* Put MVPE's into 'configuration state' */
+	set_c0_mvpcontrol(MVPCONTROL_VPC);
+
+	mvpconf0 = read_c0_mvpconf0();
+	ntc = (mvpconf0 & MVPCONF0_PTC) >> MVPCONF0_PTC_SHIFT;
+
+	nvpe = ((mvpconf0 & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT) + 1;
+	smp_num_siblings = nvpe;
+
+	/* we'll always have more TC's than VPE's, so loop setting everything
+	   to a sensible state */
+	for (tc = 0; tc <= ntc; tc++) {
+		settc(tc);
+
+		smvp_tc_init(tc, mvpconf0);
+		ncpu = smvp_vpe_init(tc, mvpconf0, ncpu);
+	}
+
+	/* Release config state */
+	clear_c0_mvpcontrol(MVPCONTROL_VPC);
+
+	/* We'll wait until starting the secondaries before starting MVPE */
+
+	printk(KERN_INFO "Detected %i available secondary CPU(s)\n", ncpu);
+}
+
+static void __init vsmp_prepare_cpus(unsigned int max_cpus)
+{
+	mips_mt_set_cpuoptions();
+}
+
+const struct plat_smp_ops vsmp_smp_ops = {
+	.send_ipi_single	= mips_smp_send_ipi_single,
+	.send_ipi_mask		= mips_smp_send_ipi_mask,
+	.init_secondary		= vsmp_init_secondary,
+	.smp_finish		= vsmp_smp_finish,
+	.boot_secondary		= vsmp_boot_secondary,
+	.smp_setup		= vsmp_smp_setup,
+	.prepare_cpus		= vsmp_prepare_cpus,
+};
+
diff --git a/arch/mips/kernel/smp-up.c b/arch/mips/kernel/smp-up.c
new file mode 100644
index 000000000..525d3196f
--- /dev/null
+++ b/arch/mips/kernel/smp-up.c
@@ -0,0 +1,79 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2006, 07 by Ralf Baechle (ralf@linux-mips.org)
+ *
+ * Symmetric Uniprocessor (TM) Support
+ */
+#include <linux/kernel.h>
+#include <linux/sched.h>
+
+/*
+ * Send inter-processor interrupt
+ */
+static void up_send_ipi_single(int cpu, unsigned int action)
+{
+	panic(KERN_ERR "%s called", __func__);
+}
+
+static inline void up_send_ipi_mask(const struct cpumask *mask,
+				    unsigned int action)
+{
+	panic(KERN_ERR "%s called", __func__);
+}
+
+/*
+ *  After we've done initial boot, this function is called to allow the
+ *  board code to clean up state, if needed
+ */
+static void up_init_secondary(void)
+{
+}
+
+static void up_smp_finish(void)
+{
+}
+
+/*
+ * Firmware CPU startup hook
+ */
+static int up_boot_secondary(int cpu, struct task_struct *idle)
+{
+	return 0;
+}
+
+static void __init up_smp_setup(void)
+{
+}
+
+static void __init up_prepare_cpus(unsigned int max_cpus)
+{
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+static int up_cpu_disable(void)
+{
+	return -ENOSYS;
+}
+
+static void up_cpu_die(unsigned int cpu)
+{
+	BUG();
+}
+#endif
+
+const struct plat_smp_ops up_smp_ops = {
+	.send_ipi_single	= up_send_ipi_single,
+	.send_ipi_mask		= up_send_ipi_mask,
+	.init_secondary		= up_init_secondary,
+	.smp_finish		= up_smp_finish,
+	.boot_secondary		= up_boot_secondary,
+	.smp_setup		= up_smp_setup,
+	.prepare_cpus		= up_prepare_cpus,
+#ifdef CONFIG_HOTPLUG_CPU
+	.cpu_disable		= up_cpu_disable,
+	.cpu_die		= up_cpu_die,
+#endif
+};
diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
new file mode 100644
index 000000000..14db66dbc
--- /dev/null
+++ b/arch/mips/kernel/smp.c
@@ -0,0 +1,721 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ *
+ * Copyright (C) 2000, 2001 Kanoj Sarcar
+ * Copyright (C) 2000, 2001 Ralf Baechle
+ * Copyright (C) 2000, 2001 Silicon Graphics, Inc.
+ * Copyright (C) 2000, 2001, 2003 Broadcom Corporation
+ */
+#include <linux/cache.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/smp.h>
+#include <linux/spinlock.h>
+#include <linux/threads.h>
+#include <linux/export.h>
+#include <linux/time.h>
+#include <linux/timex.h>
+#include <linux/sched/mm.h>
+#include <linux/cpumask.h>
+#include <linux/cpu.h>
+#include <linux/err.h>
+#include <linux/ftrace.h>
+#include <linux/irqdomain.h>
+#include <linux/of.h>
+#include <linux/of_irq.h>
+
+#include <linux/atomic.h>
+#include <asm/cpu.h>
+#include <asm/ginvt.h>
+#include <asm/processor.h>
+#include <asm/idle.h>
+#include <asm/r4k-timer.h>
+#include <asm/mips-cps.h>
+#include <asm/mmu_context.h>
+#include <asm/time.h>
+#include <asm/setup.h>
+#include <asm/maar.h>
+
+int __cpu_number_map[CONFIG_MIPS_NR_CPU_NR_MAP];   /* Map physical to logical */
+EXPORT_SYMBOL(__cpu_number_map);
+
+int __cpu_logical_map[NR_CPUS];		/* Map logical to physical */
+EXPORT_SYMBOL(__cpu_logical_map);
+
+/* Number of TCs (or siblings in Intel speak) per CPU core */
+int smp_num_siblings = 1;
+EXPORT_SYMBOL(smp_num_siblings);
+
+/* representing the TCs (or siblings in Intel speak) of each logical CPU */
+cpumask_t cpu_sibling_map[NR_CPUS] __read_mostly;
+EXPORT_SYMBOL(cpu_sibling_map);
+
+/* representing the core map of multi-core chips of each logical CPU */
+cpumask_t cpu_core_map[NR_CPUS] __read_mostly;
+EXPORT_SYMBOL(cpu_core_map);
+
+static DECLARE_COMPLETION(cpu_starting);
+static DECLARE_COMPLETION(cpu_running);
+
+/*
+ * A logcal cpu mask containing only one VPE per core to
+ * reduce the number of IPIs on large MT systems.
+ */
+cpumask_t cpu_foreign_map[NR_CPUS] __read_mostly;
+EXPORT_SYMBOL(cpu_foreign_map);
+
+/* representing cpus for which sibling maps can be computed */
+static cpumask_t cpu_sibling_setup_map;
+
+/* representing cpus for which core maps can be computed */
+static cpumask_t cpu_core_setup_map;
+
+cpumask_t cpu_coherent_mask;
+
+#ifdef CONFIG_GENERIC_IRQ_IPI
+static struct irq_desc *call_desc;
+static struct irq_desc *sched_desc;
+#endif
+
+static inline void set_cpu_sibling_map(int cpu)
+{
+	int i;
+
+	cpumask_set_cpu(cpu, &cpu_sibling_setup_map);
+
+	if (smp_num_siblings > 1) {
+		for_each_cpu(i, &cpu_sibling_setup_map) {
+			if (cpus_are_siblings(cpu, i)) {
+				cpumask_set_cpu(i, &cpu_sibling_map[cpu]);
+				cpumask_set_cpu(cpu, &cpu_sibling_map[i]);
+			}
+		}
+	} else
+		cpumask_set_cpu(cpu, &cpu_sibling_map[cpu]);
+}
+
+static inline void set_cpu_core_map(int cpu)
+{
+	int i;
+
+	cpumask_set_cpu(cpu, &cpu_core_setup_map);
+
+	for_each_cpu(i, &cpu_core_setup_map) {
+		if (cpu_data[cpu].package == cpu_data[i].package) {
+			cpumask_set_cpu(i, &cpu_core_map[cpu]);
+			cpumask_set_cpu(cpu, &cpu_core_map[i]);
+		}
+	}
+}
+
+/*
+ * Calculate a new cpu_foreign_map mask whenever a
+ * new cpu appears or disappears.
+ */
+void calculate_cpu_foreign_map(void)
+{
+	int i, k, core_present;
+	cpumask_t temp_foreign_map;
+
+	/* Re-calculate the mask */
+	cpumask_clear(&temp_foreign_map);
+	for_each_online_cpu(i) {
+		core_present = 0;
+		for_each_cpu(k, &temp_foreign_map)
+			if (cpus_are_siblings(i, k))
+				core_present = 1;
+		if (!core_present)
+			cpumask_set_cpu(i, &temp_foreign_map);
+	}
+
+	for_each_online_cpu(i)
+		cpumask_andnot(&cpu_foreign_map[i],
+			       &temp_foreign_map, &cpu_sibling_map[i]);
+}
+
+const struct plat_smp_ops *mp_ops;
+EXPORT_SYMBOL(mp_ops);
+
+void register_smp_ops(const struct plat_smp_ops *ops)
+{
+	if (mp_ops)
+		printk(KERN_WARNING "Overriding previously set SMP ops\n");
+
+	mp_ops = ops;
+}
+
+#ifdef CONFIG_GENERIC_IRQ_IPI
+void mips_smp_send_ipi_single(int cpu, unsigned int action)
+{
+	mips_smp_send_ipi_mask(cpumask_of(cpu), action);
+}
+
+void mips_smp_send_ipi_mask(const struct cpumask *mask, unsigned int action)
+{
+	unsigned long flags;
+	unsigned int core;
+	int cpu;
+
+	local_irq_save(flags);
+
+	switch (action) {
+	case SMP_CALL_FUNCTION:
+		__ipi_send_mask(call_desc, mask);
+		break;
+
+	case SMP_RESCHEDULE_YOURSELF:
+		__ipi_send_mask(sched_desc, mask);
+		break;
+
+	default:
+		BUG();
+	}
+
+	if (mips_cpc_present()) {
+		for_each_cpu(cpu, mask) {
+			if (cpus_are_siblings(cpu, smp_processor_id()))
+				continue;
+
+			core = cpu_core(&cpu_data[cpu]);
+
+			while (!cpumask_test_cpu(cpu, &cpu_coherent_mask)) {
+				mips_cm_lock_other_cpu(cpu, CM_GCR_Cx_OTHER_BLOCK_LOCAL);
+				mips_cpc_lock_other(core);
+				write_cpc_co_cmd(CPC_Cx_CMD_PWRUP);
+				mips_cpc_unlock_other();
+				mips_cm_unlock_other();
+			}
+		}
+	}
+
+	local_irq_restore(flags);
+}
+
+
+static irqreturn_t ipi_resched_interrupt(int irq, void *dev_id)
+{
+	scheduler_ipi();
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t ipi_call_interrupt(int irq, void *dev_id)
+{
+	generic_smp_call_function_interrupt();
+
+	return IRQ_HANDLED;
+}
+
+static void smp_ipi_init_one(unsigned int virq, const char *name,
+			     irq_handler_t handler)
+{
+	int ret;
+
+	irq_set_handler(virq, handle_percpu_irq);
+	ret = request_irq(virq, handler, IRQF_PERCPU, name, NULL);
+	BUG_ON(ret);
+}
+
+static unsigned int call_virq, sched_virq;
+
+int mips_smp_ipi_allocate(const struct cpumask *mask)
+{
+	int virq;
+	struct irq_domain *ipidomain;
+	struct device_node *node;
+
+	node = of_irq_find_parent(of_root);
+	ipidomain = irq_find_matching_host(node, DOMAIN_BUS_IPI);
+
+	/*
+	 * Some platforms have half DT setup. So if we found irq node but
+	 * didn't find an ipidomain, try to search for one that is not in the
+	 * DT.
+	 */
+	if (node && !ipidomain)
+		ipidomain = irq_find_matching_host(NULL, DOMAIN_BUS_IPI);
+
+	/*
+	 * There are systems which use IPI IRQ domains, but only have one
+	 * registered when some runtime condition is met. For example a Malta
+	 * kernel may include support for GIC & CPU interrupt controller IPI
+	 * IRQ domains, but if run on a system with no GIC & no MT ASE then
+	 * neither will be supported or registered.
+	 *
+	 * We only have a problem if we're actually using multiple CPUs so fail
+	 * loudly if that is the case. Otherwise simply return, skipping IPI
+	 * setup, if we're running with only a single CPU.
+	 */
+	if (!ipidomain) {
+		BUG_ON(num_present_cpus() > 1);
+		return 0;
+	}
+
+	virq = irq_reserve_ipi(ipidomain, mask);
+	BUG_ON(!virq);
+	if (!call_virq)
+		call_virq = virq;
+
+	virq = irq_reserve_ipi(ipidomain, mask);
+	BUG_ON(!virq);
+	if (!sched_virq)
+		sched_virq = virq;
+
+	if (irq_domain_is_ipi_per_cpu(ipidomain)) {
+		int cpu;
+
+		for_each_cpu(cpu, mask) {
+			smp_ipi_init_one(call_virq + cpu, "IPI call",
+					 ipi_call_interrupt);
+			smp_ipi_init_one(sched_virq + cpu, "IPI resched",
+					 ipi_resched_interrupt);
+		}
+	} else {
+		smp_ipi_init_one(call_virq, "IPI call", ipi_call_interrupt);
+		smp_ipi_init_one(sched_virq, "IPI resched",
+				 ipi_resched_interrupt);
+	}
+
+	return 0;
+}
+
+int mips_smp_ipi_free(const struct cpumask *mask)
+{
+	struct irq_domain *ipidomain;
+	struct device_node *node;
+
+	node = of_irq_find_parent(of_root);
+	ipidomain = irq_find_matching_host(node, DOMAIN_BUS_IPI);
+
+	/*
+	 * Some platforms have half DT setup. So if we found irq node but
+	 * didn't find an ipidomain, try to search for one that is not in the
+	 * DT.
+	 */
+	if (node && !ipidomain)
+		ipidomain = irq_find_matching_host(NULL, DOMAIN_BUS_IPI);
+
+	BUG_ON(!ipidomain);
+
+	if (irq_domain_is_ipi_per_cpu(ipidomain)) {
+		int cpu;
+
+		for_each_cpu(cpu, mask) {
+			free_irq(call_virq + cpu, NULL);
+			free_irq(sched_virq + cpu, NULL);
+		}
+	}
+	irq_destroy_ipi(call_virq, mask);
+	irq_destroy_ipi(sched_virq, mask);
+	return 0;
+}
+
+
+static int __init mips_smp_ipi_init(void)
+{
+	if (num_possible_cpus() == 1)
+		return 0;
+
+	mips_smp_ipi_allocate(cpu_possible_mask);
+
+	call_desc = irq_to_desc(call_virq);
+	sched_desc = irq_to_desc(sched_virq);
+
+	return 0;
+}
+early_initcall(mips_smp_ipi_init);
+#endif
+
+/*
+ * First C code run on the secondary CPUs after being started up by
+ * the master.
+ */
+asmlinkage void start_secondary(void)
+{
+	unsigned int cpu;
+
+	cpu_probe();
+	per_cpu_trap_init(false);
+	mips_clockevent_init();
+	mp_ops->init_secondary();
+	cpu_report();
+	maar_init();
+
+	/*
+	 * XXX parity protection should be folded in here when it's converted
+	 * to an option instead of something based on .cputype
+	 */
+
+	calibrate_delay();
+	cpu = smp_processor_id();
+	cpu_data[cpu].udelay_val = loops_per_jiffy;
+
+	set_cpu_sibling_map(cpu);
+	set_cpu_core_map(cpu);
+
+	cpumask_set_cpu(cpu, &cpu_coherent_mask);
+	notify_cpu_starting(cpu);
+
+	/* Notify boot CPU that we're starting & ready to sync counters */
+	complete(&cpu_starting);
+
+	synchronise_count_slave(cpu);
+
+	/* The CPU is running and counters synchronised, now mark it online */
+	set_cpu_online(cpu, true);
+
+	calculate_cpu_foreign_map();
+
+	/*
+	 * Notify boot CPU that we're up & online and it can safely return
+	 * from __cpu_up
+	 */
+	complete(&cpu_running);
+
+	/*
+	 * irq will be enabled in ->smp_finish(), enabling it too early
+	 * is dangerous.
+	 */
+	WARN_ON_ONCE(!irqs_disabled());
+	mp_ops->smp_finish();
+
+	cpu_startup_entry(CPUHP_AP_ONLINE_IDLE);
+}
+
+static void stop_this_cpu(void *dummy)
+{
+	/*
+	 * Remove this CPU:
+	 */
+
+	set_cpu_online(smp_processor_id(), false);
+	calculate_cpu_foreign_map();
+	local_irq_disable();
+	while (1);
+}
+
+void smp_send_stop(void)
+{
+	smp_call_function(stop_this_cpu, NULL, 0);
+}
+
+void __init smp_cpus_done(unsigned int max_cpus)
+{
+}
+
+/* called from main before smp_init() */
+void __init smp_prepare_cpus(unsigned int max_cpus)
+{
+	init_new_context(current, &init_mm);
+	current_thread_info()->cpu = 0;
+	mp_ops->prepare_cpus(max_cpus);
+	set_cpu_sibling_map(0);
+	set_cpu_core_map(0);
+	calculate_cpu_foreign_map();
+#ifndef CONFIG_HOTPLUG_CPU
+	init_cpu_present(cpu_possible_mask);
+#endif
+	cpumask_copy(&cpu_coherent_mask, cpu_possible_mask);
+}
+
+/* preload SMP state for boot cpu */
+void smp_prepare_boot_cpu(void)
+{
+	if (mp_ops->prepare_boot_cpu)
+		mp_ops->prepare_boot_cpu();
+	set_cpu_possible(0, true);
+	set_cpu_online(0, true);
+}
+
+int __cpu_up(unsigned int cpu, struct task_struct *tidle)
+{
+	int err;
+
+	err = mp_ops->boot_secondary(cpu, tidle);
+	if (err)
+		return err;
+
+	/* Wait for CPU to start and be ready to sync counters */
+	if (!wait_for_completion_timeout(&cpu_starting,
+					 msecs_to_jiffies(1000))) {
+		pr_crit("CPU%u: failed to start\n", cpu);
+		return -EIO;
+	}
+
+	synchronise_count_master(cpu);
+
+	/* Wait for CPU to finish startup & mark itself online before return */
+	wait_for_completion(&cpu_running);
+	return 0;
+}
+
+/* Not really SMP stuff ... */
+int setup_profiling_timer(unsigned int multiplier)
+{
+	return 0;
+}
+
+static void flush_tlb_all_ipi(void *info)
+{
+	local_flush_tlb_all();
+}
+
+void flush_tlb_all(void)
+{
+	if (cpu_has_mmid) {
+		htw_stop();
+		ginvt_full();
+		sync_ginv();
+		instruction_hazard();
+		htw_start();
+		return;
+	}
+
+	on_each_cpu(flush_tlb_all_ipi, NULL, 1);
+}
+
+static void flush_tlb_mm_ipi(void *mm)
+{
+	drop_mmu_context((struct mm_struct *)mm);
+}
+
+/*
+ * Special Variant of smp_call_function for use by TLB functions:
+ *
+ *  o No return value
+ *  o collapses to normal function call on UP kernels
+ *  o collapses to normal function call on systems with a single shared
+ *    primary cache.
+ */
+static inline void smp_on_other_tlbs(void (*func) (void *info), void *info)
+{
+	smp_call_function(func, info, 1);
+}
+
+static inline void smp_on_each_tlb(void (*func) (void *info), void *info)
+{
+	preempt_disable();
+
+	smp_on_other_tlbs(func, info);
+	func(info);
+
+	preempt_enable();
+}
+
+/*
+ * The following tlb flush calls are invoked when old translations are
+ * being torn down, or pte attributes are changing. For single threaded
+ * address spaces, a new context is obtained on the current cpu, and tlb
+ * context on other cpus are invalidated to force a new context allocation
+ * at switch_mm time, should the mm ever be used on other cpus. For
+ * multithreaded address spaces, intercpu interrupts have to be sent.
+ * Another case where intercpu interrupts are required is when the target
+ * mm might be active on another cpu (eg debuggers doing the flushes on
+ * behalf of debugees, kswapd stealing pages from another process etc).
+ * Kanoj 07/00.
+ */
+
+void flush_tlb_mm(struct mm_struct *mm)
+{
+	preempt_disable();
+
+	if (cpu_has_mmid) {
+		/*
+		 * No need to worry about other CPUs - the ginvt in
+		 * drop_mmu_context() will be globalized.
+		 */
+	} else if ((atomic_read(&mm->mm_users) != 1) || (current->mm != mm)) {
+		smp_on_other_tlbs(flush_tlb_mm_ipi, mm);
+	} else {
+		unsigned int cpu;
+
+		for_each_online_cpu(cpu) {
+			if (cpu != smp_processor_id() && cpu_context(cpu, mm))
+				set_cpu_context(cpu, mm, 0);
+		}
+	}
+	drop_mmu_context(mm);
+
+	preempt_enable();
+}
+
+struct flush_tlb_data {
+	struct vm_area_struct *vma;
+	unsigned long addr1;
+	unsigned long addr2;
+};
+
+static void flush_tlb_range_ipi(void *info)
+{
+	struct flush_tlb_data *fd = info;
+
+	local_flush_tlb_range(fd->vma, fd->addr1, fd->addr2);
+}
+
+void flush_tlb_range(struct vm_area_struct *vma, unsigned long start, unsigned long end)
+{
+	struct mm_struct *mm = vma->vm_mm;
+	unsigned long addr;
+	u32 old_mmid;
+
+	preempt_disable();
+	if (cpu_has_mmid) {
+		htw_stop();
+		old_mmid = read_c0_memorymapid();
+		write_c0_memorymapid(cpu_asid(0, mm));
+		mtc0_tlbw_hazard();
+		addr = round_down(start, PAGE_SIZE * 2);
+		end = round_up(end, PAGE_SIZE * 2);
+		do {
+			ginvt_va_mmid(addr);
+			sync_ginv();
+			addr += PAGE_SIZE * 2;
+		} while (addr < end);
+		write_c0_memorymapid(old_mmid);
+		instruction_hazard();
+		htw_start();
+	} else if ((atomic_read(&mm->mm_users) != 1) || (current->mm != mm)) {
+		struct flush_tlb_data fd = {
+			.vma = vma,
+			.addr1 = start,
+			.addr2 = end,
+		};
+
+		smp_on_other_tlbs(flush_tlb_range_ipi, &fd);
+		local_flush_tlb_range(vma, start, end);
+	} else {
+		unsigned int cpu;
+		int exec = vma->vm_flags & VM_EXEC;
+
+		for_each_online_cpu(cpu) {
+			/*
+			 * flush_cache_range() will only fully flush icache if
+			 * the VMA is executable, otherwise we must invalidate
+			 * ASID without it appearing to has_valid_asid() as if
+			 * mm has been completely unused by that CPU.
+			 */
+			if (cpu != smp_processor_id() && cpu_context(cpu, mm))
+				set_cpu_context(cpu, mm, !exec);
+		}
+		local_flush_tlb_range(vma, start, end);
+	}
+	preempt_enable();
+}
+
+static void flush_tlb_kernel_range_ipi(void *info)
+{
+	struct flush_tlb_data *fd = info;
+
+	local_flush_tlb_kernel_range(fd->addr1, fd->addr2);
+}
+
+void flush_tlb_kernel_range(unsigned long start, unsigned long end)
+{
+	struct flush_tlb_data fd = {
+		.addr1 = start,
+		.addr2 = end,
+	};
+
+	on_each_cpu(flush_tlb_kernel_range_ipi, &fd, 1);
+}
+
+static void flush_tlb_page_ipi(void *info)
+{
+	struct flush_tlb_data *fd = info;
+
+	local_flush_tlb_page(fd->vma, fd->addr1);
+}
+
+void flush_tlb_page(struct vm_area_struct *vma, unsigned long page)
+{
+	u32 old_mmid;
+
+	preempt_disable();
+	if (cpu_has_mmid) {
+		htw_stop();
+		old_mmid = read_c0_memorymapid();
+		write_c0_memorymapid(cpu_asid(0, vma->vm_mm));
+		mtc0_tlbw_hazard();
+		ginvt_va_mmid(page);
+		sync_ginv();
+		write_c0_memorymapid(old_mmid);
+		instruction_hazard();
+		htw_start();
+	} else if ((atomic_read(&vma->vm_mm->mm_users) != 1) ||
+		   (current->mm != vma->vm_mm)) {
+		struct flush_tlb_data fd = {
+			.vma = vma,
+			.addr1 = page,
+		};
+
+		smp_on_other_tlbs(flush_tlb_page_ipi, &fd);
+		local_flush_tlb_page(vma, page);
+	} else {
+		unsigned int cpu;
+
+		for_each_online_cpu(cpu) {
+			/*
+			 * flush_cache_page() only does partial flushes, so
+			 * invalidate ASID without it appearing to
+			 * has_valid_asid() as if mm has been completely unused
+			 * by that CPU.
+			 */
+			if (cpu != smp_processor_id() && cpu_context(cpu, vma->vm_mm))
+				set_cpu_context(cpu, vma->vm_mm, 1);
+		}
+		local_flush_tlb_page(vma, page);
+	}
+	preempt_enable();
+}
+
+static void flush_tlb_one_ipi(void *info)
+{
+	unsigned long vaddr = (unsigned long) info;
+
+	local_flush_tlb_one(vaddr);
+}
+
+void flush_tlb_one(unsigned long vaddr)
+{
+	smp_on_each_tlb(flush_tlb_one_ipi, (void *) vaddr);
+}
+
+EXPORT_SYMBOL(flush_tlb_page);
+EXPORT_SYMBOL(flush_tlb_one);
+
+#ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST
+
+static DEFINE_PER_CPU(call_single_data_t, tick_broadcast_csd);
+
+void tick_broadcast(const struct cpumask *mask)
+{
+	call_single_data_t *csd;
+	int cpu;
+
+	for_each_cpu(cpu, mask) {
+		csd = &per_cpu(tick_broadcast_csd, cpu);
+		smp_call_function_single_async(cpu, csd);
+	}
+}
+
+static void tick_broadcast_callee(void *info)
+{
+	tick_receive_broadcast();
+}
+
+static int __init tick_broadcast_init(void)
+{
+	call_single_data_t *csd;
+	int cpu;
+
+	for (cpu = 0; cpu < NR_CPUS; cpu++) {
+		csd = &per_cpu(tick_broadcast_csd, cpu);
+		csd->func = tick_broadcast_callee;
+	}
+
+	return 0;
+}
+early_initcall(tick_broadcast_init);
+
+#endif /* CONFIG_GENERIC_CLOCKEVENTS_BROADCAST */
diff --git a/arch/mips/kernel/spinlock_test.c b/arch/mips/kernel/spinlock_test.c
new file mode 100644
index 000000000..ab4e3e1b1
--- /dev/null
+++ b/arch/mips/kernel/spinlock_test.c
@@ -0,0 +1,127 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/init.h>
+#include <linux/kthread.h>
+#include <linux/hrtimer.h>
+#include <linux/fs.h>
+#include <linux/debugfs.h>
+#include <linux/export.h>
+#include <linux/spinlock.h>
+#include <asm/debug.h>
+
+static int ss_get(void *data, u64 *val)
+{
+	ktime_t start, finish;
+	int loops;
+	int cont;
+	DEFINE_RAW_SPINLOCK(ss_spin);
+
+	loops = 1000000;
+	cont = 1;
+
+	start = ktime_get();
+
+	while (cont) {
+		raw_spin_lock(&ss_spin);
+		loops--;
+		if (loops == 0)
+			cont = 0;
+		raw_spin_unlock(&ss_spin);
+	}
+
+	finish = ktime_get();
+
+	*val = ktime_us_delta(finish, start);
+
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(fops_ss, ss_get, NULL, "%llu\n");
+
+
+
+struct spin_multi_state {
+	raw_spinlock_t lock;
+	atomic_t start_wait;
+	atomic_t enter_wait;
+	atomic_t exit_wait;
+	int loops;
+};
+
+struct spin_multi_per_thread {
+	struct spin_multi_state *state;
+	ktime_t start;
+};
+
+static int multi_other(void *data)
+{
+	int loops;
+	int cont;
+	struct spin_multi_per_thread *pt = data;
+	struct spin_multi_state *s = pt->state;
+
+	loops = s->loops;
+	cont = 1;
+
+	atomic_dec(&s->enter_wait);
+
+	while (atomic_read(&s->enter_wait))
+		; /* spin */
+
+	pt->start = ktime_get();
+
+	atomic_dec(&s->start_wait);
+
+	while (atomic_read(&s->start_wait))
+		; /* spin */
+
+	while (cont) {
+		raw_spin_lock(&s->lock);
+		loops--;
+		if (loops == 0)
+			cont = 0;
+		raw_spin_unlock(&s->lock);
+	}
+
+	atomic_dec(&s->exit_wait);
+	while (atomic_read(&s->exit_wait))
+		; /* spin */
+	return 0;
+}
+
+static int multi_get(void *data, u64 *val)
+{
+	ktime_t finish;
+	struct spin_multi_state ms;
+	struct spin_multi_per_thread t1, t2;
+
+	ms.lock = __RAW_SPIN_LOCK_UNLOCKED("multi_get");
+	ms.loops = 1000000;
+
+	atomic_set(&ms.start_wait, 2);
+	atomic_set(&ms.enter_wait, 2);
+	atomic_set(&ms.exit_wait, 2);
+	t1.state = &ms;
+	t2.state = &ms;
+
+	kthread_run(multi_other, &t2, "multi_get");
+
+	multi_other(&t1);
+
+	finish = ktime_get();
+
+	*val = ktime_us_delta(finish, t1.start);
+
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(fops_multi, multi_get, NULL, "%llu\n");
+
+static int __init spinlock_test(void)
+{
+	debugfs_create_file("spin_single", S_IRUGO, mips_debugfs_dir, NULL,
+			    &fops_ss);
+	debugfs_create_file("spin_multi", S_IRUGO, mips_debugfs_dir, NULL,
+			    &fops_multi);
+	return 0;
+}
+device_initcall(spinlock_test);
diff --git a/arch/mips/kernel/spram.c b/arch/mips/kernel/spram.c
new file mode 100644
index 000000000..d5d96214c
--- /dev/null
+++ b/arch/mips/kernel/spram.c
@@ -0,0 +1,220 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * MIPS SPRAM support
+ *
+ * Copyright (C) 2007, 2008 MIPS Technologies, Inc.
+ */
+#include <linux/kernel.h>
+#include <linux/ptrace.h>
+#include <linux/stddef.h>
+
+#include <asm/fpu.h>
+#include <asm/mipsregs.h>
+#include <asm/r4kcache.h>
+#include <asm/hazards.h>
+
+/*
+ * These definitions are correct for the 24K/34K/74K SPRAM sample
+ * implementation. The 4KS interpreted the tags differently...
+ */
+#define SPRAM_TAG0_ENABLE	0x00000080
+#define SPRAM_TAG0_PA_MASK	0xfffff000
+#define SPRAM_TAG1_SIZE_MASK	0xfffff000
+
+#define SPRAM_TAG_STRIDE	8
+
+#define ERRCTL_SPRAM		(1 << 28)
+
+/* errctl access */
+#define read_c0_errctl(x) read_c0_ecc(x)
+#define write_c0_errctl(x) write_c0_ecc(x)
+
+/*
+ * Different semantics to the set_c0_* function built by __BUILD_SET_C0
+ */
+static unsigned int bis_c0_errctl(unsigned int set)
+{
+	unsigned int res;
+	res = read_c0_errctl();
+	write_c0_errctl(res | set);
+	return res;
+}
+
+static void ispram_store_tag(unsigned int offset, unsigned int data)
+{
+	unsigned int errctl;
+
+	/* enable SPRAM tag access */
+	errctl = bis_c0_errctl(ERRCTL_SPRAM);
+	ehb();
+
+	write_c0_taglo(data);
+	ehb();
+
+	cache_op(Index_Store_Tag_I, CKSEG0|offset);
+	ehb();
+
+	write_c0_errctl(errctl);
+	ehb();
+}
+
+
+static unsigned int ispram_load_tag(unsigned int offset)
+{
+	unsigned int data;
+	unsigned int errctl;
+
+	/* enable SPRAM tag access */
+	errctl = bis_c0_errctl(ERRCTL_SPRAM);
+	ehb();
+	cache_op(Index_Load_Tag_I, CKSEG0 | offset);
+	ehb();
+	data = read_c0_taglo();
+	ehb();
+	write_c0_errctl(errctl);
+	ehb();
+
+	return data;
+}
+
+static void dspram_store_tag(unsigned int offset, unsigned int data)
+{
+	unsigned int errctl;
+
+	/* enable SPRAM tag access */
+	errctl = bis_c0_errctl(ERRCTL_SPRAM);
+	ehb();
+	write_c0_dtaglo(data);
+	ehb();
+	cache_op(Index_Store_Tag_D, CKSEG0 | offset);
+	ehb();
+	write_c0_errctl(errctl);
+	ehb();
+}
+
+
+static unsigned int dspram_load_tag(unsigned int offset)
+{
+	unsigned int data;
+	unsigned int errctl;
+
+	errctl = bis_c0_errctl(ERRCTL_SPRAM);
+	ehb();
+	cache_op(Index_Load_Tag_D, CKSEG0 | offset);
+	ehb();
+	data = read_c0_dtaglo();
+	ehb();
+	write_c0_errctl(errctl);
+	ehb();
+
+	return data;
+}
+
+static void probe_spram(char *type,
+	    unsigned int base,
+	    unsigned int (*read)(unsigned int),
+	    void (*write)(unsigned int, unsigned int))
+{
+	unsigned int firstsize = 0, lastsize = 0;
+	unsigned int firstpa = 0, lastpa = 0, pa = 0;
+	unsigned int offset = 0;
+	unsigned int size, tag0, tag1;
+	unsigned int enabled;
+	int i;
+
+	/*
+	 * The limit is arbitrary but avoids the loop running away if
+	 * the SPRAM tags are implemented differently
+	 */
+
+	for (i = 0; i < 8; i++) {
+		tag0 = read(offset);
+		tag1 = read(offset+SPRAM_TAG_STRIDE);
+		pr_debug("DBG %s%d: tag0=%08x tag1=%08x\n",
+			 type, i, tag0, tag1);
+
+		size = tag1 & SPRAM_TAG1_SIZE_MASK;
+
+		if (size == 0)
+			break;
+
+		if (i != 0) {
+			/* tags may repeat... */
+			if ((pa == firstpa && size == firstsize) ||
+			    (pa == lastpa && size == lastsize))
+				break;
+		}
+
+		/* Align base with size */
+		base = (base + size - 1) & ~(size-1);
+
+		/* reprogram the base address base address and enable */
+		tag0 = (base & SPRAM_TAG0_PA_MASK) | SPRAM_TAG0_ENABLE;
+		write(offset, tag0);
+
+		base += size;
+
+		/* reread the tag */
+		tag0 = read(offset);
+		pa = tag0 & SPRAM_TAG0_PA_MASK;
+		enabled = tag0 & SPRAM_TAG0_ENABLE;
+
+		if (i == 0) {
+			firstpa = pa;
+			firstsize = size;
+		}
+
+		lastpa = pa;
+		lastsize = size;
+
+		if (strcmp(type, "DSPRAM") == 0) {
+			unsigned int *vp = (unsigned int *)(CKSEG1 | pa);
+			unsigned int v;
+#define TDAT	0x5a5aa5a5
+			vp[0] = TDAT;
+			vp[1] = ~TDAT;
+
+			mb();
+
+			v = vp[0];
+			if (v != TDAT)
+				printk(KERN_ERR "vp=%p wrote=%08x got=%08x\n",
+				       vp, TDAT, v);
+			v = vp[1];
+			if (v != ~TDAT)
+				printk(KERN_ERR "vp=%p wrote=%08x got=%08x\n",
+				       vp+1, ~TDAT, v);
+		}
+
+		pr_info("%s%d: PA=%08x,Size=%08x%s\n",
+			type, i, pa, size, enabled ? ",enabled" : "");
+		offset += 2 * SPRAM_TAG_STRIDE;
+	}
+}
+void spram_config(void)
+{
+	unsigned int config0;
+
+	switch (current_cpu_type()) {
+	case CPU_24K:
+	case CPU_34K:
+	case CPU_74K:
+	case CPU_1004K:
+	case CPU_1074K:
+	case CPU_INTERAPTIV:
+	case CPU_PROAPTIV:
+	case CPU_P5600:
+	case CPU_QEMU_GENERIC:
+	case CPU_I6400:
+	case CPU_P6600:
+		config0 = read_c0_config();
+		/* FIXME: addresses are Malta specific */
+		if (config0 & MIPS_CONF_ISP) {
+			probe_spram("ISPRAM", 0x1c000000,
+				    &ispram_load_tag, &ispram_store_tag);
+		}
+		if (config0 & MIPS_CONF_DSP)
+			probe_spram("DSPRAM", 0x1c100000,
+				    &dspram_load_tag, &dspram_store_tag);
+	}
+}
diff --git a/arch/mips/kernel/stacktrace.c b/arch/mips/kernel/stacktrace.c
new file mode 100644
index 000000000..f2e720940
--- /dev/null
+++ b/arch/mips/kernel/stacktrace.c
@@ -0,0 +1,93 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Stack trace management functions
+ *
+ *  Copyright (C) 2006 Atsushi Nemoto <anemo@mba.ocn.ne.jp>
+ */
+#include <linux/sched.h>
+#include <linux/sched/debug.h>
+#include <linux/sched/task_stack.h>
+#include <linux/stacktrace.h>
+#include <linux/export.h>
+#include <asm/stacktrace.h>
+
+/*
+ * Save stack-backtrace addresses into a stack_trace buffer:
+ */
+static void save_raw_context_stack(struct stack_trace *trace,
+	unsigned long reg29, int savesched)
+{
+	unsigned long *sp = (unsigned long *)reg29;
+	unsigned long addr;
+
+	while (!kstack_end(sp)) {
+		addr = *sp++;
+		if (__kernel_text_address(addr) &&
+		    (savesched || !in_sched_functions(addr))) {
+			if (trace->skip > 0)
+				trace->skip--;
+			else
+				trace->entries[trace->nr_entries++] = addr;
+			if (trace->nr_entries >= trace->max_entries)
+				break;
+		}
+	}
+}
+
+static void save_context_stack(struct stack_trace *trace,
+	struct task_struct *tsk, struct pt_regs *regs, int savesched)
+{
+	unsigned long sp = regs->regs[29];
+#ifdef CONFIG_KALLSYMS
+	unsigned long ra = regs->regs[31];
+	unsigned long pc = regs->cp0_epc;
+
+	if (raw_show_trace || !__kernel_text_address(pc)) {
+		unsigned long stack_page =
+			(unsigned long)task_stack_page(tsk);
+		if (stack_page && sp >= stack_page &&
+		    sp <= stack_page + THREAD_SIZE - 32)
+			save_raw_context_stack(trace, sp, savesched);
+		return;
+	}
+	do {
+		if (savesched || !in_sched_functions(pc)) {
+			if (trace->skip > 0)
+				trace->skip--;
+			else
+				trace->entries[trace->nr_entries++] = pc;
+			if (trace->nr_entries >= trace->max_entries)
+				break;
+		}
+		pc = unwind_stack(tsk, &sp, pc, &ra);
+	} while (pc);
+#else
+	save_raw_context_stack(trace, sp, savesched);
+#endif
+}
+
+/*
+ * Save stack-backtrace addresses into a stack_trace buffer.
+ */
+void save_stack_trace(struct stack_trace *trace)
+{
+	save_stack_trace_tsk(current, trace);
+}
+EXPORT_SYMBOL_GPL(save_stack_trace);
+
+void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
+{
+	struct pt_regs dummyregs;
+	struct pt_regs *regs = &dummyregs;
+
+	WARN_ON(trace->nr_entries || !trace->max_entries);
+
+	if (tsk != current) {
+		regs->regs[29] = tsk->thread.reg29;
+		regs->regs[31] = 0;
+		regs->cp0_epc = tsk->thread.reg31;
+	} else
+		prepare_frametrace(regs);
+	save_context_stack(trace, tsk, regs, tsk == current);
+}
+EXPORT_SYMBOL_GPL(save_stack_trace_tsk);
diff --git a/arch/mips/kernel/sync-r4k.c b/arch/mips/kernel/sync-r4k.c
new file mode 100644
index 000000000..abdd7aaa3
--- /dev/null
+++ b/arch/mips/kernel/sync-r4k.c
@@ -0,0 +1,122 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Count register synchronisation.
+ *
+ * All CPUs will have their count registers synchronised to the CPU0 next time
+ * value. This can cause a small timewarp for CPU0. All other CPU's should
+ * not have done anything significant (but they may have had interrupts
+ * enabled briefly - prom_smp_finish() should not be responsible for enabling
+ * interrupts...)
+ */
+
+#include <linux/kernel.h>
+#include <linux/irqflags.h>
+#include <linux/cpumask.h>
+
+#include <asm/r4k-timer.h>
+#include <linux/atomic.h>
+#include <asm/barrier.h>
+#include <asm/mipsregs.h>
+
+static unsigned int initcount = 0;
+static atomic_t count_count_start = ATOMIC_INIT(0);
+static atomic_t count_count_stop = ATOMIC_INIT(0);
+
+#define COUNTON 100
+#define NR_LOOPS 3
+
+void synchronise_count_master(int cpu)
+{
+	int i;
+	unsigned long flags;
+
+	pr_info("Synchronize counters for CPU %u: ", cpu);
+
+	local_irq_save(flags);
+
+	/*
+	 * We loop a few times to get a primed instruction cache,
+	 * then the last pass is more or less synchronised and
+	 * the master and slaves each set their cycle counters to a known
+	 * value all at once. This reduces the chance of having random offsets
+	 * between the processors, and guarantees that the maximum
+	 * delay between the cycle counters is never bigger than
+	 * the latency of information-passing (cachelines) between
+	 * two CPUs.
+	 */
+
+	for (i = 0; i < NR_LOOPS; i++) {
+		/* slaves loop on '!= 2' */
+		while (atomic_read(&count_count_start) != 1)
+			mb();
+		atomic_set(&count_count_stop, 0);
+		smp_wmb();
+
+		/* Let the slave writes its count register */
+		atomic_inc(&count_count_start);
+
+		/* Count will be initialised to current timer */
+		if (i == 1)
+			initcount = read_c0_count();
+
+		/*
+		 * Everyone initialises count in the last loop:
+		 */
+		if (i == NR_LOOPS-1)
+			write_c0_count(initcount);
+
+		/*
+		 * Wait for slave to leave the synchronization point:
+		 */
+		while (atomic_read(&count_count_stop) != 1)
+			mb();
+		atomic_set(&count_count_start, 0);
+		smp_wmb();
+		atomic_inc(&count_count_stop);
+	}
+	/* Arrange for an interrupt in a short while */
+	write_c0_compare(read_c0_count() + COUNTON);
+
+	local_irq_restore(flags);
+
+	/*
+	 * i386 code reported the skew here, but the
+	 * count registers were almost certainly out of sync
+	 * so no point in alarming people
+	 */
+	pr_cont("done.\n");
+}
+
+void synchronise_count_slave(int cpu)
+{
+	int i;
+	unsigned long flags;
+
+	local_irq_save(flags);
+
+	/*
+	 * Not every cpu is online at the time this gets called,
+	 * so we first wait for the master to say everyone is ready
+	 */
+
+	for (i = 0; i < NR_LOOPS; i++) {
+		atomic_inc(&count_count_start);
+		while (atomic_read(&count_count_start) != 2)
+			mb();
+
+		/*
+		 * Everyone initialises count in the last loop:
+		 */
+		if (i == NR_LOOPS-1)
+			write_c0_count(initcount);
+
+		atomic_inc(&count_count_stop);
+		while (atomic_read(&count_count_stop) != 2)
+			mb();
+	}
+	/* Arrange for an interrupt in a short while */
+	write_c0_compare(read_c0_count() + COUNTON);
+
+	local_irq_restore(flags);
+}
+#undef NR_LOOPS
diff --git a/arch/mips/kernel/syscall.c b/arch/mips/kernel/syscall.c
new file mode 100644
index 000000000..5512cd586
--- /dev/null
+++ b/arch/mips/kernel/syscall.c
@@ -0,0 +1,242 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1995, 1996, 1997, 2000, 2001, 05 by Ralf Baechle
+ * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ * Copyright (C) 2001 MIPS Technologies, Inc.
+ */
+#include <linux/capability.h>
+#include <linux/errno.h>
+#include <linux/linkage.h>
+#include <linux/fs.h>
+#include <linux/smp.h>
+#include <linux/ptrace.h>
+#include <linux/string.h>
+#include <linux/syscalls.h>
+#include <linux/file.h>
+#include <linux/utsname.h>
+#include <linux/unistd.h>
+#include <linux/sem.h>
+#include <linux/msg.h>
+#include <linux/shm.h>
+#include <linux/compiler.h>
+#include <linux/ipc.h>
+#include <linux/uaccess.h>
+#include <linux/slab.h>
+#include <linux/elf.h>
+#include <linux/sched/task_stack.h>
+
+#include <asm/asm.h>
+#include <asm/asm-eva.h>
+#include <asm/branch.h>
+#include <asm/cachectl.h>
+#include <asm/cacheflush.h>
+#include <asm/asm-offsets.h>
+#include <asm/signal.h>
+#include <asm/sim.h>
+#include <asm/shmparam.h>
+#include <asm/sync.h>
+#include <asm/sysmips.h>
+#include <asm/switch_to.h>
+
+/*
+ * For historic reasons the pipe(2) syscall on MIPS has an unusual calling
+ * convention.	It returns results in registers $v0 / $v1 which means there
+ * is no need for it to do verify the validity of a userspace pointer
+ * argument.  Historically that used to be expensive in Linux.	These days
+ * the performance advantage is negligible.
+ */
+asmlinkage int sysm_pipe(void)
+{
+	int fd[2];
+	int error = do_pipe_flags(fd, 0);
+	if (error)
+		return error;
+	current_pt_regs()->regs[3] = fd[1];
+	return fd[0];
+}
+
+SYSCALL_DEFINE6(mips_mmap, unsigned long, addr, unsigned long, len,
+	unsigned long, prot, unsigned long, flags, unsigned long,
+	fd, off_t, offset)
+{
+	if (offset & ~PAGE_MASK)
+		return -EINVAL;
+	return ksys_mmap_pgoff(addr, len, prot, flags, fd,
+			       offset >> PAGE_SHIFT);
+}
+
+SYSCALL_DEFINE6(mips_mmap2, unsigned long, addr, unsigned long, len,
+	unsigned long, prot, unsigned long, flags, unsigned long, fd,
+	unsigned long, pgoff)
+{
+	if (pgoff & (~PAGE_MASK >> 12))
+		return -EINVAL;
+
+	return ksys_mmap_pgoff(addr, len, prot, flags, fd,
+			       pgoff >> (PAGE_SHIFT - 12));
+}
+
+save_static_function(sys_fork);
+save_static_function(sys_clone);
+save_static_function(sys_clone3);
+
+SYSCALL_DEFINE1(set_thread_area, unsigned long, addr)
+{
+	struct thread_info *ti = task_thread_info(current);
+
+	ti->tp_value = addr;
+	if (cpu_has_userlocal)
+		write_c0_userlocal(addr);
+
+	return 0;
+}
+
+static inline int mips_atomic_set(unsigned long addr, unsigned long new)
+{
+	unsigned long old, tmp;
+	struct pt_regs *regs;
+	unsigned int err;
+
+	if (unlikely(addr & 3))
+		return -EINVAL;
+
+	if (unlikely(!access_ok((const void __user *)addr, 4)))
+		return -EINVAL;
+
+	if (cpu_has_llsc && IS_ENABLED(CONFIG_WAR_R10000_LLSC)) {
+		__asm__ __volatile__ (
+		"	.set	push					\n"
+		"	.set	arch=r4000				\n"
+		"	li	%[err], 0				\n"
+		"1:	ll	%[old], (%[addr])			\n"
+		"	move	%[tmp], %[new]				\n"
+		"2:	sc	%[tmp], (%[addr])			\n"
+		"	beqzl	%[tmp], 1b				\n"
+		"3:							\n"
+		"	.insn						\n"
+		"	.section .fixup,\"ax\"				\n"
+		"4:	li	%[err], %[efault]			\n"
+		"	j	3b					\n"
+		"	.previous					\n"
+		"	.section __ex_table,\"a\"			\n"
+		"	"STR(PTR)"	1b, 4b				\n"
+		"	"STR(PTR)"	2b, 4b				\n"
+		"	.previous					\n"
+		"	.set	pop					\n"
+		: [old] "=&r" (old),
+		  [err] "=&r" (err),
+		  [tmp] "=&r" (tmp)
+		: [addr] "r" (addr),
+		  [new] "r" (new),
+		  [efault] "i" (-EFAULT)
+		: "memory");
+	} else if (cpu_has_llsc) {
+		__asm__ __volatile__ (
+		"	.set	push					\n"
+		"	.set	"MIPS_ISA_ARCH_LEVEL"			\n"
+		"	li	%[err], 0				\n"
+		"1:							\n"
+		"	" __SYNC(full, loongson3_war) "			\n"
+		user_ll("%[old]", "(%[addr])")
+		"	move	%[tmp], %[new]				\n"
+		"2:							\n"
+		user_sc("%[tmp]", "(%[addr])")
+		"	beqz	%[tmp], 1b				\n"
+		"3:							\n"
+		"	.insn						\n"
+		"	.section .fixup,\"ax\"				\n"
+		"5:	li	%[err], %[efault]			\n"
+		"	j	3b					\n"
+		"	.previous					\n"
+		"	.section __ex_table,\"a\"			\n"
+		"	"STR(PTR)"	1b, 5b				\n"
+		"	"STR(PTR)"	2b, 5b				\n"
+		"	.previous					\n"
+		"	.set	pop					\n"
+		: [old] "=&r" (old),
+		  [err] "=&r" (err),
+		  [tmp] "=&r" (tmp)
+		: [addr] "r" (addr),
+		  [new] "r" (new),
+		  [efault] "i" (-EFAULT)
+		: "memory");
+	} else {
+		do {
+			preempt_disable();
+			ll_bit = 1;
+			ll_task = current;
+			preempt_enable();
+
+			err = __get_user(old, (unsigned int *) addr);
+			err |= __put_user(new, (unsigned int *) addr);
+			if (err)
+				break;
+			rmb();
+		} while (!ll_bit);
+	}
+
+	if (unlikely(err))
+		return err;
+
+	regs = current_pt_regs();
+	regs->regs[2] = old;
+	regs->regs[7] = 0;	/* No error */
+
+	/*
+	 * Don't let your children do this ...
+	 */
+	__asm__ __volatile__(
+	"	move	$29, %0						\n"
+	"	j	syscall_exit					\n"
+	: /* no outputs */
+	: "r" (regs));
+
+	/* unreached.  Honestly.  */
+	unreachable();
+}
+
+/*
+ * mips_atomic_set() normally returns directly via syscall_exit potentially
+ * clobbering static registers, so be sure to preserve them.
+ */
+save_static_function(sys_sysmips);
+
+SYSCALL_DEFINE3(sysmips, long, cmd, long, arg1, long, arg2)
+{
+	switch (cmd) {
+	case MIPS_ATOMIC_SET:
+		return mips_atomic_set(arg1, arg2);
+
+	case MIPS_FIXADE:
+		if (arg1 & ~3)
+			return -EINVAL;
+
+		if (arg1 & 1)
+			set_thread_flag(TIF_FIXADE);
+		else
+			clear_thread_flag(TIF_FIXADE);
+		if (arg1 & 2)
+			set_thread_flag(TIF_LOGADE);
+		else
+			clear_thread_flag(TIF_LOGADE);
+
+		return 0;
+
+	case FLUSH_CACHE:
+		__flush_cache_all();
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
+/*
+ * No implemented yet ...
+ */
+SYSCALL_DEFINE3(cachectl, char *, addr, int, nbytes, int, op)
+{
+	return -ENOSYS;
+}
diff --git a/arch/mips/kernel/syscalls/Makefile b/arch/mips/kernel/syscalls/Makefile
new file mode 100644
index 000000000..6efb2f688
--- /dev/null
+++ b/arch/mips/kernel/syscalls/Makefile
@@ -0,0 +1,96 @@
+# SPDX-License-Identifier: GPL-2.0
+kapi := arch/$(SRCARCH)/include/generated/asm
+uapi := arch/$(SRCARCH)/include/generated/uapi/asm
+
+_dummy := $(shell [ -d '$(uapi)' ] || mkdir -p '$(uapi)')	\
+	  $(shell [ -d '$(kapi)' ] || mkdir -p '$(kapi)')
+
+syscalln32 := $(srctree)/$(src)/syscall_n32.tbl
+syscalln64 := $(srctree)/$(src)/syscall_n64.tbl
+syscallo32 := $(srctree)/$(src)/syscall_o32.tbl
+syshdr := $(srctree)/$(src)/syscallhdr.sh
+sysnr := $(srctree)/$(src)/syscallnr.sh
+systbl := $(srctree)/$(src)/syscalltbl.sh
+
+quiet_cmd_syshdr = SYSHDR  $@
+      cmd_syshdr = $(CONFIG_SHELL) '$(syshdr)' '$<' '$@'	\
+		   '$(syshdr_abis_$(basetarget))'		\
+		   '$(syshdr_pfx_$(basetarget))'		\
+		   '$(syshdr_offset_$(basetarget))'
+
+quiet_cmd_sysnr = SYSNR   $@
+      cmd_sysnr = $(CONFIG_SHELL) '$(sysnr)' '$<' '$@'		\
+		  '$(sysnr_abis_$(basetarget))'			\
+		  '$(sysnr_pfx_$(basetarget))'			\
+		  '$(sysnr_offset_$(basetarget))'
+
+quiet_cmd_systbl = SYSTBL  $@
+      cmd_systbl = $(CONFIG_SHELL) '$(systbl)' '$<' '$@'	\
+		   '$(systbl_abis_$(basetarget))'		\
+		   '$(systbl_abi_$(basetarget))'		\
+		   '$(systbl_offset_$(basetarget))'
+
+syshdr_offset_unistd_n32 := __NR_Linux
+$(uapi)/unistd_n32.h: $(syscalln32) $(syshdr)
+	$(call if_changed,syshdr)
+
+syshdr_offset_unistd_n64 := __NR_Linux
+$(uapi)/unistd_n64.h: $(syscalln64) $(syshdr)
+	$(call if_changed,syshdr)
+
+syshdr_offset_unistd_o32 := __NR_Linux
+$(uapi)/unistd_o32.h: $(syscallo32) $(syshdr)
+	$(call if_changed,syshdr)
+
+sysnr_pfx_unistd_nr_n32 := N32
+sysnr_offset_unistd_nr_n32 := 6000
+$(uapi)/unistd_nr_n32.h: $(syscalln32) $(sysnr)
+	$(call if_changed,sysnr)
+
+sysnr_pfx_unistd_nr_n64 := 64
+sysnr_offset_unistd_nr_n64 := 5000
+$(uapi)/unistd_nr_n64.h: $(syscalln64) $(sysnr)
+	$(call if_changed,sysnr)
+
+sysnr_pfx_unistd_nr_o32 := O32
+sysnr_offset_unistd_nr_o32 := 4000
+$(uapi)/unistd_nr_o32.h: $(syscallo32) $(sysnr)
+	$(call if_changed,sysnr)
+
+systbl_abi_syscall_table_32_o32 := 32_o32
+systbl_offset_syscall_table_32_o32 := 4000
+$(kapi)/syscall_table_32_o32.h: $(syscallo32) $(systbl)
+	$(call if_changed,systbl)
+
+systbl_abi_syscall_table_64_n32 := 64_n32
+systbl_offset_syscall_table_64_n32 := 6000
+$(kapi)/syscall_table_64_n32.h: $(syscalln32) $(systbl)
+	$(call if_changed,systbl)
+
+systbl_abi_syscall_table_64_n64 := 64_n64
+systbl_offset_syscall_table_64_n64 := 5000
+$(kapi)/syscall_table_64_n64.h: $(syscalln64) $(systbl)
+	$(call if_changed,systbl)
+
+systbl_abi_syscall_table_64_o32 := 64_o32
+systbl_offset_syscall_table_64_o32 := 4000
+$(kapi)/syscall_table_64_o32.h: $(syscallo32) $(systbl)
+	$(call if_changed,systbl)
+
+uapisyshdr-y		+= unistd_n32.h			\
+			   unistd_n64.h			\
+			   unistd_o32.h			\
+			   unistd_nr_n32.h		\
+			   unistd_nr_n64.h		\
+			   unistd_nr_o32.h
+kapisyshdr-y		+= syscall_table_32_o32.h	\
+			   syscall_table_64_n32.h	\
+			   syscall_table_64_n64.h	\
+			   syscall_table_64_o32.h
+
+targets	+= $(uapisyshdr-y) $(kapisyshdr-y)
+
+PHONY += all
+all: $(addprefix $(uapi)/,$(uapisyshdr-y))
+all: $(addprefix $(kapi)/,$(kapisyshdr-y))
+	@:
diff --git a/arch/mips/kernel/syscalls/syscall_n32.tbl b/arch/mips/kernel/syscalls/syscall_n32.tbl
new file mode 100644
index 000000000..32817c954
--- /dev/null
+++ b/arch/mips/kernel/syscalls/syscall_n32.tbl
@@ -0,0 +1,381 @@
+# SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note
+#
+# system call numbers and entry vectors for mips
+#
+# The format is:
+# <number> <abi> <name> <entry point> <compat entry point>
+#
+# The <abi> is always "n32" for this file.
+#
+0	n32	read				sys_read
+1	n32	write				sys_write
+2	n32	open				sys_open
+3	n32	close				sys_close
+4	n32	stat				sys_newstat
+5	n32	fstat				sys_newfstat
+6	n32	lstat				sys_newlstat
+7	n32	poll				sys_poll
+8	n32	lseek				sys_lseek
+9	n32	mmap				sys_mips_mmap
+10	n32	mprotect			sys_mprotect
+11	n32	munmap				sys_munmap
+12	n32	brk				sys_brk
+13	n32	rt_sigaction			compat_sys_rt_sigaction
+14	n32	rt_sigprocmask			compat_sys_rt_sigprocmask
+15	n32	ioctl				compat_sys_ioctl
+16	n32	pread64				sys_pread64
+17	n32	pwrite64			sys_pwrite64
+18	n32	readv				sys_readv
+19	n32	writev				sys_writev
+20	n32	access				sys_access
+21	n32	pipe				sysm_pipe
+22	n32	_newselect			compat_sys_select
+23	n32	sched_yield			sys_sched_yield
+24	n32	mremap				sys_mremap
+25	n32	msync				sys_msync
+26	n32	mincore				sys_mincore
+27	n32	madvise				sys_madvise
+28	n32	shmget				sys_shmget
+29	n32	shmat				sys_shmat
+30	n32	shmctl				compat_sys_old_shmctl
+31	n32	dup				sys_dup
+32	n32	dup2				sys_dup2
+33	n32	pause				sys_pause
+34	n32	nanosleep			sys_nanosleep_time32
+35	n32	getitimer			compat_sys_getitimer
+36	n32	setitimer			compat_sys_setitimer
+37	n32	alarm				sys_alarm
+38	n32	getpid				sys_getpid
+39	n32	sendfile			compat_sys_sendfile
+40	n32	socket				sys_socket
+41	n32	connect				sys_connect
+42	n32	accept				sys_accept
+43	n32	sendto				sys_sendto
+44	n32	recvfrom			compat_sys_recvfrom
+45	n32	sendmsg				compat_sys_sendmsg
+46	n32	recvmsg				compat_sys_recvmsg
+47	n32	shutdown			sys_shutdown
+48	n32	bind				sys_bind
+49	n32	listen				sys_listen
+50	n32	getsockname			sys_getsockname
+51	n32	getpeername			sys_getpeername
+52	n32	socketpair			sys_socketpair
+53	n32	setsockopt			sys_setsockopt
+54	n32	getsockopt			sys_getsockopt
+55	n32	clone				__sys_clone
+56	n32	fork				__sys_fork
+57	n32	execve				compat_sys_execve
+58	n32	exit				sys_exit
+59	n32	wait4				compat_sys_wait4
+60	n32	kill				sys_kill
+61	n32	uname				sys_newuname
+62	n32	semget				sys_semget
+63	n32	semop				sys_semop
+64	n32	semctl				compat_sys_old_semctl
+65	n32	shmdt				sys_shmdt
+66	n32	msgget				sys_msgget
+67	n32	msgsnd				compat_sys_msgsnd
+68	n32	msgrcv				compat_sys_msgrcv
+69	n32	msgctl				compat_sys_old_msgctl
+70	n32	fcntl				compat_sys_fcntl
+71	n32	flock				sys_flock
+72	n32	fsync				sys_fsync
+73	n32	fdatasync			sys_fdatasync
+74	n32	truncate			sys_truncate
+75	n32	ftruncate			sys_ftruncate
+76	n32	getdents			compat_sys_getdents
+77	n32	getcwd				sys_getcwd
+78	n32	chdir				sys_chdir
+79	n32	fchdir				sys_fchdir
+80	n32	rename				sys_rename
+81	n32	mkdir				sys_mkdir
+82	n32	rmdir				sys_rmdir
+83	n32	creat				sys_creat
+84	n32	link				sys_link
+85	n32	unlink				sys_unlink
+86	n32	symlink				sys_symlink
+87	n32	readlink			sys_readlink
+88	n32	chmod				sys_chmod
+89	n32	fchmod				sys_fchmod
+90	n32	chown				sys_chown
+91	n32	fchown				sys_fchown
+92	n32	lchown				sys_lchown
+93	n32	umask				sys_umask
+94	n32	gettimeofday			compat_sys_gettimeofday
+95	n32	getrlimit			compat_sys_getrlimit
+96	n32	getrusage			compat_sys_getrusage
+97	n32	sysinfo				compat_sys_sysinfo
+98	n32	times				compat_sys_times
+99	n32	ptrace				compat_sys_ptrace
+100	n32	getuid				sys_getuid
+101	n32	syslog				sys_syslog
+102	n32	getgid				sys_getgid
+103	n32	setuid				sys_setuid
+104	n32	setgid				sys_setgid
+105	n32	geteuid				sys_geteuid
+106	n32	getegid				sys_getegid
+107	n32	setpgid				sys_setpgid
+108	n32	getppid				sys_getppid
+109	n32	getpgrp				sys_getpgrp
+110	n32	setsid				sys_setsid
+111	n32	setreuid			sys_setreuid
+112	n32	setregid			sys_setregid
+113	n32	getgroups			sys_getgroups
+114	n32	setgroups			sys_setgroups
+115	n32	setresuid			sys_setresuid
+116	n32	getresuid			sys_getresuid
+117	n32	setresgid			sys_setresgid
+118	n32	getresgid			sys_getresgid
+119	n32	getpgid				sys_getpgid
+120	n32	setfsuid			sys_setfsuid
+121	n32	setfsgid			sys_setfsgid
+122	n32	getsid				sys_getsid
+123	n32	capget				sys_capget
+124	n32	capset				sys_capset
+125	n32	rt_sigpending			compat_sys_rt_sigpending
+126	n32	rt_sigtimedwait			compat_sys_rt_sigtimedwait_time32
+127	n32	rt_sigqueueinfo			compat_sys_rt_sigqueueinfo
+128	n32	rt_sigsuspend			compat_sys_rt_sigsuspend
+129	n32	sigaltstack			compat_sys_sigaltstack
+130	n32	utime				sys_utime32
+131	n32	mknod				sys_mknod
+132	n32	personality			sys_32_personality
+133	n32	ustat				compat_sys_ustat
+134	n32	statfs				compat_sys_statfs
+135	n32	fstatfs				compat_sys_fstatfs
+136	n32	sysfs				sys_sysfs
+137	n32	getpriority			sys_getpriority
+138	n32	setpriority			sys_setpriority
+139	n32	sched_setparam			sys_sched_setparam
+140	n32	sched_getparam			sys_sched_getparam
+141	n32	sched_setscheduler		sys_sched_setscheduler
+142	n32	sched_getscheduler		sys_sched_getscheduler
+143	n32	sched_get_priority_max		sys_sched_get_priority_max
+144	n32	sched_get_priority_min		sys_sched_get_priority_min
+145	n32	sched_rr_get_interval		sys_sched_rr_get_interval_time32
+146	n32	mlock				sys_mlock
+147	n32	munlock				sys_munlock
+148	n32	mlockall			sys_mlockall
+149	n32	munlockall			sys_munlockall
+150	n32	vhangup				sys_vhangup
+151	n32	pivot_root			sys_pivot_root
+152	n32	_sysctl				sys_ni_syscall
+153	n32	prctl				sys_prctl
+154	n32	adjtimex			sys_adjtimex_time32
+155	n32	setrlimit			compat_sys_setrlimit
+156	n32	chroot				sys_chroot
+157	n32	sync				sys_sync
+158	n32	acct				sys_acct
+159	n32	settimeofday			compat_sys_settimeofday
+160	n32	mount				sys_mount
+161	n32	umount2				sys_umount
+162	n32	swapon				sys_swapon
+163	n32	swapoff				sys_swapoff
+164	n32	reboot				sys_reboot
+165	n32	sethostname			sys_sethostname
+166	n32	setdomainname			sys_setdomainname
+167	n32	create_module			sys_ni_syscall
+168	n32	init_module			sys_init_module
+169	n32	delete_module			sys_delete_module
+170	n32	get_kernel_syms			sys_ni_syscall
+171	n32	query_module			sys_ni_syscall
+172	n32	quotactl			sys_quotactl
+173	n32	nfsservctl			sys_ni_syscall
+174	n32	getpmsg				sys_ni_syscall
+175	n32	putpmsg				sys_ni_syscall
+176	n32	afs_syscall			sys_ni_syscall
+# 177 reserved for security
+177	n32	reserved177			sys_ni_syscall
+178	n32	gettid				sys_gettid
+179	n32	readahead			sys_readahead
+180	n32	setxattr			sys_setxattr
+181	n32	lsetxattr			sys_lsetxattr
+182	n32	fsetxattr			sys_fsetxattr
+183	n32	getxattr			sys_getxattr
+184	n32	lgetxattr			sys_lgetxattr
+185	n32	fgetxattr			sys_fgetxattr
+186	n32	listxattr			sys_listxattr
+187	n32	llistxattr			sys_llistxattr
+188	n32	flistxattr			sys_flistxattr
+189	n32	removexattr			sys_removexattr
+190	n32	lremovexattr			sys_lremovexattr
+191	n32	fremovexattr			sys_fremovexattr
+192	n32	tkill				sys_tkill
+193	n32	reserved193			sys_ni_syscall
+194	n32	futex				sys_futex_time32
+195	n32	sched_setaffinity		compat_sys_sched_setaffinity
+196	n32	sched_getaffinity		compat_sys_sched_getaffinity
+197	n32	cacheflush			sys_cacheflush
+198	n32	cachectl			sys_cachectl
+199	n32	sysmips				__sys_sysmips
+200	n32	io_setup			compat_sys_io_setup
+201	n32	io_destroy			sys_io_destroy
+202	n32	io_getevents			sys_io_getevents_time32
+203	n32	io_submit			compat_sys_io_submit
+204	n32	io_cancel			sys_io_cancel
+205	n32	exit_group			sys_exit_group
+206	n32	lookup_dcookie			sys_lookup_dcookie
+207	n32	epoll_create			sys_epoll_create
+208	n32	epoll_ctl			sys_epoll_ctl
+209	n32	epoll_wait			sys_epoll_wait
+210	n32	remap_file_pages		sys_remap_file_pages
+211	n32	rt_sigreturn			sysn32_rt_sigreturn
+212	n32	fcntl64				compat_sys_fcntl64
+213	n32	set_tid_address			sys_set_tid_address
+214	n32	restart_syscall			sys_restart_syscall
+215	n32	semtimedop			sys_semtimedop_time32
+216	n32	fadvise64			sys_fadvise64_64
+217	n32	statfs64			compat_sys_statfs64
+218	n32	fstatfs64			compat_sys_fstatfs64
+219	n32	sendfile64			sys_sendfile64
+220	n32	timer_create			compat_sys_timer_create
+221	n32	timer_settime			sys_timer_settime32
+222	n32	timer_gettime			sys_timer_gettime32
+223	n32	timer_getoverrun		sys_timer_getoverrun
+224	n32	timer_delete			sys_timer_delete
+225	n32	clock_settime			sys_clock_settime32
+226	n32	clock_gettime			sys_clock_gettime32
+227	n32	clock_getres			sys_clock_getres_time32
+228	n32	clock_nanosleep			sys_clock_nanosleep_time32
+229	n32	tgkill				sys_tgkill
+230	n32	utimes				sys_utimes_time32
+231	n32	mbind				compat_sys_mbind
+232	n32	get_mempolicy			compat_sys_get_mempolicy
+233	n32	set_mempolicy			compat_sys_set_mempolicy
+234	n32	mq_open				compat_sys_mq_open
+235	n32	mq_unlink			sys_mq_unlink
+236	n32	mq_timedsend			sys_mq_timedsend_time32
+237	n32	mq_timedreceive			sys_mq_timedreceive_time32
+238	n32	mq_notify			compat_sys_mq_notify
+239	n32	mq_getsetattr			compat_sys_mq_getsetattr
+240	n32	vserver				sys_ni_syscall
+241	n32	waitid				compat_sys_waitid
+# 242 was sys_setaltroot
+243	n32	add_key				sys_add_key
+244	n32	request_key			sys_request_key
+245	n32	keyctl				compat_sys_keyctl
+246	n32	set_thread_area			sys_set_thread_area
+247	n32	inotify_init			sys_inotify_init
+248	n32	inotify_add_watch		sys_inotify_add_watch
+249	n32	inotify_rm_watch		sys_inotify_rm_watch
+250	n32	migrate_pages			compat_sys_migrate_pages
+251	n32	openat				sys_openat
+252	n32	mkdirat				sys_mkdirat
+253	n32	mknodat				sys_mknodat
+254	n32	fchownat			sys_fchownat
+255	n32	futimesat			sys_futimesat_time32
+256	n32	newfstatat			sys_newfstatat
+257	n32	unlinkat			sys_unlinkat
+258	n32	renameat			sys_renameat
+259	n32	linkat				sys_linkat
+260	n32	symlinkat			sys_symlinkat
+261	n32	readlinkat			sys_readlinkat
+262	n32	fchmodat			sys_fchmodat
+263	n32	faccessat			sys_faccessat
+264	n32	pselect6			compat_sys_pselect6_time32
+265	n32	ppoll				compat_sys_ppoll_time32
+266	n32	unshare				sys_unshare
+267	n32	splice				sys_splice
+268	n32	sync_file_range			sys_sync_file_range
+269	n32	tee				sys_tee
+270	n32	vmsplice			sys_vmsplice
+271	n32	move_pages			compat_sys_move_pages
+272	n32	set_robust_list			compat_sys_set_robust_list
+273	n32	get_robust_list			compat_sys_get_robust_list
+274	n32	kexec_load			compat_sys_kexec_load
+275	n32	getcpu				sys_getcpu
+276	n32	epoll_pwait			compat_sys_epoll_pwait
+277	n32	ioprio_set			sys_ioprio_set
+278	n32	ioprio_get			sys_ioprio_get
+279	n32	utimensat			sys_utimensat_time32
+280	n32	signalfd			compat_sys_signalfd
+281	n32	timerfd				sys_ni_syscall
+282	n32	eventfd				sys_eventfd
+283	n32	fallocate			sys_fallocate
+284	n32	timerfd_create			sys_timerfd_create
+285	n32	timerfd_gettime			sys_timerfd_gettime32
+286	n32	timerfd_settime			sys_timerfd_settime32
+287	n32	signalfd4			compat_sys_signalfd4
+288	n32	eventfd2			sys_eventfd2
+289	n32	epoll_create1			sys_epoll_create1
+290	n32	dup3				sys_dup3
+291	n32	pipe2				sys_pipe2
+292	n32	inotify_init1			sys_inotify_init1
+293	n32	preadv				compat_sys_preadv
+294	n32	pwritev				compat_sys_pwritev
+295	n32	rt_tgsigqueueinfo		compat_sys_rt_tgsigqueueinfo
+296	n32	perf_event_open			sys_perf_event_open
+297	n32	accept4				sys_accept4
+298	n32	recvmmsg			compat_sys_recvmmsg_time32
+299	n32	getdents64			sys_getdents64
+300	n32	fanotify_init			sys_fanotify_init
+301	n32	fanotify_mark			sys_fanotify_mark
+302	n32	prlimit64			sys_prlimit64
+303	n32	name_to_handle_at		sys_name_to_handle_at
+304	n32	open_by_handle_at		sys_open_by_handle_at
+305	n32	clock_adjtime			sys_clock_adjtime32
+306	n32	syncfs				sys_syncfs
+307	n32	sendmmsg			compat_sys_sendmmsg
+308	n32	setns				sys_setns
+309	n32	process_vm_readv		sys_process_vm_readv
+310	n32	process_vm_writev		sys_process_vm_writev
+311	n32	kcmp				sys_kcmp
+312	n32	finit_module			sys_finit_module
+313	n32	sched_setattr			sys_sched_setattr
+314	n32	sched_getattr			sys_sched_getattr
+315	n32	renameat2			sys_renameat2
+316	n32	seccomp				sys_seccomp
+317	n32	getrandom			sys_getrandom
+318	n32	memfd_create			sys_memfd_create
+319	n32	bpf				sys_bpf
+320	n32	execveat			compat_sys_execveat
+321	n32	userfaultfd			sys_userfaultfd
+322	n32	membarrier			sys_membarrier
+323	n32	mlock2				sys_mlock2
+324	n32	copy_file_range			sys_copy_file_range
+325	n32	preadv2				compat_sys_preadv2
+326	n32	pwritev2			compat_sys_pwritev2
+327	n32	pkey_mprotect			sys_pkey_mprotect
+328	n32	pkey_alloc			sys_pkey_alloc
+329	n32	pkey_free			sys_pkey_free
+330	n32	statx				sys_statx
+331	n32	rseq				sys_rseq
+332	n32	io_pgetevents			compat_sys_io_pgetevents
+# 333 through 402 are unassigned to sync up with generic numbers
+403	n32	clock_gettime64			sys_clock_gettime
+404	n32	clock_settime64			sys_clock_settime
+405	n32	clock_adjtime64			sys_clock_adjtime
+406	n32	clock_getres_time64		sys_clock_getres
+407	n32	clock_nanosleep_time64		sys_clock_nanosleep
+408	n32	timer_gettime64			sys_timer_gettime
+409	n32	timer_settime64			sys_timer_settime
+410	n32	timerfd_gettime64		sys_timerfd_gettime
+411	n32	timerfd_settime64		sys_timerfd_settime
+412	n32	utimensat_time64		sys_utimensat
+413	n32	pselect6_time64			compat_sys_pselect6_time64
+414	n32	ppoll_time64			compat_sys_ppoll_time64
+416	n32	io_pgetevents_time64		sys_io_pgetevents
+417	n32	recvmmsg_time64			compat_sys_recvmmsg_time64
+418	n32	mq_timedsend_time64		sys_mq_timedsend
+419	n32	mq_timedreceive_time64		sys_mq_timedreceive
+420	n32	semtimedop_time64		sys_semtimedop
+421	n32	rt_sigtimedwait_time64		compat_sys_rt_sigtimedwait_time64
+422	n32	futex_time64			sys_futex
+423	n32	sched_rr_get_interval_time64	sys_sched_rr_get_interval
+424	n32	pidfd_send_signal		sys_pidfd_send_signal
+425	n32	io_uring_setup			sys_io_uring_setup
+426	n32	io_uring_enter			sys_io_uring_enter
+427	n32	io_uring_register		sys_io_uring_register
+428	n32	open_tree			sys_open_tree
+429	n32	move_mount			sys_move_mount
+430	n32	fsopen				sys_fsopen
+431	n32	fsconfig			sys_fsconfig
+432	n32	fsmount				sys_fsmount
+433	n32	fspick				sys_fspick
+434	n32	pidfd_open			sys_pidfd_open
+435	n32	clone3				__sys_clone3
+436	n32	close_range			sys_close_range
+437	n32	openat2				sys_openat2
+438	n32	pidfd_getfd			sys_pidfd_getfd
+439	n32	faccessat2			sys_faccessat2
+440	n32	process_madvise			sys_process_madvise
diff --git a/arch/mips/kernel/syscalls/syscall_n64.tbl b/arch/mips/kernel/syscalls/syscall_n64.tbl
new file mode 100644
index 000000000..9e4ea3c31
--- /dev/null
+++ b/arch/mips/kernel/syscalls/syscall_n64.tbl
@@ -0,0 +1,357 @@
+# SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note
+#
+# system call numbers and entry vectors for mips
+#
+# The format is:
+# <number> <abi> <name> <entry point>
+#
+# The <abi> is always "n64" for this file.
+#
+0	n64	read				sys_read
+1	n64	write				sys_write
+2	n64	open				sys_open
+3	n64	close				sys_close
+4	n64	stat				sys_newstat
+5	n64	fstat				sys_newfstat
+6	n64	lstat				sys_newlstat
+7	n64	poll				sys_poll
+8	n64	lseek				sys_lseek
+9	n64	mmap				sys_mips_mmap
+10	n64	mprotect			sys_mprotect
+11	n64	munmap				sys_munmap
+12	n64	brk				sys_brk
+13	n64	rt_sigaction			sys_rt_sigaction
+14	n64	rt_sigprocmask			sys_rt_sigprocmask
+15	n64	ioctl				sys_ioctl
+16	n64	pread64				sys_pread64
+17	n64	pwrite64			sys_pwrite64
+18	n64	readv				sys_readv
+19	n64	writev				sys_writev
+20	n64	access				sys_access
+21	n64	pipe				sysm_pipe
+22	n64	_newselect			sys_select
+23	n64	sched_yield			sys_sched_yield
+24	n64	mremap				sys_mremap
+25	n64	msync				sys_msync
+26	n64	mincore				sys_mincore
+27	n64	madvise				sys_madvise
+28	n64	shmget				sys_shmget
+29	n64	shmat				sys_shmat
+30	n64	shmctl				sys_old_shmctl
+31	n64	dup				sys_dup
+32	n64	dup2				sys_dup2
+33	n64	pause				sys_pause
+34	n64	nanosleep			sys_nanosleep
+35	n64	getitimer			sys_getitimer
+36	n64	setitimer			sys_setitimer
+37	n64	alarm				sys_alarm
+38	n64	getpid				sys_getpid
+39	n64	sendfile			sys_sendfile64
+40	n64	socket				sys_socket
+41	n64	connect				sys_connect
+42	n64	accept				sys_accept
+43	n64	sendto				sys_sendto
+44	n64	recvfrom			sys_recvfrom
+45	n64	sendmsg				sys_sendmsg
+46	n64	recvmsg				sys_recvmsg
+47	n64	shutdown			sys_shutdown
+48	n64	bind				sys_bind
+49	n64	listen				sys_listen
+50	n64	getsockname			sys_getsockname
+51	n64	getpeername			sys_getpeername
+52	n64	socketpair			sys_socketpair
+53	n64	setsockopt			sys_setsockopt
+54	n64	getsockopt			sys_getsockopt
+55	n64	clone				__sys_clone
+56	n64	fork				__sys_fork
+57	n64	execve				sys_execve
+58	n64	exit				sys_exit
+59	n64	wait4				sys_wait4
+60	n64	kill				sys_kill
+61	n64	uname				sys_newuname
+62	n64	semget				sys_semget
+63	n64	semop				sys_semop
+64	n64	semctl				sys_old_semctl
+65	n64	shmdt				sys_shmdt
+66	n64	msgget				sys_msgget
+67	n64	msgsnd				sys_msgsnd
+68	n64	msgrcv				sys_msgrcv
+69	n64	msgctl				sys_old_msgctl
+70	n64	fcntl				sys_fcntl
+71	n64	flock				sys_flock
+72	n64	fsync				sys_fsync
+73	n64	fdatasync			sys_fdatasync
+74	n64	truncate			sys_truncate
+75	n64	ftruncate			sys_ftruncate
+76	n64	getdents			sys_getdents
+77	n64	getcwd				sys_getcwd
+78	n64	chdir				sys_chdir
+79	n64	fchdir				sys_fchdir
+80	n64	rename				sys_rename
+81	n64	mkdir				sys_mkdir
+82	n64	rmdir				sys_rmdir
+83	n64	creat				sys_creat
+84	n64	link				sys_link
+85	n64	unlink				sys_unlink
+86	n64	symlink				sys_symlink
+87	n64	readlink			sys_readlink
+88	n64	chmod				sys_chmod
+89	n64	fchmod				sys_fchmod
+90	n64	chown				sys_chown
+91	n64	fchown				sys_fchown
+92	n64	lchown				sys_lchown
+93	n64	umask				sys_umask
+94	n64	gettimeofday			sys_gettimeofday
+95	n64	getrlimit			sys_getrlimit
+96	n64	getrusage			sys_getrusage
+97	n64	sysinfo				sys_sysinfo
+98	n64	times				sys_times
+99	n64	ptrace				sys_ptrace
+100	n64	getuid				sys_getuid
+101	n64	syslog				sys_syslog
+102	n64	getgid				sys_getgid
+103	n64	setuid				sys_setuid
+104	n64	setgid				sys_setgid
+105	n64	geteuid				sys_geteuid
+106	n64	getegid				sys_getegid
+107	n64	setpgid				sys_setpgid
+108	n64	getppid				sys_getppid
+109	n64	getpgrp				sys_getpgrp
+110	n64	setsid				sys_setsid
+111	n64	setreuid			sys_setreuid
+112	n64	setregid			sys_setregid
+113	n64	getgroups			sys_getgroups
+114	n64	setgroups			sys_setgroups
+115	n64	setresuid			sys_setresuid
+116	n64	getresuid			sys_getresuid
+117	n64	setresgid			sys_setresgid
+118	n64	getresgid			sys_getresgid
+119	n64	getpgid				sys_getpgid
+120	n64	setfsuid			sys_setfsuid
+121	n64	setfsgid			sys_setfsgid
+122	n64	getsid				sys_getsid
+123	n64	capget				sys_capget
+124	n64	capset				sys_capset
+125	n64	rt_sigpending			sys_rt_sigpending
+126	n64	rt_sigtimedwait			sys_rt_sigtimedwait
+127	n64	rt_sigqueueinfo			sys_rt_sigqueueinfo
+128	n64	rt_sigsuspend			sys_rt_sigsuspend
+129	n64	sigaltstack			sys_sigaltstack
+130	n64	utime				sys_utime
+131	n64	mknod				sys_mknod
+132	n64	personality			sys_personality
+133	n64	ustat				sys_ustat
+134	n64	statfs				sys_statfs
+135	n64	fstatfs				sys_fstatfs
+136	n64	sysfs				sys_sysfs
+137	n64	getpriority			sys_getpriority
+138	n64	setpriority			sys_setpriority
+139	n64	sched_setparam			sys_sched_setparam
+140	n64	sched_getparam			sys_sched_getparam
+141	n64	sched_setscheduler		sys_sched_setscheduler
+142	n64	sched_getscheduler		sys_sched_getscheduler
+143	n64	sched_get_priority_max		sys_sched_get_priority_max
+144	n64	sched_get_priority_min		sys_sched_get_priority_min
+145	n64	sched_rr_get_interval		sys_sched_rr_get_interval
+146	n64	mlock				sys_mlock
+147	n64	munlock				sys_munlock
+148	n64	mlockall			sys_mlockall
+149	n64	munlockall			sys_munlockall
+150	n64	vhangup				sys_vhangup
+151	n64	pivot_root			sys_pivot_root
+152	n64	_sysctl				sys_ni_syscall
+153	n64	prctl				sys_prctl
+154	n64	adjtimex			sys_adjtimex
+155	n64	setrlimit			sys_setrlimit
+156	n64	chroot				sys_chroot
+157	n64	sync				sys_sync
+158	n64	acct				sys_acct
+159	n64	settimeofday			sys_settimeofday
+160	n64	mount				sys_mount
+161	n64	umount2				sys_umount
+162	n64	swapon				sys_swapon
+163	n64	swapoff				sys_swapoff
+164	n64	reboot				sys_reboot
+165	n64	sethostname			sys_sethostname
+166	n64	setdomainname			sys_setdomainname
+167	n64	create_module			sys_ni_syscall
+168	n64	init_module			sys_init_module
+169	n64	delete_module			sys_delete_module
+170	n64	get_kernel_syms			sys_ni_syscall
+171	n64	query_module			sys_ni_syscall
+172	n64	quotactl			sys_quotactl
+173	n64	nfsservctl			sys_ni_syscall
+174	n64	getpmsg				sys_ni_syscall
+175	n64	putpmsg				sys_ni_syscall
+176	n64	afs_syscall			sys_ni_syscall
+# 177 reserved for security
+177	n64	reserved177			sys_ni_syscall
+178	n64	gettid				sys_gettid
+179	n64	readahead			sys_readahead
+180	n64	setxattr			sys_setxattr
+181	n64	lsetxattr			sys_lsetxattr
+182	n64	fsetxattr			sys_fsetxattr
+183	n64	getxattr			sys_getxattr
+184	n64	lgetxattr			sys_lgetxattr
+185	n64	fgetxattr			sys_fgetxattr
+186	n64	listxattr			sys_listxattr
+187	n64	llistxattr			sys_llistxattr
+188	n64	flistxattr			sys_flistxattr
+189	n64	removexattr			sys_removexattr
+190	n64	lremovexattr			sys_lremovexattr
+191	n64	fremovexattr			sys_fremovexattr
+192	n64	tkill				sys_tkill
+193	n64	reserved193			sys_ni_syscall
+194	n64	futex				sys_futex
+195	n64	sched_setaffinity		sys_sched_setaffinity
+196	n64	sched_getaffinity		sys_sched_getaffinity
+197	n64	cacheflush			sys_cacheflush
+198	n64	cachectl			sys_cachectl
+199	n64	sysmips				__sys_sysmips
+200	n64	io_setup			sys_io_setup
+201	n64	io_destroy			sys_io_destroy
+202	n64	io_getevents			sys_io_getevents
+203	n64	io_submit			sys_io_submit
+204	n64	io_cancel			sys_io_cancel
+205	n64	exit_group			sys_exit_group
+206	n64	lookup_dcookie			sys_lookup_dcookie
+207	n64	epoll_create			sys_epoll_create
+208	n64	epoll_ctl			sys_epoll_ctl
+209	n64	epoll_wait			sys_epoll_wait
+210	n64	remap_file_pages		sys_remap_file_pages
+211	n64	rt_sigreturn			sys_rt_sigreturn
+212	n64	set_tid_address			sys_set_tid_address
+213	n64	restart_syscall			sys_restart_syscall
+214	n64	semtimedop			sys_semtimedop
+215	n64	fadvise64			sys_fadvise64_64
+216	n64	timer_create			sys_timer_create
+217	n64	timer_settime			sys_timer_settime
+218	n64	timer_gettime			sys_timer_gettime
+219	n64	timer_getoverrun		sys_timer_getoverrun
+220	n64	timer_delete			sys_timer_delete
+221	n64	clock_settime			sys_clock_settime
+222	n64	clock_gettime			sys_clock_gettime
+223	n64	clock_getres			sys_clock_getres
+224	n64	clock_nanosleep			sys_clock_nanosleep
+225	n64	tgkill				sys_tgkill
+226	n64	utimes				sys_utimes
+227	n64	mbind				sys_mbind
+228	n64	get_mempolicy			sys_get_mempolicy
+229	n64	set_mempolicy			sys_set_mempolicy
+230	n64	mq_open				sys_mq_open
+231	n64	mq_unlink			sys_mq_unlink
+232	n64	mq_timedsend			sys_mq_timedsend
+233	n64	mq_timedreceive			sys_mq_timedreceive
+234	n64	mq_notify			sys_mq_notify
+235	n64	mq_getsetattr			sys_mq_getsetattr
+236	n64	vserver				sys_ni_syscall
+237	n64	waitid				sys_waitid
+# 238 was sys_setaltroot
+239	n64	add_key				sys_add_key
+240	n64	request_key			sys_request_key
+241	n64	keyctl				sys_keyctl
+242	n64	set_thread_area			sys_set_thread_area
+243	n64	inotify_init			sys_inotify_init
+244	n64	inotify_add_watch		sys_inotify_add_watch
+245	n64	inotify_rm_watch		sys_inotify_rm_watch
+246	n64	migrate_pages			sys_migrate_pages
+247	n64	openat				sys_openat
+248	n64	mkdirat				sys_mkdirat
+249	n64	mknodat				sys_mknodat
+250	n64	fchownat			sys_fchownat
+251	n64	futimesat			sys_futimesat
+252	n64	newfstatat			sys_newfstatat
+253	n64	unlinkat			sys_unlinkat
+254	n64	renameat			sys_renameat
+255	n64	linkat				sys_linkat
+256	n64	symlinkat			sys_symlinkat
+257	n64	readlinkat			sys_readlinkat
+258	n64	fchmodat			sys_fchmodat
+259	n64	faccessat			sys_faccessat
+260	n64	pselect6			sys_pselect6
+261	n64	ppoll				sys_ppoll
+262	n64	unshare				sys_unshare
+263	n64	splice				sys_splice
+264	n64	sync_file_range			sys_sync_file_range
+265	n64	tee				sys_tee
+266	n64	vmsplice			sys_vmsplice
+267	n64	move_pages			sys_move_pages
+268	n64	set_robust_list			sys_set_robust_list
+269	n64	get_robust_list			sys_get_robust_list
+270	n64	kexec_load			sys_kexec_load
+271	n64	getcpu				sys_getcpu
+272	n64	epoll_pwait			sys_epoll_pwait
+273	n64	ioprio_set			sys_ioprio_set
+274	n64	ioprio_get			sys_ioprio_get
+275	n64	utimensat			sys_utimensat
+276	n64	signalfd			sys_signalfd
+277	n64	timerfd				sys_ni_syscall
+278	n64	eventfd				sys_eventfd
+279	n64	fallocate			sys_fallocate
+280	n64	timerfd_create			sys_timerfd_create
+281	n64	timerfd_gettime			sys_timerfd_gettime
+282	n64	timerfd_settime			sys_timerfd_settime
+283	n64	signalfd4			sys_signalfd4
+284	n64	eventfd2			sys_eventfd2
+285	n64	epoll_create1			sys_epoll_create1
+286	n64	dup3				sys_dup3
+287	n64	pipe2				sys_pipe2
+288	n64	inotify_init1			sys_inotify_init1
+289	n64	preadv				sys_preadv
+290	n64	pwritev				sys_pwritev
+291	n64	rt_tgsigqueueinfo		sys_rt_tgsigqueueinfo
+292	n64	perf_event_open			sys_perf_event_open
+293	n64	accept4				sys_accept4
+294	n64	recvmmsg			sys_recvmmsg
+295	n64	fanotify_init			sys_fanotify_init
+296	n64	fanotify_mark			sys_fanotify_mark
+297	n64	prlimit64			sys_prlimit64
+298	n64	name_to_handle_at		sys_name_to_handle_at
+299	n64	open_by_handle_at		sys_open_by_handle_at
+300	n64	clock_adjtime			sys_clock_adjtime
+301	n64	syncfs				sys_syncfs
+302	n64	sendmmsg			sys_sendmmsg
+303	n64	setns				sys_setns
+304	n64	process_vm_readv		sys_process_vm_readv
+305	n64	process_vm_writev		sys_process_vm_writev
+306	n64	kcmp				sys_kcmp
+307	n64	finit_module			sys_finit_module
+308	n64	getdents64			sys_getdents64
+309	n64	sched_setattr			sys_sched_setattr
+310	n64	sched_getattr			sys_sched_getattr
+311	n64	renameat2			sys_renameat2
+312	n64	seccomp				sys_seccomp
+313	n64	getrandom			sys_getrandom
+314	n64	memfd_create			sys_memfd_create
+315	n64	bpf				sys_bpf
+316	n64	execveat			sys_execveat
+317	n64	userfaultfd			sys_userfaultfd
+318	n64	membarrier			sys_membarrier
+319	n64	mlock2				sys_mlock2
+320	n64	copy_file_range			sys_copy_file_range
+321	n64	preadv2				sys_preadv2
+322	n64	pwritev2			sys_pwritev2
+323	n64	pkey_mprotect			sys_pkey_mprotect
+324	n64	pkey_alloc			sys_pkey_alloc
+325	n64	pkey_free			sys_pkey_free
+326	n64	statx				sys_statx
+327	n64	rseq				sys_rseq
+328	n64	io_pgetevents			sys_io_pgetevents
+# 329 through 423 are reserved to sync up with other architectures
+424	n64	pidfd_send_signal		sys_pidfd_send_signal
+425	n64	io_uring_setup			sys_io_uring_setup
+426	n64	io_uring_enter			sys_io_uring_enter
+427	n64	io_uring_register		sys_io_uring_register
+428	n64	open_tree			sys_open_tree
+429	n64	move_mount			sys_move_mount
+430	n64	fsopen				sys_fsopen
+431	n64	fsconfig			sys_fsconfig
+432	n64	fsmount				sys_fsmount
+433	n64	fspick				sys_fspick
+434	n64	pidfd_open			sys_pidfd_open
+435	n64	clone3				__sys_clone3
+436	n64	close_range			sys_close_range
+437	n64	openat2				sys_openat2
+438	n64	pidfd_getfd			sys_pidfd_getfd
+439	n64	faccessat2			sys_faccessat2
+440	n64	process_madvise			sys_process_madvise
diff --git a/arch/mips/kernel/syscalls/syscall_o32.tbl b/arch/mips/kernel/syscalls/syscall_o32.tbl
new file mode 100644
index 000000000..29f5f28cf
--- /dev/null
+++ b/arch/mips/kernel/syscalls/syscall_o32.tbl
@@ -0,0 +1,430 @@
+# SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note
+#
+# system call numbers and entry vectors for mips
+#
+# The format is:
+# <number> <abi> <name> <entry point> <compat entry point>
+#
+# The <abi> is always "o32" for this file.
+#
+0	o32	syscall				sys_syscall			sys32_syscall
+1	o32	exit				sys_exit
+2	o32	fork				__sys_fork
+3	o32	read				sys_read
+4	o32	write				sys_write
+5	o32	open				sys_open			compat_sys_open
+6	o32	close				sys_close
+7	o32	waitpid				sys_waitpid
+8	o32	creat				sys_creat
+9	o32	link				sys_link
+10	o32	unlink				sys_unlink
+11	o32	execve				sys_execve			compat_sys_execve
+12	o32	chdir				sys_chdir
+13	o32	time				sys_time32
+14	o32	mknod				sys_mknod
+15	o32	chmod				sys_chmod
+16	o32	lchown				sys_lchown
+17	o32	break				sys_ni_syscall
+# 18 was sys_stat
+18	o32	unused18			sys_ni_syscall
+19	o32	lseek				sys_lseek
+20	o32	getpid				sys_getpid
+21	o32	mount				sys_mount
+22	o32	umount				sys_oldumount
+23	o32	setuid				sys_setuid
+24	o32	getuid				sys_getuid
+25	o32	stime				sys_stime32
+26	o32	ptrace				sys_ptrace			compat_sys_ptrace
+27	o32	alarm				sys_alarm
+# 28 was sys_fstat
+28	o32	unused28			sys_ni_syscall
+29	o32	pause				sys_pause
+30	o32	utime				sys_utime32
+31	o32	stty				sys_ni_syscall
+32	o32	gtty				sys_ni_syscall
+33	o32	access				sys_access
+34	o32	nice				sys_nice
+35	o32	ftime				sys_ni_syscall
+36	o32	sync				sys_sync
+37	o32	kill				sys_kill
+38	o32	rename				sys_rename
+39	o32	mkdir				sys_mkdir
+40	o32	rmdir				sys_rmdir
+41	o32	dup				sys_dup
+42	o32	pipe				sysm_pipe
+43	o32	times				sys_times			compat_sys_times
+44	o32	prof				sys_ni_syscall
+45	o32	brk				sys_brk
+46	o32	setgid				sys_setgid
+47	o32	getgid				sys_getgid
+48	o32	signal				sys_ni_syscall
+49	o32	geteuid				sys_geteuid
+50	o32	getegid				sys_getegid
+51	o32	acct				sys_acct
+52	o32	umount2				sys_umount
+53	o32	lock				sys_ni_syscall
+54	o32	ioctl				sys_ioctl			compat_sys_ioctl
+55	o32	fcntl				sys_fcntl			compat_sys_fcntl
+56	o32	mpx				sys_ni_syscall
+57	o32	setpgid				sys_setpgid
+58	o32	ulimit				sys_ni_syscall
+59	o32	unused59			sys_olduname
+60	o32	umask				sys_umask
+61	o32	chroot				sys_chroot
+62	o32	ustat				sys_ustat			compat_sys_ustat
+63	o32	dup2				sys_dup2
+64	o32	getppid				sys_getppid
+65	o32	getpgrp				sys_getpgrp
+66	o32	setsid				sys_setsid
+67	o32	sigaction			sys_sigaction			sys_32_sigaction
+68	o32	sgetmask			sys_sgetmask
+69	o32	ssetmask			sys_ssetmask
+70	o32	setreuid			sys_setreuid
+71	o32	setregid			sys_setregid
+72	o32	sigsuspend			sys_sigsuspend			sys32_sigsuspend
+73	o32	sigpending			sys_sigpending			compat_sys_sigpending
+74	o32	sethostname			sys_sethostname
+75	o32	setrlimit			sys_setrlimit			compat_sys_setrlimit
+76	o32	getrlimit			sys_getrlimit			compat_sys_getrlimit
+77	o32	getrusage			sys_getrusage			compat_sys_getrusage
+78	o32	gettimeofday			sys_gettimeofday		compat_sys_gettimeofday
+79	o32	settimeofday			sys_settimeofday		compat_sys_settimeofday
+80	o32	getgroups			sys_getgroups
+81	o32	setgroups			sys_setgroups
+# 82 was old_select
+82	o32	reserved82			sys_ni_syscall
+83	o32	symlink				sys_symlink
+# 84 was sys_lstat
+84	o32	unused84			sys_ni_syscall
+85	o32	readlink			sys_readlink
+86	o32	uselib				sys_uselib
+87	o32	swapon				sys_swapon
+88	o32	reboot				sys_reboot
+89	o32	readdir				sys_old_readdir			compat_sys_old_readdir
+90	o32	mmap				sys_mips_mmap
+91	o32	munmap				sys_munmap
+92	o32	truncate			sys_truncate			compat_sys_truncate
+93	o32	ftruncate			sys_ftruncate			compat_sys_ftruncate
+94	o32	fchmod				sys_fchmod
+95	o32	fchown				sys_fchown
+96	o32	getpriority			sys_getpriority
+97	o32	setpriority			sys_setpriority
+98	o32	profil				sys_ni_syscall
+99	o32	statfs				sys_statfs			compat_sys_statfs
+100	o32	fstatfs				sys_fstatfs			compat_sys_fstatfs
+101	o32	ioperm				sys_ni_syscall
+102	o32	socketcall			sys_socketcall			compat_sys_socketcall
+103	o32	syslog				sys_syslog
+104	o32	setitimer			sys_setitimer			compat_sys_setitimer
+105	o32	getitimer			sys_getitimer			compat_sys_getitimer
+106	o32	stat				sys_newstat			compat_sys_newstat
+107	o32	lstat				sys_newlstat			compat_sys_newlstat
+108	o32	fstat				sys_newfstat			compat_sys_newfstat
+109	o32	unused109			sys_uname
+110	o32	iopl				sys_ni_syscall
+111	o32	vhangup				sys_vhangup
+112	o32	idle				sys_ni_syscall
+113	o32	vm86				sys_ni_syscall
+114	o32	wait4				sys_wait4			compat_sys_wait4
+115	o32	swapoff				sys_swapoff
+116	o32	sysinfo				sys_sysinfo			compat_sys_sysinfo
+117	o32	ipc				sys_ipc				compat_sys_ipc
+118	o32	fsync				sys_fsync
+119	o32	sigreturn			sys_sigreturn			sys32_sigreturn
+120	o32	clone				__sys_clone
+121	o32	setdomainname			sys_setdomainname
+122	o32	uname				sys_newuname
+123	o32	modify_ldt			sys_ni_syscall
+124	o32	adjtimex			sys_adjtimex_time32
+125	o32	mprotect			sys_mprotect
+126	o32	sigprocmask			sys_sigprocmask			compat_sys_sigprocmask
+127	o32	create_module			sys_ni_syscall
+128	o32	init_module			sys_init_module
+129	o32	delete_module			sys_delete_module
+130	o32	get_kernel_syms			sys_ni_syscall
+131	o32	quotactl			sys_quotactl
+132	o32	getpgid				sys_getpgid
+133	o32	fchdir				sys_fchdir
+134	o32	bdflush				sys_bdflush
+135	o32	sysfs				sys_sysfs
+136	o32	personality			sys_personality			sys_32_personality
+137	o32	afs_syscall			sys_ni_syscall
+138	o32	setfsuid			sys_setfsuid
+139	o32	setfsgid			sys_setfsgid
+140	o32	_llseek				sys_llseek			sys_32_llseek
+141	o32	getdents			sys_getdents			compat_sys_getdents
+142	o32	_newselect			sys_select			compat_sys_select
+143	o32	flock				sys_flock
+144	o32	msync				sys_msync
+145	o32	readv				sys_readv
+146	o32	writev				sys_writev
+147	o32	cacheflush			sys_cacheflush
+148	o32	cachectl			sys_cachectl
+149	o32	sysmips				__sys_sysmips
+150	o32	unused150			sys_ni_syscall
+151	o32	getsid				sys_getsid
+152	o32	fdatasync			sys_fdatasync
+153	o32	_sysctl				sys_ni_syscall
+154	o32	mlock				sys_mlock
+155	o32	munlock				sys_munlock
+156	o32	mlockall			sys_mlockall
+157	o32	munlockall			sys_munlockall
+158	o32	sched_setparam			sys_sched_setparam
+159	o32	sched_getparam			sys_sched_getparam
+160	o32	sched_setscheduler		sys_sched_setscheduler
+161	o32	sched_getscheduler		sys_sched_getscheduler
+162	o32	sched_yield			sys_sched_yield
+163	o32	sched_get_priority_max		sys_sched_get_priority_max
+164	o32	sched_get_priority_min		sys_sched_get_priority_min
+165	o32	sched_rr_get_interval		sys_sched_rr_get_interval_time32
+166	o32	nanosleep			sys_nanosleep_time32
+167	o32	mremap				sys_mremap
+168	o32	accept				sys_accept
+169	o32	bind				sys_bind
+170	o32	connect				sys_connect
+171	o32	getpeername			sys_getpeername
+172	o32	getsockname			sys_getsockname
+173	o32	getsockopt			sys_getsockopt			sys_getsockopt
+174	o32	listen				sys_listen
+175	o32	recv				sys_recv			compat_sys_recv
+176	o32	recvfrom			sys_recvfrom			compat_sys_recvfrom
+177	o32	recvmsg				sys_recvmsg			compat_sys_recvmsg
+178	o32	send				sys_send
+179	o32	sendmsg				sys_sendmsg			compat_sys_sendmsg
+180	o32	sendto				sys_sendto
+181	o32	setsockopt			sys_setsockopt			sys_setsockopt
+182	o32	shutdown			sys_shutdown
+183	o32	socket				sys_socket
+184	o32	socketpair			sys_socketpair
+185	o32	setresuid			sys_setresuid
+186	o32	getresuid			sys_getresuid
+187	o32	query_module			sys_ni_syscall
+188	o32	poll				sys_poll
+189	o32	nfsservctl			sys_ni_syscall
+190	o32	setresgid			sys_setresgid
+191	o32	getresgid			sys_getresgid
+192	o32	prctl				sys_prctl
+193	o32	rt_sigreturn			sys_rt_sigreturn		sys32_rt_sigreturn
+194	o32	rt_sigaction			sys_rt_sigaction		compat_sys_rt_sigaction
+195	o32	rt_sigprocmask			sys_rt_sigprocmask		compat_sys_rt_sigprocmask
+196	o32	rt_sigpending			sys_rt_sigpending		compat_sys_rt_sigpending
+197	o32	rt_sigtimedwait			sys_rt_sigtimedwait_time32	compat_sys_rt_sigtimedwait_time32
+198	o32	rt_sigqueueinfo			sys_rt_sigqueueinfo		compat_sys_rt_sigqueueinfo
+199	o32	rt_sigsuspend			sys_rt_sigsuspend		compat_sys_rt_sigsuspend
+200	o32	pread64				sys_pread64			sys_32_pread
+201	o32	pwrite64			sys_pwrite64			sys_32_pwrite
+202	o32	chown				sys_chown
+203	o32	getcwd				sys_getcwd
+204	o32	capget				sys_capget
+205	o32	capset				sys_capset
+206	o32	sigaltstack			sys_sigaltstack			compat_sys_sigaltstack
+207	o32	sendfile			sys_sendfile			compat_sys_sendfile
+208	o32	getpmsg				sys_ni_syscall
+209	o32	putpmsg				sys_ni_syscall
+210	o32	mmap2				sys_mips_mmap2
+211	o32	truncate64			sys_truncate64			sys_32_truncate64
+212	o32	ftruncate64			sys_ftruncate64			sys_32_ftruncate64
+213	o32	stat64				sys_stat64			sys_newstat
+214	o32	lstat64				sys_lstat64			sys_newlstat
+215	o32	fstat64				sys_fstat64			sys_newfstat
+216	o32	pivot_root			sys_pivot_root
+217	o32	mincore				sys_mincore
+218	o32	madvise				sys_madvise
+219	o32	getdents64			sys_getdents64
+220	o32	fcntl64				sys_fcntl64			compat_sys_fcntl64
+221	o32	reserved221			sys_ni_syscall
+222	o32	gettid				sys_gettid
+223	o32	readahead			sys_readahead			sys32_readahead
+224	o32	setxattr			sys_setxattr
+225	o32	lsetxattr			sys_lsetxattr
+226	o32	fsetxattr			sys_fsetxattr
+227	o32	getxattr			sys_getxattr
+228	o32	lgetxattr			sys_lgetxattr
+229	o32	fgetxattr			sys_fgetxattr
+230	o32	listxattr			sys_listxattr
+231	o32	llistxattr			sys_llistxattr
+232	o32	flistxattr			sys_flistxattr
+233	o32	removexattr			sys_removexattr
+234	o32	lremovexattr			sys_lremovexattr
+235	o32	fremovexattr			sys_fremovexattr
+236	o32	tkill				sys_tkill
+237	o32	sendfile64			sys_sendfile64
+238	o32	futex				sys_futex_time32
+239	o32	sched_setaffinity		sys_sched_setaffinity		compat_sys_sched_setaffinity
+240	o32	sched_getaffinity		sys_sched_getaffinity		compat_sys_sched_getaffinity
+241	o32	io_setup			sys_io_setup			compat_sys_io_setup
+242	o32	io_destroy			sys_io_destroy
+243	o32	io_getevents			sys_io_getevents_time32
+244	o32	io_submit			sys_io_submit			compat_sys_io_submit
+245	o32	io_cancel			sys_io_cancel
+246	o32	exit_group			sys_exit_group
+247	o32	lookup_dcookie			sys_lookup_dcookie		compat_sys_lookup_dcookie
+248	o32	epoll_create			sys_epoll_create
+249	o32	epoll_ctl			sys_epoll_ctl
+250	o32	epoll_wait			sys_epoll_wait
+251	o32	remap_file_pages		sys_remap_file_pages
+252	o32	set_tid_address			sys_set_tid_address
+253	o32	restart_syscall			sys_restart_syscall
+254	o32	fadvise64			sys_fadvise64_64		sys32_fadvise64_64
+255	o32	statfs64			sys_statfs64			compat_sys_statfs64
+256	o32	fstatfs64			sys_fstatfs64			compat_sys_fstatfs64
+257	o32	timer_create			sys_timer_create		compat_sys_timer_create
+258	o32	timer_settime			sys_timer_settime32
+259	o32	timer_gettime			sys_timer_gettime32
+260	o32	timer_getoverrun		sys_timer_getoverrun
+261	o32	timer_delete			sys_timer_delete
+262	o32	clock_settime			sys_clock_settime32
+263	o32	clock_gettime			sys_clock_gettime32
+264	o32	clock_getres			sys_clock_getres_time32
+265	o32	clock_nanosleep			sys_clock_nanosleep_time32
+266	o32	tgkill				sys_tgkill
+267	o32	utimes				sys_utimes_time32
+268	o32	mbind				sys_mbind			compat_sys_mbind
+269	o32	get_mempolicy			sys_get_mempolicy		compat_sys_get_mempolicy
+270	o32	set_mempolicy			sys_set_mempolicy		compat_sys_set_mempolicy
+271	o32	mq_open				sys_mq_open			compat_sys_mq_open
+272	o32	mq_unlink			sys_mq_unlink
+273	o32	mq_timedsend			sys_mq_timedsend_time32
+274	o32	mq_timedreceive			sys_mq_timedreceive_time32
+275	o32	mq_notify			sys_mq_notify			compat_sys_mq_notify
+276	o32	mq_getsetattr			sys_mq_getsetattr		compat_sys_mq_getsetattr
+277	o32	vserver				sys_ni_syscall
+278	o32	waitid				sys_waitid			compat_sys_waitid
+# 279 was sys_setaltroot
+280	o32	add_key				sys_add_key
+281	o32	request_key			sys_request_key
+282	o32	keyctl				sys_keyctl			compat_sys_keyctl
+283	o32	set_thread_area			sys_set_thread_area
+284	o32	inotify_init			sys_inotify_init
+285	o32	inotify_add_watch		sys_inotify_add_watch
+286	o32	inotify_rm_watch		sys_inotify_rm_watch
+287	o32	migrate_pages			sys_migrate_pages		compat_sys_migrate_pages
+288	o32	openat				sys_openat			compat_sys_openat
+289	o32	mkdirat				sys_mkdirat
+290	o32	mknodat				sys_mknodat
+291	o32	fchownat			sys_fchownat
+292	o32	futimesat			sys_futimesat_time32
+293	o32	fstatat64			sys_fstatat64			sys_newfstatat
+294	o32	unlinkat			sys_unlinkat
+295	o32	renameat			sys_renameat
+296	o32	linkat				sys_linkat
+297	o32	symlinkat			sys_symlinkat
+298	o32	readlinkat			sys_readlinkat
+299	o32	fchmodat			sys_fchmodat
+300	o32	faccessat			sys_faccessat
+301	o32	pselect6			sys_pselect6_time32		compat_sys_pselect6_time32
+302	o32	ppoll				sys_ppoll_time32		compat_sys_ppoll_time32
+303	o32	unshare				sys_unshare
+304	o32	splice				sys_splice
+305	o32	sync_file_range			sys_sync_file_range		sys32_sync_file_range
+306	o32	tee				sys_tee
+307	o32	vmsplice			sys_vmsplice
+308	o32	move_pages			sys_move_pages			compat_sys_move_pages
+309	o32	set_robust_list			sys_set_robust_list		compat_sys_set_robust_list
+310	o32	get_robust_list			sys_get_robust_list		compat_sys_get_robust_list
+311	o32	kexec_load			sys_kexec_load			compat_sys_kexec_load
+312	o32	getcpu				sys_getcpu
+313	o32	epoll_pwait			sys_epoll_pwait			compat_sys_epoll_pwait
+314	o32	ioprio_set			sys_ioprio_set
+315	o32	ioprio_get			sys_ioprio_get
+316	o32	utimensat			sys_utimensat_time32
+317	o32	signalfd			sys_signalfd			compat_sys_signalfd
+318	o32	timerfd				sys_ni_syscall
+319	o32	eventfd				sys_eventfd
+320	o32	fallocate			sys_fallocate			sys32_fallocate
+321	o32	timerfd_create			sys_timerfd_create
+322	o32	timerfd_gettime			sys_timerfd_gettime32
+323	o32	timerfd_settime			sys_timerfd_settime32
+324	o32	signalfd4			sys_signalfd4			compat_sys_signalfd4
+325	o32	eventfd2			sys_eventfd2
+326	o32	epoll_create1			sys_epoll_create1
+327	o32	dup3				sys_dup3
+328	o32	pipe2				sys_pipe2
+329	o32	inotify_init1			sys_inotify_init1
+330	o32	preadv				sys_preadv			compat_sys_preadv
+331	o32	pwritev				sys_pwritev			compat_sys_pwritev
+332	o32	rt_tgsigqueueinfo		sys_rt_tgsigqueueinfo		compat_sys_rt_tgsigqueueinfo
+333	o32	perf_event_open			sys_perf_event_open
+334	o32	accept4				sys_accept4
+335	o32	recvmmsg			sys_recvmmsg_time32		compat_sys_recvmmsg_time32
+336	o32	fanotify_init			sys_fanotify_init
+337	o32	fanotify_mark			sys_fanotify_mark		compat_sys_fanotify_mark
+338	o32	prlimit64			sys_prlimit64
+339	o32	name_to_handle_at		sys_name_to_handle_at
+340	o32	open_by_handle_at		sys_open_by_handle_at		compat_sys_open_by_handle_at
+341	o32	clock_adjtime			sys_clock_adjtime32
+342	o32	syncfs				sys_syncfs
+343	o32	sendmmsg			sys_sendmmsg			compat_sys_sendmmsg
+344	o32	setns				sys_setns
+345	o32	process_vm_readv		sys_process_vm_readv
+346	o32	process_vm_writev		sys_process_vm_writev
+347	o32	kcmp				sys_kcmp
+348	o32	finit_module			sys_finit_module
+349	o32	sched_setattr			sys_sched_setattr
+350	o32	sched_getattr			sys_sched_getattr
+351	o32	renameat2			sys_renameat2
+352	o32	seccomp				sys_seccomp
+353	o32	getrandom			sys_getrandom
+354	o32	memfd_create			sys_memfd_create
+355	o32	bpf				sys_bpf
+356	o32	execveat			sys_execveat			compat_sys_execveat
+357	o32	userfaultfd			sys_userfaultfd
+358	o32	membarrier			sys_membarrier
+359	o32	mlock2				sys_mlock2
+360	o32	copy_file_range			sys_copy_file_range
+361	o32	preadv2				sys_preadv2			compat_sys_preadv2
+362	o32	pwritev2			sys_pwritev2			compat_sys_pwritev2
+363	o32	pkey_mprotect			sys_pkey_mprotect
+364	o32	pkey_alloc			sys_pkey_alloc
+365	o32	pkey_free			sys_pkey_free
+366	o32	statx				sys_statx
+367	o32	rseq				sys_rseq
+368	o32	io_pgetevents			sys_io_pgetevents_time32	compat_sys_io_pgetevents
+# room for arch specific calls
+393	o32	semget				sys_semget
+394	o32	semctl				sys_semctl			compat_sys_semctl
+395	o32	shmget				sys_shmget
+396	o32	shmctl				sys_shmctl			compat_sys_shmctl
+397	o32	shmat				sys_shmat			compat_sys_shmat
+398	o32	shmdt				sys_shmdt
+399	o32	msgget				sys_msgget
+400	o32	msgsnd				sys_msgsnd			compat_sys_msgsnd
+401	o32	msgrcv				sys_msgrcv			compat_sys_msgrcv
+402	o32	msgctl				sys_msgctl			compat_sys_msgctl
+403	o32	clock_gettime64			sys_clock_gettime		sys_clock_gettime
+404	o32	clock_settime64			sys_clock_settime		sys_clock_settime
+405	o32	clock_adjtime64			sys_clock_adjtime		sys_clock_adjtime
+406	o32	clock_getres_time64		sys_clock_getres		sys_clock_getres
+407	o32	clock_nanosleep_time64		sys_clock_nanosleep		sys_clock_nanosleep
+408	o32	timer_gettime64			sys_timer_gettime		sys_timer_gettime
+409	o32	timer_settime64			sys_timer_settime		sys_timer_settime
+410	o32	timerfd_gettime64		sys_timerfd_gettime		sys_timerfd_gettime
+411	o32	timerfd_settime64		sys_timerfd_settime		sys_timerfd_settime
+412	o32	utimensat_time64		sys_utimensat			sys_utimensat
+413	o32	pselect6_time64			sys_pselect6			compat_sys_pselect6_time64
+414	o32	ppoll_time64			sys_ppoll			compat_sys_ppoll_time64
+416	o32	io_pgetevents_time64		sys_io_pgetevents		sys_io_pgetevents
+417	o32	recvmmsg_time64			sys_recvmmsg			compat_sys_recvmmsg_time64
+418	o32	mq_timedsend_time64		sys_mq_timedsend		sys_mq_timedsend
+419	o32	mq_timedreceive_time64		sys_mq_timedreceive		sys_mq_timedreceive
+420	o32	semtimedop_time64		sys_semtimedop			sys_semtimedop
+421	o32	rt_sigtimedwait_time64		sys_rt_sigtimedwait		compat_sys_rt_sigtimedwait_time64
+422	o32	futex_time64			sys_futex			sys_futex
+423	o32	sched_rr_get_interval_time64	sys_sched_rr_get_interval	sys_sched_rr_get_interval
+424	o32	pidfd_send_signal		sys_pidfd_send_signal
+425	o32	io_uring_setup			sys_io_uring_setup
+426	o32	io_uring_enter			sys_io_uring_enter
+427	o32	io_uring_register		sys_io_uring_register
+428	o32	open_tree			sys_open_tree
+429	o32	move_mount			sys_move_mount
+430	o32	fsopen				sys_fsopen
+431	o32	fsconfig			sys_fsconfig
+432	o32	fsmount				sys_fsmount
+433	o32	fspick				sys_fspick
+434	o32	pidfd_open			sys_pidfd_open
+435	o32	clone3				__sys_clone3
+436	o32	close_range			sys_close_range
+437	o32	openat2				sys_openat2
+438	o32	pidfd_getfd			sys_pidfd_getfd
+439	o32	faccessat2			sys_faccessat2
+440	o32	process_madvise			sys_process_madvise
diff --git a/arch/mips/kernel/syscalls/syscallhdr.sh b/arch/mips/kernel/syscalls/syscallhdr.sh
new file mode 100644
index 000000000..2e241e713
--- /dev/null
+++ b/arch/mips/kernel/syscalls/syscallhdr.sh
@@ -0,0 +1,36 @@
+#!/bin/sh
+# SPDX-License-Identifier: GPL-2.0
+
+in="$1"
+out="$2"
+my_abis=`echo "($3)" | tr ',' '|'`
+prefix="$4"
+offset="$5"
+
+fileguard=_UAPI_ASM_MIPS_`basename "$out" | sed \
+	-e 'y/abcdefghijklmnopqrstuvwxyz/ABCDEFGHIJKLMNOPQRSTUVWXYZ/' \
+	-e 's/[^A-Z0-9_]/_/g' -e 's/__/_/g'`
+grep -E "^[0-9A-Fa-fXx]+[[:space:]]+${my_abis}" "$in" | sort -n | (
+	printf "#ifndef %s\n" "${fileguard}"
+	printf "#define %s\n" "${fileguard}"
+	printf "\n"
+
+	nxt=0
+	while read nr abi name entry compat ; do
+		if [ -z "$offset" ]; then
+			printf "#define __NR_%s%s\t%s\n" \
+				"${prefix}" "${name}" "${nr}"
+		else
+			printf "#define __NR_%s%s\t(%s + %s)\n" \
+				"${prefix}" "${name}" "${offset}" "${nr}"
+		fi
+		nxt=$((nr+1))
+	done
+
+	printf "\n"
+	printf "#ifdef __KERNEL__\n"
+	printf "#define __NR_syscalls\t%s\n" "${nxt}"
+	printf "#endif\n"
+	printf "\n"
+	printf "#endif /* %s */\n" "${fileguard}"
+) > "$out"
diff --git a/arch/mips/kernel/syscalls/syscallnr.sh b/arch/mips/kernel/syscalls/syscallnr.sh
new file mode 100644
index 000000000..60bbdb3fe
--- /dev/null
+++ b/arch/mips/kernel/syscalls/syscallnr.sh
@@ -0,0 +1,28 @@
+#!/bin/sh
+# SPDX-License-Identifier: GPL-2.0
+
+in="$1"
+out="$2"
+my_abis=`echo "($3)" | tr ',' '|'`
+prefix="$4"
+offset="$5"
+
+fileguard=_UAPI_ASM_MIPS_`basename "$out" | sed \
+	-e 'y/abcdefghijklmnopqrstuvwxyz/ABCDEFGHIJKLMNOPQRSTUVWXYZ/' \
+	-e 's/[^A-Z0-9_]/_/g' -e 's/__/_/g'`
+grep -E "^[0-9A-Fa-fXx]+[[:space:]]+${my_abis}" "$in" | sort -n | (
+	printf "#ifndef %s\n" "${fileguard}"
+	printf "#define %s\n" "${fileguard}"
+	printf "\n"
+
+	nxt=0
+	while read nr abi name entry compat ; do
+		nxt=$((nr+1))
+	done
+
+	printf "#define __NR_%s_Linux\t%s\n" "${prefix}" "${offset}"
+	printf "#define __NR_%s_Linux_syscalls\t%s\n" "${prefix}" "${nxt}"
+	printf "\n"
+	printf "#endif /* %s */" "${fileguard}"
+	printf "\n"
+) > "$out"
diff --git a/arch/mips/kernel/syscalls/syscalltbl.sh b/arch/mips/kernel/syscalls/syscalltbl.sh
new file mode 100644
index 000000000..1e2570740
--- /dev/null
+++ b/arch/mips/kernel/syscalls/syscalltbl.sh
@@ -0,0 +1,36 @@
+#!/bin/sh
+# SPDX-License-Identifier: GPL-2.0
+
+in="$1"
+out="$2"
+my_abis=`echo "($3)" | tr ',' '|'`
+my_abi="$4"
+offset="$5"
+
+emit() {
+	t_nxt="$1"
+	t_nr="$2"
+	t_entry="$3"
+
+	while [ $t_nxt -lt $t_nr ]; do
+		printf "__SYSCALL(%s,sys_ni_syscall)\n" "${t_nxt}"
+		t_nxt=$((t_nxt+1))
+	done
+	printf "__SYSCALL(%s,%s)\n" "${t_nxt}" "${t_entry}"
+}
+
+grep -E "^[0-9A-Fa-fXx]+[[:space:]]+${my_abis}" "$in" | sort -n | (
+	nxt=0
+	if [ -z "$offset" ]; then
+		offset=0
+	fi
+
+	while read nr abi name entry compat ; do
+		if [ "$my_abi" = "64_o32" ] && [ ! -z "$compat" ]; then
+			emit $((nxt+offset)) $((nr+offset)) $compat
+		else
+			emit $((nxt+offset)) $((nr+offset)) $entry
+		fi
+		nxt=$((nr+1))
+	done
+) > "$out"
diff --git a/arch/mips/kernel/sysrq.c b/arch/mips/kernel/sysrq.c
new file mode 100644
index 000000000..9c1a20191
--- /dev/null
+++ b/arch/mips/kernel/sysrq.c
@@ -0,0 +1,66 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * MIPS specific sysrq operations.
+ *
+ * Copyright (C) 2015 Imagination Technologies Ltd.
+ */
+#include <linux/init.h>
+#include <linux/smp.h>
+#include <linux/spinlock.h>
+#include <linux/sysrq.h>
+#include <linux/workqueue.h>
+
+#include <asm/cpu-features.h>
+#include <asm/mipsregs.h>
+#include <asm/tlbdebug.h>
+
+/*
+ * Dump TLB entries on all CPUs.
+ */
+
+static DEFINE_SPINLOCK(show_lock);
+
+static void sysrq_tlbdump_single(void *dummy)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&show_lock, flags);
+
+	pr_info("CPU%d:\n", smp_processor_id());
+	dump_tlb_regs();
+	pr_info("\n");
+	dump_tlb_all();
+	pr_info("\n");
+
+	spin_unlock_irqrestore(&show_lock, flags);
+}
+
+#ifdef CONFIG_SMP
+static void sysrq_tlbdump_othercpus(struct work_struct *dummy)
+{
+	smp_call_function(sysrq_tlbdump_single, NULL, 0);
+}
+
+static DECLARE_WORK(sysrq_tlbdump, sysrq_tlbdump_othercpus);
+#endif
+
+static void sysrq_handle_tlbdump(int key)
+{
+	sysrq_tlbdump_single(NULL);
+#ifdef CONFIG_SMP
+	schedule_work(&sysrq_tlbdump);
+#endif
+}
+
+static const struct sysrq_key_op sysrq_tlbdump_op = {
+	.handler        = sysrq_handle_tlbdump,
+	.help_msg       = "show-tlbs(x)",
+	.action_msg     = "Show TLB entries",
+	.enable_mask	= SYSRQ_ENABLE_DUMP,
+};
+
+static int __init mips_sysrq_init(void)
+{
+	return register_sysrq_key('x', &sysrq_tlbdump_op);
+}
+arch_initcall(mips_sysrq_init);
diff --git a/arch/mips/kernel/time.c b/arch/mips/kernel/time.c
new file mode 100644
index 000000000..ed339d797
--- /dev/null
+++ b/arch/mips/kernel/time.c
@@ -0,0 +1,167 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright 2001 MontaVista Software Inc.
+ * Author: Jun Sun, jsun@mvista.com or jsun@junsun.net
+ * Copyright (c) 2003, 2004  Maciej W. Rozycki
+ *
+ * Common time service routines for MIPS machines.
+ */
+#include <linux/bug.h>
+#include <linux/clockchips.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/param.h>
+#include <linux/time.h>
+#include <linux/timex.h>
+#include <linux/smp.h>
+#include <linux/spinlock.h>
+#include <linux/export.h>
+#include <linux/cpufreq.h>
+#include <linux/delay.h>
+
+#include <asm/cpu-features.h>
+#include <asm/cpu-type.h>
+#include <asm/div64.h>
+#include <asm/time.h>
+
+#ifdef CONFIG_CPU_FREQ
+
+static DEFINE_PER_CPU(unsigned long, pcp_lpj_ref);
+static DEFINE_PER_CPU(unsigned long, pcp_lpj_ref_freq);
+static unsigned long glb_lpj_ref;
+static unsigned long glb_lpj_ref_freq;
+
+static int cpufreq_callback(struct notifier_block *nb,
+			    unsigned long val, void *data)
+{
+	struct cpufreq_freqs *freq = data;
+	struct cpumask *cpus = freq->policy->cpus;
+	unsigned long lpj;
+	int cpu;
+
+	/*
+	 * Skip lpj numbers adjustment if the CPU-freq transition is safe for
+	 * the loops delay. (Is this possible?)
+	 */
+	if (freq->flags & CPUFREQ_CONST_LOOPS)
+		return NOTIFY_OK;
+
+	/* Save the initial values of the lpjes for future scaling. */
+	if (!glb_lpj_ref) {
+		glb_lpj_ref = boot_cpu_data.udelay_val;
+		glb_lpj_ref_freq = freq->old;
+
+		for_each_online_cpu(cpu) {
+			per_cpu(pcp_lpj_ref, cpu) =
+				cpu_data[cpu].udelay_val;
+			per_cpu(pcp_lpj_ref_freq, cpu) = freq->old;
+		}
+	}
+
+	/*
+	 * Adjust global lpj variable and per-CPU udelay_val number in
+	 * accordance with the new CPU frequency.
+	 */
+	if ((val == CPUFREQ_PRECHANGE  && freq->old < freq->new) ||
+	    (val == CPUFREQ_POSTCHANGE && freq->old > freq->new)) {
+		loops_per_jiffy = cpufreq_scale(glb_lpj_ref,
+						glb_lpj_ref_freq,
+						freq->new);
+
+		for_each_cpu(cpu, cpus) {
+			lpj = cpufreq_scale(per_cpu(pcp_lpj_ref, cpu),
+					    per_cpu(pcp_lpj_ref_freq, cpu),
+					    freq->new);
+			cpu_data[cpu].udelay_val = (unsigned int)lpj;
+		}
+	}
+
+	return NOTIFY_OK;
+}
+
+static struct notifier_block cpufreq_notifier = {
+	.notifier_call  = cpufreq_callback,
+};
+
+static int __init register_cpufreq_notifier(void)
+{
+	return cpufreq_register_notifier(&cpufreq_notifier,
+					 CPUFREQ_TRANSITION_NOTIFIER);
+}
+core_initcall(register_cpufreq_notifier);
+
+#endif /* CONFIG_CPU_FREQ */
+
+/*
+ * forward reference
+ */
+DEFINE_SPINLOCK(rtc_lock);
+EXPORT_SYMBOL(rtc_lock);
+
+static int null_perf_irq(void)
+{
+	return 0;
+}
+
+int (*perf_irq)(void) = null_perf_irq;
+
+EXPORT_SYMBOL(perf_irq);
+
+/*
+ * time_init() - it does the following things.
+ *
+ * 1) plat_time_init() -
+ *	a) (optional) set up RTC routines,
+ *	b) (optional) calibrate and set the mips_hpt_frequency
+ *	    (only needed if you intended to use cpu counter as timer interrupt
+ *	     source)
+ * 2) calculate a couple of cached variables for later usage
+ */
+
+unsigned int mips_hpt_frequency;
+EXPORT_SYMBOL_GPL(mips_hpt_frequency);
+
+static __init int cpu_has_mfc0_count_bug(void)
+{
+	switch (current_cpu_type()) {
+	case CPU_R4000PC:
+	case CPU_R4000SC:
+	case CPU_R4000MC:
+		/*
+		 * V3.0 is documented as suffering from the mfc0 from count bug.
+		 * Afaik this is the last version of the R4000.	 Later versions
+		 * were marketed as R4400.
+		 */
+		return 1;
+
+	case CPU_R4400PC:
+	case CPU_R4400SC:
+	case CPU_R4400MC:
+		/*
+		 * The published errata for the R4400 up to 3.0 say the CPU
+		 * has the mfc0 from count bug.  This seems the last version
+		 * produced.
+		 */
+		return 1;
+	}
+
+	return 0;
+}
+
+void __init time_init(void)
+{
+	plat_time_init();
+
+	/*
+	 * The use of the R4k timer as a clock event takes precedence;
+	 * if reading the Count register might interfere with the timer
+	 * interrupt, then we don't use the timer as a clock source.
+	 * We may still use the timer as a clock source though if the
+	 * timer interrupt isn't reliable; the interference doesn't
+	 * matter then, because we don't use the interrupt.
+	 */
+	if (mips_clockevent_init() != 0 || !cpu_has_mfc0_count_bug())
+		init_mips_clocksource();
+}
diff --git a/arch/mips/kernel/topology.c b/arch/mips/kernel/topology.c
new file mode 100644
index 000000000..08ad6371f
--- /dev/null
+++ b/arch/mips/kernel/topology.c
@@ -0,0 +1,33 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/cpu.h>
+#include <linux/cpumask.h>
+#include <linux/init.h>
+#include <linux/node.h>
+#include <linux/nodemask.h>
+#include <linux/percpu.h>
+
+static DEFINE_PER_CPU(struct cpu, cpu_devices);
+
+static int __init topology_init(void)
+{
+	int i, ret;
+
+#ifdef CONFIG_NUMA
+	for_each_online_node(i)
+		register_one_node(i);
+#endif /* CONFIG_NUMA */
+
+	for_each_present_cpu(i) {
+		struct cpu *c = &per_cpu(cpu_devices, i);
+
+		c->hotpluggable = !!i;
+		ret = register_cpu(c, i);
+		if (ret)
+			printk(KERN_WARNING "topology_init: register_cpu %d "
+			       "failed (%d)\n", i, ret);
+	}
+
+	return 0;
+}
+
+subsys_initcall(topology_init);
diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c
new file mode 100644
index 000000000..ebd0101f0
--- /dev/null
+++ b/arch/mips/kernel/traps.c
@@ -0,0 +1,2571 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1994 - 1999, 2000, 01, 06 Ralf Baechle
+ * Copyright (C) 1995, 1996 Paul M. Antoine
+ * Copyright (C) 1998 Ulf Carlsson
+ * Copyright (C) 1999 Silicon Graphics, Inc.
+ * Kevin D. Kissell, kevink@mips.com and Carsten Langgaard, carstenl@mips.com
+ * Copyright (C) 2002, 2003, 2004, 2005, 2007  Maciej W. Rozycki
+ * Copyright (C) 2000, 2001, 2012 MIPS Technologies, Inc.  All rights reserved.
+ * Copyright (C) 2014, Imagination Technologies Ltd.
+ */
+#include <linux/bitops.h>
+#include <linux/bug.h>
+#include <linux/compiler.h>
+#include <linux/context_tracking.h>
+#include <linux/cpu_pm.h>
+#include <linux/kexec.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/extable.h>
+#include <linux/mm.h>
+#include <linux/sched/mm.h>
+#include <linux/sched/debug.h>
+#include <linux/smp.h>
+#include <linux/spinlock.h>
+#include <linux/kallsyms.h>
+#include <linux/memblock.h>
+#include <linux/interrupt.h>
+#include <linux/ptrace.h>
+#include <linux/kgdb.h>
+#include <linux/kdebug.h>
+#include <linux/kprobes.h>
+#include <linux/notifier.h>
+#include <linux/kdb.h>
+#include <linux/irq.h>
+#include <linux/perf_event.h>
+
+#include <asm/addrspace.h>
+#include <asm/bootinfo.h>
+#include <asm/branch.h>
+#include <asm/break.h>
+#include <asm/cop2.h>
+#include <asm/cpu.h>
+#include <asm/cpu-type.h>
+#include <asm/dsp.h>
+#include <asm/fpu.h>
+#include <asm/fpu_emulator.h>
+#include <asm/idle.h>
+#include <asm/isa-rev.h>
+#include <asm/mips-cps.h>
+#include <asm/mips-r2-to-r6-emul.h>
+#include <asm/mipsregs.h>
+#include <asm/mipsmtregs.h>
+#include <asm/module.h>
+#include <asm/msa.h>
+#include <asm/ptrace.h>
+#include <asm/sections.h>
+#include <asm/siginfo.h>
+#include <asm/tlbdebug.h>
+#include <asm/traps.h>
+#include <linux/uaccess.h>
+#include <asm/watch.h>
+#include <asm/mmu_context.h>
+#include <asm/types.h>
+#include <asm/stacktrace.h>
+#include <asm/tlbex.h>
+#include <asm/uasm.h>
+
+#include <asm/mach-loongson64/cpucfg-emul.h>
+
+extern void check_wait(void);
+extern asmlinkage void rollback_handle_int(void);
+extern asmlinkage void handle_int(void);
+extern asmlinkage void handle_adel(void);
+extern asmlinkage void handle_ades(void);
+extern asmlinkage void handle_ibe(void);
+extern asmlinkage void handle_dbe(void);
+extern asmlinkage void handle_sys(void);
+extern asmlinkage void handle_bp(void);
+extern asmlinkage void handle_ri(void);
+extern asmlinkage void handle_ri_rdhwr_tlbp(void);
+extern asmlinkage void handle_ri_rdhwr(void);
+extern asmlinkage void handle_cpu(void);
+extern asmlinkage void handle_ov(void);
+extern asmlinkage void handle_tr(void);
+extern asmlinkage void handle_msa_fpe(void);
+extern asmlinkage void handle_fpe(void);
+extern asmlinkage void handle_ftlb(void);
+extern asmlinkage void handle_gsexc(void);
+extern asmlinkage void handle_msa(void);
+extern asmlinkage void handle_mdmx(void);
+extern asmlinkage void handle_watch(void);
+extern asmlinkage void handle_mt(void);
+extern asmlinkage void handle_dsp(void);
+extern asmlinkage void handle_mcheck(void);
+extern asmlinkage void handle_reserved(void);
+extern void tlb_do_page_fault_0(void);
+
+void (*board_be_init)(void);
+int (*board_be_handler)(struct pt_regs *regs, int is_fixup);
+void (*board_nmi_handler_setup)(void);
+void (*board_ejtag_handler_setup)(void);
+void (*board_bind_eic_interrupt)(int irq, int regset);
+void (*board_ebase_setup)(void);
+void(*board_cache_error_setup)(void);
+
+static void show_raw_backtrace(unsigned long reg29, const char *loglvl)
+{
+	unsigned long *sp = (unsigned long *)(reg29 & ~3);
+	unsigned long addr;
+
+	printk("%sCall Trace:", loglvl);
+#ifdef CONFIG_KALLSYMS
+	printk("%s\n", loglvl);
+#endif
+	while (!kstack_end(sp)) {
+		unsigned long __user *p =
+			(unsigned long __user *)(unsigned long)sp++;
+		if (__get_user(addr, p)) {
+			printk("%s (Bad stack address)", loglvl);
+			break;
+		}
+		if (__kernel_text_address(addr))
+			print_ip_sym(loglvl, addr);
+	}
+	printk("%s\n", loglvl);
+}
+
+#ifdef CONFIG_KALLSYMS
+int raw_show_trace;
+static int __init set_raw_show_trace(char *str)
+{
+	raw_show_trace = 1;
+	return 1;
+}
+__setup("raw_show_trace", set_raw_show_trace);
+#endif
+
+static void show_backtrace(struct task_struct *task, const struct pt_regs *regs,
+			   const char *loglvl)
+{
+	unsigned long sp = regs->regs[29];
+	unsigned long ra = regs->regs[31];
+	unsigned long pc = regs->cp0_epc;
+
+	if (!task)
+		task = current;
+
+	if (raw_show_trace || user_mode(regs) || !__kernel_text_address(pc)) {
+		show_raw_backtrace(sp, loglvl);
+		return;
+	}
+	printk("%sCall Trace:\n", loglvl);
+	do {
+		print_ip_sym(loglvl, pc);
+		pc = unwind_stack(task, &sp, pc, &ra);
+	} while (pc);
+	pr_cont("\n");
+}
+
+/*
+ * This routine abuses get_user()/put_user() to reference pointers
+ * with at least a bit of error checking ...
+ */
+static void show_stacktrace(struct task_struct *task,
+	const struct pt_regs *regs, const char *loglvl)
+{
+	const int field = 2 * sizeof(unsigned long);
+	long stackdata;
+	int i;
+	unsigned long __user *sp = (unsigned long __user *)regs->regs[29];
+
+	printk("%sStack :", loglvl);
+	i = 0;
+	while ((unsigned long) sp & (PAGE_SIZE - 1)) {
+		if (i && ((i % (64 / field)) == 0)) {
+			pr_cont("\n");
+			printk("%s       ", loglvl);
+		}
+		if (i > 39) {
+			pr_cont(" ...");
+			break;
+		}
+
+		if (__get_user(stackdata, sp++)) {
+			pr_cont(" (Bad stack address)");
+			break;
+		}
+
+		pr_cont(" %0*lx", field, stackdata);
+		i++;
+	}
+	pr_cont("\n");
+	show_backtrace(task, regs, loglvl);
+}
+
+void show_stack(struct task_struct *task, unsigned long *sp, const char *loglvl)
+{
+	struct pt_regs regs;
+	mm_segment_t old_fs = get_fs();
+
+	regs.cp0_status = KSU_KERNEL;
+	if (sp) {
+		regs.regs[29] = (unsigned long)sp;
+		regs.regs[31] = 0;
+		regs.cp0_epc = 0;
+	} else {
+		if (task && task != current) {
+			regs.regs[29] = task->thread.reg29;
+			regs.regs[31] = 0;
+			regs.cp0_epc = task->thread.reg31;
+		} else {
+			prepare_frametrace(&regs);
+		}
+	}
+	/*
+	 * show_stack() deals exclusively with kernel mode, so be sure to access
+	 * the stack in the kernel (not user) address space.
+	 */
+	set_fs(KERNEL_DS);
+	show_stacktrace(task, &regs, loglvl);
+	set_fs(old_fs);
+}
+
+static void show_code(unsigned int __user *pc)
+{
+	long i;
+	unsigned short __user *pc16 = NULL;
+
+	printk("Code:");
+
+	if ((unsigned long)pc & 1)
+		pc16 = (unsigned short __user *)((unsigned long)pc & ~1);
+	for(i = -3 ; i < 6 ; i++) {
+		unsigned int insn;
+		if (pc16 ? __get_user(insn, pc16 + i) : __get_user(insn, pc + i)) {
+			pr_cont(" (Bad address in epc)\n");
+			break;
+		}
+		pr_cont("%c%0*x%c", (i?' ':'<'), pc16 ? 4 : 8, insn, (i?' ':'>'));
+	}
+	pr_cont("\n");
+}
+
+static void __show_regs(const struct pt_regs *regs)
+{
+	const int field = 2 * sizeof(unsigned long);
+	unsigned int cause = regs->cp0_cause;
+	unsigned int exccode;
+	int i;
+
+	show_regs_print_info(KERN_DEFAULT);
+
+	/*
+	 * Saved main processor registers
+	 */
+	for (i = 0; i < 32; ) {
+		if ((i % 4) == 0)
+			printk("$%2d   :", i);
+		if (i == 0)
+			pr_cont(" %0*lx", field, 0UL);
+		else if (i == 26 || i == 27)
+			pr_cont(" %*s", field, "");
+		else
+			pr_cont(" %0*lx", field, regs->regs[i]);
+
+		i++;
+		if ((i % 4) == 0)
+			pr_cont("\n");
+	}
+
+#ifdef CONFIG_CPU_HAS_SMARTMIPS
+	printk("Acx    : %0*lx\n", field, regs->acx);
+#endif
+	if (MIPS_ISA_REV < 6) {
+		printk("Hi    : %0*lx\n", field, regs->hi);
+		printk("Lo    : %0*lx\n", field, regs->lo);
+	}
+
+	/*
+	 * Saved cp0 registers
+	 */
+	printk("epc   : %0*lx %pS\n", field, regs->cp0_epc,
+	       (void *) regs->cp0_epc);
+	printk("ra    : %0*lx %pS\n", field, regs->regs[31],
+	       (void *) regs->regs[31]);
+
+	printk("Status: %08x	", (uint32_t) regs->cp0_status);
+
+	if (cpu_has_3kex) {
+		if (regs->cp0_status & ST0_KUO)
+			pr_cont("KUo ");
+		if (regs->cp0_status & ST0_IEO)
+			pr_cont("IEo ");
+		if (regs->cp0_status & ST0_KUP)
+			pr_cont("KUp ");
+		if (regs->cp0_status & ST0_IEP)
+			pr_cont("IEp ");
+		if (regs->cp0_status & ST0_KUC)
+			pr_cont("KUc ");
+		if (regs->cp0_status & ST0_IEC)
+			pr_cont("IEc ");
+	} else if (cpu_has_4kex) {
+		if (regs->cp0_status & ST0_KX)
+			pr_cont("KX ");
+		if (regs->cp0_status & ST0_SX)
+			pr_cont("SX ");
+		if (regs->cp0_status & ST0_UX)
+			pr_cont("UX ");
+		switch (regs->cp0_status & ST0_KSU) {
+		case KSU_USER:
+			pr_cont("USER ");
+			break;
+		case KSU_SUPERVISOR:
+			pr_cont("SUPERVISOR ");
+			break;
+		case KSU_KERNEL:
+			pr_cont("KERNEL ");
+			break;
+		default:
+			pr_cont("BAD_MODE ");
+			break;
+		}
+		if (regs->cp0_status & ST0_ERL)
+			pr_cont("ERL ");
+		if (regs->cp0_status & ST0_EXL)
+			pr_cont("EXL ");
+		if (regs->cp0_status & ST0_IE)
+			pr_cont("IE ");
+	}
+	pr_cont("\n");
+
+	exccode = (cause & CAUSEF_EXCCODE) >> CAUSEB_EXCCODE;
+	printk("Cause : %08x (ExcCode %02x)\n", cause, exccode);
+
+	if (1 <= exccode && exccode <= 5)
+		printk("BadVA : %0*lx\n", field, regs->cp0_badvaddr);
+
+	printk("PrId  : %08x (%s)\n", read_c0_prid(),
+	       cpu_name_string());
+}
+
+/*
+ * FIXME: really the generic show_regs should take a const pointer argument.
+ */
+void show_regs(struct pt_regs *regs)
+{
+	__show_regs(regs);
+	dump_stack();
+}
+
+void show_registers(struct pt_regs *regs)
+{
+	const int field = 2 * sizeof(unsigned long);
+	mm_segment_t old_fs = get_fs();
+
+	__show_regs(regs);
+	print_modules();
+	printk("Process %s (pid: %d, threadinfo=%p, task=%p, tls=%0*lx)\n",
+	       current->comm, current->pid, current_thread_info(), current,
+	      field, current_thread_info()->tp_value);
+	if (cpu_has_userlocal) {
+		unsigned long tls;
+
+		tls = read_c0_userlocal();
+		if (tls != current_thread_info()->tp_value)
+			printk("*HwTLS: %0*lx\n", field, tls);
+	}
+
+	if (!user_mode(regs))
+		/* Necessary for getting the correct stack content */
+		set_fs(KERNEL_DS);
+	show_stacktrace(current, regs, KERN_DEFAULT);
+	show_code((unsigned int __user *) regs->cp0_epc);
+	printk("\n");
+	set_fs(old_fs);
+}
+
+static DEFINE_RAW_SPINLOCK(die_lock);
+
+void __noreturn die(const char *str, struct pt_regs *regs)
+{
+	static int die_counter;
+	int sig = SIGSEGV;
+
+	oops_enter();
+
+	if (notify_die(DIE_OOPS, str, regs, 0, current->thread.trap_nr,
+		       SIGSEGV) == NOTIFY_STOP)
+		sig = 0;
+
+	console_verbose();
+	raw_spin_lock_irq(&die_lock);
+	bust_spinlocks(1);
+
+	printk("%s[#%d]:\n", str, ++die_counter);
+	show_registers(regs);
+	add_taint(TAINT_DIE, LOCKDEP_NOW_UNRELIABLE);
+	raw_spin_unlock_irq(&die_lock);
+
+	oops_exit();
+
+	if (in_interrupt())
+		panic("Fatal exception in interrupt");
+
+	if (panic_on_oops)
+		panic("Fatal exception");
+
+	if (regs && kexec_should_crash(current))
+		crash_kexec(regs);
+
+	make_task_dead(sig);
+}
+
+extern struct exception_table_entry __start___dbe_table[];
+extern struct exception_table_entry __stop___dbe_table[];
+
+__asm__(
+"	.section	__dbe_table, \"a\"\n"
+"	.previous			\n");
+
+/* Given an address, look for it in the exception tables. */
+static const struct exception_table_entry *search_dbe_tables(unsigned long addr)
+{
+	const struct exception_table_entry *e;
+
+	e = search_extable(__start___dbe_table,
+			   __stop___dbe_table - __start___dbe_table, addr);
+	if (!e)
+		e = search_module_dbetables(addr);
+	return e;
+}
+
+asmlinkage void do_be(struct pt_regs *regs)
+{
+	const int field = 2 * sizeof(unsigned long);
+	const struct exception_table_entry *fixup = NULL;
+	int data = regs->cp0_cause & 4;
+	int action = MIPS_BE_FATAL;
+	enum ctx_state prev_state;
+
+	prev_state = exception_enter();
+	/* XXX For now.	 Fixme, this searches the wrong table ...  */
+	if (data && !user_mode(regs))
+		fixup = search_dbe_tables(exception_epc(regs));
+
+	if (fixup)
+		action = MIPS_BE_FIXUP;
+
+	if (board_be_handler)
+		action = board_be_handler(regs, fixup != NULL);
+	else
+		mips_cm_error_report();
+
+	switch (action) {
+	case MIPS_BE_DISCARD:
+		goto out;
+	case MIPS_BE_FIXUP:
+		if (fixup) {
+			regs->cp0_epc = fixup->nextinsn;
+			goto out;
+		}
+		break;
+	default:
+		break;
+	}
+
+	/*
+	 * Assume it would be too dangerous to continue ...
+	 */
+	printk(KERN_ALERT "%s bus error, epc == %0*lx, ra == %0*lx\n",
+	       data ? "Data" : "Instruction",
+	       field, regs->cp0_epc, field, regs->regs[31]);
+	if (notify_die(DIE_OOPS, "bus error", regs, 0, current->thread.trap_nr,
+		       SIGBUS) == NOTIFY_STOP)
+		goto out;
+
+	die_if_kernel("Oops", regs);
+	force_sig(SIGBUS);
+
+out:
+	exception_exit(prev_state);
+}
+
+/*
+ * ll/sc, rdhwr, sync emulation
+ */
+
+#define OPCODE 0xfc000000
+#define BASE   0x03e00000
+#define RT     0x001f0000
+#define OFFSET 0x0000ffff
+#define LL     0xc0000000
+#define SC     0xe0000000
+#define SPEC0  0x00000000
+#define SPEC3  0x7c000000
+#define RD     0x0000f800
+#define FUNC   0x0000003f
+#define SYNC   0x0000000f
+#define RDHWR  0x0000003b
+
+/*  microMIPS definitions   */
+#define MM_POOL32A_FUNC 0xfc00ffff
+#define MM_RDHWR        0x00006b3c
+#define MM_RS           0x001f0000
+#define MM_RT           0x03e00000
+
+/*
+ * The ll_bit is cleared by r*_switch.S
+ */
+
+unsigned int ll_bit;
+struct task_struct *ll_task;
+
+static inline int simulate_ll(struct pt_regs *regs, unsigned int opcode)
+{
+	unsigned long value, __user *vaddr;
+	long offset;
+
+	/*
+	 * analyse the ll instruction that just caused a ri exception
+	 * and put the referenced address to addr.
+	 */
+
+	/* sign extend offset */
+	offset = opcode & OFFSET;
+	offset <<= 16;
+	offset >>= 16;
+
+	vaddr = (unsigned long __user *)
+		((unsigned long)(regs->regs[(opcode & BASE) >> 21]) + offset);
+
+	if ((unsigned long)vaddr & 3)
+		return SIGBUS;
+	if (get_user(value, vaddr))
+		return SIGSEGV;
+
+	preempt_disable();
+
+	if (ll_task == NULL || ll_task == current) {
+		ll_bit = 1;
+	} else {
+		ll_bit = 0;
+	}
+	ll_task = current;
+
+	preempt_enable();
+
+	regs->regs[(opcode & RT) >> 16] = value;
+
+	return 0;
+}
+
+static inline int simulate_sc(struct pt_regs *regs, unsigned int opcode)
+{
+	unsigned long __user *vaddr;
+	unsigned long reg;
+	long offset;
+
+	/*
+	 * analyse the sc instruction that just caused a ri exception
+	 * and put the referenced address to addr.
+	 */
+
+	/* sign extend offset */
+	offset = opcode & OFFSET;
+	offset <<= 16;
+	offset >>= 16;
+
+	vaddr = (unsigned long __user *)
+		((unsigned long)(regs->regs[(opcode & BASE) >> 21]) + offset);
+	reg = (opcode & RT) >> 16;
+
+	if ((unsigned long)vaddr & 3)
+		return SIGBUS;
+
+	preempt_disable();
+
+	if (ll_bit == 0 || ll_task != current) {
+		regs->regs[reg] = 0;
+		preempt_enable();
+		return 0;
+	}
+
+	preempt_enable();
+
+	if (put_user(regs->regs[reg], vaddr))
+		return SIGSEGV;
+
+	regs->regs[reg] = 1;
+
+	return 0;
+}
+
+/*
+ * ll uses the opcode of lwc0 and sc uses the opcode of swc0.  That is both
+ * opcodes are supposed to result in coprocessor unusable exceptions if
+ * executed on ll/sc-less processors.  That's the theory.  In practice a
+ * few processors such as NEC's VR4100 throw reserved instruction exceptions
+ * instead, so we're doing the emulation thing in both exception handlers.
+ */
+static int simulate_llsc(struct pt_regs *regs, unsigned int opcode)
+{
+	if ((opcode & OPCODE) == LL) {
+		perf_sw_event(PERF_COUNT_SW_EMULATION_FAULTS,
+				1, regs, 0);
+		return simulate_ll(regs, opcode);
+	}
+	if ((opcode & OPCODE) == SC) {
+		perf_sw_event(PERF_COUNT_SW_EMULATION_FAULTS,
+				1, regs, 0);
+		return simulate_sc(regs, opcode);
+	}
+
+	return -1;			/* Must be something else ... */
+}
+
+/*
+ * Simulate trapping 'rdhwr' instructions to provide user accessible
+ * registers not implemented in hardware.
+ */
+static int simulate_rdhwr(struct pt_regs *regs, int rd, int rt)
+{
+	struct thread_info *ti = task_thread_info(current);
+
+	perf_sw_event(PERF_COUNT_SW_EMULATION_FAULTS,
+			1, regs, 0);
+	switch (rd) {
+	case MIPS_HWR_CPUNUM:		/* CPU number */
+		regs->regs[rt] = smp_processor_id();
+		return 0;
+	case MIPS_HWR_SYNCISTEP:	/* SYNCI length */
+		regs->regs[rt] = min(current_cpu_data.dcache.linesz,
+				     current_cpu_data.icache.linesz);
+		return 0;
+	case MIPS_HWR_CC:		/* Read count register */
+		regs->regs[rt] = read_c0_count();
+		return 0;
+	case MIPS_HWR_CCRES:		/* Count register resolution */
+		switch (current_cpu_type()) {
+		case CPU_20KC:
+		case CPU_25KF:
+			regs->regs[rt] = 1;
+			break;
+		default:
+			regs->regs[rt] = 2;
+		}
+		return 0;
+	case MIPS_HWR_ULR:		/* Read UserLocal register */
+		regs->regs[rt] = ti->tp_value;
+		return 0;
+	default:
+		return -1;
+	}
+}
+
+static int simulate_rdhwr_normal(struct pt_regs *regs, unsigned int opcode)
+{
+	if ((opcode & OPCODE) == SPEC3 && (opcode & FUNC) == RDHWR) {
+		int rd = (opcode & RD) >> 11;
+		int rt = (opcode & RT) >> 16;
+
+		simulate_rdhwr(regs, rd, rt);
+		return 0;
+	}
+
+	/* Not ours.  */
+	return -1;
+}
+
+static int simulate_rdhwr_mm(struct pt_regs *regs, unsigned int opcode)
+{
+	if ((opcode & MM_POOL32A_FUNC) == MM_RDHWR) {
+		int rd = (opcode & MM_RS) >> 16;
+		int rt = (opcode & MM_RT) >> 21;
+		simulate_rdhwr(regs, rd, rt);
+		return 0;
+	}
+
+	/* Not ours.  */
+	return -1;
+}
+
+static int simulate_sync(struct pt_regs *regs, unsigned int opcode)
+{
+	if ((opcode & OPCODE) == SPEC0 && (opcode & FUNC) == SYNC) {
+		perf_sw_event(PERF_COUNT_SW_EMULATION_FAULTS,
+				1, regs, 0);
+		return 0;
+	}
+
+	return -1;			/* Must be something else ... */
+}
+
+/*
+ * Loongson-3 CSR instructions emulation
+ */
+
+#ifdef CONFIG_CPU_LOONGSON3_CPUCFG_EMULATION
+
+#define LWC2             0xc8000000
+#define RS               BASE
+#define CSR_OPCODE2      0x00000118
+#define CSR_OPCODE2_MASK 0x000007ff
+#define CSR_FUNC_MASK    RT
+#define CSR_FUNC_CPUCFG  0x8
+
+static int simulate_loongson3_cpucfg(struct pt_regs *regs,
+				     unsigned int opcode)
+{
+	int op = opcode & OPCODE;
+	int op2 = opcode & CSR_OPCODE2_MASK;
+	int csr_func = (opcode & CSR_FUNC_MASK) >> 16;
+
+	if (op == LWC2 && op2 == CSR_OPCODE2 && csr_func == CSR_FUNC_CPUCFG) {
+		int rd = (opcode & RD) >> 11;
+		int rs = (opcode & RS) >> 21;
+		__u64 sel = regs->regs[rs];
+
+		perf_sw_event(PERF_COUNT_SW_EMULATION_FAULTS, 1, regs, 0);
+
+		/* Do not emulate on unsupported core models. */
+		preempt_disable();
+		if (!loongson3_cpucfg_emulation_enabled(&current_cpu_data)) {
+			preempt_enable();
+			return -1;
+		}
+		regs->regs[rd] = loongson3_cpucfg_read_synthesized(
+			&current_cpu_data, sel);
+		preempt_enable();
+		return 0;
+	}
+
+	/* Not ours.  */
+	return -1;
+}
+#endif /* CONFIG_CPU_LOONGSON3_CPUCFG_EMULATION */
+
+asmlinkage void do_ov(struct pt_regs *regs)
+{
+	enum ctx_state prev_state;
+
+	prev_state = exception_enter();
+	die_if_kernel("Integer overflow", regs);
+
+	force_sig_fault(SIGFPE, FPE_INTOVF, (void __user *)regs->cp0_epc);
+	exception_exit(prev_state);
+}
+
+#ifdef CONFIG_MIPS_FP_SUPPORT
+
+/*
+ * Send SIGFPE according to FCSR Cause bits, which must have already
+ * been masked against Enable bits.  This is impotant as Inexact can
+ * happen together with Overflow or Underflow, and `ptrace' can set
+ * any bits.
+ */
+void force_fcr31_sig(unsigned long fcr31, void __user *fault_addr,
+		     struct task_struct *tsk)
+{
+	int si_code = FPE_FLTUNK;
+
+	if (fcr31 & FPU_CSR_INV_X)
+		si_code = FPE_FLTINV;
+	else if (fcr31 & FPU_CSR_DIV_X)
+		si_code = FPE_FLTDIV;
+	else if (fcr31 & FPU_CSR_OVF_X)
+		si_code = FPE_FLTOVF;
+	else if (fcr31 & FPU_CSR_UDF_X)
+		si_code = FPE_FLTUND;
+	else if (fcr31 & FPU_CSR_INE_X)
+		si_code = FPE_FLTRES;
+
+	force_sig_fault_to_task(SIGFPE, si_code, fault_addr, tsk);
+}
+
+int process_fpemu_return(int sig, void __user *fault_addr, unsigned long fcr31)
+{
+	int si_code;
+	struct vm_area_struct *vma;
+
+	switch (sig) {
+	case 0:
+		return 0;
+
+	case SIGFPE:
+		force_fcr31_sig(fcr31, fault_addr, current);
+		return 1;
+
+	case SIGBUS:
+		force_sig_fault(SIGBUS, BUS_ADRERR, fault_addr);
+		return 1;
+
+	case SIGSEGV:
+		mmap_read_lock(current->mm);
+		vma = find_vma(current->mm, (unsigned long)fault_addr);
+		if (vma && (vma->vm_start <= (unsigned long)fault_addr))
+			si_code = SEGV_ACCERR;
+		else
+			si_code = SEGV_MAPERR;
+		mmap_read_unlock(current->mm);
+		force_sig_fault(SIGSEGV, si_code, fault_addr);
+		return 1;
+
+	default:
+		force_sig(sig);
+		return 1;
+	}
+}
+
+static int simulate_fp(struct pt_regs *regs, unsigned int opcode,
+		       unsigned long old_epc, unsigned long old_ra)
+{
+	union mips_instruction inst = { .word = opcode };
+	void __user *fault_addr;
+	unsigned long fcr31;
+	int sig;
+
+	/* If it's obviously not an FP instruction, skip it */
+	switch (inst.i_format.opcode) {
+	case cop1_op:
+	case cop1x_op:
+	case lwc1_op:
+	case ldc1_op:
+	case swc1_op:
+	case sdc1_op:
+		break;
+
+	default:
+		return -1;
+	}
+
+	/*
+	 * do_ri skipped over the instruction via compute_return_epc, undo
+	 * that for the FPU emulator.
+	 */
+	regs->cp0_epc = old_epc;
+	regs->regs[31] = old_ra;
+
+	/* Run the emulator */
+	sig = fpu_emulator_cop1Handler(regs, &current->thread.fpu, 1,
+				       &fault_addr);
+
+	/*
+	 * We can't allow the emulated instruction to leave any
+	 * enabled Cause bits set in $fcr31.
+	 */
+	fcr31 = mask_fcr31_x(current->thread.fpu.fcr31);
+	current->thread.fpu.fcr31 &= ~fcr31;
+
+	/* Restore the hardware register state */
+	own_fpu(1);
+
+	/* Send a signal if required.  */
+	process_fpemu_return(sig, fault_addr, fcr31);
+
+	return 0;
+}
+
+/*
+ * XXX Delayed fp exceptions when doing a lazy ctx switch XXX
+ */
+asmlinkage void do_fpe(struct pt_regs *regs, unsigned long fcr31)
+{
+	enum ctx_state prev_state;
+	void __user *fault_addr;
+	int sig;
+
+	prev_state = exception_enter();
+	if (notify_die(DIE_FP, "FP exception", regs, 0, current->thread.trap_nr,
+		       SIGFPE) == NOTIFY_STOP)
+		goto out;
+
+	/* Clear FCSR.Cause before enabling interrupts */
+	write_32bit_cp1_register(CP1_STATUS, fcr31 & ~mask_fcr31_x(fcr31));
+	local_irq_enable();
+
+	die_if_kernel("FP exception in kernel code", regs);
+
+	if (fcr31 & FPU_CSR_UNI_X) {
+		/*
+		 * Unimplemented operation exception.  If we've got the full
+		 * software emulator on-board, let's use it...
+		 *
+		 * Force FPU to dump state into task/thread context.  We're
+		 * moving a lot of data here for what is probably a single
+		 * instruction, but the alternative is to pre-decode the FP
+		 * register operands before invoking the emulator, which seems
+		 * a bit extreme for what should be an infrequent event.
+		 */
+
+		/* Run the emulator */
+		sig = fpu_emulator_cop1Handler(regs, &current->thread.fpu, 1,
+					       &fault_addr);
+
+		/*
+		 * We can't allow the emulated instruction to leave any
+		 * enabled Cause bits set in $fcr31.
+		 */
+		fcr31 = mask_fcr31_x(current->thread.fpu.fcr31);
+		current->thread.fpu.fcr31 &= ~fcr31;
+
+		/* Restore the hardware register state */
+		own_fpu(1);	/* Using the FPU again.	 */
+	} else {
+		sig = SIGFPE;
+		fault_addr = (void __user *) regs->cp0_epc;
+	}
+
+	/* Send a signal if required.  */
+	process_fpemu_return(sig, fault_addr, fcr31);
+
+out:
+	exception_exit(prev_state);
+}
+
+/*
+ * MIPS MT processors may have fewer FPU contexts than CPU threads. If we've
+ * emulated more than some threshold number of instructions, force migration to
+ * a "CPU" that has FP support.
+ */
+static void mt_ase_fp_affinity(void)
+{
+#ifdef CONFIG_MIPS_MT_FPAFF
+	if (mt_fpemul_threshold > 0 &&
+	     ((current->thread.emulated_fp++ > mt_fpemul_threshold))) {
+		/*
+		 * If there's no FPU present, or if the application has already
+		 * restricted the allowed set to exclude any CPUs with FPUs,
+		 * we'll skip the procedure.
+		 */
+		if (cpumask_intersects(&current->cpus_mask, &mt_fpu_cpumask)) {
+			cpumask_t tmask;
+
+			current->thread.user_cpus_allowed
+				= current->cpus_mask;
+			cpumask_and(&tmask, &current->cpus_mask,
+				    &mt_fpu_cpumask);
+			set_cpus_allowed_ptr(current, &tmask);
+			set_thread_flag(TIF_FPUBOUND);
+		}
+	}
+#endif /* CONFIG_MIPS_MT_FPAFF */
+}
+
+#else /* !CONFIG_MIPS_FP_SUPPORT */
+
+static int simulate_fp(struct pt_regs *regs, unsigned int opcode,
+		       unsigned long old_epc, unsigned long old_ra)
+{
+	return -1;
+}
+
+#endif /* !CONFIG_MIPS_FP_SUPPORT */
+
+void do_trap_or_bp(struct pt_regs *regs, unsigned int code, int si_code,
+	const char *str)
+{
+	char b[40];
+
+#ifdef CONFIG_KGDB_LOW_LEVEL_TRAP
+	if (kgdb_ll_trap(DIE_TRAP, str, regs, code, current->thread.trap_nr,
+			 SIGTRAP) == NOTIFY_STOP)
+		return;
+#endif /* CONFIG_KGDB_LOW_LEVEL_TRAP */
+
+	if (notify_die(DIE_TRAP, str, regs, code, current->thread.trap_nr,
+		       SIGTRAP) == NOTIFY_STOP)
+		return;
+
+	/*
+	 * A short test says that IRIX 5.3 sends SIGTRAP for all trap
+	 * insns, even for trap and break codes that indicate arithmetic
+	 * failures.  Weird ...
+	 * But should we continue the brokenness???  --macro
+	 */
+	switch (code) {
+	case BRK_OVERFLOW:
+	case BRK_DIVZERO:
+		scnprintf(b, sizeof(b), "%s instruction in kernel code", str);
+		die_if_kernel(b, regs);
+		force_sig_fault(SIGFPE,
+				code == BRK_DIVZERO ? FPE_INTDIV : FPE_INTOVF,
+				(void __user *) regs->cp0_epc);
+		break;
+	case BRK_BUG:
+		die_if_kernel("Kernel bug detected", regs);
+		force_sig(SIGTRAP);
+		break;
+	case BRK_MEMU:
+		/*
+		 * This breakpoint code is used by the FPU emulator to retake
+		 * control of the CPU after executing the instruction from the
+		 * delay slot of an emulated branch.
+		 *
+		 * Terminate if exception was recognized as a delay slot return
+		 * otherwise handle as normal.
+		 */
+		if (do_dsemulret(regs))
+			return;
+
+		die_if_kernel("Math emu break/trap", regs);
+		force_sig(SIGTRAP);
+		break;
+	default:
+		scnprintf(b, sizeof(b), "%s instruction in kernel code", str);
+		die_if_kernel(b, regs);
+		if (si_code) {
+			force_sig_fault(SIGTRAP, si_code, NULL);
+		} else {
+			force_sig(SIGTRAP);
+		}
+	}
+}
+
+asmlinkage void do_bp(struct pt_regs *regs)
+{
+	unsigned long epc = msk_isa16_mode(exception_epc(regs));
+	unsigned int opcode, bcode;
+	enum ctx_state prev_state;
+	mm_segment_t seg;
+
+	seg = get_fs();
+	if (!user_mode(regs))
+		set_fs(KERNEL_DS);
+
+	prev_state = exception_enter();
+	current->thread.trap_nr = (regs->cp0_cause >> 2) & 0x1f;
+	if (get_isa16_mode(regs->cp0_epc)) {
+		u16 instr[2];
+
+		if (__get_user(instr[0], (u16 __user *)epc))
+			goto out_sigsegv;
+
+		if (!cpu_has_mmips) {
+			/* MIPS16e mode */
+			bcode = (instr[0] >> 5) & 0x3f;
+		} else if (mm_insn_16bit(instr[0])) {
+			/* 16-bit microMIPS BREAK */
+			bcode = instr[0] & 0xf;
+		} else {
+			/* 32-bit microMIPS BREAK */
+			if (__get_user(instr[1], (u16 __user *)(epc + 2)))
+				goto out_sigsegv;
+			opcode = (instr[0] << 16) | instr[1];
+			bcode = (opcode >> 6) & ((1 << 20) - 1);
+		}
+	} else {
+		if (__get_user(opcode, (unsigned int __user *)epc))
+			goto out_sigsegv;
+		bcode = (opcode >> 6) & ((1 << 20) - 1);
+	}
+
+	/*
+	 * There is the ancient bug in the MIPS assemblers that the break
+	 * code starts left to bit 16 instead to bit 6 in the opcode.
+	 * Gas is bug-compatible, but not always, grrr...
+	 * We handle both cases with a simple heuristics.  --macro
+	 */
+	if (bcode >= (1 << 10))
+		bcode = ((bcode & ((1 << 10) - 1)) << 10) | (bcode >> 10);
+
+	/*
+	 * notify the kprobe handlers, if instruction is likely to
+	 * pertain to them.
+	 */
+	switch (bcode) {
+	case BRK_UPROBE:
+		if (notify_die(DIE_UPROBE, "uprobe", regs, bcode,
+			       current->thread.trap_nr, SIGTRAP) == NOTIFY_STOP)
+			goto out;
+		else
+			break;
+	case BRK_UPROBE_XOL:
+		if (notify_die(DIE_UPROBE_XOL, "uprobe_xol", regs, bcode,
+			       current->thread.trap_nr, SIGTRAP) == NOTIFY_STOP)
+			goto out;
+		else
+			break;
+	case BRK_KPROBE_BP:
+		if (notify_die(DIE_BREAK, "debug", regs, bcode,
+			       current->thread.trap_nr, SIGTRAP) == NOTIFY_STOP)
+			goto out;
+		else
+			break;
+	case BRK_KPROBE_SSTEPBP:
+		if (notify_die(DIE_SSTEPBP, "single_step", regs, bcode,
+			       current->thread.trap_nr, SIGTRAP) == NOTIFY_STOP)
+			goto out;
+		else
+			break;
+	default:
+		break;
+	}
+
+	do_trap_or_bp(regs, bcode, TRAP_BRKPT, "Break");
+
+out:
+	set_fs(seg);
+	exception_exit(prev_state);
+	return;
+
+out_sigsegv:
+	force_sig(SIGSEGV);
+	goto out;
+}
+
+asmlinkage void do_tr(struct pt_regs *regs)
+{
+	u32 opcode, tcode = 0;
+	enum ctx_state prev_state;
+	u16 instr[2];
+	mm_segment_t seg;
+	unsigned long epc = msk_isa16_mode(exception_epc(regs));
+
+	seg = get_fs();
+	if (!user_mode(regs))
+		set_fs(KERNEL_DS);
+
+	prev_state = exception_enter();
+	current->thread.trap_nr = (regs->cp0_cause >> 2) & 0x1f;
+	if (get_isa16_mode(regs->cp0_epc)) {
+		if (__get_user(instr[0], (u16 __user *)(epc + 0)) ||
+		    __get_user(instr[1], (u16 __user *)(epc + 2)))
+			goto out_sigsegv;
+		opcode = (instr[0] << 16) | instr[1];
+		/* Immediate versions don't provide a code.  */
+		if (!(opcode & OPCODE))
+			tcode = (opcode >> 12) & ((1 << 4) - 1);
+	} else {
+		if (__get_user(opcode, (u32 __user *)epc))
+			goto out_sigsegv;
+		/* Immediate versions don't provide a code.  */
+		if (!(opcode & OPCODE))
+			tcode = (opcode >> 6) & ((1 << 10) - 1);
+	}
+
+	do_trap_or_bp(regs, tcode, 0, "Trap");
+
+out:
+	set_fs(seg);
+	exception_exit(prev_state);
+	return;
+
+out_sigsegv:
+	force_sig(SIGSEGV);
+	goto out;
+}
+
+asmlinkage void do_ri(struct pt_regs *regs)
+{
+	unsigned int __user *epc = (unsigned int __user *)exception_epc(regs);
+	unsigned long old_epc = regs->cp0_epc;
+	unsigned long old31 = regs->regs[31];
+	enum ctx_state prev_state;
+	unsigned int opcode = 0;
+	int status = -1;
+
+	/*
+	 * Avoid any kernel code. Just emulate the R2 instruction
+	 * as quickly as possible.
+	 */
+	if (mipsr2_emulation && cpu_has_mips_r6 &&
+	    likely(user_mode(regs)) &&
+	    likely(get_user(opcode, epc) >= 0)) {
+		unsigned long fcr31 = 0;
+
+		status = mipsr2_decoder(regs, opcode, &fcr31);
+		switch (status) {
+		case 0:
+		case SIGEMT:
+			return;
+		case SIGILL:
+			goto no_r2_instr;
+		default:
+			process_fpemu_return(status,
+					     &current->thread.cp0_baduaddr,
+					     fcr31);
+			return;
+		}
+	}
+
+no_r2_instr:
+
+	prev_state = exception_enter();
+	current->thread.trap_nr = (regs->cp0_cause >> 2) & 0x1f;
+
+	if (notify_die(DIE_RI, "RI Fault", regs, 0, current->thread.trap_nr,
+		       SIGILL) == NOTIFY_STOP)
+		goto out;
+
+	die_if_kernel("Reserved instruction in kernel code", regs);
+
+	if (unlikely(compute_return_epc(regs) < 0))
+		goto out;
+
+	if (!get_isa16_mode(regs->cp0_epc)) {
+		if (unlikely(get_user(opcode, epc) < 0))
+			status = SIGSEGV;
+
+		if (!cpu_has_llsc && status < 0)
+			status = simulate_llsc(regs, opcode);
+
+		if (status < 0)
+			status = simulate_rdhwr_normal(regs, opcode);
+
+		if (status < 0)
+			status = simulate_sync(regs, opcode);
+
+		if (status < 0)
+			status = simulate_fp(regs, opcode, old_epc, old31);
+
+#ifdef CONFIG_CPU_LOONGSON3_CPUCFG_EMULATION
+		if (status < 0)
+			status = simulate_loongson3_cpucfg(regs, opcode);
+#endif
+	} else if (cpu_has_mmips) {
+		unsigned short mmop[2] = { 0 };
+
+		if (unlikely(get_user(mmop[0], (u16 __user *)epc + 0) < 0))
+			status = SIGSEGV;
+		if (unlikely(get_user(mmop[1], (u16 __user *)epc + 1) < 0))
+			status = SIGSEGV;
+		opcode = mmop[0];
+		opcode = (opcode << 16) | mmop[1];
+
+		if (status < 0)
+			status = simulate_rdhwr_mm(regs, opcode);
+	}
+
+	if (status < 0)
+		status = SIGILL;
+
+	if (unlikely(status > 0)) {
+		regs->cp0_epc = old_epc;		/* Undo skip-over.  */
+		regs->regs[31] = old31;
+		force_sig(status);
+	}
+
+out:
+	exception_exit(prev_state);
+}
+
+/*
+ * No lock; only written during early bootup by CPU 0.
+ */
+static RAW_NOTIFIER_HEAD(cu2_chain);
+
+int __ref register_cu2_notifier(struct notifier_block *nb)
+{
+	return raw_notifier_chain_register(&cu2_chain, nb);
+}
+
+int cu2_notifier_call_chain(unsigned long val, void *v)
+{
+	return raw_notifier_call_chain(&cu2_chain, val, v);
+}
+
+static int default_cu2_call(struct notifier_block *nfb, unsigned long action,
+	void *data)
+{
+	struct pt_regs *regs = data;
+
+	die_if_kernel("COP2: Unhandled kernel unaligned access or invalid "
+			      "instruction", regs);
+	force_sig(SIGILL);
+
+	return NOTIFY_OK;
+}
+
+#ifdef CONFIG_MIPS_FP_SUPPORT
+
+static int enable_restore_fp_context(int msa)
+{
+	int err, was_fpu_owner, prior_msa;
+	bool first_fp;
+
+	/* Initialize context if it hasn't been used already */
+	first_fp = init_fp_ctx(current);
+
+	if (first_fp) {
+		preempt_disable();
+		err = own_fpu_inatomic(1);
+		if (msa && !err) {
+			enable_msa();
+			/*
+			 * with MSA enabled, userspace can see MSACSR
+			 * and MSA regs, but the values in them are from
+			 * other task before current task, restore them
+			 * from saved fp/msa context
+			 */
+			write_msa_csr(current->thread.fpu.msacsr);
+			/*
+			 * own_fpu_inatomic(1) just restore low 64bit,
+			 * fix the high 64bit
+			 */
+			init_msa_upper();
+			set_thread_flag(TIF_USEDMSA);
+			set_thread_flag(TIF_MSA_CTX_LIVE);
+		}
+		preempt_enable();
+		return err;
+	}
+
+	/*
+	 * This task has formerly used the FP context.
+	 *
+	 * If this thread has no live MSA vector context then we can simply
+	 * restore the scalar FP context. If it has live MSA vector context
+	 * (that is, it has or may have used MSA since last performing a
+	 * function call) then we'll need to restore the vector context. This
+	 * applies even if we're currently only executing a scalar FP
+	 * instruction. This is because if we were to later execute an MSA
+	 * instruction then we'd either have to:
+	 *
+	 *  - Restore the vector context & clobber any registers modified by
+	 *    scalar FP instructions between now & then.
+	 *
+	 * or
+	 *
+	 *  - Not restore the vector context & lose the most significant bits
+	 *    of all vector registers.
+	 *
+	 * Neither of those options is acceptable. We cannot restore the least
+	 * significant bits of the registers now & only restore the most
+	 * significant bits later because the most significant bits of any
+	 * vector registers whose aliased FP register is modified now will have
+	 * been zeroed. We'd have no way to know that when restoring the vector
+	 * context & thus may load an outdated value for the most significant
+	 * bits of a vector register.
+	 */
+	if (!msa && !thread_msa_context_live())
+		return own_fpu(1);
+
+	/*
+	 * This task is using or has previously used MSA. Thus we require
+	 * that Status.FR == 1.
+	 */
+	preempt_disable();
+	was_fpu_owner = is_fpu_owner();
+	err = own_fpu_inatomic(0);
+	if (err)
+		goto out;
+
+	enable_msa();
+	write_msa_csr(current->thread.fpu.msacsr);
+	set_thread_flag(TIF_USEDMSA);
+
+	/*
+	 * If this is the first time that the task is using MSA and it has
+	 * previously used scalar FP in this time slice then we already nave
+	 * FP context which we shouldn't clobber. We do however need to clear
+	 * the upper 64b of each vector register so that this task has no
+	 * opportunity to see data left behind by another.
+	 */
+	prior_msa = test_and_set_thread_flag(TIF_MSA_CTX_LIVE);
+	if (!prior_msa && was_fpu_owner) {
+		init_msa_upper();
+
+		goto out;
+	}
+
+	if (!prior_msa) {
+		/*
+		 * Restore the least significant 64b of each vector register
+		 * from the existing scalar FP context.
+		 */
+		_restore_fp(current);
+
+		/*
+		 * The task has not formerly used MSA, so clear the upper 64b
+		 * of each vector register such that it cannot see data left
+		 * behind by another task.
+		 */
+		init_msa_upper();
+	} else {
+		/* We need to restore the vector context. */
+		restore_msa(current);
+
+		/* Restore the scalar FP control & status register */
+		if (!was_fpu_owner)
+			write_32bit_cp1_register(CP1_STATUS,
+						 current->thread.fpu.fcr31);
+	}
+
+out:
+	preempt_enable();
+
+	return 0;
+}
+
+#else /* !CONFIG_MIPS_FP_SUPPORT */
+
+static int enable_restore_fp_context(int msa)
+{
+	return SIGILL;
+}
+
+#endif /* CONFIG_MIPS_FP_SUPPORT */
+
+asmlinkage void do_cpu(struct pt_regs *regs)
+{
+	enum ctx_state prev_state;
+	unsigned int __user *epc;
+	unsigned long old_epc, old31;
+	unsigned int opcode;
+	unsigned int cpid;
+	int status;
+
+	prev_state = exception_enter();
+	cpid = (regs->cp0_cause >> CAUSEB_CE) & 3;
+
+	if (cpid != 2)
+		die_if_kernel("do_cpu invoked from kernel context!", regs);
+
+	switch (cpid) {
+	case 0:
+		epc = (unsigned int __user *)exception_epc(regs);
+		old_epc = regs->cp0_epc;
+		old31 = regs->regs[31];
+		opcode = 0;
+		status = -1;
+
+		if (unlikely(compute_return_epc(regs) < 0))
+			break;
+
+		if (!get_isa16_mode(regs->cp0_epc)) {
+			if (unlikely(get_user(opcode, epc) < 0))
+				status = SIGSEGV;
+
+			if (!cpu_has_llsc && status < 0)
+				status = simulate_llsc(regs, opcode);
+		}
+
+		if (status < 0)
+			status = SIGILL;
+
+		if (unlikely(status > 0)) {
+			regs->cp0_epc = old_epc;	/* Undo skip-over.  */
+			regs->regs[31] = old31;
+			force_sig(status);
+		}
+
+		break;
+
+#ifdef CONFIG_MIPS_FP_SUPPORT
+	case 3:
+		/*
+		 * The COP3 opcode space and consequently the CP0.Status.CU3
+		 * bit and the CP0.Cause.CE=3 encoding have been removed as
+		 * of the MIPS III ISA.  From the MIPS IV and MIPS32r2 ISAs
+		 * up the space has been reused for COP1X instructions, that
+		 * are enabled by the CP0.Status.CU1 bit and consequently
+		 * use the CP0.Cause.CE=1 encoding for Coprocessor Unusable
+		 * exceptions.  Some FPU-less processors that implement one
+		 * of these ISAs however use this code erroneously for COP1X
+		 * instructions.  Therefore we redirect this trap to the FP
+		 * emulator too.
+		 */
+		if (raw_cpu_has_fpu || !cpu_has_mips_4_5_64_r2_r6) {
+			force_sig(SIGILL);
+			break;
+		}
+		fallthrough;
+	case 1: {
+		void __user *fault_addr;
+		unsigned long fcr31;
+		int err, sig;
+
+		err = enable_restore_fp_context(0);
+
+		if (raw_cpu_has_fpu && !err)
+			break;
+
+		sig = fpu_emulator_cop1Handler(regs, &current->thread.fpu, 0,
+					       &fault_addr);
+
+		/*
+		 * We can't allow the emulated instruction to leave
+		 * any enabled Cause bits set in $fcr31.
+		 */
+		fcr31 = mask_fcr31_x(current->thread.fpu.fcr31);
+		current->thread.fpu.fcr31 &= ~fcr31;
+
+		/* Send a signal if required.  */
+		if (!process_fpemu_return(sig, fault_addr, fcr31) && !err)
+			mt_ase_fp_affinity();
+
+		break;
+	}
+#else /* CONFIG_MIPS_FP_SUPPORT */
+	case 1:
+	case 3:
+		force_sig(SIGILL);
+		break;
+#endif /* CONFIG_MIPS_FP_SUPPORT */
+
+	case 2:
+		raw_notifier_call_chain(&cu2_chain, CU2_EXCEPTION, regs);
+		break;
+	}
+
+	exception_exit(prev_state);
+}
+
+asmlinkage void do_msa_fpe(struct pt_regs *regs, unsigned int msacsr)
+{
+	enum ctx_state prev_state;
+
+	prev_state = exception_enter();
+	current->thread.trap_nr = (regs->cp0_cause >> 2) & 0x1f;
+	if (notify_die(DIE_MSAFP, "MSA FP exception", regs, 0,
+		       current->thread.trap_nr, SIGFPE) == NOTIFY_STOP)
+		goto out;
+
+	/* Clear MSACSR.Cause before enabling interrupts */
+	write_msa_csr(msacsr & ~MSA_CSR_CAUSEF);
+	local_irq_enable();
+
+	die_if_kernel("do_msa_fpe invoked from kernel context!", regs);
+	force_sig(SIGFPE);
+out:
+	exception_exit(prev_state);
+}
+
+asmlinkage void do_msa(struct pt_regs *regs)
+{
+	enum ctx_state prev_state;
+	int err;
+
+	prev_state = exception_enter();
+
+	if (!cpu_has_msa || test_thread_flag(TIF_32BIT_FPREGS)) {
+		force_sig(SIGILL);
+		goto out;
+	}
+
+	die_if_kernel("do_msa invoked from kernel context!", regs);
+
+	err = enable_restore_fp_context(1);
+	if (err)
+		force_sig(SIGILL);
+out:
+	exception_exit(prev_state);
+}
+
+asmlinkage void do_mdmx(struct pt_regs *regs)
+{
+	enum ctx_state prev_state;
+
+	prev_state = exception_enter();
+	force_sig(SIGILL);
+	exception_exit(prev_state);
+}
+
+/*
+ * Called with interrupts disabled.
+ */
+asmlinkage void do_watch(struct pt_regs *regs)
+{
+	enum ctx_state prev_state;
+
+	prev_state = exception_enter();
+	/*
+	 * Clear WP (bit 22) bit of cause register so we don't loop
+	 * forever.
+	 */
+	clear_c0_cause(CAUSEF_WP);
+
+	/*
+	 * If the current thread has the watch registers loaded, save
+	 * their values and send SIGTRAP.  Otherwise another thread
+	 * left the registers set, clear them and continue.
+	 */
+	if (test_tsk_thread_flag(current, TIF_LOAD_WATCH)) {
+		mips_read_watch_registers();
+		local_irq_enable();
+		force_sig_fault(SIGTRAP, TRAP_HWBKPT, NULL);
+	} else {
+		mips_clear_watch_registers();
+		local_irq_enable();
+	}
+	exception_exit(prev_state);
+}
+
+asmlinkage void do_mcheck(struct pt_regs *regs)
+{
+	int multi_match = regs->cp0_status & ST0_TS;
+	enum ctx_state prev_state;
+	mm_segment_t old_fs = get_fs();
+
+	prev_state = exception_enter();
+	show_regs(regs);
+
+	if (multi_match) {
+		dump_tlb_regs();
+		pr_info("\n");
+		dump_tlb_all();
+	}
+
+	if (!user_mode(regs))
+		set_fs(KERNEL_DS);
+
+	show_code((unsigned int __user *) regs->cp0_epc);
+
+	set_fs(old_fs);
+
+	/*
+	 * Some chips may have other causes of machine check (e.g. SB1
+	 * graduation timer)
+	 */
+	panic("Caught Machine Check exception - %scaused by multiple "
+	      "matching entries in the TLB.",
+	      (multi_match) ? "" : "not ");
+}
+
+asmlinkage void do_mt(struct pt_regs *regs)
+{
+	int subcode;
+
+	subcode = (read_vpe_c0_vpecontrol() & VPECONTROL_EXCPT)
+			>> VPECONTROL_EXCPT_SHIFT;
+	switch (subcode) {
+	case 0:
+		printk(KERN_DEBUG "Thread Underflow\n");
+		break;
+	case 1:
+		printk(KERN_DEBUG "Thread Overflow\n");
+		break;
+	case 2:
+		printk(KERN_DEBUG "Invalid YIELD Qualifier\n");
+		break;
+	case 3:
+		printk(KERN_DEBUG "Gating Storage Exception\n");
+		break;
+	case 4:
+		printk(KERN_DEBUG "YIELD Scheduler Exception\n");
+		break;
+	case 5:
+		printk(KERN_DEBUG "Gating Storage Scheduler Exception\n");
+		break;
+	default:
+		printk(KERN_DEBUG "*** UNKNOWN THREAD EXCEPTION %d ***\n",
+			subcode);
+		break;
+	}
+	die_if_kernel("MIPS MT Thread exception in kernel", regs);
+
+	force_sig(SIGILL);
+}
+
+
+asmlinkage void do_dsp(struct pt_regs *regs)
+{
+	if (cpu_has_dsp)
+		panic("Unexpected DSP exception");
+
+	force_sig(SIGILL);
+}
+
+asmlinkage void do_reserved(struct pt_regs *regs)
+{
+	/*
+	 * Game over - no way to handle this if it ever occurs.	 Most probably
+	 * caused by a new unknown cpu type or after another deadly
+	 * hard/software error.
+	 */
+	show_regs(regs);
+	panic("Caught reserved exception %ld - should not happen.",
+	      (regs->cp0_cause & 0x7f) >> 2);
+}
+
+static int __initdata l1parity = 1;
+static int __init nol1parity(char *s)
+{
+	l1parity = 0;
+	return 1;
+}
+__setup("nol1par", nol1parity);
+static int __initdata l2parity = 1;
+static int __init nol2parity(char *s)
+{
+	l2parity = 0;
+	return 1;
+}
+__setup("nol2par", nol2parity);
+
+/*
+ * Some MIPS CPUs can enable/disable for cache parity detection, but do
+ * it different ways.
+ */
+static inline __init void parity_protection_init(void)
+{
+#define ERRCTL_PE	0x80000000
+#define ERRCTL_L2P	0x00800000
+
+	if (mips_cm_revision() >= CM_REV_CM3) {
+		ulong gcr_ectl, cp0_ectl;
+
+		/*
+		 * With CM3 systems we need to ensure that the L1 & L2
+		 * parity enables are set to the same value, since this
+		 * is presumed by the hardware engineers.
+		 *
+		 * If the user disabled either of L1 or L2 ECC checking,
+		 * disable both.
+		 */
+		l1parity &= l2parity;
+		l2parity &= l1parity;
+
+		/* Probe L1 ECC support */
+		cp0_ectl = read_c0_ecc();
+		write_c0_ecc(cp0_ectl | ERRCTL_PE);
+		back_to_back_c0_hazard();
+		cp0_ectl = read_c0_ecc();
+
+		/* Probe L2 ECC support */
+		gcr_ectl = read_gcr_err_control();
+
+		if (!(gcr_ectl & CM_GCR_ERR_CONTROL_L2_ECC_SUPPORT) ||
+		    !(cp0_ectl & ERRCTL_PE)) {
+			/*
+			 * One of L1 or L2 ECC checking isn't supported,
+			 * so we cannot enable either.
+			 */
+			l1parity = l2parity = 0;
+		}
+
+		/* Configure L1 ECC checking */
+		if (l1parity)
+			cp0_ectl |= ERRCTL_PE;
+		else
+			cp0_ectl &= ~ERRCTL_PE;
+		write_c0_ecc(cp0_ectl);
+		back_to_back_c0_hazard();
+		WARN_ON(!!(read_c0_ecc() & ERRCTL_PE) != l1parity);
+
+		/* Configure L2 ECC checking */
+		if (l2parity)
+			gcr_ectl |= CM_GCR_ERR_CONTROL_L2_ECC_EN;
+		else
+			gcr_ectl &= ~CM_GCR_ERR_CONTROL_L2_ECC_EN;
+		write_gcr_err_control(gcr_ectl);
+		gcr_ectl = read_gcr_err_control();
+		gcr_ectl &= CM_GCR_ERR_CONTROL_L2_ECC_EN;
+		WARN_ON(!!gcr_ectl != l2parity);
+
+		pr_info("Cache parity protection %sabled\n",
+			l1parity ? "en" : "dis");
+		return;
+	}
+
+	switch (current_cpu_type()) {
+	case CPU_24K:
+	case CPU_34K:
+	case CPU_74K:
+	case CPU_1004K:
+	case CPU_1074K:
+	case CPU_INTERAPTIV:
+	case CPU_PROAPTIV:
+	case CPU_P5600:
+	case CPU_QEMU_GENERIC:
+	case CPU_P6600:
+		{
+			unsigned long errctl;
+			unsigned int l1parity_present, l2parity_present;
+
+			errctl = read_c0_ecc();
+			errctl &= ~(ERRCTL_PE|ERRCTL_L2P);
+
+			/* probe L1 parity support */
+			write_c0_ecc(errctl | ERRCTL_PE);
+			back_to_back_c0_hazard();
+			l1parity_present = (read_c0_ecc() & ERRCTL_PE);
+
+			/* probe L2 parity support */
+			write_c0_ecc(errctl|ERRCTL_L2P);
+			back_to_back_c0_hazard();
+			l2parity_present = (read_c0_ecc() & ERRCTL_L2P);
+
+			if (l1parity_present && l2parity_present) {
+				if (l1parity)
+					errctl |= ERRCTL_PE;
+				if (l1parity ^ l2parity)
+					errctl |= ERRCTL_L2P;
+			} else if (l1parity_present) {
+				if (l1parity)
+					errctl |= ERRCTL_PE;
+			} else if (l2parity_present) {
+				if (l2parity)
+					errctl |= ERRCTL_L2P;
+			} else {
+				/* No parity available */
+			}
+
+			printk(KERN_INFO "Writing ErrCtl register=%08lx\n", errctl);
+
+			write_c0_ecc(errctl);
+			back_to_back_c0_hazard();
+			errctl = read_c0_ecc();
+			printk(KERN_INFO "Readback ErrCtl register=%08lx\n", errctl);
+
+			if (l1parity_present)
+				printk(KERN_INFO "Cache parity protection %sabled\n",
+				       (errctl & ERRCTL_PE) ? "en" : "dis");
+
+			if (l2parity_present) {
+				if (l1parity_present && l1parity)
+					errctl ^= ERRCTL_L2P;
+				printk(KERN_INFO "L2 cache parity protection %sabled\n",
+				       (errctl & ERRCTL_L2P) ? "en" : "dis");
+			}
+		}
+		break;
+
+	case CPU_5KC:
+	case CPU_5KE:
+	case CPU_LOONGSON32:
+		write_c0_ecc(0x80000000);
+		back_to_back_c0_hazard();
+		/* Set the PE bit (bit 31) in the c0_errctl register. */
+		printk(KERN_INFO "Cache parity protection %sabled\n",
+		       (read_c0_ecc() & 0x80000000) ? "en" : "dis");
+		break;
+	case CPU_20KC:
+	case CPU_25KF:
+		/* Clear the DE bit (bit 16) in the c0_status register. */
+		printk(KERN_INFO "Enable cache parity protection for "
+		       "MIPS 20KC/25KF CPUs.\n");
+		clear_c0_status(ST0_DE);
+		break;
+	default:
+		break;
+	}
+}
+
+asmlinkage void cache_parity_error(void)
+{
+	const int field = 2 * sizeof(unsigned long);
+	unsigned int reg_val;
+
+	/* For the moment, report the problem and hang. */
+	printk("Cache error exception:\n");
+	printk("cp0_errorepc == %0*lx\n", field, read_c0_errorepc());
+	reg_val = read_c0_cacheerr();
+	printk("c0_cacheerr == %08x\n", reg_val);
+
+	printk("Decoded c0_cacheerr: %s cache fault in %s reference.\n",
+	       reg_val & (1<<30) ? "secondary" : "primary",
+	       reg_val & (1<<31) ? "data" : "insn");
+	if ((cpu_has_mips_r2_r6) &&
+	    ((current_cpu_data.processor_id & 0xff0000) == PRID_COMP_MIPS)) {
+		pr_err("Error bits: %s%s%s%s%s%s%s%s\n",
+			reg_val & (1<<29) ? "ED " : "",
+			reg_val & (1<<28) ? "ET " : "",
+			reg_val & (1<<27) ? "ES " : "",
+			reg_val & (1<<26) ? "EE " : "",
+			reg_val & (1<<25) ? "EB " : "",
+			reg_val & (1<<24) ? "EI " : "",
+			reg_val & (1<<23) ? "E1 " : "",
+			reg_val & (1<<22) ? "E0 " : "");
+	} else {
+		pr_err("Error bits: %s%s%s%s%s%s%s\n",
+			reg_val & (1<<29) ? "ED " : "",
+			reg_val & (1<<28) ? "ET " : "",
+			reg_val & (1<<26) ? "EE " : "",
+			reg_val & (1<<25) ? "EB " : "",
+			reg_val & (1<<24) ? "EI " : "",
+			reg_val & (1<<23) ? "E1 " : "",
+			reg_val & (1<<22) ? "E0 " : "");
+	}
+	printk("IDX: 0x%08x\n", reg_val & ((1<<22)-1));
+
+#if defined(CONFIG_CPU_MIPS32) || defined(CONFIG_CPU_MIPS64)
+	if (reg_val & (1<<22))
+		printk("DErrAddr0: 0x%0*lx\n", field, read_c0_derraddr0());
+
+	if (reg_val & (1<<23))
+		printk("DErrAddr1: 0x%0*lx\n", field, read_c0_derraddr1());
+#endif
+
+	panic("Can't handle the cache error!");
+}
+
+asmlinkage void do_ftlb(void)
+{
+	const int field = 2 * sizeof(unsigned long);
+	unsigned int reg_val;
+
+	/* For the moment, report the problem and hang. */
+	if ((cpu_has_mips_r2_r6) &&
+	    (((current_cpu_data.processor_id & 0xff0000) == PRID_COMP_MIPS) ||
+	    ((current_cpu_data.processor_id & 0xff0000) == PRID_COMP_LOONGSON))) {
+		pr_err("FTLB error exception, cp0_ecc=0x%08x:\n",
+		       read_c0_ecc());
+		pr_err("cp0_errorepc == %0*lx\n", field, read_c0_errorepc());
+		reg_val = read_c0_cacheerr();
+		pr_err("c0_cacheerr == %08x\n", reg_val);
+
+		if ((reg_val & 0xc0000000) == 0xc0000000) {
+			pr_err("Decoded c0_cacheerr: FTLB parity error\n");
+		} else {
+			pr_err("Decoded c0_cacheerr: %s cache fault in %s reference.\n",
+			       reg_val & (1<<30) ? "secondary" : "primary",
+			       reg_val & (1<<31) ? "data" : "insn");
+		}
+	} else {
+		pr_err("FTLB error exception\n");
+	}
+	/* Just print the cacheerr bits for now */
+	cache_parity_error();
+}
+
+asmlinkage void do_gsexc(struct pt_regs *regs, u32 diag1)
+{
+	u32 exccode = (diag1 & LOONGSON_DIAG1_EXCCODE) >>
+			LOONGSON_DIAG1_EXCCODE_SHIFT;
+	enum ctx_state prev_state;
+
+	prev_state = exception_enter();
+
+	switch (exccode) {
+	case 0x08:
+		/* Undocumented exception, will trigger on certain
+		 * also-undocumented instructions accessible from userspace.
+		 * Processor state is not otherwise corrupted, but currently
+		 * we don't know how to proceed. Maybe there is some
+		 * undocumented control flag to enable the instructions?
+		 */
+		force_sig(SIGILL);
+		break;
+
+	default:
+		/* None of the other exceptions, documented or not, have
+		 * further details given; none are encountered in the wild
+		 * either. Panic in case some of them turn out to be fatal.
+		 */
+		show_regs(regs);
+		panic("Unhandled Loongson exception - GSCause = %08x", diag1);
+	}
+
+	exception_exit(prev_state);
+}
+
+/*
+ * SDBBP EJTAG debug exception handler.
+ * We skip the instruction and return to the next instruction.
+ */
+void ejtag_exception_handler(struct pt_regs *regs)
+{
+	const int field = 2 * sizeof(unsigned long);
+	unsigned long depc, old_epc, old_ra;
+	unsigned int debug;
+
+	printk(KERN_DEBUG "SDBBP EJTAG debug exception - not handled yet, just ignored!\n");
+	depc = read_c0_depc();
+	debug = read_c0_debug();
+	printk(KERN_DEBUG "c0_depc = %0*lx, DEBUG = %08x\n", field, depc, debug);
+	if (debug & 0x80000000) {
+		/*
+		 * In branch delay slot.
+		 * We cheat a little bit here and use EPC to calculate the
+		 * debug return address (DEPC). EPC is restored after the
+		 * calculation.
+		 */
+		old_epc = regs->cp0_epc;
+		old_ra = regs->regs[31];
+		regs->cp0_epc = depc;
+		compute_return_epc(regs);
+		depc = regs->cp0_epc;
+		regs->cp0_epc = old_epc;
+		regs->regs[31] = old_ra;
+	} else
+		depc += 4;
+	write_c0_depc(depc);
+
+#if 0
+	printk(KERN_DEBUG "\n\n----- Enable EJTAG single stepping ----\n\n");
+	write_c0_debug(debug | 0x100);
+#endif
+}
+
+/*
+ * NMI exception handler.
+ * No lock; only written during early bootup by CPU 0.
+ */
+static RAW_NOTIFIER_HEAD(nmi_chain);
+
+int register_nmi_notifier(struct notifier_block *nb)
+{
+	return raw_notifier_chain_register(&nmi_chain, nb);
+}
+
+void __noreturn nmi_exception_handler(struct pt_regs *regs)
+{
+	char str[100];
+
+	nmi_enter();
+	raw_notifier_call_chain(&nmi_chain, 0, regs);
+	bust_spinlocks(1);
+	snprintf(str, 100, "CPU%d NMI taken, CP0_EPC=%lx\n",
+		 smp_processor_id(), regs->cp0_epc);
+	regs->cp0_epc = read_c0_errorepc();
+	die(str, regs);
+	nmi_exit();
+}
+
+#define VECTORSPACING 0x100	/* for EI/VI mode */
+
+unsigned long ebase;
+EXPORT_SYMBOL_GPL(ebase);
+unsigned long exception_handlers[32];
+unsigned long vi_handlers[64];
+
+void __init *set_except_vector(int n, void *addr)
+{
+	unsigned long handler = (unsigned long) addr;
+	unsigned long old_handler;
+
+#ifdef CONFIG_CPU_MICROMIPS
+	/*
+	 * Only the TLB handlers are cache aligned with an even
+	 * address. All other handlers are on an odd address and
+	 * require no modification. Otherwise, MIPS32 mode will
+	 * be entered when handling any TLB exceptions. That
+	 * would be bad...since we must stay in microMIPS mode.
+	 */
+	if (!(handler & 0x1))
+		handler |= 1;
+#endif
+	old_handler = xchg(&exception_handlers[n], handler);
+
+	if (n == 0 && cpu_has_divec) {
+#ifdef CONFIG_CPU_MICROMIPS
+		unsigned long jump_mask = ~((1 << 27) - 1);
+#else
+		unsigned long jump_mask = ~((1 << 28) - 1);
+#endif
+		u32 *buf = (u32 *)(ebase + 0x200);
+		unsigned int k0 = 26;
+		if ((handler & jump_mask) == ((ebase + 0x200) & jump_mask)) {
+			uasm_i_j(&buf, handler & ~jump_mask);
+			uasm_i_nop(&buf);
+		} else {
+			UASM_i_LA(&buf, k0, handler);
+			uasm_i_jr(&buf, k0);
+			uasm_i_nop(&buf);
+		}
+		local_flush_icache_range(ebase + 0x200, (unsigned long)buf);
+	}
+	return (void *)old_handler;
+}
+
+static void do_default_vi(void)
+{
+	show_regs(get_irq_regs());
+	panic("Caught unexpected vectored interrupt.");
+}
+
+static void *set_vi_srs_handler(int n, vi_handler_t addr, int srs)
+{
+	unsigned long handler;
+	unsigned long old_handler = vi_handlers[n];
+	int srssets = current_cpu_data.srsets;
+	u16 *h;
+	unsigned char *b;
+
+	BUG_ON(!cpu_has_veic && !cpu_has_vint);
+
+	if (addr == NULL) {
+		handler = (unsigned long) do_default_vi;
+		srs = 0;
+	} else
+		handler = (unsigned long) addr;
+	vi_handlers[n] = handler;
+
+	b = (unsigned char *)(ebase + 0x200 + n*VECTORSPACING);
+
+	if (srs >= srssets)
+		panic("Shadow register set %d not supported", srs);
+
+	if (cpu_has_veic) {
+		if (board_bind_eic_interrupt)
+			board_bind_eic_interrupt(n, srs);
+	} else if (cpu_has_vint) {
+		/* SRSMap is only defined if shadow sets are implemented */
+		if (srssets > 1)
+			change_c0_srsmap(0xf << n*4, srs << n*4);
+	}
+
+	if (srs == 0) {
+		/*
+		 * If no shadow set is selected then use the default handler
+		 * that does normal register saving and standard interrupt exit
+		 */
+		extern const u8 except_vec_vi[], except_vec_vi_lui[];
+		extern const u8 except_vec_vi_ori[], except_vec_vi_end[];
+		extern const u8 rollback_except_vec_vi[];
+		const u8 *vec_start = using_rollback_handler() ?
+				      rollback_except_vec_vi : except_vec_vi;
+#if defined(CONFIG_CPU_MICROMIPS) || defined(CONFIG_CPU_BIG_ENDIAN)
+		const int lui_offset = except_vec_vi_lui - vec_start + 2;
+		const int ori_offset = except_vec_vi_ori - vec_start + 2;
+#else
+		const int lui_offset = except_vec_vi_lui - vec_start;
+		const int ori_offset = except_vec_vi_ori - vec_start;
+#endif
+		const int handler_len = except_vec_vi_end - vec_start;
+
+		if (handler_len > VECTORSPACING) {
+			/*
+			 * Sigh... panicing won't help as the console
+			 * is probably not configured :(
+			 */
+			panic("VECTORSPACING too small");
+		}
+
+		set_handler(((unsigned long)b - ebase), vec_start,
+#ifdef CONFIG_CPU_MICROMIPS
+				(handler_len - 1));
+#else
+				handler_len);
+#endif
+		h = (u16 *)(b + lui_offset);
+		*h = (handler >> 16) & 0xffff;
+		h = (u16 *)(b + ori_offset);
+		*h = (handler & 0xffff);
+		local_flush_icache_range((unsigned long)b,
+					 (unsigned long)(b+handler_len));
+	}
+	else {
+		/*
+		 * In other cases jump directly to the interrupt handler. It
+		 * is the handler's responsibility to save registers if required
+		 * (eg hi/lo) and return from the exception using "eret".
+		 */
+		u32 insn;
+
+		h = (u16 *)b;
+		/* j handler */
+#ifdef CONFIG_CPU_MICROMIPS
+		insn = 0xd4000000 | (((u32)handler & 0x07ffffff) >> 1);
+#else
+		insn = 0x08000000 | (((u32)handler & 0x0fffffff) >> 2);
+#endif
+		h[0] = (insn >> 16) & 0xffff;
+		h[1] = insn & 0xffff;
+		h[2] = 0;
+		h[3] = 0;
+		local_flush_icache_range((unsigned long)b,
+					 (unsigned long)(b+8));
+	}
+
+	return (void *)old_handler;
+}
+
+void *set_vi_handler(int n, vi_handler_t addr)
+{
+	return set_vi_srs_handler(n, addr, 0);
+}
+
+extern void tlb_init(void);
+
+/*
+ * Timer interrupt
+ */
+int cp0_compare_irq;
+EXPORT_SYMBOL_GPL(cp0_compare_irq);
+int cp0_compare_irq_shift;
+
+/*
+ * Performance counter IRQ or -1 if shared with timer
+ */
+int cp0_perfcount_irq;
+EXPORT_SYMBOL_GPL(cp0_perfcount_irq);
+
+/*
+ * Fast debug channel IRQ or -1 if not present
+ */
+int cp0_fdc_irq;
+EXPORT_SYMBOL_GPL(cp0_fdc_irq);
+
+static int noulri;
+
+static int __init ulri_disable(char *s)
+{
+	pr_info("Disabling ulri\n");
+	noulri = 1;
+
+	return 1;
+}
+__setup("noulri", ulri_disable);
+
+/* configure STATUS register */
+static void configure_status(void)
+{
+	/*
+	 * Disable coprocessors and select 32-bit or 64-bit addressing
+	 * and the 16/32 or 32/32 FPR register model.  Reset the BEV
+	 * flag that some firmware may have left set and the TS bit (for
+	 * IP27).  Set XX for ISA IV code to work.
+	 */
+	unsigned int status_set = ST0_KERNEL_CUMASK;
+#ifdef CONFIG_64BIT
+	status_set |= ST0_FR|ST0_KX|ST0_SX|ST0_UX;
+#endif
+	if (current_cpu_data.isa_level & MIPS_CPU_ISA_IV)
+		status_set |= ST0_XX;
+	if (cpu_has_dsp)
+		status_set |= ST0_MX;
+
+	change_c0_status(ST0_CU|ST0_MX|ST0_RE|ST0_FR|ST0_BEV|ST0_TS|ST0_KX|ST0_SX|ST0_UX,
+			 status_set);
+	back_to_back_c0_hazard();
+}
+
+unsigned int hwrena;
+EXPORT_SYMBOL_GPL(hwrena);
+
+/* configure HWRENA register */
+static void configure_hwrena(void)
+{
+	hwrena = cpu_hwrena_impl_bits;
+
+	if (cpu_has_mips_r2_r6)
+		hwrena |= MIPS_HWRENA_CPUNUM |
+			  MIPS_HWRENA_SYNCISTEP |
+			  MIPS_HWRENA_CC |
+			  MIPS_HWRENA_CCRES;
+
+	if (!noulri && cpu_has_userlocal)
+		hwrena |= MIPS_HWRENA_ULR;
+
+	if (hwrena)
+		write_c0_hwrena(hwrena);
+}
+
+static void configure_exception_vector(void)
+{
+	if (cpu_has_mips_r2_r6) {
+		unsigned long sr = set_c0_status(ST0_BEV);
+		/* If available, use WG to set top bits of EBASE */
+		if (cpu_has_ebase_wg) {
+#ifdef CONFIG_64BIT
+			write_c0_ebase_64(ebase | MIPS_EBASE_WG);
+#else
+			write_c0_ebase(ebase | MIPS_EBASE_WG);
+#endif
+		}
+		write_c0_ebase(ebase);
+		write_c0_status(sr);
+	}
+	if (cpu_has_veic || cpu_has_vint) {
+		/* Setting vector spacing enables EI/VI mode  */
+		change_c0_intctl(0x3e0, VECTORSPACING);
+	}
+	if (cpu_has_divec) {
+		if (cpu_has_mipsmt) {
+			unsigned int vpflags = dvpe();
+			set_c0_cause(CAUSEF_IV);
+			evpe(vpflags);
+		} else
+			set_c0_cause(CAUSEF_IV);
+	}
+}
+
+void per_cpu_trap_init(bool is_boot_cpu)
+{
+	unsigned int cpu = smp_processor_id();
+
+	configure_status();
+	configure_hwrena();
+
+	configure_exception_vector();
+
+	/*
+	 * Before R2 both interrupt numbers were fixed to 7, so on R2 only:
+	 *
+	 *  o read IntCtl.IPTI to determine the timer interrupt
+	 *  o read IntCtl.IPPCI to determine the performance counter interrupt
+	 *  o read IntCtl.IPFDC to determine the fast debug channel interrupt
+	 */
+	if (cpu_has_mips_r2_r6) {
+		cp0_compare_irq_shift = CAUSEB_TI - CAUSEB_IP;
+		cp0_compare_irq = (read_c0_intctl() >> INTCTLB_IPTI) & 7;
+		cp0_perfcount_irq = (read_c0_intctl() >> INTCTLB_IPPCI) & 7;
+		cp0_fdc_irq = (read_c0_intctl() >> INTCTLB_IPFDC) & 7;
+		if (!cp0_fdc_irq)
+			cp0_fdc_irq = -1;
+
+	} else {
+		cp0_compare_irq = CP0_LEGACY_COMPARE_IRQ;
+		cp0_compare_irq_shift = CP0_LEGACY_PERFCNT_IRQ;
+		cp0_perfcount_irq = -1;
+		cp0_fdc_irq = -1;
+	}
+
+	if (cpu_has_mmid)
+		cpu_data[cpu].asid_cache = 0;
+	else if (!cpu_data[cpu].asid_cache)
+		cpu_data[cpu].asid_cache = asid_first_version(cpu);
+
+	mmgrab(&init_mm);
+	current->active_mm = &init_mm;
+	BUG_ON(current->mm);
+	enter_lazy_tlb(&init_mm, current);
+
+	/* Boot CPU's cache setup in setup_arch(). */
+	if (!is_boot_cpu)
+		cpu_cache_init();
+	tlb_init();
+	TLBMISS_HANDLER_SETUP();
+}
+
+/* Install CPU exception handler */
+void set_handler(unsigned long offset, const void *addr, unsigned long size)
+{
+#ifdef CONFIG_CPU_MICROMIPS
+	memcpy((void *)(ebase + offset), ((unsigned char *)addr - 1), size);
+#else
+	memcpy((void *)(ebase + offset), addr, size);
+#endif
+	local_flush_icache_range(ebase + offset, ebase + offset + size);
+}
+
+static const char panic_null_cerr[] =
+	"Trying to set NULL cache error exception handler\n";
+
+/*
+ * Install uncached CPU exception handler.
+ * This is suitable only for the cache error exception which is the only
+ * exception handler that is being run uncached.
+ */
+void set_uncached_handler(unsigned long offset, void *addr,
+	unsigned long size)
+{
+	unsigned long uncached_ebase = CKSEG1ADDR(ebase);
+
+	if (!addr)
+		panic(panic_null_cerr);
+
+	memcpy((void *)(uncached_ebase + offset), addr, size);
+}
+
+static int __initdata rdhwr_noopt;
+static int __init set_rdhwr_noopt(char *str)
+{
+	rdhwr_noopt = 1;
+	return 1;
+}
+
+__setup("rdhwr_noopt", set_rdhwr_noopt);
+
+void __init trap_init(void)
+{
+	extern char except_vec3_generic;
+	extern char except_vec4;
+	extern char except_vec3_r4000;
+	unsigned long i, vec_size;
+	phys_addr_t ebase_pa;
+
+	check_wait();
+
+	if (!cpu_has_mips_r2_r6) {
+		ebase = CAC_BASE;
+		ebase_pa = virt_to_phys((void *)ebase);
+		vec_size = 0x400;
+
+		memblock_reserve(ebase_pa, vec_size);
+	} else {
+		if (cpu_has_veic || cpu_has_vint)
+			vec_size = 0x200 + VECTORSPACING*64;
+		else
+			vec_size = PAGE_SIZE;
+
+		ebase_pa = memblock_phys_alloc(vec_size, 1 << fls(vec_size));
+		if (!ebase_pa)
+			panic("%s: Failed to allocate %lu bytes align=0x%x\n",
+			      __func__, vec_size, 1 << fls(vec_size));
+
+		/*
+		 * Try to ensure ebase resides in KSeg0 if possible.
+		 *
+		 * It shouldn't generally be in XKPhys on MIPS64 to avoid
+		 * hitting a poorly defined exception base for Cache Errors.
+		 * The allocation is likely to be in the low 512MB of physical,
+		 * in which case we should be able to convert to KSeg0.
+		 *
+		 * EVA is special though as it allows segments to be rearranged
+		 * and to become uncached during cache error handling.
+		 */
+		if (!IS_ENABLED(CONFIG_EVA) && !WARN_ON(ebase_pa >= 0x20000000))
+			ebase = CKSEG0ADDR(ebase_pa);
+		else
+			ebase = (unsigned long)phys_to_virt(ebase_pa);
+	}
+
+	if (cpu_has_mmips) {
+		unsigned int config3 = read_c0_config3();
+
+		if (IS_ENABLED(CONFIG_CPU_MICROMIPS))
+			write_c0_config3(config3 | MIPS_CONF3_ISA_OE);
+		else
+			write_c0_config3(config3 & ~MIPS_CONF3_ISA_OE);
+	}
+
+	if (board_ebase_setup)
+		board_ebase_setup();
+	per_cpu_trap_init(true);
+	memblock_set_bottom_up(false);
+
+	/*
+	 * Copy the generic exception handlers to their final destination.
+	 * This will be overridden later as suitable for a particular
+	 * configuration.
+	 */
+	set_handler(0x180, &except_vec3_generic, 0x80);
+
+	/*
+	 * Setup default vectors
+	 */
+	for (i = 0; i <= 31; i++)
+		set_except_vector(i, handle_reserved);
+
+	/*
+	 * Copy the EJTAG debug exception vector handler code to it's final
+	 * destination.
+	 */
+	if (cpu_has_ejtag && board_ejtag_handler_setup)
+		board_ejtag_handler_setup();
+
+	/*
+	 * Only some CPUs have the watch exceptions.
+	 */
+	if (cpu_has_watch)
+		set_except_vector(EXCCODE_WATCH, handle_watch);
+
+	/*
+	 * Initialise interrupt handlers
+	 */
+	if (cpu_has_veic || cpu_has_vint) {
+		int nvec = cpu_has_veic ? 64 : 8;
+		for (i = 0; i < nvec; i++)
+			set_vi_handler(i, NULL);
+	}
+	else if (cpu_has_divec)
+		set_handler(0x200, &except_vec4, 0x8);
+
+	/*
+	 * Some CPUs can enable/disable for cache parity detection, but does
+	 * it different ways.
+	 */
+	parity_protection_init();
+
+	/*
+	 * The Data Bus Errors / Instruction Bus Errors are signaled
+	 * by external hardware.  Therefore these two exceptions
+	 * may have board specific handlers.
+	 */
+	if (board_be_init)
+		board_be_init();
+
+	set_except_vector(EXCCODE_INT, using_rollback_handler() ?
+					rollback_handle_int : handle_int);
+	set_except_vector(EXCCODE_MOD, handle_tlbm);
+	set_except_vector(EXCCODE_TLBL, handle_tlbl);
+	set_except_vector(EXCCODE_TLBS, handle_tlbs);
+
+	set_except_vector(EXCCODE_ADEL, handle_adel);
+	set_except_vector(EXCCODE_ADES, handle_ades);
+
+	set_except_vector(EXCCODE_IBE, handle_ibe);
+	set_except_vector(EXCCODE_DBE, handle_dbe);
+
+	set_except_vector(EXCCODE_SYS, handle_sys);
+	set_except_vector(EXCCODE_BP, handle_bp);
+
+	if (rdhwr_noopt)
+		set_except_vector(EXCCODE_RI, handle_ri);
+	else {
+		if (cpu_has_vtag_icache)
+			set_except_vector(EXCCODE_RI, handle_ri_rdhwr_tlbp);
+		else if (current_cpu_type() == CPU_LOONGSON64)
+			set_except_vector(EXCCODE_RI, handle_ri_rdhwr_tlbp);
+		else
+			set_except_vector(EXCCODE_RI, handle_ri_rdhwr);
+	}
+
+	set_except_vector(EXCCODE_CPU, handle_cpu);
+	set_except_vector(EXCCODE_OV, handle_ov);
+	set_except_vector(EXCCODE_TR, handle_tr);
+	set_except_vector(EXCCODE_MSAFPE, handle_msa_fpe);
+
+	if (board_nmi_handler_setup)
+		board_nmi_handler_setup();
+
+	if (cpu_has_fpu && !cpu_has_nofpuex)
+		set_except_vector(EXCCODE_FPE, handle_fpe);
+
+	if (cpu_has_ftlbparex)
+		set_except_vector(MIPS_EXCCODE_TLBPAR, handle_ftlb);
+
+	if (cpu_has_gsexcex)
+		set_except_vector(LOONGSON_EXCCODE_GSEXC, handle_gsexc);
+
+	if (cpu_has_rixiex) {
+		set_except_vector(EXCCODE_TLBRI, tlb_do_page_fault_0);
+		set_except_vector(EXCCODE_TLBXI, tlb_do_page_fault_0);
+	}
+
+	set_except_vector(EXCCODE_MSADIS, handle_msa);
+	set_except_vector(EXCCODE_MDMX, handle_mdmx);
+
+	if (cpu_has_mcheck)
+		set_except_vector(EXCCODE_MCHECK, handle_mcheck);
+
+	if (cpu_has_mipsmt)
+		set_except_vector(EXCCODE_THREAD, handle_mt);
+
+	set_except_vector(EXCCODE_DSPDIS, handle_dsp);
+
+	if (board_cache_error_setup)
+		board_cache_error_setup();
+
+	if (cpu_has_vce)
+		/* Special exception: R4[04]00 uses also the divec space. */
+		set_handler(0x180, &except_vec3_r4000, 0x100);
+	else if (cpu_has_4kex)
+		set_handler(0x180, &except_vec3_generic, 0x80);
+	else
+		set_handler(0x080, &except_vec3_generic, 0x80);
+
+	local_flush_icache_range(ebase, ebase + vec_size);
+
+	sort_extable(__start___dbe_table, __stop___dbe_table);
+
+	cu2_notifier(default_cu2_call, 0x80000000);	/* Run last  */
+}
+
+static int trap_pm_notifier(struct notifier_block *self, unsigned long cmd,
+			    void *v)
+{
+	switch (cmd) {
+	case CPU_PM_ENTER_FAILED:
+	case CPU_PM_EXIT:
+		configure_status();
+		configure_hwrena();
+		configure_exception_vector();
+
+		/* Restore register with CPU number for TLB handlers */
+		TLBMISS_HANDLER_RESTORE();
+
+		break;
+	}
+
+	return NOTIFY_OK;
+}
+
+static struct notifier_block trap_pm_notifier_block = {
+	.notifier_call = trap_pm_notifier,
+};
+
+static int __init trap_pm_init(void)
+{
+	return cpu_pm_register_notifier(&trap_pm_notifier_block);
+}
+arch_initcall(trap_pm_init);
diff --git a/arch/mips/kernel/unaligned.c b/arch/mips/kernel/unaligned.c
new file mode 100644
index 000000000..126a5f3f4
--- /dev/null
+++ b/arch/mips/kernel/unaligned.c
@@ -0,0 +1,1610 @@
+/*
+ * Handle unaligned accesses by emulation.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1996, 1998, 1999, 2002 by Ralf Baechle
+ * Copyright (C) 1999 Silicon Graphics, Inc.
+ * Copyright (C) 2014 Imagination Technologies Ltd.
+ *
+ * This file contains exception handler for address error exception with the
+ * special capability to execute faulting instructions in software.  The
+ * handler does not try to handle the case when the program counter points
+ * to an address not aligned to a word boundary.
+ *
+ * Putting data to unaligned addresses is a bad practice even on Intel where
+ * only the performance is affected.  Much worse is that such code is non-
+ * portable.  Due to several programs that die on MIPS due to alignment
+ * problems I decided to implement this handler anyway though I originally
+ * didn't intend to do this at all for user code.
+ *
+ * For now I enable fixing of address errors by default to make life easier.
+ * I however intend to disable this somewhen in the future when the alignment
+ * problems with user programs have been fixed.	 For programmers this is the
+ * right way to go.
+ *
+ * Fixing address errors is a per process option.  The option is inherited
+ * across fork(2) and execve(2) calls.	If you really want to use the
+ * option in your user programs - I discourage the use of the software
+ * emulation strongly - use the following code in your userland stuff:
+ *
+ * #include <sys/sysmips.h>
+ *
+ * ...
+ * sysmips(MIPS_FIXADE, x);
+ * ...
+ *
+ * The argument x is 0 for disabling software emulation, enabled otherwise.
+ *
+ * Below a little program to play around with this feature.
+ *
+ * #include <stdio.h>
+ * #include <sys/sysmips.h>
+ *
+ * struct foo {
+ *	   unsigned char bar[8];
+ * };
+ *
+ * main(int argc, char *argv[])
+ * {
+ *	   struct foo x = {0, 1, 2, 3, 4, 5, 6, 7};
+ *	   unsigned int *p = (unsigned int *) (x.bar + 3);
+ *	   int i;
+ *
+ *	   if (argc > 1)
+ *		   sysmips(MIPS_FIXADE, atoi(argv[1]));
+ *
+ *	   printf("*p = %08lx\n", *p);
+ *
+ *	   *p = 0xdeadface;
+ *
+ *	   for(i = 0; i <= 7; i++)
+ *	   printf("%02x ", x.bar[i]);
+ *	   printf("\n");
+ * }
+ *
+ * Coprocessor loads are not supported; I think this case is unimportant
+ * in the practice.
+ *
+ * TODO: Handle ndc (attempted store to doubleword in uncached memory)
+ *	 exception for the R6000.
+ *	 A store crossing a page boundary might be executed only partially.
+ *	 Undo the partial store in this case.
+ */
+#include <linux/context_tracking.h>
+#include <linux/mm.h>
+#include <linux/signal.h>
+#include <linux/smp.h>
+#include <linux/sched.h>
+#include <linux/debugfs.h>
+#include <linux/perf_event.h>
+
+#include <asm/asm.h>
+#include <asm/branch.h>
+#include <asm/byteorder.h>
+#include <asm/cop2.h>
+#include <asm/debug.h>
+#include <asm/fpu.h>
+#include <asm/fpu_emulator.h>
+#include <asm/inst.h>
+#include <asm/unaligned-emul.h>
+#include <asm/mmu_context.h>
+#include <linux/uaccess.h>
+
+enum {
+	UNALIGNED_ACTION_QUIET,
+	UNALIGNED_ACTION_SIGNAL,
+	UNALIGNED_ACTION_SHOW,
+};
+#ifdef CONFIG_DEBUG_FS
+static u32 unaligned_instructions;
+static u32 unaligned_action;
+#else
+#define unaligned_action UNALIGNED_ACTION_QUIET
+#endif
+extern void show_registers(struct pt_regs *regs);
+
+static void emulate_load_store_insn(struct pt_regs *regs,
+	void __user *addr, unsigned int __user *pc)
+{
+	unsigned long origpc, orig31, value;
+	union mips_instruction insn;
+	unsigned int res;
+#ifdef	CONFIG_EVA
+	mm_segment_t seg;
+#endif
+	origpc = (unsigned long)pc;
+	orig31 = regs->regs[31];
+
+	perf_sw_event(PERF_COUNT_SW_EMULATION_FAULTS, 1, regs, 0);
+
+	/*
+	 * This load never faults.
+	 */
+	__get_user(insn.word, pc);
+
+	switch (insn.i_format.opcode) {
+		/*
+		 * These are instructions that a compiler doesn't generate.  We
+		 * can assume therefore that the code is MIPS-aware and
+		 * really buggy.  Emulating these instructions would break the
+		 * semantics anyway.
+		 */
+	case ll_op:
+	case lld_op:
+	case sc_op:
+	case scd_op:
+
+		/*
+		 * For these instructions the only way to create an address
+		 * error is an attempted access to kernel/supervisor address
+		 * space.
+		 */
+	case ldl_op:
+	case ldr_op:
+	case lwl_op:
+	case lwr_op:
+	case sdl_op:
+	case sdr_op:
+	case swl_op:
+	case swr_op:
+	case lb_op:
+	case lbu_op:
+	case sb_op:
+		goto sigbus;
+
+		/*
+		 * The remaining opcodes are the ones that are really of
+		 * interest.
+		 */
+	case spec3_op:
+		if (insn.dsp_format.func == lx_op) {
+			switch (insn.dsp_format.op) {
+			case lwx_op:
+				if (!access_ok(addr, 4))
+					goto sigbus;
+				LoadW(addr, value, res);
+				if (res)
+					goto fault;
+				compute_return_epc(regs);
+				regs->regs[insn.dsp_format.rd] = value;
+				break;
+			case lhx_op:
+				if (!access_ok(addr, 2))
+					goto sigbus;
+				LoadHW(addr, value, res);
+				if (res)
+					goto fault;
+				compute_return_epc(regs);
+				regs->regs[insn.dsp_format.rd] = value;
+				break;
+			default:
+				goto sigill;
+			}
+		}
+#ifdef CONFIG_EVA
+		else {
+			/*
+			 * we can land here only from kernel accessing user
+			 * memory, so we need to "switch" the address limit to
+			 * user space, so that address check can work properly.
+			 */
+			seg = force_uaccess_begin();
+			switch (insn.spec3_format.func) {
+			case lhe_op:
+				if (!access_ok(addr, 2)) {
+					force_uaccess_end(seg);
+					goto sigbus;
+				}
+				LoadHWE(addr, value, res);
+				if (res) {
+					force_uaccess_end(seg);
+					goto fault;
+				}
+				compute_return_epc(regs);
+				regs->regs[insn.spec3_format.rt] = value;
+				break;
+			case lwe_op:
+				if (!access_ok(addr, 4)) {
+					force_uaccess_end(seg);
+					goto sigbus;
+				}
+				LoadWE(addr, value, res);
+				if (res) {
+					force_uaccess_end(seg);
+					goto fault;
+				}
+				compute_return_epc(regs);
+				regs->regs[insn.spec3_format.rt] = value;
+				break;
+			case lhue_op:
+				if (!access_ok(addr, 2)) {
+					force_uaccess_end(seg);
+					goto sigbus;
+				}
+				LoadHWUE(addr, value, res);
+				if (res) {
+					force_uaccess_end(seg);
+					goto fault;
+				}
+				compute_return_epc(regs);
+				regs->regs[insn.spec3_format.rt] = value;
+				break;
+			case she_op:
+				if (!access_ok(addr, 2)) {
+					force_uaccess_end(seg);
+					goto sigbus;
+				}
+				compute_return_epc(regs);
+				value = regs->regs[insn.spec3_format.rt];
+				StoreHWE(addr, value, res);
+				if (res) {
+					force_uaccess_end(seg);
+					goto fault;
+				}
+				break;
+			case swe_op:
+				if (!access_ok(addr, 4)) {
+					force_uaccess_end(seg);
+					goto sigbus;
+				}
+				compute_return_epc(regs);
+				value = regs->regs[insn.spec3_format.rt];
+				StoreWE(addr, value, res);
+				if (res) {
+					force_uaccess_end(seg);
+					goto fault;
+				}
+				break;
+			default:
+				force_uaccess_end(seg);
+				goto sigill;
+			}
+			force_uaccess_end(seg);
+		}
+#endif
+		break;
+	case lh_op:
+		if (!access_ok(addr, 2))
+			goto sigbus;
+
+		if (IS_ENABLED(CONFIG_EVA)) {
+			if (uaccess_kernel())
+				LoadHW(addr, value, res);
+			else
+				LoadHWE(addr, value, res);
+		} else {
+			LoadHW(addr, value, res);
+		}
+
+		if (res)
+			goto fault;
+		compute_return_epc(regs);
+		regs->regs[insn.i_format.rt] = value;
+		break;
+
+	case lw_op:
+		if (!access_ok(addr, 4))
+			goto sigbus;
+
+		if (IS_ENABLED(CONFIG_EVA)) {
+			if (uaccess_kernel())
+				LoadW(addr, value, res);
+			else
+				LoadWE(addr, value, res);
+		} else {
+			LoadW(addr, value, res);
+		}
+
+		if (res)
+			goto fault;
+		compute_return_epc(regs);
+		regs->regs[insn.i_format.rt] = value;
+		break;
+
+	case lhu_op:
+		if (!access_ok(addr, 2))
+			goto sigbus;
+
+		if (IS_ENABLED(CONFIG_EVA)) {
+			if (uaccess_kernel())
+				LoadHWU(addr, value, res);
+			else
+				LoadHWUE(addr, value, res);
+		} else {
+			LoadHWU(addr, value, res);
+		}
+
+		if (res)
+			goto fault;
+		compute_return_epc(regs);
+		regs->regs[insn.i_format.rt] = value;
+		break;
+
+	case lwu_op:
+#ifdef CONFIG_64BIT
+		/*
+		 * A 32-bit kernel might be running on a 64-bit processor.  But
+		 * if we're on a 32-bit processor and an i-cache incoherency
+		 * or race makes us see a 64-bit instruction here the sdl/sdr
+		 * would blow up, so for now we don't handle unaligned 64-bit
+		 * instructions on 32-bit kernels.
+		 */
+		if (!access_ok(addr, 4))
+			goto sigbus;
+
+		LoadWU(addr, value, res);
+		if (res)
+			goto fault;
+		compute_return_epc(regs);
+		regs->regs[insn.i_format.rt] = value;
+		break;
+#endif /* CONFIG_64BIT */
+
+		/* Cannot handle 64-bit instructions in 32-bit kernel */
+		goto sigill;
+
+	case ld_op:
+#ifdef CONFIG_64BIT
+		/*
+		 * A 32-bit kernel might be running on a 64-bit processor.  But
+		 * if we're on a 32-bit processor and an i-cache incoherency
+		 * or race makes us see a 64-bit instruction here the sdl/sdr
+		 * would blow up, so for now we don't handle unaligned 64-bit
+		 * instructions on 32-bit kernels.
+		 */
+		if (!access_ok(addr, 8))
+			goto sigbus;
+
+		LoadDW(addr, value, res);
+		if (res)
+			goto fault;
+		compute_return_epc(regs);
+		regs->regs[insn.i_format.rt] = value;
+		break;
+#endif /* CONFIG_64BIT */
+
+		/* Cannot handle 64-bit instructions in 32-bit kernel */
+		goto sigill;
+
+	case sh_op:
+		if (!access_ok(addr, 2))
+			goto sigbus;
+
+		compute_return_epc(regs);
+		value = regs->regs[insn.i_format.rt];
+
+		if (IS_ENABLED(CONFIG_EVA)) {
+			if (uaccess_kernel())
+				StoreHW(addr, value, res);
+			else
+				StoreHWE(addr, value, res);
+		} else {
+			StoreHW(addr, value, res);
+		}
+
+		if (res)
+			goto fault;
+		break;
+
+	case sw_op:
+		if (!access_ok(addr, 4))
+			goto sigbus;
+
+		compute_return_epc(regs);
+		value = regs->regs[insn.i_format.rt];
+
+		if (IS_ENABLED(CONFIG_EVA)) {
+			if (uaccess_kernel())
+				StoreW(addr, value, res);
+			else
+				StoreWE(addr, value, res);
+		} else {
+			StoreW(addr, value, res);
+		}
+
+		if (res)
+			goto fault;
+		break;
+
+	case sd_op:
+#ifdef CONFIG_64BIT
+		/*
+		 * A 32-bit kernel might be running on a 64-bit processor.  But
+		 * if we're on a 32-bit processor and an i-cache incoherency
+		 * or race makes us see a 64-bit instruction here the sdl/sdr
+		 * would blow up, so for now we don't handle unaligned 64-bit
+		 * instructions on 32-bit kernels.
+		 */
+		if (!access_ok(addr, 8))
+			goto sigbus;
+
+		compute_return_epc(regs);
+		value = regs->regs[insn.i_format.rt];
+		StoreDW(addr, value, res);
+		if (res)
+			goto fault;
+		break;
+#endif /* CONFIG_64BIT */
+
+		/* Cannot handle 64-bit instructions in 32-bit kernel */
+		goto sigill;
+
+#ifdef CONFIG_MIPS_FP_SUPPORT
+
+	case lwc1_op:
+	case ldc1_op:
+	case swc1_op:
+	case sdc1_op:
+	case cop1x_op: {
+		void __user *fault_addr = NULL;
+
+		die_if_kernel("Unaligned FP access in kernel code", regs);
+		BUG_ON(!used_math());
+
+		res = fpu_emulator_cop1Handler(regs, &current->thread.fpu, 1,
+					       &fault_addr);
+		own_fpu(1);	/* Restore FPU state. */
+
+		/* Signal if something went wrong. */
+		process_fpemu_return(res, fault_addr, 0);
+
+		if (res == 0)
+			break;
+		return;
+	}
+#endif /* CONFIG_MIPS_FP_SUPPORT */
+
+#ifdef CONFIG_CPU_HAS_MSA
+
+	case msa_op: {
+		unsigned int wd, preempted;
+		enum msa_2b_fmt df;
+		union fpureg *fpr;
+
+		if (!cpu_has_msa)
+			goto sigill;
+
+		/*
+		 * If we've reached this point then userland should have taken
+		 * the MSA disabled exception & initialised vector context at
+		 * some point in the past.
+		 */
+		BUG_ON(!thread_msa_context_live());
+
+		df = insn.msa_mi10_format.df;
+		wd = insn.msa_mi10_format.wd;
+		fpr = &current->thread.fpu.fpr[wd];
+
+		switch (insn.msa_mi10_format.func) {
+		case msa_ld_op:
+			if (!access_ok(addr, sizeof(*fpr)))
+				goto sigbus;
+
+			do {
+				/*
+				 * If we have live MSA context keep track of
+				 * whether we get preempted in order to avoid
+				 * the register context we load being clobbered
+				 * by the live context as it's saved during
+				 * preemption. If we don't have live context
+				 * then it can't be saved to clobber the value
+				 * we load.
+				 */
+				preempted = test_thread_flag(TIF_USEDMSA);
+
+				res = __copy_from_user_inatomic(fpr, addr,
+								sizeof(*fpr));
+				if (res)
+					goto fault;
+
+				/*
+				 * Update the hardware register if it is in use
+				 * by the task in this quantum, in order to
+				 * avoid having to save & restore the whole
+				 * vector context.
+				 */
+				preempt_disable();
+				if (test_thread_flag(TIF_USEDMSA)) {
+					write_msa_wr(wd, fpr, df);
+					preempted = 0;
+				}
+				preempt_enable();
+			} while (preempted);
+			break;
+
+		case msa_st_op:
+			if (!access_ok(addr, sizeof(*fpr)))
+				goto sigbus;
+
+			/*
+			 * Update from the hardware register if it is in use by
+			 * the task in this quantum, in order to avoid having to
+			 * save & restore the whole vector context.
+			 */
+			preempt_disable();
+			if (test_thread_flag(TIF_USEDMSA))
+				read_msa_wr(wd, fpr, df);
+			preempt_enable();
+
+			res = __copy_to_user_inatomic(addr, fpr, sizeof(*fpr));
+			if (res)
+				goto fault;
+			break;
+
+		default:
+			goto sigbus;
+		}
+
+		compute_return_epc(regs);
+		break;
+	}
+#endif /* CONFIG_CPU_HAS_MSA */
+
+#ifndef CONFIG_CPU_MIPSR6
+	/*
+	 * COP2 is available to implementor for application specific use.
+	 * It's up to applications to register a notifier chain and do
+	 * whatever they have to do, including possible sending of signals.
+	 *
+	 * This instruction has been reallocated in Release 6
+	 */
+	case lwc2_op:
+		cu2_notifier_call_chain(CU2_LWC2_OP, regs);
+		break;
+
+	case ldc2_op:
+		cu2_notifier_call_chain(CU2_LDC2_OP, regs);
+		break;
+
+	case swc2_op:
+		cu2_notifier_call_chain(CU2_SWC2_OP, regs);
+		break;
+
+	case sdc2_op:
+		cu2_notifier_call_chain(CU2_SDC2_OP, regs);
+		break;
+#endif
+	default:
+		/*
+		 * Pheeee...  We encountered an yet unknown instruction or
+		 * cache coherence problem.  Die sucker, die ...
+		 */
+		goto sigill;
+	}
+
+#ifdef CONFIG_DEBUG_FS
+	unaligned_instructions++;
+#endif
+
+	return;
+
+fault:
+	/* roll back jump/branch */
+	regs->cp0_epc = origpc;
+	regs->regs[31] = orig31;
+	/* Did we have an exception handler installed? */
+	if (fixup_exception(regs))
+		return;
+
+	die_if_kernel("Unhandled kernel unaligned access", regs);
+	force_sig(SIGSEGV);
+
+	return;
+
+sigbus:
+	die_if_kernel("Unhandled kernel unaligned access", regs);
+	force_sig(SIGBUS);
+
+	return;
+
+sigill:
+	die_if_kernel
+	    ("Unhandled kernel unaligned access or invalid instruction", regs);
+	force_sig(SIGILL);
+}
+
+/* Recode table from 16-bit register notation to 32-bit GPR. */
+const int reg16to32[] = { 16, 17, 2, 3, 4, 5, 6, 7 };
+
+/* Recode table from 16-bit STORE register notation to 32-bit GPR. */
+static const int reg16to32st[] = { 0, 17, 2, 3, 4, 5, 6, 7 };
+
+static void emulate_load_store_microMIPS(struct pt_regs *regs,
+					 void __user *addr)
+{
+	unsigned long value;
+	unsigned int res;
+	int i;
+	unsigned int reg = 0, rvar;
+	unsigned long orig31;
+	u16 __user *pc16;
+	u16 halfword;
+	unsigned int word;
+	unsigned long origpc, contpc;
+	union mips_instruction insn;
+	struct mm_decoded_insn mminsn;
+
+	origpc = regs->cp0_epc;
+	orig31 = regs->regs[31];
+
+	mminsn.micro_mips_mode = 1;
+
+	/*
+	 * This load never faults.
+	 */
+	pc16 = (unsigned short __user *)msk_isa16_mode(regs->cp0_epc);
+	__get_user(halfword, pc16);
+	pc16++;
+	contpc = regs->cp0_epc + 2;
+	word = ((unsigned int)halfword << 16);
+	mminsn.pc_inc = 2;
+
+	if (!mm_insn_16bit(halfword)) {
+		__get_user(halfword, pc16);
+		pc16++;
+		contpc = regs->cp0_epc + 4;
+		mminsn.pc_inc = 4;
+		word |= halfword;
+	}
+	mminsn.insn = word;
+
+	if (get_user(halfword, pc16))
+		goto fault;
+	mminsn.next_pc_inc = 2;
+	word = ((unsigned int)halfword << 16);
+
+	if (!mm_insn_16bit(halfword)) {
+		pc16++;
+		if (get_user(halfword, pc16))
+			goto fault;
+		mminsn.next_pc_inc = 4;
+		word |= halfword;
+	}
+	mminsn.next_insn = word;
+
+	insn = (union mips_instruction)(mminsn.insn);
+	if (mm_isBranchInstr(regs, mminsn, &contpc))
+		insn = (union mips_instruction)(mminsn.next_insn);
+
+	/*  Parse instruction to find what to do */
+
+	switch (insn.mm_i_format.opcode) {
+
+	case mm_pool32a_op:
+		switch (insn.mm_x_format.func) {
+		case mm_lwxs_op:
+			reg = insn.mm_x_format.rd;
+			goto loadW;
+		}
+
+		goto sigbus;
+
+	case mm_pool32b_op:
+		switch (insn.mm_m_format.func) {
+		case mm_lwp_func:
+			reg = insn.mm_m_format.rd;
+			if (reg == 31)
+				goto sigbus;
+
+			if (!access_ok(addr, 8))
+				goto sigbus;
+
+			LoadW(addr, value, res);
+			if (res)
+				goto fault;
+			regs->regs[reg] = value;
+			addr += 4;
+			LoadW(addr, value, res);
+			if (res)
+				goto fault;
+			regs->regs[reg + 1] = value;
+			goto success;
+
+		case mm_swp_func:
+			reg = insn.mm_m_format.rd;
+			if (reg == 31)
+				goto sigbus;
+
+			if (!access_ok(addr, 8))
+				goto sigbus;
+
+			value = regs->regs[reg];
+			StoreW(addr, value, res);
+			if (res)
+				goto fault;
+			addr += 4;
+			value = regs->regs[reg + 1];
+			StoreW(addr, value, res);
+			if (res)
+				goto fault;
+			goto success;
+
+		case mm_ldp_func:
+#ifdef CONFIG_64BIT
+			reg = insn.mm_m_format.rd;
+			if (reg == 31)
+				goto sigbus;
+
+			if (!access_ok(addr, 16))
+				goto sigbus;
+
+			LoadDW(addr, value, res);
+			if (res)
+				goto fault;
+			regs->regs[reg] = value;
+			addr += 8;
+			LoadDW(addr, value, res);
+			if (res)
+				goto fault;
+			regs->regs[reg + 1] = value;
+			goto success;
+#endif /* CONFIG_64BIT */
+
+			goto sigill;
+
+		case mm_sdp_func:
+#ifdef CONFIG_64BIT
+			reg = insn.mm_m_format.rd;
+			if (reg == 31)
+				goto sigbus;
+
+			if (!access_ok(addr, 16))
+				goto sigbus;
+
+			value = regs->regs[reg];
+			StoreDW(addr, value, res);
+			if (res)
+				goto fault;
+			addr += 8;
+			value = regs->regs[reg + 1];
+			StoreDW(addr, value, res);
+			if (res)
+				goto fault;
+			goto success;
+#endif /* CONFIG_64BIT */
+
+			goto sigill;
+
+		case mm_lwm32_func:
+			reg = insn.mm_m_format.rd;
+			rvar = reg & 0xf;
+			if ((rvar > 9) || !reg)
+				goto sigill;
+			if (reg & 0x10) {
+				if (!access_ok(addr, 4 * (rvar + 1)))
+					goto sigbus;
+			} else {
+				if (!access_ok(addr, 4 * rvar))
+					goto sigbus;
+			}
+			if (rvar == 9)
+				rvar = 8;
+			for (i = 16; rvar; rvar--, i++) {
+				LoadW(addr, value, res);
+				if (res)
+					goto fault;
+				addr += 4;
+				regs->regs[i] = value;
+			}
+			if ((reg & 0xf) == 9) {
+				LoadW(addr, value, res);
+				if (res)
+					goto fault;
+				addr += 4;
+				regs->regs[30] = value;
+			}
+			if (reg & 0x10) {
+				LoadW(addr, value, res);
+				if (res)
+					goto fault;
+				regs->regs[31] = value;
+			}
+			goto success;
+
+		case mm_swm32_func:
+			reg = insn.mm_m_format.rd;
+			rvar = reg & 0xf;
+			if ((rvar > 9) || !reg)
+				goto sigill;
+			if (reg & 0x10) {
+				if (!access_ok(addr, 4 * (rvar + 1)))
+					goto sigbus;
+			} else {
+				if (!access_ok(addr, 4 * rvar))
+					goto sigbus;
+			}
+			if (rvar == 9)
+				rvar = 8;
+			for (i = 16; rvar; rvar--, i++) {
+				value = regs->regs[i];
+				StoreW(addr, value, res);
+				if (res)
+					goto fault;
+				addr += 4;
+			}
+			if ((reg & 0xf) == 9) {
+				value = regs->regs[30];
+				StoreW(addr, value, res);
+				if (res)
+					goto fault;
+				addr += 4;
+			}
+			if (reg & 0x10) {
+				value = regs->regs[31];
+				StoreW(addr, value, res);
+				if (res)
+					goto fault;
+			}
+			goto success;
+
+		case mm_ldm_func:
+#ifdef CONFIG_64BIT
+			reg = insn.mm_m_format.rd;
+			rvar = reg & 0xf;
+			if ((rvar > 9) || !reg)
+				goto sigill;
+			if (reg & 0x10) {
+				if (!access_ok(addr, 8 * (rvar + 1)))
+					goto sigbus;
+			} else {
+				if (!access_ok(addr, 8 * rvar))
+					goto sigbus;
+			}
+			if (rvar == 9)
+				rvar = 8;
+
+			for (i = 16; rvar; rvar--, i++) {
+				LoadDW(addr, value, res);
+				if (res)
+					goto fault;
+				addr += 4;
+				regs->regs[i] = value;
+			}
+			if ((reg & 0xf) == 9) {
+				LoadDW(addr, value, res);
+				if (res)
+					goto fault;
+				addr += 8;
+				regs->regs[30] = value;
+			}
+			if (reg & 0x10) {
+				LoadDW(addr, value, res);
+				if (res)
+					goto fault;
+				regs->regs[31] = value;
+			}
+			goto success;
+#endif /* CONFIG_64BIT */
+
+			goto sigill;
+
+		case mm_sdm_func:
+#ifdef CONFIG_64BIT
+			reg = insn.mm_m_format.rd;
+			rvar = reg & 0xf;
+			if ((rvar > 9) || !reg)
+				goto sigill;
+			if (reg & 0x10) {
+				if (!access_ok(addr, 8 * (rvar + 1)))
+					goto sigbus;
+			} else {
+				if (!access_ok(addr, 8 * rvar))
+					goto sigbus;
+			}
+			if (rvar == 9)
+				rvar = 8;
+
+			for (i = 16; rvar; rvar--, i++) {
+				value = regs->regs[i];
+				StoreDW(addr, value, res);
+				if (res)
+					goto fault;
+				addr += 8;
+			}
+			if ((reg & 0xf) == 9) {
+				value = regs->regs[30];
+				StoreDW(addr, value, res);
+				if (res)
+					goto fault;
+				addr += 8;
+			}
+			if (reg & 0x10) {
+				value = regs->regs[31];
+				StoreDW(addr, value, res);
+				if (res)
+					goto fault;
+			}
+			goto success;
+#endif /* CONFIG_64BIT */
+
+			goto sigill;
+
+			/*  LWC2, SWC2, LDC2, SDC2 are not serviced */
+		}
+
+		goto sigbus;
+
+	case mm_pool32c_op:
+		switch (insn.mm_m_format.func) {
+		case mm_lwu_func:
+			reg = insn.mm_m_format.rd;
+			goto loadWU;
+		}
+
+		/*  LL,SC,LLD,SCD are not serviced */
+		goto sigbus;
+
+#ifdef CONFIG_MIPS_FP_SUPPORT
+	case mm_pool32f_op:
+		switch (insn.mm_x_format.func) {
+		case mm_lwxc1_func:
+		case mm_swxc1_func:
+		case mm_ldxc1_func:
+		case mm_sdxc1_func:
+			goto fpu_emul;
+		}
+
+		goto sigbus;
+
+	case mm_ldc132_op:
+	case mm_sdc132_op:
+	case mm_lwc132_op:
+	case mm_swc132_op: {
+		void __user *fault_addr = NULL;
+
+fpu_emul:
+		/* roll back jump/branch */
+		regs->cp0_epc = origpc;
+		regs->regs[31] = orig31;
+
+		die_if_kernel("Unaligned FP access in kernel code", regs);
+		BUG_ON(!used_math());
+		BUG_ON(!is_fpu_owner());
+
+		res = fpu_emulator_cop1Handler(regs, &current->thread.fpu, 1,
+					       &fault_addr);
+		own_fpu(1);	/* restore FPU state */
+
+		/* If something went wrong, signal */
+		process_fpemu_return(res, fault_addr, 0);
+
+		if (res == 0)
+			goto success;
+		return;
+	}
+#endif /* CONFIG_MIPS_FP_SUPPORT */
+
+	case mm_lh32_op:
+		reg = insn.mm_i_format.rt;
+		goto loadHW;
+
+	case mm_lhu32_op:
+		reg = insn.mm_i_format.rt;
+		goto loadHWU;
+
+	case mm_lw32_op:
+		reg = insn.mm_i_format.rt;
+		goto loadW;
+
+	case mm_sh32_op:
+		reg = insn.mm_i_format.rt;
+		goto storeHW;
+
+	case mm_sw32_op:
+		reg = insn.mm_i_format.rt;
+		goto storeW;
+
+	case mm_ld32_op:
+		reg = insn.mm_i_format.rt;
+		goto loadDW;
+
+	case mm_sd32_op:
+		reg = insn.mm_i_format.rt;
+		goto storeDW;
+
+	case mm_pool16c_op:
+		switch (insn.mm16_m_format.func) {
+		case mm_lwm16_op:
+			reg = insn.mm16_m_format.rlist;
+			rvar = reg + 1;
+			if (!access_ok(addr, 4 * rvar))
+				goto sigbus;
+
+			for (i = 16; rvar; rvar--, i++) {
+				LoadW(addr, value, res);
+				if (res)
+					goto fault;
+				addr += 4;
+				regs->regs[i] = value;
+			}
+			LoadW(addr, value, res);
+			if (res)
+				goto fault;
+			regs->regs[31] = value;
+
+			goto success;
+
+		case mm_swm16_op:
+			reg = insn.mm16_m_format.rlist;
+			rvar = reg + 1;
+			if (!access_ok(addr, 4 * rvar))
+				goto sigbus;
+
+			for (i = 16; rvar; rvar--, i++) {
+				value = regs->regs[i];
+				StoreW(addr, value, res);
+				if (res)
+					goto fault;
+				addr += 4;
+			}
+			value = regs->regs[31];
+			StoreW(addr, value, res);
+			if (res)
+				goto fault;
+
+			goto success;
+
+		}
+
+		goto sigbus;
+
+	case mm_lhu16_op:
+		reg = reg16to32[insn.mm16_rb_format.rt];
+		goto loadHWU;
+
+	case mm_lw16_op:
+		reg = reg16to32[insn.mm16_rb_format.rt];
+		goto loadW;
+
+	case mm_sh16_op:
+		reg = reg16to32st[insn.mm16_rb_format.rt];
+		goto storeHW;
+
+	case mm_sw16_op:
+		reg = reg16to32st[insn.mm16_rb_format.rt];
+		goto storeW;
+
+	case mm_lwsp16_op:
+		reg = insn.mm16_r5_format.rt;
+		goto loadW;
+
+	case mm_swsp16_op:
+		reg = insn.mm16_r5_format.rt;
+		goto storeW;
+
+	case mm_lwgp16_op:
+		reg = reg16to32[insn.mm16_r3_format.rt];
+		goto loadW;
+
+	default:
+		goto sigill;
+	}
+
+loadHW:
+	if (!access_ok(addr, 2))
+		goto sigbus;
+
+	LoadHW(addr, value, res);
+	if (res)
+		goto fault;
+	regs->regs[reg] = value;
+	goto success;
+
+loadHWU:
+	if (!access_ok(addr, 2))
+		goto sigbus;
+
+	LoadHWU(addr, value, res);
+	if (res)
+		goto fault;
+	regs->regs[reg] = value;
+	goto success;
+
+loadW:
+	if (!access_ok(addr, 4))
+		goto sigbus;
+
+	LoadW(addr, value, res);
+	if (res)
+		goto fault;
+	regs->regs[reg] = value;
+	goto success;
+
+loadWU:
+#ifdef CONFIG_64BIT
+	/*
+	 * A 32-bit kernel might be running on a 64-bit processor.  But
+	 * if we're on a 32-bit processor and an i-cache incoherency
+	 * or race makes us see a 64-bit instruction here the sdl/sdr
+	 * would blow up, so for now we don't handle unaligned 64-bit
+	 * instructions on 32-bit kernels.
+	 */
+	if (!access_ok(addr, 4))
+		goto sigbus;
+
+	LoadWU(addr, value, res);
+	if (res)
+		goto fault;
+	regs->regs[reg] = value;
+	goto success;
+#endif /* CONFIG_64BIT */
+
+	/* Cannot handle 64-bit instructions in 32-bit kernel */
+	goto sigill;
+
+loadDW:
+#ifdef CONFIG_64BIT
+	/*
+	 * A 32-bit kernel might be running on a 64-bit processor.  But
+	 * if we're on a 32-bit processor and an i-cache incoherency
+	 * or race makes us see a 64-bit instruction here the sdl/sdr
+	 * would blow up, so for now we don't handle unaligned 64-bit
+	 * instructions on 32-bit kernels.
+	 */
+	if (!access_ok(addr, 8))
+		goto sigbus;
+
+	LoadDW(addr, value, res);
+	if (res)
+		goto fault;
+	regs->regs[reg] = value;
+	goto success;
+#endif /* CONFIG_64BIT */
+
+	/* Cannot handle 64-bit instructions in 32-bit kernel */
+	goto sigill;
+
+storeHW:
+	if (!access_ok(addr, 2))
+		goto sigbus;
+
+	value = regs->regs[reg];
+	StoreHW(addr, value, res);
+	if (res)
+		goto fault;
+	goto success;
+
+storeW:
+	if (!access_ok(addr, 4))
+		goto sigbus;
+
+	value = regs->regs[reg];
+	StoreW(addr, value, res);
+	if (res)
+		goto fault;
+	goto success;
+
+storeDW:
+#ifdef CONFIG_64BIT
+	/*
+	 * A 32-bit kernel might be running on a 64-bit processor.  But
+	 * if we're on a 32-bit processor and an i-cache incoherency
+	 * or race makes us see a 64-bit instruction here the sdl/sdr
+	 * would blow up, so for now we don't handle unaligned 64-bit
+	 * instructions on 32-bit kernels.
+	 */
+	if (!access_ok(addr, 8))
+		goto sigbus;
+
+	value = regs->regs[reg];
+	StoreDW(addr, value, res);
+	if (res)
+		goto fault;
+	goto success;
+#endif /* CONFIG_64BIT */
+
+	/* Cannot handle 64-bit instructions in 32-bit kernel */
+	goto sigill;
+
+success:
+	regs->cp0_epc = contpc;	/* advance or branch */
+
+#ifdef CONFIG_DEBUG_FS
+	unaligned_instructions++;
+#endif
+	return;
+
+fault:
+	/* roll back jump/branch */
+	regs->cp0_epc = origpc;
+	regs->regs[31] = orig31;
+	/* Did we have an exception handler installed? */
+	if (fixup_exception(regs))
+		return;
+
+	die_if_kernel("Unhandled kernel unaligned access", regs);
+	force_sig(SIGSEGV);
+
+	return;
+
+sigbus:
+	die_if_kernel("Unhandled kernel unaligned access", regs);
+	force_sig(SIGBUS);
+
+	return;
+
+sigill:
+	die_if_kernel
+	    ("Unhandled kernel unaligned access or invalid instruction", regs);
+	force_sig(SIGILL);
+}
+
+static void emulate_load_store_MIPS16e(struct pt_regs *regs, void __user * addr)
+{
+	unsigned long value;
+	unsigned int res;
+	int reg;
+	unsigned long orig31;
+	u16 __user *pc16;
+	unsigned long origpc;
+	union mips16e_instruction mips16inst, oldinst;
+	unsigned int opcode;
+	int extended = 0;
+
+	origpc = regs->cp0_epc;
+	orig31 = regs->regs[31];
+	pc16 = (unsigned short __user *)msk_isa16_mode(origpc);
+	/*
+	 * This load never faults.
+	 */
+	__get_user(mips16inst.full, pc16);
+	oldinst = mips16inst;
+
+	/* skip EXTEND instruction */
+	if (mips16inst.ri.opcode == MIPS16e_extend_op) {
+		extended = 1;
+		pc16++;
+		__get_user(mips16inst.full, pc16);
+	} else if (delay_slot(regs)) {
+		/*  skip jump instructions */
+		/*  JAL/JALX are 32 bits but have OPCODE in first short int */
+		if (mips16inst.ri.opcode == MIPS16e_jal_op)
+			pc16++;
+		pc16++;
+		if (get_user(mips16inst.full, pc16))
+			goto sigbus;
+	}
+
+	opcode = mips16inst.ri.opcode;
+	switch (opcode) {
+	case MIPS16e_i64_op:	/* I64 or RI64 instruction */
+		switch (mips16inst.i64.func) {	/* I64/RI64 func field check */
+		case MIPS16e_ldpc_func:
+		case MIPS16e_ldsp_func:
+			reg = reg16to32[mips16inst.ri64.ry];
+			goto loadDW;
+
+		case MIPS16e_sdsp_func:
+			reg = reg16to32[mips16inst.ri64.ry];
+			goto writeDW;
+
+		case MIPS16e_sdrasp_func:
+			reg = 29;	/* GPRSP */
+			goto writeDW;
+		}
+
+		goto sigbus;
+
+	case MIPS16e_swsp_op:
+		reg = reg16to32[mips16inst.ri.rx];
+		if (extended && cpu_has_mips16e2)
+			switch (mips16inst.ri.imm >> 5) {
+			case 0:		/* SWSP */
+			case 1:		/* SWGP */
+				break;
+			case 2:		/* SHGP */
+				opcode = MIPS16e_sh_op;
+				break;
+			default:
+				goto sigbus;
+			}
+		break;
+
+	case MIPS16e_lwpc_op:
+		reg = reg16to32[mips16inst.ri.rx];
+		break;
+
+	case MIPS16e_lwsp_op:
+		reg = reg16to32[mips16inst.ri.rx];
+		if (extended && cpu_has_mips16e2)
+			switch (mips16inst.ri.imm >> 5) {
+			case 0:		/* LWSP */
+			case 1:		/* LWGP */
+				break;
+			case 2:		/* LHGP */
+				opcode = MIPS16e_lh_op;
+				break;
+			case 4:		/* LHUGP */
+				opcode = MIPS16e_lhu_op;
+				break;
+			default:
+				goto sigbus;
+			}
+		break;
+
+	case MIPS16e_i8_op:
+		if (mips16inst.i8.func != MIPS16e_swrasp_func)
+			goto sigbus;
+		reg = 29;	/* GPRSP */
+		break;
+
+	default:
+		reg = reg16to32[mips16inst.rri.ry];
+		break;
+	}
+
+	switch (opcode) {
+
+	case MIPS16e_lb_op:
+	case MIPS16e_lbu_op:
+	case MIPS16e_sb_op:
+		goto sigbus;
+
+	case MIPS16e_lh_op:
+		if (!access_ok(addr, 2))
+			goto sigbus;
+
+		LoadHW(addr, value, res);
+		if (res)
+			goto fault;
+		MIPS16e_compute_return_epc(regs, &oldinst);
+		regs->regs[reg] = value;
+		break;
+
+	case MIPS16e_lhu_op:
+		if (!access_ok(addr, 2))
+			goto sigbus;
+
+		LoadHWU(addr, value, res);
+		if (res)
+			goto fault;
+		MIPS16e_compute_return_epc(regs, &oldinst);
+		regs->regs[reg] = value;
+		break;
+
+	case MIPS16e_lw_op:
+	case MIPS16e_lwpc_op:
+	case MIPS16e_lwsp_op:
+		if (!access_ok(addr, 4))
+			goto sigbus;
+
+		LoadW(addr, value, res);
+		if (res)
+			goto fault;
+		MIPS16e_compute_return_epc(regs, &oldinst);
+		regs->regs[reg] = value;
+		break;
+
+	case MIPS16e_lwu_op:
+#ifdef CONFIG_64BIT
+		/*
+		 * A 32-bit kernel might be running on a 64-bit processor.  But
+		 * if we're on a 32-bit processor and an i-cache incoherency
+		 * or race makes us see a 64-bit instruction here the sdl/sdr
+		 * would blow up, so for now we don't handle unaligned 64-bit
+		 * instructions on 32-bit kernels.
+		 */
+		if (!access_ok(addr, 4))
+			goto sigbus;
+
+		LoadWU(addr, value, res);
+		if (res)
+			goto fault;
+		MIPS16e_compute_return_epc(regs, &oldinst);
+		regs->regs[reg] = value;
+		break;
+#endif /* CONFIG_64BIT */
+
+		/* Cannot handle 64-bit instructions in 32-bit kernel */
+		goto sigill;
+
+	case MIPS16e_ld_op:
+loadDW:
+#ifdef CONFIG_64BIT
+		/*
+		 * A 32-bit kernel might be running on a 64-bit processor.  But
+		 * if we're on a 32-bit processor and an i-cache incoherency
+		 * or race makes us see a 64-bit instruction here the sdl/sdr
+		 * would blow up, so for now we don't handle unaligned 64-bit
+		 * instructions on 32-bit kernels.
+		 */
+		if (!access_ok(addr, 8))
+			goto sigbus;
+
+		LoadDW(addr, value, res);
+		if (res)
+			goto fault;
+		MIPS16e_compute_return_epc(regs, &oldinst);
+		regs->regs[reg] = value;
+		break;
+#endif /* CONFIG_64BIT */
+
+		/* Cannot handle 64-bit instructions in 32-bit kernel */
+		goto sigill;
+
+	case MIPS16e_sh_op:
+		if (!access_ok(addr, 2))
+			goto sigbus;
+
+		MIPS16e_compute_return_epc(regs, &oldinst);
+		value = regs->regs[reg];
+		StoreHW(addr, value, res);
+		if (res)
+			goto fault;
+		break;
+
+	case MIPS16e_sw_op:
+	case MIPS16e_swsp_op:
+	case MIPS16e_i8_op:	/* actually - MIPS16e_swrasp_func */
+		if (!access_ok(addr, 4))
+			goto sigbus;
+
+		MIPS16e_compute_return_epc(regs, &oldinst);
+		value = regs->regs[reg];
+		StoreW(addr, value, res);
+		if (res)
+			goto fault;
+		break;
+
+	case MIPS16e_sd_op:
+writeDW:
+#ifdef CONFIG_64BIT
+		/*
+		 * A 32-bit kernel might be running on a 64-bit processor.  But
+		 * if we're on a 32-bit processor and an i-cache incoherency
+		 * or race makes us see a 64-bit instruction here the sdl/sdr
+		 * would blow up, so for now we don't handle unaligned 64-bit
+		 * instructions on 32-bit kernels.
+		 */
+		if (!access_ok(addr, 8))
+			goto sigbus;
+
+		MIPS16e_compute_return_epc(regs, &oldinst);
+		value = regs->regs[reg];
+		StoreDW(addr, value, res);
+		if (res)
+			goto fault;
+		break;
+#endif /* CONFIG_64BIT */
+
+		/* Cannot handle 64-bit instructions in 32-bit kernel */
+		goto sigill;
+
+	default:
+		/*
+		 * Pheeee...  We encountered an yet unknown instruction or
+		 * cache coherence problem.  Die sucker, die ...
+		 */
+		goto sigill;
+	}
+
+#ifdef CONFIG_DEBUG_FS
+	unaligned_instructions++;
+#endif
+
+	return;
+
+fault:
+	/* roll back jump/branch */
+	regs->cp0_epc = origpc;
+	regs->regs[31] = orig31;
+	/* Did we have an exception handler installed? */
+	if (fixup_exception(regs))
+		return;
+
+	die_if_kernel("Unhandled kernel unaligned access", regs);
+	force_sig(SIGSEGV);
+
+	return;
+
+sigbus:
+	die_if_kernel("Unhandled kernel unaligned access", regs);
+	force_sig(SIGBUS);
+
+	return;
+
+sigill:
+	die_if_kernel
+	    ("Unhandled kernel unaligned access or invalid instruction", regs);
+	force_sig(SIGILL);
+}
+
+asmlinkage void do_ade(struct pt_regs *regs)
+{
+	enum ctx_state prev_state;
+	unsigned int __user *pc;
+	mm_segment_t seg;
+
+	prev_state = exception_enter();
+	perf_sw_event(PERF_COUNT_SW_ALIGNMENT_FAULTS,
+			1, regs, regs->cp0_badvaddr);
+	/*
+	 * Did we catch a fault trying to load an instruction?
+	 */
+	if (regs->cp0_badvaddr == regs->cp0_epc)
+		goto sigbus;
+
+	if (user_mode(regs) && !test_thread_flag(TIF_FIXADE))
+		goto sigbus;
+	if (unaligned_action == UNALIGNED_ACTION_SIGNAL)
+		goto sigbus;
+
+	/*
+	 * Do branch emulation only if we didn't forward the exception.
+	 * This is all so but ugly ...
+	 */
+
+	/*
+	 * Are we running in microMIPS mode?
+	 */
+	if (get_isa16_mode(regs->cp0_epc)) {
+		/*
+		 * Did we catch a fault trying to load an instruction in
+		 * 16-bit mode?
+		 */
+		if (regs->cp0_badvaddr == msk_isa16_mode(regs->cp0_epc))
+			goto sigbus;
+		if (unaligned_action == UNALIGNED_ACTION_SHOW)
+			show_registers(regs);
+
+		if (cpu_has_mmips) {
+			seg = get_fs();
+			if (!user_mode(regs))
+				set_fs(KERNEL_DS);
+			emulate_load_store_microMIPS(regs,
+				(void __user *)regs->cp0_badvaddr);
+			set_fs(seg);
+
+			return;
+		}
+
+		if (cpu_has_mips16) {
+			seg = get_fs();
+			if (!user_mode(regs))
+				set_fs(KERNEL_DS);
+			emulate_load_store_MIPS16e(regs,
+				(void __user *)regs->cp0_badvaddr);
+			set_fs(seg);
+
+			return;
+		}
+
+		goto sigbus;
+	}
+
+	if (unaligned_action == UNALIGNED_ACTION_SHOW)
+		show_registers(regs);
+	pc = (unsigned int __user *)exception_epc(regs);
+
+	seg = get_fs();
+	if (!user_mode(regs))
+		set_fs(KERNEL_DS);
+	emulate_load_store_insn(regs, (void __user *)regs->cp0_badvaddr, pc);
+	set_fs(seg);
+
+	return;
+
+sigbus:
+	die_if_kernel("Kernel unaligned instruction access", regs);
+	force_sig(SIGBUS);
+
+	/*
+	 * XXX On return from the signal handler we should advance the epc
+	 */
+	exception_exit(prev_state);
+}
+
+#ifdef CONFIG_DEBUG_FS
+static int __init debugfs_unaligned(void)
+{
+	debugfs_create_u32("unaligned_instructions", S_IRUGO, mips_debugfs_dir,
+			   &unaligned_instructions);
+	debugfs_create_u32("unaligned_action", S_IRUGO | S_IWUSR,
+			   mips_debugfs_dir, &unaligned_action);
+	return 0;
+}
+arch_initcall(debugfs_unaligned);
+#endif
diff --git a/arch/mips/kernel/uprobes.c b/arch/mips/kernel/uprobes.c
new file mode 100644
index 000000000..6dbe4eab0
--- /dev/null
+++ b/arch/mips/kernel/uprobes.c
@@ -0,0 +1,262 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/highmem.h>
+#include <linux/kdebug.h>
+#include <linux/types.h>
+#include <linux/notifier.h>
+#include <linux/sched.h>
+#include <linux/uprobes.h>
+
+#include <asm/branch.h>
+#include <asm/cpu-features.h>
+#include <asm/ptrace.h>
+
+#include "probes-common.h"
+
+static inline int insn_has_delay_slot(const union mips_instruction insn)
+{
+	return __insn_has_delay_slot(insn);
+}
+
+/**
+ * arch_uprobe_analyze_insn - instruction analysis including validity and fixups.
+ * @mm: the probed address space.
+ * @arch_uprobe: the probepoint information.
+ * @addr: virtual address at which to install the probepoint
+ * Return 0 on success or a -ve number on error.
+ */
+int arch_uprobe_analyze_insn(struct arch_uprobe *aup,
+	struct mm_struct *mm, unsigned long addr)
+{
+	union mips_instruction inst;
+
+	/*
+	 * For the time being this also blocks attempts to use uprobes with
+	 * MIPS16 and microMIPS.
+	 */
+	if (addr & 0x03)
+		return -EINVAL;
+
+	inst.word = aup->insn[0];
+
+	if (__insn_is_compact_branch(inst)) {
+		pr_notice("Uprobes for compact branches are not supported\n");
+		return -EINVAL;
+	}
+
+	aup->ixol[0] = aup->insn[insn_has_delay_slot(inst)];
+	aup->ixol[1] = UPROBE_BRK_UPROBE_XOL;		/* NOP  */
+
+	return 0;
+}
+
+/**
+ * is_trap_insn - check if the instruction is a trap variant
+ * @insn: instruction to be checked.
+ * Returns true if @insn is a trap variant.
+ *
+ * This definition overrides the weak definition in kernel/events/uprobes.c.
+ * and is needed for the case where an architecture has multiple trap
+ * instructions (like PowerPC or MIPS).  We treat BREAK just like the more
+ * modern conditional trap instructions.
+ */
+bool is_trap_insn(uprobe_opcode_t *insn)
+{
+	union mips_instruction inst;
+
+	inst.word = *insn;
+
+	switch (inst.i_format.opcode) {
+	case spec_op:
+		switch (inst.r_format.func) {
+		case break_op:
+		case teq_op:
+		case tge_op:
+		case tgeu_op:
+		case tlt_op:
+		case tltu_op:
+		case tne_op:
+			return 1;
+		}
+		break;
+
+	case bcond_op:	/* Yes, really ...  */
+		switch (inst.u_format.rt) {
+		case teqi_op:
+		case tgei_op:
+		case tgeiu_op:
+		case tlti_op:
+		case tltiu_op:
+		case tnei_op:
+			return 1;
+		}
+		break;
+	}
+
+	return 0;
+}
+
+#define UPROBE_TRAP_NR	ULONG_MAX
+
+/*
+ * arch_uprobe_pre_xol - prepare to execute out of line.
+ * @auprobe: the probepoint information.
+ * @regs: reflects the saved user state of current task.
+ */
+int arch_uprobe_pre_xol(struct arch_uprobe *aup, struct pt_regs *regs)
+{
+	struct uprobe_task *utask = current->utask;
+
+	/*
+	 * Now find the EPC where to resume after the breakpoint has been
+	 * dealt with.  This may require emulation of a branch.
+	 */
+	aup->resume_epc = regs->cp0_epc + 4;
+	if (insn_has_delay_slot((union mips_instruction) aup->insn[0])) {
+		__compute_return_epc_for_insn(regs,
+			(union mips_instruction) aup->insn[0]);
+		aup->resume_epc = regs->cp0_epc;
+	}
+	utask->autask.saved_trap_nr = current->thread.trap_nr;
+	current->thread.trap_nr = UPROBE_TRAP_NR;
+	regs->cp0_epc = current->utask->xol_vaddr;
+
+	return 0;
+}
+
+int arch_uprobe_post_xol(struct arch_uprobe *aup, struct pt_regs *regs)
+{
+	struct uprobe_task *utask = current->utask;
+
+	current->thread.trap_nr = utask->autask.saved_trap_nr;
+	regs->cp0_epc = aup->resume_epc;
+
+	return 0;
+}
+
+/*
+ * If xol insn itself traps and generates a signal(Say,
+ * SIGILL/SIGSEGV/etc), then detect the case where a singlestepped
+ * instruction jumps back to its own address. It is assumed that anything
+ * like do_page_fault/do_trap/etc sets thread.trap_nr != -1.
+ *
+ * arch_uprobe_pre_xol/arch_uprobe_post_xol save/restore thread.trap_nr,
+ * arch_uprobe_xol_was_trapped() simply checks that ->trap_nr is not equal to
+ * UPROBE_TRAP_NR == -1 set by arch_uprobe_pre_xol().
+ */
+bool arch_uprobe_xol_was_trapped(struct task_struct *tsk)
+{
+	if (tsk->thread.trap_nr != UPROBE_TRAP_NR)
+		return true;
+
+	return false;
+}
+
+int arch_uprobe_exception_notify(struct notifier_block *self,
+	unsigned long val, void *data)
+{
+	struct die_args *args = data;
+	struct pt_regs *regs = args->regs;
+
+	/* regs == NULL is a kernel bug */
+	if (WARN_ON(!regs))
+		return NOTIFY_DONE;
+
+	/* We are only interested in userspace traps */
+	if (!user_mode(regs))
+		return NOTIFY_DONE;
+
+	switch (val) {
+	case DIE_UPROBE:
+		if (uprobe_pre_sstep_notifier(regs))
+			return NOTIFY_STOP;
+		break;
+	case DIE_UPROBE_XOL:
+		if (uprobe_post_sstep_notifier(regs))
+			return NOTIFY_STOP;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+/*
+ * This function gets called when XOL instruction either gets trapped or
+ * the thread has a fatal signal. Reset the instruction pointer to its
+ * probed address for the potential restart or for post mortem analysis.
+ */
+void arch_uprobe_abort_xol(struct arch_uprobe *aup,
+	struct pt_regs *regs)
+{
+	struct uprobe_task *utask = current->utask;
+
+	instruction_pointer_set(regs, utask->vaddr);
+}
+
+unsigned long arch_uretprobe_hijack_return_addr(
+	unsigned long trampoline_vaddr, struct pt_regs *regs)
+{
+	unsigned long ra;
+
+	ra = regs->regs[31];
+
+	/* Replace the return address with the trampoline address */
+	regs->regs[31] = trampoline_vaddr;
+
+	return ra;
+}
+
+/**
+ * set_swbp - store breakpoint at a given address.
+ * @auprobe: arch specific probepoint information.
+ * @mm: the probed process address space.
+ * @vaddr: the virtual address to insert the opcode.
+ *
+ * For mm @mm, store the breakpoint instruction at @vaddr.
+ * Return 0 (success) or a negative errno.
+ *
+ * This version overrides the weak version in kernel/events/uprobes.c.
+ * It is required to handle MIPS16 and microMIPS.
+ */
+int __weak set_swbp(struct arch_uprobe *auprobe, struct mm_struct *mm,
+	unsigned long vaddr)
+{
+	return uprobe_write_opcode(auprobe, mm, vaddr, UPROBE_SWBP_INSN);
+}
+
+void arch_uprobe_copy_ixol(struct page *page, unsigned long vaddr,
+				  void *src, unsigned long len)
+{
+	unsigned long kaddr, kstart;
+
+	/* Initialize the slot */
+	kaddr = (unsigned long)kmap_atomic(page);
+	kstart = kaddr + (vaddr & ~PAGE_MASK);
+	memcpy((void *)kstart, src, len);
+	flush_icache_range(kstart, kstart + len);
+	kunmap_atomic((void *)kaddr);
+}
+
+/**
+ * uprobe_get_swbp_addr - compute address of swbp given post-swbp regs
+ * @regs: Reflects the saved state of the task after it has hit a breakpoint
+ * instruction.
+ * Return the address of the breakpoint instruction.
+ *
+ * This overrides the weak version in kernel/events/uprobes.c.
+ */
+unsigned long uprobe_get_swbp_addr(struct pt_regs *regs)
+{
+	return instruction_pointer(regs);
+}
+
+/*
+ * See if the instruction can be emulated.
+ * Returns true if instruction was emulated, false otherwise.
+ *
+ * For now we always emulate so this function just returns 0.
+ */
+bool arch_uprobe_skip_sstep(struct arch_uprobe *auprobe, struct pt_regs *regs)
+{
+	return 0;
+}
diff --git a/arch/mips/kernel/vdso.c b/arch/mips/kernel/vdso.c
new file mode 100644
index 000000000..242dc5e83
--- /dev/null
+++ b/arch/mips/kernel/vdso.c
@@ -0,0 +1,192 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2015 Imagination Technologies
+ * Author: Alex Smith <alex.smith@imgtec.com>
+ */
+
+#include <linux/binfmts.h>
+#include <linux/elf.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/ioport.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/random.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/timekeeper_internal.h>
+
+#include <asm/abi.h>
+#include <asm/mips-cps.h>
+#include <asm/page.h>
+#include <asm/vdso.h>
+#include <vdso/helpers.h>
+#include <vdso/vsyscall.h>
+
+/* Kernel-provided data used by the VDSO. */
+static union mips_vdso_data mips_vdso_data __page_aligned_data;
+struct vdso_data *vdso_data = mips_vdso_data.data;
+
+/*
+ * Mapping for the VDSO data/GIC pages. The real pages are mapped manually, as
+ * what we map and where within the area they are mapped is determined at
+ * runtime.
+ */
+static struct page *no_pages[] = { NULL };
+static struct vm_special_mapping vdso_vvar_mapping = {
+	.name = "[vvar]",
+	.pages = no_pages,
+};
+
+static void __init init_vdso_image(struct mips_vdso_image *image)
+{
+	unsigned long num_pages, i;
+	unsigned long data_pfn;
+
+	BUG_ON(!PAGE_ALIGNED(image->data));
+	BUG_ON(!PAGE_ALIGNED(image->size));
+
+	num_pages = image->size / PAGE_SIZE;
+
+	data_pfn = __phys_to_pfn(__pa_symbol(image->data));
+	for (i = 0; i < num_pages; i++)
+		image->mapping.pages[i] = pfn_to_page(data_pfn + i);
+}
+
+static int __init init_vdso(void)
+{
+	init_vdso_image(&vdso_image);
+
+#ifdef CONFIG_MIPS32_O32
+	init_vdso_image(&vdso_image_o32);
+#endif
+
+#ifdef CONFIG_MIPS32_N32
+	init_vdso_image(&vdso_image_n32);
+#endif
+
+	return 0;
+}
+subsys_initcall(init_vdso);
+
+static unsigned long vdso_base(void)
+{
+	unsigned long base = STACK_TOP;
+
+	if (IS_ENABLED(CONFIG_MIPS_FP_SUPPORT)) {
+		/* Skip the delay slot emulation page */
+		base += PAGE_SIZE;
+	}
+
+	if (current->flags & PF_RANDOMIZE) {
+		base += get_random_int() & (VDSO_RANDOMIZE_SIZE - 1);
+		base = PAGE_ALIGN(base);
+	}
+
+	return base;
+}
+
+int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp)
+{
+	struct mips_vdso_image *image = current->thread.abi->vdso;
+	struct mm_struct *mm = current->mm;
+	unsigned long gic_size, vvar_size, size, base, data_addr, vdso_addr, gic_pfn;
+	struct vm_area_struct *vma;
+	int ret;
+
+	if (mmap_write_lock_killable(mm))
+		return -EINTR;
+
+	if (IS_ENABLED(CONFIG_MIPS_FP_SUPPORT)) {
+		/* Map delay slot emulation page */
+		base = mmap_region(NULL, STACK_TOP, PAGE_SIZE,
+				VM_READ | VM_EXEC |
+				VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC,
+				0, NULL);
+		if (IS_ERR_VALUE(base)) {
+			ret = base;
+			goto out;
+		}
+	}
+
+	/*
+	 * Determine total area size. This includes the VDSO data itself, the
+	 * data page, and the GIC user page if present. Always create a mapping
+	 * for the GIC user area if the GIC is present regardless of whether it
+	 * is the current clocksource, in case it comes into use later on. We
+	 * only map a page even though the total area is 64K, as we only need
+	 * the counter registers at the start.
+	 */
+	gic_size = mips_gic_present() ? PAGE_SIZE : 0;
+	vvar_size = gic_size + PAGE_SIZE;
+	size = vvar_size + image->size;
+
+	/*
+	 * Find a region that's large enough for us to perform the
+	 * colour-matching alignment below.
+	 */
+	if (cpu_has_dc_aliases)
+		size += shm_align_mask + 1;
+
+	base = get_unmapped_area(NULL, vdso_base(), size, 0, 0);
+	if (IS_ERR_VALUE(base)) {
+		ret = base;
+		goto out;
+	}
+
+	/*
+	 * If we suffer from dcache aliasing, ensure that the VDSO data page
+	 * mapping is coloured the same as the kernel's mapping of that memory.
+	 * This ensures that when the kernel updates the VDSO data userland
+	 * will observe it without requiring cache invalidations.
+	 */
+	if (cpu_has_dc_aliases) {
+		base = __ALIGN_MASK(base, shm_align_mask);
+		base += ((unsigned long)vdso_data - gic_size) & shm_align_mask;
+	}
+
+	data_addr = base + gic_size;
+	vdso_addr = data_addr + PAGE_SIZE;
+
+	vma = _install_special_mapping(mm, base, vvar_size,
+				       VM_READ | VM_MAYREAD,
+				       &vdso_vvar_mapping);
+	if (IS_ERR(vma)) {
+		ret = PTR_ERR(vma);
+		goto out;
+	}
+
+	/* Map GIC user page. */
+	if (gic_size) {
+		gic_pfn = virt_to_phys(mips_gic_base + MIPS_GIC_USER_OFS) >> PAGE_SHIFT;
+
+		ret = io_remap_pfn_range(vma, base, gic_pfn, gic_size,
+					 pgprot_noncached(PAGE_READONLY));
+		if (ret)
+			goto out;
+	}
+
+	/* Map data page. */
+	ret = remap_pfn_range(vma, data_addr,
+			      virt_to_phys(vdso_data) >> PAGE_SHIFT,
+			      PAGE_SIZE, PAGE_READONLY);
+	if (ret)
+		goto out;
+
+	/* Map VDSO image. */
+	vma = _install_special_mapping(mm, vdso_addr, image->size,
+				       VM_READ | VM_EXEC |
+				       VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC,
+				       &image->mapping);
+	if (IS_ERR(vma)) {
+		ret = PTR_ERR(vma);
+		goto out;
+	}
+
+	mm->context.vdso = (void *)vdso_addr;
+	ret = 0;
+
+out:
+	mmap_write_unlock(mm);
+	return ret;
+}
diff --git a/arch/mips/kernel/vmlinux.lds.S b/arch/mips/kernel/vmlinux.lds.S
new file mode 100644
index 000000000..64afe075d
--- /dev/null
+++ b/arch/mips/kernel/vmlinux.lds.S
@@ -0,0 +1,230 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#include <asm/asm-offsets.h>
+#include <asm/thread_info.h>
+
+#define PAGE_SIZE _PAGE_SIZE
+
+/*
+ * Put .bss..swapper_pg_dir as the first thing in .bss. This will
+ * ensure that it has .bss alignment (64K).
+ */
+#define BSS_FIRST_SECTIONS *(.bss..swapper_pg_dir)
+
+/* Cavium Octeon should not have a separate PT_NOTE Program Header. */
+#ifndef CONFIG_CAVIUM_OCTEON_SOC
+#define EMITS_PT_NOTE
+#endif
+
+#define RUNTIME_DISCARD_EXIT
+
+#include <asm-generic/vmlinux.lds.h>
+
+#undef mips
+#define mips mips
+OUTPUT_ARCH(mips)
+ENTRY(kernel_entry)
+PHDRS {
+	text PT_LOAD FLAGS(7);	/* RWX */
+#ifndef CONFIG_CAVIUM_OCTEON_SOC
+	note PT_NOTE FLAGS(4);	/* R__ */
+#endif /* CAVIUM_OCTEON_SOC */
+}
+
+#ifdef CONFIG_32BIT
+	#ifdef CONFIG_CPU_LITTLE_ENDIAN
+		jiffies	 = jiffies_64;
+	#else
+		jiffies	 = jiffies_64 + 4;
+	#endif
+#else
+	jiffies	 = jiffies_64;
+#endif
+
+SECTIONS
+{
+#ifdef CONFIG_BOOT_ELF64
+	/* Read-only sections, merged into text segment: */
+	/* . = 0xc000000000000000; */
+
+	/* This is the value for an Origin kernel, taken from an IRIX kernel.  */
+	/* . = 0xc00000000001c000; */
+
+	/* Set the vaddr for the text segment to a value
+	 *   >= 0xa800 0000 0001 9000 if no symmon is going to configured
+	 *   >= 0xa800 0000 0030 0000 otherwise
+	 */
+
+	/* . = 0xa800000000300000; */
+	. = 0xffffffff80300000;
+#endif
+	. = LINKER_LOAD_ADDRESS;
+	/* read-only */
+	_text = .;	/* Text and read-only data */
+	.text : {
+		TEXT_TEXT
+		SCHED_TEXT
+		CPUIDLE_TEXT
+		LOCK_TEXT
+		KPROBES_TEXT
+		IRQENTRY_TEXT
+		SOFTIRQENTRY_TEXT
+		*(.text.*)
+		*(.fixup)
+		*(.gnu.warning)
+	} :text = 0
+	_etext = .;	/* End of text section */
+
+	EXCEPTION_TABLE(16)
+
+	/* Exception table for data bus errors */
+	__dbe_table : {
+		__start___dbe_table = .;
+		KEEP(*(__dbe_table))
+		__stop___dbe_table = .;
+	}
+
+	_sdata = .;			/* Start of data section */
+	RO_DATA(4096)
+
+	/* writeable */
+	.data : {	/* Data */
+		. = . + DATAOFFSET;		/* for CONFIG_MAPPED_KERNEL */
+
+		INIT_TASK_DATA(THREAD_SIZE)
+		NOSAVE_DATA
+		PAGE_ALIGNED_DATA(PAGE_SIZE)
+		CACHELINE_ALIGNED_DATA(1 << CONFIG_MIPS_L1_CACHE_SHIFT)
+		READ_MOSTLY_DATA(1 << CONFIG_MIPS_L1_CACHE_SHIFT)
+		DATA_DATA
+		CONSTRUCTORS
+	}
+	BUG_TABLE
+	_gp = . + 0x8000;
+	.lit8 : {
+		*(.lit8)
+	}
+	.lit4 : {
+		*(.lit4)
+	}
+	/* We want the small data sections together, so single-instruction offsets
+	   can access them all, and initialized data all before uninitialized, so
+	   we can shorten the on-disk segment size.  */
+	.sdata : {
+		*(.sdata)
+	}
+	_edata =  .;			/* End of data section */
+
+	/* will be freed after init */
+	. = ALIGN(PAGE_SIZE);		/* Init code and data */
+	__init_begin = .;
+	INIT_TEXT_SECTION(PAGE_SIZE)
+	INIT_DATA_SECTION(16)
+
+	. = ALIGN(4);
+	.mips.machines.init : AT(ADDR(.mips.machines.init) - LOAD_OFFSET) {
+		__mips_machines_start = .;
+		KEEP(*(.mips.machines.init))
+		__mips_machines_end = .;
+	}
+
+	/* .exit.text is discarded at runtime, not link time, to deal with
+	 * references from .rodata
+	 */
+	.exit.text : {
+		EXIT_TEXT
+	}
+	.exit.data : {
+		EXIT_DATA
+	}
+#ifdef CONFIG_SMP
+	PERCPU_SECTION(1 << CONFIG_MIPS_L1_CACHE_SHIFT)
+#endif
+
+#ifdef CONFIG_MIPS_ELF_APPENDED_DTB
+	.appended_dtb : AT(ADDR(.appended_dtb) - LOAD_OFFSET) {
+		*(.appended_dtb)
+		KEEP(*(.appended_dtb))
+	}
+#endif
+
+#ifdef CONFIG_RELOCATABLE
+	. = ALIGN(4);
+
+	.data.reloc : {
+		_relocation_start = .;
+		/*
+		 * Space for relocation table
+		 * This needs to be filled so that the
+		 * relocs tool can overwrite the content.
+		 * An invalid value is left at the start of the
+		 * section to abort relocation if the table
+		 * has not been filled in.
+		 */
+		LONG(0xFFFFFFFF);
+		FILL(0);
+		. += CONFIG_RELOCATION_TABLE_SIZE - 4;
+		_relocation_end = .;
+	}
+#endif
+
+#ifdef CONFIG_MIPS_RAW_APPENDED_DTB
+	__appended_dtb = .;
+	/* leave space for appended DTB */
+	. += 0x100000;
+#endif
+	/*
+	 * Align to 64K in attempt to eliminate holes before the
+	 * .bss..swapper_pg_dir section at the start of .bss.  This
+	 * also satisfies PAGE_SIZE alignment as the largest page size
+	 * allowed is 64K.
+	 */
+	. = ALIGN(0x10000);
+	__init_end = .;
+	/* freed after init ends here */
+
+	/*
+	 * Force .bss to 64K alignment so that .bss..swapper_pg_dir
+	 * gets that alignment.	 .sbss should be empty, so there will be
+	 * no holes after __init_end. */
+	BSS_SECTION(0, 0x10000, 8)
+
+	_end = . ;
+
+	/* These mark the ABI of the kernel for debuggers.  */
+	.mdebug.abi32 : {
+		KEEP(*(.mdebug.abi32))
+	}
+	.mdebug.abi64 : {
+		KEEP(*(.mdebug.abi64))
+	}
+
+	/* This is the MIPS specific mdebug section.  */
+	.mdebug : {
+		*(.mdebug)
+	}
+
+	STABS_DEBUG
+	DWARF_DEBUG
+	ELF_DETAILS
+
+	/* These must appear regardless of  .  */
+	.gptab.sdata : {
+		*(.gptab.data)
+		*(.gptab.sdata)
+	}
+	.gptab.sbss : {
+		*(.gptab.bss)
+		*(.gptab.sbss)
+	}
+
+	/* Sections to be discarded */
+	DISCARDS
+	/DISCARD/ : {
+		/* ABI crap starts here */
+		*(.MIPS.abiflags)
+		*(.MIPS.options)
+		*(.options)
+		*(.pdr)
+		*(.reginfo)
+	}
+}
diff --git a/arch/mips/kernel/vpe-cmp.c b/arch/mips/kernel/vpe-cmp.c
new file mode 100644
index 000000000..903c07bdc
--- /dev/null
+++ b/arch/mips/kernel/vpe-cmp.c
@@ -0,0 +1,180 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2004, 2005 MIPS Technologies, Inc.  All rights reserved.
+ * Copyright (C) 2013 Imagination Technologies Ltd.
+ */
+#include <linux/kernel.h>
+#include <linux/device.h>
+#include <linux/fs.h>
+#include <linux/slab.h>
+#include <linux/export.h>
+
+#include <asm/vpe.h>
+
+static int major;
+
+void cleanup_tc(struct tc *tc)
+{
+
+}
+
+static ssize_t store_kill(struct device *dev, struct device_attribute *attr,
+			  const char *buf, size_t len)
+{
+	struct vpe *vpe = get_vpe(aprp_cpu_index());
+	struct vpe_notifications *notifier;
+
+	list_for_each_entry(notifier, &vpe->notify, list)
+		notifier->stop(aprp_cpu_index());
+
+	release_progmem(vpe->load_addr);
+	vpe->state = VPE_STATE_UNUSED;
+
+	return len;
+}
+static DEVICE_ATTR(kill, S_IWUSR, NULL, store_kill);
+
+static ssize_t ntcs_show(struct device *cd, struct device_attribute *attr,
+			 char *buf)
+{
+	struct vpe *vpe = get_vpe(aprp_cpu_index());
+
+	return sprintf(buf, "%d\n", vpe->ntcs);
+}
+
+static ssize_t ntcs_store(struct device *dev, struct device_attribute *attr,
+			  const char *buf, size_t len)
+{
+	struct vpe *vpe = get_vpe(aprp_cpu_index());
+	unsigned long new;
+	int ret;
+
+	ret = kstrtoul(buf, 0, &new);
+	if (ret < 0)
+		return ret;
+
+	/* APRP can only reserve one TC in a VPE and no more. */
+	if (new != 1)
+		return -EINVAL;
+
+	vpe->ntcs = new;
+
+	return len;
+}
+static DEVICE_ATTR_RW(ntcs);
+
+static struct attribute *vpe_attrs[] = {
+	&dev_attr_kill.attr,
+	&dev_attr_ntcs.attr,
+	NULL,
+};
+ATTRIBUTE_GROUPS(vpe);
+
+static void vpe_device_release(struct device *cd)
+{
+}
+
+static struct class vpe_class = {
+	.name = "vpe",
+	.owner = THIS_MODULE,
+	.dev_release = vpe_device_release,
+	.dev_groups = vpe_groups,
+};
+
+static struct device vpe_device;
+
+int __init vpe_module_init(void)
+{
+	struct vpe *v = NULL;
+	struct tc *t;
+	int err;
+
+	if (!cpu_has_mipsmt) {
+		pr_warn("VPE loader: not a MIPS MT capable processor\n");
+		return -ENODEV;
+	}
+
+	if (num_possible_cpus() - aprp_cpu_index() < 1) {
+		pr_warn("No VPEs reserved for AP/SP, not initialize VPE loader\n"
+			"Pass maxcpus=<n> argument as kernel argument\n");
+		return -ENODEV;
+	}
+
+	major = register_chrdev(0, VPE_MODULE_NAME, &vpe_fops);
+	if (major < 0) {
+		pr_warn("VPE loader: unable to register character device\n");
+		return major;
+	}
+
+	err = class_register(&vpe_class);
+	if (err) {
+		pr_err("vpe_class registration failed\n");
+		goto out_chrdev;
+	}
+
+	device_initialize(&vpe_device);
+	vpe_device.class	= &vpe_class,
+	vpe_device.parent	= NULL,
+	dev_set_name(&vpe_device, "vpe_sp");
+	vpe_device.devt = MKDEV(major, VPE_MODULE_MINOR);
+	err = device_add(&vpe_device);
+	if (err) {
+		pr_err("Adding vpe_device failed\n");
+		goto out_class;
+	}
+
+	t = alloc_tc(aprp_cpu_index());
+	if (!t) {
+		pr_warn("VPE: unable to allocate TC\n");
+		err = -ENOMEM;
+		goto out_dev;
+	}
+
+	/* VPE */
+	v = alloc_vpe(aprp_cpu_index());
+	if (v == NULL) {
+		pr_warn("VPE: unable to allocate VPE\n");
+		kfree(t);
+		err = -ENOMEM;
+		goto out_dev;
+	}
+
+	v->ntcs = 1;
+
+	/* add the tc to the list of this vpe's tc's. */
+	list_add(&t->tc, &v->tc);
+
+	/* TC */
+	t->pvpe = v;	/* set the parent vpe */
+
+	return 0;
+
+out_dev:
+	device_del(&vpe_device);
+
+out_class:
+	put_device(&vpe_device);
+	class_unregister(&vpe_class);
+
+out_chrdev:
+	unregister_chrdev(major, VPE_MODULE_NAME);
+
+	return err;
+}
+
+void __exit vpe_module_exit(void)
+{
+	struct vpe *v, *n;
+
+	device_unregister(&vpe_device);
+	class_unregister(&vpe_class);
+	unregister_chrdev(major, VPE_MODULE_NAME);
+
+	/* No locking needed here */
+	list_for_each_entry_safe(v, n, &vpecontrol.vpe_list, list)
+		if (v->state != VPE_STATE_UNUSED)
+			release_vpe(v);
+}
diff --git a/arch/mips/kernel/vpe-mt.c b/arch/mips/kernel/vpe-mt.c
new file mode 100644
index 000000000..496ed8f36
--- /dev/null
+++ b/arch/mips/kernel/vpe-mt.c
@@ -0,0 +1,520 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2004, 2005 MIPS Technologies, Inc.  All rights reserved.
+ * Copyright (C) 2013 Imagination Technologies Ltd.
+ */
+#include <linux/kernel.h>
+#include <linux/device.h>
+#include <linux/fs.h>
+#include <linux/slab.h>
+#include <linux/export.h>
+
+#include <asm/mipsregs.h>
+#include <asm/mipsmtregs.h>
+#include <asm/mips_mt.h>
+#include <asm/vpe.h>
+
+static int major;
+
+/* The number of TCs and VPEs physically available on the core */
+static int hw_tcs, hw_vpes;
+
+/* We are prepared so configure and start the VPE... */
+int vpe_run(struct vpe *v)
+{
+	unsigned long flags, val, dmt_flag;
+	struct vpe_notifications *notifier;
+	unsigned int vpeflags;
+	struct tc *t;
+
+	/* check we are the Master VPE */
+	local_irq_save(flags);
+	val = read_c0_vpeconf0();
+	if (!(val & VPECONF0_MVP)) {
+		pr_warn("VPE loader: only Master VPE's are able to config MT\n");
+		local_irq_restore(flags);
+
+		return -1;
+	}
+
+	dmt_flag = dmt();
+	vpeflags = dvpe();
+
+	if (list_empty(&v->tc)) {
+		evpe(vpeflags);
+		emt(dmt_flag);
+		local_irq_restore(flags);
+
+		pr_warn("VPE loader: No TC's associated with VPE %d\n",
+			v->minor);
+
+		return -ENOEXEC;
+	}
+
+	t = list_first_entry(&v->tc, struct tc, tc);
+
+	/* Put MVPE's into 'configuration state' */
+	set_c0_mvpcontrol(MVPCONTROL_VPC);
+
+	settc(t->index);
+
+	/* should check it is halted, and not activated */
+	if ((read_tc_c0_tcstatus() & TCSTATUS_A) ||
+	   !(read_tc_c0_tchalt() & TCHALT_H)) {
+		evpe(vpeflags);
+		emt(dmt_flag);
+		local_irq_restore(flags);
+
+		pr_warn("VPE loader: TC %d is already active!\n",
+			t->index);
+
+		return -ENOEXEC;
+	}
+
+	/*
+	 * Write the address we want it to start running from in the TCPC
+	 * register.
+	 */
+	write_tc_c0_tcrestart((unsigned long)v->__start);
+	write_tc_c0_tccontext((unsigned long)0);
+
+	/*
+	 * Mark the TC as activated, not interrupt exempt and not dynamically
+	 * allocatable
+	 */
+	val = read_tc_c0_tcstatus();
+	val = (val & ~(TCSTATUS_DA | TCSTATUS_IXMT)) | TCSTATUS_A;
+	write_tc_c0_tcstatus(val);
+
+	write_tc_c0_tchalt(read_tc_c0_tchalt() & ~TCHALT_H);
+
+	/*
+	 * We don't pass the memsize here, so VPE programs need to be
+	 * compiled with DFLT_STACK_SIZE and DFLT_HEAP_SIZE defined.
+	 */
+	mttgpr(7, 0);
+	mttgpr(6, v->ntcs);
+
+	/* set up VPE1 */
+	/*
+	 * bind the TC to VPE 1 as late as possible so we only have the final
+	 * VPE registers to set up, and so an EJTAG probe can trigger on it
+	 */
+	write_tc_c0_tcbind((read_tc_c0_tcbind() & ~TCBIND_CURVPE) | 1);
+
+	write_vpe_c0_vpeconf0(read_vpe_c0_vpeconf0() & ~(VPECONF0_VPA));
+
+	back_to_back_c0_hazard();
+
+	/* Set up the XTC bit in vpeconf0 to point at our tc */
+	write_vpe_c0_vpeconf0((read_vpe_c0_vpeconf0() & ~(VPECONF0_XTC))
+			      | (t->index << VPECONF0_XTC_SHIFT));
+
+	back_to_back_c0_hazard();
+
+	/* enable this VPE */
+	write_vpe_c0_vpeconf0(read_vpe_c0_vpeconf0() | VPECONF0_VPA);
+
+	/* clear out any left overs from a previous program */
+	write_vpe_c0_status(0);
+	write_vpe_c0_cause(0);
+
+	/* take system out of configuration state */
+	clear_c0_mvpcontrol(MVPCONTROL_VPC);
+
+	/*
+	 * SMVP kernels manage VPE enable independently, but uniprocessor
+	 * kernels need to turn it on, even if that wasn't the pre-dvpe() state.
+	 */
+#ifdef CONFIG_SMP
+	evpe(vpeflags);
+#else
+	evpe(EVPE_ENABLE);
+#endif
+	emt(dmt_flag);
+	local_irq_restore(flags);
+
+	list_for_each_entry(notifier, &v->notify, list)
+		notifier->start(VPE_MODULE_MINOR);
+
+	return 0;
+}
+
+void cleanup_tc(struct tc *tc)
+{
+	unsigned long flags;
+	unsigned int mtflags, vpflags;
+	int tmp;
+
+	local_irq_save(flags);
+	mtflags = dmt();
+	vpflags = dvpe();
+	/* Put MVPE's into 'configuration state' */
+	set_c0_mvpcontrol(MVPCONTROL_VPC);
+
+	settc(tc->index);
+	tmp = read_tc_c0_tcstatus();
+
+	/* mark not allocated and not dynamically allocatable */
+	tmp &= ~(TCSTATUS_A | TCSTATUS_DA);
+	tmp |= TCSTATUS_IXMT;	/* interrupt exempt */
+	write_tc_c0_tcstatus(tmp);
+
+	write_tc_c0_tchalt(TCHALT_H);
+	mips_ihb();
+
+	clear_c0_mvpcontrol(MVPCONTROL_VPC);
+	evpe(vpflags);
+	emt(mtflags);
+	local_irq_restore(flags);
+}
+
+/* module wrapper entry points */
+/* give me a vpe */
+void *vpe_alloc(void)
+{
+	int i;
+	struct vpe *v;
+
+	/* find a vpe */
+	for (i = 1; i < MAX_VPES; i++) {
+		v = get_vpe(i);
+		if (v != NULL) {
+			v->state = VPE_STATE_INUSE;
+			return v;
+		}
+	}
+	return NULL;
+}
+EXPORT_SYMBOL(vpe_alloc);
+
+/* start running from here */
+int vpe_start(void *vpe, unsigned long start)
+{
+	struct vpe *v = vpe;
+
+	v->__start = start;
+	return vpe_run(v);
+}
+EXPORT_SYMBOL(vpe_start);
+
+/* halt it for now */
+int vpe_stop(void *vpe)
+{
+	struct vpe *v = vpe;
+	struct tc *t;
+	unsigned int evpe_flags;
+
+	evpe_flags = dvpe();
+
+	t = list_entry(v->tc.next, struct tc, tc);
+	if (t != NULL) {
+		settc(t->index);
+		write_vpe_c0_vpeconf0(read_vpe_c0_vpeconf0() & ~VPECONF0_VPA);
+	}
+
+	evpe(evpe_flags);
+
+	return 0;
+}
+EXPORT_SYMBOL(vpe_stop);
+
+/* I've done with it thank you */
+int vpe_free(void *vpe)
+{
+	struct vpe *v = vpe;
+	struct tc *t;
+	unsigned int evpe_flags;
+
+	t = list_entry(v->tc.next, struct tc, tc);
+	if (t == NULL)
+		return -ENOEXEC;
+
+	evpe_flags = dvpe();
+
+	/* Put MVPE's into 'configuration state' */
+	set_c0_mvpcontrol(MVPCONTROL_VPC);
+
+	settc(t->index);
+	write_vpe_c0_vpeconf0(read_vpe_c0_vpeconf0() & ~VPECONF0_VPA);
+
+	/* halt the TC */
+	write_tc_c0_tchalt(TCHALT_H);
+	mips_ihb();
+
+	/* mark the TC unallocated */
+	write_tc_c0_tcstatus(read_tc_c0_tcstatus() & ~TCSTATUS_A);
+
+	v->state = VPE_STATE_UNUSED;
+
+	clear_c0_mvpcontrol(MVPCONTROL_VPC);
+	evpe(evpe_flags);
+
+	return 0;
+}
+EXPORT_SYMBOL(vpe_free);
+
+static ssize_t store_kill(struct device *dev, struct device_attribute *attr,
+			  const char *buf, size_t len)
+{
+	struct vpe *vpe = get_vpe(aprp_cpu_index());
+	struct vpe_notifications *notifier;
+
+	list_for_each_entry(notifier, &vpe->notify, list)
+		notifier->stop(aprp_cpu_index());
+
+	release_progmem(vpe->load_addr);
+	cleanup_tc(get_tc(aprp_cpu_index()));
+	vpe_stop(vpe);
+	vpe_free(vpe);
+
+	return len;
+}
+static DEVICE_ATTR(kill, S_IWUSR, NULL, store_kill);
+
+static ssize_t ntcs_show(struct device *cd, struct device_attribute *attr,
+			 char *buf)
+{
+	struct vpe *vpe = get_vpe(aprp_cpu_index());
+
+	return sprintf(buf, "%d\n", vpe->ntcs);
+}
+
+static ssize_t ntcs_store(struct device *dev, struct device_attribute *attr,
+			  const char *buf, size_t len)
+{
+	struct vpe *vpe = get_vpe(aprp_cpu_index());
+	unsigned long new;
+	int ret;
+
+	ret = kstrtoul(buf, 0, &new);
+	if (ret < 0)
+		return ret;
+
+	if (new == 0 || new > (hw_tcs - aprp_cpu_index()))
+		return -EINVAL;
+
+	vpe->ntcs = new;
+
+	return len;
+}
+static DEVICE_ATTR_RW(ntcs);
+
+static struct attribute *vpe_attrs[] = {
+	&dev_attr_kill.attr,
+	&dev_attr_ntcs.attr,
+	NULL,
+};
+ATTRIBUTE_GROUPS(vpe);
+
+static void vpe_device_release(struct device *cd)
+{
+}
+
+static struct class vpe_class = {
+	.name = "vpe",
+	.owner = THIS_MODULE,
+	.dev_release = vpe_device_release,
+	.dev_groups = vpe_groups,
+};
+
+static struct device vpe_device;
+
+int __init vpe_module_init(void)
+{
+	unsigned int mtflags, vpflags;
+	unsigned long flags, val;
+	struct vpe *v = NULL;
+	struct tc *t;
+	int tc, err;
+
+	if (!cpu_has_mipsmt) {
+		pr_warn("VPE loader: not a MIPS MT capable processor\n");
+		return -ENODEV;
+	}
+
+	if (vpelimit == 0) {
+		pr_warn("No VPEs reserved for AP/SP, not initialize VPE loader\n"
+			"Pass maxvpes=<n> argument as kernel argument\n");
+
+		return -ENODEV;
+	}
+
+	if (aprp_cpu_index() == 0) {
+		pr_warn("No TCs reserved for AP/SP, not initialize VPE loader\n"
+			"Pass maxtcs=<n> argument as kernel argument\n");
+
+		return -ENODEV;
+	}
+
+	major = register_chrdev(0, VPE_MODULE_NAME, &vpe_fops);
+	if (major < 0) {
+		pr_warn("VPE loader: unable to register character device\n");
+		return major;
+	}
+
+	err = class_register(&vpe_class);
+	if (err) {
+		pr_err("vpe_class registration failed\n");
+		goto out_chrdev;
+	}
+
+	device_initialize(&vpe_device);
+	vpe_device.class	= &vpe_class,
+	vpe_device.parent	= NULL,
+	dev_set_name(&vpe_device, "vpe1");
+	vpe_device.devt = MKDEV(major, VPE_MODULE_MINOR);
+	err = device_add(&vpe_device);
+	if (err) {
+		pr_err("Adding vpe_device failed\n");
+		goto out_class;
+	}
+
+	local_irq_save(flags);
+	mtflags = dmt();
+	vpflags = dvpe();
+
+	/* Put MVPE's into 'configuration state' */
+	set_c0_mvpcontrol(MVPCONTROL_VPC);
+
+	val = read_c0_mvpconf0();
+	hw_tcs = (val & MVPCONF0_PTC) + 1;
+	hw_vpes = ((val & MVPCONF0_PVPE) >> MVPCONF0_PVPE_SHIFT) + 1;
+
+	for (tc = aprp_cpu_index(); tc < hw_tcs; tc++) {
+		/*
+		 * Must re-enable multithreading temporarily or in case we
+		 * reschedule send IPIs or similar we might hang.
+		 */
+		clear_c0_mvpcontrol(MVPCONTROL_VPC);
+		evpe(vpflags);
+		emt(mtflags);
+		local_irq_restore(flags);
+		t = alloc_tc(tc);
+		if (!t) {
+			err = -ENOMEM;
+			goto out_dev;
+		}
+
+		local_irq_save(flags);
+		mtflags = dmt();
+		vpflags = dvpe();
+		set_c0_mvpcontrol(MVPCONTROL_VPC);
+
+		/* VPE's */
+		if (tc < hw_tcs) {
+			settc(tc);
+
+			v = alloc_vpe(tc);
+			if (v == NULL) {
+				pr_warn("VPE: unable to allocate VPE\n");
+				goto out_reenable;
+			}
+
+			v->ntcs = hw_tcs - aprp_cpu_index();
+
+			/* add the tc to the list of this vpe's tc's. */
+			list_add(&t->tc, &v->tc);
+
+			/* deactivate all but vpe0 */
+			if (tc >= aprp_cpu_index()) {
+				unsigned long tmp = read_vpe_c0_vpeconf0();
+
+				tmp &= ~VPECONF0_VPA;
+
+				/* master VPE */
+				tmp |= VPECONF0_MVP;
+				write_vpe_c0_vpeconf0(tmp);
+			}
+
+			/* disable multi-threading with TC's */
+			write_vpe_c0_vpecontrol(read_vpe_c0_vpecontrol() &
+						~VPECONTROL_TE);
+
+			if (tc >= vpelimit) {
+				/*
+				 * Set config to be the same as vpe0,
+				 * particularly kseg0 coherency alg
+				 */
+				write_vpe_c0_config(read_c0_config());
+			}
+		}
+
+		/* TC's */
+		t->pvpe = v;	/* set the parent vpe */
+
+		if (tc >= aprp_cpu_index()) {
+			unsigned long tmp;
+
+			settc(tc);
+
+			/*
+			 * A TC that is bound to any other VPE gets bound to
+			 * VPE0, ideally I'd like to make it homeless but it
+			 * doesn't appear to let me bind a TC to a non-existent
+			 * VPE. Which is perfectly reasonable.
+			 *
+			 * The (un)bound state is visible to an EJTAG probe so
+			 * may notify GDB...
+			 */
+			tmp = read_tc_c0_tcbind();
+			if (tmp & TCBIND_CURVPE) {
+				/* tc is bound >vpe0 */
+				write_tc_c0_tcbind(tmp & ~TCBIND_CURVPE);
+
+				t->pvpe = get_vpe(0);	/* set the parent vpe */
+			}
+
+			/* halt the TC */
+			write_tc_c0_tchalt(TCHALT_H);
+			mips_ihb();
+
+			tmp = read_tc_c0_tcstatus();
+
+			/* mark not activated and not dynamically allocatable */
+			tmp &= ~(TCSTATUS_A | TCSTATUS_DA);
+			tmp |= TCSTATUS_IXMT;	/* interrupt exempt */
+			write_tc_c0_tcstatus(tmp);
+		}
+	}
+
+out_reenable:
+	/* release config state */
+	clear_c0_mvpcontrol(MVPCONTROL_VPC);
+
+	evpe(vpflags);
+	emt(mtflags);
+	local_irq_restore(flags);
+
+	return 0;
+
+out_dev:
+	device_del(&vpe_device);
+
+out_class:
+	put_device(&vpe_device);
+	class_unregister(&vpe_class);
+
+out_chrdev:
+	unregister_chrdev(major, VPE_MODULE_NAME);
+
+	return err;
+}
+
+void __exit vpe_module_exit(void)
+{
+	struct vpe *v, *n;
+
+	device_unregister(&vpe_device);
+	class_unregister(&vpe_class);
+	unregister_chrdev(major, VPE_MODULE_NAME);
+
+	/* No locking needed here */
+	list_for_each_entry_safe(v, n, &vpecontrol.vpe_list, list) {
+		if (v->state != VPE_STATE_UNUSED)
+			release_vpe(v);
+	}
+}
diff --git a/arch/mips/kernel/vpe.c b/arch/mips/kernel/vpe.c
new file mode 100644
index 000000000..d0d832ab3
--- /dev/null
+++ b/arch/mips/kernel/vpe.c
@@ -0,0 +1,933 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2004, 2005 MIPS Technologies, Inc.  All rights reserved.
+ * Copyright (C) 2013 Imagination Technologies Ltd.
+ *
+ * VPE spport module for loading a MIPS SP program into VPE1. The SP
+ * environment is rather simple since there are no TLBs. It needs
+ * to be relocatable (or partiall linked). Initialize your stack in
+ * the startup-code. The loader looks for the symbol __start and sets
+ * up the execution to resume from there. To load and run, simply do
+ * a cat SP 'binary' to the /dev/vpe1 device.
+ */
+#include <linux/kernel.h>
+#include <linux/device.h>
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/list.h>
+#include <linux/vmalloc.h>
+#include <linux/elf.h>
+#include <linux/seq_file.h>
+#include <linux/syscalls.h>
+#include <linux/moduleloader.h>
+#include <linux/interrupt.h>
+#include <linux/poll.h>
+#include <linux/memblock.h>
+#include <asm/mipsregs.h>
+#include <asm/mipsmtregs.h>
+#include <asm/cacheflush.h>
+#include <linux/atomic.h>
+#include <asm/mips_mt.h>
+#include <asm/processor.h>
+#include <asm/vpe.h>
+
+#ifndef ARCH_SHF_SMALL
+#define ARCH_SHF_SMALL 0
+#endif
+
+/* If this is set, the section belongs in the init part of the module */
+#define INIT_OFFSET_MASK (1UL << (BITS_PER_LONG-1))
+
+struct vpe_control vpecontrol = {
+	.vpe_list_lock	= __SPIN_LOCK_UNLOCKED(vpe_list_lock),
+	.vpe_list	= LIST_HEAD_INIT(vpecontrol.vpe_list),
+	.tc_list_lock	= __SPIN_LOCK_UNLOCKED(tc_list_lock),
+	.tc_list	= LIST_HEAD_INIT(vpecontrol.tc_list)
+};
+
+/* get the vpe associated with this minor */
+struct vpe *get_vpe(int minor)
+{
+	struct vpe *res, *v;
+
+	if (!cpu_has_mipsmt)
+		return NULL;
+
+	res = NULL;
+	spin_lock(&vpecontrol.vpe_list_lock);
+	list_for_each_entry(v, &vpecontrol.vpe_list, list) {
+		if (v->minor == VPE_MODULE_MINOR) {
+			res = v;
+			break;
+		}
+	}
+	spin_unlock(&vpecontrol.vpe_list_lock);
+
+	return res;
+}
+
+/* get the vpe associated with this minor */
+struct tc *get_tc(int index)
+{
+	struct tc *res, *t;
+
+	res = NULL;
+	spin_lock(&vpecontrol.tc_list_lock);
+	list_for_each_entry(t, &vpecontrol.tc_list, list) {
+		if (t->index == index) {
+			res = t;
+			break;
+		}
+	}
+	spin_unlock(&vpecontrol.tc_list_lock);
+
+	return res;
+}
+
+/* allocate a vpe and associate it with this minor (or index) */
+struct vpe *alloc_vpe(int minor)
+{
+	struct vpe *v;
+
+	v = kzalloc(sizeof(struct vpe), GFP_KERNEL);
+	if (v == NULL)
+		goto out;
+
+	INIT_LIST_HEAD(&v->tc);
+	spin_lock(&vpecontrol.vpe_list_lock);
+	list_add_tail(&v->list, &vpecontrol.vpe_list);
+	spin_unlock(&vpecontrol.vpe_list_lock);
+
+	INIT_LIST_HEAD(&v->notify);
+	v->minor = VPE_MODULE_MINOR;
+
+out:
+	return v;
+}
+
+/* allocate a tc. At startup only tc0 is running, all other can be halted. */
+struct tc *alloc_tc(int index)
+{
+	struct tc *tc;
+
+	tc = kzalloc(sizeof(struct tc), GFP_KERNEL);
+	if (tc == NULL)
+		goto out;
+
+	INIT_LIST_HEAD(&tc->tc);
+	tc->index = index;
+
+	spin_lock(&vpecontrol.tc_list_lock);
+	list_add_tail(&tc->list, &vpecontrol.tc_list);
+	spin_unlock(&vpecontrol.tc_list_lock);
+
+out:
+	return tc;
+}
+
+/* clean up and free everything */
+void release_vpe(struct vpe *v)
+{
+	list_del(&v->list);
+	if (v->load_addr)
+		release_progmem(v->load_addr);
+	kfree(v);
+}
+
+/* Find some VPE program space */
+void *alloc_progmem(unsigned long len)
+{
+	void *addr;
+
+#ifdef CONFIG_MIPS_VPE_LOADER_TOM
+	/*
+	 * This means you must tell Linux to use less memory than you
+	 * physically have, for example by passing a mem= boot argument.
+	 */
+	addr = pfn_to_kaddr(max_low_pfn);
+	memset(addr, 0, len);
+#else
+	/* simple grab some mem for now */
+	addr = kzalloc(len, GFP_KERNEL);
+#endif
+
+	return addr;
+}
+
+void release_progmem(void *ptr)
+{
+#ifndef CONFIG_MIPS_VPE_LOADER_TOM
+	kfree(ptr);
+#endif
+}
+
+/* Update size with this section: return offset. */
+static long get_offset(unsigned long *size, Elf_Shdr *sechdr)
+{
+	long ret;
+
+	ret = ALIGN(*size, sechdr->sh_addralign ? : 1);
+	*size = ret + sechdr->sh_size;
+	return ret;
+}
+
+/* Lay out the SHF_ALLOC sections in a way not dissimilar to how ld
+   might -- code, read-only data, read-write data, small data.	Tally
+   sizes, and place the offsets into sh_entsize fields: high bit means it
+   belongs in init. */
+static void layout_sections(struct module *mod, const Elf_Ehdr *hdr,
+			    Elf_Shdr *sechdrs, const char *secstrings)
+{
+	static unsigned long const masks[][2] = {
+		/* NOTE: all executable code must be the first section
+		 * in this array; otherwise modify the text_size
+		 * finder in the two loops below */
+		{SHF_EXECINSTR | SHF_ALLOC, ARCH_SHF_SMALL},
+		{SHF_ALLOC, SHF_WRITE | ARCH_SHF_SMALL},
+		{SHF_WRITE | SHF_ALLOC, ARCH_SHF_SMALL},
+		{ARCH_SHF_SMALL | SHF_ALLOC, 0}
+	};
+	unsigned int m, i;
+
+	for (i = 0; i < hdr->e_shnum; i++)
+		sechdrs[i].sh_entsize = ~0UL;
+
+	for (m = 0; m < ARRAY_SIZE(masks); ++m) {
+		for (i = 0; i < hdr->e_shnum; ++i) {
+			Elf_Shdr *s = &sechdrs[i];
+
+			if ((s->sh_flags & masks[m][0]) != masks[m][0]
+			    || (s->sh_flags & masks[m][1])
+			    || s->sh_entsize != ~0UL)
+				continue;
+			s->sh_entsize =
+				get_offset((unsigned long *)&mod->core_layout.size, s);
+		}
+
+		if (m == 0)
+			mod->core_layout.text_size = mod->core_layout.size;
+
+	}
+}
+
+/* from module-elf32.c, but subverted a little */
+
+struct mips_hi16 {
+	struct mips_hi16 *next;
+	Elf32_Addr *addr;
+	Elf32_Addr value;
+};
+
+static struct mips_hi16 *mips_hi16_list;
+static unsigned int gp_offs, gp_addr;
+
+static int apply_r_mips_none(struct module *me, uint32_t *location,
+			     Elf32_Addr v)
+{
+	return 0;
+}
+
+static int apply_r_mips_gprel16(struct module *me, uint32_t *location,
+				Elf32_Addr v)
+{
+	int rel;
+
+	if (!(*location & 0xffff)) {
+		rel = (int)v - gp_addr;
+	} else {
+		/* .sbss + gp(relative) + offset */
+		/* kludge! */
+		rel =  (int)(short)((int)v + gp_offs +
+				    (int)(short)(*location & 0xffff) - gp_addr);
+	}
+
+	if ((rel > 32768) || (rel < -32768)) {
+		pr_debug("VPE loader: apply_r_mips_gprel16: relative address 0x%x out of range of gp register\n",
+			 rel);
+		return -ENOEXEC;
+	}
+
+	*location = (*location & 0xffff0000) | (rel & 0xffff);
+
+	return 0;
+}
+
+static int apply_r_mips_pc16(struct module *me, uint32_t *location,
+			     Elf32_Addr v)
+{
+	int rel;
+	rel = (((unsigned int)v - (unsigned int)location));
+	rel >>= 2; /* because the offset is in _instructions_ not bytes. */
+	rel -= 1;  /* and one instruction less due to the branch delay slot. */
+
+	if ((rel > 32768) || (rel < -32768)) {
+		pr_debug("VPE loader: apply_r_mips_pc16: relative address out of range 0x%x\n",
+			 rel);
+		return -ENOEXEC;
+	}
+
+	*location = (*location & 0xffff0000) | (rel & 0xffff);
+
+	return 0;
+}
+
+static int apply_r_mips_32(struct module *me, uint32_t *location,
+			   Elf32_Addr v)
+{
+	*location += v;
+
+	return 0;
+}
+
+static int apply_r_mips_26(struct module *me, uint32_t *location,
+			   Elf32_Addr v)
+{
+	if (v % 4) {
+		pr_debug("VPE loader: apply_r_mips_26: unaligned relocation\n");
+		return -ENOEXEC;
+	}
+
+/*
+ * Not desperately convinced this is a good check of an overflow condition
+ * anyway. But it gets in the way of handling undefined weak symbols which
+ * we want to set to zero.
+ * if ((v & 0xf0000000) != (((unsigned long)location + 4) & 0xf0000000)) {
+ * printk(KERN_ERR
+ * "module %s: relocation overflow\n",
+ * me->name);
+ * return -ENOEXEC;
+ * }
+ */
+
+	*location = (*location & ~0x03ffffff) |
+		((*location + (v >> 2)) & 0x03ffffff);
+	return 0;
+}
+
+static int apply_r_mips_hi16(struct module *me, uint32_t *location,
+			     Elf32_Addr v)
+{
+	struct mips_hi16 *n;
+
+	/*
+	 * We cannot relocate this one now because we don't know the value of
+	 * the carry we need to add.  Save the information, and let LO16 do the
+	 * actual relocation.
+	 */
+	n = kmalloc(sizeof(*n), GFP_KERNEL);
+	if (!n)
+		return -ENOMEM;
+
+	n->addr = location;
+	n->value = v;
+	n->next = mips_hi16_list;
+	mips_hi16_list = n;
+
+	return 0;
+}
+
+static int apply_r_mips_lo16(struct module *me, uint32_t *location,
+			     Elf32_Addr v)
+{
+	unsigned long insnlo = *location;
+	Elf32_Addr val, vallo;
+	struct mips_hi16 *l, *next;
+
+	/* Sign extend the addend we extract from the lo insn.	*/
+	vallo = ((insnlo & 0xffff) ^ 0x8000) - 0x8000;
+
+	if (mips_hi16_list != NULL) {
+
+		l = mips_hi16_list;
+		while (l != NULL) {
+			unsigned long insn;
+
+			/*
+			 * The value for the HI16 had best be the same.
+			 */
+			if (v != l->value) {
+				pr_debug("VPE loader: apply_r_mips_lo16/hi16: inconsistent value information\n");
+				goto out_free;
+			}
+
+			/*
+			 * Do the HI16 relocation.  Note that we actually don't
+			 * need to know anything about the LO16 itself, except
+			 * where to find the low 16 bits of the addend needed
+			 * by the LO16.
+			 */
+			insn = *l->addr;
+			val = ((insn & 0xffff) << 16) + vallo;
+			val += v;
+
+			/*
+			 * Account for the sign extension that will happen in
+			 * the low bits.
+			 */
+			val = ((val >> 16) + ((val & 0x8000) != 0)) & 0xffff;
+
+			insn = (insn & ~0xffff) | val;
+			*l->addr = insn;
+
+			next = l->next;
+			kfree(l);
+			l = next;
+		}
+
+		mips_hi16_list = NULL;
+	}
+
+	/*
+	 * Ok, we're done with the HI16 relocs.	 Now deal with the LO16.
+	 */
+	val = v + vallo;
+	insnlo = (insnlo & ~0xffff) | (val & 0xffff);
+	*location = insnlo;
+
+	return 0;
+
+out_free:
+	while (l != NULL) {
+		next = l->next;
+		kfree(l);
+		l = next;
+	}
+	mips_hi16_list = NULL;
+
+	return -ENOEXEC;
+}
+
+static int (*reloc_handlers[]) (struct module *me, uint32_t *location,
+				Elf32_Addr v) = {
+	[R_MIPS_NONE]	= apply_r_mips_none,
+	[R_MIPS_32]	= apply_r_mips_32,
+	[R_MIPS_26]	= apply_r_mips_26,
+	[R_MIPS_HI16]	= apply_r_mips_hi16,
+	[R_MIPS_LO16]	= apply_r_mips_lo16,
+	[R_MIPS_GPREL16] = apply_r_mips_gprel16,
+	[R_MIPS_PC16] = apply_r_mips_pc16
+};
+
+static char *rstrs[] = {
+	[R_MIPS_NONE]	= "MIPS_NONE",
+	[R_MIPS_32]	= "MIPS_32",
+	[R_MIPS_26]	= "MIPS_26",
+	[R_MIPS_HI16]	= "MIPS_HI16",
+	[R_MIPS_LO16]	= "MIPS_LO16",
+	[R_MIPS_GPREL16] = "MIPS_GPREL16",
+	[R_MIPS_PC16] = "MIPS_PC16"
+};
+
+static int apply_relocations(Elf32_Shdr *sechdrs,
+		      const char *strtab,
+		      unsigned int symindex,
+		      unsigned int relsec,
+		      struct module *me)
+{
+	Elf32_Rel *rel = (void *) sechdrs[relsec].sh_addr;
+	Elf32_Sym *sym;
+	uint32_t *location;
+	unsigned int i;
+	Elf32_Addr v;
+	int res;
+
+	for (i = 0; i < sechdrs[relsec].sh_size / sizeof(*rel); i++) {
+		Elf32_Word r_info = rel[i].r_info;
+
+		/* This is where to make the change */
+		location = (void *)sechdrs[sechdrs[relsec].sh_info].sh_addr
+			+ rel[i].r_offset;
+		/* This is the symbol it is referring to */
+		sym = (Elf32_Sym *)sechdrs[symindex].sh_addr
+			+ ELF32_R_SYM(r_info);
+
+		if (!sym->st_value) {
+			pr_debug("%s: undefined weak symbol %s\n",
+				 me->name, strtab + sym->st_name);
+			/* just print the warning, dont barf */
+		}
+
+		v = sym->st_value;
+
+		res = reloc_handlers[ELF32_R_TYPE(r_info)](me, location, v);
+		if (res) {
+			char *r = rstrs[ELF32_R_TYPE(r_info)];
+			pr_warn("VPE loader: .text+0x%x relocation type %s for symbol \"%s\" failed\n",
+				rel[i].r_offset, r ? r : "UNKNOWN",
+				strtab + sym->st_name);
+			return res;
+		}
+	}
+
+	return 0;
+}
+
+static inline void save_gp_address(unsigned int secbase, unsigned int rel)
+{
+	gp_addr = secbase + rel;
+	gp_offs = gp_addr - (secbase & 0xffff0000);
+}
+/* end module-elf32.c */
+
+/* Change all symbols so that sh_value encodes the pointer directly. */
+static void simplify_symbols(Elf_Shdr *sechdrs,
+			    unsigned int symindex,
+			    const char *strtab,
+			    const char *secstrings,
+			    unsigned int nsecs, struct module *mod)
+{
+	Elf_Sym *sym = (void *)sechdrs[symindex].sh_addr;
+	unsigned long secbase, bssbase = 0;
+	unsigned int i, n = sechdrs[symindex].sh_size / sizeof(Elf_Sym);
+	int size;
+
+	/* find the .bss section for COMMON symbols */
+	for (i = 0; i < nsecs; i++) {
+		if (strncmp(secstrings + sechdrs[i].sh_name, ".bss", 4) == 0) {
+			bssbase = sechdrs[i].sh_addr;
+			break;
+		}
+	}
+
+	for (i = 1; i < n; i++) {
+		switch (sym[i].st_shndx) {
+		case SHN_COMMON:
+			/* Allocate space for the symbol in the .bss section.
+			   st_value is currently size.
+			   We want it to have the address of the symbol. */
+
+			size = sym[i].st_value;
+			sym[i].st_value = bssbase;
+
+			bssbase += size;
+			break;
+
+		case SHN_ABS:
+			/* Don't need to do anything */
+			break;
+
+		case SHN_UNDEF:
+			/* ret = -ENOENT; */
+			break;
+
+		case SHN_MIPS_SCOMMON:
+			pr_debug("simplify_symbols: ignoring SHN_MIPS_SCOMMON symbol <%s> st_shndx %d\n",
+				 strtab + sym[i].st_name, sym[i].st_shndx);
+			/* .sbss section */
+			break;
+
+		default:
+			secbase = sechdrs[sym[i].st_shndx].sh_addr;
+
+			if (strncmp(strtab + sym[i].st_name, "_gp", 3) == 0)
+				save_gp_address(secbase, sym[i].st_value);
+
+			sym[i].st_value += secbase;
+			break;
+		}
+	}
+}
+
+#ifdef DEBUG_ELFLOADER
+static void dump_elfsymbols(Elf_Shdr *sechdrs, unsigned int symindex,
+			    const char *strtab, struct module *mod)
+{
+	Elf_Sym *sym = (void *)sechdrs[symindex].sh_addr;
+	unsigned int i, n = sechdrs[symindex].sh_size / sizeof(Elf_Sym);
+
+	pr_debug("dump_elfsymbols: n %d\n", n);
+	for (i = 1; i < n; i++) {
+		pr_debug(" i %d name <%s> 0x%x\n", i, strtab + sym[i].st_name,
+			 sym[i].st_value);
+	}
+}
+#endif
+
+static int find_vpe_symbols(struct vpe *v, Elf_Shdr *sechdrs,
+				      unsigned int symindex, const char *strtab,
+				      struct module *mod)
+{
+	Elf_Sym *sym = (void *)sechdrs[symindex].sh_addr;
+	unsigned int i, n = sechdrs[symindex].sh_size / sizeof(Elf_Sym);
+
+	for (i = 1; i < n; i++) {
+		if (strcmp(strtab + sym[i].st_name, "__start") == 0)
+			v->__start = sym[i].st_value;
+
+		if (strcmp(strtab + sym[i].st_name, "vpe_shared") == 0)
+			v->shared_ptr = (void *)sym[i].st_value;
+	}
+
+	if ((v->__start == 0) || (v->shared_ptr == NULL))
+		return -1;
+
+	return 0;
+}
+
+/*
+ * Allocates a VPE with some program code space(the load address), copies the
+ * contents of the program (p)buffer performing relocatations/etc, free's it
+ * when finished.
+ */
+static int vpe_elfload(struct vpe *v)
+{
+	Elf_Ehdr *hdr;
+	Elf_Shdr *sechdrs;
+	long err = 0;
+	char *secstrings, *strtab = NULL;
+	unsigned int len, i, symindex = 0, strindex = 0, relocate = 0;
+	struct module mod; /* so we can re-use the relocations code */
+
+	memset(&mod, 0, sizeof(struct module));
+	strcpy(mod.name, "VPE loader");
+
+	hdr = (Elf_Ehdr *) v->pbuffer;
+	len = v->plen;
+
+	/* Sanity checks against insmoding binaries or wrong arch,
+	   weird elf version */
+	if (memcmp(hdr->e_ident, ELFMAG, SELFMAG) != 0
+	    || (hdr->e_type != ET_REL && hdr->e_type != ET_EXEC)
+	    || !elf_check_arch(hdr)
+	    || hdr->e_shentsize != sizeof(*sechdrs)) {
+		pr_warn("VPE loader: program wrong arch or weird elf version\n");
+
+		return -ENOEXEC;
+	}
+
+	if (hdr->e_type == ET_REL)
+		relocate = 1;
+
+	if (len < hdr->e_shoff + hdr->e_shnum * sizeof(Elf_Shdr)) {
+		pr_err("VPE loader: program length %u truncated\n", len);
+
+		return -ENOEXEC;
+	}
+
+	/* Convenience variables */
+	sechdrs = (void *)hdr + hdr->e_shoff;
+	secstrings = (void *)hdr + sechdrs[hdr->e_shstrndx].sh_offset;
+	sechdrs[0].sh_addr = 0;
+
+	/* And these should exist, but gcc whinges if we don't init them */
+	symindex = strindex = 0;
+
+	if (relocate) {
+		for (i = 1; i < hdr->e_shnum; i++) {
+			if ((sechdrs[i].sh_type != SHT_NOBITS) &&
+			    (len < sechdrs[i].sh_offset + sechdrs[i].sh_size)) {
+				pr_err("VPE program length %u truncated\n",
+				       len);
+				return -ENOEXEC;
+			}
+
+			/* Mark all sections sh_addr with their address in the
+			   temporary image. */
+			sechdrs[i].sh_addr = (size_t) hdr +
+				sechdrs[i].sh_offset;
+
+			/* Internal symbols and strings. */
+			if (sechdrs[i].sh_type == SHT_SYMTAB) {
+				symindex = i;
+				strindex = sechdrs[i].sh_link;
+				strtab = (char *)hdr +
+					sechdrs[strindex].sh_offset;
+			}
+		}
+		layout_sections(&mod, hdr, sechdrs, secstrings);
+	}
+
+	v->load_addr = alloc_progmem(mod.core_layout.size);
+	if (!v->load_addr)
+		return -ENOMEM;
+
+	pr_info("VPE loader: loading to %p\n", v->load_addr);
+
+	if (relocate) {
+		for (i = 0; i < hdr->e_shnum; i++) {
+			void *dest;
+
+			if (!(sechdrs[i].sh_flags & SHF_ALLOC))
+				continue;
+
+			dest = v->load_addr + sechdrs[i].sh_entsize;
+
+			if (sechdrs[i].sh_type != SHT_NOBITS)
+				memcpy(dest, (void *)sechdrs[i].sh_addr,
+				       sechdrs[i].sh_size);
+			/* Update sh_addr to point to copy in image. */
+			sechdrs[i].sh_addr = (unsigned long)dest;
+
+			pr_debug(" section sh_name %s sh_addr 0x%x\n",
+				 secstrings + sechdrs[i].sh_name,
+				 sechdrs[i].sh_addr);
+		}
+
+		/* Fix up syms, so that st_value is a pointer to location. */
+		simplify_symbols(sechdrs, symindex, strtab, secstrings,
+				 hdr->e_shnum, &mod);
+
+		/* Now do relocations. */
+		for (i = 1; i < hdr->e_shnum; i++) {
+			const char *strtab = (char *)sechdrs[strindex].sh_addr;
+			unsigned int info = sechdrs[i].sh_info;
+
+			/* Not a valid relocation section? */
+			if (info >= hdr->e_shnum)
+				continue;
+
+			/* Don't bother with non-allocated sections */
+			if (!(sechdrs[info].sh_flags & SHF_ALLOC))
+				continue;
+
+			if (sechdrs[i].sh_type == SHT_REL)
+				err = apply_relocations(sechdrs, strtab,
+							symindex, i, &mod);
+			else if (sechdrs[i].sh_type == SHT_RELA)
+				err = apply_relocate_add(sechdrs, strtab,
+							 symindex, i, &mod);
+			if (err < 0)
+				return err;
+
+		}
+	} else {
+		struct elf_phdr *phdr = (struct elf_phdr *)
+						((char *)hdr + hdr->e_phoff);
+
+		for (i = 0; i < hdr->e_phnum; i++) {
+			if (phdr->p_type == PT_LOAD) {
+				memcpy((void *)phdr->p_paddr,
+				       (char *)hdr + phdr->p_offset,
+				       phdr->p_filesz);
+				memset((void *)phdr->p_paddr + phdr->p_filesz,
+				       0, phdr->p_memsz - phdr->p_filesz);
+		    }
+		    phdr++;
+		}
+
+		for (i = 0; i < hdr->e_shnum; i++) {
+			/* Internal symbols and strings. */
+			if (sechdrs[i].sh_type == SHT_SYMTAB) {
+				symindex = i;
+				strindex = sechdrs[i].sh_link;
+				strtab = (char *)hdr +
+					sechdrs[strindex].sh_offset;
+
+				/*
+				 * mark symtab's address for when we try
+				 * to find the magic symbols
+				 */
+				sechdrs[i].sh_addr = (size_t) hdr +
+					sechdrs[i].sh_offset;
+			}
+		}
+	}
+
+	/* make sure it's physically written out */
+	flush_icache_range((unsigned long)v->load_addr,
+			   (unsigned long)v->load_addr + v->len);
+
+	if ((find_vpe_symbols(v, sechdrs, symindex, strtab, &mod)) < 0) {
+		if (v->__start == 0) {
+			pr_warn("VPE loader: program does not contain a __start symbol\n");
+			return -ENOEXEC;
+		}
+
+		if (v->shared_ptr == NULL)
+			pr_warn("VPE loader: program does not contain vpe_shared symbol.\n"
+				" Unable to use AMVP (AP/SP) facilities.\n");
+	}
+
+	pr_info(" elf loaded\n");
+	return 0;
+}
+
+static int getcwd(char *buff, int size)
+{
+	mm_segment_t old_fs;
+	int ret;
+
+	old_fs = get_fs();
+	set_fs(KERNEL_DS);
+
+	ret = sys_getcwd(buff, size);
+
+	set_fs(old_fs);
+
+	return ret;
+}
+
+/* checks VPE is unused and gets ready to load program	*/
+static int vpe_open(struct inode *inode, struct file *filp)
+{
+	enum vpe_state state;
+	struct vpe_notifications *notifier;
+	struct vpe *v;
+	int ret;
+
+	if (VPE_MODULE_MINOR != iminor(inode)) {
+		/* assume only 1 device at the moment. */
+		pr_warn("VPE loader: only vpe1 is supported\n");
+
+		return -ENODEV;
+	}
+
+	v = get_vpe(aprp_cpu_index());
+	if (v == NULL) {
+		pr_warn("VPE loader: unable to get vpe\n");
+
+		return -ENODEV;
+	}
+
+	state = xchg(&v->state, VPE_STATE_INUSE);
+	if (state != VPE_STATE_UNUSED) {
+		pr_debug("VPE loader: tc in use dumping regs\n");
+
+		list_for_each_entry(notifier, &v->notify, list)
+			notifier->stop(aprp_cpu_index());
+
+		release_progmem(v->load_addr);
+		cleanup_tc(get_tc(aprp_cpu_index()));
+	}
+
+	/* this of-course trashes what was there before... */
+	v->pbuffer = vmalloc(P_SIZE);
+	if (!v->pbuffer) {
+		pr_warn("VPE loader: unable to allocate memory\n");
+		return -ENOMEM;
+	}
+	v->plen = P_SIZE;
+	v->load_addr = NULL;
+	v->len = 0;
+
+	v->cwd[0] = 0;
+	ret = getcwd(v->cwd, VPE_PATH_MAX);
+	if (ret < 0)
+		pr_warn("VPE loader: open, getcwd returned %d\n", ret);
+
+	v->shared_ptr = NULL;
+	v->__start = 0;
+
+	return 0;
+}
+
+static int vpe_release(struct inode *inode, struct file *filp)
+{
+#if defined(CONFIG_MIPS_VPE_LOADER_MT) || defined(CONFIG_MIPS_VPE_LOADER_CMP)
+	struct vpe *v;
+	Elf_Ehdr *hdr;
+	int ret = 0;
+
+	v = get_vpe(aprp_cpu_index());
+	if (v == NULL)
+		return -ENODEV;
+
+	hdr = (Elf_Ehdr *) v->pbuffer;
+	if (memcmp(hdr->e_ident, ELFMAG, SELFMAG) == 0) {
+		if (vpe_elfload(v) >= 0) {
+			vpe_run(v);
+		} else {
+			pr_warn("VPE loader: ELF load failed.\n");
+			ret = -ENOEXEC;
+		}
+	} else {
+		pr_warn("VPE loader: only elf files are supported\n");
+		ret = -ENOEXEC;
+	}
+
+	/* It's good to be able to run the SP and if it chokes have a look at
+	   the /dev/rt?. But if we reset the pointer to the shared struct we
+	   lose what has happened. So perhaps if garbage is sent to the vpe
+	   device, use it as a trigger for the reset. Hopefully a nice
+	   executable will be along shortly. */
+	if (ret < 0)
+		v->shared_ptr = NULL;
+
+	vfree(v->pbuffer);
+	v->plen = 0;
+
+	return ret;
+#else
+	pr_warn("VPE loader: ELF load failed.\n");
+	return -ENOEXEC;
+#endif
+}
+
+static ssize_t vpe_write(struct file *file, const char __user *buffer,
+			 size_t count, loff_t *ppos)
+{
+	size_t ret = count;
+	struct vpe *v;
+
+	if (iminor(file_inode(file)) != VPE_MODULE_MINOR)
+		return -ENODEV;
+
+	v = get_vpe(aprp_cpu_index());
+
+	if (v == NULL)
+		return -ENODEV;
+
+	if ((count + v->len) > v->plen) {
+		pr_warn("VPE loader: elf size too big. Perhaps strip unneeded symbols\n");
+		return -ENOMEM;
+	}
+
+	count -= copy_from_user(v->pbuffer + v->len, buffer, count);
+	if (!count)
+		return -EFAULT;
+
+	v->len += count;
+	return ret;
+}
+
+const struct file_operations vpe_fops = {
+	.owner = THIS_MODULE,
+	.open = vpe_open,
+	.release = vpe_release,
+	.write = vpe_write,
+	.llseek = noop_llseek,
+};
+
+void *vpe_get_shared(int index)
+{
+	struct vpe *v = get_vpe(index);
+
+	if (v == NULL)
+		return NULL;
+
+	return v->shared_ptr;
+}
+EXPORT_SYMBOL(vpe_get_shared);
+
+int vpe_notify(int index, struct vpe_notifications *notify)
+{
+	struct vpe *v = get_vpe(index);
+
+	if (v == NULL)
+		return -1;
+
+	list_add(&notify->list, &v->notify);
+	return 0;
+}
+EXPORT_SYMBOL(vpe_notify);
+
+char *vpe_getcwd(int index)
+{
+	struct vpe *v = get_vpe(index);
+
+	if (v == NULL)
+		return NULL;
+
+	return v->cwd;
+}
+EXPORT_SYMBOL(vpe_getcwd);
+
+module_init(vpe_module_init);
+module_exit(vpe_module_exit);
+MODULE_DESCRIPTION("MIPS VPE Loader");
+MODULE_AUTHOR("Elizabeth Oldham, MIPS Technologies, Inc.");
+MODULE_LICENSE("GPL");
diff --git a/arch/mips/kernel/watch.c b/arch/mips/kernel/watch.c
new file mode 100644
index 000000000..c9263b95c
--- /dev/null
+++ b/arch/mips/kernel/watch.c
@@ -0,0 +1,211 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2008 David Daney
+ */
+
+#include <linux/sched.h>
+
+#include <asm/processor.h>
+#include <asm/watch.h>
+
+/*
+ * Install the watch registers for the current thread.	A maximum of
+ * four registers are installed although the machine may have more.
+ */
+void mips_install_watch_registers(struct task_struct *t)
+{
+	struct mips3264_watch_reg_state *watches = &t->thread.watch.mips3264;
+	unsigned int watchhi = MIPS_WATCHHI_G |		/* Trap all ASIDs */
+			       MIPS_WATCHHI_IRW;	/* Clear result bits */
+
+	switch (current_cpu_data.watch_reg_use_cnt) {
+	default:
+		BUG();
+	case 4:
+		write_c0_watchlo3(watches->watchlo[3]);
+		write_c0_watchhi3(watchhi | watches->watchhi[3]);
+		fallthrough;
+	case 3:
+		write_c0_watchlo2(watches->watchlo[2]);
+		write_c0_watchhi2(watchhi | watches->watchhi[2]);
+		fallthrough;
+	case 2:
+		write_c0_watchlo1(watches->watchlo[1]);
+		write_c0_watchhi1(watchhi | watches->watchhi[1]);
+		fallthrough;
+	case 1:
+		write_c0_watchlo0(watches->watchlo[0]);
+		write_c0_watchhi0(watchhi | watches->watchhi[0]);
+	}
+}
+
+/*
+ * Read back the watchhi registers so the user space debugger has
+ * access to the I, R, and W bits.  A maximum of four registers are
+ * read although the machine may have more.
+ */
+void mips_read_watch_registers(void)
+{
+	struct mips3264_watch_reg_state *watches =
+		&current->thread.watch.mips3264;
+	unsigned int watchhi_mask = MIPS_WATCHHI_MASK | MIPS_WATCHHI_IRW;
+
+	switch (current_cpu_data.watch_reg_use_cnt) {
+	default:
+		BUG();
+	case 4:
+		watches->watchhi[3] = (read_c0_watchhi3() & watchhi_mask);
+		fallthrough;
+	case 3:
+		watches->watchhi[2] = (read_c0_watchhi2() & watchhi_mask);
+		fallthrough;
+	case 2:
+		watches->watchhi[1] = (read_c0_watchhi1() & watchhi_mask);
+		fallthrough;
+	case 1:
+		watches->watchhi[0] = (read_c0_watchhi0() & watchhi_mask);
+	}
+	if (current_cpu_data.watch_reg_use_cnt == 1 &&
+	    (watches->watchhi[0] & MIPS_WATCHHI_IRW) == 0) {
+		/* Pathological case of release 1 architecture that
+		 * doesn't set the condition bits.  We assume that
+		 * since we got here, the watch condition was met and
+		 * signal that the conditions requested in watchlo
+		 * were met.  */
+		watches->watchhi[0] |= (watches->watchlo[0] & MIPS_WATCHHI_IRW);
+	}
+ }
+
+/*
+ * Disable all watch registers.	 Although only four registers are
+ * installed, all are cleared to eliminate the possibility of endless
+ * looping in the watch handler.
+ */
+void mips_clear_watch_registers(void)
+{
+	switch (current_cpu_data.watch_reg_count) {
+	default:
+		BUG();
+	case 8:
+		write_c0_watchlo7(0);
+		fallthrough;
+	case 7:
+		write_c0_watchlo6(0);
+		fallthrough;
+	case 6:
+		write_c0_watchlo5(0);
+		fallthrough;
+	case 5:
+		write_c0_watchlo4(0);
+		fallthrough;
+	case 4:
+		write_c0_watchlo3(0);
+		fallthrough;
+	case 3:
+		write_c0_watchlo2(0);
+		fallthrough;
+	case 2:
+		write_c0_watchlo1(0);
+		fallthrough;
+	case 1:
+		write_c0_watchlo0(0);
+	}
+}
+
+void mips_probe_watch_registers(struct cpuinfo_mips *c)
+{
+	unsigned int t;
+
+	if ((c->options & MIPS_CPU_WATCH) == 0)
+		return;
+	/*
+	 * Check which of the I,R and W bits are supported, then
+	 * disable the register.
+	 */
+	write_c0_watchlo0(MIPS_WATCHLO_IRW);
+	back_to_back_c0_hazard();
+	t = read_c0_watchlo0();
+	write_c0_watchlo0(0);
+	c->watch_reg_masks[0] = t & MIPS_WATCHLO_IRW;
+
+	/* Write the mask bits and read them back to determine which
+	 * can be used. */
+	c->watch_reg_count = 1;
+	c->watch_reg_use_cnt = 1;
+	t = read_c0_watchhi0();
+	write_c0_watchhi0(t | MIPS_WATCHHI_MASK);
+	back_to_back_c0_hazard();
+	t = read_c0_watchhi0();
+	c->watch_reg_masks[0] |= (t & MIPS_WATCHHI_MASK);
+	if ((t & MIPS_WATCHHI_M) == 0)
+		return;
+
+	write_c0_watchlo1(MIPS_WATCHLO_IRW);
+	back_to_back_c0_hazard();
+	t = read_c0_watchlo1();
+	write_c0_watchlo1(0);
+	c->watch_reg_masks[1] = t & MIPS_WATCHLO_IRW;
+
+	c->watch_reg_count = 2;
+	c->watch_reg_use_cnt = 2;
+	t = read_c0_watchhi1();
+	write_c0_watchhi1(t | MIPS_WATCHHI_MASK);
+	back_to_back_c0_hazard();
+	t = read_c0_watchhi1();
+	c->watch_reg_masks[1] |= (t & MIPS_WATCHHI_MASK);
+	if ((t & MIPS_WATCHHI_M) == 0)
+		return;
+
+	write_c0_watchlo2(MIPS_WATCHLO_IRW);
+	back_to_back_c0_hazard();
+	t = read_c0_watchlo2();
+	write_c0_watchlo2(0);
+	c->watch_reg_masks[2] = t & MIPS_WATCHLO_IRW;
+
+	c->watch_reg_count = 3;
+	c->watch_reg_use_cnt = 3;
+	t = read_c0_watchhi2();
+	write_c0_watchhi2(t | MIPS_WATCHHI_MASK);
+	back_to_back_c0_hazard();
+	t = read_c0_watchhi2();
+	c->watch_reg_masks[2] |= (t & MIPS_WATCHHI_MASK);
+	if ((t & MIPS_WATCHHI_M) == 0)
+		return;
+
+	write_c0_watchlo3(MIPS_WATCHLO_IRW);
+	back_to_back_c0_hazard();
+	t = read_c0_watchlo3();
+	write_c0_watchlo3(0);
+	c->watch_reg_masks[3] = t & MIPS_WATCHLO_IRW;
+
+	c->watch_reg_count = 4;
+	c->watch_reg_use_cnt = 4;
+	t = read_c0_watchhi3();
+	write_c0_watchhi3(t | MIPS_WATCHHI_MASK);
+	back_to_back_c0_hazard();
+	t = read_c0_watchhi3();
+	c->watch_reg_masks[3] |= (t & MIPS_WATCHHI_MASK);
+	if ((t & MIPS_WATCHHI_M) == 0)
+		return;
+
+	/* We use at most 4, but probe and report up to 8. */
+	c->watch_reg_count = 5;
+	t = read_c0_watchhi4();
+	if ((t & MIPS_WATCHHI_M) == 0)
+		return;
+
+	c->watch_reg_count = 6;
+	t = read_c0_watchhi5();
+	if ((t & MIPS_WATCHHI_M) == 0)
+		return;
+
+	c->watch_reg_count = 7;
+	t = read_c0_watchhi6();
+	if ((t & MIPS_WATCHHI_M) == 0)
+		return;
+
+	c->watch_reg_count = 8;
+}