diff options
Diffstat (limited to 'arch/mips/kernel/r4k_fpu.S')
| -rw-r--r-- | arch/mips/kernel/r4k_fpu.S | 561 |
1 files changed, 561 insertions, 0 deletions
diff --git a/arch/mips/kernel/r4k_fpu.S b/arch/mips/kernel/r4k_fpu.S new file mode 100644 index 000000000..8e3a6020c --- /dev/null +++ b/arch/mips/kernel/r4k_fpu.S @@ -0,0 +1,561 @@ +/* + * 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_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_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 + +/* + * Load the FPU with signalling NANS. This bit pattern we're using has + * the property that no matter whether considered as single or as double + * precision represents signaling NANS. + * + * The value to initialize fcr31 to comes in $a0. + */ + + .set push + SET_HARDFLOAT + +LEAF(_init_fpu) + mfc0 t0, CP0_STATUS + li t1, ST0_CU1 + or t0, t1 + mtc0 t0, CP0_STATUS + enable_fpu_hazard + + ctc1 a0, fcr31 + + li t1, -1 # SNaN + +#ifdef CONFIG_64BIT + sll t0, t0, 5 + bgez t0, 1f # 16 / 32 register mode? + + dmtc1 t1, $f1 + dmtc1 t1, $f3 + dmtc1 t1, $f5 + dmtc1 t1, $f7 + dmtc1 t1, $f9 + dmtc1 t1, $f11 + dmtc1 t1, $f13 + dmtc1 t1, $f15 + dmtc1 t1, $f17 + dmtc1 t1, $f19 + dmtc1 t1, $f21 + dmtc1 t1, $f23 + dmtc1 t1, $f25 + dmtc1 t1, $f27 + dmtc1 t1, $f29 + dmtc1 t1, $f31 +1: +#endif + +#ifdef CONFIG_CPU_MIPS32 + mtc1 t1, $f0 + mtc1 t1, $f1 + mtc1 t1, $f2 + mtc1 t1, $f3 + mtc1 t1, $f4 + mtc1 t1, $f5 + mtc1 t1, $f6 + mtc1 t1, $f7 + mtc1 t1, $f8 + mtc1 t1, $f9 + mtc1 t1, $f10 + mtc1 t1, $f11 + mtc1 t1, $f12 + mtc1 t1, $f13 + mtc1 t1, $f14 + mtc1 t1, $f15 + mtc1 t1, $f16 + mtc1 t1, $f17 + mtc1 t1, $f18 + mtc1 t1, $f19 + mtc1 t1, $f20 + mtc1 t1, $f21 + mtc1 t1, $f22 + mtc1 t1, $f23 + mtc1 t1, $f24 + mtc1 t1, $f25 + mtc1 t1, $f26 + mtc1 t1, $f27 + mtc1 t1, $f28 + mtc1 t1, $f29 + mtc1 t1, $f30 + mtc1 t1, $f31 + +#if defined(CONFIG_CPU_MIPS32_R2) || defined(CONFIG_CPU_MIPS32_R6) + .set push + .set MIPS_ISA_LEVEL_RAW + .set fp=64 + sll t0, t0, 5 # is Status.FR set? + bgez t0, 1f # no: skip setting upper 32b + + mthc1 t1, $f0 + mthc1 t1, $f1 + mthc1 t1, $f2 + mthc1 t1, $f3 + mthc1 t1, $f4 + mthc1 t1, $f5 + mthc1 t1, $f6 + mthc1 t1, $f7 + mthc1 t1, $f8 + mthc1 t1, $f9 + mthc1 t1, $f10 + mthc1 t1, $f11 + mthc1 t1, $f12 + mthc1 t1, $f13 + mthc1 t1, $f14 + mthc1 t1, $f15 + mthc1 t1, $f16 + mthc1 t1, $f17 + mthc1 t1, $f18 + mthc1 t1, $f19 + mthc1 t1, $f20 + mthc1 t1, $f21 + mthc1 t1, $f22 + mthc1 t1, $f23 + mthc1 t1, $f24 + mthc1 t1, $f25 + mthc1 t1, $f26 + mthc1 t1, $f27 + mthc1 t1, $f28 + mthc1 t1, $f29 + mthc1 t1, $f30 + mthc1 t1, $f31 +1: .set pop +#endif /* CONFIG_CPU_MIPS32_R2 || CONFIG_CPU_MIPS32_R6 */ +#else + .set MIPS_ISA_ARCH_LEVEL_RAW + dmtc1 t1, $f0 + dmtc1 t1, $f2 + dmtc1 t1, $f4 + dmtc1 t1, $f6 + dmtc1 t1, $f8 + dmtc1 t1, $f10 + dmtc1 t1, $f12 + dmtc1 t1, $f14 + dmtc1 t1, $f16 + dmtc1 t1, $f18 + dmtc1 t1, $f20 + dmtc1 t1, $f22 + dmtc1 t1, $f24 + dmtc1 t1, $f26 + dmtc1 t1, $f28 + dmtc1 t1, $f30 +#endif + jr ra + END(_init_fpu) + + .set pop /* SET_HARDFLOAT */ + + .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_MIPSR6) + .set push + SET_HARDFLOAT +#ifdef CONFIG_CPU_MIPSR2 + .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_MIPSR6) + .set push + SET_HARDFLOAT +#ifdef CONFIG_CPU_MIPSR2 + .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 |