Adding upstream version 5.10.209.upstream/5.10.209

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

1 files changed, 965 insertions, 0 deletions

diff --git a/arch/parisc/kernel/syscall.S b/arch/parisc/kernel/syscall.S
new file mode 100644
index 000000000..4e53515cf
--- /dev/null
+++ b/arch/parisc/kernel/syscall.S
@@ -0,0 +1,965 @@
+/* 
+ * Linux/PA-RISC Project (http://www.parisc-linux.org/)
+ * 
+ * System call entry code / Linux gateway page
+ * Copyright (c) Matthew Wilcox 1999 <willy@infradead.org>
+ * Licensed under the GNU GPL.
+ * thanks to Philipp Rumpf, Mike Shaver and various others
+ * sorry about the wall, puffin..
+ */
+
+/*
+How does the Linux gateway page on PA-RISC work?
+------------------------------------------------
+The Linux gateway page on PA-RISC is "special".
+It actually has PAGE_GATEWAY bits set (this is linux terminology; in parisc
+terminology it's Execute, promote to PL0) in the page map.  So anything
+executing on this page executes with kernel level privilege (there's more to it
+than that: to have this happen, you also have to use a branch with a ,gate
+completer to activate the privilege promotion).  The upshot is that everything
+that runs on the gateway page runs at kernel privilege but with the current
+user process address space (although you have access to kernel space via %sr2).
+For the 0x100 syscall entry, we redo the space registers to point to the kernel
+address space (preserving the user address space in %sr3), move to wide mode if
+required, save the user registers and branch into the kernel syscall entry
+point.  For all the other functions, we execute at kernel privilege but don't
+flip address spaces. The basic upshot of this is that these code snippets are
+executed atomically (because the kernel can't be pre-empted) and they may
+perform architecturally forbidden (to PL3) operations (like setting control
+registers).
+*/
+
+
+#include <asm/asm-offsets.h>
+#include <asm/unistd.h>
+#include <asm/errno.h>
+#include <asm/page.h>
+#include <asm/psw.h>
+#include <asm/thread_info.h>
+#include <asm/assembly.h>
+#include <asm/processor.h>
+#include <asm/cache.h>
+
+#include <linux/linkage.h>
+
+	/* We fill the empty parts of the gateway page with
+ 	 * something that will kill the kernel or a
+ 	 * userspace application.
+	 */
+#define KILL_INSN	break	0,0
+
+	.level          PA_ASM_LEVEL
+
+	.text
+
+	.import syscall_exit,code
+	.import syscall_exit_rfi,code
+
+	/* Linux gateway page is aliased to virtual page 0 in the kernel
+	 * address space. Since it is a gateway page it cannot be
+	 * dereferenced, so null pointers will still fault. We start
+	 * the actual entry point at 0x100. We put break instructions
+	 * at the beginning of the page to trap null indirect function
+	 * pointers.
+	 */
+
+	.align PAGE_SIZE
+ENTRY(linux_gateway_page)
+
+        /* ADDRESS 0x00 to 0xb0 = 176 bytes / 4 bytes per insn = 44 insns */
+	.rept 44
+	KILL_INSN
+	.endr
+
+	/* ADDRESS 0xb0 to 0xb8, lws uses two insns for entry */
+	/* Light-weight-syscall entry must always be located at 0xb0 */
+	/* WARNING: Keep this number updated with table size changes */
+#define __NR_lws_entries (3)
+
+lws_entry:
+	gate	lws_start, %r0		/* increase privilege */
+	depi	3, 31, 2, %r31		/* Ensure we return into user mode. */
+
+	/* Fill from 0xb8 to 0xe0 */
+	.rept 10
+	KILL_INSN
+	.endr
+
+	/* This function MUST be located at 0xe0 for glibc's threading 
+	mechanism to work. DO NOT MOVE THIS CODE EVER! */
+set_thread_pointer:
+	gate	.+8, %r0		/* increase privilege */
+	depi	3, 31, 2, %r31		/* Ensure we return into user mode. */
+	be	0(%sr7,%r31)		/* return to user space */
+	mtctl	%r26, %cr27		/* move arg0 to the control register */
+
+	/* Increase the chance of trapping if random jumps occur to this
+	address, fill from 0xf0 to 0x100 */
+	.rept 4
+	KILL_INSN
+	.endr
+
+/* This address must remain fixed at 0x100 for glibc's syscalls to work */
+	.align LINUX_GATEWAY_ADDR
+linux_gateway_entry:
+	gate	.+8, %r0			/* become privileged */
+	mtsp	%r0,%sr4			/* get kernel space into sr4 */
+	mtsp	%r0,%sr5			/* get kernel space into sr5 */
+	mtsp	%r0,%sr6			/* get kernel space into sr6 */
+
+#ifdef CONFIG_64BIT
+	/* Store W bit on entry to the syscall in case it's a wide userland
+	 * process. */
+	ssm	PSW_SM_W, %r1
+	extrd,u	%r1,PSW_W_BIT,1,%r1
+	/* sp must be aligned on 4, so deposit the W bit setting into
+	 * the bottom of sp temporarily */
+	or,ev	%r1,%r30,%r30
+	b,n	1f
+	/* The top halves of argument registers must be cleared on syscall
+	 * entry from narrow executable.
+	 */
+	depdi	0, 31, 32, %r26
+	depdi	0, 31, 32, %r25
+	depdi	0, 31, 32, %r24
+	depdi	0, 31, 32, %r23
+	depdi	0, 31, 32, %r22
+	depdi	0, 31, 32, %r21
+1:	
+#endif
+
+	/* We use a rsm/ssm pair to prevent sr3 from being clobbered
+	 * by external interrupts.
+	 */
+	mfsp    %sr7,%r1                        /* save user sr7 */
+	rsm	PSW_SM_I, %r0			/* disable interrupts */
+	mtsp    %r1,%sr3                        /* and store it in sr3 */
+
+	mfctl   %cr30,%r1
+	xor     %r1,%r30,%r30                   /* ye olde xor trick */
+	xor     %r1,%r30,%r1
+	xor     %r1,%r30,%r30
+	
+	ldo     THREAD_SZ_ALGN+FRAME_SIZE(%r30),%r30  /* set up kernel stack */
+
+	/* N.B.: It is critical that we don't set sr7 to 0 until r30
+	 *       contains a valid kernel stack pointer. It is also
+	 *       critical that we don't start using the kernel stack
+	 *       until after sr7 has been set to 0.
+	 */
+
+	mtsp	%r0,%sr7			/* get kernel space into sr7 */
+	ssm	PSW_SM_I, %r0			/* enable interrupts */
+	STREGM	%r1,FRAME_SIZE(%r30)		/* save r1 (usp) here for now */
+	mfctl	%cr30,%r1			/* get task ptr in %r1 */
+	LDREG	TI_TASK(%r1),%r1
+
+	/* Save some registers for sigcontext and potential task
+	   switch (see entry.S for the details of which ones are
+	   saved/restored).  TASK_PT_PSW is zeroed so we can see whether
+	   a process is on a syscall or not.  For an interrupt the real
+	   PSW value is stored.  This is needed for gdb and sys_ptrace. */
+	STREG	%r0,  TASK_PT_PSW(%r1)
+	STREG	%r2,  TASK_PT_GR2(%r1)		/* preserve rp */
+	STREG	%r19, TASK_PT_GR19(%r1)
+
+	LDREGM	-FRAME_SIZE(%r30), %r2		/* get users sp back */
+#ifdef CONFIG_64BIT
+	extrd,u	%r2,63,1,%r19			/* W hidden in bottom bit */
+#if 0
+	xor	%r19,%r2,%r2			/* clear bottom bit */
+	depd,z	%r19,1,1,%r19
+	std	%r19,TASK_PT_PSW(%r1)
+#endif
+#endif
+	STREG	%r2,  TASK_PT_GR30(%r1)		/* ... and save it */
+	
+	STREG	%r20, TASK_PT_GR20(%r1)		/* Syscall number */
+	STREG	%r21, TASK_PT_GR21(%r1)
+	STREG	%r22, TASK_PT_GR22(%r1)
+	STREG	%r23, TASK_PT_GR23(%r1)		/* 4th argument */
+	STREG	%r24, TASK_PT_GR24(%r1)		/* 3rd argument */
+	STREG	%r25, TASK_PT_GR25(%r1)		/* 2nd argument */
+	STREG	%r26, TASK_PT_GR26(%r1)	 	/* 1st argument */
+	STREG	%r27, TASK_PT_GR27(%r1)		/* user dp */
+	STREG   %r28, TASK_PT_GR28(%r1)         /* return value 0 */
+	STREG   %r0, TASK_PT_ORIG_R28(%r1)      /* don't prohibit restarts */
+	STREG	%r29, TASK_PT_GR29(%r1)		/* return value 1 */
+	STREG	%r31, TASK_PT_GR31(%r1)		/* preserve syscall return ptr */
+	
+	ldo	TASK_PT_FR0(%r1), %r27		/* save fpregs from the kernel */
+	save_fp	%r27				/* or potential task switch  */
+
+	mfctl	%cr11, %r27			/* i.e. SAR */
+	STREG	%r27, TASK_PT_SAR(%r1)
+
+	loadgp
+
+#ifdef CONFIG_64BIT
+	ldo	-16(%r30),%r29			/* Reference param save area */
+	copy	%r19,%r2			/* W bit back to r2 */
+#else
+	/* no need to save these on stack in wide mode because the first 8
+	 * args are passed in registers */
+	stw     %r22, -52(%r30)                 /* 5th argument */
+	stw     %r21, -56(%r30)                 /* 6th argument */
+#endif
+
+	/* Are we being ptraced? */
+	mfctl	%cr30, %r1
+	LDREG	TI_FLAGS(%r1),%r1
+	ldi	_TIF_SYSCALL_TRACE_MASK, %r19
+	and,COND(=) %r1, %r19, %r0
+	b,n	.Ltracesys
+	
+	/* Note!  We cannot use the syscall table that is mapped
+	nearby since the gateway page is mapped execute-only. */
+
+#ifdef CONFIG_64BIT
+	ldil	L%sys_call_table, %r1
+	or,=	%r2,%r2,%r2
+	addil	L%(sys_call_table64-sys_call_table), %r1
+	ldo	R%sys_call_table(%r1), %r19
+	or,=	%r2,%r2,%r2
+	ldo	R%sys_call_table64(%r1), %r19
+#else
+	load32	sys_call_table, %r19
+#endif	
+	comiclr,>>	__NR_Linux_syscalls, %r20, %r0
+	b,n	.Lsyscall_nosys
+	
+	LDREGX  %r20(%r19), %r19
+
+	/* If this is a sys_rt_sigreturn call, and the signal was received
+	 * when not in_syscall, then we want to return via syscall_exit_rfi,
+	 * not syscall_exit.  Signal no. in r20, in_syscall in r25 (see
+	 * trampoline code in signal.c).
+	 */
+	ldi	__NR_rt_sigreturn,%r2
+	comb,=	%r2,%r20,.Lrt_sigreturn
+.Lin_syscall:
+	ldil	L%syscall_exit,%r2
+	be      0(%sr7,%r19)
+	ldo	R%syscall_exit(%r2),%r2
+.Lrt_sigreturn:
+	comib,<> 0,%r25,.Lin_syscall
+	ldil	L%syscall_exit_rfi,%r2
+	be      0(%sr7,%r19)
+	ldo	R%syscall_exit_rfi(%r2),%r2
+
+	/* Note!  Because we are not running where we were linked, any
+	calls to functions external to this file must be indirect.  To
+	be safe, we apply the opposite rule to functions within this
+	file, with local labels given to them to ensure correctness. */
+	
+.Lsyscall_nosys:
+syscall_nosys:
+	ldil	L%syscall_exit,%r1
+	be	R%syscall_exit(%sr7,%r1)
+	ldo	-ENOSYS(%r0),%r28		   /* set errno */
+
+
+/* Warning! This trace code is a virtual duplicate of the code above so be
+ * sure to maintain both! */
+.Ltracesys:
+tracesys:
+	/* Need to save more registers so the debugger can see where we
+	 * are.  This saves only the lower 8 bits of PSW, so that the C
+	 * bit is still clear on syscalls, and the D bit is set if this
+	 * full register save path has been executed.  We check the D
+	 * bit on syscall_return_rfi to determine which registers to
+	 * restore.  An interrupt results in a full PSW saved with the
+	 * C bit set, a non-straced syscall entry results in C and D clear
+	 * in the saved PSW.
+	 */
+	ldo     -THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r1      /* get task ptr */
+	LDREG	TI_TASK(%r1), %r1
+	ssm	0,%r2
+	STREG	%r2,TASK_PT_PSW(%r1)		/* Lower 8 bits only!! */
+	mfsp	%sr0,%r2
+	STREG	%r2,TASK_PT_SR0(%r1)
+	mfsp	%sr1,%r2
+	STREG	%r2,TASK_PT_SR1(%r1)
+	mfsp	%sr2,%r2
+	STREG	%r2,TASK_PT_SR2(%r1)
+	mfsp	%sr3,%r2
+	STREG	%r2,TASK_PT_SR3(%r1)
+	STREG	%r2,TASK_PT_SR4(%r1)
+	STREG	%r2,TASK_PT_SR5(%r1)
+	STREG	%r2,TASK_PT_SR6(%r1)
+	STREG	%r2,TASK_PT_SR7(%r1)
+	STREG	%r2,TASK_PT_IASQ0(%r1)
+	STREG	%r2,TASK_PT_IASQ1(%r1)
+	LDREG	TASK_PT_GR31(%r1),%r2
+	STREG	%r2,TASK_PT_IAOQ0(%r1)
+	ldo	4(%r2),%r2
+	STREG	%r2,TASK_PT_IAOQ1(%r1)
+	ldo	TASK_REGS(%r1),%r2
+	/* reg_save %r2 */
+	STREG	%r3,PT_GR3(%r2)
+	STREG	%r4,PT_GR4(%r2)
+	STREG	%r5,PT_GR5(%r2)
+	STREG	%r6,PT_GR6(%r2)
+	STREG	%r7,PT_GR7(%r2)
+	STREG	%r8,PT_GR8(%r2)
+	STREG	%r9,PT_GR9(%r2)
+	STREG	%r10,PT_GR10(%r2)
+	STREG	%r11,PT_GR11(%r2)
+	STREG	%r12,PT_GR12(%r2)
+	STREG	%r13,PT_GR13(%r2)
+	STREG	%r14,PT_GR14(%r2)
+	STREG	%r15,PT_GR15(%r2)
+	STREG	%r16,PT_GR16(%r2)
+	STREG	%r17,PT_GR17(%r2)
+	STREG	%r18,PT_GR18(%r2)
+	/* Finished saving things for the debugger */
+
+	copy	%r2,%r26
+	ldil	L%do_syscall_trace_enter,%r1
+	ldil	L%tracesys_next,%r2
+	be	R%do_syscall_trace_enter(%sr7,%r1)
+	ldo	R%tracesys_next(%r2),%r2
+	
+tracesys_next:
+	/* do_syscall_trace_enter either returned the syscallno, or -1L,
+	 *  so we skip restoring the PT_GR20 below, since we pulled it from
+	 *  task->thread.regs.gr[20] above.
+	 */
+	copy	%ret0,%r20
+
+	ldo     -THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r1      /* get task ptr */
+	LDREG	TI_TASK(%r1), %r1
+	LDREG   TASK_PT_GR28(%r1), %r28		/* Restore return value */
+	LDREG   TASK_PT_GR26(%r1), %r26		/* Restore the users args */
+	LDREG   TASK_PT_GR25(%r1), %r25
+	LDREG   TASK_PT_GR24(%r1), %r24
+	LDREG   TASK_PT_GR23(%r1), %r23
+	LDREG   TASK_PT_GR22(%r1), %r22
+	LDREG   TASK_PT_GR21(%r1), %r21
+#ifdef CONFIG_64BIT
+	ldo	-16(%r30),%r29			/* Reference param save area */
+#else
+	stw     %r22, -52(%r30)                 /* 5th argument */
+	stw     %r21, -56(%r30)                 /* 6th argument */
+#endif
+
+	cmpib,COND(=),n -1,%r20,tracesys_exit /* seccomp may have returned -1 */
+	comiclr,>>	__NR_Linux_syscalls, %r20, %r0
+	b,n	.Ltracesys_nosys
+
+	/* Note!  We cannot use the syscall table that is mapped
+	nearby since the gateway page is mapped execute-only. */
+
+#ifdef CONFIG_64BIT
+	LDREG	TASK_PT_GR30(%r1), %r19		/* get users sp back */
+	extrd,u	%r19,63,1,%r2			/* W hidden in bottom bit */
+
+	ldil	L%sys_call_table, %r1
+	or,=	%r2,%r2,%r2
+	addil	L%(sys_call_table64-sys_call_table), %r1
+	ldo	R%sys_call_table(%r1), %r19
+	or,=	%r2,%r2,%r2
+	ldo	R%sys_call_table64(%r1), %r19
+#else
+	load32	sys_call_table, %r19
+#endif
+
+	LDREGX  %r20(%r19), %r19
+
+	/* If this is a sys_rt_sigreturn call, and the signal was received
+	 * when not in_syscall, then we want to return via syscall_exit_rfi,
+	 * not syscall_exit.  Signal no. in r20, in_syscall in r25 (see
+	 * trampoline code in signal.c).
+	 */
+	ldi	__NR_rt_sigreturn,%r2
+	comb,=	%r2,%r20,.Ltrace_rt_sigreturn
+.Ltrace_in_syscall:
+	ldil	L%tracesys_exit,%r2
+	be      0(%sr7,%r19)
+	ldo	R%tracesys_exit(%r2),%r2
+
+.Ltracesys_nosys:
+	ldo	-ENOSYS(%r0),%r28		/* set errno */
+
+	/* Do *not* call this function on the gateway page, because it
+	makes a direct call to syscall_trace. */
+	
+tracesys_exit:
+	ldo     -THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r1      /* get task ptr */
+	LDREG	TI_TASK(%r1), %r1
+#ifdef CONFIG_64BIT
+	ldo	-16(%r30),%r29			/* Reference param save area */
+#endif
+	ldo	TASK_REGS(%r1),%r26
+	BL	do_syscall_trace_exit,%r2
+	STREG   %r28,TASK_PT_GR28(%r1)          /* save return value now */
+	ldo     -THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r1      /* get task ptr */
+	LDREG	TI_TASK(%r1), %r1
+	LDREG   TASK_PT_GR28(%r1), %r28		/* Restore return val. */
+
+	ldil	L%syscall_exit,%r1
+	be,n	R%syscall_exit(%sr7,%r1)
+
+.Ltrace_rt_sigreturn:
+	comib,<> 0,%r25,.Ltrace_in_syscall
+	ldil	L%tracesys_sigexit,%r2
+	be      0(%sr7,%r19)
+	ldo	R%tracesys_sigexit(%r2),%r2
+
+tracesys_sigexit:
+	ldo     -THREAD_SZ_ALGN-FRAME_SIZE(%r30),%r1      /* get task ptr */
+	LDREG	TI_TASK(%r1), %r1
+#ifdef CONFIG_64BIT
+	ldo	-16(%r30),%r29			/* Reference param save area */
+#endif
+	BL	do_syscall_trace_exit,%r2
+	ldo	TASK_REGS(%r1),%r26
+
+	ldil	L%syscall_exit_rfi,%r1
+	be,n	R%syscall_exit_rfi(%sr7,%r1)
+
+
+	/*********************************************************
+		32/64-bit Light-Weight-Syscall ABI
+
+		* - Indicates a hint for userspace inline asm
+		implementations.
+
+		Syscall number (caller-saves)
+	        - %r20
+	        * In asm clobber.
+
+		Argument registers (caller-saves)
+	        - %r26, %r25, %r24, %r23, %r22
+	        * In asm input.
+
+		Return registers (caller-saves)
+	        - %r28 (return), %r21 (errno)
+	        * In asm output.
+
+		Caller-saves registers
+	        - %r1, %r27, %r29
+	        - %r2 (return pointer)
+	        - %r31 (ble link register)
+	        * In asm clobber.
+
+		Callee-saves registers
+	        - %r3-%r18
+	        - %r30 (stack pointer)
+	        * Not in asm clobber.
+
+		If userspace is 32-bit:
+		Callee-saves registers
+	        - %r19 (32-bit PIC register)
+
+		Differences from 32-bit calling convention:
+		- Syscall number in %r20
+		- Additional argument register %r22 (arg4)
+		- Callee-saves %r19.
+
+		If userspace is 64-bit:
+		Callee-saves registers
+		- %r27 (64-bit PIC register)
+
+		Differences from 64-bit calling convention:
+		- Syscall number in %r20
+		- Additional argument register %r22 (arg4)
+		- Callee-saves %r27.
+
+		Error codes returned by entry path:
+
+		ENOSYS - r20 was an invalid LWS number.
+
+	*********************************************************/
+lws_start:
+
+#ifdef CONFIG_64BIT
+	ssm	PSW_SM_W, %r1
+	extrd,u	%r1,PSW_W_BIT,1,%r1
+	/* sp must be aligned on 4, so deposit the W bit setting into
+	 * the bottom of sp temporarily */
+	or,od	%r1,%r30,%r30
+
+	/* Clip LWS number to a 32-bit value for 32-bit processes */
+	depdi	0, 31, 32, %r20
+#endif	
+
+        /* Is the lws entry number valid? */
+	comiclr,>>	__NR_lws_entries, %r20, %r0
+	b,n	lws_exit_nosys
+
+	/* Load table start */
+	ldil	L%lws_table, %r1
+	ldo	R%lws_table(%r1), %r28	/* Scratch use of r28 */
+	LDREGX	%r20(%sr2,r28), %r21	/* Scratch use of r21 */
+
+	/* Jump to lws, lws table pointers already relocated */
+	be,n	0(%sr2,%r21)
+
+lws_exit_nosys:
+	ldo	-ENOSYS(%r0),%r21		   /* set errno */
+	/* Fall through: Return to userspace */
+
+lws_exit:
+#ifdef CONFIG_64BIT
+	/* decide whether to reset the wide mode bit
+	 *
+	 * For a syscall, the W bit is stored in the lowest bit
+	 * of sp.  Extract it and reset W if it is zero */
+	extrd,u,*<>	%r30,63,1,%r1
+	rsm	PSW_SM_W, %r0
+	/* now reset the lowest bit of sp if it was set */
+	xor	%r30,%r1,%r30
+#endif
+	be,n	0(%sr7, %r31)
+
+
+	
+	/***************************************************
+		Implementing 32bit CAS as an atomic operation:
+
+		%r26 - Address to examine
+		%r25 - Old value to check (old)
+		%r24 - New value to set (new)
+		%r28 - Return prev through this register.
+		%r21 - Kernel error code
+
+		If debugging is DISabled:
+
+		%r21 has the following meanings:
+
+		EAGAIN - CAS is busy, ldcw failed, try again.
+		EFAULT - Read or write failed.		
+
+		If debugging is enabled:
+
+		EDEADLOCK - CAS called recursively.
+		EAGAIN && r28 == 1 - CAS is busy. Lock contended.
+		EAGAIN && r28 == 2 - CAS is busy. ldcw failed.
+		EFAULT - Read or write failed.
+
+		Scratch: r20, r28, r1
+
+	****************************************************/
+
+	/* Do not enable LWS debugging */
+#define ENABLE_LWS_DEBUG 0 
+
+	/* ELF64 Process entry path */
+lws_compare_and_swap64:
+#ifdef CONFIG_64BIT
+	b,n	lws_compare_and_swap
+#else
+	/* If we are not a 64-bit kernel, then we don't
+	 * have 64-bit input registers, and calling
+	 * the 64-bit LWS CAS returns ENOSYS.
+	 */
+	b,n	lws_exit_nosys
+#endif
+
+	/* ELF32 Process entry path */
+lws_compare_and_swap32:
+#ifdef CONFIG_64BIT
+	/* Clip all the input registers */
+	depdi	0, 31, 32, %r26
+	depdi	0, 31, 32, %r25
+	depdi	0, 31, 32, %r24
+#endif
+
+lws_compare_and_swap:
+	/* Load start of lock table */
+	ldil	L%lws_lock_start, %r20
+	ldo	R%lws_lock_start(%r20), %r28
+
+	/* Extract eight bits from r26 and hash lock (Bits 3-11) */
+	extru  %r26, 28, 8, %r20
+
+	/* Find lock to use, the hash is either one of 0 to
+	   15, multiplied by 16 (keep it 16-byte aligned)
+	   and add to the lock table offset. */
+	shlw	%r20, 4, %r20
+	add	%r20, %r28, %r20
+
+# if ENABLE_LWS_DEBUG
+	/*	
+		DEBUG, check for deadlock! 
+		If the thread register values are the same
+		then we were the one that locked it last and
+		this is a recurisve call that will deadlock.
+		We *must* giveup this call and fail.
+	*/
+	ldw	4(%sr2,%r20), %r28			/* Load thread register */
+	/* WARNING: If cr27 cycles to the same value we have problems */
+	mfctl	%cr27, %r21				/* Get current thread register */
+	cmpb,<>,n	%r21, %r28, cas_lock		/* Called recursive? */
+	b	lws_exit				/* Return error! */
+	ldo	-EDEADLOCK(%r0), %r21
+cas_lock:
+	cmpb,=,n	%r0, %r28, cas_nocontend	/* Is nobody using it? */
+	ldo	1(%r0), %r28				/* 1st case */
+	b	lws_exit				/* Contended... */
+	ldo	-EAGAIN(%r0), %r21			/* Spin in userspace */
+cas_nocontend:
+# endif
+/* ENABLE_LWS_DEBUG */
+
+	rsm	PSW_SM_I, %r0				/* Disable interrupts */
+	/* COW breaks can cause contention on UP systems */
+	LDCW	0(%sr2,%r20), %r28			/* Try to acquire the lock */
+	cmpb,<>,n	%r0, %r28, cas_action		/* Did we get it? */
+cas_wouldblock:
+	ldo	2(%r0), %r28				/* 2nd case */
+	ssm	PSW_SM_I, %r0
+	b	lws_exit				/* Contended... */
+	ldo	-EAGAIN(%r0), %r21			/* Spin in userspace */
+
+	/*
+		prev = *addr;
+		if ( prev == old )
+		  *addr = new;
+		return prev;
+	*/
+
+	/* NOTES:
+		This all works becuse intr_do_signal
+		and schedule both check the return iasq
+		and see that we are on the kernel page
+		so this process is never scheduled off
+		or is ever sent any signal of any sort,
+		thus it is wholly atomic from usrspaces
+		perspective
+	*/
+cas_action:
+#if defined CONFIG_SMP && ENABLE_LWS_DEBUG
+	/* DEBUG */
+	mfctl	%cr27, %r1
+	stw	%r1, 4(%sr2,%r20)
+#endif
+	/* The load and store could fail */
+1:	ldw	0(%r26), %r28
+	sub,<>	%r28, %r25, %r0
+2:	stw	%r24, 0(%r26)
+	/* Free lock */
+	stw,ma	%r20, 0(%sr2,%r20)
+#if ENABLE_LWS_DEBUG
+	/* Clear thread register indicator */
+	stw	%r0, 4(%sr2,%r20)
+#endif
+	/* Enable interrupts */
+	ssm	PSW_SM_I, %r0
+	/* Return to userspace, set no error */
+	b	lws_exit
+	copy	%r0, %r21
+
+3:		
+	/* Error occurred on load or store */
+	/* Free lock */
+	stw,ma	%r20, 0(%sr2,%r20)
+#if ENABLE_LWS_DEBUG
+	stw	%r0, 4(%sr2,%r20)
+#endif
+	ssm	PSW_SM_I, %r0
+	b	lws_exit
+	ldo	-EFAULT(%r0),%r21	/* set errno */
+	nop
+	nop
+	nop
+	nop
+
+	/* Two exception table entries, one for the load,
+	   the other for the store. Either return -EFAULT.
+	   Each of the entries must be relocated. */
+	ASM_EXCEPTIONTABLE_ENTRY(1b-linux_gateway_page, 3b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(2b-linux_gateway_page, 3b-linux_gateway_page)
+
+
+	/***************************************************
+		New CAS implementation which uses pointers and variable size
+		information. The value pointed by old and new MUST NOT change
+		while performing CAS. The lock only protect the value at %r26.
+
+		%r26 - Address to examine
+		%r25 - Pointer to the value to check (old)
+		%r24 - Pointer to the value to set (new)
+		%r23 - Size of the variable (0/1/2/3 for 8/16/32/64 bit)
+		%r28 - Return non-zero on failure
+		%r21 - Kernel error code
+
+		%r21 has the following meanings:
+
+		EAGAIN - CAS is busy, ldcw failed, try again.
+		EFAULT - Read or write failed.
+
+		Scratch: r20, r22, r28, r29, r1, fr4 (32bit for 64bit CAS only)
+
+	****************************************************/
+
+	/* ELF32 Process entry path */
+lws_compare_and_swap_2:
+#ifdef CONFIG_64BIT
+	/* Clip the input registers. We don't need to clip %r23 as we
+	   only use it for word operations */
+	depdi	0, 31, 32, %r26
+	depdi	0, 31, 32, %r25
+	depdi	0, 31, 32, %r24
+#endif
+
+	/* Check the validity of the size pointer */
+	subi,>>= 3, %r23, %r0
+	b,n	lws_exit_nosys
+
+	/* Jump to the functions which will load the old and new values into
+	   registers depending on the their size */
+	shlw	%r23, 2, %r29
+	blr	%r29, %r0
+	nop
+
+	/* 8bit load */
+4:	ldb	0(%r25), %r25
+	b	cas2_lock_start
+5:	ldb	0(%r24), %r24
+	nop
+	nop
+	nop
+	nop
+	nop
+
+	/* 16bit load */
+6:	ldh	0(%r25), %r25
+	b	cas2_lock_start
+7:	ldh	0(%r24), %r24
+	nop
+	nop
+	nop
+	nop
+	nop
+
+	/* 32bit load */
+8:	ldw	0(%r25), %r25
+	b	cas2_lock_start
+9:	ldw	0(%r24), %r24
+	nop
+	nop
+	nop
+	nop
+	nop
+
+	/* 64bit load */
+#ifdef CONFIG_64BIT
+10:	ldd	0(%r25), %r25
+11:	ldd	0(%r24), %r24
+#else
+	/* Load old value into r22/r23 - high/low */
+10:	ldw	0(%r25), %r22
+11:	ldw	4(%r25), %r23
+	/* Load new value into fr4 for atomic store later */
+12:	flddx	0(%r24), %fr4
+#endif
+
+cas2_lock_start:
+	/* Load start of lock table */
+	ldil	L%lws_lock_start, %r20
+	ldo	R%lws_lock_start(%r20), %r28
+
+	/* Extract eight bits from r26 and hash lock (Bits 3-11) */
+	extru  %r26, 28, 8, %r20
+
+	/* Find lock to use, the hash is either one of 0 to
+	   15, multiplied by 16 (keep it 16-byte aligned)
+	   and add to the lock table offset. */
+	shlw	%r20, 4, %r20
+	add	%r20, %r28, %r20
+
+	rsm	PSW_SM_I, %r0			/* Disable interrupts */
+	/* COW breaks can cause contention on UP systems */
+	LDCW	0(%sr2,%r20), %r28		/* Try to acquire the lock */
+	cmpb,<>,n	%r0, %r28, cas2_action	/* Did we get it? */
+cas2_wouldblock:
+	ldo	2(%r0), %r28			/* 2nd case */
+	ssm	PSW_SM_I, %r0
+	b	lws_exit			/* Contended... */
+	ldo	-EAGAIN(%r0), %r21		/* Spin in userspace */
+
+	/*
+		prev = *addr;
+		if ( prev == old )
+		  *addr = new;
+		return prev;
+	*/
+
+	/* NOTES:
+		This all works becuse intr_do_signal
+		and schedule both check the return iasq
+		and see that we are on the kernel page
+		so this process is never scheduled off
+		or is ever sent any signal of any sort,
+		thus it is wholly atomic from usrspaces
+		perspective
+	*/
+cas2_action:
+	/* Jump to the correct function */
+	blr	%r29, %r0
+	/* Set %r28 as non-zero for now */
+	ldo	1(%r0),%r28
+
+	/* 8bit CAS */
+13:	ldb	0(%r26), %r29
+	sub,=	%r29, %r25, %r0
+	b,n	cas2_end
+14:	stb	%r24, 0(%r26)
+	b	cas2_end
+	copy	%r0, %r28
+	nop
+	nop
+
+	/* 16bit CAS */
+15:	ldh	0(%r26), %r29
+	sub,=	%r29, %r25, %r0
+	b,n	cas2_end
+16:	sth	%r24, 0(%r26)
+	b	cas2_end
+	copy	%r0, %r28
+	nop
+	nop
+
+	/* 32bit CAS */
+17:	ldw	0(%r26), %r29
+	sub,=	%r29, %r25, %r0
+	b,n	cas2_end
+18:	stw	%r24, 0(%r26)
+	b	cas2_end
+	copy	%r0, %r28
+	nop
+	nop
+
+	/* 64bit CAS */
+#ifdef CONFIG_64BIT
+19:	ldd	0(%r26), %r29
+	sub,*=	%r29, %r25, %r0
+	b,n	cas2_end
+20:	std	%r24, 0(%r26)
+	copy	%r0, %r28
+#else
+	/* Compare first word */
+19:	ldw	0(%r26), %r29
+	sub,=	%r29, %r22, %r0
+	b,n	cas2_end
+	/* Compare second word */
+20:	ldw	4(%r26), %r29
+	sub,=	%r29, %r23, %r0
+	b,n	cas2_end
+	/* Perform the store */
+21:	fstdx	%fr4, 0(%r26)
+	copy	%r0, %r28
+#endif
+
+cas2_end:
+	/* Free lock */
+	stw,ma	%r20, 0(%sr2,%r20)
+	/* Enable interrupts */
+	ssm	PSW_SM_I, %r0
+	/* Return to userspace, set no error */
+	b	lws_exit
+	copy	%r0, %r21
+
+22:
+	/* Error occurred on load or store */
+	/* Free lock */
+	stw,ma	%r20, 0(%sr2,%r20)
+	ssm	PSW_SM_I, %r0
+	ldo	1(%r0),%r28
+	b	lws_exit
+	ldo	-EFAULT(%r0),%r21	/* set errno */
+	nop
+	nop
+	nop
+
+	/* Exception table entries, for the load and store, return EFAULT.
+	   Each of the entries must be relocated. */
+	ASM_EXCEPTIONTABLE_ENTRY(4b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(5b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(6b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(7b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(8b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(9b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(10b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(11b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(13b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(14b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(15b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(16b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(17b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(18b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(19b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(20b-linux_gateway_page, 22b-linux_gateway_page)
+#ifndef CONFIG_64BIT
+	ASM_EXCEPTIONTABLE_ENTRY(12b-linux_gateway_page, 22b-linux_gateway_page)
+	ASM_EXCEPTIONTABLE_ENTRY(21b-linux_gateway_page, 22b-linux_gateway_page)
+#endif
+
+	/* Make sure nothing else is placed on this page */
+	.align PAGE_SIZE
+END(linux_gateway_page)
+ENTRY(end_linux_gateway_page)
+
+	/* Relocate symbols assuming linux_gateway_page is mapped
+	   to virtual address 0x0 */
+
+#define LWS_ENTRY(_name_) ASM_ULONG_INSN (lws_##_name_ - linux_gateway_page)
+
+	.section .rodata,"a"
+
+	.align 8
+	/* Light-weight-syscall table */
+	/* Start of lws table. */
+ENTRY(lws_table)
+	LWS_ENTRY(compare_and_swap32)		/* 0 - ELF32 Atomic 32bit CAS */
+	LWS_ENTRY(compare_and_swap64)		/* 1 - ELF64 Atomic 32bit CAS */
+	LWS_ENTRY(compare_and_swap_2)		/* 2 - ELF32 Atomic 64bit CAS */
+END(lws_table)
+	/* End of lws table */
+
+#define __SYSCALL(nr, entry)	ASM_ULONG_INSN entry
+	.align 8
+ENTRY(sys_call_table)
+	.export sys_call_table,data
+#ifdef CONFIG_64BIT
+#include <asm/syscall_table_c32.h>   /* Compat syscalls */
+#else
+#include <asm/syscall_table_32.h>    /* 32-bit native syscalls */
+#endif
+END(sys_call_table)
+
+#ifdef CONFIG_64BIT
+	.align 8
+ENTRY(sys_call_table64)
+#include <asm/syscall_table_64.h>    /* 64-bit native syscalls */
+END(sys_call_table64)
+#endif
+#undef __SYSCALL
+
+	/*
+		All light-weight-syscall atomic operations 
+		will use this set of locks 
+
+		NOTE: The lws_lock_start symbol must be
+		at least 16-byte aligned for safe use
+		with ldcw.
+	*/
+	.section .data
+	.align	L1_CACHE_BYTES
+ENTRY(lws_lock_start)
+	/* lws locks */
+	.rept 256
+	/* Keep locks aligned at 16-bytes */
+	.word 1
+	.word 0 
+	.word 0
+	.word 0
+	.endr
+END(lws_lock_start)
+	.previous
+
+.end
+
+