korg/linux
0
0
Fork 0
mirror of https://mirrors.bfsu.edu.cn/git/linux.git synced 2024-12-14 06:24:53 +08:00
Code Issues Packages Projects Releases Wiki Activity
3e233385ef
linux/arch/x86/lib/atomic64_386_32.S

192 lines
2.5 KiB
ArmAsm
Raw Normal View History

x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
/*
* atomic64_t for 386/486
*
* Copyright © 2010 Luca Barbieri
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*/
#include <linux/linkage.h>
#include <asm/alternative-asm.h>
/* if you want SMP support, implement these with real spinlocks */
.macro LOCK reg
x86/debug: Remove perpetually broken, unmaintainable dwarf annotations So the dwarf2 annotations in low level assembly code have become an increasing hindrance: unreadable, messy macros mixed into some of the most security sensitive code paths of the Linux kernel. These debug info annotations don't even buy the upstream kernel anything: dwarf driven stack unwinding has caused problems in the past so it's out of tree, and the upstream kernel only uses the much more robust framepointers based stack unwinding method. In addition to that there's a steady, slow bitrot going on with these annotations, requiring frequent fixups. There's no tooling and no functionality upstream that keeps it correct. So burn down the sick forest, allowing new, healthier growth: 27 files changed, 350 insertions(+), 1101 deletions(-) Someone who has the willingness and time to do this properly can attempt to reintroduce dwarf debuginfo in x86 assembly code plus dwarf unwinding from first principles, with the following conditions: - it should be maximally readable, and maximally low-key to 'ordinary' code reading and maintenance. - find a build time method to insert dwarf annotations automatically in the most common cases, for pop/push instructions that manipulate the stack pointer. This could be done for example via a preprocessing step that just looks for common patterns - plus special annotations for the few cases where we want to depart from the default. We have hundreds of CFI annotations, so automating most of that makes sense. - it should come with build tooling checks that ensure that CFI annotations are sensible. We've seen such efforts from the framepointer side, and there's no reason it couldn't be done on the dwarf side. Cc: Andy Lutomirski <luto@amacapital.net> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: Frédéric Weisbecker <fweisbec@gmail.com Cc: H. Peter Anvin <hpa@zytor.com> Cc: Jan Beulich <JBeulich@suse.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-05-28 18:21:47 +08:00
pushfl
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
cli
.endm
.macro UNLOCK reg
x86/debug: Remove perpetually broken, unmaintainable dwarf annotations So the dwarf2 annotations in low level assembly code have become an increasing hindrance: unreadable, messy macros mixed into some of the most security sensitive code paths of the Linux kernel. These debug info annotations don't even buy the upstream kernel anything: dwarf driven stack unwinding has caused problems in the past so it's out of tree, and the upstream kernel only uses the much more robust framepointers based stack unwinding method. In addition to that there's a steady, slow bitrot going on with these annotations, requiring frequent fixups. There's no tooling and no functionality upstream that keeps it correct. So burn down the sick forest, allowing new, healthier growth: 27 files changed, 350 insertions(+), 1101 deletions(-) Someone who has the willingness and time to do this properly can attempt to reintroduce dwarf debuginfo in x86 assembly code plus dwarf unwinding from first principles, with the following conditions: - it should be maximally readable, and maximally low-key to 'ordinary' code reading and maintenance. - find a build time method to insert dwarf annotations automatically in the most common cases, for pop/push instructions that manipulate the stack pointer. This could be done for example via a preprocessing step that just looks for common patterns - plus special annotations for the few cases where we want to depart from the default. We have hundreds of CFI annotations, so automating most of that makes sense. - it should come with build tooling checks that ensure that CFI annotations are sensible. We've seen such efforts from the framepointer side, and there's no reason it couldn't be done on the dwarf side. Cc: Andy Lutomirski <luto@amacapital.net> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: Frédéric Weisbecker <fweisbec@gmail.com Cc: H. Peter Anvin <hpa@zytor.com> Cc: Jan Beulich <JBeulich@suse.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-05-28 18:21:47 +08:00
popfl
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
.endm
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#define BEGIN(op) \
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
.macro endp; \
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
ENDPROC(atomic64_##op##_386); \
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
.purgem endp; \
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
.endm; \
ENTRY(atomic64_##op##_386); \
LOCK v;
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
#define ENDP endp
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#define RET \
UNLOCK v; \
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
ret
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
#define RET_ENDP \
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
RET; \
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
ENDP
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#define v %ecx
BEGIN(read)
movl (v), %eax
movl 4(v), %edx
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
RET_ENDP
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#undef v
#define v %esi
BEGIN(set)
movl %ebx, (v)
movl %ecx, 4(v)
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
RET_ENDP
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#undef v
#define v %esi
BEGIN(xchg)
movl (v), %eax
movl 4(v), %edx
movl %ebx, (v)
movl %ecx, 4(v)
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
RET_ENDP
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#undef v
#define v %ecx
BEGIN(add)
addl %eax, (v)
adcl %edx, 4(v)
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
RET_ENDP
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#undef v
#define v %ecx
BEGIN(add_return)
addl (v), %eax
adcl 4(v), %edx
movl %eax, (v)
movl %edx, 4(v)
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
RET_ENDP
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#undef v
#define v %ecx
BEGIN(sub)
subl %eax, (v)
sbbl %edx, 4(v)
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
RET_ENDP
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#undef v
#define v %ecx
BEGIN(sub_return)
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
negl %edx
negl %eax
sbbl $0, %edx
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
addl (v), %eax
adcl 4(v), %edx
movl %eax, (v)
movl %edx, 4(v)
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
RET_ENDP
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#undef v
#define v %esi
BEGIN(inc)
addl $1, (v)
adcl $0, 4(v)
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
RET_ENDP
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#undef v
#define v %esi
BEGIN(inc_return)
movl (v), %eax
movl 4(v), %edx
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
addl $1, %eax
adcl $0, %edx
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
movl %eax, (v)
movl %edx, 4(v)
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
RET_ENDP
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#undef v
#define v %esi
BEGIN(dec)
subl $1, (v)
sbbl $0, 4(v)
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
RET_ENDP
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#undef v
#define v %esi
BEGIN(dec_return)
movl (v), %eax
movl 4(v), %edx
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
subl $1, %eax
sbbl $0, %edx
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
movl %eax, (v)
movl %edx, 4(v)
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
RET_ENDP
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#undef v
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
x86: atomic64 assembly improvements In the "xchg" implementation, %ebx and %ecx don't need to be copied into %eax and %edx respectively (this is only necessary when desiring to only read the stored value). In the "add_unless" implementation, swapping the use of %ecx and %esi for passing arguments allows %esi to become an input only (i.e. permitting the register to be re-used to address the same object without reload). In "{add,sub}_return", doing the initial read64 through the passed in %ecx decreases a register dependency. In "inc_not_zero", a branch can be eliminated by or-ing together the two halves of the current (64-bit) value, and code size can be further reduced by adjusting the arithmetic slightly. v2: Undo the folding of "xchg" and "set". Signed-off-by: Jan Beulich <jbeulich@suse.com> Link: http://lkml.kernel.org/r/4F19A2BC020000780006E0DC@nat28.tlf.novell.com Cc: Luca Barbieri <luca@luca-barbieri.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2012-01-21 00:22:04 +08:00
#define v %esi
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
BEGIN(add_unless)
x86: atomic64 assembly improvements In the "xchg" implementation, %ebx and %ecx don't need to be copied into %eax and %edx respectively (this is only necessary when desiring to only read the stored value). In the "add_unless" implementation, swapping the use of %ecx and %esi for passing arguments allows %esi to become an input only (i.e. permitting the register to be re-used to address the same object without reload). In "{add,sub}_return", doing the initial read64 through the passed in %ecx decreases a register dependency. In "inc_not_zero", a branch can be eliminated by or-ing together the two halves of the current (64-bit) value, and code size can be further reduced by adjusting the arithmetic slightly. v2: Undo the folding of "xchg" and "set". Signed-off-by: Jan Beulich <jbeulich@suse.com> Link: http://lkml.kernel.org/r/4F19A2BC020000780006E0DC@nat28.tlf.novell.com Cc: Luca Barbieri <luca@luca-barbieri.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2012-01-21 00:22:04 +08:00
addl %eax, %ecx
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
adcl %edx, %edi
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
addl (v), %eax
adcl 4(v), %edx
x86: atomic64 assembly improvements In the "xchg" implementation, %ebx and %ecx don't need to be copied into %eax and %edx respectively (this is only necessary when desiring to only read the stored value). In the "add_unless" implementation, swapping the use of %ecx and %esi for passing arguments allows %esi to become an input only (i.e. permitting the register to be re-used to address the same object without reload). In "{add,sub}_return", doing the initial read64 through the passed in %ecx decreases a register dependency. In "inc_not_zero", a branch can be eliminated by or-ing together the two halves of the current (64-bit) value, and code size can be further reduced by adjusting the arithmetic slightly. v2: Undo the folding of "xchg" and "set". Signed-off-by: Jan Beulich <jbeulich@suse.com> Link: http://lkml.kernel.org/r/4F19A2BC020000780006E0DC@nat28.tlf.novell.com Cc: Luca Barbieri <luca@luca-barbieri.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2012-01-21 00:22:04 +08:00
cmpl %eax, %ecx
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
je 3f
1:
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
movl %eax, (v)
movl %edx, 4(v)
x86-32: Fix atomic64_add_unless return value convention atomic64_add_unless must return 1 if it perfomed the add and 0 otherwise. The implementation did the opposite thing. Reported-by: H. Peter Anvin <hpa@zytor.com> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267469749-11878-3-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-03-02 02:55:46 +08:00
movl $1, %eax
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
2:
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
RET
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
3:
cmpl %edx, %edi
jne 1b
x86-32: Fix atomic64_add_unless return value convention atomic64_add_unless must return 1 if it perfomed the add and 0 otherwise. The implementation did the opposite thing. Reported-by: H. Peter Anvin <hpa@zytor.com> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267469749-11878-3-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-03-02 02:55:46 +08:00
xorl %eax, %eax
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
jmp 2b
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
ENDP
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#undef v
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#define v %esi
BEGIN(inc_not_zero)
movl (v), %eax
movl 4(v), %edx
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
testl %eax, %eax
je 3f
1:
addl $1, %eax
adcl $0, %edx
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
movl %eax, (v)
movl %edx, 4(v)
x86-32: Fix atomic64_inc_not_zero return value convention atomic64_inc_not_zero must return 1 if it perfomed the add and 0 otherwise. It was doing the opposite thing. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267469749-11878-6-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-03-02 02:55:49 +08:00
movl $1, %eax
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
2:
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
RET
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
3:
testl %edx, %edx
jne 1b
jmp 2b
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
ENDP
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#undef v
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#define v %esi
BEGIN(dec_if_positive)
movl (v), %eax
movl 4(v), %edx
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
subl $1, %eax
sbbl $0, %edx
js 1f
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
movl %eax, (v)
movl %edx, 4(v)
x86-32: Rewrite 32-bit atomic64 functions in assembly This patch replaces atomic64_32.c with two assembly implementations, one for 386/486 machines using pushf/cli/popf and one for 586+ machines using cmpxchg8b. The cmpxchg8b implementation provides the following advantages over the current one: 1. Implements atomic64_add_unless, atomic64_dec_if_positive and atomic64_inc_not_zero 2. Uses the ZF flag changed by cmpxchg8b instead of doing a comparison 3. Uses custom register calling conventions that reduce or eliminate register moves to suit cmpxchg8b 4. Reads the initial value instead of using cmpxchg8b to do that. Currently we use lock xaddl and movl, which seems the fastest. 5. Does not use the lock prefix for atomic64_set 64-bit writes are already atomic, so we don't need that. We still need it for atomic64_read to avoid restoring a value changed in the meantime. 6. Allocates registers as well or better than gcc The 386 implementation provides support for 386 and 486 machines. 386/486 SMP is not supported (we dropped it), but such support can be added easily if desired. A pure assembly implementation is required due to the custom calling conventions, and desire to use %ebp in atomic64_add_return (we need 7 registers...), as well as the ability to use pushf/popf in the 386 code without an intermediate pop/push. The parameter names are changed to match the convention in atomic_64.h Changes in v3 (due to rebasing to tip/x86/asm): - Patches atomic64_32.h instead of atomic_32.h - Uses the CALL alternative mechanism from commit 1b1d9258181bae199dc940f4bd0298126b9a73d9 Changes in v2: - Merged 386 and cx8 support in the same patch - 386 support now done in assembly, C code no longer used at all - cmpxchg64 is used for atomic64_cmpxchg - stop using macros, use one-line inline functions instead - miscellanous changes and improvements Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <1267005265-27958-5-git-send-email-luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-02-24 17:54:25 +08:00
1:
x86, asm: Use a lower case name for the end macro in atomic64_386_32.S Use a lowercase name for the end macro, which somehow fixes a binutils 2.16 problem. Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> LKML-Reference: <tip-30246557a06bb20618bed906a06d1e1e0faa8bb4@git.kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2010-08-12 22:00:35 +08:00
RET_ENDP
x86, asm: Refactor atomic64_386_32.S to support old binutils and be cleaner The old code didn't work on binutils 2.12 because setting a symbol to a register apparently requires a fairly recent version. This commit refactors the code to use the C preprocessor instead, and in the process makes the whole code a bit easier to understand. The object code produced is unchanged as expected. This fixes kernel bugzilla 16506. Reported-by: Dieter Stussy <kd6lvw+software@kd6lvw.ampr.org> Signed-off-by: Luca Barbieri <luca@luca-barbieri.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com> Cc: <stable@kernel.org> 2.6.35 LKML-Reference: <tip-*@git.kernel.org>
2010-08-06 10:04:38 +08:00
#undef v
Reference in New Issue Copy Permalink