diff options
| author |   Uros Bizjak <ubizjak@gmail.com> | 2018-06-05 18:39:09 +0200 |
|---|---|---|
| committer |   Ingo Molnar <mingo@kernel.org> | 2018-06-21 15:21:47 +0200 |
| commit | 1966c5e5bd9802cf62d3744ef4d2d6d32e22604d (patch) | |
| tree | 926b83555049513976c105f3a85af46e9dfcc87e /arch/x86/include/asm/percpu.h | |
| parent | Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma (diff) | |
| download | linux-dev-1966c5e5bd9802cf62d3744ef4d2d6d32e22604d.tar.xz linux-dev-1966c5e5bd9802cf62d3744ef4d2d6d32e22604d.zip | |
x86/asm: Use CC_SET/CC_OUT in percpu_cmpxchg8b_double() to micro-optimize code generation
Use CC_SET(z)/CC_OUT(z) instead of explicit SETZ instruction.
Using these two defines, the compiler that supports generation of
condition code outputs from inline assembly flags generates e.g.:
cmpxchg8b %fs:(%esi)
jne 172255 <__kmalloc+0x65>
instead of:
cmpxchg8b %fs:(%esi)
sete %al
test %al,%al
je 172255 <__kmalloc+0x65>
Note that older compilers now generate:
cmpxchg8b %fs:(%esi)
sete %cl
test %cl,%cl
je 173a85 <__kmalloc+0x65>
since we have to mark that cmpxchg8b instruction outputs to %eax
register and this way clobbers the value in the register.
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.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>
Link: https://lore.kernel.org/lkml/20180605163910.13015-1-ubizjak@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Diffstat (limited to 'arch/x86/include/asm/percpu.h')
| -rw-r--r-- | arch/x86/include/asm/percpu.h | 7 |
1 files changed, 4 insertions, 3 deletions
diff --git a/arch/x86/include/asm/percpu.h b/arch/x86/include/asm/percpu.h index a06b07399d17..e9202a0de8f0 100644 --- a/arch/x86/include/asm/percpu.h +++ b/arch/x86/include/asm/percpu.h @@ -450,9 +450,10 @@ do { \ bool __ret; \ typeof(pcp1) __o1 = (o1), __n1 = (n1); \ typeof(pcp2) __o2 = (o2), __n2 = (n2); \ - asm volatile("cmpxchg8b "__percpu_arg(1)"\n\tsetz %0\n\t" \ - : "=a" (__ret), "+m" (pcp1), "+m" (pcp2), "+d" (__o2) \ - : "b" (__n1), "c" (__n2), "a" (__o1)); \ + asm volatile("cmpxchg8b "__percpu_arg(1) \ + CC_SET(z) \ + : CC_OUT(z) (__ret), "+m" (pcp1), "+m" (pcp2), "+a" (__o1), "+d" (__o2) \ + : "b" (__n1), "c" (__n2)); \ __ret; \ }) |