diff options
| author | Will Deacon <will.deacon@arm.com> | 2014-02-21 17:01:48 +0100 |
|---|---|---|
| committer | Russell King <rmk+kernel@arm.linux.org.uk> | 2014-02-25 11:30:20 +0000 |
| commit | c32ffce0f66e5d1d4856254516e24f5ef275cd00 (patch) | |
| tree | 125229cdd38bfd6e7e62cff7eb8771a34cc999a7 /arch/arm/include/asm/futex.h | |
| parent | 6ea41c80115f49e7d8b80312ffc99973d283471f (diff) | |
ARM: 7984/1: prefetch: add prefetchw invocations for barriered atomics
After a bunch of benchmarking on the interaction between dmb and pldw,
it turns out that issuing the pldw *after* the dmb instruction can
give modest performance gains (~3% atomic_add_return improvement on a
dual A15).
This patch adds prefetchw invocations to our barriered atomic operations
including cmpxchg, test_and_xxx and futexes.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Diffstat (limited to 'arch/arm/include/asm/futex.h')
| -rw-r--r-- | arch/arm/include/asm/futex.h | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/arch/arm/include/asm/futex.h b/arch/arm/include/asm/futex.h index 2aff798fbef4..53e69dae796f 100644 --- a/arch/arm/include/asm/futex.h +++ b/arch/arm/include/asm/futex.h @@ -23,6 +23,7 @@ #define __futex_atomic_op(insn, ret, oldval, tmp, uaddr, oparg) \ smp_mb(); \ + prefetchw(uaddr); \ __asm__ __volatile__( \ "1: ldrex %1, [%3]\n" \ " " insn "\n" \ @@ -46,6 +47,8 @@ futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr, return -EFAULT; smp_mb(); + /* Prefetching cannot fault */ + prefetchw(uaddr); __asm__ __volatile__("@futex_atomic_cmpxchg_inatomic\n" "1: ldrex %1, [%4]\n" " teq %1, %2\n" |