diff options
| author | Will Deacon <will.deacon@arm.com> | 2014-02-07 19:12:32 +0100 |
|---|---|---|
| committer | Jiri Slaby <jslaby@suse.cz> | 2014-03-05 17:13:40 +0100 |
| commit | 103a9391af1e014f6cc1c1e224a4f51bf831768f (patch) | |
| tree | e05111546ba0bc104a49c509e34dc9ab68b7de97 /arch | |
| parent | 160d1d210a8cc5b29722580484f5256882dc275e (diff) | |
ARM: 7955/1: spinlock: ensure we have a compiler barrier before sev
commit 7c8746a9eb287642deaad0e7c2cdf482dce5e4be upstream.
When unlocking a spinlock, we require the following, strictly ordered
sequence of events:
<barrier> /* dmb */
<unlock>
<barrier> /* dsb */
<sev>
Whilst the code does indeed reflect this in terms of the architecture,
the final <barrier> + <sev> have been contracted into a single inline
asm without a "memory" clobber, therefore the compiler is at liberty to
reorder the unlock to the end of the above sequence. In such a case,
a waiting CPU may be woken up before the lock has been unlocked, leading
to extremely poor performance.
This patch reworks the dsb_sev() function to make use of the dsb()
macro and ensure ordering against the unlock.
Reported-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Diffstat (limited to 'arch')
| -rw-r--r-- | arch/arm/include/asm/spinlock.h | 15 |
1 files changed, 3 insertions, 12 deletions
diff --git a/arch/arm/include/asm/spinlock.h b/arch/arm/include/asm/spinlock.h index 4f2c28060c9..05f8066255d 100644 --- a/arch/arm/include/asm/spinlock.h +++ b/arch/arm/include/asm/spinlock.h @@ -44,18 +44,9 @@ static inline void dsb_sev(void) { -#if __LINUX_ARM_ARCH__ >= 7 - __asm__ __volatile__ ( - "dsb ishst\n" - SEV - ); -#else - __asm__ __volatile__ ( - "mcr p15, 0, %0, c7, c10, 4\n" - SEV - : : "r" (0) - ); -#endif + + dsb(ishst); + __asm__(SEV); } /* |