Messages in this thread Patch in this message | | | From | Andy Lutomirski <> | Subject | [PATCH v5 8/8] x86-64: Optimize vDSO time() | Date | Mon, 23 May 2011 09:31:31 -0400 |
| |
This function just reads a 64-bit variable that's updated atomically, so we don't need any locks.
Signed-off-by: Andy Lutomirski <luto@mit.edu> --- arch/x86/vdso/vclock_gettime.c | 8 ++------ 1 files changed, 2 insertions(+), 6 deletions(-)
diff --git a/arch/x86/vdso/vclock_gettime.c b/arch/x86/vdso/vclock_gettime.c index e6e9f90..a69929e 100644 --- a/arch/x86/vdso/vclock_gettime.c +++ b/arch/x86/vdso/vclock_gettime.c @@ -180,12 +180,8 @@ notrace time_t __vdso_time(time_t *t) if (unlikely(!VVAR(vsyscall_gtod_data).sysctl_enabled)) return time_syscall(t); - do { - seq = read_seqbegin(&VVAR(vsyscall_gtod_data).lock); - - result = VVAR(vsyscall_gtod_data).wall_time_sec; - - } while (read_seqretry(&VVAR(vsyscall_gtod_data).lock, seq)); + /* This is atomic on x86_64 so we don't need any locks. */ + result = ACCESS_ONCE(VVAR(vsyscall_gtod_data).wall_time_sec); if (t) *t = result; -- 1.7.5.1
| |