x86-64: Simplify and optimize vdso clock_gettime monotonic variants
authorAndy Lutomirski <luto@mit.edu>
Fri, 23 Mar 2012 04:15:51 +0000 (21:15 -0700)
committerJohn Stultz <john.stultz@linaro.org>
Fri, 23 Mar 2012 23:49:33 +0000 (16:49 -0700)
commit91ec87d57fc38c529034e853687dfb7756de5406
tree4f88d6a7e0221bdd791a1de6d4c36b17da8337a2
parent88b28adf6fcdd6d10a1cfc7765bb200d7366a265
x86-64: Simplify and optimize vdso clock_gettime monotonic variants

We used to store the wall-to-monotonic offset and the realtime base.
It's faster to precompute the monotonic base.

This is about a 3% speedup on Sandy Bridge for CLOCK_MONOTONIC.
It's much more impressive for CLOCK_MONOTONIC_COARSE.

Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: John Stultz <john.stultz@linaro.org>
arch/x86/include/asm/vgtod.h
arch/x86/kernel/vsyscall_64.c
arch/x86/vdso/vclock_gettime.c