From: Andy Lutomirski Date: Mon, 4 Jan 2016 23:14:28 +0000 (-0800) Subject: x86/vdso/pvclock: Protect STABLE check with the seqcount X-Git-Url: https://git.stricted.de/?a=commitdiff_plain;h=78fd8c7288e0a4bba3ad1d69caf9396a6b69cb00;p=GitHub%2Fmoto-9609%2Fandroid_kernel_motorola_exynos9610.git x86/vdso/pvclock: Protect STABLE check with the seqcount If the clock becomes unstable while we're reading it, we need to bail. We can do this by simply moving the check into the seqcount loop. Reported-by: Marcelo Tosatti Signed-off-by: Andy Lutomirski Cc: Alexander Graf Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Linus Torvalds Cc: Paolo Bonzini Cc: Peter Zijlstra Cc: Radim Krcmar Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/755dcedb17269e1d7ce12a9a713dea303835137e.1451949191.git.luto@kernel.org Signed-off-by: Ingo Molnar --- diff --git a/arch/x86/entry/vdso/vclock_gettime.c b/arch/x86/entry/vdso/vclock_gettime.c index 8602f06c759f..1a50e09c945b 100644 --- a/arch/x86/entry/vdso/vclock_gettime.c +++ b/arch/x86/entry/vdso/vclock_gettime.c @@ -126,23 +126,23 @@ static notrace cycle_t vread_pvclock(int *mode) * * On Xen, we don't appear to have that guarantee, but Xen still * supplies a valid seqlock using the version field. - + * * We only do pvclock vdso timing at all if * PVCLOCK_TSC_STABLE_BIT is set, and we interpret that bit to * mean that all vCPUs have matching pvti and that the TSC is * synced, so we can just look at vCPU 0's pvti. */ - if (unlikely(!(pvti->flags & PVCLOCK_TSC_STABLE_BIT))) { - *mode = VCLOCK_NONE; - return 0; - } - do { version = pvti->version; smp_rmb(); + if (unlikely(!(pvti->flags & PVCLOCK_TSC_STABLE_BIT))) { + *mode = VCLOCK_NONE; + return 0; + } + tsc = rdtsc_ordered(); pvti_tsc_to_system_mul = pvti->tsc_to_system_mul; pvti_tsc_shift = pvti->tsc_shift;