diff mbox series

[RFC,3/4] arm64: vdso: Fix nsec handling for CLOCK_MONOTONIC_RAW

Message ID 1495856035-6622-4-git-send-email-john.stultz@linaro.org
State Superseded
Headers show
Series Fixes for two recently found timekeeping bugs | expand

Commit Message

John Stultz May 27, 2017, 3:33 a.m. UTC 
From: Will Deacon <will.deacon@arm.com>


Commit 45a7905fc48f ("arm64: vdso: defer shifting of nanosecond
component of timespec") fixed sub-ns inaccuracies in our vDSO
clock_gettime implementation by deferring the right-shift of the
nanoseconds components until after the timespec addition, which
operates on left-shifted values. That worked nicely until
support for CLOCK_MONOTONIC_RAW was added in 49eea433b326
("arm64: Add support for CLOCK_MONOTONIC_RAW in clock_gettime()
vDSO"). Noticing that the core timekeeping code never set
tkr_raw.xtime_nsec, the vDSO implementation didn't bother
exposing it via the data page and instead took the unshifted
tk->raw_time.tv_nsec value which was then immediately shifted
left in the vDSO code.

Now that the core code is actually setting tkr_raw.xtime_nsec,
we need to take that into account in the vDSO by adding it to
the shifted raw_time value. Rather than do that at each use (and
expand the data page in the process), instead perform the
shift/addition operation when populating the data page and
remove the shift from the vDSO code entirely.

Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Miroslav Lichvar <mlichvar@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Stephen Boyd <stephen.boyd@linaro.org>
Cc: Kevin Brodsky <kevin.brodsky@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Daniel Mentz <danielmentz@google.com>
Reported-by: John Stultz <john.stultz@linaro.org>
Acked-by: Acked-by: Kevin Brodsky <kevin.brodsky@arm.com>

Signed-off-by: Will Deacon <will.deacon@arm.com>

[jstultz: minor whitespace tweak]
Signed-off-by: John Stultz <john.stultz@linaro.org>

---
 arch/arm64/kernel/vdso.c              | 5 +++--
 arch/arm64/kernel/vdso/gettimeofday.S | 1 -
 2 files changed, 3 insertions(+), 3 deletions(-)

-- 
2.7.4

Comments

Will Deacon May 30, 2017, 9:38 a.m. UTC | #1 
On Fri, May 26, 2017 at 08:33:54PM -0700, John Stultz wrote:
> From: Will Deacon <will.deacon@arm.com>

> 

> Commit 45a7905fc48f ("arm64: vdso: defer shifting of nanosecond

> component of timespec") fixed sub-ns inaccuracies in our vDSO

> clock_gettime implementation by deferring the right-shift of the

> nanoseconds components until after the timespec addition, which

> operates on left-shifted values. That worked nicely until

> support for CLOCK_MONOTONIC_RAW was added in 49eea433b326

> ("arm64: Add support for CLOCK_MONOTONIC_RAW in clock_gettime()

> vDSO"). Noticing that the core timekeeping code never set

> tkr_raw.xtime_nsec, the vDSO implementation didn't bother

> exposing it via the data page and instead took the unshifted

> tk->raw_time.tv_nsec value which was then immediately shifted

> left in the vDSO code.

> 

> Now that the core code is actually setting tkr_raw.xtime_nsec,

> we need to take that into account in the vDSO by adding it to

> the shifted raw_time value. Rather than do that at each use (and

> expand the data page in the process), instead perform the

> shift/addition operation when populating the data page and

> remove the shift from the vDSO code entirely.

> 

> Cc: Thomas Gleixner <tglx@linutronix.de>

> Cc: Ingo Molnar <mingo@kernel.org>

> Cc: Miroslav Lichvar <mlichvar@redhat.com>

> Cc: Richard Cochran <richardcochran@gmail.com>

> Cc: Prarit Bhargava <prarit@redhat.com>

> Cc: Stephen Boyd <stephen.boyd@linaro.org>

> Cc: Kevin Brodsky <kevin.brodsky@arm.com>

> Cc: Will Deacon <will.deacon@arm.com>

> Cc: Daniel Mentz <danielmentz@google.com>

> Reported-by: John Stultz <john.stultz@linaro.org>

> Acked-by: Acked-by: Kevin Brodsky <kevin.brodsky@arm.com>


I don't think Kevin liked it *that* much ^^

Will
diff mbox series

Patch

diff --git a/arch/arm64/kernel/vdso.c b/arch/arm64/kernel/vdso.c
index 41b6e31..d0cb007 100644
--- a/arch/arm64/kernel/vdso.c
+++ b/arch/arm64/kernel/vdso.c
@@ -221,10 +221,11 @@  void update_vsyscall(struct timekeeper *tk)
 		/* tkr_mono.cycle_last == tkr_raw.cycle_last */
 		vdso_data->cs_cycle_last	= tk->tkr_mono.cycle_last;
 		vdso_data->raw_time_sec		= tk->raw_time.tv_sec;
-		vdso_data->raw_time_nsec	= tk->raw_time.tv_nsec;
+		vdso_data->raw_time_nsec	= (tk->raw_time.tv_nsec <<
+						   tk->tkr_raw.shift) +
+						  tk->tkr_raw.xtime_nsec;
 		vdso_data->xtime_clock_sec	= tk->xtime_sec;
 		vdso_data->xtime_clock_nsec	= tk->tkr_mono.xtime_nsec;
-		/* tkr_raw.xtime_nsec == 0 */
 		vdso_data->cs_mono_mult		= tk->tkr_mono.mult;
 		vdso_data->cs_raw_mult		= tk->tkr_raw.mult;
 		/* tkr_mono.shift == tkr_raw.shift */
diff --git a/arch/arm64/kernel/vdso/gettimeofday.S b/arch/arm64/kernel/vdso/gettimeofday.S
index e00b467..76320e9 100644
--- a/arch/arm64/kernel/vdso/gettimeofday.S
+++ b/arch/arm64/kernel/vdso/gettimeofday.S
@@ -256,7 +256,6 @@  monotonic_raw:
 	seqcnt_check fail=monotonic_raw
 
 	/* All computations are done with left-shifted nsecs. */
-	lsl	x14, x14, x12
 	get_nsec_per_sec res=x9
 	lsl	x9, x9, x12