Mohammed reports (https://bugzilla.kernel.org/show_bug.cgi?id=213029)
the commit e4ab4658f1cf ("clocksource/drivers/hyper-v: Handle vDSO
differences inline") broke vDSO on x86. The problem appears to be that
VDSO_CLOCKMODE_HVCLOCK is an enum value in 'enum vdso_clock_mode' and
'#ifdef VDSO_CLOCKMODE_HVCLOCK' branch evaluates to false (it is not
a define). Use a dedicated HAVE_VDSO_CLOCKMODE_HVCLOCK define instead.
Suggested-by: Thomas Gleixner <[email protected]>
Reported-by: Mohammed Gamal <[email protected]>
Fixes: e4ab4658f1cf ("clocksource/drivers/hyper-v: Handle vDSO differences inline")
Signed-off-by: Vitaly Kuznetsov <[email protected]>
---
arch/x86/include/asm/vdso/clocksource.h | 2 ++
drivers/clocksource/hyperv_timer.c | 4 ++--
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/arch/x86/include/asm/vdso/clocksource.h b/arch/x86/include/asm/vdso/clocksource.h
index 119ac8612d89..136e5e57cfe1 100644
--- a/arch/x86/include/asm/vdso/clocksource.h
+++ b/arch/x86/include/asm/vdso/clocksource.h
@@ -7,4 +7,6 @@
VDSO_CLOCKMODE_PVCLOCK, \
VDSO_CLOCKMODE_HVCLOCK
+#define HAVE_VDSO_CLOCKMODE_HVCLOCK
+
#endif /* __ASM_VDSO_CLOCKSOURCE_H */
diff --git a/drivers/clocksource/hyperv_timer.c b/drivers/clocksource/hyperv_timer.c
index 977fd05ac35f..d6ece7bbce89 100644
--- a/drivers/clocksource/hyperv_timer.c
+++ b/drivers/clocksource/hyperv_timer.c
@@ -419,7 +419,7 @@ static void resume_hv_clock_tsc(struct clocksource *arg)
hv_set_register(HV_REGISTER_REFERENCE_TSC, tsc_msr);
}
-#ifdef VDSO_CLOCKMODE_HVCLOCK
+#ifdef HAVE_VDSO_CLOCKMODE_HVCLOCK
static int hv_cs_enable(struct clocksource *cs)
{
vclocks_set_used(VDSO_CLOCKMODE_HVCLOCK);
@@ -435,7 +435,7 @@ static struct clocksource hyperv_cs_tsc = {
.flags = CLOCK_SOURCE_IS_CONTINUOUS,
.suspend= suspend_hv_clock_tsc,
.resume = resume_hv_clock_tsc,
-#ifdef VDSO_CLOCKMODE_HVCLOCK
+#ifdef HAVE_VDSO_CLOCKMODE_HVCLOCK
.enable = hv_cs_enable,
.vdso_clock_mode = VDSO_CLOCKMODE_HVCLOCK,
#else
--
2.31.1
From: Vitaly Kuznetsov <[email protected]> Sent: Thursday, May 13, 2021 12:33 AM
>
> Mohammed reports (https://bugzilla.kernel.org/show_bug.cgi?id=213029)
> the commit e4ab4658f1cf ("clocksource/drivers/hyper-v: Handle vDSO
> differences inline") broke vDSO on x86. The problem appears to be that
> VDSO_CLOCKMODE_HVCLOCK is an enum value in 'enum vdso_clock_mode' and
> '#ifdef VDSO_CLOCKMODE_HVCLOCK' branch evaluates to false (it is not
> a define). Use a dedicated HAVE_VDSO_CLOCKMODE_HVCLOCK define instead.
>
> Suggested-by: Thomas Gleixner <[email protected]>
> Reported-by: Mohammed Gamal <[email protected]>
> Fixes: e4ab4658f1cf ("clocksource/drivers/hyper-v: Handle vDSO differences inline")
> Signed-off-by: Vitaly Kuznetsov <[email protected]>
> ---
> arch/x86/include/asm/vdso/clocksource.h | 2 ++
> drivers/clocksource/hyperv_timer.c | 4 ++--
> 2 files changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/arch/x86/include/asm/vdso/clocksource.h
> b/arch/x86/include/asm/vdso/clocksource.h
> index 119ac8612d89..136e5e57cfe1 100644
> --- a/arch/x86/include/asm/vdso/clocksource.h
> +++ b/arch/x86/include/asm/vdso/clocksource.h
> @@ -7,4 +7,6 @@
> VDSO_CLOCKMODE_PVCLOCK, \
> VDSO_CLOCKMODE_HVCLOCK
>
> +#define HAVE_VDSO_CLOCKMODE_HVCLOCK
> +
> #endif /* __ASM_VDSO_CLOCKSOURCE_H */
> diff --git a/drivers/clocksource/hyperv_timer.c b/drivers/clocksource/hyperv_timer.c
> index 977fd05ac35f..d6ece7bbce89 100644
> --- a/drivers/clocksource/hyperv_timer.c
> +++ b/drivers/clocksource/hyperv_timer.c
> @@ -419,7 +419,7 @@ static void resume_hv_clock_tsc(struct clocksource *arg)
> hv_set_register(HV_REGISTER_REFERENCE_TSC, tsc_msr);
> }
>
> -#ifdef VDSO_CLOCKMODE_HVCLOCK
> +#ifdef HAVE_VDSO_CLOCKMODE_HVCLOCK
> static int hv_cs_enable(struct clocksource *cs)
> {
> vclocks_set_used(VDSO_CLOCKMODE_HVCLOCK);
> @@ -435,7 +435,7 @@ static struct clocksource hyperv_cs_tsc = {
> .flags = CLOCK_SOURCE_IS_CONTINUOUS,
> .suspend= suspend_hv_clock_tsc,
> .resume = resume_hv_clock_tsc,
> -#ifdef VDSO_CLOCKMODE_HVCLOCK
> +#ifdef HAVE_VDSO_CLOCKMODE_HVCLOCK
> .enable = hv_cs_enable,
> .vdso_clock_mode = VDSO_CLOCKMODE_HVCLOCK,
> #else
> --
> 2.31.1
Reviewed-by: Michael Kelley <[email protected]>
The following commit has been merged into the timers/urgent branch of tip:
Commit-ID: 3486d2c9be652a31033363bdd50391b0c8a8fe21
Gitweb: https://git.kernel.org/tip/3486d2c9be652a31033363bdd50391b0c8a8fe21
Author: Vitaly Kuznetsov <[email protected]>
AuthorDate: Thu, 13 May 2021 09:32:46 +02:00
Committer: Thomas Gleixner <[email protected]>
CommitterDate: Fri, 14 May 2021 14:55:13 +02:00
clocksource/drivers/hyper-v: Re-enable VDSO_CLOCKMODE_HVCLOCK on X86
Mohammed reports (https://bugzilla.kernel.org/show_bug.cgi?id=213029)
the commit e4ab4658f1cf ("clocksource/drivers/hyper-v: Handle vDSO
differences inline") broke vDSO on x86. The problem appears to be that
VDSO_CLOCKMODE_HVCLOCK is an enum value in 'enum vdso_clock_mode' and
'#ifdef VDSO_CLOCKMODE_HVCLOCK' branch evaluates to false (it is not
a define).
Use a dedicated HAVE_VDSO_CLOCKMODE_HVCLOCK define instead.
Fixes: e4ab4658f1cf ("clocksource/drivers/hyper-v: Handle vDSO differences inline")
Reported-by: Mohammed Gamal <[email protected]>
Suggested-by: Thomas Gleixner <[email protected]>
Signed-off-by: Vitaly Kuznetsov <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Reviewed-by: Michael Kelley <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
---
arch/x86/include/asm/vdso/clocksource.h | 2 ++
drivers/clocksource/hyperv_timer.c | 4 ++--
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/arch/x86/include/asm/vdso/clocksource.h b/arch/x86/include/asm/vdso/clocksource.h
index 119ac86..136e5e5 100644
--- a/arch/x86/include/asm/vdso/clocksource.h
+++ b/arch/x86/include/asm/vdso/clocksource.h
@@ -7,4 +7,6 @@
VDSO_CLOCKMODE_PVCLOCK, \
VDSO_CLOCKMODE_HVCLOCK
+#define HAVE_VDSO_CLOCKMODE_HVCLOCK
+
#endif /* __ASM_VDSO_CLOCKSOURCE_H */
diff --git a/drivers/clocksource/hyperv_timer.c b/drivers/clocksource/hyperv_timer.c
index 977fd05..d6ece7b 100644
--- a/drivers/clocksource/hyperv_timer.c
+++ b/drivers/clocksource/hyperv_timer.c
@@ -419,7 +419,7 @@ static void resume_hv_clock_tsc(struct clocksource *arg)
hv_set_register(HV_REGISTER_REFERENCE_TSC, tsc_msr);
}
-#ifdef VDSO_CLOCKMODE_HVCLOCK
+#ifdef HAVE_VDSO_CLOCKMODE_HVCLOCK
static int hv_cs_enable(struct clocksource *cs)
{
vclocks_set_used(VDSO_CLOCKMODE_HVCLOCK);
@@ -435,7 +435,7 @@ static struct clocksource hyperv_cs_tsc = {
.flags = CLOCK_SOURCE_IS_CONTINUOUS,
.suspend= suspend_hv_clock_tsc,
.resume = resume_hv_clock_tsc,
-#ifdef VDSO_CLOCKMODE_HVCLOCK
+#ifdef HAVE_VDSO_CLOCKMODE_HVCLOCK
.enable = hv_cs_enable,
.vdso_clock_mode = VDSO_CLOCKMODE_HVCLOCK,
#else
On Thu, May 13, 2021 at 01:29:12PM +0000, Michael Kelley wrote:
> From: Vitaly Kuznetsov <[email protected]> Sent: Thursday, May 13, 2021 12:33 AM
> >
> > Mohammed reports (https://bugzilla.kernel.org/show_bug.cgi?id=213029)
> > the commit e4ab4658f1cf ("clocksource/drivers/hyper-v: Handle vDSO
> > differences inline") broke vDSO on x86. The problem appears to be that
> > VDSO_CLOCKMODE_HVCLOCK is an enum value in 'enum vdso_clock_mode' and
> > '#ifdef VDSO_CLOCKMODE_HVCLOCK' branch evaluates to false (it is not
> > a define). Use a dedicated HAVE_VDSO_CLOCKMODE_HVCLOCK define instead.
> >
> > Suggested-by: Thomas Gleixner <[email protected]>
> > Reported-by: Mohammed Gamal <[email protected]>
> > Fixes: e4ab4658f1cf ("clocksource/drivers/hyper-v: Handle vDSO differences inline")
> > Signed-off-by: Vitaly Kuznetsov <[email protected]>
[...]
>
> Reviewed-by: Michael Kelley <[email protected]>
>
Applied to hyperv-fixes. Thanks.
On Sat, May 15 2021 at 15:43, Wei Liu wrote:
> On Thu, May 13, 2021 at 01:29:12PM +0000, Michael Kelley wrote:
>> From: Vitaly Kuznetsov <[email protected]> Sent: Thursday, May 13, 2021 12:33 AM
>> >
>> > Mohammed reports (https://bugzilla.kernel.org/show_bug.cgi?id=213029)
>> > the commit e4ab4658f1cf ("clocksource/drivers/hyper-v: Handle vDSO
>> > differences inline") broke vDSO on x86. The problem appears to be that
>> > VDSO_CLOCKMODE_HVCLOCK is an enum value in 'enum vdso_clock_mode' and
>> > '#ifdef VDSO_CLOCKMODE_HVCLOCK' branch evaluates to false (it is not
>> > a define). Use a dedicated HAVE_VDSO_CLOCKMODE_HVCLOCK define instead.
>> >
>> > Suggested-by: Thomas Gleixner <[email protected]>
>> > Reported-by: Mohammed Gamal <[email protected]>
>> > Fixes: e4ab4658f1cf ("clocksource/drivers/hyper-v: Handle vDSO differences inline")
>> > Signed-off-by: Vitaly Kuznetsov <[email protected]>
> [...]
>>
>> Reviewed-by: Michael Kelley <[email protected]>
>>
> Applied to hyperv-fixes. Thanks.
It's already in the tip tree...
On Sat, May 15, 2021 at 09:06:37PM +0200, Thomas Gleixner wrote:
> On Sat, May 15 2021 at 15:43, Wei Liu wrote:
>
> > On Thu, May 13, 2021 at 01:29:12PM +0000, Michael Kelley wrote:
> >> From: Vitaly Kuznetsov <[email protected]> Sent: Thursday, May 13, 2021 12:33 AM
> >> >
> >> > Mohammed reports (https://bugzilla.kernel.org/show_bug.cgi?id=213029)
> >> > the commit e4ab4658f1cf ("clocksource/drivers/hyper-v: Handle vDSO
> >> > differences inline") broke vDSO on x86. The problem appears to be that
> >> > VDSO_CLOCKMODE_HVCLOCK is an enum value in 'enum vdso_clock_mode' and
> >> > '#ifdef VDSO_CLOCKMODE_HVCLOCK' branch evaluates to false (it is not
> >> > a define). Use a dedicated HAVE_VDSO_CLOCKMODE_HVCLOCK define instead.
> >> >
> >> > Suggested-by: Thomas Gleixner <[email protected]>
> >> > Reported-by: Mohammed Gamal <[email protected]>
> >> > Fixes: e4ab4658f1cf ("clocksource/drivers/hyper-v: Handle vDSO differences inline")
> >> > Signed-off-by: Vitaly Kuznetsov <[email protected]>
> > [...]
> >>
> >> Reviewed-by: Michael Kelley <[email protected]>
> >>
> > Applied to hyperv-fixes. Thanks.
>
> It's already in the tip tree...
Okay. I will drop it. Thanks.