* [PATCH] perf kvm: Fix arm64 VM-exit duration accounting
@ 2026-10-31 12:01 lirongqing
2026-10-31 12:05 ` sashiko-bot
2026-11-04 11:17 ` Leo Yan
0 siblings, 2 replies; 4+ messages in thread
From: lirongqing @ 2026-10-31 12:01 UTC (permalink / raw)
To: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo,
Namhyung Kim, Mark Rutland, Alexander Shishkin, Jiri Olsa,
Ian Rogers, Adrian Hunter, James Clark, Li RongQing,
Sergey Senozhatsky, Leo Yan, linux-perf-users, linux-kernel
From: Li RongQing <lirongqing@baidu.com>
Use kvm_exit as the start event and the following kvm_entry as the
end event for arm64 VM-exit statistics. Keep decoding ret and esr_ec
from the exit sample.
Before this change, arm64 measured guest runtime as VM-exit time.
Fixes: 1218838d68f5 ("perf kvm: Add kvm-stat for arm64")
Signed-off-by: Li RongQing <lirongqing@baidu.com>
---
tools/perf/util/kvm-stat-arch/kvm-stat-arm64.c | 14 +++++++-------
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/tools/perf/util/kvm-stat-arch/kvm-stat-arm64.c b/tools/perf/util/kvm-stat-arch/kvm-stat-arm64.c
index 018b0db..ff853ef 100644
--- a/tools/perf/util/kvm-stat-arch/kvm-stat-arm64.c
+++ b/tools/perf/util/kvm-stat-arch/kvm-stat-arm64.c
@@ -36,13 +36,7 @@ static void event_get_key(struct perf_sample *sample,
}
static bool event_begin(struct perf_sample *sample,
- struct event_key *key __maybe_unused)
-{
- return evsel__name_is(sample->evsel, kvm_entry_trace(EM_AARCH64));
-}
-
-static bool event_end(struct perf_sample *sample,
- struct event_key *key)
+ struct event_key *key)
{
if (evsel__name_is(sample->evsel, kvm_exit_trace(EM_AARCH64))) {
event_get_key(sample, key);
@@ -51,6 +45,12 @@ static bool event_end(struct perf_sample *sample,
return false;
}
+static bool event_end(struct perf_sample *sample,
+ struct event_key *key __maybe_unused)
+{
+ return evsel__name_is(sample->evsel, kvm_entry_trace(EM_AARCH64));
+}
+
static const struct kvm_events_ops exit_events = {
.is_begin_event = event_begin,
.is_end_event = event_end,
--
2.9.4
^ permalink raw reply related [flat|nested] 4+ messages in thread* Re: [PATCH] perf kvm: Fix arm64 VM-exit duration accounting
2026-10-31 12:01 [PATCH] perf kvm: Fix arm64 VM-exit duration accounting lirongqing
@ 2026-10-31 12:05 ` sashiko-bot
2026-11-04 11:17 ` Leo Yan
1 sibling, 0 replies; 4+ messages in thread
From: sashiko-bot @ 2026-10-31 12:05 UTC (permalink / raw)
To: lirongqing, linux-perf-users
> From: Li RongQing <lirongqing@baidu.com>
>
> Use kvm_exit as the start event and the following kvm_entry as the
> end event for arm64 VM-exit statistics. Keep decoding ret and esr_ec
> from the exit sample.
>
> Before this change, arm64 measured guest runtime as VM-exit time.
>
> Fixes: 1218838d68f5 ("perf kvm: Add kvm-stat for arm64")
> Signed-off-by: Li RongQing <lirongqing@baidu.com>
Sashiko has reviewed this patch and found no issues.
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261031120123.2312-1-lirongqing@baidu.com?part=1
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] perf kvm: Fix arm64 VM-exit duration accounting
2026-10-31 12:01 [PATCH] perf kvm: Fix arm64 VM-exit duration accounting lirongqing
2026-10-31 12:05 ` sashiko-bot
@ 2026-11-04 11:17 ` Leo Yan
2026-11-04 21:19 ` Arnaldo Carvalho de Melo
1 sibling, 1 reply; 4+ messages in thread
From: Leo Yan @ 2026-11-04 11:17 UTC (permalink / raw)
To: lirongqing
Cc: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo,
Namhyung Kim, Mark Rutland, Alexander Shishkin, Jiri Olsa,
Ian Rogers, Adrian Hunter, James Clark, Sergey Senozhatsky,
Leo Yan, linux-perf-users, linux-kernel
On Sat, Oct 31, 2026 at 08:01:23PM +0800, lirongqing wrote:
> From: Li RongQing <lirongqing@baidu.com>
>
> Use kvm_exit as the start event and the following kvm_entry as the
> end event for arm64 VM-exit statistics. Keep decoding ret and esr_ec
> from the exit sample.
>
> Before this change, arm64 measured guest runtime as VM-exit time.
>
> Fixes: 1218838d68f5 ("perf kvm: Add kvm-stat for arm64")
> Signed-off-by: Li RongQing <lirongqing@baidu.com>
Thanks for the fix!
Reviewed-by: Leo Yan <leo.yan@arm.com>
Since arm64 requires the exit reason, we cannot directly use the common
exit_event_begin()/exit_event_end() callbacks as other archs do.
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] perf kvm: Fix arm64 VM-exit duration accounting
2026-11-04 11:17 ` Leo Yan
@ 2026-11-04 21:19 ` Arnaldo Carvalho de Melo
0 siblings, 0 replies; 4+ messages in thread
From: Arnaldo Carvalho de Melo @ 2026-11-04 21:19 UTC (permalink / raw)
To: Leo Yan
Cc: lirongqing, Peter Zijlstra, Ingo Molnar, Namhyung Kim,
Mark Rutland, Alexander Shishkin, Jiri Olsa, Ian Rogers,
Adrian Hunter, James Clark, Sergey Senozhatsky, Leo Yan,
linux-perf-users, linux-kernel
On Wed, Nov 04, 2026 at 11:17:51AM +0000, Leo Yan wrote:
> On Sat, Oct 31, 2026 at 08:01:23PM +0800, lirongqing wrote:
> > From: Li RongQing <lirongqing@baidu.com>
> >
> > Use kvm_exit as the start event and the following kvm_entry as the
> > end event for arm64 VM-exit statistics. Keep decoding ret and esr_ec
> > from the exit sample.
> >
> > Before this change, arm64 measured guest runtime as VM-exit time.
> >
> > Fixes: 1218838d68f5 ("perf kvm: Add kvm-stat for arm64")
> > Signed-off-by: Li RongQing <lirongqing@baidu.com>
>
> Thanks for the fix!
>
> Reviewed-by: Leo Yan <leo.yan@arm.com>
>
> Since arm64 requires the exit reason, we cannot directly use the common
> exit_event_begin()/exit_event_end() callbacks as other archs do.
Thanks, applied for v7.4.
- Arnaldo
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2026-11-04 21:19 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-10-31 12:01 [PATCH] perf kvm: Fix arm64 VM-exit duration accounting lirongqing
2026-10-31 12:05 ` sashiko-bot
2026-11-04 11:17 ` Leo Yan
2026-11-04 21:19 ` Arnaldo Carvalho de Melo
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.