Re: BUG in "perf: Suppress AUX/OVERWRITE records"?

From: Will Deacon
Date: Tue Apr 02 2019 - 06:47:19 EST


On Wed, Mar 27, 2019 at 11:17:43AM +0200, Alexander Shishkin wrote:
> Ben Gainey <Ben.Gainey@xxxxxxx> writes:
>
> >> It was an unintentional side effect that it also
> >> happened to coincide with context switches in the overwrite mode.
> >
> > I'm not using overwrite mode, I'm opening the mmap with PROT_WRITE
> > (i.e. in truncate mode).
>
> Now I get it. Does the below fix the problem for you?
>
> From bf52320cce0e74a2c0d987db7bd571f7687b4f4f Mon Sep 17 00:00:00 2001
> From: Alexander Shishkin <alexander.shishkin@xxxxxxxxxxxxxxx>
> Date: Wed, 27 Mar 2019 11:05:53 +0200
> Subject: [PATCH] perf: Fix AUX record suppression
>
> Commit 1627314fb54a33e ("perf: Suppress AUX/OVERWRITE records") has an
> unintended side-effect of also suppressing all AUX records with no flags
> and non-zero size, so all the regular records in the full trace mode.
> This breaks some use cases for people.
>
> Fix this by restoring "regular" AUX records to their former glory.
>
> Signed-off-by: Alexander Shishkin <alexander.shishkin@xxxxxxxxxxxxxxx>
> Fixes: 1627314fb54a33e ("perf: Suppress AUX/OVERWRITE records")
> Reported-by: Ben Gainey <Ben.Gainey@xxxxxxx>
> CC: stable@xxxxxxxxxxxxxxx # v4.20+
> ---
> kernel/events/ring_buffer.c | 33 +++++++++++++++------------------
> 1 file changed, 15 insertions(+), 18 deletions(-)
>
> diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c
> index 678ccec60d8f..626256dc26c1 100644
> --- a/kernel/events/ring_buffer.c
> +++ b/kernel/events/ring_buffer.c
> @@ -455,24 +455,21 @@ void perf_aux_output_end(struct perf_output_handle *handle, unsigned long size)
> rb->aux_head += size;
> }
>
> - if (size || handle->aux_flags) {
> - /*
> - * Only send RECORD_AUX if we have something useful to communicate
> - *
> - * Note: the OVERWRITE records by themselves are not considered
> - * useful, as they don't communicate any *new* information,
> - * aside from the short-lived offset, that becomes history at
> - * the next event sched-in and therefore isn't useful.
> - * The userspace that needs to copy out AUX data in overwrite
> - * mode should know to use user_page::aux_head for the actual
> - * offset. So, from now on we don't output AUX records that
> - * have *only* OVERWRITE flag set.
> - */
> -
> - if (handle->aux_flags & ~(u64)PERF_AUX_FLAG_OVERWRITE)
> - perf_event_aux_event(handle->event, aux_head, size,
> - handle->aux_flags);
> - }
> + /*
> + * Only send RECORD_AUX if we have something useful to communicate
> + *
> + * Note: the OVERWRITE records by themselves are not considered
> + * useful, as they don't communicate any *new* information,
> + * aside from the short-lived offset, that becomes history at
> + * the next event sched-in and therefore isn't useful.
> + * The userspace that needs to copy out AUX data in overwrite
> + * mode should know to use user_page::aux_head for the actual
> + * offset. So, from now on we don't output AUX records that
> + * have *only* OVERWRITE flag set.
> + */
> + if (size || (handle->aux_flags & ~(u64)PERF_AUX_FLAG_OVERWRITE))
> + perf_event_aux_event(handle->event, aux_head, size,
> + handle->aux_flags);

Acked-by: Will Deacon <will.deacon@xxxxxxx>

Assuming this will make it into 5.1 as a fix?

Will