Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1695106 > unrolled thread

[PATCH] perf: qcom_l2: fix column exclusion check

Started byNeil Leeder <nleeder@codeaurora.org>
First post2017-07-24 23:20 +0200
Last post2017-07-26 10:30 +0200
Articles 4 — 4 participants

Back to article view | Back to linux.kernel


Contents

  [PATCH] perf: qcom_l2: fix column exclusion check Neil Leeder <nleeder@codeaurora.org> - 2017-07-24 23:20 +0200
    Re: [PATCH] perf: qcom_l2: fix column exclusion check Mark Rutland <mark.rutland@arm.com> - 2017-07-25 19:10 +0200
      Re: [PATCH] perf: qcom_l2: fix column exclusion check "Leeder, Neil" <nleeder@codeaurora.org> - 2017-07-25 21:50 +0200
        Re: [PATCH] perf: qcom_l2: fix column exclusion check Will Deacon <will.deacon@arm.com> - 2017-07-26 10:30 +0200

#1695106 — [PATCH] perf: qcom_l2: fix column exclusion check

FromNeil Leeder <nleeder@codeaurora.org>
Date2017-07-24 23:20 +0200
Subject[PATCH] perf: qcom_l2: fix column exclusion check
Message-ID<u6SVH-227-1@gated-at.bofh.it>
The check for column exclusion did not verify that the event being
checked was an L2 event, and not a software event.
Software events should not be checked for column exclusion.
This resulted in a group with both software and L2 events sometimes
incorrectly rejecting the L2 event for column exclusion and
not counting it.

Add a check for PMU type before applying column exclusion logic.

Signed-off-by: Neil Leeder <nleeder@codeaurora.org>
---
 drivers/perf/qcom_l2_pmu.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/perf/qcom_l2_pmu.c b/drivers/perf/qcom_l2_pmu.c
index c259848..b242cce 100644
--- a/drivers/perf/qcom_l2_pmu.c
+++ b/drivers/perf/qcom_l2_pmu.c
@@ -546,6 +546,7 @@ static int l2_cache_event_init(struct perf_event *event)
 	}
 
 	if ((event != event->group_leader) &&
+	    !is_software_event(event->group_leader) &&
 	    (L2_EVT_GROUP(event->group_leader->attr.config) ==
 	     L2_EVT_GROUP(event->attr.config))) {
 		dev_dbg_ratelimited(&l2cache_pmu->pdev->dev,
@@ -558,6 +559,7 @@ static int l2_cache_event_init(struct perf_event *event)
 	list_for_each_entry(sibling, &event->group_leader->sibling_list,
 			    group_entry) {
 		if ((sibling != event) &&
+		    !is_software_event(sibling) &&
 		    (L2_EVT_GROUP(sibling->attr.config) ==
 		     L2_EVT_GROUP(event->attr.config))) {
 			dev_dbg_ratelimited(&l2cache_pmu->pdev->dev,
-- 
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project.

[toc] | [next] | [standalone]


#1695955

FromMark Rutland <mark.rutland@arm.com>
Date2017-07-25 19:10 +0200
Message-ID<u7bvk-5B8-23@gated-at.bofh.it>
In reply to#1695106
On Mon, Jul 24, 2017 at 05:17:02PM -0400, Neil Leeder wrote:
> The check for column exclusion did not verify that the event being
> checked was an L2 event, and not a software event.
> Software events should not be checked for column exclusion.
> This resulted in a group with both software and L2 events sometimes
> incorrectly rejecting the L2 event for column exclusion and
> not counting it.
> 
> Add a check for PMU type before applying column exclusion logic.
> 
> Signed-off-by: Neil Leeder <nleeder@codeaurora.org>

This looks correct, so:

Acked-by: Mark Rutland <mark.rutland@arm.com>

Should this have:

Fixes: 21bdbb7102edeaeb ("perf: add qcom l2 cache perf events driver")

... ?

> ---
>  drivers/perf/qcom_l2_pmu.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/perf/qcom_l2_pmu.c b/drivers/perf/qcom_l2_pmu.c
> index c259848..b242cce 100644
> --- a/drivers/perf/qcom_l2_pmu.c
> +++ b/drivers/perf/qcom_l2_pmu.c
> @@ -546,6 +546,7 @@ static int l2_cache_event_init(struct perf_event *event)
>  	}
>  
>  	if ((event != event->group_leader) &&
> +	    !is_software_event(event->group_leader) &&
>  	    (L2_EVT_GROUP(event->group_leader->attr.config) ==
>  	     L2_EVT_GROUP(event->attr.config))) {
>  		dev_dbg_ratelimited(&l2cache_pmu->pdev->dev,
> @@ -558,6 +559,7 @@ static int l2_cache_event_init(struct perf_event *event)
>  	list_for_each_entry(sibling, &event->group_leader->sibling_list,
>  			    group_entry) {
>  		if ((sibling != event) &&
> +		    !is_software_event(sibling) &&
>  		    (L2_EVT_GROUP(sibling->attr.config) ==
>  		     L2_EVT_GROUP(event->attr.config))) {
>  			dev_dbg_ratelimited(&l2cache_pmu->pdev->dev,

It's unfortunate that we duplicate the checks for the leader and
siblings, but that's not a new problem, and we can fix that in a
follow-up patch.

Thanks,
Mark.

[toc] | [prev] | [next] | [standalone]


#1696167

From"Leeder, Neil" <nleeder@codeaurora.org>
Date2017-07-25 21:50 +0200
Message-ID<u7e0b-701-37@gated-at.bofh.it>
In reply to#1695955
On 7/25/2017 1:01 PM, Mark Rutland wrote:
> On Mon, Jul 24, 2017 at 05:17:02PM -0400, Neil Leeder wrote:
>> The check for column exclusion did not verify that the event being
>> checked was an L2 event, and not a software event.
>> Software events should not be checked for column exclusion.
>> This resulted in a group with both software and L2 events sometimes
>> incorrectly rejecting the L2 event for column exclusion and
>> not counting it.
>>
>> Add a check for PMU type before applying column exclusion logic.
>>
>> Signed-off-by: Neil Leeder <nleeder@codeaurora.org>
> 
> This looks correct, so:
> 
> Acked-by: Mark Rutland <mark.rutland@arm.com>
> 
> Should this have:
> 
> Fixes: 21bdbb7102edeaeb ("perf: add qcom l2 cache perf events driver")
> 
> ... ?
> 
Thanks. I'll re-post with the Fixes tag.

Neil
-- 
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project.

[toc] | [prev] | [next] | [standalone]


#1696876

FromWill Deacon <will.deacon@arm.com>
Date2017-07-26 10:30 +0200
Message-ID<u7pRF-6lO-25@gated-at.bofh.it>
In reply to#1696167
On Tue, Jul 25, 2017 at 03:43:54PM -0400, Leeder, Neil wrote:
> On 7/25/2017 1:01 PM, Mark Rutland wrote:
> > On Mon, Jul 24, 2017 at 05:17:02PM -0400, Neil Leeder wrote:
> >> The check for column exclusion did not verify that the event being
> >> checked was an L2 event, and not a software event.
> >> Software events should not be checked for column exclusion.
> >> This resulted in a group with both software and L2 events sometimes
> >> incorrectly rejecting the L2 event for column exclusion and
> >> not counting it.
> >>
> >> Add a check for PMU type before applying column exclusion logic.
> >>
> >> Signed-off-by: Neil Leeder <nleeder@codeaurora.org>
> > 
> > This looks correct, so:
> > 
> > Acked-by: Mark Rutland <mark.rutland@arm.com>
> > 
> > Should this have:
> > 
> > Fixes: 21bdbb7102edeaeb ("perf: add qcom l2 cache perf events driver")
> > 
> > ... ?
> > 
> Thanks. I'll re-post with the Fixes tag.

No need, I'll pick it up and add it when I apply.

Cheers,

Will

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web