Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1330629 > unrolled thread
| Started by | Thomas Gleixner <tglx@linutronix.de> |
|---|---|
| First post | 2016-02-09 21:20 +0100 |
| Last post | 2016-02-17 13:30 +0100 |
| Articles | 5 — 2 participants |
Back to article view | Back to linux.kernel
[patch 0/4] perf: Fix various cpu hotplug oddities Thomas Gleixner <tglx@linutronix.de> - 2016-02-09 21:20 +0100
[patch 2/4] perf: Remove the bogus and dangerous CPU_DOWN_FAILED hotplug state Thomas Gleixner <tglx@linutronix.de> - 2016-02-09 21:20 +0100
[tip:perf/core] perf/core: Remove the bogus and dangerous CPU_DOWN_FAILED hotplug state tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-02-17 13:30 +0100
[patch 1/4] perf: Remove bogus UP_CANCELED hotplug state Thomas Gleixner <tglx@linutronix.de> - 2016-02-09 21:20 +0100
[tip:perf/core] perf/core: Remove bogus UP_CANCELED hotplug state tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-02-17 13:30 +0100
| From | Thomas Gleixner <tglx@linutronix.de> |
|---|---|
| Date | 2016-02-09 21:20 +0100 |
| Subject | [patch 0/4] perf: Fix various cpu hotplug oddities |
| Message-ID | <r0mVr-wW-7@gated-at.bofh.it> |
The following series fixes a few oddities in the core perf code related to cpu hotplug namely memory leaks and pointless smp function calls to offline cpus. Thanks, tglx
[toc] | [next] | [standalone]
| From | Thomas Gleixner <tglx@linutronix.de> |
|---|---|
| Date | 2016-02-09 21:20 +0100 |
| Subject | [patch 2/4] perf: Remove the bogus and dangerous CPU_DOWN_FAILED hotplug state |
| Message-ID | <r0mVt-wW-27@gated-at.bofh.it> |
| In reply to | #1330629 |
If CPU_DOWN_PREPARE fails the perf hotplug notifier is called for
CPU_DOWN_FAILED and calls perf_event_init_cpu(), which checks whether the
swhash is referenced. If yes it allocates a new hash and stores the pointer in
the per cpu data structure.
But at this point the cpu is still online, so there must be a valid hash
already. By overwriting the pointer the existing hash is not longer
accessible.
Remove the CPU_DOWN_FAILED state, as there is nothing to (re)allocate.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/events/core.c | 1 -
1 file changed, 1 deletion(-)
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -9282,7 +9282,6 @@ perf_cpu_notify(struct notifier_block *s
switch (action & ~CPU_TASKS_FROZEN) {
case CPU_UP_PREPARE:
- case CPU_DOWN_FAILED:
perf_event_init_cpu(cpu);
break;
[toc] | [prev] | [next] | [standalone]
| From | tip-bot for Thomas Gleixner <tipbot@zytor.com> |
|---|---|
| Date | 2016-02-17 13:30 +0100 |
| Subject | [tip:perf/core] perf/core: Remove the bogus and dangerous CPU_DOWN_FAILED hotplug state |
| Message-ID | <r39p0-6nX-21@gated-at.bofh.it> |
| In reply to | #1330631 |
Commit-ID: 27ca9236c96f4a21b72a2b4f08260efeab951bd0
Gitweb: http://git.kernel.org/tip/27ca9236c96f4a21b72a2b4f08260efeab951bd0
Author: Thomas Gleixner <tglx@linutronix.de>
AuthorDate: Tue, 9 Feb 2016 20:11:26 +0000
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Wed, 17 Feb 2016 10:37:29 +0100
perf/core: Remove the bogus and dangerous CPU_DOWN_FAILED hotplug state
If CPU_DOWN_PREPARE fails the perf hotplug notifier is called for
CPU_DOWN_FAILED and calls perf_event_init_cpu(), which checks whether the
swhash is referenced. If yes it allocates a new hash and stores the pointer in
the per cpu data structure.
But at this point the cpu is still online, so there must be a valid hash
already. By overwriting the pointer the existing hash is not longer
accessible.
Remove the CPU_DOWN_FAILED state, as there is nothing to (re)allocate.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Link: http://lkml.kernel.org/r/20160209201007.763417379@linutronix.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
kernel/events/core.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 474ffea..4aa64a8 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -9282,7 +9282,6 @@ perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu)
switch (action & ~CPU_TASKS_FROZEN) {
case CPU_UP_PREPARE:
- case CPU_DOWN_FAILED:
perf_event_init_cpu(cpu);
break;
[toc] | [prev] | [next] | [standalone]
| From | Thomas Gleixner <tglx@linutronix.de> |
|---|---|
| Date | 2016-02-09 21:20 +0100 |
| Subject | [patch 1/4] perf: Remove bogus UP_CANCELED hotplug state |
| Message-ID | <r0mVt-wW-37@gated-at.bofh.it> |
| In reply to | #1330629 |
If CPU_UP_PREPARE fails the perf hotplug code calls perf_event_exit_cpu(), which is a pointless exercise. The cpu is not online, so the smp function calls return -ENXIO. So the result is a list walk to call noops. Remove it. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> --- kernel/events/core.c | 1 - 1 file changed, 1 deletion(-) --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -9286,7 +9286,6 @@ perf_cpu_notify(struct notifier_block *s perf_event_init_cpu(cpu); break; - case CPU_UP_CANCELED: case CPU_DOWN_PREPARE: perf_event_exit_cpu(cpu); break;
[toc] | [prev] | [next] | [standalone]
| From | tip-bot for Thomas Gleixner <tipbot@zytor.com> |
|---|---|
| Date | 2016-02-17 13:30 +0100 |
| Subject | [tip:perf/core] perf/core: Remove bogus UP_CANCELED hotplug state |
| Message-ID | <r39p0-6nX-25@gated-at.bofh.it> |
| In reply to | #1330634 |
Commit-ID: b4f75d44bed1bdbb14ac704bfc38f62a3675e591 Gitweb: http://git.kernel.org/tip/b4f75d44bed1bdbb14ac704bfc38f62a3675e591 Author: Thomas Gleixner <tglx@linutronix.de> AuthorDate: Tue, 9 Feb 2016 20:11:20 +0000 Committer: Ingo Molnar <mingo@kernel.org> CommitDate: Wed, 17 Feb 2016 10:37:28 +0100 perf/core: Remove bogus UP_CANCELED hotplug state If CPU_UP_PREPARE fails the perf hotplug code calls perf_event_exit_cpu(), which is a pointless exercise. The cpu is not online, so the smp function calls return -ENXIO. So the result is a list walk to call noops. Remove it. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Cc: Vince Weaver <vincent.weaver@maine.edu> Link: http://lkml.kernel.org/r/20160209201007.682184765@linutronix.de Signed-off-by: Ingo Molnar <mingo@kernel.org> --- kernel/events/core.c | 1 - 1 file changed, 1 deletion(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index 5946460..474ffea 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -9286,7 +9286,6 @@ perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) perf_event_init_cpu(cpu); break; - case CPU_UP_CANCELED: case CPU_DOWN_PREPARE: perf_event_exit_cpu(cpu); break;
[toc] | [prev] | [standalone]
Back to top | Article view | linux.kernel
csiph-web