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

Groups > linux.kernel > #1375001 > unrolled thread

[PATCH 3.14 75/76] perf/x86/intel: Use PAGE_SIZE for PEBS buffer size on Core2

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  [PATCH 3.14 75/76] perf/x86/intel: Use PAGE_SIZE for PEBS buffer size on Core2 Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-04-10 21:30 +0200
    Re: [PATCH 3.14 75/76] perf/x86/intel: Use PAGE_SIZE for PEBS buffer  size on Core2 Jiri Slaby <jslaby@suse.cz> - 2016-04-11 09:10 +0200
      Re: [PATCH 3.14 75/76] perf/x86/intel: Use PAGE_SIZE for PEBS buffer  size on Core2 Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-04-12 16:20 +0200

#1375001 — [PATCH 3.14 75/76] perf/x86/intel: Use PAGE_SIZE for PEBS buffer size on Core2

3.14-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Jiri Olsa <jolsa@redhat.com>

commit 3135a66b768c5ee84c8a98b21d0330dc1c1234b4 upstream.

Using PAGE_SIZE buffers makes the WRMSR to PERF_GLOBAL_CTRL in
intel_pmu_enable_all() mysteriously hang on Core2. As a workaround, we
don't do this.

The hard lockup is easily triggered by running 'perf test attr'
repeatedly. Most of the time it gets stuck on sample session with
small periods.

  # perf test attr -vv
  14: struct perf_event_attr setup                             :
  --- start ---
  ...
    'PERF_TEST_ATTR=/tmp/tmpuEKz3B /usr/bin/perf record -o /tmp/tmpuEKz3B/perf.data -c 123 kill >/dev/null 2>&1' ret 1

Reported-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/20160301190352.GA8355@krava.redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 arch/x86/kernel/cpu/perf_event.h          |    1 +
 arch/x86/kernel/cpu/perf_event_intel_ds.c |   13 +++++++++++--
 2 files changed, 12 insertions(+), 2 deletions(-)

--- a/arch/x86/kernel/cpu/perf_event.h
+++ b/arch/x86/kernel/cpu/perf_event.h
@@ -445,6 +445,7 @@ struct x86_pmu {
 			pebs_active	:1,
 			pebs_broken	:1;
 	int		pebs_record_size;
+	int		pebs_buffer_size;
 	void		(*drain_pebs)(struct pt_regs *regs);
 	struct event_constraint *pebs_constraints;
 	void		(*pebs_aliases)(struct perf_event *event);
--- a/arch/x86/kernel/cpu/perf_event_intel_ds.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c
@@ -241,7 +241,7 @@ static int alloc_pebs_buffer(int cpu)
 	if (!x86_pmu.pebs)
 		return 0;
 
-	buffer = kzalloc_node(PEBS_BUFFER_SIZE, GFP_KERNEL, node);
+	buffer = kzalloc_node(x86_pmu.pebs_buffer_size, GFP_KERNEL, node);
 	if (unlikely(!buffer))
 		return -ENOMEM;
 
@@ -258,7 +258,7 @@ static int alloc_pebs_buffer(int cpu)
 		per_cpu(insn_buffer, cpu) = ibuffer;
 	}
 
-	max = PEBS_BUFFER_SIZE / x86_pmu.pebs_record_size;
+	max = x86_pmu.pebs_buffer_size / x86_pmu.pebs_record_size;
 
 	ds->pebs_buffer_base = (u64)(unsigned long)buffer;
 	ds->pebs_index = ds->pebs_buffer_base;
@@ -1047,6 +1047,7 @@ void intel_ds_init(void)
 
 	x86_pmu.bts  = boot_cpu_has(X86_FEATURE_BTS);
 	x86_pmu.pebs = boot_cpu_has(X86_FEATURE_PEBS);
+	x86_pmu.pebs_buffer_size = PEBS_BUFFER_SIZE;
 	if (x86_pmu.pebs) {
 		char pebs_type = x86_pmu.intel_cap.pebs_trap ?  '+' : '-';
 		int format = x86_pmu.intel_cap.pebs_format;
@@ -1055,6 +1056,14 @@ void intel_ds_init(void)
 		case 0:
 			printk(KERN_CONT "PEBS fmt0%c, ", pebs_type);
 			x86_pmu.pebs_record_size = sizeof(struct pebs_record_core);
+			/*
+			 * Using >PAGE_SIZE buffers makes the WRMSR to
+			 * PERF_GLOBAL_CTRL in intel_pmu_enable_all()
+			 * mysteriously hang on Core2.
+			 *
+			 * As a workaround, we don't do this.
+			 */
+			x86_pmu.pebs_buffer_size = PAGE_SIZE;
 			x86_pmu.drain_pebs = intel_pmu_drain_pebs_core;
 			break;
 

[toc] | [next] | [standalone]

#1375530 — Re: [PATCH 3.14 75/76] perf/x86/intel: Use PAGE_SIZE for PEBS buffer size on Core2

On 04/10/2016, 08:37 PM, Greg Kroah-Hartman wrote:
> 3.14-stable review patch.  If anyone has any objections, please let me know.
> 
> ------------------
> 
> From: Jiri Olsa <jolsa@redhat.com>
> 
> commit 3135a66b768c5ee84c8a98b21d0330dc1c1234b4 upstream.

Where does this come from?

fatal: bad object 3135a66b768c5ee84c8a98b21d0330dc1c1234b4

The same for the later perf patch.

> Using PAGE_SIZE buffers makes the WRMSR to PERF_GLOBAL_CTRL in
> intel_pmu_enable_all() mysteriously hang on Core2. As a workaround, we
> don't do this.
> 
> The hard lockup is easily triggered by running 'perf test attr'
> repeatedly. Most of the time it gets stuck on sample session with
> small periods.
> 
>   # perf test attr -vv
>   14: struct perf_event_attr setup                             :
>   --- start ---
>   ...
>     'PERF_TEST_ATTR=/tmp/tmpuEKz3B /usr/bin/perf record -o /tmp/tmpuEKz3B/perf.data -c 123 kill >/dev/null 2>&1' ret 1
> 
> Reported-by: Arnaldo Carvalho de Melo <acme@redhat.com>
> Signed-off-by: Jiri Olsa <jolsa@kernel.org>
> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> Reviewed-by: Andi Kleen <ak@linux.intel.com>
> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
> Cc: Jiri Olsa <jolsa@redhat.com>
> Cc: Kan Liang <kan.liang@intel.com>
> Cc: Linus Torvalds <torvalds@linux-foundation.org>
> Cc: Peter Zijlstra <peterz@infradead.org>
> Cc: Stephane Eranian <eranian@google.com>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Vince Weaver <vincent.weaver@maine.edu>
> Cc: Wang Nan <wangnan0@huawei.com>
> Link: http://lkml.kernel.org/r/20160301190352.GA8355@krava.redhat.com
> Signed-off-by: Ingo Molnar <mingo@kernel.org>
> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> ---
>  arch/x86/kernel/cpu/perf_event.h          |    1 +
>  arch/x86/kernel/cpu/perf_event_intel_ds.c |   13 +++++++++++--
>  2 files changed, 12 insertions(+), 2 deletions(-)

thanks,
-- 
js
suse labs

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

#1376841 — Re: [PATCH 3.14 75/76] perf/x86/intel: Use PAGE_SIZE for PEBS buffer size on Core2

On Mon, Apr 11, 2016 at 09:01:33AM +0200, Jiri Slaby wrote:
> On 04/10/2016, 08:37 PM, Greg Kroah-Hartman wrote:
> > 3.14-stable review patch.  If anyone has any objections, please let me know.
> > 
> > ------------------
> > 
> > From: Jiri Olsa <jolsa@redhat.com>
> > 
> > commit 3135a66b768c5ee84c8a98b21d0330dc1c1234b4 upstream.
> 
> Where does this come from?
> 
> fatal: bad object 3135a66b768c5ee84c8a98b21d0330dc1c1234b4
> 
> The same for the later perf patch.

Crap, I got these all wrong, my fault, I'll go edit them and put the
correct id in there.

greg k-h

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web