[PATCH 01/31] perf, x86: Add PEBSv2 record support

2012-10-02 Thread Andi Kleen
From: Andi Kleen Add support for the v2 PEBS format. It has a superset of the v1 PEBS fields, but has a longer record so we need to adjust the code paths. The main advantage is the new "EventingRip" support which directly gives the instruction, not off-by-one instruction. So with precise == 2

[PATCH 01/31] perf, x86: Add PEBSv2 record support

2012-10-02 Thread Andi Kleen
From: Andi Kleen a...@linux.intel.com Add support for the v2 PEBS format. It has a superset of the v1 PEBS fields, but has a longer record so we need to adjust the code paths. The main advantage is the new EventingRip support which directly gives the instruction, not off-by-one instruction. So

Re: [PATCH 01/31] perf, x86: Add PEBSv2 record support

2012-09-28 Thread Andi Kleen
On Fri, Sep 28, 2012 at 10:43:04AM +0200, Peter Zijlstra wrote: > On Thu, 2012-09-27 at 21:31 -0700, Andi Kleen wrote: > > + if (event->attr.precise_ip > 1 && > > x86_pmu.intel_cap.pebs_format < 2) { > > Shouldn't that be: && x86_pmu.intel_cap.pebs_trap, like most other sites >

Re: [PATCH 01/31] perf, x86: Add PEBSv2 record support

2012-09-28 Thread Stephane Eranian
On Fri, Sep 28, 2012 at 11:28 AM, Peter Zijlstra wrote: > On Fri, 2012-09-28 at 10:54 +0200, Stephane Eranian wrote: >> On Fri, Sep 28, 2012 at 10:43 AM, Peter Zijlstra >> wrote: >> > On Thu, 2012-09-27 at 21:31 -0700, Andi Kleen wrote: >> >> + if (event->attr.precise_ip > 1 &&

Re: [PATCH 01/31] perf, x86: Add PEBSv2 record support

2012-09-28 Thread Peter Zijlstra
On Fri, 2012-09-28 at 10:54 +0200, Stephane Eranian wrote: > On Fri, Sep 28, 2012 at 10:43 AM, Peter Zijlstra > wrote: > > On Thu, 2012-09-27 at 21:31 -0700, Andi Kleen wrote: > >> + if (event->attr.precise_ip > 1 && > >> x86_pmu.intel_cap.pebs_format < 2) { > > > > Shouldn't that

Re: [PATCH 01/31] perf, x86: Add PEBSv2 record support

2012-09-28 Thread Stephane Eranian
On Fri, Sep 28, 2012 at 10:43 AM, Peter Zijlstra wrote: > On Thu, 2012-09-27 at 21:31 -0700, Andi Kleen wrote: >> + if (event->attr.precise_ip > 1 && >> x86_pmu.intel_cap.pebs_format < 2) { > > Shouldn't that be: && x86_pmu.intel_cap.pebs_trap, like most other sites > instead? Or

Re: [PATCH 01/31] perf, x86: Add PEBSv2 record support

2012-09-28 Thread Peter Zijlstra
On Thu, 2012-09-27 at 21:31 -0700, Andi Kleen wrote: > + if (event->attr.precise_ip > 1 && > x86_pmu.intel_cap.pebs_format < 2) { Shouldn't that be: && x86_pmu.intel_cap.pebs_trap, like most other sites instead? Or didn't they flip the trap capability on Haswell? -- To unsubscribe

Re: [PATCH 01/31] perf, x86: Add PEBSv2 record support

2012-09-28 Thread Peter Zijlstra
On Thu, 2012-09-27 at 21:31 -0700, Andi Kleen wrote: + if (event-attr.precise_ip 1 x86_pmu.intel_cap.pebs_format 2) { Shouldn't that be: x86_pmu.intel_cap.pebs_trap, like most other sites instead? Or didn't they flip the trap capability on Haswell? -- To unsubscribe from this

Re: [PATCH 01/31] perf, x86: Add PEBSv2 record support

2012-09-28 Thread Stephane Eranian
On Fri, Sep 28, 2012 at 10:43 AM, Peter Zijlstra a.p.zijls...@chello.nl wrote: On Thu, 2012-09-27 at 21:31 -0700, Andi Kleen wrote: + if (event-attr.precise_ip 1 x86_pmu.intel_cap.pebs_format 2) { Shouldn't that be: x86_pmu.intel_cap.pebs_trap, like most other sites

Re: [PATCH 01/31] perf, x86: Add PEBSv2 record support

2012-09-28 Thread Peter Zijlstra
On Fri, 2012-09-28 at 10:54 +0200, Stephane Eranian wrote: On Fri, Sep 28, 2012 at 10:43 AM, Peter Zijlstra a.p.zijls...@chello.nl wrote: On Thu, 2012-09-27 at 21:31 -0700, Andi Kleen wrote: + if (event-attr.precise_ip 1 x86_pmu.intel_cap.pebs_format 2) { Shouldn't

Re: [PATCH 01/31] perf, x86: Add PEBSv2 record support

2012-09-28 Thread Stephane Eranian
On Fri, Sep 28, 2012 at 11:28 AM, Peter Zijlstra a.p.zijls...@chello.nl wrote: On Fri, 2012-09-28 at 10:54 +0200, Stephane Eranian wrote: On Fri, Sep 28, 2012 at 10:43 AM, Peter Zijlstra a.p.zijls...@chello.nl wrote: On Thu, 2012-09-27 at 21:31 -0700, Andi Kleen wrote: + if

Re: [PATCH 01/31] perf, x86: Add PEBSv2 record support

2012-09-28 Thread Andi Kleen
On Fri, Sep 28, 2012 at 10:43:04AM +0200, Peter Zijlstra wrote: On Thu, 2012-09-27 at 21:31 -0700, Andi Kleen wrote: + if (event-attr.precise_ip 1 x86_pmu.intel_cap.pebs_format 2) { Shouldn't that be: x86_pmu.intel_cap.pebs_trap, like most other sites instead? Or didn't

[PATCH 01/31] perf, x86: Add PEBSv2 record support

2012-09-27 Thread Andi Kleen
From: Andi Kleen Add support for the v2 PEBS format. It has a superset of the v1 PEBS fields, but has a longer record so we need to adjust the code paths. The main advantage is the new "EventingRip" support which directly gives the instruction, not off-by-one instruction. So with precise == 2

[PATCH 01/31] perf, x86: Add PEBSv2 record support

2012-09-27 Thread Andi Kleen
From: Andi Kleen a...@linux.intel.com Add support for the v2 PEBS format. It has a superset of the v1 PEBS fields, but has a longer record so we need to adjust the code paths. The main advantage is the new EventingRip support which directly gives the instruction, not off-by-one instruction. So