lkml.org 
[lkml]   [2010]   [Nov]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [PATCH 1/2] perf-events: Add support for supplementary event registers v2
From
On Mon, Nov 15, 2010 at 12:06 PM, Peter Zijlstra <a.p.zijlstra@chello.nl> wrote:
> On Mon, 2010-11-15 at 12:01 +0100, Stephane Eranian wrote:
>> On Sat, Nov 13, 2010 at 11:32 AM, Peter Zijlstra <a.p.zijlstra@chello.nl> wrote:
>> > On Sat, 2010-11-13 at 11:13 +0100, Andi Kleen wrote:
>> >> > Hrmm,. this doesn't really scale right wrt multiple extra_regs.
>> >> >
>> >> > You made the extra_regs thing fairly extensible, but the above doesn't
>> >> > really work well if there's multiple extra regs (say OFFCORE and
>> >> > LBR_CONFIG -- possibly even the two OFFCORE regs present on westmere).
>> >> >
>> >> > It basically needs a per-core state for each extra_reg possible.
>> >>
>> >> It shouldn't be too hard to extend it to use a table. I didn't want
>> >> to do that for the first iteration. I guess it could be done as followons
>> >> once LBR is implemented.
>> >
>> > Isn't it also relevant for the two offcore regs for wsm?
>> >
>> But that's two distinct events (0xbb and 0xb7), though they behave
>> exactly the same and have the same set of umasks and extra MSR
>> values.
>
> Ah, my bad, I thought there were two separate msrs, ok I'll take a last
> look at these patches and if nothing else pops out I'll take them.
>
OFFCORE_RESPONSE_0, event code 0xb7 -> extra MSR@0x1a6
OFFCORE_REPPONSE_1, event code 0xbb -> extra MSR@0x1a7

Why would you want to measure both at the same time?
Because you can measure two distinct request/response in one run.

If you want to measure both you need to create two distinct perf events.


\
 
 \ /
  Last update: 2010-11-15 12:21    [W:0.043 / U:0.204 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site