lkml.org 
[lkml]   [2015]   [Feb]   [11]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [PATCH 2/2] ARM: perf: Add support for Scorpion PMUs
From
On 11 February 2015 at 13:27, Stephen Boyd <sboyd@codeaurora.org> wrote:
> On 02/10, Ashwin Chaugule wrote:
>> Hi Stephen,
>>
>> On 10 February 2015 at 20:05, Stephen Boyd <sboyd@codeaurora.org> wrote:
>> > Scorpion supports a set of local performance monitor event
>> > selection registers (LPM) sitting behind a cp15 based interface
>> > that extend the architected PMU events to include Scorpion CPU
>> > and Venum VFP specific events. To use these events the user is
>> > expected to program the lpm register with the event code shifted
>> > into the group they care about and then point the PMNx event at
>> > that region+group combo by writing a LPMn_GROUPx event. Add
>> > support for this hardware.
>> >
>> > Note: the raw event number is a pure software construct that
>> > allows us to map the multi-dimensional number space of regions,
>> > groups, and event codes into a flat event number space suitable
>> > for use by the perf framework.
>> >
>> > This is based on code originally written by Ashwin Chaugule and
>> > Neil Leeder [1] massed to become similar to the Krait PMU support
>> > code.
>>
>> Thanks for taking this up!
>> Overall this series looks good to me, but from what I faintly
>> recollect, doesn't this (and the Krait pmu code) get affected by
>> powercollapse issues anymore?
>> e.g.
>> https://www.codeaurora.org/cgit/quic/la/kernel/msm/commit/arch/arm/kernel/perf_event_msm.c?h=msm-3.4&id=b5ca687960f0fea2f4735e83ca5c9543474c19de
>>
>
> Right now there isn't any power collapse support in mainline so
> there's no immediate problem. Once we add power collapse support
> (i.e. cpuidle) to the Scorpion and Krait platforms we'll need to
> do something in the perf event code to properly maintain the
> counts across idle. I imagine it would be done by registering for
> cpu_pm notifications and then doing the save/restore on
> CPU_PM_ENTER and CPU_PM_EXIT. At least, that's what you started
> doing in this patch[1]. And then it seems the patch you mention
> came after that and actually did the save/restore of the counts.
>
> [1] https://www.codeaurora.org/cgit/quic/la/kernel/msm/commit/?h=msm-3.4&id=464983a7e991a484cac0bc0885cee4fee318d659

Right. Thats essential whenever the power collapse stuff goes in.

Thanks,
Ashwin.


\
 
 \ /
  Last update: 2015-02-12 03:21    [W:0.047 / U:1.808 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site