[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <5572cf25-b89b-1044-f3db-29bb2100be41@arm.com>
Date: Wed, 10 May 2023 07:59:37 +0100
From: Lukasz Luba <lukasz.luba@....com>
To: Pierre Gondois <pierre.gondois@....com>
Cc: dietmar.eggemann@....com, rui.zhang@...el.com,
linux-pm@...r.kernel.org, amit.kucheria@...durent.com,
amit.kachhap@...il.com, daniel.lezcano@...aro.org,
viresh.kumar@...aro.org, len.brown@...el.com, pavel@....cz,
rafael@...nel.org, ionela.voinescu@....com,
linux-kernel@...r.kernel.org, rostedt@...dmis.org,
mhiramat@...nel.org
Subject: Re: [PATCH 05/17] trace: energy_model: Add trace event for EM runtime
modifications
On 4/11/23 16:39, Pierre Gondois wrote:
> Hello Lukasz,
> Just a suggestion, maybe it would be good to trace the CPUs affected by
We print the device name in the trace event, which would be the first
CPU in the frequency domain.
> the modification. It is possible to retrieve this information by going
> to /sys/kernel/debug/energy_model/xxx/cpus, but might be simpler when
> parsing a trace.
We don't want to introduce an array of CPUs, or a name like cpus0-3,
which than has to be parsed in more complex way. Also, the other devices
like GPU might use this trace event, so I just wanted to keep it simple
and generic to handle them all.
Powered by blists - more mailing lists