Re: [PATCH v1 5/5] coresight: etm4x: save/restore state across CPU low power states

From: Mathieu Poirier <mathieu.poirier@linaro.org>
To: Sudeep Holla <sudeep.holla@arm.com>
Cc: Al Grant <Al.Grant@arm.com>,
	Suzuki K Poulose <suzuki.poulose@arm.com>,
	Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Andrew Murray <andrew.murray@arm.com>,
	linux-arm-kernel <linux-arm-kernel@lists.infradead.org>,
	Mike Leach <mike.leach@linaro.org>
Subject: Re: [PATCH v1 5/5] coresight: etm4x: save/restore state across CPU low power states
Date: Thu, 20 Jun 2019 12:00:06 -0600	[thread overview]
Message-ID: <CANLsYkyBTZHt0CcOWOpkpFMPXh+JXFdU=bAkC9VjpQHF3Xm6Fg@mail.gmail.com> (raw)
In-Reply-To: <20190620171102.GF25273@e107155-lin>

On Thu, 20 Jun 2019 at 11:11, Sudeep Holla <sudeep.holla@arm.com> wrote:
>
> On Thu, Jun 20, 2019 at 06:00:48PM +0100, Suzuki K Poulose wrote:
> >
> >
> > On 20/06/2019 17:54, Andrew Murray wrote:
> > > On Thu, Jun 20, 2019 at 10:47:38AM -0600, Mathieu Poirier wrote:
> > > > On Thu, 20 Jun 2019 at 10:34, Sudeep Holla <sudeep.holla@arm.com> wrote:
> > > > >
> > > > > On Thu, Jun 20, 2019 at 10:14:04AM -0600, Mathieu Poirier wrote:
> > > > > > On Thu, 20 Jun 2019 at 09:41, Sudeep Holla <sudeep.holla@arm.com> wrote:
> >
> > ...
> >
> > > > Then all we have to do is make the ACPI/DT property that indicate the
> > > > method used to deal with tracer idling mandatory.  That way people are
> > > > conscious of the choice they are making.  To be backward compatible
> > > > with current systems we default to the TRCPDCR.PU method but print a
> > > > warning message, just like we do for obsolete DT bindings.
> > >
> > > I'll respin the series based on this approach. I'll also flip the
> > > 'disable_pm_save' module option to 'enable_pm_save' - thus allowing any
> > > one to use software save/restore if they wish.
> >
> > If you are going to add a firmware property, please get a consensus on the
> > name here, before respinning to avoid another churn :-). How about one of :
> >
> > "arm,coresight-etm-looses-state"
> > "arm,coresight-etm-needs-save-restore"
> >
>
> Just to be more clear, I am fine with just kernel command/module parameter
> approach and DT bindings may not be required. If at all it is decided to
> take DT approach, then you really don't need command/module parameter IMO.
> I will leave that to you and Mathieu, wanted to make sure I am not
> contributing to the confusion yet again.

Both DT/ACPI and module parameters are needed.

The module parameter is there to overwrite the default setting for
debugging purposes when using the notifiers method and assumes cores
have been prevented from idling - otherwise the system will crash.
The DT/ACPI property is needed to tell coresight which method to use,
i.e architected of notifiers, to avoid latency issues when possible.

Thanks,
Mathieu

>
> --
> Regards,
> Sudeep

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel