Re: [PATCH] Revert "arm64: dts: marvell: add CPU Idle power state support on Armada 7K/8K"

From: Sudeep Holla <sudeep.holla@arm.com>
To: Gregory CLEMENT <gregory.clement@bootlin.com>
Cc: Andrew Lunn <andrew@lunn.ch>, Baruch Siach <baruch@tkos.co.il>,
	Jason Cooper <jason@lakedaemon.net>,
	Sergey Matyukevich <geomatsi@gmail.com>,
	linux-pm@vger.kernel.org, orenbh <orenbh@marvell.com>,
	Daniel Lezcano <daniel.lezcano@linaro.org>,
	"Rafael J. Wysocki" <rjw@rjwysocki.net>,
	Russell King <linux@armlinux.org.uk>,
	Sudeep Holla <sudeep.holla@arm.com>,
	linux-arm-kernel@lists.infradead.org,
	Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com>
Subject: Re: [PATCH] Revert "arm64: dts: marvell: add CPU Idle power state support on Armada 7K/8K"
Date: Tue, 18 Dec 2018 17:11:44 +0000	[thread overview]
Message-ID: <20181218171144.GA17290@e107155-lin> (raw)
In-Reply-To: <877eg6hiv9.fsf@bootlin.com>

On Tue, Dec 18, 2018 at 05:58:34PM +0100, Gregory CLEMENT wrote:
> Hi Sudeep,
>
>  On mer., déc. 05 2018, Sudeep Holla <sudeep.holla@arm.com> wrote:
>
> > On Wed, Dec 05, 2018 at 09:27:45AM +0100, Gregory CLEMENT wrote:
> >> Hi Baruch,
> >>
> >>  On mer., déc. 05 2018, Baruch Siach <baruch@tkos.co.il> wrote:
> >>
> >> > Hi Gregory,
> >> >
> >> > On Tue, Dec 04, 2018 at 06:48:43PM +0200, Baruch Siach wrote:
> >> >> On Tue, Dec 04, 2018 at 05:21:25PM +0100, Gregory CLEMENT wrote:
> >> >> >  On mar., déc. 04 2018, Baruch Siach <baruch@tkos.co.il> wrote:
> >> >> > > This reverts commit 8ed46368776b3bc93d74c1f8f2bfb9fd8a9ad805.
> >> >> > >
> >> >> > > This commit breaks boot on Armada 8K based systems. Reverting it makes
> >> >> > > affected systems boot again.
> >> >> > >
> >> >> > > Reported-by: Sergey Matyukevich <geomatsi@gmail.com>
> >> >> > > Signed-off-by: Baruch Siach <baruch@tkos.co.il>
> >> >> >
> >> >> > I would like to know more about it because on the McBin I use there is
> >> >> > no issue with 4.20-rc1.
> >> >> >
> >> >> > Do you observe this issue with the arm64 defconfig or do you have some
> >> >> > specific configuration option?
> >> >>
> >> >> I tested the attached defconfig on Clearfog GT-8K and on another custom Armada
> >> >> 8K based board.
> >> >>
> >> >> Tomorrow I'll test Macchiatobin with this config and the arm64 defconfig.
> >> >
> >> > I reproduced same issue on a Macchiatobin board with both this config and the
> >> > arm64 defconfig. The latter produced this splat:
> >>
> >> OK thanks, for this, I also got the confirmation by my colleagues. The
> >> only difference I can think of, is the firmware, but the surprising
> >> thing, is that I have an old firmware: 17.06.2 and that with a recent
> >> one, it doesn't work.
> >>
> >> So I will apply it in mvebu/fixes.
> >>
> >
> > Sorry if it's too late. But I would rather disable the states that are
> > broken on old firmware and see if it's possible for bootloader or boot
> > firmware loading DT can modify the status to "enabled" if it can detect
> > the firmware version.
>
> you mean adding status = "disable" on each idle-state subnode?
>
Yes indeed. Keep it disabled by default and can enable it in bootloader
if you can detect it that it should work with particular version of
firmware. Just a thought especially if certain set of platforms support
it while another not.

--
Regards,
Sudeep

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