Re: [PATCH v5 7/8] drivers: cpuidle: initialize Exynos driver through DT

[Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>]

Hi,

On Friday, July 18, 2014 02:15:01 PM Chander Kashyap wrote:
> Hi Lorenzo,
> 
> 
> On 17 July 2014 19:50, Lorenzo Pieralisi <lorenzo.pieralisi@arm.com> wrote:
> > On Wed, Jun 25, 2014 at 04:23:38PM +0100, Bartlomiej Zolnierkiewicz wrote:
> >>
> >> Hi,
> >>
> >> On Wednesday, June 25, 2014 03:10:20 PM Lorenzo Pieralisi wrote:
> >> > With the introduction of DT based idle states, CPUidle drivers for
> >> > ARM can now initialize idle states data through properties in the device
> >> > tree.
> >> >
> >> > This patch adds code to the Exynos CPUidle driver to dynamically
> >> > initialize idle states data through the updated device tree source
> >> > files.
> >> >
> >> > Cc: Kukjin Kim <kgene.kim@samsung.com>
> >> > Cc: Tomasz Figa <t.figa@samsung.com>
> >> > Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
> >> > ---
> >> > Compile tested, I am not sure I patched the right dts files, please check.
> >>
> >> cpuidle-exynos driver is currently working properly in deeper cpuidle
> >> mode (AFTR) on Exynos4210 and Exynos5250 (please also see the following
> >> patch from Tomasz Figa: [1]).  There is ongoing work to AFTR mode work
> >> also on Exynos4x12 and Exynos3250 but it is not complete yet.  Exynos5410
> >> OTOH should probably use the generic big little cpuidle driver (this SoC
> >> is similar to Exynos5420 one for which Chander Kashyap has developed
> >> cpuidle-big_little support [2]).
> >>
> >> Making long story short, I think that your patch should depend on patch
> >> [1] and update only exynos4210.dtsi and exynos5250.dtsi.  Also for your
> >> patch #6 there needs to be some coordination with merging of Chander's
> >> patchset ([2]).
> >>
> >> [1] http://www.spinics.net/lists/arm-kernel/msg341023.html
> >> [2] https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg664470.html
> >
> >
> > exynos4210.dtsi does not even have cpu nodes in it. Should I add them or
> > this might trigger regression (ie cpu_logical_map()) ?
> 
> Yes that can cause regression.

Yes, two patches from Tomasz Figa are needed to fix it:

- [PATCH 2/6] ARM: EXYNOS: Fix core ID used by platsmp and hotplug code
  http://www.mail-archive.com/linux-samsung-soc@vger.kernel.org/msg32811.html

- [PATCH] irqchip: gic: Fix core ID calculation when topology is read from DT
  http://www.mail-archive.com/linux-samsung-soc@vger.kernel.org/msg34277.html

> > I will post a new version soon, should I just patch 5250 for now ?

I posted patch adding CPU nodes for Exynos4 SoCs to DT:

- [PATCH] ARM: dts: add CPU nodes for Exynos4 SoCs
  http://www.mail-archive.com/linux-samsung-soc@vger.kernel.org/msg34378.html

Please make your series depend on it and add Exynos4210 support.

> > I would need help to test this patch thanks.
> 
> I can test the patch for 5250.

I can do testing on Exynos4210 if needed.

Best regards,
--
Bartlomiej Zolnierkiewicz
Samsung R&D Institute Poland
Samsung Electronics

> >
> > Lorenzo
> >
> >> >  .../devicetree/bindings/arm/exynos/idle-states.txt | 27 ++++++++++++++++++++
> >> >  arch/arm/boot/dts/exynos3250.dtsi                  | 16 ++++++++++++
> >> >  arch/arm/boot/dts/exynos5250.dtsi                  | 15 +++++++++++
> >> >  arch/arm/boot/dts/exynos5410.dtsi                  | 17 +++++++++++++
> >> >  drivers/cpuidle/Kconfig.arm                        |  1 +
> >> >  drivers/cpuidle/cpuidle-exynos.c                   | 29 +++++++++++++---------
> >> >  6 files changed, 93 insertions(+), 12 deletions(-)
> >> >  create mode 100644 Documentation/devicetree/bindings/arm/exynos/idle-states.txt