Re: [PATCH/RFC] ARM: mach-shmobile: sh7372 power domain prototype

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Friday, April 29, 2011, Rafael J. Wysocki wrote:
> On Friday, April 29, 2011, Guennadi Liakhovetski wrote:
> > On Thu, 21 Apr 2011, Guennadi Liakhovetski wrote:
> > 
> > > On Tue, 19 Apr 2011, Guennadi Liakhovetski wrote:
> > > 
> > > > On Fri, 15 Apr 2011, Guennadi Liakhovetski wrote:
> > > > 
> > > > > Hi Magnus
> > > > > 
> > > > > On Tue, 12 Apr 2011, Magnus Damm wrote:
> > > > > 
> > > > > > From: Magnus Damm <damm@opensource.se>
> > > > > > 
> > > > > > This experimental sh7372-specific Runtime PM prototype
> > > > > > adds support for the A4LC and A4MP power domains.
> > > > > > 
> > > > > > Tested on the Mackerel board together with the LCDC:
> > > > > > 
> > > > > > # blank HDMI fbdev
> > > > > > echo 3 > /sys/devices/platform/sh_mobile_lcdc_fb.1/graphics/fb1/blank
> > > > > > 
> > > > > > # blank LCD panel fbdev
> > > > > > echo 3 > /sys/devices/platform/sh_mobile_lcdc_fb.0/graphics/fb0/blank
> > > > > > # unblank LCD panel fbdev
> > > > > > echo 0 > /sys/devices/platform/sh_mobile_lcdc_fb.0/graphics/fb0/blank
> > > > > > 
> > > > > > Not-yet-signed-off-by: Magnus Damm <damm@opensource.se>
> > > > > > ---
> > > > > > 
> > > > > >  arch/arm/mach-shmobile/pm_runtime.c |  227 ++++++++++++++++++++++++++++++++++-
> > > > > >  1 file changed, 222 insertions(+), 5 deletions(-)
> > > > > 
> > > > > Your patch worked mostly, below is a small addition (to be applied on top 
> > > > > of your patch), that adds several new domains and devices and fixes a 
> > > > > couple of issues with your original patch. Notice an "#if 0" around i2c 
> > > > > devices: activating them causes problems - output on the LCDC disappears 
> > > > > and the boot process doesn't complete, but without a serial console I 
> > > > > couldn't tell, what exactly was the problem. Should be investigated 
> > > > > further.
> > > > 
> > > > Below is a fixed version of the earlier proposed fix;)
> > > 
> > > v3: I think, I've got the use-counting and power-domain initialisation 
> > > correct this time too. At least this version works without any need to 
> > > modify the runtime PM core.
> > 
> > v4: (1) add checks for BIT_ACTIVE for devices without any clocks 
> > associated with them, this fixes Oopses. (2) I think, this time I've 
> > finally found _the_ correct time and place to remove the "prd" from the 
> > list. As before, this is a work in progress, subject to change, depending 
> > on surrounding developments. Posting here just for reference.
> 
> Please have a look at these patches:
> https://patchwork.kernel.org/patch/740162/
> https://patchwork.kernel.org/patch/740152/
> which are based on the patchset from:
> git://git.kernel.org/pub/scm/linux/kernel/git/rafael/suspend-2.6.git power-domains

I had to rebase this branch due to recent modifications requested by reviewers.

Also the two patches above have been updated, but the latest versions have
been tested with the A4LC power domain and LCDC and confirmed to work.

> Do you think it would be possible to rebase your work on top of them?
> If not, please tell me what the problems are.

Thanks,
Rafael