On Thu, 8 May 2014, Peter Ujfalusi wrote: > Hi Paul, > > On 05/08/2014 03:20 AM, Paul Walmsley wrote: > > Hi Péter, > > > > On Wed, 30 Apr 2014, Peter Ujfalusi wrote: > > > >> Add HWMOD_SWSUP_SIDLE to flags. > >> > >> Signed-off-by: Peter Ujfalusi <peter.ujfalusi@xxxxxx> > > > > This patch could use a better changelog. It would be ideal to understand > > _why_ HWMOD_SWSUP_SIDLE is needed. Is there a known hardware bug? Or is > > this simply based on practical observation, e.g., something bad happens if > > the McPDM is not software-idled? > > The same issue exists on OMAP5 as on OMAP4: > > commit 12d82e4b0aa6c71f38b668a372f9a13f243207da > Author: Peter Ujfalusi <peter.ujfalusi@xxxxxx> > Date: Fri Jan 18 16:48:16 2013 -0700 > > ARM: OMAP4: hwmod_data: Correct IDLEMODE for McPDM > > McPDM need to be configured to NO_IDLE mode when it is in used otherwise > vital clocks will be gated which results 'slow motion' audio playback. > > Signed-off-by: Peter Ujfalusi <peter.ujfalusi@xxxxxx> > [paul@xxxxxxxxx: copy patch description into hwmod data comments] > Signed-off-by: Paul Walmsley <paul@xxxxxxxxx> > > diff --git a/arch/arm/mach-omap2/omap_hwmod_44xx_data.c > b/arch/arm/mach-omap2/omap_hwmod_44xx_data.c Thanks, could you please resend the patch with an updated commit message? Then I'll take it. I assume this is -rc material? - Paul
