RE: [PATCH v2 2/3] drm/i915/cx0: program DDI_CLK_VALFREQ with DDI clock frequency
"Kandpal, Suraj" <[email protected]>
| Newsgroups | org.freedesktop.lists.intel-gfx,org.freedesktop.lists.intel-xe |
|---|---|
| Message-ID | <DS4PPFE901A304FFA6E05BB74604B8235E9E3DC2@DS4PPFE901A304F.namprd11.prod.outlook.com> |
> Subject: RE: [PATCH v2 2/3] drm/i915/cx0: program DDI_CLK_VALFREQ with DDI > clock frequency > > On Wed, 12 Aug 2026, "Kandpal, Suraj" <[email protected]> wrote: > >> Subject: Re: [PATCH v2 2/3] drm/i915/cx0: program DDI_CLK_VALFREQ > >> with DDI clock frequency > >> > >> > >> On 8/11/2026 11:28 PM, Suraj Kandpal wrote: > >> > DDI_CLK_VALFREQ is programmed with the port clock, which for DP is > >> > the symbol clock computed assuming 8b/10b encoding (link_rate / > >> > 10). For DP 128b/132b (UHBR) rates and for HDMI FRL the port clock > >> > needs to be modfied. > >> > On silicon DDI_CLK_VALFREQ is a scratch register with no functional > >> > impact, so the bug is not observed. > >> > Use intel_ddi_link_symbol_clock() to write the correct DDI clock in > >> > kHz. > >> > >> Perhaps can be re-phrased as : > >> > >> DDI_CLK_VALFREQ does not configure hardware, it only records the > >> frequency software intends to set, so there is no functional impact. > >> > >> Patch LGTM. > >> > >> Reviewed-by: Ankit Nautiyal <[email protected]> > > > > Thanks for the review > > > > Pushed to din with above mentioned changes in commit message > > Process nitpick #1. This series was pushed before there were xe CI results > AFAICS. > > I understand the series is the same as v1, with just the commit message > changed. But that's process nitpick #2. Please don't send a v2 to just change the > commit message, on the same day, with no other review comments in yet. Wait > out for the review comments, and if the only thing that needs changing in the > end is the commit message, you don't have to resend anything at all. Just fix > the commit message and push. Ask in the review thread if you're unsure about > the wording. > > In this case, going slower would've gone faster, and saved a full v2 CI round on > dozens of machines, just for a commit message wording change. > Sure Jani will keep it in mind going forward Regards, Suraj Kandpal > BR, > Jani. > > > -- > Jani Nikula, Intel