Re: [PATCH v2] drm/probe-helper: Default to 640x480 if no EDID

From: Doug Anderson <dianders@chromium.org>
To: Thomas Zimmermann <tzimmermann@suse.de>
Cc: dri-devel <dri-devel@lists.freedesktop.org>,
	Sankeerth Billakanti <quic_sbillaka@quicinc.com>,
	LKML <linux-kernel@vger.kernel.org>,
	David Airlie <airlied@linux.ie>,
	linux-arm-msm <linux-arm-msm@vger.kernel.org>,
	"Abhinav Kumar (QUIC)" <quic_abhinavk@quicinc.com>,
	Stephen Boyd <swboyd@chromium.org>,
	Dmitry Baryshkov <dmitry.baryshkov@linaro.org>,
	"Aravind Venkateswaran (QUIC)" <quic_aravindh@quicinc.com>,
	"Kuogee Hsieh (QUIC)" <quic_khsieh@quicinc.com>,
	freedreno <freedreno@lists.freedesktop.org>
Subject: Re: [PATCH v2] drm/probe-helper: Default to 640x480 if no EDID
Date: Wed, 11 May 2022 14:32:26 -0700	[thread overview]
Message-ID: <CAD=FV=WoSTcSOB_reDbayNb=q7w00rd7p-zHUDt+evTkSjQ=2g@mail.gmail.com> (raw)
In-Reply-To: <db7a2b7f-3c94-d45d-98fd-7fd0b181e6aa@suse.de>

Hi,

On Wed, May 11, 2022 at 12:14 AM Thomas Zimmermann <tzimmermann@suse.de> wrote:
>
> Hi
>
> Am 10.05.22 um 22:51 schrieb Douglas Anderson:
> > If we're unable to read the EDID for a display because it's corrupt /
> > bogus / invalid then we'll add a set of standard modes for the
> > display. When userspace looks at these modes it doesn't really have a
> > good concept for which mode to pick and it'll likely pick the highest
> > resolution one by default. That's probably not ideal because the modes
> > were purely guesses on the part of the Linux kernel.
>
> I'm skeptical. Why does the kernel do a better job than userspace here?
> Only the graphics driver could possibly make such a decision.
>
> Not setting any preferred mode at least gives a clear message to userspace.

OK, that's a fair point. So I tried to find out what our userspace is
doing. I believe it's:

https://source.chromium.org/chromium/chromium/src/+/main:ui/ozone/platform/drm/common/drm_util.cc;l=529

Specifically this bit of code:

  // If we still have no preferred mode, then use the first one since it should
  // be the best mode.
  if (!*out_native_mode && !modes.empty())
    *out_native_mode = modes.front().get();

Do you agree with what our userspace is doing here, or is it wrong?

If our userspace is doing the right thing, then I guess the problem is
the call to "drm_mode_sort(&connector->modes);" at the end of
drm_helper_probe_single_connector_modes(). Would you be OK with me
_not_ sorting the modes in the "bad EDID" case? That also seems to fix
my problem...

-Doug