[Intel-gfx] skylake + drm-next - warn city

[Dave Airlie]

On 3 November 2015 at 20:47, Ville Syrjälä
<ville.syrjala at linux.intel.com> wrote:
> On Tue, Nov 03, 2015 at 12:35:58PM +0200, Imre Deak wrote:
>> On ti, 2015-11-03 at 11:42 +0200, Jani Nikula wrote:
>> > On Tue, 03 Nov 2015, Dave Airlie <airlied at gmail.com> wrote:
>> > > Just booted drm-next on a Skylake laptop that happened to be on my
>> > > desk for a few days.
>> > >
>> > > I wasn't impressed. I'm very disappointed. Doesn't anyone have any
>> > > pride in the code they write anymore.
>> > >
>> > > Initially the previous sentence had a lot of curse words and was Linus
>> > > like in it's stature, but I've been promised by twitter that being
>> > > nice will get me better results, so let's make it so.
>> >
>> > Much appreciated; we get the message.
>> >
>> > > So could someone from Intel takes some responsibility for testing the
>> > > code they send me actually you know works on the hardware it's meant
>> > > to, or at least tell me what is going so horribly wrong here.
>> > >
>> > > the lockdep trace at the end doesn't look fun.
>>
>> That's from the GuC firmware loader, so won't be fixed by the DMC
>> patches. Ville mentioned he has a fix for that.
>
> No, I don't. I probably just mentioned that I saw this lockdep spew
> already some weeks ago, and complained about it at the time. I think
> Daniel tried to ping someone specific about it, but no idea if they
> took any notice.

>From my perspective outside Intel, since the QA took a dive, I think
the pipeline is far too long now.

We've added delays to the pipeline previously to facilitate QA, but
when QA is removed we haven't shortened the pipeline again.

I'm kind of tempted to ask you guys to throw away everything queued up
in drm-intel-next on the basis that it has probably had 0 actual
testing and start again.

We have a major process failure in place here, and shoving more code
in the backend and hoping it somehow magically fixes itself between
drm-intel-next and merging to Linus's tree is clearly not working for
the past 6 months at least. I'm really unhappy about how shoddy 4.2
is, and 4.3 is clearly not shaping up to have been a winner, 4.4 is
looking even less fun.

So maybe you guys can brainstrom a bit, also when Daniel gets back.
But at the moment I think until QA is fully reestablished, I think not
merging anything to drm-intel-next for a few weeks and taking a break
on new features until some of the features that were merged broken
actually get fixed.

I'm also going to start looking at reverting skylake firmware loading,
it's clearly never been tested with lockdep enabled by anyone who
cared, which to my mind says it should never have been merged in the
first place.

Dave.