[Intel-gfx] [PATCH] drm/i915: Add two-stage ILK-style watermark programming (v11)

[Matt Roper]

On Mon, Feb 29, 2016 at 09:23:14AM +0100, Maarten Lankhorst wrote:
> Op 25-02-16 om 17:48 schreef Matt Roper:
> > I never got a CI email for this (probably because fdo was down for a while),
> > but I can see the results below in patchwork
> >
> >>  Test gem_cs_prefetch:
> >>          Subgroup basic-default:
> >>                  incomplete -> PASS       (ilk-hp8440p)
> >>  Test kms_flip:
> >>          Subgroup basic-flip-vs-dpms:
> >>                  pass       -> DMESG-WARN (ilk-hp8440p) UNSTABLE
> > Pre-existing watermark bug here:
> >         https://bugs.freedesktop.org/show_bug.cgi?id=93787
> >
> > I had hoped this patch might also fix that problem, but I guess further
> > investigation will be needed; it seems to be something very
> > ILK-specific, not seen on SNB, IVB, etc.
> >
> >
> >>  Test kms_force_connector_basic:
> >>          Subgroup force-edid:
> >>                  skip       -> PASS       (snb-x220t)
> >>                  pass       -> SKIP       (ivb-t430s)
> > Pre-existing:
> >         https://bugs.freedesktop.org/show_bug.cgi?id=93769
> >
> >>          Subgroup force-load-detect:
> >>                  dmesg-fail -> FAIL       (snb-x220t)
> >>                  dmesg-fail -> FAIL       (hsw-gt2)
> > The dmesg warning is gone now (a locking warning that I don't think my
> > patch should have affected); the remaining test failure is identical to
> > what it was before (temp->connection != DRM_MODE_UNKNOWNCONNECTION) and
> > unrelated.
> >
> > I don't see a bugzilla for this, which seems strange since it's a
> > pre-existing defect.  Am I overlooking it or do I need to file a new
> > one?
> >
> >>                  fail       -> SKIP       (ilk-hp8440p)
> > Not sure what is actually plugged into the CI machine, so the skip may
> > be legitimate/expected?  Doesn't seem related to anything in my patch
> > anyway.
> >
> >>  Test kms_pipe_crc_basic:
> >>          Subgroup suspend-read-crc-pipe-c:
> >>                  pass       -> DMESG-WARN (bsw-nuc-2)
> > This one was happening sporadically before my patch; I think it's the
> > same bug filed here:
> >
> >   https://bugs.freedesktop.org/show_bug.cgi?id=94164
> >
> > I'll update the bugzilla to indicate it also happens on this test.
> >
> >>  Test pm_rpm:
> >>          Subgroup basic-rte:
> >>                  pass       -> DMESG-WARN (bsw-nuc-2)
> >>                  fail       -> PASS       (hsw-brixbox)
> > Same bug mentioned above (94164)
> >
> Reviewed-by: Maarten Lankhorst <maarten.lankhorst at linux.intel.com>

Thanks for the review.  Pushed to dinq.


Matt

-- 
Matt Roper
Graphics Software Engineer
IoTG Platform Enabling & Development
Intel Corporation
(916) 356-2795