[Intel-gfx] ✗ failure: Fi.CI.BAT

[Jani Nikula]

On Thu, 07 Jan 2016, Daniel Vetter <daniel at ffwll.ch> wrote:
> On Thu, Jan 07, 2016 at 03:06:13PM +0200, Jani Nikula wrote:
>> On Thu, 07 Jan 2016, Dave Gordon <david.s.gordon at intel.com> wrote:
>> > On 06/01/16 10:20, Patchwork wrote:
>> >> == Summary ==
>> >>
>> >> Built on 24b053acb16b4b3b021575e4ee30ffedd3ab2920 drm-intel-nightly: 2016y-01m-06d-08h-16m-11s UTC integration manifest
>> >>
>> >> Test drv_getparams_basic:
>> >>          Subgroup basic-eu-total:
>> >>                  pass       -> DMESG-FAIL (skl-i5k-2)
>> >>                  pass       -> DMESG-FAIL (skl-i7k-2)
>> 
>> [snip]
>> 
>> >>
>> >> bdw-nuci7        total:132  pass:122  dwarn:1   dfail:0   fail:0   skip:9
>> >> bdw-ultra        total:132  pass:126  dwarn:0   dfail:0   fail:0   skip:6
>> >> bsw-nuc-2        total:135  pass:115  dwarn:0   dfail:0   fail:0   skip:20
>> >> byt-nuc          total:135  pass:121  dwarn:1   dfail:0   fail:0   skip:13
>> >> hsw-brixbox      total:135  pass:128  dwarn:0   dfail:0   fail:0   skip:7
>> >> hsw-gt2          total:135  pass:131  dwarn:0   dfail:0   fail:0   skip:4
>> >> ilk-hp8440p      total:135  pass:100  dwarn:0   dfail:0   fail:0   skip:35
>> >> ivb-t430s        total:135  pass:129  dwarn:0   dfail:0   fail:0   skip:6
>> >> skl-i5k-2        total:135  pass:2    dwarn:0   dfail:132 fail:0   skip:1
>> >> skl-i7k-2        total:135  pass:2    dwarn:0   dfail:132 fail:0   skip:1
>> >> snb-dellxps      total:135  pass:123  dwarn:0   dfail:0   fail:0   skip:12
>> >> snb-x220t        total:135  pass:123  dwarn:0   dfail:0   fail:1   skip:11
>> >>
>> >> Results at /archive/results/CI_IGT_test/Patchwork_1094/
>> >>
>> >> _______________________________________________
>> >> Intel-gfx mailing list
>> >> Intel-gfx at lists.freedesktop.org
>> >> http://lists.freedesktop.org/mailman/listinfo/intel-gfx
>> >
>> > Looks like the CI SKL doesn't have the GuC firmware installed?
>> 
>> It's tons of:
>> 
>> [   38.169461] [drm:intel_lr_context_deferred_alloc [i915]] *ERROR* ring create req: -5
>> 
>> If that gets fixed by installing the GuC firmware, the answer is *not*
>> to install the GuC firmware on the CI machines. The answer is to make
>> the driver handle missing firmware gracefully.
>
> I kinda don't want to support 2 different ways to run things on any given
> platform, because we can't even support one way properly.
>
> But since it took forever to get guc enabled people will indeed scream if
> guc isn't there, so either we enable this only for bxt and later or we
> indeed need to support both cases on skl :(

It's probably considered a regression to add a hard requirement on guc
firmware now that skl has been running fine without. Yet we probably
want all the coverage we can get for the guc case, so I don't think we
should rely on bxt alone with that.

> Either way we do need to install guc firmware first on CI boxes, since
> otherwise coverage isn't there.

How about having guc firmware on half the CI boxes, at least for
starters. It's going to blow up anyway...

BR,
Jani.


-- 
Jani Nikula, Intel Open Source Technology Center