[Intel-gfx] ✗ Fi.CI.BAT: failure for drm/i915: Set PROBE_PREFER_ASYNCHRONOUS

[Brian Norris]

On Fri, Nov 04, 2022 at 02:38:03PM +0000, Matthew Auld wrote:
> On Thu, 3 Nov 2022 at 00:14, Brian Norris <briannorris at chromium.org> wrote:
> > I'm still curious about the reported failures, but maybe they require
> > some particular sequence of tests? I also don't have the full
> > igt-gpu-tools set running, so maybe they do something a little
> > differently than my steps in [1]?
> >
> > Brian
> >
> > [1] I have a GLk system, if it matters. I figured I can run some of
> > these with any one of the following:
> >
> >   modprobe i915 live_selftests=1
> >   modprobe i915 live_selftests=1 igt__20__live_workarounds=Y
> >   modprobe i915 live_selftests=1 igt__19__live_uncore=Y
> >   modprobe i915 live_selftests=1 igt__18__live_sanitycheck=Y
> >   ...
> 
> CI should be using the IGT wrapper to run them, AFAIK. So something like:
> 
> ./build/tests/i915_selftest
> 
> Or to just run the live, mock or perf:
> 
> ./build/tests/i915_selftest --run-subtest live
> ./build/tests/i915_selftest --run-subtest mock
> ./build/tests/i915_selftest --run-subtest perf
> 
> Or if you want to run some particular selftest, like live mman tests:
> 
> ./build/tests/i915_selftest --run-subtest live --dyn mman

Thanks. I'm running through those now, and it seems like I'm doing
closer to what the CI logs show [1], but I'm still not reproducing on my
GLK. (I've now managed to run it with drm-tip; still no luck.)

So far, now I've managed to just reproduced *different* known problems:

https://lore.kernel.org/all/Y2WfpLbX1SeDtk+7@google.com/

But after working around those, I run without any similar lockup
failures.

I might poke around some more next week, but I've probably spent more
time than reasonable on this already.

Anyway, thanks for the help!

Regards,
Brian

[1] For one, I've run through a test list, in order, based on this:
https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_110277v1/fi-glk-j4005/testlist0.txt