[Intel-gfx] ✗ Fi.CI.BAT: warning for ICL display initialization, selected patches (rev3)

Tomi Sarvela tomi.p.sarvela at intel.com
Fri Feb 9 09:31:21 UTC 2018


On 02/08/18 15:34, Paulo Zanoni wrote:
> Em Ter, 2018-02-06 às 19:54 +0000, Patchwork escreveu:
>> == Series Details ==
>>
>> Series: ICL display initialization, selected patches (rev3)
>> URL   : https://patchwork.freedesktop.org/series/37668/
>> State : warning
>>
>> == Summary ==
>>
>> Series 37668v3 ICL display initialization, selected patches
>> https://patchwork.freedesktop.org/api/1.0/series/37668/revisions/3/mb
>> ox/
>>
>> Test gem_sync:
>>          Subgroup basic-all:
>>                  pass       -> SKIP       (fi-pnv-d510)
>>          Subgroup basic-each:
>>                  pass       -> SKIP       (fi-pnv-d510)
>>          Subgroup basic-many-each:
>>                  pass       -> SKIP       (fi-pnv-d510)
>>          Subgroup basic-store-all:
>>                  pass       -> SKIP       (fi-pnv-d510)
>>          Subgroup basic-store-each:
>>                  pass       -> SKIP       (fi-pnv-d510)
> 
> IGT-Version: 1.21-g3fd9b578 (x86_64) (Linux: 4.15.0-CI-Patchwork_7913+
> x86_64)

> 
> The fact that these SKIPs don't happen in the test of rev2 and the
> changes from both rev1->rev2 and rev2->rev3 are not related to pnv at
> all suggests that these skips are unrelated to this series.
> 
> A small search on the other recent CI results points me to lots of
> patches resulting in a SKIP -> pass for this test, with some others
> doing the opposite. Perhaps we could signal this a a bug in the CI
> results.

Result flipflops are handled through cibuglog which connects (test,host) 
to fd.o bug. There is no bug for PNV for this issue, it's quite rare as 
can be seen from

https://intel-gfx-ci.01.org/tree/drm-tip/fi-pnv-d510.html

(the bug seem to appear after failure on gem_ringfill at basic-default-hang)

So yes: in this case CI gives misleading result, but that can be easily 
checked (by human) from history. Work is being done towards cibuglog-ng 
that is much easier to use to detect and create bugs like this, and has 
better knowledge about previous runs.

Tomi
-- 
Intel Finland Oy - BIC 0357606-4 - Westendinkatu 7, 02160 Espoo


More information about the Intel-gfx mailing list