[Mesa-ci-intel] Intel CI. Please add new branch for running tests on iris driver
Clayton Craft
clayton.a.craft at intel.com
Tue Apr 23 16:58:32 UTC 2019
Hi Denys,
On Tue, Apr 23, 2019 at 06:07:36PM +0300, Denys Kostin wrote:
>I am sorry for bothering you again, today we discussed your answer
>inside the team and as a result - would like to ask you to add simply
>one more branch into CI.
>
>As I wrote below, right now we are investigating and fixing several
>tests in CTS/piglit, so need to run all tests quite often. Second
>branch would increase our speed.
No problem, I added your second branch to CI. CI builds for this branch will be
under a job called 'global_logic_2', and will show up under
https://mesa-ci.01.org once builds have been triggered on that branch. Both the
global_logic and global_logic_2 jobs will test i965 and Iris drivers.
>
>
>_link to repo:_ https://gitlab.freedesktop.org/GL/mesa.git
>
>_branch_: i965_second
>
>
>On 4/23/19 12:24 PM, Denys Kostin wrote:
>>Thank you very match, Clayton!
>>
>>On 4/19/19 9:04 PM, Clayton Craft wrote:
>>>Hi Denys,
>>>
>>>On Fri, Apr 19, 2019 at 12:12:12PM +0300, Denys Kostin wrote:
>>>>Hello Clayton, Mark!
>>>>
>>>>Since we got access to your CI system, we improved quality of
>>>>our patches a lot. Thank you for this opportunity one more time.
>>>>During last month we started work with new iris driver, made few
>>>>fixes there. It would very cool to get access to the iris CI so
>>>>we could also test our iris fixes before pushing them to mesa
>>>>master.
>>>
>>>The CI will already build/test your other branch, i965/ci, with
>>>Iris, so no
>>>changes to the CI are necessary to enable this for you. When you
>>>push to this
>>>branch, you will get test results for both i965 and Iris.
>>>
>>>>
>>>>Our main goal is to compare test results from i965 and Iris
>>>>driver and determine those, who require fixing. Also, as
>>>>additional setting, maybe would be helpful to "drop" all list
>>>>with "skipped" tests from i965, so we can see the hole picture
>>>>and "create" new list exactly for Iris, with it's own "Skipped"
>>>>tests after investigation of failure root cause.
>>>
>>>Probably the easiest way to see the latest, most up to date test
>>>status for i965
>>>and Iris drivers is to look at the .conf files in the CI
>>>mesa_jenkins repo:
>>>https://gitlab.freedesktop.org/Mesa_CI/mesa_jenkins
>>>
>>>There are subdirectories, *-test, which contain .conf files
>>>showing which tests
>>>are expected to crash or fail in CI, separated by platform and
>>>driver. Conf
>>>files for iris have _iris in the name. The *_blacklist files in those
>>>directories contain tests that are known to be flaky (intermittent
>>>pass/fail),
>>>or hard-hang the system.
>>>
>>>
>>>>
>>>>If it is possible, please, add this branch:
>>>>
>>>>_link to repo:_https://gitlab.freedesktop.org/GL/mesa.git
>>>>
>>>>_branch_: iris_ci
>>>>
>>>>_Email for notifications: _ Intel-Graphics-Dev at globallogic.com
>>>><mailto:Intel-Graphics-Dev at globallogic.com>
>>>>
>>>>
>>>>Thank you in advance,
>>>>
>>>>Denys.
>>>>
>>>
>>>>_______________________________________________
>>>>Mesa-ci-intel mailing list
>>>>Mesa-ci-intel at lists.freedesktop.org
>>>>https://lists.freedesktop.org/mailman/listinfo/mesa-ci-intel
>>>
>--
>Denys Kostin,
>Test Engineer at GlobalLogic
>
>_______________________________________________
>Mesa-ci-intel mailing list
>Mesa-ci-intel at lists.freedesktop.org
>https://lists.freedesktop.org/mailman/listinfo/mesa-ci-intel
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <https://lists.freedesktop.org/archives/mesa-ci-intel/attachments/20190423/a2245ac8/attachment.sig>
More information about the Mesa-ci-intel
mailing list