Created attachment 71839 [details] kms_flip.dmesg Environment: -------------------------- Kernel: (drm-intel-next-queued)079a43f67f41274b137d8be1fb6093c8f6b6a762 Some additional commit info: Author: Ben Widawsky <benjamin.widawsky@intel.com> Date: Tue Dec 18 10:31:24 2012 -0800 drm/i915: Missed conversion to gtt_pte_t Steps: --------------------------- ./kms_flip Result: -------------------------- 1.return value = 1 2.stdout: running testcase: wf_vblank-ts-check could not find requested crtc 3 Beginning wf_vblank-ts-check on crtc 4, connector 5 1024x600 60 1024 1048 1184 1438 600 603 604 628 0xa 0x48 54200 failed to page flip: Invalid argument
Can you please test with this patch applied: https://patchwork.kernel.org/patch/1887841/ Also, does this happen every time, or just sometimes?
Created attachment 71884 [details] kms_flip dmesg Test result: ------------------------ 1. return value=1 2. STD-OUT: running testcase: plain-flip Beginning plain-flip on crtc 3, connector 17 1920x1080 60 1920 2008 2052 2200 1080 1084 1089 1125 0x5 0x48 148500 select timed out or error (ret 0) Issue also exist.
(In reply to comment #1) > Can you please test with this patch applied: > > https://patchwork.kernel.org/patch/1887841/ > > Also, does this happen every time, or just sometimes? It happened every time.
This is strange - to prepare the next -testing round if just done a full igt run on g33,ivb&hsw, and safe for a few exceptions all the kms_flip subtests work here ... So a few more questions: - Does this happen for all subtests, or just plain-test? - Does this happen across all machines? I've checked nightly test results and on some machines kms_flip passes most/all testcases. Can you try to figure out what's different between these machines (os version, platforms, versions of your own tools, ...)? I'm a bit lost here since I've never seen this failure consistently, so I have no idea what's going on here ...
Ping. Please may we have a status update.
I reported a new bug: bug 59339 This bug related to kernel config : if enable most configurations in kernel hacking, this case worked nicely, if not, case will fail. All of the machines "kms_flip" fail, but it doesn't look like one reason. Now, we are working for this.
I'd like to close this bug. The reasons that cause fail are different, we are digging them, then we'll file new bug report only involved subtest.
Sounds good to me.
Closing verified+invalid.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.