On fi-hsw-4770k gem_exec_suspend@basic-s3 does not complete. E.g. for the last incomplete run CI_DRM_1635 (14-Sep-2016) dmesg tail has ... [ 355.860011] gem_exec_suspend: executing [ 355.860109] [drm:i915_gem_open] [ 355.861772] [drm:i915_gem_open] [ 355.862351] gem_exec_suspend: starting subtest basic [ 355.958711] gem_exec_suspend: exiting, ret=0 [ 356.096236] gem_exec_suspend: executing [ 356.096337] [drm:i915_gem_open] [ 356.097855] [drm:i915_gem_open] [ 356.098518] gem_exec_suspend: starting subtest basic-S3 while on the execution CI_DRM_1637 having Success the dmesg looks like ... [ 332.936844] gem_exec_suspend: executing [ 332.936940] [drm:i915_gem_open] [ 332.938575] [drm:i915_gem_open] [ 332.939160] gem_exec_suspend: starting subtest basic [ 333.035632] gem_exec_suspend: exiting, ret=0 [ 333.170070] gem_exec_suspend: executing [ 333.170166] [drm:i915_gem_open] [ 333.171718] [drm:i915_gem_open] [ 333.172307] gem_exec_suspend: starting subtest basic-S3 [ 333.686847] PM: Syncing filesystems ... done. [ 333.687901] Freezing user space processes ... (elapsed 0.001 seconds) done. [ 333.689489] Double checking all user space processes after OOM killer disable... (elapsed 0.000 seconds) [ 333.689547] Freezing remaining freezable tasks ... (elapsed 0.001 seconds) done. [ 333.690692] Suspending console(s) (use no_console_suspend to debug) [ 333.692391] sd 0:0:0:0: [sda] Synchronizing SCSI cache [ 333.692985] sd 0:0:0:0: [sda] Stopping disk [ 333.697233] serial 00:05: disabled [ 333.699109] [drm:connected_sink_compute_bpp] [CONNECTOR:44:HDMI-A-2] checking for sink bpp constrains ...
On this machine suspend/resume doesn't work even without the i915 driver being loaded. I'm still trying to narrow down the root cause.
The same symptom can be seen also with other suspend related cases ... igt@kms_pipe_crc_basic@suspend-read-crc-pipe-a igt@kms_pipe_crc_basic@suspend-read-crc-pipe-b igt@kms_pipe_crc_basic@suspend-read-crc-pipe-c
The hang happens on this particular platform (HSW/H87M-D3H) during resuming from S3 somewhere in BIOS code, before reaching the kernel. I reported the problem to the OEM, the HSW machine was replaced with another one in the CI farm.
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.