On 01/15/2016 01:26 PM, Ville Syrjälä wrote: > On Fri, Jan 15, 2016 at 11:34:08AM +0100, Vlastimil Babka wrote: >> >> I have suspected that kwin is waiting for some event, but nevertheless >> tried bisecting the kernel between 4.3 and 4.4, which lead to: >> >> # first bad commit: [4dfd64862ff852df7b1198d667dda778715ee88f] drm: Use >> vblank timestamps to guesstimate how many vblanks were missed >> >> I can confirm that 4.4 works if I revert the following commits: >> 63154ff230fc9255cc507af6277cd181943c50a1 "drm/amdgpu: Fixup hw vblank >> counter/ts for new drm_update_vblank_count() (v3)" >> >> d1145ad1e41b6c33758a856163198cb53bb96a50 "drm/radeon: Fixup hw vblank >> counter/ts for new drm_update_vblank_count() (v2)" > > The sha1s don't seem to match what I have, so not sure which kernel tree
Hm sorry, I pasted the sha1 of the reverts by mistake. Correct sha1 are: 5b5561b3660db734652fbd02b4b6cbe00434d96b "drm/radeon: Fixup hw vblank counter/ts for new drm_update_vblank_count() (v2)" fa4270d8e0257b4b76f11baa2866f4313d29aaf5 "drm: Don't zero vblank timestamps from the irq handler" 235fabe09b46469adad2c9e4cb0563758155187c "drm: Add DRM_DEBUG_VBL()" 4dfd64862ff852df7b1198d667dda778715ee88f "drm: Use vblank timestamps to guesstimate how many vblanks were missed" 8e36f9d33c134d5c6448ad65b423a9fd94e045cf "drm/amdgpu: Fixup hw vblank counter/ts for new drm_update_vblank_count() (v3)" Also, it turns out that the process actually showing the progress is "ksplashqml", not kwin. It survives killing kwin, and restarting kwin just makes it shown on top again, or something. If I force kill ksplashqml instead of kwin, the desktop works including decorations and everything. ksplashqml itself also waits in kernel in poll(). I'll try some of your suggestions, thanks!
