14:22 < berniyh> hi, plasmashell (5.7.4) is stuck after unlocking the screen 14:22 < berniyh> didn't yet restart it 14:22 < berniyh> is there a way to find out where it's stuck? 14:22 < berniyh> and is there a point? 14:22 < mgraesslin> gdb 14:25 < berniyh> gdb --pid something? 14:25 < berniyh> and then what? 14:26 < mgraesslin> yes gdb --pid 14:26 < mgraesslin> then t a a bt 14:27 < berniyh> https://bpaste.net/show/3704245d4600 14:27 < berniyh> oh, forgot about the threads 14:27 < mgraesslin> I think that's already the correct one 14:28 < berniyh> https://bpaste.net/show/484078dc935a 14:28 < mgraesslin> yeah, that erm looks bad 14:29 < mgraesslin> it's a freeze in the radeonsi driver 14:29 < mgraesslin> though I don't understand why 14:29 < mgraesslin> what looks really bad is that thread 1 and thread 2 are both blocked in xcb_conn_wait 14:29 < mgraesslin> no surprise it freezes 14:29 < mgraesslin> really bad 14:29 < berniyh> maybe it's related to the freezing bug I'm experiencing from time to time (wrote you an email about that)? 14:30 < berniyh> but in that case the whole screen freezes, while here only plasmashell is affected 14:30 < mgraesslin> quite possible it's related 14:30 < mgraesslin> in compositing the same could happen, if we go into an xcb_conn_wait in thread 1 it will freeze 14:30 < berniyh> [88765.431890] [drm:amdgpu_crtc_page_flip] *ERROR* failed to reserve new rbo buffer before flip 14:31 < berniyh> seems like there is something going on 14:31 < mgraesslin> yep that is probably related 14:31 < mgraesslin> my guess from the method names is that this is failing to allocate a back buffer 14:31 < mgraesslin> which would match the error message 14:32 < mgraesslin> that backtrace might be quite useful for the radeon devs, even without the debug symbols 14:32 < berniyh> I'll attach the bt and the log from the conversation to the bug I reported 14:33 < mgraesslin> ok 14:33 < berniyh> if I suspend and activate compositing, the task bar starts flickering 14:34 < mgraesslin> probably kwin gets a bad pixmap 14:34 < mgraesslin> which would given the freeze also not be too surprising 14:34 < berniyh> k