I get X lockups with a Navi 5700 XT and HWCursor enabled. Only seems to happen if TearFree (and VSync) are disabled. I can reliably reproduce this with The Witcher Enhanced Edition with TearFree off (the game doesn't support VSync without forcing it), run with D9VK. These are recoverable at least (just need to restart X) and it doesn't cause a GPU hang, just frozen mouse. If I run with TearFree on, a multi-monitor setup, or with SWCursor forced on, the issue goes away. I've also never seen the bug with a Vega card, although I can't say if it happens or not with the exact kernel I used. I'll also add I typically run with Freesync enabled, but it doesn't appear to affect things in either direction. The issue also happens seemingly randomly with other games (like Kingdom Come: Deliverance and the native Steam client). Reproduction steps: 1. Run "The Witcher: Enhanced Edition" with D9VK with TearFree off and HWCursor enabled, in a single monitor configuration 2. Move the mouse around during the main menu 3. X lockups with the following message: [ 314.590442] ------------[ cut here ]------------ [ 314.590487] WARNING: CPU: 1 PID: 6061 at drivers/gpu/drm/amd/amdgpu/../display/dc/dcn20/dcn20resource.c:2820 dcn20validatebandwidth.cold+0xe/0x18 [amdgpu] [ 314.590487] Modules linked in: iptablefilter iptables afpacket bridge stp llc ipv6 nfdefragipv6 zram uinput kvmamd sndhdacodecrealtek kvm joydev sndhdacodecgeneric mousedev amdgpu sndhdaintel sndvirtuoso sndhdacodec sndoxygenlib sndmpu401uart sndhwdep$ [ 314.590498] CPU: 1 PID: 6061 Comm: X Not tainted 5.3.7-kmsfix-5.3.y-ck0-extra1+ #102 [ 314.590498] Hardware name: System manufacturer System Product Name/PRIME X370-PRO, BIOS 5216 09/02/2019 [ 314.590524] RIP: 0010:dcn20validatebandwidth.cold+0xe/0x18 [amdgpu] [ 314.590526] Code: 05 d6 45 14 00 48 8b 74 24 08 0f b7 44 24 2e 80 cc 0c 66 89 44 24 2c e9 65 ad f5 ff 48 c7 c7 00 f6 45 a0 31 c0 e8 be 8e d3 e0 <0f> 0b 44 89 e0 e9 93 b2 f5 ff 48 c7 c7 00 f6 45 a0 31 c0 e8 a6 8e [ 314.590527] RSP: 0018:ffffc90001aa3ab8 EFLAGS: 0 0010246 [ 314.590529] RAX: 0000000000000024 RBX: 4079400000000000 RCX: 0000000000000000 [ 314.590530] RDX: 0000000000000000 RSI: ffff88881e6561f8 RDI: 00000000ffffffff [ 314.590531] RBP: ffff888722170000 R08: 0000000000000001 R09: 0000000000000711 [ 314.590531] R10: 0000000000000001 R11: 0000000000000000 R12: 0000000000000001 [ 314.590532] R13: ffff8888095a0000 R14: 0000000000000000 R15: ffff88881a7e6ad8 [ 314.590533] FS: 00007f61ce2a28c0(0000) GS:ffff88881e640000(0000) knlGS:0000000000000000 [ 314.590534] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 314.590535] CR2: 00000000e085f000 CR3: 000000080041f000 CR4: 00000000003406e0 [ 314.590535] Call Trace: [ 314.590563] dcvalidateglobalstate+0x21f/0x290 [amdgpu] [ 314.590593] amdgpudmatomiccheck+0x553/0x7a0 [amdgpu] [ 314.590596] drmatomiccheckonly+0x53e/0x760 [ 314.590599] ? _rawspinunlockirqrestore+0xf/0x30 [ 314.590600] drmatomiccommit+0xe/0x50 [ 314.590602] drmatomicconnectorcommitdpms+0xd1/0xf0 [ 314.590603] drmmodeobjsetpropertyioctl+0x153/0x29b [ 314.590605] ? drmconnectorsetobjprop+0x80/0x80 [ 314.590606] drmconnectorpropertysetioctl+0x34/0x50 [ 314.590608] drmioctlkernel+0x9f/0xe0 [ 314.590609] drmioctl+0x1ef/0x381 [ 314.590610] ? drmconnectorsetobjprop+0x80/0x80 [ 314.590612] ? epreadeventsproc+0xd0/0xd0 [ 314.590614] ? preemptcountadd+0x63/0x90 [ 314.590614] ? epreadeventsproc+0xd0/0xd0 [ 314.590635] amdgpudrmioctl+0x44/0x80 [amdgpu] [ 314.590637] dovfsioctl+0x3ea/0x640 [ 314.590638] ? _x64sysepollctl+0xf2/0x1050 [ 314.590639] ksysioctl+0x35/0x70 [ 314.590640] _x64sysioctl+0x11/0x20 [ 314.590642] dosyscall64+0x4e/0x1b0 [ 314.590643] ? prepareexittousermode+0x4c/0xc0 [ 314.590646] entrySYSCALL64afterhwframe+0x44/0xa9 [ 314.590647] RIP: 0033:0x7f61ceb9da37 [ 314.590648] Code: 00 00 00 75 0c 48 c7 c0 ff ff ff ff 48 83 c4 18 c3 e8 5d d8 01 00 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 29 f4 0c 00 f7 d8 64 89 01 48 [ 314.590649] RSP: 002b:00007fff2df1d388 EFLAGS: 00000246 ORIGRAX: 0000000000000010 [ 314.590650] RAX: ffffffffffffffda RBX: 000000000000000b RCX: 00007f61ceb9da37 [ 314.590651] RDX: 00007fff2df1d390 RSI: 00000000c01064ab RDI: 000000000000000b [ 314.590652] RBP: 00000000c01064ab R08: 0000000000000001 R09: 000056186e5094d0 [ 314.590652] R10: 000056186e517970 R11: 0000000000000246 R12: 00007fff2df1d390 [ 314.590653] R13: 000056186fa200d0 R14: 000056186e506600 R15: 000056186e506000 [ 314.590654] ---[ end trace 7784d0b1b9f93521 ]--- System configuration: MSI Evoke 5700 XT Mesa git amd-staging-drm-next (also tested with 5.4-rc4) xf86-video-amdgpu git
-- GitLab Migration Automatic Message -- This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity. You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/drm/amd/issues/939.
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.