Having this issue too, on Fedora though, so it's certianly upstream from you. Have the issue worst when playing Minecraft (often, not recoverable), also happens on TF2 on Steam Linux, though in that case it's recoverable. Not seeing anything odd in Xorg.0.log, but i've got this from my dmesg: [ 3067.859336] radeon 0000:01:00.0: GPU lockup CP stall for more than 10000msec [ 3067.859344] radeon 0000:01:00.0: GPU lockup (waiting for 0x0000000000099aed last fence id 0x0000000000099ae6) [ 3067.860419] radeon 0000:01:00.0: Saved 343 dwords of commands on ring 0. [ 3067.860426] radeon 0000:01:00.0: GPU softreset: 0x00000003 [ 3067.864215] radeon 0000:01:00.0: GRBM_STATUS = 0xE5500828 [ 3067.864220] radeon 0000:01:00.0: GRBM_STATUS_SE0 = 0xEC000001 [ 3067.864224] radeon 0000:01:00.0: GRBM_STATUS_SE1 = 0x00000007 [ 3067.864227] radeon 0000:01:00.0: SRBM_STATUS = 0x200000C0 [ 3067.864230] radeon 0000:01:00.0: R_008674_CP_STALLED_STAT1 = 0x00000000 [ 3067.864233] radeon 0000:01:00.0: R_008678_CP_STALLED_STAT2 = 0x00010800 [ 3067.864237] radeon 0000:01:00.0: R_00867C_CP_BUSY_STAT = 0x00028006 [ 3067.864240] radeon 0000:01:00.0: R_008680_CP_STAT = 0x80038647 [ 3067.864243] radeon 0000:01:00.0: GRBM_SOFT_RESET=0x00007F6B [ 3067.864297] radeon 0000:01:00.0: GRBM_STATUS = 0x00003828 [ 3067.864320] radeon 0000:01:00.0: GRBM_STATUS_SE0 = 0x00000007 [ 3067.864326] radeon 0000:01:00.0: GRBM_STATUS_SE1 = 0x00000007 [ 3067.864343] radeon 0000:01:00.0: SRBM_STATUS = 0x200000C0 [ 3067.864361] radeon 0000:01:00.0: R_008674_CP_STALLED_STAT1 = 0x00000000 [ 3067.864376] radeon 0000:01:00.0: R_008678_CP_STALLED_STAT2 = 0x00000000 [ 3067.864382] radeon 0000:01:00.0: R_00867C_CP_BUSY_STAT = 0x00000000 [ 3067.864387] radeon 0000:01:00.0: R_008680_CP_STAT = 0x00000000 [ 3067.881741] radeon 0000:01:00.0: GPU reset succeeded, trying to resume [ 3067.899981] [drm] probing gen 2 caps for device 8086:151 = 3/e [ 3067.899985] [drm] PCIE gen 2 link speeds already enabled [ 3067.901926] [drm] PCIE GART of 512M enabled (table at 0x0000000000040000). [ 3067.902030] radeon 0000:01:00.0: WB enabled [ 3067.902032] radeon 0000:01:00.0: fence driver on ring 0 use gpu addr 0x0000000040000c00 and cpu addr 0xffff88014235ac00 [ 3067.902033] radeon 0000:01:00.0: fence driver on ring 3 use gpu addr 0x0000000040000c0c and cpu addr 0xffff88014235ac0c [ 3067.918240] [drm] ring test on 0 succeeded in 1 usecs [ 3067.918305] [drm] ring test on 3 succeeded in 1 usecs [ 3067.930434] [drm] ib test on ring 0 succeeded in 0 usecs [ 3067.930477] [drm] ib test on ring 3 succeeded in 0 usecs
Having this issue too, on Fedora though, so it's certianly upstream from you. Have the issue worst when playing Minecraft (often, not recoverable), also happens on TF2 on Steam Linux, though in that case it's recoverable. Not seeing anything odd in Xorg.0.log, but i've got this from my dmesg: CP_STALLED_ STAT1 = 0x00000000 CP_STALLED_ STAT2 = 0x00010800 CP_BUSY_ STAT = 0x00028006 RESET=0x00007F6 B CP_STALLED_ STAT1 = 0x00000000 CP_STALLED_ STAT2 = 0x00000000 CP_BUSY_ STAT = 0x00000000 000).
[ 3067.859336] radeon 0000:01:00.0: GPU lockup CP stall for more than 10000msec
[ 3067.859344] radeon 0000:01:00.0: GPU lockup (waiting for 0x0000000000099aed last fence id 0x0000000000099ae6)
[ 3067.860419] radeon 0000:01:00.0: Saved 343 dwords of commands on ring 0.
[ 3067.860426] radeon 0000:01:00.0: GPU softreset: 0x00000003
[ 3067.864215] radeon 0000:01:00.0: GRBM_STATUS = 0xE5500828
[ 3067.864220] radeon 0000:01:00.0: GRBM_STATUS_SE0 = 0xEC000001
[ 3067.864224] radeon 0000:01:00.0: GRBM_STATUS_SE1 = 0x00000007
[ 3067.864227] radeon 0000:01:00.0: SRBM_STATUS = 0x200000C0
[ 3067.864230] radeon 0000:01:00.0: R_008674_
[ 3067.864233] radeon 0000:01:00.0: R_008678_
[ 3067.864237] radeon 0000:01:00.0: R_00867C_
[ 3067.864240] radeon 0000:01:00.0: R_008680_CP_STAT = 0x80038647
[ 3067.864243] radeon 0000:01:00.0: GRBM_SOFT_
[ 3067.864297] radeon 0000:01:00.0: GRBM_STATUS = 0x00003828
[ 3067.864320] radeon 0000:01:00.0: GRBM_STATUS_SE0 = 0x00000007
[ 3067.864326] radeon 0000:01:00.0: GRBM_STATUS_SE1 = 0x00000007
[ 3067.864343] radeon 0000:01:00.0: SRBM_STATUS = 0x200000C0
[ 3067.864361] radeon 0000:01:00.0: R_008674_
[ 3067.864376] radeon 0000:01:00.0: R_008678_
[ 3067.864382] radeon 0000:01:00.0: R_00867C_
[ 3067.864387] radeon 0000:01:00.0: R_008680_CP_STAT = 0x00000000
[ 3067.881741] radeon 0000:01:00.0: GPU reset succeeded, trying to resume
[ 3067.899981] [drm] probing gen 2 caps for device 8086:151 = 3/e
[ 3067.899985] [drm] PCIE gen 2 link speeds already enabled
[ 3067.901926] [drm] PCIE GART of 512M enabled (table at 0x0000000000040
[ 3067.902030] radeon 0000:01:00.0: WB enabled
[ 3067.902032] radeon 0000:01:00.0: fence driver on ring 0 use gpu addr 0x0000000040000c00 and cpu addr 0xffff88014235ac00
[ 3067.902033] radeon 0000:01:00.0: fence driver on ring 3 use gpu addr 0x0000000040000c0c and cpu addr 0xffff88014235ac0c
[ 3067.918240] [drm] ring test on 0 succeeded in 1 usecs
[ 3067.918305] [drm] ring test on 3 succeeded in 1 usecs
[ 3067.930434] [drm] ib test on ring 0 succeeded in 0 usecs
[ 3067.930477] [drm] ib test on ring 3 succeeded in 0 usecs