Hi, I get the next dmesg trace playing an h264 video with a custom player using libmpv-0.28.0. I'm using a debian8 with a vanilla 4.14.24 kernel with the next packages built from a padoka ppa: mesa 18.1~git1803060730.237c9c~oibaf~x xserver-xorg-video-amdgpu 1.4.99+git1803011934.7854ac~oibaf~x The video is played ok, maybe be some slow downs, but this trace is generated [ 2865.187435] ------------[ cut here ]------------ [ 2865.187520] WARNING: CPU: 0 PID: 2284 at drivers/gpu/drm/amd/amdgpu/amdgpu_object.c:1029 amdgpu_bo_gpu_offset+0x7e/0x90 [amdgpu] [ 2865.187521] Modules linked in: cfg80211 rfkill binfmt_misc nfsd auth_rpcgss oid_registry nfs_acl nfs lockd grace fscache sunrpc edac_mce_amd kvm irqbypass crct10dif_pclmul crc32_pclmul ghash_clmulni_intel pcbc evdev cdc_acm aesni_intel aes_x86_64 crypto_simd glue_helper cryptd amdgpu pcspkr snd_hda_codec_cirrus fam15h_power k10temp snd_hda_codec_hdmi snd_hda_codec_generic ttm snd_hda_intel snd_hda_codec snd_hda_core snd_hwdep drm_kms_helper snd_pcm snd_timer video snd soundcore sp5100_tco drm i2c_piix4 shpchp acpi_cpufreq button fuse autofs4 ext4 crc16 mbcache jbd2 sg sd_mod hid_generic usbhid hid crc32c_intel ehci_pci ahci ehci_hcd sdhci_pci xhci_pci sdhci libahci xhci_hcd igb libata mmc_core i2c_algo_bit i2c_core usbcore scsi_mod dca ptp usb_common pps_core [ 2865.187577] CPU: 0 PID: 2284 Comm: amdgpu_cs:0 Not tainted 4.14.24 #1 [ 2865.187578] Hardware name: Default string Default string/GCM-B610-CS, BIOS R1.00.E1 08/01/2017 [ 2865.187579] task: ffff8803e6b2f180 task.stack: ffffc9000be94000 [ 2865.187599] RIP: 0010:amdgpu_bo_gpu_offset+0x7e/0x90 [amdgpu] [ 2865.187600] RSP: 0018:ffffc9000be97a68 EFLAGS: 00010246 [ 2865.187602] RAX: ffff8803eb4c3200 RBX: ffff8803eb797000 RCX: ffff8803eb797000 [ 2865.187603] RDX: ffffc9000be97a80 RSI: 00000000001053a0 RDI: ffff8803eb4c32e8 [ 2865.187604] RBP: 00000001053a0000 R08: ffff8803e9840b90 R09: ffff8803e9840ab8 [ 2865.187604] R10: ffffffff81f26b00 R11: 0000000003f41000 R12: ffff880388445d80 [ 2865.187606] R13: ffff8803ba9e2ee0 R14: ffffffffa06079e0 R15: ffff8803ba9f2460 [ 2865.187608] FS: 00007fc14bc5b700(0000) GS:ffff8803fec00000(0000) knlGS:0000000000000000 [ 2865.187609] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 2865.187610] CR2: 00000000019c4000 CR3: 0000000398f56000 CR4: 00000000001406f0 [ 2865.187611] Call Trace: [ 2865.187640] amdgpu_uvd_cs_pass2+0x44/0x7d0 [amdgpu] [ 2865.187664] ? amdgpu_uvd_cs_pass1+0xf0/0xf0 [amdgpu] [ 2865.187685] amdgpu_uvd_cs_packets+0x153/0x170 [amdgpu] [ 2865.187706] amdgpu_uvd_ring_parse_cs+0xc9/0x140 [amdgpu] [ 2865.187726] amdgpu_cs_ioctl+0x106e/0x1a60 [amdgpu] [ 2865.187749] ? amdgpu_cs_find_mapping+0xa0/0xa0 [amdgpu] [ 2865.187764] drm_ioctl_kernel+0x65/0xb0 [drm] [ 2865.187773] drm_ioctl+0x2ce/0x380 [drm] [ 2865.187792] ? amdgpu_cs_find_mapping+0xa0/0xa0 [amdgpu] [ 2865.187797] ? update_curr+0xd8/0x130 [ 2865.187799] ? update_curr+0x5c/0x130 [ 2865.187815] amdgpu_drm_ioctl+0x46/0x80 [amdgpu] [ 2865.187826] do_vfs_ioctl+0x8b/0x5c0 [ 2865.187829] ? SyS_futex+0x71/0x170 [ 2865.187831] SyS_ioctl+0x76/0x90 [ 2865.187834] do_syscall_64+0x6c/0x1f0 [ 2865.187837] entry_SYSCALL_64_after_hwframe+0x3d/0xa2 [ 2865.187839] RIP: 0033:0x7fc15e47abe7 [ 2865.187840] RSP: 002b:00007fc14bc5abf8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [ 2865.187842] RAX: ffffffffffffffda RBX: 00007fc14bc5ad18 RCX: 00007fc15e47abe7 [ 2865.187843] RDX: 00007fc14bc5ac60 RSI: 00000000c0186444 RDI: 0000000000000007 [ 2865.187844] RBP: 00007fc14bc5ac60 R08: 00007fc14bc5ad40 R09: 00007fc14bc5ad18 [ 2865.187845] R10: 00007fc14bc5ac40 R11: 0000000000000246 R12: 00000000c0186444 [ 2865.187845] R13: 0000000000000007 R14: 0000000000000002 R15: 0000000000000002 [ 2865.187847] Code: 75 ee 0f 0b 48 8b 83 08 02 00 00 5b c3 8b 93 b8 02 00 00 85 d2 75 c1 0f 0b eb bd 48 8b bf f8 00 00 00 e8 c6 d0 ff ff 84 c0 75 02 <0f> 0b 8b 83 b4 00 00 00 eb 90 0f 0b eb 8c 0f 0b eb ae 0f 1f 44 [ 2865.187874] ---[ end trace 0b523768a49ad76c ]--- [ 2865.187921] ------------[ cut here ]------------ [ 2865.187941] WARNING: CPU: 0 PID: 2284 at drivers/gpu/drm/amd/amdgpu/amdgpu_object.c:1032 amdgpu_bo_gpu_offset+0x8c/0x90 [amdgpu] [ 2865.187946] Modules linked in: cfg80211 rfkill binfmt_misc nfsd auth_rpcgss oid_registry nfs_acl nfs lockd grace fscache sunrpc edac_mce_amd kvm irqbypass crct10dif_pclmul crc32_pclmul ghash_clmulni_intel pcbc evdev cdc_acm aesni_intel aes_x86_64 crypto_simd glue_helper cryptd amdgpu pcspkr snd_hda_codec_cirrus fam15h_power k10temp snd_hda_codec_hdmi snd_hda_codec_generic ttm snd_hda_intel snd_hda_codec snd_hda_core snd_hwdep drm_kms_helper snd_pcm snd_timer video snd soundcore sp5100_tco drm i2c_piix4 shpchp acpi_cpufreq button fuse autofs4 ext4 crc16 mbcache jbd2 sg sd_mod hid_generic usbhid hid crc32c_intel ehci_pci ahci ehci_hcd sdhci_pci xhci_pci sdhci libahci xhci_hcd igb libata mmc_core i2c_algo_bit i2c_core usbcore scsi_mod dca ptp usb_common pps_core [ 2865.187992] CPU: 0 PID: 2284 Comm: amdgpu_cs:0 Tainted: G W 4.14.24 #1 [ 2865.187997] Hardware name: Default string Default string/GCM-B610-CS, BIOS R1.00.E1 08/01/2017 [ 2865.188001] task: ffff8803e6b2f180 task.stack: ffffc9000be94000 [ 2865.188022] RIP: 0010:amdgpu_bo_gpu_offset+0x8c/0x90 [amdgpu] [ 2865.188027] RSP: 0018:ffffc9000be97a68 EFLAGS: 00010246 [ 2865.188034] RAX: 0000000000000001 RBX: ffff8803eb797000 RCX: ffff8803eb797000 [ 2865.188038] RDX: 7fffffffffffffff RSI: 00000000001053a0 RDI: ffff8803eb4c32e8 [ 2865.188042] RBP: 00000001053a0000 R08: ffff8803e9840b90 R09: ffff8803e9840ab8 [ 2865.188046] R10: ffffffff81f26b00 R11: 0000000003f41000 R12: ffff880388445d80 [ 2865.188051] R13: ffff8803ba9e2ee0 R14: ffffffffa06079e0 R15: ffff8803ba9f2460 [ 2865.188058] FS: 00007fc14bc5b700(0000) GS:ffff8803fec00000(0000) knlGS:0000000000000000 [ 2865.188062] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 2865.188067] CR2: 00000000019c4000 CR3: 0000000398f56000 CR4: 00000000001406f0 [ 2865.188073] Call Trace: [ 2865.188101] amdgpu_uvd_cs_pass2+0x44/0x7d0 [amdgpu] [ 2865.188126] ? amdgpu_uvd_cs_pass1+0xf0/0xf0 [amdgpu] [ 2865.188158] amdgpu_uvd_cs_packets+0x153/0x170 [amdgpu] [ 2865.188183] amdgpu_uvd_ring_parse_cs+0xc9/0x140 [amdgpu] [ 2865.188208] amdgpu_cs_ioctl+0x106e/0x1a60 [amdgpu] [ 2865.188234] ? amdgpu_cs_find_mapping+0xa0/0xa0 [amdgpu] [ 2865.188248] drm_ioctl_kernel+0x65/0xb0 [drm] [ 2865.188260] drm_ioctl+0x2ce/0x380 [drm] [ 2865.188290] ? amdgpu_cs_find_mapping+0xa0/0xa0 [amdgpu] [ 2865.188301] ? update_curr+0xd8/0x130 [ 2865.188308] ? update_curr+0x5c/0x130 [ 2865.188330] amdgpu_drm_ioctl+0x46/0x80 [amdgpu] [ 2865.188347] do_vfs_ioctl+0x8b/0x5c0 [ 2865.188352] ? SyS_futex+0x71/0x170 [ 2865.188359] SyS_ioctl+0x76/0x90 [ 2865.188365] do_syscall_64+0x6c/0x1f0 [ 2865.188372] entry_SYSCALL_64_after_hwframe+0x3d/0xa2 [ 2865.188377] RIP: 0033:0x7fc15e47abe7 [ 2865.188383] RSP: 002b:00007fc14bc5abf8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [ 2865.188392] RAX: ffffffffffffffda RBX: 00007fc14bc5ad18 RCX: 00007fc15e47abe7 [ 2865.188394] RDX: 00007fc14bc5ac60 RSI: 00000000c0186444 RDI: 0000000000000007 [ 2865.188395] RBP: 00007fc14bc5ac60 R08: 00007fc14bc5ad40 R09: 00007fc14bc5ad18 [ 2865.188396] R10: 00007fc14bc5ac40 R11: 0000000000000246 R12: 00000000c0186444 [ 2865.188397] R13: 0000000000000007 R14: 0000000000000002 R15: 0000000000000002 [ 2865.188398] Code: 93 b8 02 00 00 85 d2 75 c1 0f 0b eb bd 48 8b bf f8 00 00 00 e8 c6 d0 ff ff 84 c0 75 02 0f 0b 8b 83 b4 00 00 00 eb 90 0f 0b eb 8c <0f> 0b eb ae 0f 1f 44 00 00 41 57 41 56 41 55 41 54 55 53 48 89 [ 2865.188444] ---[ end trace 0b523768a49ad76d ]---
Sounds like a known issue in kernel 4.14 to me. Please test with some newer kernel as well, probably best to test with Alex amd-staging-drm-next branch.
Thanks Christian for the info! I'll try it... This issues are fixed backporting changes to a LTS kernel version? I ask in order to check in a more recent 4.14 version instead of change to a new version...
I have same issue after upgrade from 4.14.111 to 4.14.132. Use of following branch is impossible as it has DisplayCore enabled by default and video decoding works with crucial issue on my AMD GPU chip and only 4.14 kernels don't have DC enabled
(In reply to devbazilio from comment #3) > [...] only 4.14 kernels don't have DC enabled FWIW, you can always disable DC with amdgpu.dc=0 on the kernel command line.
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.