Bug 106970

Summary: [drm:drm_sched_entity_cleanup [gpu_sched]] *ERROR* fence add callback failed (-22)
Product: DRI Reporter: mikhail.v.gavrilov
Component: DRM/AMDgpuAssignee: Default DRI bug account <dri-devel>
Status: RESOLVED MOVED QA Contact:
Severity: normal    
Priority: medium CC: andrey.grodzovsky
Version: XOrg git   
Hardware: Other   
OS: All   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
dmesg
none
system log none

Description mikhail.v.gavrilov 2018-06-20 04:11:02 UTC
Created attachment 140240 [details]
dmesg

[39383.734817] WARNING: CPU: 5 PID: 30215 at drivers/dma-buf/dma-fence.c:248 dma_fence_add_callback+0x176/0x1f0
[39383.734835] Modules linked in: macvtap macvlan tap tun fuse rfcomm devlink nf_conntrack_netbios_ns nf_conntrack_broadcast xt_CT ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 xt_conntrack ip_set nfnetlink ebtable_nat ebtable_broute bridge stp llc ip6table_nat nf_conntrack_ipv6 nf_defrag_ipv6 nf_nat_ipv6 ip6table_mangle ip6table_raw ip6table_security iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_raw iptable_security ebtable_filter ebtables ip6table_filter ip6_tables cmac bnep sunrpc xfs vfat fat libcrc32c intel_rapl x86_pkg_temp_thermal hid_logitech_hidpp intel_powerclamp coretemp kvm_intel kvm iTCO_wdt iTCO_vendor_support irqbypass crct10dif_pclmul gpio_ich crc32_pclmul ppdev ghash_clmulni_intel intel_cstate intel_uncore intel_rapl_perf snd_hda_codec_realtek
[39383.734912]  snd_hda_codec_generic snd_hda_codec_hdmi snd_usb_audio btusb btrtl btbcm btintel snd_hda_intel bluetooth snd_usbmidi_lib snd_hda_codec snd_rawmidi joydev snd_hda_core snd_hwdep hid_logitech_dj snd_seq snd_seq_device snd_pcm ecdh_generic rfkill snd_timer snd mei_me soundcore i2c_i801 mei lpc_ich parport_pc parport video binfmt_misc hid_sony ff_memless amdkfd amd_iommu_v2 amdgpu chash i2c_algo_bit gpu_sched drm_kms_helper ttm drm crc32c_intel r8169 mii
[39383.734967] CPU: 5 PID: 30215 Comm: gallium_drv:0 Not tainted 4.18.0-0.rc0.git9.1.fc29.x86_64 #1
[39383.734968] Hardware name: Gigabyte Technology Co., Ltd. Z87M-D3H/Z87M-D3H, BIOS F11 08/12/2014
[39383.734972] RIP: 0010:dma_fence_add_callback+0x176/0x1f0
[39383.734973] Code: f6 74 1b 49 8b 06 49 8b 7e 08 49 83 c6 18 48 89 de e8 ce 9a 71 00 49 8b 06 48 85 c0 75 e8 65 ff 0d 8f c8 92 6c e9 0f ff ff ff <0f> 0b 41 bc ea ff ff ff e9 eb fe ff ff 65 ff 05 76 c8 92 6c 48 8b 
[39383.735039] RSP: 0018:ffffa0354f733c98 EFLAGS: 00010246
[39383.735042] RAX: 0000000000000004 RBX: ffff95348d6e3818 RCX: 00000000000000ff
[39383.735043] RDX: ffffffffc01e6e00 RSI: ffff95348d6e3818 RDI: 0000000000000000
[39383.735044] RBP: ffff9536ae501580 R08: 0000798cd856a73b R09: 0000000000000000
[39383.735046] R10: 0000000000000000 R11: 0000000000000000 R12: ffff953654aeb180
[39383.735047] R13: ffffffffc01e6e00 R14: ffff9536ae5015e8 R15: 000000000000801a
[39383.735049] FS:  0000000000000000(0000) GS:ffff953afe000000(0000) knlGS:0000000000000000
[39383.735050] CS:  0010 DS: 002b ES: 002b CR0: 0000000080050033
[39383.735052] CR2: 00000000f2488ae4 CR3: 000000053c612001 CR4: 00000000001626e0
[39383.735053] Call Trace:
[39383.735062]  ? drm_sched_entity_wakeup+0x60/0x60 [gpu_sched]
[39383.735065]  drm_sched_entity_cleanup+0x10a/0x160 [gpu_sched]
[39383.735143]  amdgpu_ctx_mgr_entity_cleanup+0xb9/0xf0 [amdgpu]
[39383.735178]  amdgpu_ctx_mgr_fini+0x35/0xb0 [amdgpu]
[39383.735203]  amdgpu_driver_postclose_kms+0x138/0x290 [amdgpu]
[39383.735208]  ? __mutex_unlock_slowpath+0x4b/0x2b0
[39383.735224]  drm_release+0x293/0x3f0 [drm]
[39383.735231]  __fput+0xca/0x250
[39383.735236]  task_work_run+0xa1/0xd0
[39383.735242]  do_exit+0x3fe/0xc70
[39383.735246]  ? trace_hardirqs_on_caller+0xed/0x180
[39383.735251]  __ia32_sys_exit+0x17/0x20
[39383.735254]  do_fast_syscall_32+0xac/0x340
[39383.735257]  entry_SYSENTER_compat+0x84/0x96
[39383.735266] irq event stamp: 12604
[39383.735269] hardirqs last  enabled at (12603): [<ffffffff9315a89c>] ktime_get+0xec/0x130
[39383.735272] hardirqs last disabled at (12604): [<ffffffff93c01389>] error_entry+0x89/0x110
[39383.735274] softirqs last  enabled at (12482): [<ffffffff93e0038c>] __do_softirq+0x38c/0x4f7
[39383.735276] softirqs last disabled at (12465): [<ffffffff930c1a51>] irq_exit+0x101/0x110
[39383.735278] ---[ end trace 403f5176c1d37060 ]---
[39383.735281] [drm:drm_sched_entity_cleanup [gpu_sched]] *ERROR* fence add callback failed (-22)
Comment 1 mikhail.v.gavrilov 2018-06-20 04:29:09 UTC
Created attachment 140241 [details]
system log
Comment 2 Andrey Grodzovsky 2018-06-23 23:39:06 UTC
What's the scenario ?

Please check if your git tree contains this change - 
drm/scheduler: Avoid using wait_event_killable for dying process

Andrey
Comment 3 Martin Peres 2019-11-19 08:41:42 UTC
-- 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/423.

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.