Bug 91172

Summary: [4.1.0][radeonsi][BONAIRE] Playing Ark Survival Evolved (new game) - IH ring buffer overflow
Product: DRI Reporter: Shawn Starr <shawn.starr>
Component: DRM/RadeonAssignee: Default DRI bug account <dri-devel>
Status: RESOLVED WONTFIX QA Contact:
Severity: normal    
Priority: medium    
Version: DRI git   
Hardware: Other   
OS: All   
Whiteboard:
i915 platform: i915 features:

Description Shawn Starr 2015-07-01 05:29:51 UTC 
Mesa git master (2015-06-30)
LLVM (3.7) trunk (2015-06-30
kernel: 4.1.0-1.fc23.x86_64

[78755.225438] radeon 0000:01:00.0: IH ring buffer overflow (0x00000020, 0x00000010, 0x00000030)
[79732.915343] [TTM] Failed to find memory space for buffer 0xffff88074fef7468 eviction
[79732.915718] [TTM] No space for ffff88074fef7468 (65536 pages, 262144K, 256M)
[79732.916085] [TTM]   placement[0]=0x00060002 (1)
[79732.916322] [TTM]     has_type: 1
[79732.916512] [TTM]     use_type: 1
[79732.916703] [TTM]     flags: 0x0000000A
[79732.916945] [TTM]     gpu_offset: 0x80000000
[79732.917209] [TTM]     size: 262144
[79732.917374] [TTM]     available_caching: 0x00070000
[79732.917609] [TTM]     default_caching: 0x00010000
[79737.559442] [TTM] Failed to find memory space for buffer 0xffff8801038e4868 eviction
[79737.559817] [TTM] No space for ffff8801038e4868 (23873 pages, 95492K, 93M)
[79737.560180] [TTM]   placement[0]=0x00060002 (1)
[79737.560489] [TTM]     has_type: 1
[79737.560653] [TTM]     use_type: 1
[79737.560815] [TTM]     flags: 0x0000000A
[79737.561027] [TTM]     gpu_offset: 0x80000000
[79737.561264] [TTM]     size: 262144
[79737.561458] [TTM]     available_caching: 0x00070000
[79737.561695] [TTM]     default_caching: 0x00010000
[79737.923045] [TTM] Failed to find memory space for buffer 0xffff88074fef7468 eviction
[79737.923424] [TTM] No space for ffff88074fef7468 (65536 pages, 262144K, 256M)
[79737.923779] [TTM]   placement[0]=0x00060002 (1)
[79737.924055] [TTM]     has_type: 1
[79737.924249] [TTM]     use_type: 1
[79737.924412] [TTM]     flags: 0x0000000A
[79737.924624] [TTM]     gpu_offset: 0x80000000
[79737.924830] [TTM]     size: 262144
[79737.924998] [TTM]     available_caching: 0x00070000
[79737.925291] [TTM]     default_caching: 0x00010000
[79738.780552] [TTM] Failed to find memory space for buffer 0xffff88074fef7468 eviction
[79738.780955] [TTM] No space for ffff88074fef7468 (65536 pages, 262144K, 256M)
[79738.781325] [TTM]   placement[0]=0x00060002 (1)
[79738.781587] [TTM]     has_type: 1
[79738.781766] [TTM]     use_type: 1
[79738.781929] [TTM]     flags: 0x0000000A
[79738.782143] [TTM]     gpu_offset: 0x80000000
[79738.782351] [TTM]     size: 262144
[79738.782515] [TTM]     available_caching: 0x00070000
[79738.782751] [TTM]     default_caching: 0x00010000


Kernel then killed CPU:

[80228.890501] NMI watchdog: BUG: soft lockup - CPU#3 stuck for 22s! [RenderThread 1:24470]
[80228.890897] Modules linked in: nfnetlink_queue nfnetlink_log nfnetlink bluetooth tun ebtable_filter ebtables ip6table_filter ip6_tables bridge pl2303 intel_rapl iosf_mbi x86_pkg_temp_thermal coretemp kvm_intel snd_hda_codec_realtek snd_hda_codec_generic uvcvideo arc4 snd_usb_audio kvm videobuf2_vmalloc videobuf2_core videobuf2_memops snd_usbmidi_lib crct10dif_pclmul iwlmvm crc32_pclmul crc32c_intel ghash_clmulni_intel snd_hda_codec_hdmi snd_hda_intel snd_hda_controller snd_hda_codec mac80211 v4l2_common videodev snd_rawmidi vfat fat snd_soc_rt5640 snd_soc_rl6231 snd_soc_core media snd_hda_core snd_compress snd_hwdep ppdev iTCO_wdt snd_pcm_dmaengine snd_seq snd_seq_device snd_pcm iTCO_vendor_support iwlwifi cfg80211 dell_wmi sparse_keymap dell_laptop dcdbas i8k snd_timer snd rfkill mei_me joydev regmap_i2c
[80228.894854]  snd_soc_sst_acpi soundcore i2c_i801 i2c_designware_platform ie31200_edac dw_dmac parport_pc lpc_ich mei parport shpchp video i2c_designware_core mfd_core edac_core dw_dmac_core wmi dell_smo8800 nfsd auth_rpcgss nfs_acl lockd grace sunrpc binfmt_misc 8021q garp stp llc mrp amdkfd amd_iommu_v2 radeon i2c_algo_bit drm_kms_helper ttm e1000e drm sdhci_acpi sdhci_pci serio_raw ptp sdhci pps_core mmc_core i2c_hid
[80228.896947] CPU: 3 PID: 24470 Comm: RenderThread 1 Not tainted 4.1.0-1.fc23.x86_64 #1
[80228.897323] Hardware name: Dell Inc. Precision M6800/05NG6V, BIOS A14 05/24/2015
[80228.897677] task: ffff8806df86eca0 ti: ffff8800c6718000 task.ti: ffff8800c6718000
[80228.898037] RIP: 0010:[<ffffffff817a08a2>]  [<ffffffff817a08a2>] _raw_spin_unlock_irqrestore+0x12/0x20
[80228.898495] RSP: 0018:ffff8800c671b2e8  EFLAGS: 00000257
[80228.898751] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
[80228.899096] RDX: 0000000000000002 RSI: 0000000000000257 RDI: 0000000000000257
[80228.899438] RBP: ffff8800c671b2e8 R08: 00000000000cec30 R09: 0000000000000002
[80228.899782] R10: 0000000000001fc8 R11: ffffc900048a1538 R12: 0000000000000001
[80228.900125] R13: 0000000000000000 R14: 0000000100400040 R15: 000000000082e000
[80228.900471] FS:  00007f3f8d8ed700(0000) GS:ffff88082dcc0000(0000) knlGS:0000000000000000
[80228.900858] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[80228.901136] CR2: 00000000eef4f000 CR3: 00000002106de000 CR4: 00000000001407e0
[80228.901479] Stack:
[80228.901582]  ffff8800c671b368 ffffffff814cfe90 0000000000000257 0000000000000002
[80228.901972]  0000000000000002 ffff88022a50c480 00000000000fffff 01ffffff00000001
[80228.902363]  ffff88022a50c480 00000000fc39fd85 00000000000cec31 0000007ffffff000
[80228.902752] Call Trace:
[80228.902879]  [<ffffffff814cfe90>] alloc_iova+0x220/0x290
[80228.903141]  [<ffffffff814dae35>] intel_alloc_iova+0x95/0xe0
[80228.903416]  [<ffffffff814dec37>] __intel_map_single+0x97/0x190
[80228.903704]  [<ffffffff814ded69>] intel_map_page+0x39/0x40
[80228.903987]  [<ffffffffa018af75>] radeon_ttm_tt_populate+0x105/0x2b0 [radeon]
[80228.904337]  [<ffffffffa012adf0>] ttm_tt_bind+0x40/0x80 [ttm]
[80228.904620]  [<ffffffffa012d1df>] ttm_bo_handle_move_mem+0x5df/0x670 [ttm]
[80228.904961]  [<ffffffffa012da6e>] ? ttm_bo_mem_space+0x28e/0x3b0 [ttm]
[80228.905286]  [<ffffffffa012d3ca>] ttm_bo_evict+0x15a/0x350 [ttm]
[80228.905589]  [<ffffffffa00961d3>] ? drm_vma_offset_add+0xc3/0xe0 [drm]
[80228.905915]  [<ffffffffa012d766>] ttm_mem_evict_first+0x1a6/0x220 [ttm]
[80228.906242]  [<ffffffffa012da92>] ttm_bo_mem_space+0x2b2/0x3b0 [ttm]
[80228.906559]  [<ffffffffa012dffa>] ttm_bo_validate+0x1aa/0x220 [ttm]
[80228.906869]  [<ffffffffa01320fb>] ? ttm_eu_reserve_buffers+0x16b/0x320 [ttm]
[80228.907223]  [<ffffffffa018cd57>] radeon_bo_list_validate+0xd7/0x230 [radeon]
[80228.907577]  [<ffffffffa01a57db>] radeon_cs_parser_relocs+0x3ab/0x480 [radeon]
[80228.907944]  [<ffffffffa01a6263>] radeon_cs_ioctl+0x273/0x7b0 [radeon]
[80228.908265]  [<ffffffff8120b3df>] ? __slab_free+0xbf/0x260
[80228.908530]  [<ffffffff811ee582>] ? __vunmap+0xa2/0x100
[80228.908793]  [<ffffffffa007b9df>] drm_ioctl+0x12f/0x620 [drm]
[80228.909086]  [<ffffffffa01a5ff0>] ? radeon_cs_parser_init+0x4d0/0x4d0 [radeon]
[80228.909438]  [<ffffffff811df7d6>] ? handle_mm_fault+0x506/0x1840
[80228.909737]  [<ffffffffa016b04c>] radeon_drm_ioctl+0x4c/0x80 [radeon]
[80228.910054]  [<ffffffff8123fc66>] do_vfs_ioctl+0x2c6/0x4d0
[80228.910319]  [<ffffffff81068c41>] ? __do_page_fault+0x1b1/0x440
[80228.910607]  [<ffffffff8123fef1>] SyS_ioctl+0x81/0xa0
[80228.910852]  [<ffffffff81068eff>] ? do_page_fault+0x2f/0x80
[80228.911124]  [<ffffffff817a0f2e>] system_call_fastpath+0x12/0x71
[80228.911414] Code: d0 f0 0f b1 0f 39 d0 75 e5 b8 01 00 00 00 5d c3 66 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 55 48 89 e5 66 83 07 01 48 89 f7 57 9d <0f> 1f 44 00 00 5d c3 0f 1f 80 00 00 00 00 0f 1f 44 00 00 55 48 
[80256.911785] NMI watchdog: BUG: soft lockup - CPU#3 stuck for 23s! [RenderThread 1:24470]
[80256.912181] Modules linked in: nfnetlink_queue nfnetlink_log nfnetlink bluetooth tun ebtable_filter ebtables ip6table_filter ip6_tables bridge pl2303 intel_rapl iosf_mbi x86_pkg_temp_thermal coretemp kvm_intel snd_hda_codec_realtek snd_hda_codec_generic uvcvideo arc4 snd_usb_audio kvm videobuf2_vmalloc videobuf2_core videobuf2_memops snd_usbmidi_lib crct10dif_pclmul iwlmvm crc32_pclmul crc32c_intel ghash_clmulni_intel snd_hda_codec_hdmi snd_hda_intel snd_hda_controller snd_hda_codec mac80211 v4l2_common videodev snd_rawmidi vfat fat snd_soc_rt5640 snd_soc_rl6231 snd_soc_core media snd_hda_core snd_compress snd_hwdep ppdev iTCO_wdt snd_pcm_dmaengine snd_seq snd_seq_device snd_pcm iTCO_vendor_support iwlwifi cfg80211 dell_wmi sparse_keymap dell_laptop dcdbas i8k snd_timer snd rfkill mei_me joydev regmap_i2c
[80256.916196]  snd_soc_sst_acpi soundcore i2c_i801 i2c_designware_platform ie31200_edac dw_dmac parport_pc lpc_ich mei parport shpchp video i2c_designware_core mfd_core edac_core dw_dmac_core wmi dell_smo8800 nfsd auth_rpcgss nfs_acl lockd grace sunrpc binfmt_misc 8021q garp stp llc mrp amdkfd amd_iommu_v2 radeon i2c_algo_bit drm_kms_helper ttm e1000e drm sdhci_acpi sdhci_pci serio_raw ptp sdhci pps_core mmc_core i2c_hid
[80256.918314] CPU: 3 PID: 24470 Comm: RenderThread 1 Tainted: G             L  4.1.0-1.fc23.x86_64 #1
[80256.918759] Hardware name: Dell Inc. Precision M6800/05NG6V, BIOS A14 05/24/2015
[80256.919120] task: ffff8806df86eca0 ti: ffff8800c6718000 task.ti: ffff8800c6718000
[80256.919483] RIP: 0010:[<ffffffff817a08a2>]  [<ffffffff817a08a2>] _raw_spin_unlock_irqrestore+0x12/0x20
[80256.919951] RSP: 0018:ffff8800c671b2e8  EFLAGS: 00000257
[80256.920211] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
[80256.920559] RDX: 0000000000000002 RSI: 0000000000000257 RDI: 0000000000000257
[80256.920906] RBP: ffff8800c671b2e8 R08: 00000000a0670a01 R09: 0000000000000002
[80256.921254] R10: ffff8807a0670a40 R11: ffffc900048a2000 R12: 0000000000000001
[80256.921600] R13: 0000000000000000 R14: 0000000180400040 R15: 000000000082e000
[80256.921949] FS:  00007f3f8d8ed700(0000) GS:ffff88082dcc0000(0000) knlGS:0000000000000000
[80256.922343] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[80256.922624] CR2: 00000000eef4f000 CR3: 00000002106de000 CR4: 00000000001407e0
[80256.922971] Stack:
[80256.923077]  ffff8800c671b368 ffffffff814cfe90 0000000000000257 0000000000000002
[80256.923470]  0000000000000002 ffff88022a50c480 00000000000fffff 01ffffff00000001
[80256.923861]  ffff88022a50c480 00000000fc39fd85 0000000007ff9f53 0000007ffffff000
[80256.924254] Call Trace:
[80256.924385]  [<ffffffff814cfe90>] alloc_iova+0x220/0x290
[80256.924652]  [<ffffffff814dae35>] intel_alloc_iova+0x95/0xe0
[80256.924932]  [<ffffffff814dec37>] __intel_map_single+0x97/0x190
[80256.925230]  [<ffffffff814ded69>] intel_map_page+0x39/0x40
[80256.925530]  [<ffffffffa018af75>] radeon_ttm_tt_populate+0x105/0x2b0 [radeon]
[80256.925893]  [<ffffffffa012adf0>] ttm_tt_bind+0x40/0x80 [ttm]
[80256.926187]  [<ffffffffa012d1df>] ttm_bo_handle_move_mem+0x5df/0x670 [ttm]
[80256.926534]  [<ffffffffa012da6e>] ? ttm_bo_mem_space+0x28e/0x3b0 [ttm]
[80256.926857]  [<ffffffffa012d3ca>] ttm_bo_evict+0x15a/0x350 [ttm]
[80256.927173]  [<ffffffffa00961d3>] ? drm_vma_offset_add+0xc3/0xe0 [drm]
[80256.927506]  [<ffffffffa012d766>] ttm_mem_evict_first+0x1a6/0x220 [ttm]
[80256.927837]  [<ffffffffa012da92>] ttm_bo_mem_space+0x2b2/0x3b0 [ttm]
[80256.928160]  [<ffffffffa012dffa>] ttm_bo_validate+0x1aa/0x220 [ttm]
[80256.928478]  [<ffffffffa01320fb>] ? ttm_eu_reserve_buffers+0x16b/0x320 [ttm]
[80256.928836]  [<ffffffffa018cd57>] radeon_bo_list_validate+0xd7/0x230 [radeon]
[80256.929202]  [<ffffffffa01a57db>] radeon_cs_parser_relocs+0x3ab/0x480 [radeon]
[80256.929575]  [<ffffffffa01a6263>] radeon_cs_ioctl+0x273/0x7b0 [radeon]
[80256.929896]  [<ffffffff8120b3df>] ? __slab_free+0xbf/0x260
[80256.930165]  [<ffffffff811ee582>] ? __vunmap+0xa2/0x100
[80256.930436]  [<ffffffffa007b9df>] drm_ioctl+0x12f/0x620 [drm]
[80256.930736]  [<ffffffffa01a5ff0>] ? radeon_cs_parser_init+0x4d0/0x4d0 [radeon]
[80256.931096]  [<ffffffff811df7d6>] ? handle_mm_fault+0x506/0x1840
[80256.931402]  [<ffffffffa016b04c>] radeon_drm_ioctl+0x4c/0x80 [radeon]
[80256.931724]  [<ffffffff8123fc66>] do_vfs_ioctl+0x2c6/0x4d0
[80256.931994]  [<ffffffff81068c41>] ? __do_page_fault+0x1b1/0x440
[80256.932283]  [<ffffffff8123fef1>] SyS_ioctl+0x81/0xa0
[80256.932536]  [<ffffffff81068eff>] ? do_page_fault+0x2f/0x80
[80256.932814]  [<ffffffff817a0f2e>] system_call_fastpath+0x12/0x71
[80256.933115] Code: d0 f0 0f b1 0f 39 d0 75 e5 b8 01 00 00 00 5d c3 66 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 55 48 89 e5 66 83 07 01 48 89 f7 57 9d <0f> 1f 44 00 00 5d c3 0f 1f 80 00 00 00 00 0f 1f 44 00 00 55 48
Comment 1 Shawn Starr 2015-07-02 19:40:32 UTC 
A workaround mentioned on IRC is to set the GART size to 2048. This stops the overflow.

There is a fix to make this 2048 for GPUs with bigger than 1GB of vram coming down the pipe so closing this.

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.