[Bug 100460] AMDGPU system hang with Slic3r Prusa Edition

Thu Mar 30 00:05:13 UTC 2017

https://bugs.freedesktop.org/show_bug.cgi?id=100460

            Bug ID: 100460
           Summary: AMDGPU system hang with Slic3r Prusa Edition
           Product: DRI
           Version: unspecified
          Hardware: x86-64 (AMD64)
                OS: Linux (All)
            Status: NEW
          Severity: major
          Priority: medium
         Component: DRM/AMDgpu
          Assignee: dri-devel at lists.freedesktop.org
          Reporter: greg at chown.ath.cx

The 3D printing software Slic3r Prusa Edition uses OpenGL to render 3D views of
objects and a preview of the tool path.

The tool path preview causes system hangs for me with AMDGPU and RadeonSI, with
the latest mainline code of all components involved (Linux 4.11rc4, Mesa git,
LLVM svn). It seems to be memory management related. It looks like there may be
a deadlock of some sort.

Here's how to reproduce. I wasn't able to record a working trace.

1. Download Slic3r Prusa Edition 1.33.8 from
https://github.com/prusa3d/Slic3r/releases/tag/version_1.33.8 (AppImage is the
easy choice)
2. Get the model "LASER CAT - Voronoi Style" from
http://www.thingiverse.com/thing:179266/#files (e.g.
Laser_Cat_-_Voronoi_coarse.stl)
3. Start Slic3r, abort setup wizard. Default settings are fine.
4. Press "Add..." and select the .stl file
5. Press "Slice now" in the right panel
6. Select "Preview" in the bottom tab list

This results in a hanging system. The kernel's hang detection is triggered
after a while:

66929.634087] NMI watchdog: BUG: soft lockup - CPU#1 stuck for 23s!
[perl5.22.0:21405]
[66929.634087] Modules linked in: ufs qnx4 hfsplus hfs minix ntfs msdos jfs xfs
libcrc32c ccm nvram f71882fg arc4 rt2800usb rt2x00usb rt2800lib rt2x00lib
mac80211 cfg80211 input_leds joydev pci_stub vboxpci(OE) vboxnetadp(OE)
vboxnetflt(OE) vboxdrv(OE) binfmt_misc nls_iso8859_1 edac_mce_amd edac_core
kvm_amd kvm irqbypass crct10dif_pclmul crc32_pclmul ghash_clmulni_intel
snd_seq_midi pcbc snd_seq_midi_event snd_rawmidi snd_hda_codec_realtek
snd_hda_codec_generic snd_hda_codec_hdmi snd_seq snd_hda_intel snd_hda_codec
snd_hda_core snd_hwdep aesni_intel aes_x86_64 crypto_simd glue_helper cryptd
snd_pcm fam15h_power k10temp snd_seq_device i2c_piix4 snd_timer snd soundcore
shpchp tpm_infineon mac_hid cpuid msr parport_pc ppdev lp parport autofs4 btrfs
xor raid6_pq hid_generic usbhid hid uas usb_storage amdkfd
[66929.634087]  amd_iommu_v2 amdgpu i2c_algo_bit ttm drm_kms_helper syscopyarea
sysfillrect sysimgblt fb_sys_fops drm ahci r8169 libahci mii wmi fjes video
[66929.634087] CPU: 1 PID: 21405 Comm: perl5.22.0 Tainted: G           OE  
4.11.0-041100rc1-generic #201703051731
[66929.634087] Hardware name: MSI MS-7721/A68HM-P33 (MS-7721), BIOS V34.4
12/15/2014
[66929.634087] task: ffffa0f8eb574380 task.stack: ffffb6f146c1c000
[66929.634087] RIP: 0010:_raw_spin_unlock_irqrestore+0x15/0x20
[66929.634087] RSP: 0018:ffffb6f146c1f780 EFLAGS: 00000282 ORIG_RAX:
ffffffffffffff10
[66929.634087] RAX: 0000000000000002 RBX: 0000000000000000 RCX:
0000000000000000
[66929.634087] RDX: 0000000000000000 RSI: 0000000000000282 RDI:
0000000000000282
[66929.634087] RBP: ffffb6f146c1f780 R08: 0000000000000000 R09:
ffffa0f8cd1bd380
[66929.634087] R10: ffffb6f14571e628 R11: 000000000001ef18 R12:
0000000000000001
[66929.634087] R13: ffffa0f8cd1bd380 R14: ffffa0f8eb512940 R15:
0000000000000001
[66929.634087] FS:  00007f6b3c771700(0000) GS:ffffa0f8fec80000(0000)
knlGS:0000000000000000
[66929.634087] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[66929.634087] CR2: 00007f69a3195000 CR3: 00000003d9548000 CR4:
00000000000406e0
[66929.634087] Call Trace:
[66929.634087]  alloc_iova+0x21c/0x240
[66929.634087]  alloc_iova_fast+0xa6/0x200
[66929.634087]  dma_ops_alloc_iova.isra.23+0x6b/0x80
[66929.634087]  __map_single.isra.24+0x49/0x1b0
[66929.634087]  map_page+0x64/0x80
[66929.634087]  amdgpu_ttm_tt_populate+0xe9/0x270 [amdgpu]
[66929.634087]  ttm_tt_bind+0x2b/0x60 [ttm]
[66929.634087]  ttm_bo_handle_move_mem+0x535/0x5b0 [ttm]
[66929.634087]  ? shmem_alloc_inode+0x1a/0x30
[66929.634087]  ttm_bo_validate+0x13e/0x150 [ttm]
[66929.634087]  ttm_bo_init+0x243/0x430 [ttm]
[66929.634087]  amdgpu_bo_create_restricted+0x4ae/0x5b0 [amdgpu]
[66929.634087]  ? amdgpu_update_memory_usage+0xe0/0xe0 [amdgpu]
[66929.634087]  amdgpu_bo_create+0xed/0x1f0 [amdgpu]
[66929.634087]  amdgpu_gem_object_create+0xba/0x150 [amdgpu]
[66929.634087]  amdgpu_gem_create_ioctl+0xa4/0x130 [amdgpu]
[66929.634087]  drm_ioctl+0x209/0x4c0 [drm]
[66929.634087]  ? amdgpu_gem_object_close+0x130/0x130 [amdgpu]
[66929.634087]  ? __handle_mm_fault+0x953/0x10e0
[66929.634087]  ? do_mmap+0x445/0x510
[66929.634087]  ? common_mmap+0x45/0x50
[66929.634087]  amdgpu_drm_ioctl+0x4f/0x90 [amdgpu]
[66929.634087]  do_vfs_ioctl+0xa3/0x600
[66929.634087]  ? handle_mm_fault+0xd0/0x240
[66929.634087]  ? __check_object_size+0x100/0x19d
[66929.634087]  SyS_ioctl+0x79/0x90
[66929.634087]  entry_SYSCALL_64_fastpath+0x1e/0xad
[66929.634087] RIP: 0033:0x7f6b3b31e357
[66929.634087] RSP: 002b:00007fff75cc5b98 EFLAGS: 00000246 ORIG_RAX:
0000000000000010
[66929.634087] RAX: ffffffffffffffda RBX: 0000000000000000 RCX:
00007f6b3b31e357
[66929.634087] RDX: 00007fff75cc5be0 RSI: 00000000c0206440 RDI:
000000000000000e
[66929.634087] RBP: 0000000000c3c000 R08: 00000000072c11c0 R09:
0000000000000001
[66929.634087] R10: 00007f6b3b5e5bc8 R11: 0000000000000246 R12:
000000000000c000
[66929.634087] R13: 0000000006bac130 R14: 00007f6a30c3a010 R15:
0000000000000002
[66929.634087] Code: 08 74 0b 65 81 25 58 dc b3 59 ff ff ff 7f 89 d0 5d c3 90
90 90 90 0f 1f 44 00 00 55 48 89 e5 c6 07 00 0f 1f 40 00 48 89 f7 57 9d <0f> 1f
44 00 00 5d c3 0f 1f 40 00 0f 1f 44 00 00 55 48 89 e5 c6

The application possibly might not behave well, but this should never happen
nonetheless.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/dri-devel/attachments/20170330/2b7fe60c/attachment-0001.html>