Bug 91315 - When running JohnTheRipper OpenCL tests: The kernel rejected CS, see dmesg for more information. / [mi] EQ overflowing. Additional events will be discarded until existing events are processed.
Summary: When running JohnTheRipper OpenCL tests: The kernel rejected CS, see dmesg fo...
Status: NEW
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/Gallium/radeonsi (show other bugs)
Version: 10.6
Hardware: Other All
: medium normal
Assignee: Default DRI bug account
QA Contact: Default DRI bug account
URL:
Whiteboard:
Keywords:
Depends on:
Blocks: 99553
  Show dependency treegraph
 
Reported: 2015-07-12 09:18 UTC by Dennis Schridde
Modified: 2017-03-15 14:16 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments
Last lines of output from: john --test=0 --verbosity=5 (15.88 KB, text/plain)
2015-07-12 09:18 UTC, Dennis Schridde
Details
Patch to JohnTheRipper (689 bytes, patch)
2015-07-12 09:20 UTC, Dennis Schridde
Details | Splinter Review
Xorg.0.log (324.60 KB, text/plain)
2015-07-12 09:21 UTC, Dennis Schridde
Details
dmesg (251.95 KB, text/plain)
2015-07-12 09:21 UTC, Dennis Schridde
Details
lspci -v (10.97 KB, text/plain)
2015-07-12 09:22 UTC, Dennis Schridde
Details
Linux kernel config (100.51 KB, text/plain)
2015-07-12 09:22 UTC, Dennis Schridde
Details
emerge --info media-libs/mesa (7.28 KB, text/plain)
2015-07-12 09:23 UTC, Dennis Schridde
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Dennis Schridde 2015-07-12 09:18:36 UTC
Created attachment 117064 [details]
Last lines of output from: john --test=0 --verbosity=5

I ran `john --test=0 --verbosity=5` from JohnTheRipper@e0e854e01e46e3e58dcec89d0e87b651c448cdff (plus a local patch as suggested in https://github.com/magnumripper/JohnTheRipper/issues/1470#issuecomment-120698629) as a regular non-root user, which made X crash.

The last lines of output from `john --test=0 --verbosity=5`, when run as a regular non-root user):
```
[...]
Testing: mscash-opencl, M$ Cache Hash [MD4 OpenCL]... Options used: -I /usr/share/john/kernels -cl-mad-enable -D__GPU__ -DDEVICE_INFO=10 -DDEV_VER_MAJOR=10 -DDEV_VER_MINOR=6 -D_OPENCL_COMPILER -D NUM_INT_KEYS=1 -D IS_STATIC_GPU_MASK=0 -D $
Build log: input.cl:365:18: warning: unknown attribute 'max_constant_size' ignored
input.cl:375:18: warning: unknown attribute 'max_constant_size' ignored

Self test GWS: 64, LWS: 8
FAILED (cmp_all(0))

Options used: -I /usr/share/john/kernels -cl-mad-enable -D__GPU__ -DDEVICE_INFO=10 -DDEV_VER_MAJOR=10 -DDEV_VER_MINOR=6 -D_OPENCL_COMPILER -D SALT_BUFFER_SIZE=260
Build log: input.cl:586:17: warning: unknown attribute 'max_constant_size' ignored

binary size 107038
Error creating binary file $JOHN/kernels/pbkdf2_kernel_-D_SALT_BUFFER_SIZE=260_-DDEV_VER_MAJOR=10_-DDEV_VER_MINOR=6AMD_KAVERI_0.bin
Device 0  GWS: 8192, LWS: 8
Testing: mscash2-opencl, MS Cache Hash 2 (DCC2) [PBKDF2-SHA1 OpenCL]... radeon: The kernel rejected CS, see dmesg for more information.
radeon: The kernel rejected CS, see dmesg for more information.
radeon: The kernel rejected CS, see dmesg for more information.
radeon: The kernel rejected CS, see dmesg for more information.
radeon: The kernel rejected CS, see dmesg for more information.
radeon: The kernel rejected CS, see dmesg for more information.
radeon: The kernel rejected CS, see dmesg for more information.
radeon: Failed to allocate virtual address for buffer:
radeon:    size      : 131072 bytes
radeon:    alignment : 4096 bytes
radeon:    domains   : 2
radeon:    va        : 0x00000000020b2000
radeon: Failed to allocate virtual address for buffer:
radeon:    size      : 131072 bytes
radeon:    alignment : 4096 bytes
radeon:    domains   : 2
radeon:    va        : 0x00000000020b2000
Bus error
```

X logs:
```
[...]
(EE) [mi] EQ overflowing.  Additional events will be discarded until existing events are processed.
(EE)
(EE) Backtrace:
(EE) 0: /usr/bin/X (xorg_backtrace+0x4f) [0x586aff]
(EE) 1: /usr/bin/X (mieqEnqueue+0x23b) [0x5690fb]
(EE) 2: /usr/bin/X (QueuePointerEvents+0x5a) [0x44f57a]
(EE) 3: /usr/lib64/xorg/modules/input/evdev_drv.so (0x7fa4741c7000+0x5f17) [0x7fa4741ccf17]
(EE) 4: /usr/lib64/xorg/modules/input/evdev_drv.so (0x7fa4741c7000+0x679d) [0x7fa4741cd79d]
(EE) 5: /usr/bin/X (0x400000+0x74978) [0x474978]
(EE) 6: /usr/bin/X (0x400000+0x9d350) [0x49d350]
(EE) 7: /lib64/libc.so.6 (0x308b000000+0x336a0) [0x308b0336a0]
(EE) 8: /lib64/libc.so.6 (ioctl+0x7) [0x308b0e0c67]
(EE) 9: /usr/lib64/libdrm.so.2 (drmIoctl+0x28) [0x3bb1c04428]
(EE) 10: /usr/lib64/libdrm.so.2 (drmCommandWrite+0x1b) [0x3bb1c06ff3]
(EE) 11: /usr/lib64/dri/radeonsi_dri.so (0x7fa478037000+0x466944) [0x7fa47849d944]
(EE) 12: /usr/lib64/dri/radeonsi_dri.so (0x7fa478037000+0x467ff2) [0x7fa47849eff2]
(EE) 13: /usr/lib64/dri/radeonsi_dri.so (0x7fa478037000+0x4794ff) [0x7fa4784b04ff]
(EE) 14: /usr/lib64/dri/radeonsi_dri.so (0x7fa478037000+0x1bdec1) [0x7fa4781f4ec1]
(EE) 15: /usr/lib64/dri/radeonsi_dri.so (0x7fa478037000+0x13bb1c) [0x7fa478172b1c]
(EE) 16: /usr/lib64/dri/radeonsi_dri.so (0x7fa478037000+0x1c0f2e) [0x7fa4781f7f2e]
(EE) 17: /usr/lib64/dri/radeonsi_dri.so (0x7fa478037000+0x13cdbc) [0x7fa478173dbc]
(EE) 18: /usr/lib64/dri/radeonsi_dri.so (0x7fa478037000+0x13d012) [0x7fa478174012]
(EE) 19: /usr/lib64/xorg/modules/libglamoregl.so (0x7fa478a47000+0x215b6) [0x7fa478a685b6]
(EE) 20: /usr/lib64/xorg/modules/libglamoregl.so (0x7fa478a47000+0xb2df) [0x7fa478a522df]
(EE) 21: /usr/lib64/xorg/modules/libglamoregl.so (0x7fa478a47000+0xbb6a) [0x7fa478a52b6a]
(EE) 22: /usr/bin/X (miCopyRegion+0x1a7) [0x5675a7]
(EE) 23: /usr/bin/X (miDoCopy+0x43e) [0x567b2e]
(EE) 24: /usr/lib64/xorg/modules/libglamoregl.so (0x7fa478a47000+0xbc04) [0x7fa478a52c04]
(EE) 25: /usr/bin/X (0x400000+0x114a72) [0x514a72]
(EE) 26: /usr/bin/X (0x400000+0x343ae) [0x4343ae]
(EE) 27: /usr/bin/X (0x400000+0x3829f) [0x43829f]
(EE) 28: /usr/bin/X (0x400000+0x3c32a) [0x43c32a]
(EE) 29: /lib64/libc.so.6 (__libc_start_main+0xf0) [0x308b020040]
(EE) 30: /usr/bin/X (0x400000+0x26d2e) [0x426d2e]
(EE)
(EE) [mi] These backtraces from mieqEnqueue may point to a culprit higher up the stack.
(EE) [mi] mieq is *NOT* the cause.  It is a victim.
(EE) [mi] EQ overflow continuing.  100 events have been dropped.
[...]
```
Comment 1 Dennis Schridde 2015-07-12 09:20:40 UTC
Created attachment 117065 [details] [review]
Patch to JohnTheRipper

Patch to JohnTheRipper suggested by @magnumripper in https://github.com/magnumripper/JohnTheRipper/issues/1470#issuecomment-120698629 to workaround kernel/mesa crash when running OpenCL DES kernel
Comment 2 Dennis Schridde 2015-07-12 09:21:01 UTC
Created attachment 117066 [details]
Xorg.0.log
Comment 3 Dennis Schridde 2015-07-12 09:21:21 UTC
Created attachment 117067 [details]
dmesg
Comment 4 Dennis Schridde 2015-07-12 09:22:16 UTC
Created attachment 117068 [details]
lspci -v
Comment 5 Dennis Schridde 2015-07-12 09:22:43 UTC
Created attachment 117069 [details]
Linux kernel config
Comment 6 Dennis Schridde 2015-07-12 09:23:51 UTC
Created attachment 117070 [details]
emerge --info media-libs/mesa
Comment 7 Dennis Schridde 2015-07-12 09:25:26 UTC
I am running Mesa 10.6.1 on Linux 4.1.2. Please note that I was also running radeontop 0.8 at the time I ran john and when X crashed.
Comment 8 Dennis Schridde 2015-07-12 09:27:38 UTC
Unlike with bug #91305 there was no switching to a VT and kernel output this time. X (SDDM+KDE, managed by systemd) immediately restarted.
Comment 9 Dennis Schridde 2015-07-12 09:32:09 UTC
After this, without rebooting, john will immediately segfault:
```
$ /usr/sbin/john --test=0 --verbosity=5
radeon: Failed to allocate virtual address for buffer:
radeon:    size      : 4352 bytes
radeon:    alignment : 4096 bytes
radeon:    domains   : 4
radeon:    va        : 0x0000000000800000
radeon: Failed to allocate virtual address for buffer:
radeon:    size      : 4352 bytes
radeon:    alignment : 4096 bytes
radeon:    domains   : 4
radeon:    va        : 0x0000000000800000
Segmentation fault
```


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.