2021-08-16-Ubuntu updates Mesa and OCL divers -- Still errors in package
16 Aug 2021 13:02:19 UTC
Topic 225853
(moderation:
Greetings. There still is an error in the Ubuntu Package Build.
When I have time, I am trying to notify the builder of the OCL package.
data from the usual sources...
Task id
LATeah4012L00_452.0_0_0.0_1329770_1
wu id
568704954
computer
12201025
Stderr output
<core_client_version>7.16.16</core_client_version>
<![CDATA[
<message>
process exited with code 69 (0x45, -187)</message>
<stderr_txt>
08:10:15 (3351): [normal]: This Einstein@home App was built at: Jan 16 2017 08:09:16
08:10:15 (3351): [normal]: Start of BOINC application '../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati'.
08:10:15 (3351): [debug]: 1e+16 fp, 4.5e+09 fp/s, 2327040 s, 646h24m00s06
command line: ../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati --inputfile ../../projects/einstein.phys.uwm.edu/LATeah4012L00.dat --alpha 0.943218186562 --delta 1.30995332125 --skyRadius 2.617990e-08 --ldiBins 30 --f0start 444.0 --f0Band 8.0 --firstSkyPoint 0 --numSkyPoints 1 --f1dot -1e-13 --f1dotBand 1e-13 --df1dot 1.713401242e-15 --ephemdir ../../projects/einstein.phys.uwm.edu/JPLEPH --Tcoh 2097152.0 --toplist 10 --cohFollow 10 --numCells 1 --useWeights 1 --Srefinement 1 --CohSkyRef 1 --cohfullskybox 1 --mmfu 0.1 --reftime 56100 --model 0 --f0orbit 0.005 --mismatch 0.1 --demodbinary 1 --BinaryPointFile ../../projects/einstein.phys.uwm.edu/templates_LATeah4012L00_0452_1329770.dat --debug 0 --device 0 -o LATeah4012L00_452.0_0_0.0_1329770_1_0.out
output files: 'LATeah4012L00_452.0_0_0.0_1329770_1_0.out' '../../projects/einstein.phys.uwm.edu/LATeah4012L00_452.0_0_0.0_1329770_1_0' 'LATeah4012L00_452.0_0_0.0_1329770_1_0.out.cohfu' '../../projects/einstein.phys.uwm.edu/LATeah4012L00_452.0_0_0.0_1329770_1_1'
08:10:15 (3351): [debug]: Flags: X64 SSE SSE2 GNUC X86 GNUX86
08:10:15 (3351): [debug]: glibc version/release: 2.33/release
08:10:15 (3351): [debug]: Set up communication with graphics process.
boinc_get_opencl_ids returned [0x13bb788 , 0x7f9c20741800]
Using OpenCL platform provided by: Mesa
Using OpenCL device "AMD VERDE (DRM 2.50.0, 5.11.0-25-generic, LLVM 12.0.0)" by: AMD
Max allocation limit: 1503238553
Global mem size: 2147483648
OpenCL compiling FAILED! : -11 . Error message: fatal error: cannot open file '/usr/local//usr/lib/clc/verde-amdgcn-mesa-mesa3d.bc': No such file or directory
OpenCL device has no FP64 support
OpenCL compiling FAILED! : -11 . Error message: fatal error: cannot open file '/usr/local//usr/lib/clc/verde-amdgcn-mesa-mesa3d.bc': No such file or directory
Couldn't compile OpenCL kernels (error: 1)
initialize_ocl returned error [2006]
OCL context null
OCL queue null
Error generating generic FFT context object [5]
08:10:15 (3351): [CRITICAL]: ERROR: MAIN() returned with error '5'
FPU status flags:
mv: cannot stat 'LATeah4012L00_452.0_0_0.0_1329770_1_0.out': No such file or directory
mv: cannot stat 'LATeah4012L00_452.0_0_0.0_1329770_1_0.out': No such file or directory
mv: cannot stat 'LATeah4012L00_452.0_0_0.0_1329770_1_0.out': No such file or directory
mv: cannot stat 'LATeah4012L00_452.0_0_0.0_1329770_1_0.out': No such file or directory
mv: cannot stat 'LATeah4012L00_452.0_0_0.0_1329770_1_0.out': No such file or directory
mv: cannot stat 'LATeah4012L00_452.0_0_0.0_1329770_1_0.out.cohfu': No such file or directory
mv: cannot stat 'LATeah4012L00_452.0_0_0.0_1329770_1_0.out.cohfu': No such file or directory
mv: cannot stat 'LATeah4012L00_452.0_0_0.0_1329770_1_0.out.cohfu': No such file or directory
mv: cannot stat 'LATeah4012L00_452.0_0_0.0_1329770_1_0.out.cohfu': No such file or directory
mv: cannot stat 'LATeah4012L00_452.0_0_0.0_1329770_1_0.out.cohfu': No such file or directory
mv: cannot stat 'LATeah4012L00_452.0_0_0.0_1329770_1_0.out.cohfu': No such file or directory
mv: cannot stat 'LATeah4012L00_452.0_0_0.0_1329770_1_0.out.cohfu': No such file or directory
08:10:27 (3351): [normal]: done. calling boinc_finish(69).
08:10:27 (3351): called boinc_finish
</stderr_txt>
======================================
Note the mangled path:
OpenCL compiling FAILED! : -11 . Error message: fatal error: cannot open file '/usr/local//usr/lib/clc/verde-amdgcn-mesa-mesa3d.bc': No such file or directory
======================================
My account:
https://einsteinathome.org/account/142557
my computer with older Radeon/ATI video card
https://einsteinathome.org/host/12201025
Domain name:
pc-14
Local standard time:
UTC -4 hours
Name:
pc-14
Created:
21 Feb 2016 0:05:10 UTC
Total credit:
57,786,253
Average credit:
6,455.46
Cross project credit:
CPU type:
AuthenticAMD AMD FX(tm)-8150 Eight-Core Processor [Family 21 Model 1 Stepping 2]
Number of processors:
8
Coprocessors:
AMD AMD VERDE (DRM 2.50.0, 5.11.0-25-generic, LLVM 12.0.0) (2048MB)
Operating system:
Linux Ubuntu Ubuntu 21.04 [5.11.0-25-generic|libc 2.33 (Ubuntu GLIBC 2.33-0ubuntu5)]
BOINC client version:
7.16.16
Memory:
11874.96 MiB
Cache:
2048 KiB
Swap space:
9538 MiB
Total disk space:
91.17 GiB
Free disk space:
82.93 GiB
Measured floating point speed:
4512.17 million ops/sec
Measured integer speed:
29395.13 million ops/sec
Average upload rate:
35.18 KiB/sec
Average download rate:
14478.91 KiB/sec
Average turnaround time:
0.9 days
Tasks:
14
OS
Ubuntu Release 21.04 (Hirsute Hippo) 64-bit
Kernel Linux 5.11.0-25-generic x86_64
OS sees
Graphics: AMD VERDE (DRM 2.50.0, 5.11.0-25-generic, LLVM 12.0.0)
BOINC sees:
Mon 16 Aug 2021 08:08:04 AM EDT | | Starting BOINC client version 7.16.16 for x86_64-pc-linux-gnu
Mon 16 Aug 2021 08:08:04 AM EDT | | log flags: file_xfer, sched_ops, task
Mon 16 Aug 2021 08:08:04 AM EDT | | Libraries: libcurl/7.74.0 OpenSSL/1.1.1j zlib/1.2.11 brotli/1.0.9 libidn2/2.3.0 libpsl/0.21.0 (+libidn2/2.3.0) libssh/0.9.5/openssl/zlib nghttp2/1.43.0 librtmp/2.3
Mon 16 Aug 2021 08:08:04 AM EDT | | Data directory: /var/lib/boinc-client
Mon 16 Aug 2021 08:08:05 AM EDT | | OpenCL: AMD/ATI GPU 0: AMD VERDE (DRM 2.50.0, 5.11.0-25-generic, LLVM 12.0.0) (driver version 21.0.3, device version OpenCL 1.1 Mesa 21.0.3, 2048MB, 2048MB available, 512 GFLOPS peak)
Mon 16 Aug 2021 08:08:05 AM EDT | | libc: Ubuntu GLIBC 2.33-0ubuntu5 version 2.33
Mon 16 Aug 2021 08:08:05 AM EDT | | Host name: pc-14
Mon 16 Aug 2021 08:08:05 AM EDT | | Processor: 8 AuthenticAMD AMD FX(tm)-8150 Eight-Core Processor [Family 21 Model 1 Stepping 2]
Mon 16 Aug 2021 08:08:05 AM EDT | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf pni pclmulqdq monitor ssse3 cx16 sse4_1 sse4_2 popcnt aes xsave avx lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs xop skinit wdt fma4 nodeid_msr topoext perfctr_core perfctr_nb cpb hw_pstate ssbd ibpb vmmcall arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold
Mon 16 Aug 2021 08:08:05 AM EDT | | OS: Linux Ubuntu: Ubuntu 21.04 [5.11.0-25-generic|libc 2.33 (Ubuntu GLIBC 2.33-0ubuntu5)]
Mon 16 Aug 2021 08:08:05 AM EDT | | Memory: 11.60 GB physical, 9.31 GB virtual
Mon 16 Aug 2021 08:08:05 AM EDT | | Disk: 91.17 GB total, 83.19 GB free
New Mesa (including OCL) drivers installed today:
from /var/log/apt/history.log)
Start-Date: 2021-08-16 07:52:41
Requested-By: jay (1000)
Upgrade:
mesa-opencl-icd:amd64 (21.0.3-0ubuntu0.2, 21.0.3-0ubuntu0.3),
libglx-mesa0:amd64 (21.0.3-0ubuntu0.2, 21.0.3-0ubuntu0.3),
libgbm1:amd64 (21.0.3-0ubuntu0.2, 21.0.3-0ubuntu0.3),
libxatracker2:amd64 (21.0.3-0ubuntu0.2, 21.0.3-0ubuntu0.3),
mesa-va-drivers:amd64 (21.0.3-0ubuntu0.2, 21.0.3-0ubuntu0.3),
libgl1-mesa-dri:amd64 (21.0.3-0ubuntu0.2, 21.0.3-0ubuntu0.3),
libegl1-mesa-dev:amd64 (21.0.3-0ubuntu0.2, 21.0.3-0ubuntu0.3),
mesa-vulkan-drivers:amd64 (21.0.3-0ubuntu0.2, 21.0.3-0ubuntu0.3),
libglapi-mesa:amd64 (21.0.3-0ubuntu0.2, 21.0.3-0ubuntu0.3),
libegl-mesa0:amd64 (21.0.3-0ubuntu0.2, 21.0.3-0ubuntu0.3),
mesa-vdpau-drivers:amd64 (21.0.3-0ubuntu0.2, 21.0.3-0ubuntu0.3)
End-Date: 2021-08-16 07:52:54
https://reposcope.com/package/mesa-opencl-icd
says:
| Original maintainer | Debian X Strike Force |
|---|---|
| Homepage | https://mesa3d.org/ |
Greetings! There is a
)
Greetings!
There is a workaround..
As root, make some directories that match what the mangled path is looking for; and; make a link to the real file...
find /usr -print | grep verde-amdgcn-mesa-mesa3d
/usr/lib/clc/verde-amdgcn-mesa-mesa3d.bc
- this is the real file -- and this is the location on my machine...
The errored/mangled path is looking for /usr/local//usr/lib/clc/verde-amdgcn-mesa-mesa3d.bc
so
cd to /usr/local/ and as root, mkdir usr
cd to /usr/local/usr and , as root mkdir lib
cd to /usr/local/usr/lib and , as root mkdir clc
(almost there)
cd to /usr/local/usr/lib/clc
and , as root, ln -s /usr/lib/clc/verde-amdgcn-mesa-mesa3d.bc verde-amdgcn-mesa-mesa3d.bc
Your Installation may have a different paths.
So
the first argument of the 'ln-s' command is the real address.
Use the the address where the verde-amdgcn-mesa-mesa3d.bc file exists.
It is a messy kludge. It works for me on Ubuntu Mate.
Have a safe day.
[soapbox]
If you are not vaccinated yet, do it. Protect the spread to others who can not get vaccinated.
[/soapbox]
jay wrote:There is a
)
Yes, creating an appropriate symlink will work around the buggy package that creates the mangled path. You really need to alert the package maintainer to get it fixed properly, though.
There haven't been other reports I've seen and that's probably because Ubuntu users don't seem to use mesa OpenCL. AMD puts out special packages for both Ubuntu and Red Hat Linux versions under the name AMDGPU-PRO (or Radeon Software for Linux) which contain various components (including OpenCL) - some open source and some proprietary - with proper installation scripts for those two distros. With suitable options, it's possible to install just the OpenCL bits instead of mesa OpenCL. This seems to be what the majority of Ubuntu users choose to do - hence no problems for them and no reports here of packaging errors.
I noticed in your previous message where you posted some startup messages that BOINC identified your GPU as:-
The OpenCL version of 1.1 is a bit of a worry. I happen to have a very similar GPU in my junk box and I fired it up recently and installed the OpenCL bits from the Red Hat version of amdgpu-pro 20.40. By running inxi -Gxx, the device identifies as:-
Advanced Micro Devices [AMD/ATI] Cape Verde PRO [Radeon HD 7750/8740 / R7 250E]Boinc identifies this GPU in the following way:-
OpenCL: AMD/ATI GPU 0: AMD Radeon HD 7700 Series (driver version 3180.7, device version OpenCL 1.2 AMD-APP (3180.7), 3882MB,Notice the OpenCL version is 1.2. From what I've read, Cape Verde devices (which are GCN 1st gen hardware) do support OpenCL 1.2. My understanding is that 1.1 is very old and that 1.2 represented a significant upgrade. This old GPU of mine is crunching tasks nearly twice as fast as yours - around 3900 secs instead of 7000 secs in your case. You might get a significant speed improvement by upgrading to OpenCL 1.2 from amdgpu-pro.
The Einstein forums are not the place for this type of rhetoric so I'd ask you kindly not to continue posting something that is actually misinformation.
I'm fully vaccinated myself and do consider it extremely unwise for everybody able to, not to make the same choice. The misinformation is to claim you are protecting others who can't get vaccinated.
You are just protecting yourself from the worst ravages of catching the virus. That's it. Period. You can still catch the virus and have very mild symptoms or even be asymptomatic. You could therefore quite unknowingly spread it to others. If you wrongly imagined that you couldn't, you may very well take less precautions and potentially spread it more easily that way to the very people you think you are protecting.
Cheers,
Gary.