Thread 'which PPA for Ubuntu Mesa openCL for Raden 7750???'

Message boards : GPUs : which PPA for Ubuntu Mesa openCL for Raden 7750???
Message board moderation

To post messages, you must log in.

AuthorMessage
Profilejay_e

Send message
Joined: 8 Mar 07
Posts: 115
United States
Message 78145 - Posted: 22 May 2017, 21:18:07 UTC

I have a question...
Is there a post (somewhere) that addresses this: ??
Someone know an answer??

I have Radeon/ATI 7750 GPU running Ubuntu Mate 16.10 crunching E@H happily with Mesa OpenCL with a PPA.
BOINC says :(driver version 17.2.0-devel, device version OpenCL 1.1 Mesa 17.2.0-devel

I have another machine (pc-10) that used to run fglrx. I want to upgrade to 16.04 or later.
Can't find a working driver with PPA for Mesa opencl for any of
Ubuntu 17.04 (Zesty Zapus)
Ubuntu 16.10 (Yakkety Yak)
Ubuntu 16.04.2 LTS (Xenial Xerus)

(without PPA does not work either.)
BOINC can find GPU but WU errors out in 3 to 4 seconds.
E*H results file says error is:
Invalid bitcast
  %21 = bitcast %union.anon addrspace(4)* %6 to i8*
Invalid bitcast
  %22 = bitcast %union.anon addrspace(4)* %4 to i8*
Invalid bitcast
  %23 = bitcast %union.anon addrspace(4)* %2 to i8*
LLVM ERROR: Broken function found, compilation aborted!


other errors have to do with atomic insruction.

for a while, a wide range of errors are accessible at: (due to trying different drivers)
The errors on pc-10 (https://einsteinathome.org/host/12533566)
https://einsteinathome.org/task/646569916
https://einsteinathome.org/task/646873535
https://einsteinathome.org/task/647476568

Been trying for 4 days. Now time to ask.
Please help.

T H A N K Y O U,
Jay

PS
PPA that I used now supported for LTS and 17.04
I cant get any to work
BOINC sees card; but WU fails.
AMDGPU-Pro won't work for 7750

PPS
reference:
https://www.gamingonlinux.com/articles/ubuntu-now-has-a-more-official-ppa-to-get-mesa-updates.9393

PPPS
https://launchpad.net/~ubuntu-x-swat/+archive/ubuntu/updates
https://launchpad.net/~paulo-miguel-dias/+archive/ubuntu/mesa
https://launchpad.net/~paulo-miguel-dias/+archive/ubuntu/pkppa/
https://launchpad.net/~oibaf/+archive/ubuntu/graphics-drivers/
ID: 78145 · Report as offensive
ProfileJord
Volunteer tester
Help desk expert
Avatar

Send message
Joined: 29 Aug 05
Posts: 15566
Netherlands
Message 78146 - Posted: 22 May 2017, 22:49:10 UTC - in response to Message 78145.  

Have you asked at Einstein if their application is Mesa OpenCL compatible? Because if it isn't... one OpenCL isn't the equal to another.
ID: 78146 · Report as offensive
Profilejay_e

Send message
Joined: 8 Mar 07
Posts: 115
United States
Message 78149 - Posted: 23 May 2017, 2:08:57 UTC - in response to Message 78146.  
Last modified: 23 May 2017, 2:10:49 UTC

Hi Jord,
I didn't ask because I have the same board in another machine happily crunching on E@H GPU WU with
a version of OpenCL. It has been running in Ubuntu (mate) 16.10 (Yakkety YaK)

I can't get that PPA anymore since those are now developed for the LTS and current release.
I have tried the blunt force method - 5 PPA on three releases.
All errored.

I wanted to test on some SETI GPU WU but that well is apparently dry.

I hope I'm missing something.

Right now, I'm missing a beer.

How have you been? Good news you want to share?
I think the US politics is making a market in zoloft. :-)

Send a case your way?

Jay
ID: 78149 · Report as offensive
ProfileJord
Volunteer tester
Help desk expert
Avatar

Send message
Joined: 29 Aug 05
Posts: 15566
Netherlands
Message 78171 - Posted: 23 May 2017, 17:57:40 UTC - in response to Message 78149.  

Seti is down at the moment for maintenance, which is why you won't get work from there. And trouble is, the information about Mesa OpenCL not being (always) compatible with project apps came from a thread there, but since they're in maintenance, I can't link to it. But at least Seti applications won't work with Mesa either, because of it being different from Khronos OpenCL with which the apps are built.
ID: 78171 · Report as offensive
Profilejay_e

Send message
Joined: 8 Mar 07
Posts: 115
United States
Message 78179 - Posted: 24 May 2017, 4:48:36 UTC - in response to Message 78171.  

Jord,
Thank you! I am amazed.

IT may be a waste of time, but I tried something else (as well.)

I did a ppa-purge on the mesa opencl,
and re-reviewed the AMD documentation.
Either it changed or I mis read-it, or I read a different thread.
Anyway, I got the AMD script matched for 7700 and Ubuntu 16.04 and it installed the .deb files it made.
I rebooted (Yay - no black screen) and resrtarted BOINC.

BOINC recognized the GPU with OpenCL.
But the Einstein WU failed.

I'll pursue this with the Einstein Forum - since building the AMD code for the release
seems (to me) to be the least weird of the builds and the most predictable in how it could go wrong.
(Trying not to laugh.)

Here is what the WU task log said of the failure....

 https://einsteinathome.org/host/12533566
to 
https://einsteinathome.org/host/12533566/tasks
to
https://einsteinathome.org/task/648650142
which has:
00:05:03 (2383): [normal]: Start of BOINC application '../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati'.
00:05:03 (2383): [debug]: 1e+16 fp, 2.3e+09 fp/s, 4619059 s, 1283h04m19s27
command line: ../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati --inputfile ../../projects/einstein.phys.uwm.edu/LATeah0029L.dat --alpha 4.42281478648 --delta -0.0345027837249 --skyRadius 2.152570e-06 --ldiBins 15 --f0start 1076.0 --f0Band 8.0 --firstSkyPoint 0 --numSkyPoints 1 --f1dot -1e-13 --f1dotBand 1e-13 --df1dot 3.344368011e-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_LATeah0029L_1084_10804295.dat --debug 1 --device 0 -o LATeah0029L_1084.0_0_0.0_10804295_2_0.out
output files: 'LATeah0029L_1084.0_0_0.0_10804295_2_0.out' '../../projects/einstein.phys.uwm.edu/LATeah0029L_1084.0_0_0.0_10804295_2_0' 'LATeah0029L_1084.0_0_0.0_10804295_2_0.out.cohfu' '../../projects/einstein.phys.uwm.edu/LATeah0029L_1084.0_0_0.0_10804295_2_1'
00:05:03 (2383): [debug]: Flags: X64 SSE SSE2 GNUC X86 GNUX86
00:05:03 (2383): [debug]: glibc version/release: 2.23/stable
00:05:03 (2383): [debug]: Set up communication with graphics process.
boinc_get_opencl_ids returned [0x2fb43f0 , 0x7fd7e8a12e98] 
Using OpenCL platform provided by: Advanced Micro Devices, Inc.
Using OpenCL device "Capeverde" by: Advanced Micro Devices, Inc.
Max allocation limit: 1350254592

Global mem size: 2040578048
Couldn't create OpenCL command queue (error: -6)!  

OpenCL shutdown complete!
initialize_ocl returned error [2013]
OCL context null
OCL queue null
Error generating generic FFT context object [5]
00:05:03 (2383): [CRITICAL]: ERROR: MAIN() returned with error '5'

FPU status flags: 
mv: cannot stat 'LATeah0029L_1084.0_0_0.0_10804295_2_0.out': No such file or directory
mv: cannot stat 'LATeah0029L_1084.0_0_0.0_10804295_2_0.out': No such file or directory
mv: cannot stat 'LATeah0029L_1084.0_0_0.0_10804295_2_0.out': No such file or directory
mv: cannot stat 'LATeah0029L_1084.0_0_0.0_10804295_2_0.out': No such file or directory
mv: cannot stat 'LATeah0029L_1084.0_0_0.0_10804295_2_0.out': No such file or directory
mv: cannot stat 'LATeah0029L_1084.0_0_0.0_10804295_2_0.out.cohfu': No such file or directory
mv: cannot stat 'LATeah0029L_1084.0_0_0.0_10804295_2_0.out.cohfu': No such file or directory
mv: cannot stat 'LATeah0029L_1084.0_0_0.0_10804295_2_0.out.cohfu': No such file or directory
mv: cannot stat 'LATeah0029L_1084.0_0_0.0_10804295_2_0.out.cohfu': No such file or directory
mv: cannot stat 'LATeah0029L_1084.0_0_0.0_10804295_2_0.out.cohfu': No such file or directory
mv: cannot stat 'LATeah0029L_1084.0_0_0.0_10804295_2_0.out.cohfu': No such file or directory
mv: cannot stat 'LATeah0029L_1084.0_0_0.0_10804295_2_0.out.cohfu': No such file or directory
00:05:15 (2383): [normal]: done. calling boinc_finish(69).
00:05:15 (2383): called boinc_finish



Thanks again!!
Jay

PS.
Current AMD Links
http://support.amd.com/en-us/download/linux
(down to natching Ubuntu 16.04 to Radeon 7700)
to-->>
http://support.amd.com/en-us/kb-articles/Pages/AMDGPU-PRO-Driver-for-Linux-Release-Notes.aspx

AMDGPU-PRO Driver for Linux Release Notes
SEE section on Compatibility!
Article Number: RN-AMDGPU-PRO-LINUX
​​​​​This article provides information on the latest version of the AMDGPU-Pro Driver for Linux®.
(scrolled down to AMDGPU-Pro Driver Version 17.10 for Ubuntu 16.04.2) to get blob.
ID: 78179 · Report as offensive
ProfileJord
Volunteer tester
Help desk expert
Avatar

Send message
Joined: 29 Aug 05
Posts: 15566
Netherlands
Message 78185 - Posted: 24 May 2017, 9:29:27 UTC - in response to Message 78179.  

http://support.amd.com/en-us/kb-articles/Pages/AMDGPU-PRO-Driver-for-Linux-Release-Notes.aspx
I wonder....
The AMDGPU-PRO driver is OpenCL 1.2 compliant, the GPU you have does OpenCL 1.2, but Mesa is OpenCL 1.1 (incomplete) or 1.0 at best. It's probably missing some instruction(s) these days that the are there in OpenCL 1.2, but aren't in the earlier versions.

But this goes far over my head.
ID: 78185 · Report as offensive
ProfileAgentb
Avatar

Send message
Joined: 30 May 15
Posts: 265
United Kingdom
Message 78239 - Posted: 26 May 2017, 0:03:52 UTC - in response to Message 78179.  


I did a ppa-purge on the mesa opencl,
and re-reviewed the AMD documentation.

Both good ideas, running two drivers on the one device often ends badly.

BOINC recognized the GPU with OpenCL.
But the Einstein WU failed.

I'll pursue this with the Einstein Forum - since building the AMD code for the release
seems (to me) to be the least weird of the builds and the most predictable in how it could go wrong.
(Trying not to laugh.)

I'd say this probably the only way it will go right unless you are handy at c++ and feel happy to help with debugging and writing the missing 1.2 bits at mesa.


Here is what the WU task log said of the failure....

<snip up to the first error>

Using OpenCL platform provided by: Advanced Micro Devices, Inc.
Using OpenCL device "Capeverde" by: Advanced Micro Devices, Inc.
Max allocation limit: 1350254592

Global mem size: 2040578048
Couldn't create OpenCL command queue (error: -6)!

OpenCL shutdown complete!

Google for OpenCL error codes

-6 CL_OUT_OF_HOST_MEMORY if there is a failure to allocate resources required by the OpenCL implementation on the host.


The apps at E@H take at least .9GBB VRAM - clinfo may be revealing and it is reported there are some environment variables that can be set to eek a little extra, but note from the AMDGPUPRO driver page the risk of pushing OpenCL past the maximum.


Known Issues
System hangs when OpenCL™ attempts to allocate more than available system memory.


I see some valid tasks at E@H so seems good.
ID: 78239 · Report as offensive
Profilejay_e

Send message
Joined: 8 Mar 07
Posts: 115
United States
Message 78341 - Posted: 28 May 2017, 23:13:44 UTC - in response to Message 78239.  

Hi!
Having a good weekend?

... in recap of amdGPU pro on Ubuntu Release 16.04.2 LTS (Xenial Xerus) 64-bit with Radeon 7750...
Installed are:
aptitude search opencl | grep ^i
i A libopencl1-amdgpu-pro - AMD OpenCL ICD Loader library
i A libopencl1-amdgpu-pro:i386 - AMD OpenCL ICD Loader library
i A ocl-icd-libopencl1 - Generic OpenCL ICD Loader
i ocl-icd-opencl-dev - OpenCL development files
i A opencl-amdgpu-pro-icd - non-free AMD OpenCL ICD Loaders
i A opencl-amdgpu-pro-icd:i386 - non-free AMD OpenCL ICD Loaders
i A opencl-headers - OpenCL (Open Computing Language) header fi


cliinfo shows OpenCl1.2 and 2 Gig of memory on the GPU
  Device Version                                  OpenCL 1.2 AMD-APP (2348.3)
...
    Support is emulated in software               No
    Correctly-rounded divide and sqrt operations  No
  Address bits                                    32, Little-Endian
  Global memory size                              2021892096 (1.883GiB)
  Global free memory (AMD)                        <printDeviceInfo:68: get number of CL_DEVICE_GLOBAL_FREE_MEMORY_AMD : error -33>
  Global memory channels (AMD)                    4
  Global memory banks per channel (AMD)           8
  Global memory bank width (AMD)                  256 bytes
  Error Correction support                        No
  Max memory allocation                           1335263232 (1.244GiB)
  Unified memory for Host and Device              No
  

I tried to google the -33 error but NO luck for me.

( Am I correct in assuming that error message is about system memory on the CPU ?
Or is it talking abount memory on the GPU???)

I will check BOINC config and allow 95% of memory - ansd swap memory
and run vmstat when starting BOINC and starting an Einstein GPU.
(will see if SETI GPU WU available too.)


@agentB
I have an earlier installation of Ubunto 16.10 the DOES work with Mesa Opencl.
Recently ried 16.04. 16.10, and 17.05 with Ubuntu's Mesa's OpenCL packages.

I guess I should swap cards too. Rule out a hardware problem.

here is what the error report on einstein shows:
1) Dashboard showd two computers (both have 7750s
https://einsteinathome.org/account/dashboard

2) pc-10 is Computer 12533566
https://einsteinathome.org/host/12533566
3) It's tasks
https://einsteinathome.org/host/12533566/tasks

4) This particular result ran on the AMD-GPU Pro driver
https://einsteinathome.org/task/648987795
which is/was:
<stderr_txt>
01:21:22 (4217): [normal]: This Einstein@home App was built at: Jan 16 2017 08:09:16

01:21:22 (4217): [normal]: Start of BOINC application '../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati'.
01:21:22 (4217): [debug]: 1e+16 fp, 2.3e+09 fp/s, 4622774 s, 1284h06m13s66
command line: ../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati --inputfile ../../projects/einstein.phys.uwm.edu/LATeah0029L.dat --alpha 4.42281478648 --delta -0.0345027837249 --skyRadius 2.152570e-06 --ldiBins 15 --f0start 1172.0 --f0Band 8.0 --firstSkyPoint 0 --numSkyPoints 1 --f1dot -1e-13 --f1dotBand 1e-13 --df1dot 3.344368011e-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_LATeah0029L_1180_9959680.dat --debug 1 --device 0 -o LATeah0029L_1180.0_0_0.0_9959680_0_0.out
output files: 'LATeah0029L_1180.0_0_0.0_9959680_0_0.out' '../../projects/einstein.phys.uwm.edu/LATeah0029L_1180.0_0_0.0_9959680_0_0' 'LATeah0029L_1180.0_0_0.0_9959680_0_0.out.cohfu' '../../projects/einstein.phys.uwm.edu/LATeah0029L_1180.0_0_0.0_9959680_0_1'
01:21:22 (4217): [debug]: Flags: X64 SSE SSE2 GNUC X86 GNUX86
01:21:22 (4217): [debug]: glibc version/release: 2.23/stable
01:21:22 (4217): [debug]: Set up communication with graphics process.
boinc_get_opencl_ids returned [0x1fb53f0 , 0x7f1246cdae98]
Using OpenCL platform provided by: Advanced Micro Devices, Inc.
Using OpenCL device "Capeverde" by: Advanced Micro Devices, Inc.
Max allocation limit: 1349726208
Global mem size: 2040963072
Couldn't create OpenCL command queue (error: -6)!
OpenCL shutdown complete!
initialize_ocl returned error [2013]
OCL context null
OCL queue null
Error generating generic FFT context object [5]
01:21:23 (4217): [CRITICAL]: ERROR: MAIN() returned with error '5'
FPU status flags:
mv: cannot stat 'LATeah0029L_1180.0_0_0.0_9959680_0_0.out': No such file or directory

===================================================================
Trying again with more CPU mem avail configuration

was

Page/swap file: use at most: 70 % of total
Limit the swap space (page file) used by BOINC.

Memory: when computer is in use, use at most: 80 % of total
Limit the memory used by BOINC when you're using the computer.

Memory: when computer is not in use, use at most: 80 % of total
Limit the memory used by BOINC when you're not using the computer.

change to
Page/swap file: use at most: 95 % of total
Limit the swap space (page file) used by BOINC.

Memory: when computer is in use, use at most: 95 % of total
Limit the memory used by BOINC when you're using the computer.

Memory: when computer is not in use, use at most: 95 % of total
Limit the memory used by BOINC when you're not using the computer.

vnstat data
boinc off
jay@pc-10:~$ vmstat 2 200
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu-----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
 0  0      0 4016920 138624 1179664    0    0    69    39  295  516  3  1 95  1  0
 2  0      0 4016928 138624 1179620    0    0     0     0 1194 2162  3  1 96  0  0
 0  0      0 4016160 138624 1179628    0    0     0     0  973 1502  2  0 98  0  0


boinc started but no einstein wu yet
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu-----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
 0  0      0 3918636 138852 1226820    0    0     0     0 1726 2862  3  1 96  0  0
 0  0      0 3918660 138852 1226808    0    0     0     0 1254 2082  2  1 97  0  0
 0  0      0 3918760 138852 1226808    0    0     0     0  403  732  1  0 99  0  0

Einstein WU downloading, running, aborting
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu-----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
 1  0      0 3882588 138936 1258328    0    0    68    37  293  511  3  1 95  1  0
 0  0      0 3881804 138944 1259204    0    0     0    46 2500 4232  6  1 93  1  0
 0  0      0 3881832 138944 1259920    0    0     0     0 2038 2577  1  1 98  0  0
 2  0      0 3863300 138944 1260444    0    0     0    58 2584 4141 12  2 86  0  0
 0  0      0 3862236 138944 1260508    0    0     0     0 1215 2141  3  1 97  0  0
 0  0      0 3862052 138944 1260464    0    0     0     0 1178 2084  3  1 97  0  0
 0  0      0 3861796 138944 1260464    0    0     0     0  485  924  1  0 99  0  0
 0  0      0 3861680 138952 1260464    0    0     0    28  502  867  1  0 98  0  0
 0  0      0 3861432 138952 1260500    0    0     0     0  503  864  0  1 99  0  0
 0  0      0 3860952 138952 1260684    0    0     0  1682  521  862  1  0 99  0  0
 0  0      0 3880792 138952 1260620    0    0     0    60  435  790  1  0 99  0  0
 0  0      0 3880892 138952 1260520    0    0     0     0  409  780  1  0 99  0  0
 0  0      0 3880892 138952 1260524    0    0     0     0  367  670  0  0 99  0  0
 0  0      0 3880744 138952 1260528    0    0     0     0  962 1567  2  0 98  0  0
 0  0      0 3880700 138952 1260544    0    0     0     0  617 1054  1  0 98  0  0
 0  0      0 3880700 138952 1260548    0    0     0     0  389  717  0  0 100  0  0
 0  0      0 3880700 138952 1260548    0    0     0     0  429  810  1  0 99  0  0


Enstein shows task reult:
boinc_get_opencl_ids returned [0x277f3f0 , 0x7efd3a92ae98]
Using OpenCL platform provided by: Advanced Micro Devices, Inc.
Using OpenCL device "Capeverde" by: Advanced Micro Devices, Inc.
Max allocation limit: 1349271552
Global mem size: 2039652352
Couldn't create OpenCL command queue (error: -6)!
OpenCL shutdown complete!
initialize_ocl returned error [2013]
OCL context null
OCL queue null
Error generating generic FFT context object [5]
18:58:12 (3376): [CRITICAL]: ERROR: MAIN() returned with error '5'
---
I guess the difference in Global mem size is the change od data that the display driver is using
---
will post this. shut down; swap cards; and try again.
ID: 78341 · Report as offensive
Profilejay_e

Send message
Joined: 8 Mar 07
Posts: 115
United States
Message 78343 - Posted: 29 May 2017, 5:17:37 UTC - in response to Message 78341.  

In pc-10, the Machine running Ubuntu 16.04 with the AMDGPU-Pro Diver, I swapped the GPU card with the one that can successfully crunch Einstein GPUs. (Test the small chance there is a hardware error.)

The WU fails in 5 seconds with the same error.

---------------------------
Also tried a SETI GPU WU
The BOINC event Log says:
Mon 29 May 2017 12:43:31 AM EDT | SETI@home | Starting task 20se08aa.12169.12856.15.42.175_1
Mon 29 May 2017 12:43:32 AM EDT | SETI@home | work fetch suspended by user
Mon 29 May 2017 12:43:32 AM EDT | SETI@home | task postponed 30.000000 sec:
Mon 29 May 2017 12:44:34 AM EDT | SETI@home | task postponed 30.000000 sec:

<stderr_txt>
OpenCL platform detected: Advanced Micro Devices, Inc.
Number of OpenCL devices found : 1
BOINC assigns slot on device #0.
Info: BOINC provided OpenCL device ID used
ERROR: OpenCL kernel/call 'Creating Command Queue. (clCreateCommandQueue)' call failed (-6) in file ../../src/GPU_lock.cpp near line 1359.
Waiting 30 sec before restart...
OpenCL platform detected: Advanced Micro Devices, Inc.
Number of OpenCL devices found : 1
BOINC assigns slot on device #0.
Info: BOINC provided OpenCL device ID used
(Repeating message until WU aborted.)

Plan:
Try mesa - OpenCL in 16.04
wipe disk. Start over with 16.10. Post results
ID: 78343 · Report as offensive
Profilejay_e

Send message
Joined: 8 Mar 07
Posts: 115
United States
Message 78344 - Posted: 29 May 2017, 6:20:18 UTC - in response to Message 78343.  

Next step
Try MESA OopenCl - in ubuntu 16.04

---
removed AMDGPU-Pro
install the Mesa OpenCL
- No PPA
------

BOINC Sees:
 
Mon 29 May 2017 01:44:30 AM EDT |  | Starting BOINC client version 7.6.31 for x86_64-pc-linux-gnu
Mon 29 May 2017 01:44:30 AM EDT |  | log flags: file_xfer, sched_ops, task
Mon 29 May 2017 01:44:30 AM EDT |  | Libraries: libcurl/7.47.0 OpenSSL/1.0.2g zlib/1.2.8 libidn/1.32 librtmp/2.3
Mon 29 May 2017 01:44:30 AM EDT |  | Data directory: /var/lib/boinc-client
Mon 29 May 2017 01:44:30 AM EDT |  | OpenCL: AMD/ATI GPU 0: AMD CAPE VERDE (DRM 2.43.0 / 4.4.0-78-generic, LLVM 3.8.0) (driver version 12.0.6, device version OpenCL 1.1 Mesa 12.0.6, 1024MB, 1024MB available, 512 GFLOPS peak)
Mon 29 May 2017 01:44:30 AM EDT |  | Host name: pc-10
Mon 29 May 2017 01:44:30 AM EDT |  | Processor: 4 AuthenticAMD AMD Phenom(tm) 9550 Quad-Core Processor [Family 16 Model 2 Stepping 3]
Mon 29 May 2017 01:44:30 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 3dnowext 3dnow constant_tsc rep_good nopl nonstop_tsc extd_apicid pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs hw_pstate vmmcall npt lbrv svm_lock
Mon 29 May 2017 01:44:30 AM EDT |  | OS: Linux: 4.4.0-78-generic
Mon 29 May 2017 01:44:30 AM EDT |  | Memory: 5.83 GB physical, 9.77 GB virtual
Mon 29 May 2017 01:44:30 AM EDT |  | Disk: 57.67 GB total, 48.91 GB free
Mon 29 May 2017 01:44:30 AM EDT |  | Local time is UTC -4 hours



Get the error:
LLVM ERROR: Cannot select: 0x3f04af0: i32,ch = AtomicCmpSwap<Volatile

as in: https://einsteinathome.org/task/650217120
<stderr_txt>
01:49:24 (2361): [normal]: This Einstein@home App was built at: Jan 16 2017 08:09:16

01:49:24 (2361): [normal]: Start of BOINC application '../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati'.
01:49:24 (2361): [debug]: 1e+16 fp, 2.3e+09 fp/s, 4622774 s, 1284h06m13s66
command line: ../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati --inputfile ../../projects/einstein.phys.uwm.edu/LATeah0030L.dat --alpha 4.42281478648 --delta -0.0345027837249 --skyRadius 2.152570e-06 --ldiBins 15 --f0start 804.0 --f0Band 8.0 --firstSkyPoint 0 --numSkyPoints 1 --f1dot -1e-13 --f1dotBand 1e-13 --df1dot 3.344368011e-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_LATeah0030L_0812_1039140.dat --debug 1 --device 0 -o LATeah0030L_812.0_0_0.0_1039140_0_0.out
output files: 'LATeah0030L_812.0_0_0.0_1039140_0_0.out' '../../projects/einstein.phys.uwm.edu/LATeah0030L_812.0_0_0.0_1039140_0_0' 'LATeah0030L_812.0_0_0.0_1039140_0_0.out.cohfu' '../../projects/einstein.phys.uwm.edu/LATeah0030L_812.0_0_0.0_1039140_0_1'
01:49:24 (2361): [debug]: Flags: X64 SSE SSE2 GNUC X86 GNUX86
01:49:24 (2361): [debug]: glibc version/release: 2.23/stable
01:49:24 (2361): [debug]: Set up communication with graphics process.
boinc_get_opencl_ids returned [0x2791cf8 , 0x7f0676a40ae0] 
Using OpenCL platform provided by: Mesa
Using OpenCL device "AMD CAPE VERDE (DRM 2.43.0 / 4.4.0-78-generic, LLVM 3.8.0)" by: AMD
Max allocation limit: 268435456
Global mem size: 1073741824
OpenCL device has FP64 support
LLVM ERROR: Cannot select: 0x3f04af0: i32,ch = AtomicCmpSwap<Volatile LDST4[%1592(addrspace=1)]> 0x4c03fa0, 0x4be1ad0, 0x4c4e4b0, 0x37cf510
  0x4be1ad0: i64,ch = CopyFromReg 0x4c03fa0, Register:i64 %vreg241
    0x37d0c40: i64 = Register %vreg241
  0x4c4e4b0: i32,ch = CopyFromReg 0x4c03fa0, Register:i32 %vreg244
    0x3f01c90: i32 = Register %vreg244
  0x37cf510: i32 = bitcast 0x2946c20
    0x2946c20: f32 = fadd 0x4cd8bd0, 0x4b4bc00
      0x4cd8bd0: f32,ch = CopyFromReg 0x4c03fa0, Register:f32 %vreg236
        0x37d84a0: f32 = Register %vreg236
      0x4b4bc00: f32 = bitcast 0x4c4e4b0
        0x4c4e4b0: i32,ch = CopyFromReg 0x4c03fa0, Register:i32 %vreg244
          0x3f01c90: i32 = Register %vreg244
In function: kernel_ts_2_phase_diff_sorted

</stderr_txt>


==============
NEXT: install PPA for Mesa
see https://laanwj.github.io/2016/05/06/opencl-ubuntu1604.html
and
https://www.gamingonlinux.com/articles/ubuntu-now-has-a-more-official-ppa-to-get-mesa-updates.9393
says

The first says:
$ sudo add-apt-repository ppa:paulo-miguel-dias/mesa
$ sudo apt-get update
$ sudo apt-get install libclc-amdgcn mesa-opencl-icd
Note: this is a development PPA - NOT the Stable PPA

The second says:
sudo add-apt-repository ppa:ubuntu-x-swat/updates


will try x-swat this time.

does
sudo aptitude dist-upgrade
The following NEW packages will be installed:
  libclang-common-4.0-dev{a} libllvm4.0{a} 
The following packages will be REMOVED:
  libclang-common-3.7-dev{u} libllvm3.7{u} 
The following packages will be upgraded:
  libclc-amdgcn libclc-dev libclc-r600 libdrm-amdgpu1 libdrm-dev libdrm-intel1 libdrm-nouveau2 libdrm-radeon1 libdrm2 libegl1-mesa libegl1-mesa-drivers 
  libgbm1 libgl1-mesa-dev libgl1-mesa-dri libgl1-mesa-glx libglapi-mesa libgles1-mesa{b} libgles2-mesa libwayland-client0 libwayland-cursor0 
  libwayland-egl1-mesa libwayland-server0 libxatracker2 mesa-common-dev mesa-opencl-icd mesa-va-drivers 
26 packages upgraded, 2 newly installed, 2 to remove and 0 not upgraded.
Need to get 66.6 MB of archives. After unpacking 166 MB will be used.
The following packages have unmet dependencies:
 libgles1-mesa : Depends: libglapi-mesa (= 13.0.4-1ubuntu1~16.04~0.1) but 17.1.0-1ubuntu1~16.04~1 is to be installed.
The following actions will resolve these dependencies:

     Remove the following packages:
1)     libgles1-mesa         

..... and .....

The following NEW packages will be installed:
  libclang-common-4.0-dev{a} libllvm4.0{a} 
The following packages will be REMOVED:
  libclang-common-3.7-dev{u} libgles1-mesa{a} libllvm3.7{u} 
The following packages will be upgraded:
  libclc-amdgcn libclc-dev libclc-r600 libdrm-amdgpu1 libdrm-dev libdrm-intel1 libdrm-nouveau2 libdrm-radeon1 libdrm2 libegl1-mesa libegl1-mesa-drivers 
  libgbm1 libgl1-mesa-dev libgl1-mesa-dri libgl1-mesa-glx libglapi-mesa libgles2-mesa libwayland-client0 libwayland-cursor0 libwayland-egl1-mesa 
  libwayland-server0 libxatracker2 mesa-common-dev mesa-opencl-icd mesa-va-drivers 
25 packages upgraded, 2 newly installed, 3 to remove and 0 not upgraded.
Need to get 66.6 MB of archives. After unpacking 166 MB will be used.


Plan: Install; reboot; test
ID: 78344 · Report as offensive
Profilejay_e

Send message
Joined: 8 Mar 07
Posts: 115
United States
Message 78347 - Posted: 29 May 2017, 8:27:59 UTC - in response to Message 78344.  

test of x-swat PPA on Ubuntu 16.04 with Radeon 7750 with Mesa open CL

Boinc Sees:
Mon 29 May 2017 02:43:38 AM EDT |  | Starting BOINC client version 7.6.31 for x86_64-pc-linux-gnu
Mon 29 May 2017 02:43:38 AM EDT |  | log flags: file_xfer, sched_ops, task
Mon 29 May 2017 02:43:38 AM EDT |  | Libraries: libcurl/7.47.0 OpenSSL/1.0.2g zlib/1.2.8 libidn/1.32 librtmp/2.3
Mon 29 May 2017 02:43:38 AM EDT |  | Data directory: /var/lib/boinc-client
Mon 29 May 2017 02:43:38 AM EDT |  | OpenCL: AMD/ATI GPU 0: AMD CAPE VERDE (DRM 2.43.0 / 4.4.0-78-generic, LLVM 4.0.0) (driver version 17.1.0, device version OpenCL 1.1 Mesa 17.1.0, 2048MB, 2048MB available, 512 GFLOPS peak)
Mon 29 May 2017 02:43:38 AM EDT |  | Host name: pc-10
Mon 29 May 2017 02:43:38 AM EDT |  | Processor: 4 AuthenticAMD AMD Phenom(tm) 9550 Quad-Core Processor [Family 16 Model 2 Stepping 3]
Mon 29 May 2017 02:43:38 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 3dnowext 3dnow constant_tsc rep_good nopl nonstop_tsc extd_apicid pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs hw_pstate vmmcall npt lbrv svm_lock
Mon 29 May 2017 02:43:38 AM EDT |  | OS: Linux: 4.4.0-78-generic
Mon 29 May 2017 02:43:38 AM EDT |  | Memory: 5.83 GB physical, 9.77 GB virtual
Mon 29 May 2017 02:43:38 AM EDT |  | Disk: 57.67 GB total, 48.70 GB free
Mon 29 May 2017 02:43:38 AM EDT |  | Local time is UTC -4 hours


Einstein WU errors out and sees:
Stderr output
<core_client_version>7.6.31</core_client_version>
<![CDATA[
<message>
process exited with code 11 (0xb, -245)
</message>
<stderr_txt>
02:45:18 (2285): [normal]: This Einstein@home App was built at: Jan 16 2017 08:09:16

02:45:18 (2285): [normal]: Start of BOINC application '../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati'.
02:45:18 (2285): [debug]: 1e+16 fp, 2.3e+09 fp/s, 4622774 s, 1284h06m13s66
command line: ../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati --inputfile ../../projects/einstein.phys.uwm.edu/LATeah0030L.dat --alpha 4.42281478648 --delta -0.0345027837249 --skyRadius 2.152570e-06 --ldiBins 15 --f0start 804.0 --f0Band 8.0 --firstSkyPoint 0 --numSkyPoints 1 --f1dot -1e-13 --f1dotBand 1e-13 --df1dot 3.344368011e-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_LATeah0030L_0812_2385755.dat --debug 1 --device 0 -o LATeah0030L_812.0_0_0.0_2385755_0_0.out
output files: 'LATeah0030L_812.0_0_0.0_2385755_0_0.out' '../../projects/einstein.phys.uwm.edu/LATeah0030L_812.0_0_0.0_2385755_0_0' 'LATeah0030L_812.0_0_0.0_2385755_0_0.out.cohfu' '../../projects/einstein.phys.uwm.edu/LATeah0030L_812.0_0_0.0_2385755_0_1'
02:45:18 (2285): [debug]: Flags: X64 SSE SSE2 GNUC X86 GNUX86
02:45:18 (2285): [debug]: glibc version/release: 2.23/stable
02:45:18 (2285): [debug]: Set up communication with graphics process.
boinc_get_opencl_ids returned [0x11bd898 , 0x7f55173d5ec0] 
Using OpenCL platform provided by: Mesa
Using OpenCL device "AMD CAPE VERDE (DRM 2.43.0 / 4.4.0-78-generic, LLVM 4.0.0)" by: AMD
Max allocation limit: 1503238553
Global mem size: 2147483648
OpenCL device has FP64 support

-- signal handler called: signal 1
2 stack frames obtained for this thread:
Frame 22:
	Binary file: ../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati (0x48b101)
	Source file: hs_boinc_extras.c (Function: sighandler / Line: 291)
Frame 21:
	Binary file: /usr/lib/x86_64-linux-gnu/libLLVM-4.0.so.1 (0x7f551315ac71)
	Offset info: _ZN4llvm9LiveRange4findENS_9SlotIndexE+0x21
Frame 20:
	Binary file: /usr/lib/x86_64-linux-gnu/libLLVM-4.0.so.1 (0x7f551315ac71)
	Offset info: _ZN4llvm9LiveRange4findENS_9SlotIndexE+0x21
Frame 19:
	Binary file: /usr/lib/x86_64-linux-gnu/libLLVM-4.0.so.1 (0x7f55132940ec)
	Offset info: +0xa070ec
Frame 18:
	Binary file: /usr/lib/x86_64-linux-gnu/libLLVM-4.0.so.1 (0x7f55132986e9)
	Offset info: +0xa0b6e9
Frame 17:
	Binary file: /usr/lib/x86_64-linux-gnu/libLLVM-4.0.so.1 (0x7f551329a9db)
	Offset info: +0xa0d9db
Frame 16:
	Binary file: /usr/lib/x86_64-linux-gnu/libLLVM-4.0.so.1 (0x7f551329b81e)
	Offset info: +0xa0e81e
Frame 15:
	Binary file: /usr/lib/x86_64-linux-gnu/libLLVM-4.0.so.1 (0x7f55131cc8a1)
	Offset info: _ZN4llvm19MachineFunctionPass13runOnFunctionERNS_8FunctionE+0x91
Frame 14:
	Binary file: /usr/lib/x86_64-linux-gnu/libLLVM-4.0.so.1 (0x7f5513065f92)
	Offset info: _ZN4llvm13FPPassManager13runOnFunctionERNS_8FunctionE+0x1f2
Frame 13:
	Binary file: /usr/lib/x86_64-linux-gnu/libLLVM-4.0.so.1 (0x7f5513066043)
	Offset info: _ZN4llvm13FPPassManager11runOnModuleERNS_6ModuleE+0x33
Frame 12:
	Binary file: /usr/lib/x86_64-linux-gnu/libLLVM-4.0.so.1 (0x7f5513065bdf)
	Offset info: _ZN4llvm6legacy15PassManagerImpl3runERNS_6ModuleE+0x30f
Frame 11:
	Binary file: /usr/lib/x86_64-linux-gnu/libMesaOpenCL.so.1 (0x7f5516280b91)
	Offset info: +0x240b91
Frame 10:
	Binary file: /usr/lib/x86_64-linux-gnu/libMesaOpenCL.so.1 (0x7f5516281210)
	Offset info: +0x241210
Frame 9:
	Binary file: /usr/lib/x86_64-linux-gnu/libMesaOpenCL.so.1 (0x7f551627d16a)
	Offset info: +0x23d16a
Frame 8:
	Binary file: /usr/lib/x86_64-linux-gnu/libMesaOpenCL.so.1 (0x7f551626da4f)
	Offset info: +0x22da4f
Frame 7:
	Binary file: /usr/lib/x86_64-linux-gnu/libMesaOpenCL.so.1 (0x7f551624bdeb)
	Offset info: +0x20bdeb
Frame 6:
	Binary file: ../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati (0x48d609)
	Offset info: initialize_ocl+0x9a9
	Source file: unknown (Function: initialize_ocl / Line: 0)
Frame 5:
	Binary file: ../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati (0x48dc7b)
	Offset info: gen_fft_get_ctx+0x14b
	Source file: unknown (Function: gen_fft_get_ctx / Line: 0)
Frame 4:
	Binary file: ../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati (0x4795fc)
	Offset info: MAIN+0x15c
	Source file: HSgammaPulsar.c (Function: MAIN / Line: 4230)
Frame 3:
	Binary file: ../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati (0x46c06f)
	Offset info: main+0x5ff
	Source file: hs_boinc_extras.c (Function: worker / Line: 833)
	Source file: hs_boinc_extras.c (Function: main / Line: 1039)
Frame 2:
	Binary file: /lib/x86_64-linux-gnu/libc.so.6 (0x7f55173fa830)
	Offset info: __libc_start_main+0xf0
Frame 1:
	Binary file: ../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati (0x46e569)
	Source file: unknown (Function: _start / Line: 0)

End of stcaktrace
02:45:21 (2285): called boinc_finish
</stderr_txt>


at least this is a different error. :-)

======

try a seti WU.

the error details in the task say:
https://setiathome.berkeley.edu/result.php?resultid=5771467818
<stderr_txt>
OpenCL platform detected: Advanced Micro Devices, Inc.
Number of OpenCL devices found : 1
BOINC assigns slot on device #0.
Info: BOINC provided OpenCL device ID used
ERROR: OpenCL kernel/call 'Creating Command Queue. (clCreateCommandQueue)' call failed (-6) in file ../../src/GPU_lock.cpp near line 1359.
Waiting 30 sec before restart...

...repeating...


pan: Try the other PPA; the wi-pe dik; try Ubuntu 16.17. Look for Driver version 2.1 (that worked on the other machine.(running on 16.10))
ID: 78347 · Report as offensive
Profilejay_e

Send message
Joined: 8 Mar 07
Posts: 115
United States
Message 78366 - Posted: 30 May 2017, 7:30:59 UTC - in response to Message 78347.  

On Ubuntu, try another PPA - get a different error

sudo add-apt-repository ppa:ubuntu-x-swat/updates
see https://launchpad.net/~ubuntu-x-swat/+archive/ubuntu/updates

skip

retry on 16.04
working ---- Padoka PPA
https://launchpad.net/~paulo-miguel-dias/+archive/ubuntu/mesa
-
sudo add-apt-repository ppa:paulo-miguel-dias/mesa
sudo apt-get update


jay@pc-10:~$ sudo aptitude update
Get: 1 file:/var/opt/amdgpu-pro-local ./ InRelease
Ign file:/var/opt/amdgpu-pro-local ./ InRelease
Get: 2 file:/var/opt/amdgpu-pro-local ./ Release [814 B]
Get: 3 file:/var/opt/amdgpu-pro-local ./ Release [814 B]
Get: 4 file:/var/opt/amdgpu-pro-local ./ Release.gpg
Ign file:/var/opt/amdgpu-pro-local ./ Release.gpg
Hit http://us.archive.ubuntu.com/ubuntu xenial InRelease
Hit http://archive.canonical.com/ubuntu xenial InRelease
Get: 5 http://security.ubuntu.com/ubuntu xenial-security InRelease [102 kB]
Get: 6 http://us.archive.ubuntu.com/ubuntu xenial-updates InRelease [102 kB]
Get: 7 http://us.archive.ubuntu.com/ubuntu xenial-backports InRelease [102 kB]
Get: 8 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial InRelease [17.5 kB]
Get: 9 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 Packages [24.5 kB]
Get: 10 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main i386 Packages [24.5 kB]
Get: 11 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main Translation-en [10.4 kB]


Current status: 32 (+32) upgradable, 283 (+181) new.



time sudo aptitude upgrade
Resolving dependencies...
The following NEW packages will be installed:
libclang-common-4.0-dev{a} libllvm4.0{a} libllvm5.0{a} xserver-xorg-video-mach64{a} xserver-xorg-video-r128{a}
The following packages will be REMOVED:
libclang-common-3.7-dev{u} libllvm3.7{u}
The following packages will be upgraded:
libclc-amdgcn libclc-dev libclc-r600 libdrm-amdgpu1 libdrm-dev libdrm-intel1 libdrm-nouveau2 libdrm-radeon1 libdrm2 libegl1-mesa libegl1-mesa-drivers
libgbm1 libgl1-mesa-dev libgl1-mesa-dri libgl1-mesa-glx libglapi-mesa libgles2-mesa libva-drm1 libva-x11-1 libva1 libwayland-client0
libwayland-cursor0 libwayland-egl1-mesa libwayland-server0 libxatracker2 mesa-common-dev mesa-opencl-icd xserver-xorg-video-amdgpu
xserver-xorg-video-ati xserver-xorg-video-intel xserver-xorg-video-nouveau xserver-xorg-video-radeon
32 packages upgraded, 5 newly installed, 2 to remove and 0 not upgraded.
Need to get 52.4 MB of archives. After unpacking 111 MB will be used.
Do you want to continue? [Y/n/?] y


list
-----
Get: 1 http://us.archive.ubuntu.com/ubuntu xenial/universe amd64 xserver-xorg-video-r128 amd64 6.10.0-1build2 [47.9 kB]
Get: 2 http://us.archive.ubuntu.com/ubuntu xenial/universe amd64 xserver-xorg-video-mach64 amd64 6.9.5-1build2 [58.2 kB]
Get: 3 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libllvm4.0 amd64 1:4.0~svn291444-0~x~padoka0 [12.2 MB]
Get: 4 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libclang-common-4.0-dev amd64 1:4.0~svn291444-0~x~padoka0 [2,918 kB]         
Get: 5 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libclc-amdgcn all 0.2.0+git20170302.0107.17648cd~x~padoka0 [2,438 kB]        
Get: 6 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libclc-r600 all 0.2.0+git20170302.0107.17648cd~x~padoka0 [604 kB]            
Get: 7 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libclc-dev all 0.2.0+git20170302.0107.17648cd~x~padoka0 [24.8 kB]            
Get: 8 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libdrm-dev amd64 2.4.81+git1705261543.81312f3~x~padoka0 [233 kB]             
Get: 9 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libdrm2 amd64 2.4.81+git1705261543.81312f3~x~padoka0 [37.3 kB]               
Get: 10 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libdrm-intel1 amd64 2.4.81+git1705261543.81312f3~x~padoka0 [68.0 kB]        
Get: 11 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libdrm-radeon1 amd64 2.4.81+git1705261543.81312f3~x~padoka0 [29.7 kB]       
Get: 12 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libdrm-nouveau2 amd64 2.4.81+git1705261543.81312f3~x~padoka0 [24.5 kB]      
Get: 13 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libdrm-amdgpu1 amd64 2.4.81+git1705261543.81312f3~x~padoka0 [26.2 kB]       
Get: 14 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libwayland-egl1-mesa amd64 1:17.2~git170526160800.a958a30~x~padoka0 [44.9 kB]
Get: 15 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libwayland-client0 amd64 1.13.0+git201705130017.0eefe99~x~padoka0 [24.4 kB] 
Get: 16 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libwayland-server0 amd64 1.13.0+git201705130017.0eefe99~x~padoka0 [30.4 kB] 
Get: 17 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libllvm5.0 amd64 1:5.0~svn303494-0~x~padoka0 [13.2 MB]                      
Get: 18 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 mesa-opencl-icd amd64 1:17.2~git170526160800.a958a30~x~padoka0 [7,743 kB]   
Get: 19 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libgl1-mesa-dev amd64 1:17.2~git170526160800.a958a30~x~padoka0 [43.6 kB]    
Get: 20 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 mesa-common-dev amd64 1:17.2~git170526160800.a958a30~x~padoka0 [548 kB]     
Get: 21 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libgl1-mesa-glx amd64 1:17.2~git170526160800.a958a30~x~padoka0 [1,281 kB]   
Get: 22 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libgles2-mesa amd64 1:17.2~git170526160800.a958a30~x~padoka0 [52.8 kB]      
Get: 23 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libglapi-mesa amd64 1:17.2~git170526160800.a958a30~x~padoka0 [61.3 kB]      
Get: 24 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libegl1-mesa amd64 1:17.2~git170526160800.a958a30~x~padoka0 [115 kB]        
Get: 25 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libgl1-mesa-dri amd64 1:17.2~git170526160800.a958a30~x~padoka0 [7,158 kB]   
Get: 26 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libgbm1 amd64 1:17.2~git170526160800.a958a30~x~padoka0 [62.6 kB]            
Get: 27 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libegl1-mesa-drivers amd64 1:17.2~git170526160800.a958a30~x~padoka0 [42.9 kB]
Get: 28 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libva-drm1 amd64 1.8.3+git201705261546-0~x~padoka0 [14.6 kB]                
Get: 29 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libva1 amd64 1.8.3+git201705261546-0~x~padoka0 [58.2 kB]                    
Get: 30 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libva-x11-1 amd64 1.8.3+git201705261546-0~x~padoka0 [19.3 kB]               
Get: 31 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libwayland-cursor0 amd64 1.13.0+git201705130017.0eefe99~x~padoka0 [13.2 kB] 
Get: 32 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 libxatracker2 amd64 1:17.2~git170526160800.a958a30~x~padoka0 [1,203 kB]     
Get: 33 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 xserver-xorg-video-amdgpu amd64 1.3.99+git1705261614.000e5ea~x~padoka0 [54.4 kB]
Get: 34 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 xserver-xorg-video-radeon amd64 1:7.9.99+git1705261616.d80d01a~x~padoka0 [164 kB]
Get: 35 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 xserver-xorg-video-ati amd64 1:7.9.99+git1705261616.d80d01a~x~padoka0 [27.0 kB]
Get: 36 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 xserver-xorg-video-intel amd64 2:2.99.917+git1704220121.c72bb27~x~padoka0 [1,589 kB]
Get: 37 http://ppa.launchpad.net/paulo-miguel-dias/mesa/ubuntu xenial/main amd64 xserver-xorg-video-nouveau amd64 1:1.0.15+git1704222053.a7c190e~x~padoka0 [122 kB]
Fetched 52.4 MB in 1min 51s (472 kB/s) 


etc


Current status: 0 (-32) upgradable.

real 2m59.823s
user 0m26.228s
sys 0m6.440s

C

reboot and test


Tue 30 May 2017 02:54:33 AM EDT |  | Starting BOINC client version 7.6.31 for x86_64-pc-linux-gnu
Tue 30 May 2017 02:54:33 AM EDT |  | log flags: file_xfer, sched_ops, task
Tue 30 May 2017 02:54:33 AM EDT |  | Libraries: libcurl/7.47.0 OpenSSL/1.0.2g zlib/1.2.8 libidn/1.32 librtmp/2.3
Tue 30 May 2017 02:54:33 AM EDT |  | Data directory: /var/lib/boinc-client
Tue 30 May 2017 02:54:33 AM EDT |  | OpenCL: AMD/ATI GPU 0: AMD CAPE VERDE (DRM 2.43.0 / 4.4.0-78-generic, LLVM 5.0.0) (driver version 17.2.0-devel - padoka PPA, device version OpenCL 1.1 Mesa 17.2.0-devel - padoka PPA, 2048MB, 2048MB available, 512 GFLOPS peak)
Tue 30 May 2017 02:54:33 AM EDT |  | Host name: pc-10
Tue 30 May 2017 02:54:33 AM EDT |  | Processor: 4 AuthenticAMD AMD Phenom(tm) 9550 Quad-Core Processor [Family 16 Model 2 Stepping 3]
Tue 30 May 2017 02:54:33 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 3dnowext 3dnow constant_tsc rep_good nopl nonstop_tsc extd_apicid pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs hw_pstate vmmcall npt lbrv svm_lock
Tue 30 May 2017 02:54:33 AM EDT |  | OS: Linux: 4.4.0-78-generic
Tue 30 May 2017 02:54:33 AM EDT |  | Memory: 5.83 GB physical, 9.77 GB virtual
Tue 30 May 2017 02:54:33 AM EDT |  | Disk: 57.67 GB total, 48.71 GB free
Tue 30 May 2017 02:54:33 AM EDT |  | Local time is UTC -4 hours


get einstein
errors out in two seconds, BUT different error. Getting closer??

https://einsteinathome.org/host/12533566/tasks
https://einsteinathome.org/task/650574264
[code]
Stderr output

<core_client_version>7.6.31</core_client_version>
<![CDATA[
<message>
process exited with code 1 (0x1, -255)
</message>
<stderr_txt>
03:00:06 (3893): [normal]: This Einstein@home App was built at: Jan 16 2017 08:09:16

03:00:06 (3893): [normal]: Start of BOINC application '../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati'.
03:00:06 (3893): [debug]: 1e+16 fp, 2.3e+09 fp/s, 4622774 s, 1284h06m13s66
command line: ../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati --inputfile ../../projects/einstein.phys.uwm.edu/LATeah0030L.dat --alpha 4.42281478648 --delta -0.0345027837249 --skyRadius 2.152570e-06 --ldiBins 15 --f0start 876.0 --f0Band 8.0 --firstSkyPoint 0 --numSkyPoints 1 --f1dot -1e-13 --f1dotBand 1e-13 --df1dot 3.344368011e-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_LATeah0030L_0884_4619655.dat --debug 1 --device 0 -o LATeah0030L_884.0_0_0.0_4619655_1_0.out
output files: 'LATeah0030L_884.0_0_0.0_4619655_1_0.out' '../../projects/einstein.phys.uwm.edu/LATeah0030L_884.0_0_0.0_4619655_1_0' 'LATeah0030L_884.0_0_0.0_4619655_1_0.17.2.0-devel - padoka PPA, device version OpenCL 1.1 Mesa 17.2.0-devel - padoka PPA, 2048MB, 2048MB available, 512 GFLOPS peak)
Tue 30 May 2017 02:54:33 AM EDT | | Host name: pc-10
Tue 30 May 2017 02:54:33 AM EDT | | Processor: 4 AuthenticAMD AMD Phenom(tm) 9550 Quad-Core Processor [Family 16 Model 2 Stepping 3]
Tue 30 May 2017 02:54:33 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 3dnowext 3dnow constant_tsc rep_good nopl nonstop_tsc extd_apicid pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs hw_pstate vmmcall npt lbrv svm_lock
Tue 30 May 2017 02:54:33 AM EDT | | OS: Linux: 4.4.0-78-generic
Tue 30 May 2017 02:54:33 AM EDT | | Memory: 5.83 GB physical, 9.77 GB virtual
Tue 30 May 2017 02:54:33 AM EDT | | Disk: 57.67 GB total, 48.71 GB free
Tue 30 May 2017 02:54:33 AM EDT | | Local time is UTC -4 hours
[/code]

get einstein
errors out in two seconds, BUT different error. Getting closer??

https://einsteinathome.org/host/12533566/tasks
https://einsteinathome.org/task/650574264
[code]
Stderr output

<core_client_version>7.6.31</core_client_version>
<![CDATA[
<message>
process exited with code 1 (0x1, -255)
</message>
<stderr_txt>
03:00:06 (3893): [normal]: This Einstein@home App was built at: Jan 16 2017 08:09:16

03:00:06 (3893): [normal]: Start of BOINC application '../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati'.
03:00:06 (3893): [debug]: 1e+16 fp, 2.3e+09 fp/s, 4622774 s, 1284h06m13s66
command line: ../../projects/einstein.phys.uwm.edu/hsgamma_FGRPB1G_1.18_x86_64-pc-linux-gnu__FGRPopencl1K-ati --inputfile ../../projects/einstein.phys.uwm.edu/LATeah0030L.dat --alpha 4.42281478648 --delta -0.0345027837249 --skyRadius 2.152570e-06 --ldiBins 15 --f0start 876.0 --f0Band 8.0 --firstSkyPoint 0 --numSkyPoints 1 --f1dot -1e-13 --f1dotBand 1e-13 --df1dot 3.344368011e-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_LATeah0030L_0884_4619655.dat --debug 1 --device 0 -o LATeah0030L_884.0_0_0.0_4619655_1_0.out
output files: 'LATeah0030L_884.0_0_0.0_4619655_1_0.out' '../../projects/einstein.phys.uwm.edu/LATeah0030L_884.0_0_0.0_4619655_1_0' 'LATeah0030L_884.0_0_0.0_4619655_1_0.out.cohfu' '../../projects/einstein.phys.uwm.edu/LATeah0030L_884.0_0_0.0_4619655_1_1'
03:00:06 (3893): [debug]: Flags: X64 SSE SSE2 GNUC X86 GNUX86
03:00:06 (3893): [debug]: glibc version/release: 2.23/stable
03:00:06 (3893): [debug]: Set up communication with graphics process.
boinc_get_opencl_ids returned [0x2de7278 , 0x7f852b992f80]
Using OpenCL platform provided by: Mesa
Using OpenCL device "AMD CAPE VERDE (DRM 2.43.0 / 4.4.0-78-generic, LLVM 5.0.0)" by: AMD
Max allocation limit: 1503238553
Global mem size: 2147483648
OpenCL device has FP64 support
Invalid bitcast
%21 = bitcast %union.anon addrspace(4)* %6 to i8*
Invalid bitcast
%22 = bitcast %union.anon addrspace(4)* %4 to i8*
Invalid bitcast
%23 = bitcast %union.anon addrspace(4)* %2 to i8*
LLVM ERROR: Broken function found, compilation aborted!

</stderr_txt>
[/code]

Note above:
vice version OpenCL 1.1 Mesa 17.2.0-devel - padoka PPA,
ID: 78366 · Report as offensive
Profilejay_e

Send message
Joined: 8 Mar 07
Posts: 115
United States
Message 78397 - Posted: 30 May 2017, 19:13:29 UTC - in response to Message 78366.  
Last modified: 30 May 2017, 19:30:43 UTC

Today, Tuesday is The SETI maintenance day. So, no SET WU to test.
will clean disk and start afresh with Ububtu 16.10.
I do have a 16.10 working on another machine with a 7750.
I'll edit this post and add what it says.

-->>> THIS DRIVER WORKS!!! <<<---

boinc says:
Tue 30 May 2017 09:51:16 AM EDT |  | Starting BOINC client version 7.6.33 for x86_64-pc-linux-gnu
Tue 30 May 2017 09:51:16 AM EDT |  | log flags: file_xfer, sched_ops, task
Tue 30 May 2017 09:51:16 AM EDT |  | Libraries: libcurl/7.50.1 OpenSSL/1.0.2g zlib/1.2.8 libidn/1.33 librtmp/2.3
Tue 30 May 2017 09:51:16 AM EDT |  | Data directory: /var/lib/boinc-client
Tue 30 May 2017 09:51:17 AM EDT |  | OpenCL: AMD/ATI GPU 0: AMD CAPE VERDE (DRM 2.46.0 / 4.8.0-53-generic, LLVM 4.0.1) (driver version 17.2.0-devel, device version OpenCL 1.1 Mesa 17.2.0-devel, 2048MB, 2048MB available, 512 GFLOPS peak)
Tue 30 May 2017 09:51:17 AM EDT |  | Host name: pc-14-large
Tue 30 May 2017 09:51:17 AM EDT |  | Processor: 8 AuthenticAMD AMD FX(tm)-8150 Eight-Core Processor [Family 21 Model 1 Stepping 2]
Tue 30 May 2017 09:51:17 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 extd_apicid aperfmperf eagerfpu 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 lwp fma4 nodeid_msr topoext perfctr_core perfctr_nb cpb hw_pstate vmmcall arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold
Tue 30 May 2017 09:51:17 AM EDT |  | OS: Linux: 4.8.0-53-generic
Tue 30 May 2017 09:51:17 AM EDT |  | Memory: 11.64 GB physical, 48.83 GB virtual
Tue 30 May 2017 09:51:17 AM EDT |  | Disk: 134.57 GB total, 126.32 GB free
Tue 30 May 2017 09:51:17 AM EDT |  | Local time is UTC -4 hours
Tue 30 May 2017 09:51:17 AM EDT |  | Config: GUI RPCs allowed from:
Tue 30 May 2017 09:51:17 AM EDT | Einstein@Home | URL http://einstein.phys.uwm.edu/; Computer ID 12201025; resource share 5
Tue 30 May 2017 09:51:17 AM EDT | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 8268458; resource share 5
Tue 30 May 2017 09:51:17 AM EDT | World Community Grid | URL http://www.worldcommunitygrid.org/; Computer ID 3197561; resource share 95
Tue 30 May 2017 09:51:17 AM EDT | World Community Grid | General prefs: from World Community Grid (last modified 26-May-2017 05:16:17)
Tue 30 May 2017 09:51:17 AM EDT | World Community Grid | Host location: none
Tue 30 May 2017 09:51:17 AM EDT | World Community Grid | General prefs: using your defaults
Tue 30 May 2017 09:51:17 AM EDT |  | Preferences:


Einstein says:

Task 650725687
Name:
LATeah0030L_908.0_0_0.0_8798805_1
Workunit ID:
298339438
Created:
30 May 2017 16:10:31 GMT
Sent:
30 May 2017 16:34:21 GMT
Report deadline:
13 Jun 2017 16:34:21 GMT
Received:
30 May 2017 18:54:14 GMT
Server state:
Over
Outcome:
Success
Client state:
Done
Exit status:
0 (0x00000000)
Computer:
12201025
Run time (sec):
4,756.65
CPU time (sec):
188.76
Peak working set size (MB):
284.64
Peak swap size (MB):
857.83
Peak disk usage (MB):
0.33
Validation state:
Initial
Claimed credit:
6.73
Granted credit:
0.00
Application:
Gamma-ray pulsar binary search #1 on GPUs v1.18 (FGRPopencl1K-ati)
x86_64-pc-linux-gnu 


clinfo says:
Number of platforms                               1
  Platform Name                                   Clover
  Platform Vendor                                 Mesa
  Platform Version                                OpenCL 1.1 Mesa 17.2.0-devel
  Platform Profile                                FULL_PROFILE
  Platform Extensions                             cl_khr_icd
  Platform Extensions function suffix             MESA

  Platform Name                                   Clover
Number of devices                                 1
  Device Name                                     AMD CAPE VERDE (DRM 2.46.0 / 4.8.0-53-generic, LLVM 4.0.1)
  Device Vendor                                   AMD
  Device Vendor ID                                0x1002
  Device Version                                  OpenCL 1.1 Mesa 17.2.0-devel
  Driver Version                                  17.2.0-devel
  Device OpenCL C Version                         OpenCL C 1.1 
  Device Type                                     GPU
  Device Profile                                  FULL_PROFILE
  Max compute units                               8
  Max clock frequency                             800MHz
  Max work item dimensions                        3
  Max work item sizes                             256x256x256
  Max work group size                             256
  Preferred work group size multiple              64
  Preferred / native vector sizes                 
    char                                                16 / 16      
    short                                                8 / 8       
    int                                                  4 / 4       
    long                                                 2 / 2       
    half                                                 0 / 0        (n/a)
    float                                                4 / 4       
    double                                               2 / 2        (cl_khr_fp64)
  Half-precision Floating-point support           (n/a)
  Single-precision Floating-point support         (core)
    Denormals                                     No
    Infinity and NANs                             Yes
    Round to nearest                              Yes
    Round to zero                                 No
    Round to infinity                             No
    IEEE754-2008 fused multiply-add               No
    Support is emulated in software               No
    Correctly-rounded divide and sqrt operations  No
  Double-precision Floating-point support         (cl_khr_fp64)
    Denormals                                     Yes
    Infinity and NANs                             Yes
    Round to nearest                              Yes
    Round to zero                                 Yes
    Round to infinity                             Yes
    IEEE754-2008 fused multiply-add               Yes
    Support is emulated in software               No
    Correctly-rounded divide and sqrt operations  No
  Address bits                                    64, Little-Endian
  Global memory size                              2147483648 (2GiB)
  Error Correction support                        No
  Max memory allocation                           1503238553 (1.4GiB)
  Unified memory for Host and Device              Yes
  Minimum alignment for any data type             128 bytes
  Alignment of base address                       1024 bits (128 bytes)
  Global Memory cache type                        None
  Image support                                   No
  Local memory type                               Local
  Local memory size                               32768 (32KiB)
  Max constant buffer size                        1503238553 (1.4GiB)
  Max number of constant args                     16
  Max size of kernel argument                     1024
  Queue properties                                
    Out-of-order execution                        No
    Profiling                                     Yes
  Profiling timer resolution                      0ns
  Execution capabilities                          
    Run OpenCL kernels                            Yes
    Run native kernels                            No
  Device Available                                Yes
  Compiler Available                              Yes
  Device Extensions                               cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_byte_addressable_store cl_khr_fp64

NULL platform behavior
  clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...)  Clover
  clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...)   Success [MESA]
  clCreateContext(NULL, ...) [default]            Success [MESA]
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU)  Success (1)
    Platform Name                                 Clover
    Device Name                                   AMD CAPE VERDE (DRM 2.46.0 / 4.8.0-53-generic, LLVM 4.0.1)
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL)  Success (1)
    Platform Name                                 Clover
    Device Name                                   AMD CAPE VERDE (DRM 2.46.0 / 4.8.0-53-generic, LLVM 4.0.1)

ICD loader properties
  ICD loader Name                                 OpenCL ICD Loader
  ICD loader Vendor                               OCL Icd free software
  ICD loader Version                              2.2.9
  ICD loader Profile                              OpenCL 2.1
ID: 78397 · Report as offensive
Profilejay_e

Send message
Joined: 8 Mar 07
Posts: 115
United States
Message 78435 - Posted: 31 May 2017, 18:34:54 UTC - in response to Message 78397.  

Greetings,
A bit of a recap,
I have two machines - both running Ubuntu Mate - Both with Radeon 7750.
One is running Ubuntu Mate 16.10 - which is nether the LTS release nor the Latest release (it is between them).
I can crunch Einstien and SETI GPU WU on it.- and have been for several months.

So , I wanted to upgrade the other machine, runng fglrx drivers with Ubuntu 14.something.
But, I could not find the correct driver - paired with a release for it.
It *appears* that the AMDGPU-Pro driver does crunch on any of the 3 latest Ubuntu releases.
I did a hit and miss test of 16.04, 16.10, and 17.04 - all failed.
So I started over with a better test plan.
(I am not a tester. I should have started with a better trail of results - so I started over, again.)

In trial and error fashion, for each release, I am trying the drivers released by Ubuntu and PPAs:
- The AMDGPU-pro driver ( a script-built set of packages.)
-The mesa-opencl-icd package
- the x-swat PPA
-The stable Padoka Stable Mesa PPA -- https://launchpad.net/~paulo-miguel-dias/+archive/ubuntu/pkppa/
- The unstale (lates release) Padoka PPA -- https://launchpad.net/~paulo-miguel-dias/+archive/ubuntu/mesa
- Other PPAs as -- https://launchpad.net/~oibaf/+archive/ubuntu/graphics-drivers

(Note: Dias and Oibaf have worked together putting releases into Launchpad.)
The last post in this thread was a recap that works - but I didn't save the info on which PPA it was.
That release was installed when 16.10 WAS the latest release. It no longer is, and Most PPAs only support the LTS and the latest.
So I am now testing Ubuntu Mate 16.10 with the various drivers available.
(Looking at the output of clinfo shows differences; but I don't know which are significant to crunch a WU.
Probably depends on the WU.)

The results so far on this release: (I'm leaving out the details today.)
- Tuesday May 30 2017 Mesa opencl 'out-of-the box' : Fails
- mesa-opencl-icd mesa-common-dev libopencl-2.1-1 ocl-icd-libopencl1 : Fails
- try the Stable -- https://launchpad.net/~paulo-miguel-dias/+archive/ubuntu/pkppa : Fails
- try Updated and Optimized Open Graphics Drivers -- https://launchpad.net/~oibaf/+archive/ubuntu/graphics-drivers/
(sudo add-apt-repository ppa:oibaf/graphics-drivers)

It has been running for an hour and has 20 minutes to go on an Einstein GPU WU.
I'll do another posting when there is one successful completion. (yay!)
ID: 78435 · Report as offensive
Profilejay_e

Send message
Joined: 8 Mar 07
Posts: 115
United States
Message 78437 - Posted: 31 May 2017, 19:05:36 UTC - in response to Message 78435.  

Einstein WU completed and waiting on Validation

Yay

https://einsteinathome.org/task/651051489

Stderr output

<core_client_version>7.6.33</core_client_version>
<![CDATA[
<stderr_txt>
1e-15 f1dot_start: -1e-13 f1dot_band: 1e-13
% Filling array of photon pairs
etc., etc. etc.

.
% Time spent on semicoherent stage: 4653.1636s
% Writing semicoherent output file.

% Following up candidate number: 1
% Refining in S
% Following-up in P
Error in OpenCL context: Shader Stats: SGPRS: 11 VGPRS: 5 Code Size: 36374 LDS: 0 Scratch: 0 Max Waves: 10 Spilled SGPRs: 0 Spilled VGPRs: 0 PrivMem VGPRs: 0

% Following up candidate number: 2
% Refining in S
% Following-up in P

% Following up candidate number: 3
% Refining in S
% Following-up in P

% Following up candidate number: 4
% Refining in S
% Following-up in P

% Following up candidate number: 5
% Refining in S
% Following-up in P

% Following up candidate number: 6
% Refining in S
% Following-up in P

% Following up candidate number: 7
% Refining in S
% Following-up in P
% C 8 1262

% Following up candidate number: 8
% Refining in S
% Following-up in P

% Following up candidate number: 9
% Refining in S
% Following-up in P

% Following up candidate number: 10
% Refining in S
% Following-up in P
% Writing follow-up output file.
FPU status flags:
14:35:02 (4835): [normal]: done. calling boinc_finish(0).
14:35:02 (4835): called boinc_finish

</stderr_txt>



so far, so good.
will see if seti is available.
ID: 78437 · Report as offensive

Message boards : GPUs : which PPA for Ubuntu Mesa openCL for Raden 7750???

Copyright © 2024 University of California.
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.