| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Beta Testing
Forum: Beta Test Support Forum
Thread: OpenPandemics GPU Beta Test - March 12 2021 [ Issues Thread ] |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 160
|
|
| Author |
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges: 
|
is a cc_config.xml really needed for BOINC to use all GPUs available? Yes And FWIW the HCC GPU app did not play well on machines both Nvidia and AMD cards trying to both run at the same time. Driver conflicts. There were no Intel GPUs at that time. Don't know if that may be a problem here now but based on past experiences it could be.
In 1969 I took an oath to defend and protect the U S Constitution against all enemies, both foreign and Domestic. There was no expiration date.
----------------------------------------  [Edit 1 times, last edit by nanoprobe at Mar 12, 2021 10:24:59 PM] |
||
|
|
ThreadRipper
Veteran Cruncher Sweden Joined: Apr 26, 2007 Post Count: 1317 Status: Offline Project Badges: 
|
Another question I have: That same machine has three GPUs: 7850K iGPU, 1050Ti and GT 710. BOINC runs WUs on the iGPU and 1050Ti simultaneously, but it does not want to try to run on the GT710, despite it being available. Are the WUs assiges specifically to a certain GPU when they get downloaded ot why would it not try to run any on that GPU? Do you have <use_all_gpus>1</use_all_gpus> in your cc_config. xml? Hi, I did not have a cc_config.xml file, but now I creatd one and put it in boinc data directory. However, now I do not have any Nvidia beta WUs left in cache so I'll have to wait and see if this works when more GPU work is available. Maybe @Uplinger knows and can confirm: is a cc_config.xml really needed for BOINC to use all GPUs available? Thanks! At least this is what BOINC config page says: <use_all_gpus>0|1</use_all_gpus> If 1, use all GPUs (otherwise only the most capable ones are used). Requires a client restart. Okay, thanks! Now I restared boic client after creating the cc_config file and the log says this, especially the last line below should be an indication that the file settings have been read? Now only to wait for more task to see if it helped. 2021-03-12 23:47:25 | | OpenCL: NVIDIA GPU 0: GeForce GTX 1050 Ti (driver version 461.72, device version OpenCL 1.2 CUDA, 4096MB, 3376MB available, 2138 GFLOPS peak) 2021-03-12 23:47:25 | | OpenCL: NVIDIA GPU 0: GeForce GTX 1050 Ti (driver version 461.72, device version OpenCL 1.2 CUDA, 4096MB, 3376MB available, 2138 GFLOPS peak) 2021-03-12 23:47:25 | | OpenCL: NVIDIA GPU 1: GeForce GT 710 (driver version 461.72, device version OpenCL 1.2 CUDA, 1024MB, 827MB available, 366 GFLOPS peak) 2021-03-12 23:47:25 | | OpenCL: NVIDIA GPU 1: GeForce GT 710 (driver version 461.72, device version OpenCL 1.2 CUDA, 1024MB, 827MB available, 366 GFLOPS peak) 2021-03-12 23:47:25 | | OpenCL: AMD/ATI GPU 0: AMD Radeon(TM) R7 Graphics (driver version 3188.4, device version OpenCL 2.0 AMD-APP (3188.4), 3176MB, 3176MB available, 922 GFLOPS peak) 2021-03-12 23:47:25 | | Memory: 7.44 GB physical, 11.43 GB virtual 2021-03-12 23:47:25 | | Disk: 237.87 GB total, 35.17 GB free 2021-03-12 23:47:25 | | Local time is UTC +1 hours 2021-03-12 23:47:25 | | No WSL found. 2021-03-12 23:47:25 | | Config: use all coprocessors  Join The International Team: https://www.worldcommunitygrid.org/team/viewTeamInfo.do?teamId=CK9RP1BKX1 AMD TR2990WX @ PBO, 64GB Quad 3200MHz 14-17-17-17-1T, RX6900XT @ Stock AMD 3800X @ PBO AMD 2700X @ 4GHz |
||
|
|
ThreadRipper
Veteran Cruncher Sweden Joined: Apr 26, 2007 Post Count: 1317 Status: Offline Project Badges:
|
is a cc_config.xml really needed for BOINC to use all GPUs available? Yes And FWIW the HCC GPU app did not play well on machines both Nvidia and AMD cards trying to both run at the same time. Driver conflicts. There were no Intel GPUs at that time. Don't know if that may be a problem here now but based on past experiences it could be. I see, yes the HCC was such a long time ago that I do no recall the issues from that time. But right now the PC had one AMD and one NVIDIA GPU task running at the same time. It just did not want to use both NVIDIA GPUs at the same time. But we'll see if the cc_config.xml helped or not once more tasks are released into the wild. Join The International Team: https://www.worldcommunitygrid.org/team/viewTeamInfo.do?teamId=CK9RP1BKX1 AMD TR2990WX @ PBO, 64GB Quad 3200MHz 14-17-17-17-1T, RX6900XT @ Stock AMD 3800X @ PBO AMD 2700X @ 4GHz |
||
|
|
Jorlin
Advanced Cruncher Deutschland Joined: Jan 22, 2020 Post Count: 89 Status: Offline Project Badges: 
|
is a cc_config.xml really needed for BOINC to use all GPUs available? Yes And FWIW the HCC GPU app did not play well on machines both Nvidia and AMD cards trying to both run at the same time. Driver conflicts. There were no Intel GPUs at that time. Don't know if that may be a problem here now but based on past experiences it could be. I see, yes the HCC was such a long time ago that I do no recall the issues from that time. But right now the PC had one AMD and one NVIDIA GPU task running at the same time. It just did not want to use both NVIDIA GPUs at the same time. But we'll see if the cc_config.xml helped or not once more tasks are released into the wild. for testing you could opt in to Einstein@Home... that's what i'm running when nothing else is available for GPU. You can exclude GPUs from certain tasks or anything at all. Maybe this thread is helpful: https://boinc.berkeley.edu/forum_thread.php?id=11531  [Edit 1 times, last edit by Jorlin at Mar 12, 2021 11:08:24 PM] |
||
|
|
widdershins
Veteran Cruncher Scotland Joined: Apr 30, 2007 Post Count: 674 Status: Offline Project Badges: 
|
Me again, still no joy. Same error crops up about an access violation
Access Violation (0xc0000005) at address 0x000000013FF9DB30 read attempt to address 0x00000000 |
||
|
|
Grumpy Swede
Master Cruncher Svíþjóð Joined: Apr 10, 2020 Post Count: 2068 Status: Recently Active Project Badges:
|
Me again, still no joy. Same error crops up about an access violation Access Violation (0xc0000005) at address 0x000000013FF9DB30 read attempt to address 0x00000000 Same error "Access Violation (0xc0000005)" (but at address 0x00007FF7911BDB30) for me on my GTX660M. The same error and at the exact same address for all 4 of my error results. So, app version 7.27 and the fix for the DP issue, did not help. Ah well, the GTX660M is no speed monster, so I think WCG can live without it.  |
||
|
|
widdershins
Veteran Cruncher Scotland Joined: Apr 30, 2007 Post Count: 674 Status: Offline Project Badges:
|
One thing I noticed when looking at the error log was in this part. Is the similarity (and one digit difference) between the two items highlighted significant?
- Unhandled Exception Record - Reason: Access Violation (0xc0000005) at address 0x00000001 3F72DB30 read attempt to address 0x00000000 - Registers - rax=0000000000000030 rbx=0000000000000032 rcx=0000000000000000 rdx=0000000000000030 rsi=0000000000000000 rdi=0000000000a0daf0 r8=0000000000000008 r9=00000000e4fa0000 r10=0000000000000032 r11=0000000000a0dc50 r12=0000000000a0dc50 r13=0000000000000000 r14=0000000000000000 r15=0000000000000002 rip=00000000 3f72db30 rsp=000000000092fd30 rbp=000000000092fe30 cs=0033 ss=002b ds=002b es=002b fs=0053 gs=002b efl=00010206 - Callstack - ChildEBP RetAddr Args to Child 00a0d860 3f725be5 3f7cc158 00000001 00a0bb38 00badeb2 wcgrid_beta29_autodockgpu_7!+0x0 00aa6bd0 3f71c2fe 00aa6c98 3f7ae160 00b979a0 00b79d70 wcgrid_beta29_autodockgpu_7!+0x0 00aa7240 3f71da78 00b8a4e0 00b8a4e0 00b8e030 00b8a4e0 wcgrid_beta29_autodockgpu_7!+0x0 00abf850 3f752061 3f7f3fa8 00000000 00000000 3f7f3fb0 wcgrid_beta29_autodockgpu_7!+0x0 00abf890 771b556d 00000000 00000000 00000000 00000000 wcgrid_beta29_autodockgpu_7!+0x0 00abf8c0 7741372d 00000000 00000000 00000000 00000000 kernel32!BaseThreadInitThunk+0x0 00abf910 00000000 00000000 00000000 00000000 00000000 ntdll!RtlUserThreadStart+0x0 |
||
|
|
Grumpy Swede
Master Cruncher Svíþjóð Joined: Apr 10, 2020 Post Count: 2068 Status: Recently Active Project Badges:
|
The only thing I can say about those errors, is that a card that is NOT OpenCL 1.2 capable, throws the same error (and also like my GTX660M at the same time resets the Display Driver.)
At the beginning of the GPU Beta, WU's were sent out to non OpenCL 1.2 capable cards, and I got the same errors with my GeForce 315M, as with the GTX660M GeForce 315M, is not OpenCl 1.2 capable only OpenCL 1.0 (1.1). So, I'm leaning more and more toward the idea that GTX660M, and other older cards that declare themselves as OpenCL 1.2, in fact isn't fully OpenCL 1.2 |
||
|
|
Speedy51
Veteran Cruncher New Zealand Joined: Nov 4, 2005 Post Count: 1259 Status: Offline Project Badges: 
|
Grumpy Swede wrote
----------------------------------------Well, for what I see both from my GTX980 and iGPU HD4600, these workunits are not any harder than those in the 7.26 Beta. On the contrary they crunch faster than the previous ones. I have to agree with him in the fact that the longest running result I have had so far is 0.05 hours (61 jobs)on a 2070 Uplinger if you wanted to look into it at my host is DESKTOP-FVL1L8F Longest task that I can see in my task list was 0.10 hours this had 77 jobs App version 7.26 also times reflected above are with one CPU core dedicated to the GPU  [Edit 1 times, last edit by Speedy51 at Mar 13, 2021 7:10:25 PM] |
||
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 1928 Status: Offline Project Badges: 
|
Got a total of 9 WUs today.
----------------------------------------The 3 on a Dell i7 with an NVidia card finished on about 20min, strangely now claiming 0.0 credits. Got 6 on my MacBookPro i7 with Intel GD3000/AMD Radeon HD6750M and all of them crash immediately...  Ralf  |
||
|
|
|