Message boards :
Number crunching :
Milkyway@Home 1.02 (opencl_AMD_ati) all erroring out
Message board moderation
Author | Message |
---|---|
Send message Joined: 19 Aug 11 Posts: 38 Credit: 50,018,810 RAC: 0 |
Hey folks, whats happening to me ? Any guesses ? Using AMD 13.12 and HD 6950 gpu. Modified Fit works perfectly. Greetings |
Send message Joined: 18 Jul 09 Posts: 300 Credit: 303,562,776 RAC: 0 |
I've seen this before with another user. For now, just run Modified Fit WUs. Try a detach/reattach and see what happens then. Are running with an app_config file? If so, please post it. Also, can you downgrade to CCC 13.1? or another known good driver? |
Send message Joined: 19 Aug 11 Posts: 38 Credit: 50,018,810 RAC: 0 |
<app_config> <app> <name>milkyway</name> <gpu_versions> <gpu_usage>1</gpu_usage> <cpu_usage>.5</cpu_usage> </gpu_versions> </app> <app> <name>milkyway_separation__modified_fit</name> <gpu_versions> <gpu_usage>1</gpu_usage> <cpu_usage>.5</cpu_usage> </gpu_versions> </app> </app_config> Found only legacy driver for 13.1 !? Before that i tried 14.4 and beta 14.7 , same issue. |
Send message Joined: 19 Aug 11 Posts: 38 Credit: 50,018,810 RAC: 0 |
I reattached the project, no success :( still using 13.12 AMD drivers. I will rise the cpu usage to .75 , and will see...no success... |
Send message Joined: 18 Jul 09 Posts: 300 Credit: 303,562,776 RAC: 0 |
Are you getting Blue Screens? |
Send message Joined: 18 Jul 11 Posts: 22 Credit: 4,160,197 RAC: 0 |
Just rejoined MW after long absence from BOINC. Also got computation error on first wu's I tried. Also 6950, same version of AMD drivers as Rantanplan. Frankly, I've forgotten most of the tweaks I used to use, so I'll wait to see if there's a resolution to this thread before trying a bunch of stuff. |
Send message Joined: 18 Jul 11 Posts: 22 Credit: 4,160,197 RAC: 0 |
I went back to an older driver, but opencl still errors out. The other task seems to run okay. |
Send message Joined: 19 Aug 11 Posts: 38 Credit: 50,018,810 RAC: 0 |
Are you getting Blue Screens? no, nothing happens when it appears. |
Send message Joined: 8 May 09 Posts: 3321 Credit: 520,562,938 RAC: 28,764 |
I went back to an older driver, but opencl still errors out. The other task seems to run okay. Two things MilkyWay likes the 'release' version of drivers, not always the latest and greatest Beta drivers, nor the older ones either. The other thing is are you leaving a cpu core free just for the gpu to use? The gpu apps need alot of cpu power to keep the gpu fed and working optimally, if it is fighting for cpu time with a cpu workunit it can have problems. |
Send message Joined: 19 Aug 11 Posts: 38 Credit: 50,018,810 RAC: 0 |
4 cores free to crunsh, error in result :( Using SSE4.1 path Found 2 platforms Platform 0 information: Name: Intel(R) OpenCL Version: OpenCL 1.1 Vendor: Intel(R) Corporation Extensions: cl_khr_fp64 cl_khr_icd 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_intel_printf cl_ext_device_fission cl_intel_exec_by_local_thread cl_khr_gl_sharing cl_intel_dx9_media_sharing Profile: FULL_PROFILE Platform 1 information: Name: AMD Accelerated Parallel Processing Version: OpenCL 1.2 AMD-APP (1526.3) Vendor: Advanced Micro Devices, Inc. interesting...how can i disable platform 0 from OpenCL crunshing ? |
Send message Joined: 8 May 09 Posts: 3321 Credit: 520,562,938 RAC: 28,764 |
4 cores free to crunsh, error in result :( Thru an cc_config.xml file kinda sorta like this: <cc_config> <options> <use_all_gpus>1</use_all_gpus> <exclude_gpu> <url>http://milkyway.cs.rpi.edu/milkyway/</url> <device_num>0</device_num> </exclude_gpu> </options> </cc_config> Instead of using the <device_num>0</device_num> you could use this instead: <type>intel_gpu</type> Syntax is critical here, meaning if you get it wrong it will NOT work!! Capital letters DO count!! Here is the wiki on how it all works: http://boinc.berkeley.edu/wiki/Client_configuration |
Send message Joined: 19 Aug 11 Posts: 38 Credit: 50,018,810 RAC: 0 |
ok, gpu related, i know this, but CPU ? |
Send message Joined: 8 May 09 Posts: 3321 Credit: 520,562,938 RAC: 28,764 |
ok, gpu related, i know this, but CPU ? No I can't think of a way right now. |
Send message Joined: 18 Jul 11 Posts: 22 Credit: 4,160,197 RAC: 0 |
Open cl ATI tasks still have computation error for me using 6950. The modified fit task seems to work fine, so I guess I'll just run that. Have no other cpu tasks running at all, so cpu load is not a problem. |
Send message Joined: 22 Jun 13 Posts: 44 Credit: 64,258,609 RAC: 0 |
I am running a 6950 with Catalyst 13.1 drivers and both types of work units are working fine. I can't remember exactly what it was but there is something quirky about the 6950 and it required the older drivers. Hope that helps. |
Send message Joined: 30 Jan 11 Posts: 23 Credit: 33,424,579 RAC: 0 |
I also get this issue on my two 5970's. I am using the latest drivers on my windows 8.1 install. |
Send message Joined: 22 Jan 11 Posts: 375 Credit: 64,657,871 RAC: 0 |
I am running a 6950 with Catalyst 13.1 drivers and both types of work units are working fine. I can't remember exactly what it was but there is something quirky about the 6950 and it required the older drivers. Yep, see the thread I just bumped called 'All Milkyway@Home 1.02 tasks ending in computation error on HD6950'. Is also relevant to owners of 5800s & 5900s. Without the above fix you can do mod fit WUs but not the MW v1.02 WUs. Mikey Not having a free CPU core won't cause the WUs to error out, I know I've done it ;). He probably needs the 13.1 driver & a proper driver clean out as per above thread. Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit 2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7 |
Send message Joined: 28 Mar 10 Posts: 13 Credit: 13,311,528 RAC: 0 |
Hm, rejoined MW just recently and right now I have a different issue. My HD7790 doesn't get any WorkUnits at all - has the work distribution for ATI OpenCL been halted maybe ? |
Send message Joined: 22 Jan 11 Posts: 375 Credit: 64,657,871 RAC: 0 |
Sure you've got the GPU apps ticked in your online MW preferences? Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit 2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7 |
Send message Joined: 28 Mar 10 Posts: 13 Credit: 13,311,528 RAC: 0 |
Yep, another Host is chugging along just fine (NVidia GPU). Looking at the Event Log, the AMD GPU Host was requesting GPU tasks and the reponse always was "Got 0 new tasks, Scheduler request complete". For now it's busy with another project but I'll try again to get it some work tomorrow. But if that doesn't work, I'm a bit at a loss there. I already toggled the option back & forth but it made no difference. Also, all Applications are allowed to run without restrictions. So far, the host got Collatz and Einstein tasks without issues. Only as I wanted to move to MilkyWay, I noted it not getting any work for the card... |
©2024 Astroinformatics Group