Message boards :
News :
Scheduled Maintenance Concluded
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 . . . 13 · Next
Author | Message |
---|---|
Send message Joined: 27 Jun 09 Posts: 12 Credit: 148,038,330 RAC: 0 |
<core_client_version>7.6.22</core_client_version> <![CDATA[ <message> aborted by user </message> <stderr_txt> <search_application> milkyway_separation 1.42 Windows x86_64 double </search_application> Reading preferences ended prematurely BOINC GPU type suggests using OpenCL vendor 'Advanced Micro Devices, Inc.' Error loading Lua script 'astronomy_parameters.txt': [string "number_parameters: 4..."]:1: '<name>' expected near '4' Switching to Parameter File 'astronomy_parameters.txt' <number_WUs> 5 </number_WUs> <number_params_per_WU> 20 </number_params_per_WU> Using AVX path Integral 0 time = 641.348502 s Running likelihood with 84046 stars Likelihood time = 1.657994 s <background_integral> 0.000069192102838 </background_integral> <stream_integral> 87.919837480121885 1231.763016660424500 218.680946939705140 </stream_integral> <background_likelihood> -3.558790208380722 </background_likelihood> <stream_only_likelihood> -19.405544659459434 -5.167057809848690 -3.549714346561023 </stream_only_likelihood> <search_likelihood> -3.135608747542042 </search_likelihood> Using AVX path |
Send message Joined: 25 Feb 13 Posts: 580 Credit: 94,200,158 RAC: 0 |
Peciak, That's weird. The version info doesn't say its an OpenCL app. Let me see if I forgot to set the flag when I recompiled it. Sorry about this, I haven't slept much over the last couple days trying to get this update done. Jake |
Send message Joined: 14 Nov 14 Posts: 9 Credit: 214,644,261 RAC: 0 |
Looks like the new version isn't using the GPU. My 750Ti show no GPU usage and the CPU task is being used fully. The first two WU's validated but used the CPU. Here is from the standard error file if this will help: <core_client_version>7.6.33</core_client_version> <![CDATA[ <stderr_txt> <search_application> milkyway_separation 1.42 Windows x86_64 double </search_application> Reading preferences ended prematurely BOINC GPU type suggests using OpenCL vendor 'NVIDIA Corporation' Error loading Lua script 'astronomy_parameters.txt': [string "number_parameters: 4..."]:1: '<name>' expected near '4' Switching to Parameter File 'astronomy_parameters.txt' <number_WUs> 1 </number_WUs> <number_params_per_WU> 20 </number_params_per_WU> Using AVX path Integral 0 time = 836.284111 s Running likelihood with 84046 stars Likelihood time = 1.842669 s <background_integral> 0.000183211346476 </background_integral> <stream_integral> 35.170988580602945 220.771463270831330 56.153686511532385 </stream_integral> <background_likelihood> -3.782092719257110 </background_likelihood> <stream_only_likelihood> -50.039987133739942 -4.114360911786672 -3.380701603567321 </stream_only_likelihood> <search_likelihood> -2.964758398505063 </search_likelihood> 16:52:25 (20784): called boinc_finish(0) </stderr_txt> |
Send message Joined: 25 Feb 13 Posts: 580 Credit: 94,200,158 RAC: 0 |
Rich, I see it, it was my fault during building of the program. I have a fix made, just double and triple checking to make sure I don't make another silly mistake this time. Will be released soon. Jake |
Send message Joined: 14 Nov 14 Posts: 9 Credit: 214,644,261 RAC: 0 |
Thanks, lol, I'll let the 12 WU's finish up with the CPU and then let more units start loading. I've done many mistakes in my time, fully understand. Keep up the good work. |
Send message Joined: 6 Oct 14 Posts: 46 Credit: 20,017,425 RAC: 0 |
|
Send message Joined: 24 Oct 16 Posts: 12 Credit: 56,127,036 RAC: 0 |
Hey Jake The GPU units are crunching again after downloading ver. 1.42 :) 11/11/2016 3:28:56 PM | Milkyway@Home | Started download of milkyway_1.42_windows_x86_64__opencl_ati_101.exe 11/11/2016 3:29:33 PM | Milkyway@Home | Finished download of milkyway_1.42_windows_x86_64__opencl_ati_101.exe 11/11/2016 3:29:33 PM | Milkyway@Home | Starting task de_modfit_fast_19_3s_136_bundle5_ModfitConstraints3_4_1478900148_13142_0 Thanks again Les Intel(R) Core(TM) i7-2700K CPU @ 3.50GHz [Family 6 Model 42 Stepping 7] (8 processors) AMD Radeon HD 6900 series (Cayman) (2048MB) driver: 1.4.1848 OpenCL: 1.2 Microsoft Windows 7 Pro x64 Service Pack 1, (06.01.7601.00) |
Send message Joined: 25 Feb 13 Posts: 580 Credit: 94,200,158 RAC: 0 |
Hey Everyone, I'm still working on a fix. Having issues with other the Windows OpenCL version but not Linux OpenCL. Need to see if my drivers are back on the machine I'm testing with or if there is actually an issue with the client. Might take a couple hours for a fix. Sorry, Jake |
Send message Joined: 19 Feb 08 Posts: 350 Credit: 141,284,369 RAC: 0 |
Well, yes, they start crunching, but at ~70% they restart with 0%. In 45 minutes not a single gpu wu finished. |
Send message Joined: 2 Oct 16 Posts: 167 Credit: 1,008,062,758 RAC: 1,016 |
My 280x in Linux is completing WUs in about 150-155 seconds at 4x. Same time per task as before. Still lots of validation inconclusive but I had a lot of those before. It still has app v1.4. |
Send message Joined: 14 May 14 Posts: 2 Credit: 3,945,126 RAC: 0 |
Yes, On both my machines CPU wu's are also resetting at about 60% |
Send message Joined: 24 Oct 16 Posts: 12 Credit: 56,127,036 RAC: 0 |
OOOPs.....similar here Werkstatt. 1st GPU bundle5 unit jumped back to 0% when it reached ~40% complete. CPU units are still crunching through to completion. Intel(R) Core(TM) i7-2700K CPU @ 3.50GHz [Family 6 Model 42 Stepping 7] (8 processors) AMD Radeon HD 6900 series (Cayman) (2048MB) driver: 1.4.1848 OpenCL: 1.2 Microsoft Windows 7 Pro x64 Service Pack 1, (06.01.7601.00) |
Send message Joined: 30 Apr 14 Posts: 67 Credit: 160,674,488 RAC: 0 |
Looks like Milkyway@Home 1.42 (opencl_ati_101) app is CPU only app. No GPU usage (at all). 4 WUs have had gone back to 0% around 77-78%. (Copied from slots of one of WU after it went back to 0%)
|
Send message Joined: 24 Jan 11 Posts: 715 Credit: 555,881,294 RAC: 42,655 |
My client doesn't like something about the downloaded 1.42 file. Had the wrong size error when it automatically downloaded it. Had the same error when I manually downloaded it from the directory. 11/11/2016 3:03:52 PM | Milkyway@Home | Resetting file projects/milkyway.cs.rpi.edu_milkyway/milkyway_1.42_windows_x86_64__opencl_nvidia_101.exe: wrong size 11/11/2016 3:03:52 PM | Milkyway@Home | Fetching scheduler list 11/11/2016 3:03:54 PM | Milkyway@Home | Started download of milkyway_1.42_windows_x86_64__opencl_nvidia_101.exe 11/11/2016 3:03:55 PM | Milkyway@Home | Master file download succeeded 11/11/2016 3:03:59 PM | Milkyway@Home | Finished download of milkyway_1.42_windows_x86_64__opencl_nvidia_101.exe 11/11/2016 3:03:59 PM | Milkyway@Home | [error] File milkyway_1.42_windows_x86_64__opencl_nvidia_101.exe has wrong size: expected 1292288, got 1316864 11/11/2016 3:03:59 PM | Milkyway@Home | [error] Checksum or signature error for milkyway_1.42_windows_x86_64__opencl_nvidia_101.exe 11/11/2016 3:04:00 PM | Milkyway@Home | Sending scheduler request: To report completed tasks. 11/11/2016 3:04:00 PM | Milkyway@Home | Reporting 102 completed tasks Anybody else having issues trying to run the new work units with the supplied 1.42 Nvidia app? |
Send message Joined: 22 Jan 08 Posts: 29 Credit: 242,730,423 RAC: 0 |
All tasks with version <= 1.41 produce errors, only. Engagig on v1.42 tasks results in immediate GPU core clock reduction to 300 Mhz and GPU RAM clock to 150 MHz on both 280X and 290X AMD graphics boards. When reaching 100% of estimated run time, the bundle tasks on my 290X reset to zero progress and appear to restart although the total runtime duration is not reset (I guess from the file name that it might be a bundle of 5, so this will hopefully repeat 5-fold and then upload). By contrast, the constraints tasks do complete and upload in the expected manner on my 280X. Runtime with reduced core and RAM clock is 741.63 secs as opposed to 9 secs with standard clocks. The result file first ends up in the "inconclusive" bunch - as is usual. So, there is still something wrong with both of these tasks types with respect to clock resetting. I checked with Einstein: Once a new Einstein WU starts after having finished a MW one, the core and RAM clocks go up to regular speed. So, the down clocking is conducted by MW client. Please inform us once you have solved the clocking issue. Michael. President of Rechenkraft.net e.V. - This planet's first and largest distributed computing organization. |
Send message Joined: 18 Aug 09 Posts: 123 Credit: 21,151,470 RAC: 2,217 |
Just cleared out 66 V1.40 tasks all with failures. Also just download a series of V1.42 tasks now, see how they hold up. |
Send message Joined: 4 Feb 11 Posts: 86 Credit: 60,913,150 RAC: 0 |
Work units that are generated with the old version of MilkyWay@home that need additional processing are causing errors. See https://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=1386764939 for an example. It was generated for the old version of MilkyWay@home, and it needs additional runs for validation. Either the new version needs to be able to deal with results from old work units, or the new version should be listed as a new application so that old work units continue to be processed with the old version. |
Send message Joined: 24 Oct 16 Posts: 12 Credit: 56,127,036 RAC: 0 |
1 GPU bundle5 WU made the trip. The 1st GPU bundle5 unit was "ready to report" and reported after 1:17:09 and several progress resets to 0% at ~40%. The 2nd GPU bundle5 progress reset to 0% @ 38% with 15 min. elapsed. Will see if it finishes. ~15 minutes X 5 = ~1 hour and 15 minutes.....is the progress simply resetting to 0% after each unit in the bundle of 5 units is completed? Intel(R) Core(TM) i7-2700K CPU @ 3.50GHz [Family 6 Model 42 Stepping 7] (8 processors) AMD Radeon HD 6900 series (Cayman) (2048MB) driver: 1.4.1848 OpenCL: 1.2 Microsoft Windows 7 Pro x64 Service Pack 1, (06.01.7601.00) |
Send message Joined: 30 Dec 08 Posts: 30 Credit: 6,999,702 RAC: 0 |
It definitely is a downclocking issue. Work units reset at appx 50%. Aborted 76 work units and sent back. Will wait until issue is resolved before getting anymore. |
Send message Joined: 24 Oct 16 Posts: 12 Credit: 56,127,036 RAC: 0 |
My 2nd GPU bundle5 unit "ready to report" and reported....no error messages for both. Like the 1st GPU bundle5 unit, the total elapsed time was 1 hour 17 minutes. The unit progress returned to 0% @ 40% after ~15 minutes 4 times (the % before returning to 0% increased each time up to 58%). After 4 resets (elapsed time 1hr 2 minutes), the progress % continued up to 100%...."ready to report" & reported. Looks like a reset of the progress indicator takes place after completion of each of the 1st 4 units in a bundle....then the indicator shows a % for the entire bundle5 unit. Prior to the update and bundle5 units, each GPU unit would complete in 19 to 33 seconds....on this 'puter.... S/B ~ 2 min 30 sec for 5 old units. Are the individual units in the new bundle of 5 units bigger than the old units? Intel(R) Core(TM) i7-2700K CPU @ 3.50GHz [Family 6 Model 42 Stepping 7] (8 processors) AMD Radeon HD 6900 series (Cayman) (2048MB) driver: 1.4.1848 OpenCL: 1.2 Microsoft Windows 7 Pro x64 Service Pack 1, (06.01.7601.00) |
©2024 Astroinformatics Group