"Must update drivers"
log in

Advanced search

Message boards : Number crunching : "Must update drivers"

Post to thread Subscribe Sort
1 · 2 · Next
Author Message
pls
Send message
Joined: 18 Jul 20
Posts: 6
Credit: 27,307,980
RAC: 0
Message 9869 - Posted: 28 Mar 2024, 20:02:49 UTC

I am no longer receiving any GPU tasks for SRBase. I am getting a message in the debug log

2024-03-28 11:54:39 | SRBase | Message from server: NVIDIA GPU: Upgrade to the latest driver to process tasks using your computer's GPU

I have the latest drivers from the NVIDIA web site. What exactly are you looking for?

Windows 10
Quadro K2000
Driver 30.0.14.7482

Thanks

ID: 9869 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7709
Credit: 44,264,218
RAC: 0
Message 9870 - Posted: 28 Mar 2024, 20:24:22 UTC - in response to Message 9869.
Last modified: 28 Mar 2024, 20:25:37 UTC

I am no longer receiving any GPU tasks for SRBase. I am getting a message in the debug log

2024-03-28 11:54:39 | SRBase | Message from server: NVIDIA GPU: Upgrade to the latest driver to process tasks using your computer's GPU

I have the latest drivers from the NVIDIA web site. What exactly are you looking for?

Windows 10
Quadro K2000
Driver 30.0.14.7482

Thanks


What cuda version do you have installed? cc6.0+ is supported, looks like your card is too old. I see you have some badges. When did you get the last WUs?

ID: 9870 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile vaughan
Send message
Joined: 9 Dec 14
Posts: 90
Credit: 4,184,397,515
RAC: 13,401,752
Message 9871 - Posted: 28 Mar 2024, 22:53:44 UTC - in response to Message 9870.

... looks like your card is too old.


What is "too old", what GPUs are supported?
____________

ID: 9871 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7709
Credit: 44,264,218
RAC: 0
Message 9873 - Posted: 29 Mar 2024, 0:03:34 UTC - in response to Message 9871.

... looks like your card is too old.


What is "too old", what GPUs are supported?


I have found some specs by google with cc5 if this is right.

ID: 9873 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
pls
Send message
Joined: 18 Jul 20
Posts: 6
Credit: 27,307,980
RAC: 0
Message 9875 - Posted: 29 Mar 2024, 4:18:33 UTC - in response to Message 9870.

These are the relevant lines from the BOINC event log. I think you're right, you've "upgraded" the SRBASE software past the capabilities of my video card.

2024-03-28 03:04:37 | | CUDA: NVIDIA GPU 0: Quadro K2000 (driver version 474.82, CUDA version 11.4, compute capability 3.0, 2048MB, 2048MB available, 100 GFLOPS peak)
2024-03-28 03:04:37 | | OpenCL: NVIDIA GPU 0: Quadro K2000 (driver version 474.82, device version OpenCL 3.0 CUDA, 2048MB, 2048MB available, 100 GFLOPS peak)

ID: 9875 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7709
Credit: 44,264,218
RAC: 0
Message 9877 - Posted: 29 Mar 2024, 8:58:17 UTC - in response to Message 9875.
Last modified: 29 Mar 2024, 8:58:32 UTC

These are the relevant lines from the BOINC event log. I think you're right, you've "upgraded" the SRBASE software past the capabilities of my video card.

2024-03-28 03:04:37 | | CUDA: NVIDIA GPU 0: Quadro K2000 (driver version 474.82, CUDA version 11.4, compute capability 3.0, 2048MB, 2048MB available, 100 GFLOPS peak)
2024-03-28 03:04:37 | | OpenCL: NVIDIA GPU 0: Quadro K2000 (driver version 474.82, device version OpenCL 3.0 CUDA, 2048MB, 2048MB available, 100 GFLOPS peak)


Its a Kepler card with cc3.0, I have changed to cc3.0 in plan_class for cuda10. If its not working for some cards I will change it back.

ID: 9877 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
pls
Send message
Joined: 18 Jul 20
Posts: 6
Credit: 27,307,980
RAC: 0
Message 9878 - Posted: 31 Mar 2024, 11:41:03 UTC - in response to Message 9877.

Thank you for that. I had to reset the project to get new work. So far, 3 tasks have completed and validated.

ID: 9878 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Doug
Send message
Joined: 18 Jul 17
Posts: 11
Credit: 466,051,712
RAC: 2,539,282
Message 10527 - Posted: 16 Mar 2025, 19:04:07 UTC - in response to Message 9877.

@rebirther, do you know if this was changed back? I'm adding some old K80 cards and get the same error:

2025-03-16T19:04:37.768554+00:00 vm01 boinc[11654]: 16-Mar-2025 19:04:37 [SRBase] NVIDIA GPU: Upgrade to the latest driver to process tasks using your computer's GPU


Here are the lines from my log with the CUDA version/etc..:

2025-03-16T18:56:28.815046+00:00 vm01 boinc[11654]: 16-Mar-2025 18:56:28 [---] CUDA: NVIDIA GPU 0: Tesla K80 (driver version 470.99, CUDA version 11.4, compute capability 3.7, 11441MB, 11441MB available, 4111 GFLOPS peak) 2025-03-16T18:56:28.815357+00:00 vm01 boinc[11654]: 16-Mar-2025 18:56:28 [---] CUDA: NVIDIA GPU 1: Tesla K80 (driver version 470.99, CUDA version 11.4, compute capability 3.7, 11441MB, 11441MB available, 4111 GFLOPS peak)

ID: 10527 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7709
Credit: 44,264,218
RAC: 0
Message 10528 - Posted: 16 Mar 2025, 19:29:08 UTC - in response to Message 10527.

Do you have updated to cuda 11? Can you use cuda 10?

ID: 10528 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Doug
Send message
Joined: 18 Jul 17
Posts: 11
Credit: 466,051,712
RAC: 2,539,282
Message 10529 - Posted: 16 Mar 2025, 20:15:03 UTC - in response to Message 10528.

So far I haven't had any luck downgrading to cuda 10 using the 450.66 driver (based on https://docs.nvidia.com/deploy/cuda-compatibility/index.html)

I'm running ubuntu 24 on this node so I'll try and downgrade to Ubuntu 22 or 20 to see if that allows me to install the cuda 10 driver.

ID: 10529 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7709
Credit: 44,264,218
RAC: 0
Message 10530 - Posted: 16 Mar 2025, 20:18:30 UTC - in response to Message 10529.

As info from google K80 is not running with cuda11, unfortunately.

ID: 10530 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Doug
Send message
Joined: 18 Jul 17
Posts: 11
Credit: 466,051,712
RAC: 2,539,282
Message 10531 - Posted: 17 Mar 2025, 4:39:18 UTC - in response to Message 10530.

Hmm where do you see that? K80s are compute capability 3.7 so I thought they are supported up to CUDA 11.8, just deprecated:

https://docs.nvidia.com/cuda/archive/11.8.0/cuda-toolkit-release-notes/index.html#deprecated-features

I finally got a CUDA 10 driver installed with an Ubuntu 20 machine:

$ nvidia-smi Mon Mar 17 04:34:33 2025 +-----------------------------------------------------------------------------+ | NVIDIA-SMI 440.100 Driver Version: 440.100 CUDA Version: 10.2 | |-------------------------------+----------------------+----------------------+


But I still see the same error in the logs, although it seems like the project downloads the files:

Mar 17 04:29:17 vm01 boinc[46003]: 17-Mar-2025 04:29:17 Initialization completed Mar 17 04:29:17 vm01 boinc[46003]: 17-Mar-2025 04:29:17 [SRBase] Sending scheduler request: To fetch work. Mar 17 04:29:17 vm01 boinc[46003]: 17-Mar-2025 04:29:17 [SRBase] Requesting new tasks for NVIDIA GPU Mar 17 04:29:19 vm01 boinc[46003]: 17-Mar-2025 04:29:19 [SRBase] Scheduler request completed: got 47 new tasks Mar 17 04:29:19 vm01 boinc[46003]: 17-Mar-2025 04:29:19 [SRBase] NVIDIA GPU: Upgrade to the latest driver to process tasks using your computer's GPU Mar 17 04:29:19 vm01 boinc[46003]: 17-Mar-2025 04:29:19 [SRBase] Project requested delay of 7 seconds Mar 17 04:29:21 vm01 boinc[46003]: 17-Mar-2025 04:29:21 [SRBase] Started download of wrapper_26018_linux_x86-64 Mar 17 04:29:21 vm01 boinc[46003]: 17-Mar-2025 04:29:21 [SRBase] Started download of mfaktc-linux64-cuda100v1.zip Mar 17 04:29:24 vm01 boinc[46003]: 17-Mar-2025 04:29:24 [SRBase] Finished download of mfaktc-linux64-cuda100v1.zip Mar 17 04:29:24 vm01 boinc[46003]: 17-Mar-2025 04:29:24 [SRBase] Started download of job_TF_l64c1_00043.xml Mar 17 04:29:25 vm01 boinc[46003]: 17-Mar-2025 04:29:25 [SRBase] Finished download of job_TF_l64c1_00043.xml .... Mar 17 04:29:32 vm01 boinc[46003]: 17-Mar-2025 04:29:32 [SRBase] Started download of worktodo13a382_0077223.txt Mar 17 04:29:32 vm01 boinc[46003]: 17-Mar-2025 04:29:32 [SRBase] Computation for task TF_75-76_272-277M_wu_78821_0 finished Mar 17 04:29:32 vm01 boinc[46003]: 17-Mar-2025 04:29:32 [SRBase] Output file TF_75-76_272-277M_wu_78821_0_0 for task TF_75-76_272-277M_wu_78821_0 absent Mar 17 04:29:32 vm01 boinc[46003]: 17-Mar-2025 04:29:32 [SRBase] Starting task TF_75-76_272-277M_wu_77377_1 Mar 17 04:29:32 vm01 boinc[46003]: 17-Mar-2025 04:29:32 [SRBase] Starting task TF_75-76_272-277M_wu_78854_0 Mar 17 04:29:32 vm01 boinc[46003]: 17-Mar-2025 04:29:32 [SRBase] Scheduler request completed: got 47 new tasks Mar 17 04:29:32 vm01 boinc[46003]: 17-Mar-2025 04:29:32 [SRBase] NVIDIA GPU: Upgrade to the latest driver to process tasks using your computer's GPU Mar 17 04:29:32 vm01 boinc[46003]: 17-Mar-2025 04:29:32 [SRBase] Project requested delay of 7 seconds Mar 17 04:29:33 vm01 boinc[46324]: mv: cannot stat 'slots/2/results.json.txt': No such file or directory Mar 17 04:29:33 vm01 boinc[46003]: 17-Mar-2025 04:29:33 [SRBase] Computation for task TF_75-76_272-277M_wu_78853_0 finished Mar 17 04:29:33 vm01 boinc[46003]: 17-Mar-2025 04:29:33 [SRBase] Output file TF_75-76_272-277M_wu_78853_0_0 for task TF_75-76_272-277M_wu_78853_0 absent


On a side note, I'm running einstien at home with other K80s with the same 470.256.02 driver without a problem; looking at their projects directory, they run with libcudart.so.10.2.

If it's too much effort/doesn't make sense to get K80s to work with SRBase, I can just allocate this machine to something else.

ID: 10531 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Doug
Send message
Joined: 18 Jul 17
Posts: 11
Credit: 466,051,712
RAC: 2,539,282
Message 10532 - Posted: 17 Mar 2025, 4:42:42 UTC - in response to Message 10531.

Also, it looks like the jobs are failing because they are looking for libcudart.so.10.1 but libcudart.so.10.0 is deployed:


# cat /var/lib/boinc/slots/1/stderr.txt 2025-03-17 04:29:49 (46400): wrapper (7.24.26018): starting 2025-03-17 04:29:49 (46400): wrapper (7.24.26018): starting 2025-03-17 04:29:49 (46400): wrapper: running ./mfaktc.exe (-d 1) 2025-03-17 04:29:49 (46400): wrapper: created child process 46404 ./mfaktc.exe: error while loading shared libraries: libcudart.so.10.1: cannot open shared object file: No such file or directory 2025-03-17 04:29:50 (46400): ./mfaktc.exe exited; CPU time 0.001167 2025-03-17 04:29:50 (46400): app exit status: 0x7f 2025-03-17 04:29:50 (46400): called boinc_finish(195)


# ll /var/lib/boinc/slots/1/ total 3276 drwxrwx--x 2 boinc boinc 4096 Mar 17 04:29 ./ drwxr-xr-x 6 boinc boinc 4096 Mar 17 04:29 ../ -rw-r--r-- 1 boinc boinc 35147 Oct 23 19:09 COPYING -rw-r--r-- 1 boinc boinc 19480 Oct 23 19:09 Changelog.txt -rw-r--r-- 1 boinc boinc 14134 Oct 23 19:09 README.txt -rw-r--r-- 1 boinc boinc 4 Mar 17 04:29 boinc_finish_called -rw-r--r-- 1 boinc boinc 8192 Mar 17 04:29 boinc_mmap_file -rw-r--r-- 1 boinc boinc 9046 Mar 17 04:29 init_data.xml -rw-r--r-- 1 boinc boinc 88 Mar 17 04:29 job.xml -rw-r--r-- 1 boinc boinc 495736 Dec 29 01:16 libcudart.so.10.0 -rwxrwxrwx 1 boinc boinc 2715824 Sep 4 2019 mfaktc.exe* -rw-r--r-- 1 boinc boinc 9103 Jan 9 10:01 mfaktc.ini -rw-r--r-- 1 boinc boinc 554 Mar 17 04:29 stderr.txt -rw-r--r-- 1 boinc boinc 57 Mar 17 04:29 worktodo.txt -rw-r--r-- 1 boinc boinc 92 Mar 17 04:29 wrapper_26018_linux_x86-64

ID: 10532 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7709
Credit: 44,264,218
RAC: 0
Message 10533 - Posted: 17 Mar 2025, 5:26:46 UTC - in response to Message 10532.
Last modified: 17 Mar 2025, 5:55:14 UTC

https://srbase.my-firewall.org/sr5/download/libcudart.so.10.1

Update:
I have asked deepseek and cuda <11.4 is supported. If you have Ubuntu24 still in line I can try to change the plan_class to support the driver.

Changed the driver to min 470.25 in plan_class.

ID: 10533 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Doug
Send message
Joined: 18 Jul 17
Posts: 11
Credit: 466,051,712
RAC: 2,539,282
Message 10537 - Posted: 19 Mar 2025, 17:24:02 UTC - in response to Message 10533.

I redeployed Ubuntu 24 and reinstalled the 470.256.02 driver but still seeing:

[SRBase] NVIDIA GPU: Upgrade to the latest driver to process tasks using your computer's GPU

vm01:~$ nvidia-smi Wed Mar 19 17:25:30 2025 +-----------------------------------------------------------------------------+ | NVIDIA-SMI 470.256.02 Driver Version: 470.256.02 CUDA Version: 11.4 | |-------------------------------+----------------------+----------------------+

ID: 10537 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7709
Credit: 44,264,218
RAC: 0
Message 10539 - Posted: 19 Mar 2025, 18:15:50 UTC

I see, we had 2 different plan_classes for cuda11, win and linux, I have changed one and set it to cc3.0 min, should be working now, pls try again...

ID: 10539 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Doug
Send message
Joined: 18 Jul 17
Posts: 11
Credit: 466,051,712
RAC: 2,539,282
Message 10540 - Posted: 20 Mar 2025, 3:46:01 UTC - in response to Message 10539.

Interestingly, it still has the "Upgrade driver" message but it proceeds and downloads files; but it looks like its getting an error in the tasks:

2025-03-20T03:41:34.322868+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:34 [SRBase] NVIDIA GPU: Upgrade to the latest driver to process tasks using your computer's GPU 2025-03-20T03:41:34.322947+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:34 [SRBase] Project requested delay of 7 seconds 2025-03-20T03:41:36.372613+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:36 [SRBase] Started download of worktodo13a383_0026378.txt 2025-03-20T03:41:36.372742+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:36 [SRBase] Started download of worktodo13a383_0026415.txt 2025-03-20T03:41:38.382452+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:38 [SRBase] Finished download of worktodo13a383_0026378.txt (57 bytes) 2025-03-20T03:41:38.382733+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:38 [SRBase] Finished download of worktodo13a383_0026415.txt (57 bytes) 2025-03-20T03:41:38.382798+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:38 [SRBase] Started download of worktodo13a383_0026416.txt 2025-03-20T03:41:38.382917+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:38 [SRBase] Started download of worktodo13a383_0026417.txt 2025-03-20T03:41:38.386071+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:38 [SRBase] Starting task TF_75-76_236-282M_wu_26415_0 2025-03-20T03:41:38.389121+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:38 [SRBase] Starting task TF_75-76_236-282M_wu_26378_0 2025-03-20T03:41:40.422707+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:40 [SRBase] Finished download of worktodo13a383_0026416.txt (57 bytes) 2025-03-20T03:41:40.422892+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:40 [SRBase] Finished download of worktodo13a383_0026417.txt (57 bytes) 2025-03-20T03:41:40.423007+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:40 [SRBase] Started download of worktodo13a383_0026418.txt 2025-03-20T03:41:40.423070+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:40 [SRBase] Started download of worktodo13a383_0026529.txt 2025-03-20T03:41:40.426282+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:40 [SRBase] Starting task TF_75-76_236-282M_wu_26416_0 2025-03-20T03:41:40.429498+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:40 [SRBase] Starting task TF_75-76_236-282M_wu_26417_0 2025-03-20T03:41:42.466372+00:00 vm01 boinc[47982]: mv: cannot stat 'slots/0/results.json.txt': No such file or directory 2025-03-20T03:41:42.469145+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:42 [SRBase] Finished download of worktodo13a383_0026418.txt (57 bytes) 2025-03-20T03:41:42.469362+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:42 [SRBase] Finished download of worktodo13a383_0026529.txt (57 bytes) 2025-03-20T03:41:42.469461+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:42 [SRBase] Started download of worktodo13a383_0026375.txt 2025-03-20T03:41:42.469552+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:42 [SRBase] Started download of worktodo13a383_0026377.txt 2025-03-20T03:41:42.469629+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:42 [SRBase] Computation for task TF_75-76_236-282M_wu_26415_0 finished 2025-03-20T03:41:42.469691+00:00 vm01 boinc[47533]: 20-Mar-2025 03:41:42 [SRBase] Output file TF_75-76_236-282M_wu_26415_0_0 for task TF_75-76_236-282M_wu_26415_0 absent


$ sudo cat /var/lib/boinc/slots/0/stderr.txt 2025-03-20 03:41:50 (48039): wrapper (7.24.26018): starting 2025-03-20 03:41:50 (48039): wrapper (7.24.26018): starting 2025-03-20 03:41:50 (48039): wrapper: running ./mfaktc.exe (-d 0) 2025-03-20 03:41:50 (48039): wrapper: created child process 48044 2025-03-20 03:41:52 (48039): ./mfaktc.exe exited; CPU time 0.697094 2025-03-20 03:41:52 (48039): app exit status: 0x1 2025-03-20 03:41:52 (48039): called boinc_finish(195) free(): invalid pointer SIGABRT: abort called Stack trace (22 frames): ../../projects/srbase.my-firewall.org_sr5/wrapper_26018_linux_x86-64(+0x41eac)[0x62d17fa41eac] /lib/x86_64-linux-gnu/libc.so.6(+0x45330)[0x74590ac45330] /lib/x86_64-linux-gnu/libc.so.6(pthread_kill+0x11c)[0x74590ac9eb2c] /lib/x86_64-linux-gnu/libc.so.6(gsignal+0x1e)[0x74590ac4527e] /lib/x86_64-linux-gnu/libc.so.6(abort+0xdf)[0x74590ac288ff] /lib/x86_64-linux-gnu/libc.so.6(+0x297b6)[0x74590ac297b6] /lib/x86_64-linux-gnu/libc.so.6(+0xa8ff5)[0x74590aca8ff5] /lib/x86_64-linux-gnu/libc.so.6(+0xab38c)[0x74590acab38c] /lib/x86_64-linux-gnu/libc.so.6(__libc_free+0x7e)[0x74590acaddae] ../../projects/srbase.my-firewall.org_sr5/wrapper_26018_linux_x86-64(+0x67daa)[0x62d17fa67daa] ../../projects/srbase.my-firewall.org_sr5/wrapper_26018_linux_x86-64(+0x87334)[0x62d17fa87334] /lib/x86_64-linux-gnu/libc.so.6(+0x45330)[0x74590ac45330] /lib/x86_64-linux-gnu/libc.so.6(clock_nanosleep+0xbf)[0x74590acecadf] /lib/x86_64-linux-gnu/libc.so.6(nanosleep+0x17)[0x74590acf9a27] /lib/x86_64-linux-gnu/libc.so.6(sleep+0x43)[0x74590ad0ec63] ../../projects/srbase.my-firewall.org_sr5/wrapper_26018_linux_x86-64(+0x592d4)[0x62d17fa592d4] ../../projects/srbase.my-firewall.org_sr5/wrapper_26018_linux_x86-64(+0x3b129)[0x62d17fa3b129] ../../projects/srbase.my-firewall.org_sr5/wrapper_26018_linux_x86-64(+0x3b1cd)[0x62d17fa3b1cd] ../../projects/srbase.my-firewall.org_sr5/wrapper_26018_linux_x86-64(+0x1e4df)[0x62d17fa1e4df] /lib/x86_64-linux-gnu/libc.so.6(+0x2a1ca)[0x74590ac2a1ca] /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0x8b)[0x74590ac2a28b] ../../projects/srbase.my-firewall.org_sr5/wrapper_26018_linux_x86-64(+0x1aa2a)[0x62d17fa1aa2a] Exiting...

ID: 10540 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7709
Credit: 44,264,218
RAC: 0
Message 10541 - Posted: 20 Mar 2025, 9:29:52 UTC
Last modified: 20 Mar 2025, 9:31:20 UTC

Can you try to unzip the content of mfaktc.zip in a separate folder and run
./mfaktc -st ?

Your cuda10 was working.

ID: 10541 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Doug
Send message
Joined: 18 Jul 17
Posts: 11
Credit: 466,051,712
RAC: 2,539,282
Message 10544 - Posted: 21 Mar 2025, 15:57:46 UTC - in response to Message 10541.

Sure, here's the output:

$ ./mfaktc.exe -st mfaktc v0.23.0 (64bit built) Compiletime options THREADS_PER_BLOCK 256 SIEVE_SIZE_LIMIT 32kiB SIEVE_SIZE 193154bits SIEVE_SPLIT 250 MORE_CLASSES enabled Runtime options SievePrimes 25000 SievePrimesAdjust 1 SievePrimesMin 5000 SievePrimesMax 100000 NumStreams 3 CPUStreams 3 GridSize 3 GPU Sieving enabled GPUSievePrimes 82486 GPUSieveSize 2047Mi bits GPUSieveProcessSize 16Ki bits Checkpoints enabled CheckpointDelay 60s WorkFileAddDelay disabled Stages enabled StopAfterFactor bitlevel PrintMode compact Logging disabled V5UserID (none) ComputerID (none) AllowSleep no TimeStampInResults no CUDA version info binary compiled for CUDA 11.20 CUDA runtime version 11.20 CUDA driver version 11.40 CUDA device info name Tesla K80 compute capability 3.7 max threads per block 1024 max shared memory per MP 114688 byte number of multiprocessors 13 CUDA cores per MP 192 CUDA cores - total 2496 clock rate (CUDA cores) 823MHz memory clock rate: 2505MHz memory bus width: 384 bit Automatic parameters threads per grid 851968 GPUSievePrimes (adjusted) 82486 GPUsieve minimum exponent 1055144 ########## testcase 1/2867 ########## Starting trial factoring M50804297 from 2^67 to 2^68 (0.59 GHz-days) Using GPU kernel "75bit_mul32_gs" Date Time | class Pct | time ETA | GHz-d/day Sieve Wait Mar 21 16:01 | 3387 0.1% | 0.001 n.a. | n.a. 82485 n.a.%ERROR: cudaGetLastError() returned 209: no kernel image is available for execution on the device

ID: 10544 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7709
Credit: 44,264,218
RAC: 0
Message 10545 - Posted: 21 Mar 2025, 18:19:03 UTC
Last modified: 21 Mar 2025, 19:17:19 UTC

Thx for the test, it has been confirmed that the app was not correct compiled with cc3.5, we will try to get another fix for this.

ID: 10545 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
1 · 2 · Next
Post to thread

Message boards : Number crunching : "Must update drivers"

Main page · Your account · Message boards


Copyright © 2014-2025 BOINC Confederation / rebirther