Posts by Dirk Broer
log in
1) Message boards : Number crunching : Sudden credit loss (Message 6440)
Posted 19 May 2020 by Dirk Broer
Was it my imagination or did I lose 200 credit?
I can't access Free-DC from China, so could anyone check?


Only two members of team AMD Users lost credit (in the FreeDC view), Paul_AMD and me.
2) Message boards : Number crunching : Sudden credit loss (Message 6436)
Posted 18 May 2020 by Dirk Broer
I lost more than 100k in credit (probably around 130k just for TF) since yesterday.

Somewhat less:
TF -100,500 Sierpinski / Riesel Base - long -11,700
3) Message boards : Number crunching : Sudden credit loss (Message 6435)
Posted 18 May 2020 by Dirk Broer
I have lost 421000 of TF credit in the crash.


-360.500, according to this page
4) Message boards : Number crunching : Sudden credit loss (Message 6434)
Posted 18 May 2020 by Dirk Broer
My own loss was slight, compared to others
Sierpinski/Riesel Base -650 Sierpinski Base - short -840 Riesel Base - short -1,200 Sierpinski Base -1,960
5) Message boards : Number crunching : Sudden credit loss (Message 6433)
Posted 18 May 2020 by Dirk Broer
I crunched only TF recently and credit for this subproject was lost. Total credit before crash was around 1,950,000. I can't be more specific. Sorry. And thanks in advance.



Your TF loss was -213,500
6) Message boards : Number crunching : Trial Factoring (Message 6045)
Posted 7 Apr 2020 by Dirk Broer
with ubuntu 20.04 LTS the nvidia-cuda-toolkit 10.1 package is present in the repositories


Ubuntu 20.04 LTS is still in Beta, the official release date is April 23
7) Message boards : Number crunching : Trial Factoring (Message 6037)
Posted 6 Apr 2020 by Dirk Broer
Just copy the file libcudart.so.10.1 into your /lib64 folder (or whatever your distribution needs)


Last time 'just copy this to that' worked for me was under DOS. This is not DOS, this is Linux. libcudart.so.10.1 is totally unknown yet in the repo that supplies my software.


It goes even further. Under Linux, libcudart.so.10.1 is NOT part of the official CUDA 10.1 toolkit, not even of the official CUDA 10.2 toolkit. Both come with libcudart.so.9.1. You can only get libcudart.so.10.1 by adding an experimental PPA (Mageia) to your softwarelist. As libcudart.so.10.1 also has (inter)dependencies, merely copying the file in a directory will not be enough -but I wrote that earlier.
8) Message boards : Number crunching : Trial Factoring (Message 5975)
Posted 31 Mar 2020 by Dirk Broer
Just copy the file libcudart.so.10.1 into your /lib64 folder (or whatever your distribution needs)


Last time 'just copy this to that' worked for me was under DOS. This is not DOS, this is Linux. libcudart.so.10.1 is totally unknown yet in the repo that supplies my software.
9) Message boards : Number crunching : Trial Factoring (Message 5922)
Posted 29 Mar 2020 by Dirk Broer
KEP, every info is welcome, we need more users which running into these issues and want to help to test. Lets see how we can fix this, linux opencl has a 100% failure rate and mac is using the same config. If we can fix linux then mac too.


Something tells me that you tried to port a Windows CUDA app with a bare minimum of re-compiling. Try to make the developer start from OpenCL on a Linux system (preferably with a Radeon card) for the next OpenCL version.
10) Message boards : Number crunching : Trial Factoring (Message 5897)
Posted 28 Mar 2020 by Dirk Broer
Funny thing is that the system with the A10-5700 has three HD6670 cards, and it own IGP, and crunches four of the GPU72 WUs at the same time.

I get the idea that in Radeon-coding the option '--device 0' just doesn't work, not for Linux and not for Windows.


-d 0


That is a code/compiler thing, not something I can change


you can use an app_config file but try in standalone.


I have no problems with running four WUs at a time, I have problems with running each WU for more than 12 hours, getting a few meagre number of credits, while the same GPUs bring me 300k credits a day on e.g. Moo! and double that on Collatz.
If I want it to run on just one GPU I can use
<cc_config> <log_flags> </log_flags> <options> <use_all_gpus>0</use_all_gpus> </options> <cc_config>
, but that's not why I have three GPUs next to my APU
11) Message boards : Number crunching : Trial Factoring (Message 5895)
Posted 28 Mar 2020 by Dirk Broer
Funny thing is that the system with the A10-5700 has three HD6670 cards, and it own IGP, and crunches four of the GPU72 WUs at the same time.

I get the idea that in Radeon-coding the option '--device 0' just doesn't work, not for Linux and not for Windows.


-d 0


That is a code/compiler thing, not something I can change
12) Message boards : Number crunching : Trial Factoring (Message 5893)
Posted 28 Mar 2020 by Dirk Broer
Funny thing is that the system with the A10-5700 has three HD6670 cards, and it own IGP, and crunches four of the GPU72 WUs at the same time.

Those four WUs will have been running on just 1 of the GPUs.


No they didn't, all four AMD GPUs were busy at the same time. Normally they run four Moo!, Seti, Seti Beta or Collatz WUs, or any combination of these.
The view in BOINC Tasks was just the same when running SRBase GPU72, all GPUs running.
13) Message boards : Number crunching : Trial Factoring (Message 5889)
Posted 28 Mar 2020 by Dirk Broer
Funny thing is that the system with the A10-5700 has three HD6670 cards, and it own IGP, and crunches four of the GPU72 WUs at the same time.

I get the idea that in Radeon-coding the option '--device 0' just doesn't work, not for Linux and not for Windows.
14) Message boards : Number crunching : Trial Factoring (Message 5887)
Posted 28 Mar 2020 by Dirk Broer
GPU72 tasks for Ati on Windows keeps running forever for HD 6670, Radeon HD 7660D 'Devastator' IGP of AMD A10-5700 and HD 8400/R3 IGP of Athlon 5350.


check your stderr.txt in slot folder. The cards could be very slow.


I've aborted the lot when they weren't finished after 10 hours, 9 hours of which they were at 100% completed.
http://srbase.my-firewall.org/sr5/results.php?hostid=996
http://srbase.my-firewall.org/sr5/results.php?hostid=2148
15) Message boards : Number crunching : Trial Factoring (Message 5885)
Posted 28 Mar 2020 by Dirk Broer
GPU72 tasks for Ati on Windows keeps running forever for HD 6670, Radeon HD 7660D 'Devastator' IGP of AMD A10-5700 and HD 8400/R3 IGP of Athlon 5350.
16) Message boards : Number crunching : Trial Factoring (Message 5877)
Posted 28 Mar 2020 by Dirk Broer
That's not something I can choose, it is an error that the code comes up with


What card do you have?


Error comes up using various AMD APUs (Athlon 5350 Radeon R3, Athlon A12-9800E Radeon R7)


Can you pls post also the opencl line from the BOINC event manager so we could forward this to the developer of the app.


For the Athlon A12-9800E:
28-3-2020 01:15:45 OpenCL: AMD/ATI GPU 0: AMD CARRIZO (DRM 3.33.0, 5.3.0-42-generic, LLVM 9.0.1) (driver version 20.1.0-devel, device version OpenCL 1.1 Mesa 20.1.0-devel (git-aad0e6f 2020-03-27 bionic-oibaf-ppa), 3072MB, 3072MB available, 576 GFLOPS peak)
28-3-2020 01:15:45 OpenCL CPU: pthread-AMD A12-9800E RADEON R7, 12 COMPUTE CORES 4C+8G (OpenCL driver vendor: The pocl project, driver version 1.1, device version OpenCL 1.2 pocl HSTR: pthread-x86_64-pc-linux-gnu-bdver4)
17) Message boards : Number crunching : Trial Factoring (Message 5876)
Posted 27 Mar 2020 by Dirk Broer
That's not something I can choose, it is an error that the code comes up with


What card do you have?


Error comes up using various AMD APUs (Athlon 5350 Radeon R3, Athlon A12-9800E Radeon R7)


Can you pls post also the opencl line from the BOINC event manager so we could forward this to the developer of the app.


For the Athlon 5350:
14-8-2019 11:23:24 OpenCL: AMD/ATI GPU 0: AMD Radeon Graphics (driver version 2671.3, device version OpenCL 1.2 AMD-APP (2671.3), 2553MB, 2553MB available, 154 GFLOPS peak)
14-8-2019 11:23:24 OpenCL: AMD/ATI GPU 1: AMD KABINI (DRM 3.27.0, 5.0.0-25-generic, LLVM 8.0.0) (driver version 19.0.8, device version OpenCL 1.1 Mesa 19.0.8, 3072MB, 3072MB available, 96 GFLOPS peak)
18) Message boards : Number crunching : Trial Factoring (Message 5874)
Posted 27 Mar 2020 by Dirk Broer
Nvidia on Linux

<message>
process exited with code 195 (0xc3, -61)</message>
<stderr_txt>
22:54:06 (28652): wrapper (7.2.26012): starting
22:54:06 (28652): wrapper: running ./mfaktc.exe ( --device 0)
./mfaktc.exe: error while loading shared libraries: libcudart.so.10.1: cannot open shared object file: No such file or directory
22:54:07 (28652): ./mfaktc.exe exited; CPU time 0.001267
22:54:07 (28652): app exit status: 0x7f00
22:54:07 (28652): called boinc_finish
</stderr_txt>

This code tries running an .exe on Linux! Can be done under WINE (emulator), but not native.


Please see the FAQ, you're missing the CUDA 10.1 toolkit and you need this specific version in order for it to work.


I'll upgrade the system to Ubuntu 19.10 and install the latest Nvidia, OpenCL and CUDA files
19) Message boards : Number crunching : Trial Factoring (Message 5871)
Posted 27 Mar 2020 by Dirk Broer
That's not something I can choose, it is an error that the code comes up with


What card do you have?


Error comes up using various AMD APUs (Athlon 5350 Radeon R3, Athlon A12-9800E Radeon R7)
20) Message boards : Number crunching : Trial Factoring (Message 5868)
Posted 27 Mar 2020 by Dirk Broer
Nvidia on Linux

<message>
process exited with code 195 (0xc3, -61)</message>
<stderr_txt>
22:54:06 (28652): wrapper (7.2.26012): starting
22:54:06 (28652): wrapper: running ./mfaktc.exe ( --device 0)
./mfaktc.exe: error while loading shared libraries: libcudart.so.10.1: cannot open shared object file: No such file or directory
22:54:07 (28652): ./mfaktc.exe exited; CPU time 0.001267
22:54:07 (28652): app exit status: 0x7f00
22:54:07 (28652): called boinc_finish
</stderr_txt>

This code tries running an .exe on Linux! Can be done under WINE (emulator), but not native.


Next 20

Main page · Your account · Message boards


Copyright © 2014-2020 BOINC Confederation / rebirther