Posts by Georgi Vidinski
|
41)
(Message 7960)
Posted 16 Sep 2023 by 
Georgi Vidinski
Post: Thanks for the support guys! I'll keep you all in the loop. I just need to find some spare time. Other than that, only sky is the limit. Cheers! Georgi |
|
42)
(Message 7959)
Posted 16 Sep 2023 by
Georgi Vidinski
Post: I never imagined that computers could help with scientific analysis You're kidding, right? Why do you think they were invented on first place? ;) In the pre-silicon-computer era "computer" was a person who does compute. Cheers! |
|
43)
(Message 7956)
Posted 16 Sep 2023 by
Georgi Vidinski
Post: Check this out. This could put some enlightenment to the issue with amdgpu: The CS has been rejected (-125), but the context isn't robust. amdgpu: The process will be terminated. Amdgpu timeout with OpenCl kernel (reproducible) Random GPU crashes amdgpu across 2 different GPUs Send your feed back on your progress with that. |
|
44)
(Message 7955)
Posted 16 Sep 2023 by
Georgi Vidinski
Post: It's not exactly that. The linux app is OpenCL 1.1 compliant (in respect to the Windows app, which is OpenCL 1.2). I've tested it on my Arch Linux system using clover-mesa drivers and it was working as a charm. Well, as you remember I was using latest libraries for everything, especially I had to built the clover-mesa from the latest source in their gitlab branch. The application does not use any Image nor Texture manipulation functions, wich are the Achilles heel in most of the cases with Clover. But as I mentioned previously, integrated graphics needs special approach and will need additional research:
As your graphics are GCN5 you definitely have to move towards ROCm. Clover will be retired in the next few months: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19385 |
|
45)
(Message 7951)
Posted 15 Sep 2023 by
Georgi Vidinski
Post: Latest versions has been released. |
|
46)
(Message 7950)
Posted 15 Sep 2023 by
Georgi Vidinski
Post: Be patient guys. As you can see from here https://asteroidsathome.net/boinc/apps.php those apps are not released yet. Kyong will not be available till later today. |
|
47)
(Message 7943)
Posted 15 Sep 2023 by
Georgi Vidinski
Post: I see. The executable is definitely the wrong one. I'll talk to Kyong. Will keep you posted. |
|
48)
(Message 7940)
Posted 15 Sep 2023 by
Georgi Vidinski
Post: However it fails to run on R9 280X (OpenCL 1.2). You may try again on this one as the final app v102.17 is a different build that the previous buggy (Version: 102.18.1.0) one. You should be able to see Version: 102.18.5.0in Stderr output of the tasks. Let me know how was it. |
|
49)
(Message 7938)
Posted 14 Sep 2023 by
Georgi Vidinski
Post: Brief update: Windows - AMD OpenCL app v102.17 seams to be the most stable of all the others. Keep sending your feedback. It may help us to polish issues (or at least some of them) within the next build. Yet, the app may not work on some GPUs. Keep on mind that OpenCL apps can be highly affected by the drivers in use and there is no universal cure (version, vendor, etc.). Linux - AMD OpenCL app v102.17 is ready. Waiting for Kyong to release it. Here are some details for the system on which it was built: Ubuntu 20.04.1 LTS Release: 20.04 ldd (Ubuntu GLIBC 2.31-0ubuntu9) 2.31 I hope this will do, as I can't go lower with versions. And again, keep sending your feedback please. By the way it will be interesting to know if the app woks on Mesa platforms. As to the other questions Have you deliberately set it to use about half the GPU RAM? No. The amount of allocated memory will be different for every GPU model. Do you use something like this to run CUDA code on AMD GPUs? No. The code was written from scratch. Cheers! Georgi |
|
50)
(Message 7925)
Posted 11 Sep 2023 by
Georgi Vidinski
Post: New build with fixes is on its way. But I still have to run some tests on it first. It still may have issues with integrated AMD Graphics though. Those CPU based Graphics needs different memory alignment. Unfortunately there is now way for us to distinct them at project level from discrete GPUs. So, those of you who have such systems may want to restrict their use for now using cc_congif.xml and <exclude_gpu>tag (Client configuration). At least until we handle their specs through the code. Georgi |
|
51)
(Message 7917)
Posted 10 Sep 2023 by
Georgi Vidinski
Post: Georgi, Ian&Steve C. you're right, I was working on Arch Linux 6.1.49-1-MANJARO. And yes, except the GCC, wich I had to downgrade to v12 everything else was in final versions (GLIBC 2.38 include). And I did't realized there could be a dependency issues. Well, as I love to say - it's a constant learning curve, and thank you for pointing that out! I'll see what I can do and will recompile the lunux app. I'll change the Grid dimto dynamic value from clGetDeviceInfo()as well. |
|
52)
(Message 7908)
Posted 10 Sep 2023 by
Georgi Vidinski
Post: Hi Samuel, Thank you for your feedback! Your GLIBC version is a bit older indeed glibc-2.35.tar.gz 2022-02-03 01:35. You may want to upgrade all of your packages by running: sudo apt update && sudo apt upgrade -y Keep me posted, Georgi [/code] |
|
53)
(Message 7906)
Posted 10 Sep 2023 by
Georgi Vidinski
Post: Well, Mr P Hucker, Can you leave those machines of yours with AMD GPUs to do some more work, so we could gather more logs? If you want to help of course. Something like for a day will be enough. Same for others. Guys, Any feedback is welcome. I saw already that the app is running slow, with very small Grid dim. That was intended. Thanks, Georgi |
|
54)
(Message 7902)
Posted 9 Sep 2023 by
Georgi Vidinski
Post: Yes, I changed it to one month, this shows in the top left graph the last addition to work was much larger. I was wondering if he had another data source, so in future the work will not run out. I'm not aware of new observatory been joined to the data gathering process. It's more likely one of them just have sent bigger patch of data, but can't say for sure. |
|
55)
(Message 7898)
Posted 9 Sep 2023 by
Georgi Vidinski
Post: Well, I still believe using capital letters is just a way to say that out loud and nothing more. And they are always been heard. As we are always aware of the fact that there is a gap again. When you do science you have to be patient and you can always expect unexpected. ;) Cheers! Georgi |
|
56)
(Message 7897)
Posted 9 Sep 2023 by
Georgi Vidinski
Post: Hi Geronimo, I really appreciate your feedback. As I said, during the time all other applications was updated to the requirements added by the new WUs structure, exceeding the old limit of 1000 light points. Unfortunately the team wasn't able to react quickly enough and prepare new macOS apps. We'll do our best to fix that ASAP (in relative terms). May I contact you later for this matter, if it is OK with you of course? Georgi |
|
57)
(Message 7896)
Posted 9 Sep 2023 by
Georgi Vidinski
Post: Alpha version of the OpenCL app for AMD GPUs is on its way to be released. We'll appreciate any feedback that will help us to try to polish rough edges of the code! Cheers! Georgi |
|
58)
(Message 7893)
Posted 9 Sep 2023 by
Georgi Vidinski
Post: Who is upset? No one looks upset ;) |
|
59)
(Message 7892)
Posted 9 Sep 2023 by
Georgi Vidinski
Post: Hi, I presume you have already found a solution? Just out of curiosity what was causing it? Of course it should have nothing with our application and looks to me to have nothing with neither of BOINC apps as well. Georgi |
|
60)
(Message 7891)
Posted 8 Sep 2023 by
Georgi Vidinski
Post: bump. This is still happening and I haven't seen any engagement. Hi Donald, What do you mean by "& linux"? As I can see your issues id only with the MacOS application, isn't it? And yes there are some new WUs which carries larger number of Light points than the older limit of 1000. Having updated app for MacOS is in the Task List, but I can't say any specific time that it'll take. Meanwhile you may want to switch your Macs to a different project and get back once we have updated version. Best, Georgi |
Previous 20 · Next 20