AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |
Back to Blog
Error 417 boinc3/20/2023 ![]() The first preliminary result on my standalone test on my HD 4000 says that it's running and quite fast. I would be grateful to hear about results of running the standalone test on an Intel HD 4000. Has anyone manged to install drivers in a way that both an Intel HD GPU and an AMD GPU are visible as OpenCL cards to BOINC at the same time? If so, how? There have been numerous reports that the Intel OpenCL drivers cause problems for running OpenCL apps on other GPUs, so I will only later install a HD 7790 in this system. Does anyone know what kind of control one has to configure the Intel HD GPU, via BIOS or Intel tools? I need to explore this a bit more.ĭuring the tests, the HD 2500 was also driving the video output for the PC. Well, the GPU shares the power and thermal envelope with the CPU. I used GPU-Z to monitor the GPU, and it seems to be very aggressive in downclocking the GPU clock rate. I suspect that the dynamic memory allocation of the GPU as opposed to the dedicated video RAM of GPUs on PCIe boards may cause some problems. I got some error messages in my stderr.txt which need some investigation. I'll have to make some tests with one phycical or virtual core reserved for the GPU app. On my new PC it took about 10 hours while the CPU was under full load with regular BOINC CPU jobs. Output will be continuously written to the file stderr.txt so you can periodically check there for messages. Ideally, the BRP4 workunit will now start processsing in standalone mode. Where EXE-FILE is the name of the executable you downloaded and N is the device number (see above). Now run this in the directory you have created with all the downloaded stuff: Where N is the number you are looking for. Look up the device number of teh Intel HD GPU, there should be something the event log of BOINC like Now download the OpenCL-ATI binary from Albert to the same directory: ![]() You can download a ZIP file containing the input files for a BRP4 task (complete with 8 sub-tasks) from here: Oh well, so let's start with standalone testing. If all else fails I will build a special app version that ignores all input from the BOINC side and will run the task on the first Intel GPU it can find.for testing only, of course. I'll keep trying and will need to talk to Bernd next week when back at the office, Bernd is our expert on the server part software. ![]() The server software of doesn't yet understand work requests for Intel GPUs, so I thought some combination of app_info.xml and cc_config.xml editing might do the trick, but so far this doesn't work for me. I got it as far as requesting CPU work and then trying to run it on the GPU, but at that point it fails. I was unable so far to trick BOINC into doing work units on the HD GPU under BOINC control. So I played around with it a bit, with mixed results: Yesterday I finally got my hands on a Windows PC with Ivy Bridge CPU and a HD 2500 integrated GPU. Is anyone who is experienced in app_info.xml writing interested in such an experiment? So the Boinc client would ask for work for a CPU anonymous platform task, but the task would then in fact be calculated on the Intel GPU, so one would at least get an idea about the runtime. What would be possible at the moment: we provide (in the timeframe of a few days) an app to download that someone could test using an app_info.xml file that would pretend the app is a CPU app. Yeah, it's a shame really, especially with the next Intel CPU generation to be launched soon (Haswell) which is said to feature some relatively powerful GPUs. We (the admins and devs) have quite a lot of other stuff on our desks at the moment (to name just a few: launch of S6directed on Einstein, transition of BRP4 to a CPU only run, evaluating the early BRP5 results, a complete overhaul of the Web Homepage and forum platform, postprocessing of finished GW runs. Our BOINC server side code is not yet able to handle Intel GPUs consistently (certainly not on Einstein and not even on Albert IMHO). Any progress on this? I have successfully crunched GPU tasks for both collatz and SETI (astropulse), on my intel HD 4000.
0 Comments
Read More
Leave a Reply. |