I'm not noticing any performance improvements, but no regressions either. It seems solid enough. The only difference I see is that it's calling some new ioctl that I've never seen before, _IOC(_IOC_READ|_IOC_WRITE, 0x4b, 0x2, 0x60) which is apparently a new AMDKFD_IOC call.
However I'm also seeing less interference with day-to-day graphical tasks. I haven't had any crashes when switching virtual terminals like I did with 6.2.3 but it is possible that it is only coincidence.
Edit: Hasn't eliminated the issue with amdgpu crashing so I guess that is 100% on amdgpu's side and is not something that ROCm is doing to it. But it's not worse anyway and it doesn't interfere with folding, just GUI tasks.
Amd released Rocm 6.4
If you think it might be a driver problem, see viewforum.php?f=79
Moderators: Site Moderators, FAHC Science Team
Return to “GPU Projects and FahCores”
Jump to
- Announcements
- ↳ Announcements - Folding Consortium
- ↳ Announcements - Forum related
- New Donors
- ↳ New Donors start here
- Current Software
- ↳ v8.4.9 Public Release for Windows / Linux / macOS
- ↳ v8.4.xx Open Beta
- ↳ V8.3.xx Open Beta
- ↳ V8.1.xx Open Beta
- ↳ V7.6.x Public Release Windows/Linux/MacOS X
- ↳ V7.5.1 Public Release Windows/Linux/MacOS X [deprecated]
- ↳ Software for ARM hardware
- ↳ Q&A about unsupported distros of Linux
- ↳ New GPUs (whitelist)
- ↳ Android client from Sony (deprecated)
- GPU drivers
- ↳ Problems with NVidia drivers
- ↳ Windows 10 + NVidia
- ↳ Problems with AMD/ATI drivers
- ↳ Intel iGP Drivers
- Stats and server issues
- ↳ Issues with a specific server
- ↳ Issues with a specific WU
- ↳ CPU Projects - released FAHCores _a7 & _a8 (a4 retired)
- ↳ GPU Projects and FahCores
- ↳ Issues with this forum
- ↳ CPU Projects - beta FAHCores (Currently _a8)
- General Discussions
- ↳ Discussions of General-FAH topics
- ↳ How can we help FAH grow
- ↳ The Science of FAH -- questions/answers
- ↳ Scientific discussions (Non-FAH)
- ↳ FAH Hardware
- ↳ In memoriam
- 3rd party software
- ↳ 3rd party contributed software