Hi Christian, On 2023-10-18 04:07, Christian Kastner wrote:
On 2023-10-18 09:31, Cordell Bloor wrote:I have a Radeon Pro W6400 and I couldn't reproduce the crash. Just a data point to add.Well, that makes things even more interesting. This is with a kernel and firmware-amd-graphics package from the official Archive, right?
Yes. I've uploaded a new batch of logs from my latest round of hardware testing [1][2]. (There are lots of hipblas failures, but I believe those are caused by my use of OpenBLAS as the reference library. I wouldn't worry too much about them.)
Sincerely, Cory Bloor [1]: https://slerp.xyz/rocm/logs/full/ [2]: https://slerp.xyz/rocm/logs/full/2023-10-05-gfx1034.log