Maximilian Golub

Same issue, Macos 10.13.6 on go version go1.11.5 darwin/amd64

I’m also having this issue with 2x 4090s - it actually corrupts the model files when I use 2 GPUs. Both work fine by themselves using `CUDA_VISIBLE_DEVICES` , and both...

I Also have this issue - two rocky linux 9.2 systems, phoenix test suite 10.8.4, and nearly identical php-confs: ```bash 3,4c3,4 < PHP: 8.0.30 < PHP VERSION ID: 80030 ---...

After much print/echo debugging, I figured out how to access the actual error code from $http_response_header[0] (php has magic hidden variables, I relearned): ```php public static function http_upload_via_post($url, $to_post_data, $supports_proxy...

Also seeing this issue, on ROCm 6.0.2, Ubuntu 22.04 fresh setup when running llama.cpp. Hardware is a AMD 7940HS CPU w/ Radeon 780M GPU. When loading up larger models and...

> > 5.15.0-94-generic > > Any reason you want to run the older kernel? > > A few people are experiencing better results on the newer kernels, and the 6.0.2...

> PS: if recent kernels help, with patches not backported into the fresh-install Ubuntu kernel the AMD documentation implicitly assumes, then said documentation should include instructions on installing an updated...

> and just checking - you are on new kernels with the 6.0.2 binaries or your code compiled against the 6.0.2 libs? (yes - i have had to compile pytorch...

> Possibly thinking this is some weird interplay between threading, affinity and the cx11 ABI? Will test in a more normal configuration soon. Wow, it might actually be cxx11 abi...

I can paper over the issue by setting the cpu affinity using (fishshell incoming) `sudo pgrep pt_main_thread | while read -l pid; taskset --all-tasks -p ffff $pid; end` > That's...