Comment by nickjj
1 day ago
Swap doesn't help in this case.
It's GPU drivers freaking out when they run out of VRAM, but it might only be specific to NVIDIA cards since AMD seems to handle this better.
But, if it's driver specific that means it could be fixed at the kernel level I suppose, especially since Windows handles it seamlessly?
Can you explain a little more please. I thought that graphics processing would revert to CPU if the GPU is maxed out and so swap might ultimately help? Is there no general GPU-swap mechanism? If CPU threads are competing for GPU resource does the GPU getting paged-out?
I suppose this could be an historic thing, GPUs can have a type of DMA now, and disk access is much faster too, so paging perhaps makes sense when it didn't before?
The linked post in the original comment at https://nickjanetakis.com/blog/gpu-memory-allocation-bugs-wi... has a lot of details.
AMD's GPU do use system memory seamlessly as a fallback, but NVIDIA's don't, at least not back in January 2026. Hundreds of people reported it in their forums too, or variations of the problem.
AMD GPUs use GTT seemlessly under linux. It looks funny in some "top" applications when iGPU that has 512mb allocated to it, uses 8gb of memory: displayed as "uses 8gb out of 0.5gb"
Thanks, I'll give it a read.