[PATCH 4/6] drm/amdgpu: rework how isolation is enforced v2

Limiting the number of available VMIDs to enforce isolation causes some issues with gang submit and applying certain HW workarounds which require multiple VMIDs to work correctly. So instead start to track all submissions to the relevant engines in a per partition data structure and use the dma_fe

[PATCH 4/6] drm/amdgpu: rework how isolation is enforced

Limiting the number of available VMIDs to enforce isolation causes some issues with gang submit and applying certain HW workarounds which require multiple VMIDs to work correctly. So instead start to track all submissions to the relevant engines in a per partition data structure and use the dma_fe