Re: [PATCH] drm/xe: Handle mixed mappings and existing VRAM on atomic faults

["Ghimiray, Himal Prasad" <himal.prasad.ghimiray@intel.com>]

On 09-10-2025 18:36, Matthew Brost wrote:
> Moving to VRAM will fail if mixed mappings are present or if the page is
> already located in VRAM. Atomic faults that require a move to VRAM
> currently retry without attempting to evict mixed mappings or locate
> existing VRAM mappings.
> 
> This patch fixes the issue by attempting to evict mixed mappings or find
> existing VRAM pages when a move to VRAM fails during atomic fault
> handling.
> 
> Fixes: a9ac0fa455b0 ("drm/xe: Strict migration policy for atomic SVM faults")
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>

Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>> ---
>   drivers/gpu/drm/xe/xe_svm.c | 13 ++++++++++++-
>   1 file changed, 12 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
> index 7e2db71ff34e..b268ee0d2271 100644
> --- a/drivers/gpu/drm/xe/xe_svm.c
> +++ b/drivers/gpu/drm/xe/xe_svm.c
> @@ -1073,7 +1073,17 @@ static int __xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
>   				drm_dbg(&vm->xe->drm,
>   					"VRAM allocation failed, falling back to retrying fault, asid=%u, errno=%pe\n",
>   					vm->usm.asid, ERR_PTR(err));
> -				goto retry;
> +
> +				/*
> +				 * In the devmem-only case, mixed mappings may
> +				 * be found. The get_pages function will fix
> +				 * these up to a single location, allowing the
> +				 * page fault handler to make forward progress.
> +				 */
> +				if (ctx.devmem_only)
> +					goto get_pages;
> +				else
> +					goto retry;
>   			} else {
>   				drm_err(&vm->xe->drm,
>   					"VRAM allocation failed, retry count exceeded, asid=%u, errno=%pe\n",
> @@ -1083,6 +1093,7 @@ static int __xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
>   		}
>   	}
>   
> +get_pages:
>   	get_pages_start = xe_svm_stats_ktime_get();
>   
>   	range_debug(range, "GET PAGES");