Re: [PATCH v2] drm/xe/pt: check no-DMA huge-pte cases before DMA segment test

[Matthew Brost <matthew.brost@intel.com>]

On Mon, Jul 27, 2026 at 11:43:43AM +0530, Himal Prasad Ghimiray wrote:
> On a non-range clear, curs.size is never set, so the segment test
> (next - va_curs_start > curs->size) returns false for every level > 0
> before the clear_pt short-circuit is reached. The clear then descends to
> level 0 instead of forming a huge zero-leaf, wasting page tables and
> risking -ENOMEM on unbind.
> 
> Move the null-VMA, purged-BO and clear_pt short-circuits above the
> curs->size test. The bind path always sets curs.size, so it is unaffected.
> 
> v2
> - Also set curs.size on the clear path so the cursor stays meaningful
> during the walk. clear_pt is only reached with range == NULL, so assert
> that invariant. (Matthew Brost)
> 
> Cc: Matthew Brost <matthew.brost@intel.com>

Reviewed-by: Matthew Brost <matthew.brost@intel.com>

> Fixes: 5b658b7e89c3 ("drm/xe: Clear scratch page on vm_bind")
> Reported-by: Sashiko <sashiko-bot@kernel.org>
> Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
> ---
>  drivers/gpu/drm/xe/xe_pt.c | 13 ++++++++-----
>  1 file changed, 8 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
> index 598c6b2571e7..0e9d669620b9 100644
> --- a/drivers/gpu/drm/xe/xe_pt.c
> +++ b/drivers/gpu/drm/xe/xe_pt.c
> @@ -443,10 +443,6 @@ static bool xe_pt_hugepte_possible(u64 addr, u64 next, unsigned int level,
>  	if (!xe_pt_covers(addr, next, level, &xe_walk->base))
>  		return false;
>  
> -	/* Does the DMA segment cover the whole pte? */
> -	if (next - xe_walk->va_curs_start > xe_walk->curs->size)
> -		return false;
> -
>  	/* null VMA's and purged BO's do not have dma addresses */
>  	if (xe_vma_is_null(xe_walk->vma) || (bo && xe_bo_is_purged(bo)))
>  		return true;
> @@ -455,6 +451,10 @@ static bool xe_pt_hugepte_possible(u64 addr, u64 next, unsigned int level,
>  	if (xe_walk->clear_pt)
>  		return true;
>  
> +	/* Does the DMA segment cover the whole pte? */
> +	if (next - xe_walk->va_curs_start > xe_walk->curs->size)
> +		return false;
> +
>  	/* Is the DMA address huge PTE size aligned? */
>  	size = next - addr;
>  	dma = addr - xe_walk->va_curs_start + xe_res_dma(xe_walk->curs);
> @@ -775,8 +775,11 @@ xe_pt_stage_bind(struct xe_tile *tile, struct xe_vma *vma,
>  	}
>  
>  	xe_walk.needs_64K = (vm->flags & XE_VM_FLAG_64K);
> -	if (clear_pt)
> +	if (clear_pt) {
> +		xe_assert(xe, !range);
> +		curs.size = xe_vma_size(vma);
>  		goto walk_pt;
> +	}
>  
>  	if (vma->gpuva.flags & XE_VMA_ATOMIC_PTE_BIT) {
>  		xe_walk.default_vram_pte = xe_atomic_for_vram(vm, vma) ? XE_USM_PPGTT_PTE_AE : 0;
> -- 
> 2.43.0
>