Re: [PATCH v4 1/5] drm/xe: Add helper to extend CPU-mirrored VMA range for merge

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

On Mon, Nov 03, 2025 at 11:39:53AM +0530, Himal Prasad Ghimiray wrote:
> Introduce xe_vm_find_cpu_addr_mirror_vma_range(), which computes an
> extended range around a given range by including adjacent VMAs that are
> CPU-address-mirrored and have default memory attributes. This helper is
> useful for determining mergeable range without performing the actual merge.
> 
> v2
> - Add assert
> - Move unmap check to this patch
> 
> v3
> - Decrease offset to check by SZ_4K to avoid wrong vma return in fast
>   lookup path
> 
> Cc: Matthew Brost <matthew.brost@intel.com>
> Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
> ---
>  drivers/gpu/drm/xe/xe_vm.c | 41 ++++++++++++++++++++++++++++++++++++++
>  drivers/gpu/drm/xe/xe_vm.h |  3 +++
>  2 files changed, 44 insertions(+)
> 
> diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
> index 00f3520dec38..944eb35bb2b8 100644
> --- a/drivers/gpu/drm/xe/xe_vm.c
> +++ b/drivers/gpu/drm/xe/xe_vm.c
> @@ -4363,6 +4363,47 @@ int xe_vm_alloc_madvise_vma(struct xe_vm *vm, uint64_t start, uint64_t range)
>  	return xe_vm_alloc_vma(vm, &map_req, true);
>  }
>  
> +static bool is_cpu_addr_vma_with_default_attr(struct xe_vma *vma)
> +{
> +	return vma && xe_vma_is_cpu_addr_mirror(vma) &&
> +	       !xe_svm_has_mapping(xe_vma_vm(vma), xe_vma_start(vma), xe_vma_end(vma)) &&

Why do we need the xe_svm_has_mapping check? I think it ok to merge to
VMAs even if they have SVM ranges within the VMA as VMA should be fully
disconnected from the SVM ranges.



> +	       xe_vma_has_default_mem_attrs(vma);
> +}
> +
> +/**
> + * xe_vm_find_cpu_addr_mirror_vma_range - Extend a VMA range to include adjacent CPU-mirrored VMAs
> + * @vm: VM to search within
> + * @start: Input/output pointer to the starting address of the range
> + * @end: Input/output pointer to the end address of the range
> + *
> + * Given a range defined by @start and @range, this function checks the VMAs
> + * immediately before and after the range. If those neighboring VMAs are
> + * CPU-address-mirrored and have default memory attributes, the function
> + * updates @start and @range to include them. This extended range can then
> + * be used for merging or other operations that require a unified VMA.
> + *
> + * The function does not perform the merge itself; it only computes the
> + * mergeable boundaries.
> + */
> +void xe_vm_find_cpu_addr_mirror_vma_range(struct xe_vm *vm, u64 *start, u64 *end)
> +{
> +	struct xe_vma *prev, *next;
> +
> +	lockdep_assert_held(&vm->lock);
> +
> +	if (*start > SZ_4K) {

*start >= SZ_4K

Matt

> +		prev = xe_vm_find_vma_by_addr(vm, *start - SZ_4K);
> +		if (is_cpu_addr_vma_with_default_attr(prev))
> +			*start = xe_vma_start(prev);
> +	}
> +
> +	if (*end < vm->size) {
> +		next = xe_vm_find_vma_by_addr(vm, *end + 1);
> +		if (is_cpu_addr_vma_with_default_attr(next))
> +			*end = xe_vma_end(next);
> +	}
> +}
> +
>  /**
>   * xe_vm_alloc_cpu_addr_mirror_vma - Allocate CPU addr mirror vma
>   * @vm: Pointer to the xe_vm structure
> diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
> index ef8a5019574e..361f10b3c453 100644
> --- a/drivers/gpu/drm/xe/xe_vm.h
> +++ b/drivers/gpu/drm/xe/xe_vm.h
> @@ -68,6 +68,9 @@ xe_vm_find_overlapping_vma(struct xe_vm *vm, u64 start, u64 range);
>  
>  bool xe_vma_has_default_mem_attrs(struct xe_vma *vma);
>  
> +void xe_vm_find_cpu_addr_mirror_vma_range(struct xe_vm *vm,
> +					  u64 *start,
> +					  u64 *end);
>  /**
>   * xe_vm_has_scratch() - Whether the vm is configured for scratch PTEs
>   * @vm: The vm
> -- 
> 2.34.1
>