[PATCH 2/2] drm/amdkfd: map SVM range with correct access permission

Felix Kuehling felix.kuehling at amd.com
Thu Aug 19 21:32:38 UTC 2021


Am 2021-08-19 um 10:56 a.m. schrieb Philip Yang:
> Restore retry fault or prefetch range, or restore svm range after
> eviction to map range to GPU with correct read or write access
> permission.
>
> Range may includes multiple VMAs, update GPU page table with offset of
> prange, number of pages for each VMA according VMA access permission.
>
> Signed-off-by: Philip Yang <Philip.Yang at amd.com>

Minor nitpicks, and one question. See inline. It looks good otherwise.


> ---
>  drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 131 +++++++++++++++++----------
>  1 file changed, 84 insertions(+), 47 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
> index cf1009bb532a..94612581963f 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
> @@ -120,6 +120,7 @@ static void svm_range_remove_notifier(struct svm_range *prange)
>  
>  static int
>  svm_range_dma_map_dev(struct amdgpu_device *adev, struct svm_range *prange,
> +		      unsigned long offset, unsigned long npages,
>  		      unsigned long *hmm_pfns, uint32_t gpuidx)
>  {
>  	enum dma_data_direction dir = DMA_BIDIRECTIONAL;
> @@ -136,7 +137,8 @@ svm_range_dma_map_dev(struct amdgpu_device *adev, struct svm_range *prange,
>  		prange->dma_addr[gpuidx] = addr;
>  	}
>  
> -	for (i = 0; i < prange->npages; i++) {
> +	addr += offset;
> +	for (i = 0; i < npages; i++) {
>  		if (WARN_ONCE(addr[i] && !dma_mapping_error(dev, addr[i]),
>  			      "leaking dma mapping\n"))
>  			dma_unmap_page(dev, addr[i], PAGE_SIZE, dir);
> @@ -167,6 +169,7 @@ svm_range_dma_map_dev(struct amdgpu_device *adev, struct svm_range *prange,
>  
>  static int
>  svm_range_dma_map(struct svm_range *prange, unsigned long *bitmap,
> +		  unsigned long offset, unsigned long npages,
>  		  unsigned long *hmm_pfns)
>  {
>  	struct kfd_process *p;
> @@ -187,7 +190,8 @@ svm_range_dma_map(struct svm_range *prange, unsigned long *bitmap,
>  		}
>  		adev = (struct amdgpu_device *)pdd->dev->kgd;
>  
> -		r = svm_range_dma_map_dev(adev, prange, hmm_pfns, gpuidx);
> +		r = svm_range_dma_map_dev(adev, prange, offset, npages,
> +					  hmm_pfns, gpuidx);
>  		if (r)
>  			break;
>  	}
> @@ -1088,11 +1092,6 @@ svm_range_get_pte_flags(struct amdgpu_device *adev, struct svm_range *prange,
>  	pte_flags |= snoop ? AMDGPU_PTE_SNOOPED : 0;
>  
>  	pte_flags |= amdgpu_gem_va_map_flags(adev, mapping_flags);
> -
> -	pr_debug("svms 0x%p [0x%lx 0x%lx] vram %d PTE 0x%llx mapping 0x%x\n",
> -		 prange->svms, prange->start, prange->last,
> -		 (domain == SVM_RANGE_VRAM_DOMAIN) ? 1:0, pte_flags, mapping_flags);
> -
>  	return pte_flags;
>  }
>  
> @@ -1156,7 +1155,8 @@ svm_range_unmap_from_gpus(struct svm_range *prange, unsigned long start,
>  
>  static int
>  svm_range_map_to_gpu(struct amdgpu_device *adev, struct amdgpu_vm *vm,
> -		     struct svm_range *prange, dma_addr_t *dma_addr,
> +		     struct svm_range *prange, unsigned long offset,
> +		     unsigned long npages, bool readonly, dma_addr_t *dma_addr,
>  		     struct amdgpu_device *bo_adev, struct dma_fence **fence)
>  {
>  	struct amdgpu_bo_va bo_va;
> @@ -1167,14 +1167,15 @@ svm_range_map_to_gpu(struct amdgpu_device *adev, struct amdgpu_vm *vm,
>  	int r = 0;
>  	int64_t i;
>  
> -	pr_debug("svms 0x%p [0x%lx 0x%lx]\n", prange->svms, prange->start,
> -		 prange->last);
> +	last_start = prange->start + offset;
> +
> +	pr_debug("svms 0x%p [0x%lx 0x%lx] readonly %d\n", prange->svms,
> +		 last_start, last_start + npages - 1, readonly);
>  
>  	if (prange->svm_bo && prange->ttm_res)
>  		bo_va.is_xgmi = amdgpu_xgmi_same_hive(adev, bo_adev);
>  
> -	last_start = prange->start;
> -	for (i = 0; i < prange->npages; i++) {
> +	for (i = offset; i < offset + npages; i++) {
>  		last_domain = dma_addr[i] & SVM_RANGE_VRAM_DOMAIN;
>  		dma_addr[i] &= ~SVM_RANGE_VRAM_DOMAIN;
>  		if ((prange->start + i) < prange->last &&
> @@ -1183,13 +1184,21 @@ svm_range_map_to_gpu(struct amdgpu_device *adev, struct amdgpu_vm *vm,
>  
>  		pr_debug("Mapping range [0x%lx 0x%llx] on domain: %s\n",
>  			 last_start, prange->start + i, last_domain ? "GPU" : "CPU");
> +
>  		pte_flags = svm_range_get_pte_flags(adev, prange, last_domain);
> -		r = amdgpu_vm_bo_update_mapping(adev, bo_adev, vm, false, false, NULL,
> -						last_start,
> +		if (readonly)
> +			pte_flags &= ~AMDGPU_PTE_WRITEABLE;
> +
> +		pr_debug("svms 0x%p map [0x%lx 0x%llx] vram %d PTE 0x%llx\n",
> +			 prange->svms, last_start, prange->start + i,
> +			 (last_domain == SVM_RANGE_VRAM_DOMAIN) ? 1 : 0,
> +			 pte_flags);
> +
> +		r = amdgpu_vm_bo_update_mapping(adev, bo_adev, vm, false, false,
> +						NULL, last_start,
>  						prange->start + i, pte_flags,
>  						last_start - prange->start,
> -						NULL,
> -						dma_addr,
> +						NULL, dma_addr,
>  						&vm->last_update,
>  						&table_freed);
>  		if (r) {
> @@ -1220,8 +1229,10 @@ svm_range_map_to_gpu(struct amdgpu_device *adev, struct amdgpu_vm *vm,
>  	return r;
>  }
>  
> -static int svm_range_map_to_gpus(struct svm_range *prange,
> -				 unsigned long *bitmap, bool wait)
> +static int
> +svm_range_map_to_gpus(struct svm_range *prange, unsigned long offset,
> +		      unsigned long npages, bool readonly,
> +		      unsigned long *bitmap, bool wait)
>  {
>  	struct kfd_process_device *pdd;
>  	struct amdgpu_device *bo_adev;
> @@ -1257,7 +1268,8 @@ static int svm_range_map_to_gpus(struct svm_range *prange,
>  		}
>  
>  		r = svm_range_map_to_gpu(adev, drm_priv_to_vm(pdd->drm_priv),
> -					 prange, prange->dma_addr[gpuidx],
> +					 prange, offset, npages, readonly,
> +					 prange->dma_addr[gpuidx],
>  					 bo_adev, wait ? &fence : NULL);
>  		if (r)
>  			break;
> @@ -1390,6 +1402,7 @@ static int svm_range_validate_and_map(struct mm_struct *mm,
>  				      int32_t gpuidx, bool intr, bool wait)
>  {
>  	struct svm_validate_context ctx;
> +	unsigned long start, end, addr;
>  	struct hmm_range *hmm_range;
>  	struct kfd_process *p;
>  	void *owner;
> @@ -1448,40 +1461,64 @@ static int svm_range_validate_and_map(struct mm_struct *mm,
>  			break;
>  		}
>  	}
> -	r = amdgpu_hmm_range_get_pages(&prange->notifier, mm, NULL,
> -				       prange->start << PAGE_SHIFT,
> -				       prange->npages, &hmm_range,
> -				       false, true, owner);
> -	if (r) {
> -		pr_debug("failed %d to get svm range pages\n", r);
> -		goto unreserve_out;
> -	}
>  
> -	r = svm_range_dma_map(prange, ctx.bitmap,
> -			      hmm_range->hmm_pfns);
> -	if (r) {
> -		pr_debug("failed %d to dma map range\n", r);
> -		goto unreserve_out;
> -	}
> +	start = prange->start << PAGE_SHIFT;
> +	end = (prange->last + 1) << PAGE_SHIFT;
> +	for (addr = start; addr < end && !r; ) {
> +		struct vm_area_struct *vma;
> +		unsigned long next;
> +		unsigned long offset;
> +		unsigned long npages;
> +		bool readonly;
>  
> -	prange->validated_once = true;
> +		vma = find_vma(mm, addr);
> +		if (!vma || addr < vma->vm_start) {
> +			r = -EINVAL;

I think -EFAULT would be the appropriate error code here.


> +			goto unreserve_out;
> +		}
> +		readonly = !(vma->vm_flags & VM_WRITE);
>  
> -	svm_range_lock(prange);
> -	if (amdgpu_hmm_range_get_pages_done(hmm_range)) {
> -		pr_debug("hmm update the range, need validate again\n");
> -		r = -EAGAIN;
> -		goto unlock_out;
> -	}
> -	if (!list_empty(&prange->child_list)) {
> -		pr_debug("range split by unmap in parallel, validate again\n");
> -		r = -EAGAIN;
> -		goto unlock_out;
> -	}
> +		next = min(vma->vm_end, end);
> +		npages = (next - addr) / PAGE_SIZE;

Use >> PAGE_SHIFT for consistency.


> +		r = amdgpu_hmm_range_get_pages(&prange->notifier, mm, NULL,
> +					       addr, npages, &hmm_range,
> +					       readonly, true, owner);
> +		if (r) {
> +			pr_debug("failed %d to get svm range pages\n", r);
> +			goto unreserve_out;
> +		}
>  
> -	r = svm_range_map_to_gpus(prange, ctx.bitmap, wait);
> +		offset = (addr - start) / PAGE_SIZE;

>> PAGE_SHIFT


> +		r = svm_range_dma_map(prange, ctx.bitmap, offset, npages,
> +				      hmm_range->hmm_pfns);
> +		if (r) {
> +			pr_debug("failed %d to dma map range\n", r);
> +			goto unreserve_out;
> +		}
> +
> +		svm_range_lock(prange);
> +		if (amdgpu_hmm_range_get_pages_done(hmm_range)) {
> +			pr_debug("hmm update the range, need validate again\n");
> +			r = -EAGAIN;
> +			goto unlock_out;
> +		}
> +		if (!list_empty(&prange->child_list)) {
> +			pr_debug("range split by unmap in parallel, validate again\n");
> +			r = -EAGAIN;
> +			goto unlock_out;
> +		}
> +
> +		r = svm_range_map_to_gpus(prange, offset, npages, readonly,
> +					  ctx.bitmap, wait);
>  
>  unlock_out:
> -	svm_range_unlock(prange);
> +		svm_range_unlock(prange);
> +
> +		addr = next;
> +	}
> +
> +	prange->validated_once = true;

Should this be conditional on "!r"?

Regards,
  Felix


> +
>  unreserve_out:
>  	svm_range_unreserve_bos(&ctx);
>  


More information about the amd-gfx mailing list
  翻译: