Skip to contentSkip to conversation

drm/amdkfd: Fix partial migrate issue

PATCH In review Emily Deng 2 participants 2 messages Jan 10, 2025

Subsystems (3)
lore
Share & tools
lore

Emily Deng

author
#0 lore raw

PATCH v5 1/1 drm/amdkfd: Fix partial migrate issue

For partial migrate from ram to vram, the migrate->cpages is not equal to migrate->npages, should use migrate->npages to check all needed migrate pages which could be copied or not.

And only need to set those pages could be migrated to migrate->dst[i], or the migrate_vma_pages will migrate the wrong pages based on the migrate->dst[i].

v2:
Add mpages to break the loop earlier.

v3:
Uses MIGRATE_PFN_MIGRATE to identify whether page could be migrated.

---
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)
drivers/gpu/drm/amd/amdkfd/kfd_migrate.c +12 -8
index 4b275937d05e..edff564d6a3c 100644
@@ -278,10 +278,11 @@ svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange,
 			 struct migrate_vma *migrate, struct dma_fence **mfence,
 			 dma_addr_t *scratch, uint64_t ttm_res_offset)
 {
-	uint64_t npages = migrate->cpages;
+	uint64_t npages = migrate->npages;
 	struct amdgpu_device *adev = node->adev;
 	struct device *dev = adev->dev;
 	struct amdgpu_res_cursor cursor;
+	uint64_t mpages = 0;
 	dma_addr_t *src;
 	uint64_t *dst;
 	uint64_t i, j;
@@ -295,14 +296,16 @@ svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange,
 
 	amdgpu_res_first(prange->ttm_res, ttm_res_offset,
 			 npages << PAGE_SHIFT, &cursor);
-	for (i = j = 0; i < npages; i++) {
+	for (i = j = 0; (i < npages) && (mpages < migrate->cpages); i++) {
 		struct page *spage;
 
-		dst[i] = cursor.start + (j << PAGE_SHIFT);
-		migrate->dst[i] = svm_migrate_addr_to_pfn(adev, dst[i]);
-		svm_migrate_get_vram_page(prange, migrate->dst[i]);
-		migrate->dst[i] = migrate_pfn(migrate->dst[i]);
-
+		if (migrate->src[i] & MIGRATE_PFN_MIGRATE) {
+			dst[i] = cursor.start + (j << PAGE_SHIFT);
+			migrate->dst[i] = svm_migrate_addr_to_pfn(adev, dst[i]);
+			svm_migrate_get_vram_page(prange, migrate->dst[i]);
+			migrate->dst[i] = migrate_pfn(migrate->dst[i]);
+			mpages++;
+		}
 		spage = migrate_pfn_to_page(migrate->src[i]);
 		if (spage && !is_zone_device_page(spage)) {
 			src[i] = dma_map_page(dev, spage, 0, PAGE_SIZE,
@@ -354,7 +357,8 @@ svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange,
 	if (r) {
 		pr_debug("failed %d to copy memory to vram\n", r);
 		while (i--) {
-			svm_migrate_put_vram_page(adev, dst[i]);
+			if (dst[i])
+				svm_migrate_put_vram_page(adev, dst[i]);
 			migrate->dst[i] = 0;
 		}
 	}
Signature · 1 line
2.34.1
#1 lore raw

On 2025-01-10 09:25, Emily Deng wrote:
For partial migrate from ram to vram, the migrate->cpages is not equal to migrate->npages, should use migrate->npages to check all needed migrate pages which could be copied or not. And only need to set those pages could be migrated to migrate->dst[i], or the migrate_vma_pages will migrate the wrong pages based on the migrate->dst[i]. v2: Add mpages to break the loop earlier. v3: Uses MIGRATE_PFN_MIGRATE to identify whether page could be migrated. Signed-off-by: Emily Deng <[email protected]> --- drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 20 ++++++++++++-------- 1 file changed, 12 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c index 4b275937d05e..edff564d6a3c 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c @@ -278,10 +278,11 @@ svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange, struct migrate_vma *migrate, struct dma_fence **mfence, dma_addr_t *scratch, uint64_t ttm_res_offset) { - uint64_t npages = migrate->cpages; + uint64_t npages = migrate->npages; struct amdgpu_device *adev = node->adev; struct device *dev = adev->dev; struct amdgpu_res_cursor cursor; + uint64_t mpages = 0; dma_addr_t *src; uint64_t *dst; uint64_t i, j; @@ -295,14 +296,16 @@ svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange, amdgpu_res_first(prange->ttm_res, ttm_res_offset, npages << PAGE_SHIFT, &cursor); - for (i = j = 0; i < npages; i++) { + for (i = j = 0; (i < npages) && (mpages < migrate->cpages); i++) { struct page *spage; - dst[i] = cursor.start + (j << PAGE_SHIFT); - migrate->dst[i] = svm_migrate_addr_to_pfn(adev, dst[i]); - svm_migrate_get_vram_page(prange, migrate->dst[i]); - migrate->dst[i] = migrate_pfn(migrate->dst[i]); - + if (migrate->src[i] & MIGRATE_PFN_MIGRATE) { + dst[i] = cursor.start + (j << PAGE_SHIFT); + migrate->dst[i] = svm_migrate_addr_to_pfn(adev, dst[i]); + svm_migrate_get_vram_page(prange, migrate->dst[i]); + migrate->dst[i] = migrate_pfn(migrate->dst[i]); + mpages++; + } spage = migrate_pfn_to_page(migrate->src[i]); if (spage && !is_zone_device_page(spage)) { src[i] = dma_map_page(dev, spage, 0, PAGE_SIZE, @@ -354,7 +357,8 @@ svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange, if (r) { pr_debug("failed %d to copy memory to vram\n", r); while (i--) { - svm_migrate_put_vram_page(adev, dst[i]); + if (dst[i]) For error case if svm_migrate_copy_memory_gart failed in the middle of for loop, i is the last page index, the i-- could miss the last page. please use the change in my v4 review.
Regards,
Philip

+ svm_migrate_put_vram_page(adev, dst[i]); migrate->dst[i] = 0; } }