mm/migrate.c: remove MIGRATE_PFN_LOCKED

MIGRATE_PFN_LOCKED is used to indicate to migrate_vma_prepare() that a
source page was already locked during migrate_vma_collect().  If it
wasn't then the a second attempt is made to lock the page.  However if
the first attempt failed it's unlikely a second attempt will succeed,
and the retry adds complexity.  So clean this up by removing the retry
and MIGRATE_PFN_LOCKED flag.

Destination pages are also meant to have the MIGRATE_PFN_LOCKED flag
set, but nothing actually checks that.

Link: https://lkml.kernel.org/r/20211025041608.289017-1-apopple@nvidia.com
Signed-off-by: Alistair Popple <apopple@nvidia.com>
Reviewed-by: Ralph Campbell <rcampbell@nvidia.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Jerome Glisse <jglisse@redhat.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Zi Yan <ziy@nvidia.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Documentation/vm/hmm.rst

@@ -360,7 +360,7 @@ between device driver specific code and shared common code:
    system memory page, locks the page with ``lock_page()``, and fills in the
    ``dst`` array entry with::
 
-     dst[i] = migrate_pfn(page_to_pfn(dpage)) | MIGRATE_PFN_LOCKED;
+     dst[i] = migrate_pfn(page_to_pfn(dpage));
 
    Now that the driver knows that this page is being migrated, it can
    invalidate device private MMU mappings and copy device private memory

arch/powerpc/kvm/book3s_hv_uvmem.c

@@ -560,7 +560,7 @@ static int __kvmppc_svm_page_out(struct vm_area_struct *vma,
 				  gpa, 0, page_shift);
 
 	if (ret == U_SUCCESS)
-		*mig.dst = migrate_pfn(pfn) | MIGRATE_PFN_LOCKED;
+		*mig.dst = migrate_pfn(pfn);
 	else {
 		unlock_page(dpage);
 		__free_page(dpage);
@@ -774,7 +774,7 @@ static int kvmppc_svm_page_in(struct vm_area_struct *vma,
 		}
 	}
 
-	*mig.dst = migrate_pfn(page_to_pfn(dpage)) | MIGRATE_PFN_LOCKED;
+	*mig.dst = migrate_pfn(page_to_pfn(dpage));
 	migrate_vma_pages(&mig);
 out_finalize:
 	migrate_vma_finalize(&mig);

drivers/gpu/drm/amd/amdkfd/kfd_migrate.c

@@ -317,7 +317,6 @@ svm_migrate_copy_to_vram(struct amdgpu_device *adev, struct svm_range *prange,
 			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]);
-			migrate->dst[i] |= MIGRATE_PFN_LOCKED;
 			src[i] = dma_map_page(dev, spage, 0, PAGE_SIZE,
 					      DMA_TO_DEVICE);
 			r = dma_mapping_error(dev, src[i]);
@@ -610,7 +609,6 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 				     dst[i] >> PAGE_SHIFT, page_to_pfn(dpage));
 
 		migrate->dst[i] = migrate_pfn(page_to_pfn(dpage));
-		migrate->dst[i] |= MIGRATE_PFN_LOCKED;
 		j++;
 	}
 

drivers/gpu/drm/nouveau/nouveau_dmem.c

@@ -166,7 +166,7 @@ static vm_fault_t nouveau_dmem_fault_copy_one(struct nouveau_drm *drm,
 		goto error_dma_unmap;
 	mutex_unlock(&svmm->mutex);
 
-	args->dst[0] = migrate_pfn(page_to_pfn(dpage)) | MIGRATE_PFN_LOCKED;
+	args->dst[0] = migrate_pfn(page_to_pfn(dpage));
 	return 0;
 
 error_dma_unmap:
@@ -602,7 +602,7 @@ static unsigned long nouveau_dmem_migrate_copy_one(struct nouveau_drm *drm,
 		((paddr >> PAGE_SHIFT) << NVIF_VMM_PFNMAP_V0_ADDR_SHIFT);
 	if (src & MIGRATE_PFN_WRITE)
 		*pfn |= NVIF_VMM_PFNMAP_V0_W;
-	return migrate_pfn(page_to_pfn(dpage)) | MIGRATE_PFN_LOCKED;
+	return migrate_pfn(page_to_pfn(dpage));
 
 out_dma_unmap:
 	dma_unmap_page(dev, *dma_addr, PAGE_SIZE, DMA_BIDIRECTIONAL);

include/linux/migrate.h

@@ -110,7 +110,6 @@ static inline int migrate_misplaced_page(struct page *page,
  */
 #define MIGRATE_PFN_VALID	(1UL << 0)
 #define MIGRATE_PFN_MIGRATE	(1UL << 1)
-#define MIGRATE_PFN_LOCKED	(1UL << 2)
 #define MIGRATE_PFN_WRITE	(1UL << 3)
 #define MIGRATE_PFN_SHIFT	6
 

lib/test_hmm.c

@@ -613,8 +613,7 @@ static void dmirror_migrate_alloc_and_copy(struct migrate_vma *args,
 		 */
 		rpage->zone_device_data = dmirror;
 
-		*dst = migrate_pfn(page_to_pfn(dpage)) |
+		*dst = migrate_pfn(page_to_pfn(dpage));
-			    MIGRATE_PFN_LOCKED;
 		if ((*src & MIGRATE_PFN_WRITE) ||
 		    (!spage && args->vma->vm_flags & VM_WRITE))
 			*dst |= MIGRATE_PFN_WRITE;
@@ -1137,7 +1136,7 @@ static vm_fault_t dmirror_devmem_fault_alloc_and_copy(struct migrate_vma *args,
 		lock_page(dpage);
 		xa_erase(&dmirror->pt, addr >> PAGE_SHIFT);
 		copy_highpage(dpage, spage);
-		*dst = migrate_pfn(page_to_pfn(dpage)) | MIGRATE_PFN_LOCKED;
+		*dst = migrate_pfn(page_to_pfn(dpage));
 		if (*src & MIGRATE_PFN_WRITE)
 			*dst |= MIGRATE_PFN_WRITE;
 	}

mm/migrate.c

@@ -2362,7 +2362,6 @@ static int migrate_vma_collect_pmd(pmd_t *pmdp,
 		 * can't be dropped from it).
 		 */
 		get_page(page);
-		migrate->cpages++;
 
 		/*
 		 * Optimize for the common case where page is only mapped once
@@ -2372,7 +2371,7 @@ static int migrate_vma_collect_pmd(pmd_t *pmdp,
 		if (trylock_page(page)) {
 			pte_t swp_pte;
 
-			mpfn |= MIGRATE_PFN_LOCKED;
+			migrate->cpages++;
 			ptep_get_and_clear(mm, addr, ptep);
 
 			/* Setup special migration page table entry */
@@ -2406,6 +2405,9 @@ static int migrate_vma_collect_pmd(pmd_t *pmdp,
 
 			if (pte_present(pte))
 				unmapped++;
+		} else {
+			put_page(page);
+			mpfn = 0;
 		}
 
 next:
@@ -2510,15 +2512,17 @@ static bool migrate_vma_check_page(struct page *page)
 }
 
 /*
- * migrate_vma_prepare() - lock pages and isolate them from the lru
+ * migrate_vma_unmap() - replace page mapping with special migration pte entry
  * @migrate: migrate struct containing all migration information
  *
- * This locks pages that have been collected by migrate_vma_collect(). Once each
+ * Isolate pages from the LRU and replace mappings (CPU page table pte) with a
- * page is locked it is isolated from the lru (for non-device pages). Finally,
+ * special migration pte entry and check if it has been pinned. Pinned pages are
- * the ref taken by migrate_vma_collect() is dropped, as locked pages cannot be
+ * restored because we cannot migrate them.
- * migrated by concurrent kernel threads.
+ *
+ * This is the last step before we call the device driver callback to allocate
+ * destination memory and copy contents of original page over to new page.
  */
-static void migrate_vma_prepare(struct migrate_vma *migrate)
+static void migrate_vma_unmap(struct migrate_vma *migrate)
 {
 	const unsigned long npages = migrate->npages;
 	const unsigned long start = migrate->start;
@@ -2527,32 +2531,12 @@ static void migrate_vma_prepare(struct migrate_vma *migrate)
 
 	lru_add_drain();
 
-	for (i = 0; (i < npages) && migrate->cpages; i++) {
+	for (i = 0; i < npages; i++) {
 		struct page *page = migrate_pfn_to_page(migrate->src[i]);
-		bool remap = true;
 
 		if (!page)
 			continue;
 
-		if (!(migrate->src[i] & MIGRATE_PFN_LOCKED)) {
-			/*
-			 * Because we are migrating several pages there can be
-			 * a deadlock between 2 concurrent migration where each
-			 * are waiting on each other page lock.
-			 *
-			 * Make migrate_vma() a best effort thing and backoff
-			 * for any page we can not lock right away.
-			 */
-			if (!trylock_page(page)) {
-				migrate->src[i] = 0;
-				migrate->cpages--;
-				put_page(page);
-				continue;
-			}
-			remap = false;
-			migrate->src[i] |= MIGRATE_PFN_LOCKED;
-		}
-
 		/* ZONE_DEVICE pages are not on LRU */
 		if (!is_zone_device_page(page)) {
 			if (!PageLRU(page) && allow_drain) {
@@ -2562,16 +2546,9 @@ static void migrate_vma_prepare(struct migrate_vma *migrate)
 			}
 
 			if (isolate_lru_page(page)) {
-				if (remap) {
+				migrate->src[i] &= ~MIGRATE_PFN_MIGRATE;
-					migrate->src[i] &= ~MIGRATE_PFN_MIGRATE;
+				migrate->cpages--;
-					migrate->cpages--;
+				restore++;
-					restore++;
-				} else {
-					migrate->src[i] = 0;
-					unlock_page(page);
-					migrate->cpages--;
-					put_page(page);
-				}
 				continue;
 			}
 
@@ -2579,80 +2556,20 @@ static void migrate_vma_prepare(struct migrate_vma *migrate)
 			put_page(page);
 		}
 
-		if (!migrate_vma_check_page(page)) {
+		if (page_mapped(page))
-			if (remap) {
-				migrate->src[i] &= ~MIGRATE_PFN_MIGRATE;
-				migrate->cpages--;
-				restore++;
-
-				if (!is_zone_device_page(page)) {
-					get_page(page);
-					putback_lru_page(page);
-				}
-			} else {
-				migrate->src[i] = 0;
-				unlock_page(page);
-				migrate->cpages--;
-
-				if (!is_zone_device_page(page))
-					putback_lru_page(page);
-				else
-					put_page(page);
-			}
-		}
-	}
-
-	for (i = 0, addr = start; i < npages && restore; i++, addr += PAGE_SIZE) {
-		struct page *page = migrate_pfn_to_page(migrate->src[i]);
-
-		if (!page || (migrate->src[i] & MIGRATE_PFN_MIGRATE))
-			continue;
-
-		remove_migration_pte(page, migrate->vma, addr, page);
-
-		migrate->src[i] = 0;
-		unlock_page(page);
-		put_page(page);
-		restore--;
-	}
-}
-
-/*
- * migrate_vma_unmap() - replace page mapping with special migration pte entry
- * @migrate: migrate struct containing all migration information
- *
- * Replace page mapping (CPU page table pte) with a special migration pte entry
- * and check again if it has been pinned. Pinned pages are restored because we
- * cannot migrate them.
- *
- * This is the last step before we call the device driver callback to allocate
- * destination memory and copy contents of original page over to new page.
- */
-static void migrate_vma_unmap(struct migrate_vma *migrate)
-{
-	const unsigned long npages = migrate->npages;
-	const unsigned long start = migrate->start;
-	unsigned long addr, i, restore = 0;
-
-	for (i = 0; i < npages; i++) {
-		struct page *page = migrate_pfn_to_page(migrate->src[i]);
-
-		if (!page || !(migrate->src[i] & MIGRATE_PFN_MIGRATE))
-			continue;
-
-		if (page_mapped(page)) {
 			try_to_migrate(page, 0);
-			if (page_mapped(page))
-				goto restore;
-		}
 
-		if (migrate_vma_check_page(page))
+		if (page_mapped(page) || !migrate_vma_check_page(page)) {
+			if (!is_zone_device_page(page)) {
+				get_page(page);
+				putback_lru_page(page);
+			}
+
+			migrate->src[i] &= ~MIGRATE_PFN_MIGRATE;
+			migrate->cpages--;
+			restore++;
 			continue;
-
+		}
-restore:
-		migrate->src[i] &= ~MIGRATE_PFN_MIGRATE;
-		migrate->cpages--;
-		restore++;
 	}
 
 	for (addr = start, i = 0; i < npages && restore; addr += PAGE_SIZE, i++) {
@@ -2665,12 +2582,8 @@ static void migrate_vma_unmap(struct migrate_vma *migrate)
 
 		migrate->src[i] = 0;
 		unlock_page(page);
+		put_page(page);
 		restore--;
-
-		if (is_zone_device_page(page))
-			put_page(page);
-		else
-			putback_lru_page(page);
 	}
 }
 
@@ -2693,8 +2606,8 @@ static void migrate_vma_unmap(struct migrate_vma *migrate)
  * it for all those entries (ie with MIGRATE_PFN_VALID and MIGRATE_PFN_MIGRATE
  * flag set).  Once these are allocated and copied, the caller must update each
  * corresponding entry in the dst array with the pfn value of the destination
- * page and with the MIGRATE_PFN_VALID and MIGRATE_PFN_LOCKED flags set
+ * page and with MIGRATE_PFN_VALID. Destination pages must be locked via
- * (destination pages must have their struct pages locked, via lock_page()).
+ * lock_page().
  *
  * Note that the caller does not have to migrate all the pages that are marked
  * with MIGRATE_PFN_MIGRATE flag in src array unless this is a migration from
@@ -2763,8 +2676,6 @@ int migrate_vma_setup(struct migrate_vma *args)
 
 	migrate_vma_collect(args);
 
-	if (args->cpages)
-		migrate_vma_prepare(args);
 	if (args->cpages)
 		migrate_vma_unmap(args);