diff mbox series

[v6,07/13] mm: add generic type support to migrate_vma helpers

Message ID 20210813063150.2938-8-alex.sierra@amd.com (mailing list archive)
State New
Headers show
Series Support DEVICE_GENERIC memory in migrate_vma_* | expand

Commit Message

Sierra Guiza, Alejandro (Alex) Aug. 13, 2021, 6:31 a.m. UTC 
Device generic type case added for migrate_vma_pages and
migrate_vma_check_page helpers.
Both, generic and private device types have the same
conditions to decide to migrate pages from/to device
memory.

Signed-off-by: Alex Sierra <alex.sierra@amd.com>
---
 mm/migrate.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

Comments

Christoph Hellwig Aug. 15, 2021, 9:19 a.m. UTC | #1 
On Fri, Aug 13, 2021 at 01:31:44AM -0500, Alex Sierra wrote:
> Device generic type case added for migrate_vma_pages and
> migrate_vma_check_page helpers.
> Both, generic and private device types have the same
> conditions to decide to migrate pages from/to device
> memory.

This reas a little weird mostly because it doesn't use up the line
length nicely:

Add the device generic type case to the migrate_vma_pages and
migrate_vma_check_page helpers.  This new case is handled identically
to the existing device private case.

> +			 * We support migrating to private and generic types for device
> +			 * zone memory.

Don't spill comments over 80 characters.

Otherwise looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>
diff mbox series

Patch

diff --git a/mm/migrate.c b/mm/migrate.c
index e3a10e2a1bb3..f9e6bfa2867c 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -2565,7 +2565,7 @@  static bool migrate_vma_check_page(struct page *page)
 		 * FIXME proper solution is to rework migration_entry_wait() so
 		 * it does not need to take a reference on page.
 		 */
-		return is_device_private_page(page);
+		return is_device_page(page);
 	}
 
 	/* For file back page */
@@ -2854,7 +2854,7 @@  EXPORT_SYMBOL(migrate_vma_setup);
  *     handle_pte_fault()
  *       do_anonymous_page()
  * to map in an anonymous zero page but the struct page will be a ZONE_DEVICE
- * private page.
+ * private or generic page.
  */
 static void migrate_vma_insert_page(struct migrate_vma *migrate,
 				    unsigned long addr,
@@ -2925,10 +2925,14 @@  static void migrate_vma_insert_page(struct migrate_vma *migrate,
 
 			swp_entry = make_device_private_entry(page, vma->vm_flags & VM_WRITE);
 			entry = swp_entry_to_pte(swp_entry);
+		} else if (is_device_page(page)) {
+			entry = mk_pte(page, vma->vm_page_prot);
+			if (vma->vm_flags & VM_WRITE)
+				entry = pte_mkwrite(pte_mkdirty(entry));
 		} else {
 			/*
-			 * For now we only support migrating to un-addressable
-			 * device memory.
+			 * We support migrating to private and generic types for device
+			 * zone memory.
 			 */
 			pr_warn_once("Unsupported ZONE_DEVICE page type.\n");
 			goto abort;
@@ -3034,10 +3038,10 @@  void migrate_vma_pages(struct migrate_vma *migrate)
 		mapping = page_mapping(page);
 
 		if (is_zone_device_page(newpage)) {
-			if (is_device_private_page(newpage)) {
+			if (is_device_page(newpage)) {
 				/*
-				 * For now only support private anonymous when
-				 * migrating to un-addressable device memory.
+				 * For now only support private and generic
+				 * anonymous when migrating to device memory.
 				 */
 				if (mapping) {
 					migrate->src[i] &= ~MIGRATE_PFN_MIGRATE;