[PATCH v2] mm/migrate_device: Try to handle swapcache pages

[PATCH v2] mm/migrate_device: Try to handle swapcache pages

[mpenttil]

From: Mika Penttilä <mpenttil@redhat.com>

Migrating file pages and swapcache pages into device memory is not supported.
The decision is done based on page_mapping(). For now, swapcache pages are not migrated.

Things can however be improved, for swapcache pages. Try to get rid of the swap cache,
and if successful, go ahead as with other anonymous pages.

Cc: Alistair Popple <apopple@nvidia.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: Huang Ying <ying.huang@intel.com>
Signed-off-by: Mika Penttilä <mpenttil@redhat.com>
---

v2:
  - use folio_test_anon() (Huang, Ying)

 
 mm/migrate_device.c | 24 ++++++++++++++++++------
 1 file changed, 18 insertions(+), 6 deletions(-)

diff --git a/mm/migrate_device.c b/mm/migrate_device.c
index d30c9de60b0d..829bbc526758 100644
--- a/mm/migrate_device.c
+++ b/mm/migrate_device.c
@@ -747,13 +747,25 @@ static void __migrate_device_pages(unsigned long *src_pfns,
 
 		if (is_device_private_page(newpage) ||
 		    is_device_coherent_page(newpage)) {
-			/*
-			 * For now only support anonymous memory migrating to
-			 * device private or coherent memory.
-			 */
+
 			if (mapping) {
-				src_pfns[i] &= ~MIGRATE_PFN_MIGRATE;
-				continue;
+				struct folio *folio;
+
+				folio = page_folio(page);
+
+				/*
+				 * For now only support anonymous memory migrating to
+				 * device private or coherent memory.
+				 *
+				 * Try to get rid of swap cache if possible.
+				 * page is here file or swapcache page, could be shmem also
+				 * folio_test_anon() filters out file and shmem
+				 *
+				 */
+				if (!folio_test_anon(folio) || !folio_free_swap(folio)) {
+					src_pfns[i] &= ~MIGRATE_PFN_MIGRATE;
+					continue;
+				}
 			}
 		} else if (is_zone_device_page(newpage)) {
 			/*
-- 
2.17.1

Re: [PATCH v2] mm/migrate_device: Try to handle swapcache pages

[Huang, Ying]

mpenttil@redhat.com writes:

> From: Mika Penttilä <mpenttil@redhat.com>
>
> Migrating file pages and swapcache pages into device memory is not supported.
> The decision is done based on page_mapping(). For now, swapcache pages are not migrated.
>
> Things can however be improved, for swapcache pages. Try to get rid of the swap cache,
> and if successful, go ahead as with other anonymous pages.
>
> Cc: Alistair Popple <apopple@nvidia.com>
> Cc: John Hubbard <jhubbard@nvidia.com>
> Cc: Ralph Campbell <rcampbell@nvidia.com>
> Cc: Huang Ying <ying.huang@intel.com>
> Signed-off-by: Mika Penttilä <mpenttil@redhat.com>
> ---
>
> v2:
>   - use folio_test_anon() (Huang, Ying)
>
>  
>  mm/migrate_device.c | 24 ++++++++++++++++++------
>  1 file changed, 18 insertions(+), 6 deletions(-)
>
> diff --git a/mm/migrate_device.c b/mm/migrate_device.c
> index d30c9de60b0d..829bbc526758 100644
> --- a/mm/migrate_device.c
> +++ b/mm/migrate_device.c
> @@ -747,13 +747,25 @@ static void __migrate_device_pages(unsigned long *src_pfns,
>  
>  		if (is_device_private_page(newpage) ||
>  		    is_device_coherent_page(newpage)) {
> -			/*
> -			 * For now only support anonymous memory migrating to
> -			 * device private or coherent memory.
> -			 */
> +
>  			if (mapping) {
> -				src_pfns[i] &= ~MIGRATE_PFN_MIGRATE;
> -				continue;
> +				struct folio *folio;
> +
> +				folio = page_folio(page);
> +
> +				/*
> +				 * For now only support anonymous memory migrating to
> +				 * device private or coherent memory.
> +				 *
> +				 * Try to get rid of swap cache if possible.

I think we can delete the following 2 lines of comments.  They add
nothing except what code says already.

Otherwise looks good to me.

Reviewed-by: "Huang, Ying" <ying.huang@intel.com>

> +				 * page is here file or swapcache page, could be shmem also
> +				 * folio_test_anon() filters out file and shmem
> +				 *
> +				 */
> +				if (!folio_test_anon(folio) || !folio_free_swap(folio)) {
> +					src_pfns[i] &= ~MIGRATE_PFN_MIGRATE;
> +					continue;
> +				}
>  			}
>  		} else if (is_zone_device_page(newpage)) {
>  			/*