diff mbox series

[197/262] mm/rmap.c: avoid double faults migrating device private pages

Message ID 20211105204500.E7gZ9uZtc%akpm@linux-foundation.org (mailing list archive)
State New
Headers show
Series [001/262] scripts/spelling.txt: add more spellings to spelling.txt | expand

Commit Message

Andrew Morton Nov. 5, 2021, 8:45 p.m. UTC
From: Alistair Popple <apopple@nvidia.com>
Subject: mm/rmap.c: avoid double faults migrating device private pages

During migration special page table entries are installed for each page
being migrated.  These entries store the pfn and associated permissions of
ptes mapping the page being migarted.

Device-private pages use special swap pte entries to distinguish read-only
vs.  writeable pages which the migration code checks when creating
migration entries.  Normally this follows a fast path in
migrate_vma_collect_pmd() which correctly copies the permissions of
device-private pages over to migration entries when migrating pages back
to the CPU.

However the slow-path falls back to using try_to_migrate() which
unconditionally creates read-only migration entries for device-private
pages.  This leads to unnecessary double faults on the CPU as the new
pages are always mapped read-only even when they could be mapped
writeable.  Fix this by correctly copying device-private permissions in
try_to_migrate_one().

Link: https://lkml.kernel.org/r/20211018045247.3128058-1-apopple@nvidia.com
Signed-off-by: Alistair Popple <apopple@nvidia.com>
Reported-by: Ralph Campbell <rcampbell@nvidia.com>
Reviewed-by: John Hubbard <jhubbard@nvidia.com>
Cc: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 mm/rmap.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)
diff mbox series

Patch

--- a/mm/rmap.c~mm-rmapc-avoid-double-faults-migrating-device-private-pages
+++ a/mm/rmap.c
@@ -1807,6 +1807,7 @@  static bool try_to_migrate_one(struct pa
 		update_hiwater_rss(mm);
 
 		if (is_zone_device_page(page)) {
+			unsigned long pfn = page_to_pfn(page);
 			swp_entry_t entry;
 			pte_t swp_pte;
 
@@ -1815,8 +1816,11 @@  static bool try_to_migrate_one(struct pa
 			 * pte. do_swap_page() will wait until the migration
 			 * pte is removed and then restart fault handling.
 			 */
-			entry = make_readable_migration_entry(
-							page_to_pfn(page));
+			entry = pte_to_swp_entry(pteval);
+			if (is_writable_device_private_entry(entry))
+				entry = make_writable_migration_entry(pfn);
+			else
+				entry = make_readable_migration_entry(pfn);
 			swp_pte = swp_entry_to_pte(entry);
 
 			/*