Am 2021-11-21 um 9:40 p.m. schrieb Alistair Popple:

quoted

quoted

quoted

diff --git a/mm/migrate.c b/mm/migrate.c
index 1852d787e6ab..f74422a42192 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c

@@ -362,7 +362,7 @@ static int expected_page_refs(struct address_space *mapping, struct page *page)
 	 * Device private pages have an extra refcount as they are
 	 * ZONE_DEVICE pages.
 	 */
-	expected_count += is_device_private_page(page);
+	expected_count += is_device_page(page);
 	if (mapping)
 		expected_count += thp_nr_pages(page) + page_has_private(page);
 

@@ -2503,7 +2503,7 @@ 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.
 		 */

Note that I have posted a patch to fix this - see
https://lore.kernel.org/all/20211118020754.954425-1-apopple@nvidia.com/ (local) This
looks ok for now assuming coherent pages can never be pinned.

However that raises a question - what happens when something calls
get_user_pages() on a pfn pointing to a coherent device page? I can't see
anything in this series that prevents pinning of coherent device pages, so we
can't just assume they aren't pinned.

I agree. I think we need to depend on your patch to go in first.

I'm also wondering if we need to do something to prevent get_user_pages
from pinning device pages. And by "pin", I think migrate_vma_check_page
is not talking about FOLL_PIN, but any get_user_pages call. As far as I
can tell, there should be nothing fundamentally wrong with pinning
device pages for a short time. But I think we'll want to avoid
FOLL_LONGTERM because that would affect our memory manager's ability to
evict device memory.

Right, so long as my fix goes in I don't think there is anything wrong with
pinning device public pages. Agree that we should avoid FOLL_LONGTERM pins for
device memory though. I think the way to do that is update is_pinnable_page()
so we treat device pages the same as other unpinnable pages ie. long-term pins
will migrate the page.

I'm trying to understand check_and_migrate_movable_pages in gup.c. It
doesn't look like the right way to migrate device pages. We may have to
do something different there as well. So instead of changing
is_pinnable_page, it maybe better to explicitly check for is_device_page
or is_device_coherent_page in check_and_migrate_movable_pages to migrate
it correctly, or just fail outright.

Thanks,
  Felix

quoted

quoted

In the case of device-private pages this is enforced by the fact they never
have present pte's, so any attempt to GUP them results in a fault. But if I'm
understanding this series correctly that won't be the case for coherent device
pages right?

Right.

Regards,
  Felix

quoted

quoted

-		return is_device_private_page(page);
+		return is_device_page(page);
 	}
 
 	/* For file back page */

@@ -2791,7 +2791,7 @@ 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 coherent page.
  */
 static void migrate_vma_insert_page(struct migrate_vma *migrate,
 				    unsigned long addr,

@@ -2867,10 +2867,15 @@ static void migrate_vma_insert_page(struct migrate_vma *migrate,
 				swp_entry = make_readable_device_private_entry(
 							page_to_pfn(page));
 			entry = swp_entry_to_pte(swp_entry);
+		} else if (is_device_page(page)) {

How about adding an explicit `is_device_coherent_page()` helper? It would make
the test more explicit that this is expected to handle just coherent pages and
I bet there will be future changes that need to differentiate between private
and coherent pages anyway.

quoted

+			entry = pte_mkold(mk_pte(page,
+						 READ_ONCE(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 coherent types
+			 * for device zone memory.
 			 */
 			pr_warn_once("Unsupported ZONE_DEVICE page type.\n");
 			goto abort;

@@ -2976,10 +2981,10 @@ 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 coherent
+				 * anonymous when migrating to device memory.
 				 */
 				if (mapping) {
 					migrate->src[i] &= ~MIGRATE_PFN_MIGRATE;