On Alan's advice, rather than do a wholesale conversion on a single

architecture from page queue lock to a hashed array of page locks
(based on a patch by Jeff Roberson), I've implemented page lock
support in the MI code and have only moved vm_page's hold_count
out from under page queue mutex to page lock. This changes
pmap_extract_and_hold on all pmaps.

Supported by: Bitgravity Inc.

Discussed with: alc, jeffr, and kib

1 files changed, 6 insertions, 3 deletions

diff --git a/sys/vm/sg_pager.c b/sys/vm/sg_pager.c
index a17fe82..845401a 100644
--- a/sys/vm/sg_pager.c
+++ b/sys/vm/sg_pager.c
@@ -198,10 +198,13 @@ sg_pager_getpages(vm_object_t object, vm_page_t *m, int count, int reqpage)
 	TAILQ_INSERT_TAIL(&object->un_pager.sgp.sgp_pglist, page, pageq);
 
 	/* Free the original pages and insert this fake page into the object. */
-	vm_page_lock_queues();
-	for (i = 0; i < count; i++)
+	for (i = 0; i < count; i++) {
+		vm_page_lock(m[i]);
+		vm_page_lock_queues();
 		vm_page_free(m[i]);
-	vm_page_unlock_queues();
+		vm_page_unlock_queues();
+		vm_page_unlock(m[i]);
+	}
 	vm_page_insert(page, object, offset);
 	m[reqpage] = page;
 	page->valid = VM_PAGE_BITS_ALL;