On 03/30/2017 03:00 PM, Thomas Hellstrom wrote:

quoted

quoted

 
 	if (unlikely(nr_lazy > lazy_max_pages()))
-		try_purge_vmap_area_lazy();

Perhaps a slight optimization would be to schedule work iff
!mutex_locked(&vmap_purge_lock) below?

Makes sense, we don't need to spawn workers if we already purging.



From: Andrey Ryabinin <redacted>
Subject: mm/vmalloc: allow to call vfree() in atomic context fix

Don't spawn worker if we already purging.

Signed-off-by: Andrey Ryabinin <redacted>
---
 mm/vmalloc.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index ea1b4ab..88168b8 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c

@@ -737,7 +737,8 @@ static void free_vmap_area_noflush(struct vmap_area *va)
 	/* After this point, we may free va at any time */
 	llist_add(&va->purge_list, &vmap_purge_list);
 
-	if (unlikely(nr_lazy > lazy_max_pages()))
+	if (unlikely(nr_lazy > lazy_max_pages()) &&
+	    !mutex_is_locked(&vmap_purge_lock))
 		schedule_work(&purge_vmap_work);
 }