[Intel-gfx] [PATCH 06/13] mm, notifier: Catch sleeping/blocking for !blockable

Daniel Vetter daniel.vetter at ffwll.ch
Wed Nov 7 15:30:12 UTC 2018

We need to make sure implementations don't cheat and don't have a
possible schedule/blocking point deeply burried where review can't
catch it.

I'm not sure whether this is the best way to make sure all the
might_sleep() callsites trigger, and it's a bit ugly in the code flow.
But it gets the job done.

Cc: Andrew Morton <akpm at linux-foundation.org>
Cc: Michal Hocko <mhocko at suse.com>
Cc: David Rientjes <rientjes at google.com>
Cc: "Christian König" <christian.koenig at amd.com>
Cc: Daniel Vetter <daniel.vetter at ffwll.ch>
Cc: "Jérôme Glisse" <jglisse at redhat.com>
Cc: linux-mm at kvack.org
Signed-off-by: Daniel Vetter <daniel.vetter at intel.com>
 mm/mmu_notifier.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/mm/mmu_notifier.c b/mm/mmu_notifier.c
index 59e102589a25..4d282cfb296e 100644
--- a/mm/mmu_notifier.c
+++ b/mm/mmu_notifier.c
@@ -185,7 +185,13 @@ int __mmu_notifier_invalidate_range_start(struct mm_struct *mm,
 	id = srcu_read_lock(&srcu);
 	hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) {
 		if (mn->ops->invalidate_range_start) {
-			int _ret = mn->ops->invalidate_range_start(mn, mm, start, end, blockable);
+			int _ret;
+				preempt_disable();
+			_ret = mn->ops->invalidate_range_start(mn, mm, start, end, blockable);
+				preempt_enable();
 			if (_ret) {
 				pr_info("%pS callback failed with %d in %sblockable context.\n",
 						mn->ops->invalidate_range_start, _ret,

More information about the Intel-gfx mailing list