mm: warn about illegal __GFP_NOFAIL usage in a more appropriate location and manner

Three points for this change:

1. We should consolidate all warnings in one place. Currently, the
   order > 1 warning is in the hotpath, while others are in less
   likely scenarios. Moving all warnings to the slowpath will reduce
   the overhead for order > 1 and increase the visibility of other
   warnings.

2. We currently have two warnings for order: one for order > 1 in
   the hotpath and another for order > costly_order in the laziest
   path. I suggest standardizing on order > 1 since it's been in
   use for a long time.

3. We don't need to check for __GFP_NOWARN in this case. __GFP_NOWARN
   is meant to suppress allocation failure reports, but here we're
   dealing with bug detection, not allocation failures. So replace
   WARN_ON_ONCE_GFP by WARN_ON_ONCE.

[[email protected]: also update the doc for __GFP_NOFAIL with order > 1]
  Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Barry Song <[email protected]>
Suggested-by: Vlastimil Babka <[email protected]>
Reviewed-by: Vlastimil Babka <[email protected]>
Acked-by: David Hildenbrand <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Christoph Lameter <[email protected]>
Cc: Davidlohr Bueso <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: "Eugenio Pérez" <[email protected]>
Cc: Hailong.Liu <[email protected]>
Cc: Hyeonggon Yoo <[email protected]>
Cc: Jason Wang <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Kees Cook <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: Lorenzo Stoakes <[email protected]>
Cc: Maxime Coquelin <[email protected]>
Cc: "Michael S. Tsirkin" <[email protected]>
Cc: Pekka Enberg <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Uladzislau Rezki (Sony) <[email protected]>
Cc: Xie Yongji <[email protected]>
Cc: Xuan Zhuo <[email protected]>
Cc: Yafang Shao <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

2 files changed, 27 insertions, 26 deletions

diff --git a/include/linux/gfp_types.h b/include/linux/gfp_types.h
index 4a1fa7706b0c..65db9349f905 100644
--- a/include/linux/gfp_types.h
+++ b/include/linux/gfp_types.h
@@ -253,7 +253,8 @@ enum {
  * used only when there is no reasonable failure policy) but it is
  * definitely preferable to use the flag rather than opencode endless
  * loop around allocator.
- * Using this flag for costly allocations is _highly_ discouraged.
+ * Allocating pages from the buddy with __GFP_NOFAIL and order > 1 is
+ * not supported. Please consider using kvmalloc() instead.
  */
 #define __GFP_IO	((__force gfp_t)___GFP_IO)
 #define __GFP_FS	((__force gfp_t)___GFP_FS)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 3469ca78bf7a..23509b823e2b 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3032,12 +3032,6 @@ struct page *rmqueue(struct zone *preferred_zone,
 {
 	struct page *page;
 
-	/*
-	 * We most definitely don't want callers attempting to
-	 * allocate greater than order-1 page units with __GFP_NOFAIL.
-	 */
-	WARN_ON_ONCE((gfp_flags & __GFP_NOFAIL) && (order > 1));
-
 	if (likely(pcp_allowed_order(order))) {
 		page = rmqueue_pcplist(preferred_zone, zone, order,
 				       migratetype, alloc_flags);
@@ -4175,6 +4169,7 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
 {
 	bool can_direct_reclaim = gfp_mask & __GFP_DIRECT_RECLAIM;
 	bool can_compact = gfp_compaction_allowed(gfp_mask);
+	bool nofail = gfp_mask & __GFP_NOFAIL;
 	const bool costly_order = order > PAGE_ALLOC_COSTLY_ORDER;
 	struct page *page = NULL;
 	unsigned int alloc_flags;
@@ -4187,6 +4182,25 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
 	unsigned int zonelist_iter_cookie;
 	int reserve_flags;
 
+	if (unlikely(nofail)) {
+		/*
+		 * We most definitely don't want callers attempting to
+		 * allocate greater than order-1 page units with __GFP_NOFAIL.
+		 */
+		WARN_ON_ONCE(order > 1);
+		/*
+		 * Also we don't support __GFP_NOFAIL without __GFP_DIRECT_RECLAIM,
+		 * otherwise, we may result in lockup.
+		 */
+		WARN_ON_ONCE(!can_direct_reclaim);
+		/*
+		 * PF_MEMALLOC request from this context is rather bizarre
+		 * because we cannot reclaim anything and only can loop waiting
+		 * for somebody to do a work for us.
+		 */
+		WARN_ON_ONCE(current->flags & PF_MEMALLOC);
+	}
+
 restart:
 	compaction_retries = 0;
 	no_progress_loops = 0;
@@ -4404,30 +4418,16 @@ nopage:
 	 * Make sure that __GFP_NOFAIL request doesn't leak out and make sure
 	 * we always retry
 	 */
-	if (gfp_mask & __GFP_NOFAIL) {
+	if (unlikely(nofail)) {
 		/*
-		 * All existing users of the __GFP_NOFAIL are blockable, so warn
-		 * of any new users that actually require GFP_NOWAIT
+		 * Lacking direct_reclaim we can't do anything to reclaim memory,
+		 * we disregard these unreasonable nofail requests and still
+		 * return NULL
 		 */
-		if (WARN_ON_ONCE_GFP(!can_direct_reclaim, gfp_mask))
+		if (!can_direct_reclaim)
 			goto fail;
 
 		/*
-		 * PF_MEMALLOC request from this context is rather bizarre
-		 * because we cannot reclaim anything and only can loop waiting
-		 * for somebody to do a work for us
-		 */
-		WARN_ON_ONCE_GFP(current->flags & PF_MEMALLOC, gfp_mask);
-
-		/*
-		 * non failing costly orders are a hard requirement which we
-		 * are not prepared for much so let's warn about these users
-		 * so that we can identify them and convert them to something
-		 * else.
-		 */
-		WARN_ON_ONCE_GFP(costly_order, gfp_mask);
-
-		/*
 		 * Help non-failing allocations by giving some access to memory
 		 * reserves normally used for high priority non-blocking
 		 * allocations but do not use ALLOC_NO_WATERMARKS because this