[PATCH] shrinker->nr = LONG_MAX means deadlock for icache

With Andrew Morton <[e-mail]>

The slab scanning code tries to balance the scanning rate of slabs versus the
scanning rate of LRU pages. To do this, it retains state concerning how many
slabs have been scanned - if a particular slab shrinker didn't scan enough
objects, we remember that for next time, and scan more objects on the next
pass.

The problem with this is that with (say) a huge number of GFP_NOIO
direct-reclaim attempts, the number of objects which are to be scanned when we
finally get a GFP_KERNEL request can be huge. Because some shrinker handlers
just bail out if !__GFP_FS.

So the patch clamps the number of objects-to-be-scanned to 2* the total number
of objects in the slab cache.

Signed-off-by: Andrea Arcangeli <[e-mail]>
Signed-off-by: Andrew Morton <[e-mail]>
Signed-off-by: Linus Torvalds <[e-mail]>

mm/vmscan.c

···
 	list_for_each_entry(shrinker, &shrinker_list, list) {
 		unsigned long long delta;
 		unsigned long total_scan;
+		unsigned long max_pass = (*shrinker->shrinker)(0, gfp_mask);
 
 		delta = (4 * scanned) / shrinker->seeks;
-		delta *= (*shrinker->shrinker)(0, gfp_mask);
+		delta *= max_pass;
 		do_div(delta, lru_pages + 1);
 		shrinker->nr += delta;
-		if (shrinker->nr < 0)
-			shrinker->nr = LONG_MAX;	/* It wrapped! */
+		if (shrinker->nr < 0) {
+			printk(KERN_ERR "%s: nr=%ld\n",
+					__FUNCTION__, shrinker->nr);
+			shrinker->nr = max_pass;
+		}
+
+		/*
+		 * Avoid risking looping forever due to too large nr value:
+		 * never try to free more than twice the estimate number of
+		 * freeable entries.
+		 */
+		if (shrinker->nr > max_pass * 2)
+			shrinker->nr = max_pass * 2;
 
 		total_scan = shrinker->nr;
 		shrinker->nr = 0;