Re: 2.6.3-mm3

[Mike Fedyk]

Mike Fedyk wrote:
> Nick Piggin wrote:
>
> > IMO, shrink_slab-for-all-zones.patch and zone-balancing-fix.patch
> > should be all you need although they won't shrink the slab as
> > much as mm3. They should be pretty easy to port by hand.
>
>
> How does this patch for 2.6.3 look?

Never mind, I was brain dead at the time.

This one should be better.

Mike
--- linux-2.6.3/mm/vmscan.c	2004-02-25 20:50:35.000000000 -0800
+++ linux-2.6.3-slab-lofft/mm/vmscan.c	2004-02-25 20:41:45.000000000 -0800
@@ -885,12 +885,10 @@
 
 		/* Take a nap, wait for some writeback to complete */
 		blk_congestion_wait(WRITE, HZ/10);
-		if (zones[0] - zones[0]->zone_pgdat->node_zones < ZONE_HIGHMEM) {
-			shrink_slab(total_scanned, gfp_mask);
-			if (reclaim_state) {
-				nr_reclaimed += reclaim_state->reclaimed_slab;
-				reclaim_state->reclaimed_slab = 0;
-			}
+		shrink_slab(total_scanned, gfp_mask);
+		if (reclaim_state) {
+			nr_reclaimed += reclaim_state->reclaimed_slab;
+			reclaim_state->reclaimed_slab = 0;
 		}
 	}
 	if ((gfp_mask & __GFP_FS) && !(gfp_mask & __GFP_NORETRY))
@@ -962,11 +960,9 @@
 				max_scan = SWAP_CLUSTER_MAX;
 			to_free -= shrink_zone(zone, max_scan, GFP_KERNEL,
 					to_reclaim, &nr_mapped, ps, priority);
-			if (i < ZONE_HIGHMEM) {
-				reclaim_state->reclaimed_slab = 0;
-				shrink_slab(max_scan + nr_mapped, GFP_KERNEL);
-				to_free -= reclaim_state->reclaimed_slab;
-			}
+			reclaim_state->reclaimed_slab = 0;
+			shrink_slab(max_scan + nr_mapped, GFP_KERNEL);
+			to_free -= reclaim_state->reclaimed_slab;
 			if (zone->all_unreclaimable)
 				continue;
 			if (zone->pages_scanned > zone->present_pages * 2)