From: Richard Yao <ryao@cs.stonybrook.edu>
Date: Thu, 21 Mar 2013 17:21:11 +0000 (-0400)
Subject: Do not call cond_resched() in spl_slab_reclaim()
X-Git-Tag: zfs-0.8.0-rc1~152^2~296
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=feaf1e321da97ab0aaa60edc121fb6aa2d03952d;p=zfs

Do not call cond_resched() in spl_slab_reclaim()

Calling cond_resched() after each object is freed and then after each
slab is freed can cause slabs of objects to live for excessive periods
of time following reclaimation. This interferes with the kernel's own
memory management when called from kswapd and can cause direct reclaim
to occur in response to memory pressure that should have been resolved.

Signed-off-by: Richard Yao <ryao@cs.stonybrook.edu>
---

diff --git a/module/spl/spl-kmem.c b/module/spl/spl-kmem.c
index f9c111491..a0ca2d2be 100644
--- a/module/spl/spl-kmem.c
+++ b/module/spl/spl-kmem.c
@@ -1112,14 +1112,11 @@ spl_slab_reclaim(spl_kmem_cache_t *skc, int count, int flag)
 
 		if (skc->skc_flags & KMC_OFFSLAB)
 			kv_free(skc, sko->sko_addr, size);
-
-		cond_resched();
 	}
 
 	list_for_each_entry_safe(sks, m, &sks_list, sks_list) {
 		ASSERT(sks->sks_magic == SKS_MAGIC);
 		kv_free(skc, sks, skc->skc_slab_size);
-		cond_resched();
 	}
 
 	SEXIT;