From: Steffen Klassert <steffen.klassert@secunet.com>
Date: Thu, 29 Apr 2010 12:37:32 +0000 (+0200)
Subject: padata: Dont scale the parallel objects with the cpus
X-Git-Tag: v2.6.35-rc1~446^2~24
X-Git-Url: http://ftp.safe.ca/?p=safe%2Fjmp%2Flinux-2.6;a=commitdiff_plain;h=97e3d94aac1c3e95bd04d1b186479a4df3663ab8

padata: Dont scale the parallel objects with the cpus

Scaling the maximum number of objects in the parallel
codepath can lead to out of memory problems on bigsmp
machines.

Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---

diff --git a/kernel/padata.c b/kernel/padata.c
index 0282478..5085046 100644
--- a/kernel/padata.c
+++ b/kernel/padata.c
@@ -28,7 +28,7 @@
 #include <linux/rcupdate.h>
 
 #define MAX_SEQ_NR INT_MAX - NR_CPUS
-#define MAX_OBJ_NUM 10000 * NR_CPUS
+#define MAX_OBJ_NUM 1000
 
 static int padata_index_to_cpu(struct parallel_data *pd, int cpu_index)
 {