]> git.openfabrics.org - ~shefty/rdma-dev.git/commitdiff
slub: document setting min order with debug_guardpage_minorder > 0
authorStanislaw Gruszka <sgruszka@redhat.com>
Fri, 13 Jan 2012 01:17:39 +0000 (17:17 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Fri, 13 Jan 2012 04:13:04 +0000 (20:13 -0800)
Acked-by: David Rientjes <rientjes@google.com>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Documentation/ABI/testing/sysfs-kernel-slab
Documentation/vm/slub.txt

index 8b093f8222d318e411113735f7e841c4b52e7ae0..91bd6ca5440f32dbe724c994728a2bee18060692 100644 (file)
@@ -346,6 +346,10 @@ Description:
                number of objects per slab.  If a slab cannot be allocated
                because of fragmentation, SLUB will retry with the minimum order
                possible depending on its characteristics.
                number of objects per slab.  If a slab cannot be allocated
                because of fragmentation, SLUB will retry with the minimum order
                possible depending on its characteristics.
+               When debug_guardpage_minorder=N (N > 0) parameter is specified
+               (see Documentation/kernel-parameters.txt), the minimum possible
+               order is used and this sysfs entry can not be used to change
+               the order at run time.
 
 What:          /sys/kernel/slab/cache/order_fallback
 Date:          April 2008
 
 What:          /sys/kernel/slab/cache/order_fallback
 Date:          April 2008
index 2acdda9601b097958ad4ea732ba784a1ed4fffce..6752870c4970d73721c48a041b15aa36af929ab5 100644 (file)
@@ -131,7 +131,10 @@ slub_min_objects.
 slub_max_order specified the order at which slub_min_objects should no
 longer be checked. This is useful to avoid SLUB trying to generate
 super large order pages to fit slub_min_objects of a slab cache with
 slub_max_order specified the order at which slub_min_objects should no
 longer be checked. This is useful to avoid SLUB trying to generate
 super large order pages to fit slub_min_objects of a slab cache with
-large object sizes into one high order page.
+large object sizes into one high order page. Setting command line
+parameter debug_guardpage_minorder=N (N > 0), forces setting
+slub_max_order to 0, what cause minimum possible order of slabs
+allocation.
 
 SLUB Debug output
 -----------------
 
 SLUB Debug output
 -----------------