From: clameter@sgi.com
To: Nishanth Aravamudan <nacc@us.ibm.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>, linux-mm@kvack.org
Subject: [RFC 12/13] SLUB: minimum alignment fixes
Date: Thu, 14 Jun 2007 00:50:38 -0700 [thread overview]
Message-ID: <20070614075336.870895145@sgi.com> (raw)
In-Reply-To: <20070614075026.607300756@sgi.com>
[-- Attachment #1: slub_min_align --]
[-- Type: text/plain, Size: 3991 bytes --]
If ARCH_KMALLOC_MINALIGN is set to a value greater than 8 (SLUBs smallest
kmalloc cache) then SLUB may generate duplicate slabs in sysfs (yes again).
However, no arch sets ARCH_KMALLOC_MINALIGN larger than 8 though except mips
which for some reason wants a 128 byte alignment.
This patch increases the size of the smallest cache if ARCH_KMALLOC_MINALIGN
is greater than 8. In that case more and more of the smallest caches are
disabled.
If we do that then the count of the active general caches that is displayed
on boot is not correct anymore since we may skip elements of the kmalloc
array. So count them separately.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
---
include/linux/slub_def.h | 13 +++++++++++--
mm/slub.c | 20 +++++++++++++++-----
2 files changed, 26 insertions(+), 7 deletions(-)
Index: vps/include/linux/slub_def.h
===================================================================
--- vps.orig/include/linux/slub_def.h 2007-06-12 15:58:30.000000000 -0700
+++ vps/include/linux/slub_def.h 2007-06-12 16:00:43.000000000 -0700
@@ -28,7 +28,7 @@ struct kmem_cache {
int size; /* The size of an object including meta data */
int objsize; /* The size of an object without meta data */
int offset; /* Free pointer offset. */
- unsigned int order;
+ int order;
/*
* Avoid an extra cache line for UP, SMP and for the node local to
@@ -56,7 +56,13 @@ struct kmem_cache {
/*
* Kmalloc subsystem.
*/
-#define KMALLOC_SHIFT_LOW 3
+#if defined(ARCH_KMALLOC_MINALIGN) && ARCH_KMALLOC_MINALIGN > 8
+#define KMALLOC_MIN_SIZE ARCH_KMALLOC_MINALIGN
+#else
+#define KMALLOC_MIN_SIZE 8
+#endif
+
+#define KMALLOC_SHIFT_LOW ilog2(KMALLOC_MIN_SIZE)
/*
* We keep the general caches in an array of slab caches that are used for
@@ -76,6 +82,9 @@ static inline int kmalloc_index(size_t s
if (size > KMALLOC_MAX_SIZE)
return -1;
+ if (size <= KMALLOC_MIN_SIZE)
+ return KMALLOC_SHIFT_LOW;
+
if (size > 64 && size <= 96)
return 1;
if (size > 128 && size <= 192)
Index: vps/mm/slub.c
===================================================================
--- vps.orig/mm/slub.c 2007-06-12 15:58:37.000000000 -0700
+++ vps/mm/slub.c 2007-06-12 16:03:00.000000000 -0700
@@ -2521,6 +2521,7 @@ EXPORT_SYMBOL(krealloc);
void __init kmem_cache_init(void)
{
int i;
+ int caches = 0;
if (!page_group_by_mobility_disabled && !user_override) {
/*
@@ -2540,20 +2541,29 @@ void __init kmem_cache_init(void)
create_kmalloc_cache(&kmalloc_caches[0], "kmem_cache_node",
sizeof(struct kmem_cache_node), GFP_KERNEL);
kmalloc_caches[0].refcount = -1;
+ caches++;
#endif
/* Able to allocate the per node structures */
slab_state = PARTIAL;
/* Caches that are not of the two-to-the-power-of size */
- create_kmalloc_cache(&kmalloc_caches[1],
+ if (KMALLOC_MIN_SIZE <= 64) {
+ create_kmalloc_cache(&kmalloc_caches[1],
"kmalloc-96", 96, GFP_KERNEL);
- create_kmalloc_cache(&kmalloc_caches[2],
+ caches++;
+ }
+ if (KMALLOC_MIN_SIZE <= 128) {
+ create_kmalloc_cache(&kmalloc_caches[2],
"kmalloc-192", 192, GFP_KERNEL);
+ caches++;
+ }
- for (i = KMALLOC_SHIFT_LOW; i <= KMALLOC_SHIFT_HIGH; i++)
+ for (i = KMALLOC_SHIFT_LOW; i <= KMALLOC_SHIFT_HIGH; i++) {
create_kmalloc_cache(&kmalloc_caches[i],
"kmalloc", 1 << i, GFP_KERNEL);
+ caches++;
+ }
slab_state = UP;
@@ -2570,8 +2580,8 @@ void __init kmem_cache_init(void)
nr_cpu_ids * sizeof(struct page *);
printk(KERN_INFO "SLUB: Genslabs=%d, HWalign=%d, Order=%d-%d, MinObjects=%d,"
- " Processors=%d, Nodes=%d\n",
- KMALLOC_SHIFT_HIGH, cache_line_size(),
+ " CPUs=%d, Nodes=%d\n",
+ caches, cache_line_size(),
slub_min_order, slub_max_order, slub_min_objects,
nr_cpu_ids, nr_node_ids);
}
--
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2007-06-14 7:50 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-06-14 7:50 [RFC 00/13] RFC memoryless node handling fixes clameter
2007-06-14 7:50 ` [RFC 01/13] NUMA: introduce node_memory_map clameter
2007-06-14 7:50 ` [RFC 02/13] Fix MPOL_INTERLEAVE behavior for memoryless nodes clameter
2007-06-14 7:50 ` [RFC 03/13] OOM: use the node_memory_map instead of constructing one on the fly clameter
2007-06-14 7:50 ` [RFC 04/13] Memoryless Nodes: No need for kswapd clameter
2007-06-14 7:50 ` [RFC 05/13] Memoryless Node: Slab support clameter
2007-06-14 7:50 ` [RFC 06/13] Memoryless nodes: SLUB support clameter
2007-06-14 7:50 ` [RFC 07/13] Uncached allocator: Handle memoryless nodes clameter
2007-06-14 7:50 ` [RFC 08/13] Memoryless node: Allow profiling data to fall back to other nodes clameter
2007-06-14 7:50 ` [RFC 09/13] Memoryless nodes: Update memory policy and page migration clameter
2007-06-14 7:50 ` [RFC 10/13] Memoryless nodes: Fix GFP_THISNODE behavior clameter
2007-06-14 16:07 ` Nishanth Aravamudan
2007-06-14 16:13 ` Christoph Lameter
2007-06-18 16:47 ` Nishanth Aravamudan
2007-06-14 7:50 ` [RFC 11/13] SLUB: Ensure that the # object per slabs stays low enough clameter
2007-06-14 7:50 ` clameter [this message]
2007-06-14 7:56 ` [RFC 12/13] SLUB: minimum alignment fixes Christoph Lameter
2007-06-14 7:50 ` [RFC 13/13] I finally found a way to get rid of the nasty list of comparisions in slub_def.h. ilog2 seems to work right for constants clameter
2007-06-14 7:57 ` Christoph Lameter
2007-06-14 14:24 ` [RFC 00/13] RFC memoryless node handling fixes Nishanth Aravamudan
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20070614075336.870895145@sgi.com \
--to=clameter@sgi.com \
--cc=Lee.Schermerhorn@hp.com \
--cc=linux-mm@kvack.org \
--cc=nacc@us.ibm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox