linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Christoph Lameter <cl@linux.com>
To: Pekka Enberg <penberg@kernel.org>
Cc: linux-mm@kvack.org, David Rientjes <rientjes@google.com>,
	Matt Mackall <mpm@selenic.com>,
	Glauber Costa <glommer@parallels.com>,
	Joonsoo Kim <js1304@gmail.com>
Subject: Common 22/22] Common object size alignment
Date: Wed, 23 May 2012 15:34:55 -0500	[thread overview]
Message-ID: <20120523203517.587788826@linux.com> (raw)
In-Reply-To: <20120523203433.340661918@linux.com>

[-- Attachment #1: align_size --]
[-- Type: text/plain, Size: 2841 bytes --]

All allocators align the objects to a word boundary. Put that into
common code.

Signed-off-by: Christoph Lameter <cl@linux.com>

---
 mm/slab.c        |   10 ----------
 mm/slab_common.c |    3 ++-
 mm/slub.c        |    7 -------
 3 files changed, 2 insertions(+), 18 deletions(-)

Index: linux-2.6/mm/slab.c
===================================================================
--- linux-2.6.orig/mm/slab.c	2012-05-23 09:13:05.938664721 -0500
+++ linux-2.6/mm/slab.c	2012-05-23 09:13:30.758664204 -0500
@@ -2240,16 +2240,6 @@ int __kmem_cache_create(struct kmem_cach
 	BUG_ON(flags & ~CREATE_MASK);
 
 	/*
-	 * Check that size is in terms of words.  This is needed to avoid
-	 * unaligned accesses for some archs when redzoning is used, and makes
-	 * sure any on-slab bufctl's are also correctly aligned.
-	 */
-	if (size & (BYTES_PER_WORD - 1)) {
-		size += (BYTES_PER_WORD - 1);
-		size &= ~(BYTES_PER_WORD - 1);
-	}
-
-	/*
 	 * Redzoning and user store require word alignment or possibly larger.
 	 * Note this will be overridden by architecture or caller mandated
 	 * alignment if either is greater than BYTES_PER_WORD.
Index: linux-2.6/mm/slab_common.c
===================================================================
--- linux-2.6.orig/mm/slab_common.c	2012-05-23 09:13:05.974664718 -0500
+++ linux-2.6/mm/slab_common.c	2012-05-23 09:14:49.634662589 -0500
@@ -127,6 +127,7 @@ struct kmem_cache *kmem_cache_create(con
 	WARN_ON(strchr(name, ' '));	/* It confuses parsers */
 #endif
 
+	/* Align size to a word boundary */
 	s = __kmem_cache_alias(name, size, align, flags, ctor);
 	if (s)
 		goto oops;
@@ -144,7 +145,7 @@ struct kmem_cache *kmem_cache_create(con
 
 	s->name = n;
 	s->ctor = ctor;
-	s->size = size;
+	s->size = ALIGN(s->size, sizeof(void *));
 	s->align = calculate_alignment(flags, align, size);
 	s->flags = flags;
 	r = __kmem_cache_create(s);
Index: linux-2.6/mm/slub.c
===================================================================
--- linux-2.6.orig/mm/slub.c	2012-05-23 09:13:05.954664718 -0500
+++ linux-2.6/mm/slub.c	2012-05-23 09:13:30.762664204 -0500
@@ -2860,13 +2860,6 @@ static int calculate_sizes(struct kmem_c
 	unsigned long align = s->align;
 	int order;
 
-	/*
-	 * Round up object size to the next word boundary. We can only
-	 * place the free pointer at word boundaries and this determines
-	 * the possible location of the free pointer.
-	 */
-	size = ALIGN(size, sizeof(void *));
-
 #ifdef CONFIG_SLUB_DEBUG
 	/*
 	 * Determine if we can poison the object itself. If the user of

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2012-05-23 20:35 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-05-23 20:34 Common 00/22] Sl[auo]b: Common functionality V3 Christoph Lameter
2012-05-23 20:34 ` Common 01/22] [slob] Define page struct fields used in mm_types.h Christoph Lameter
2012-05-31 21:14   ` David Rientjes
2012-06-01 13:30     ` JoonSoo Kim
2012-05-23 20:34 ` Common 03/22] [slob] Remove various small accessors Christoph Lameter
2012-05-31 21:19   ` David Rientjes
2012-06-01 13:29     ` JoonSoo Kim
2012-05-23 20:34 ` Common 05/22] [slab] Remove some accessors Christoph Lameter
2012-05-23 20:34 ` Common 06/22] Extract common fields from struct kmem_cache Christoph Lameter
2012-05-30  6:39   ` Pekka Enberg
2012-05-30 15:29     ` Christoph Lameter
2012-05-30 18:10       ` Christoph Lameter
2012-05-31  7:52       ` Pekka Enberg
2012-05-23 20:34 ` Common 07/22] Extract common code for kmem_cache_create() Christoph Lameter
2012-05-23 20:34 ` Common 08/22] Common definition for boot state of the slab allocators Christoph Lameter
2012-05-23 20:34 ` Common 09/22] Use a common mutex definition Christoph Lameter
2012-05-23 20:34 ` Common 10/22] Move kmem_cache_create mutex handling to common code Christoph Lameter
2012-05-23 20:34 ` Common 12/22] Extract a common function for kmem_cache_destroy Christoph Lameter
2012-05-23 20:34 ` Common 13/22] Always use the name "kmem_cache" for the slab cache with the kmem_cache structure Christoph Lameter
2012-05-23 20:34 ` Common 15/22] Get rid of __kmem_cache_destroy Christoph Lameter
2012-05-23 20:34 ` Common 16/22] Move duping of slab name to slab_common.c Christoph Lameter
2012-05-23 20:34 ` Common 17/22] Do slab aliasing call from common code Christoph Lameter
2012-05-23 20:34 ` Common 19/22] Do not pass ctor to __kmem_cache_create() Christoph Lameter
2012-05-23 20:34 ` Common 20/22] Set parameters on kmem_cache instead of passing them to functions Christoph Lameter
2012-05-23 20:34 ` Common 21/22] Common alignment code Christoph Lameter
2012-05-23 20:34 ` Christoph Lameter [this message]
2012-05-30  6:28 ` Common 00/22] Sl[auo]b: Common functionality V3 Pekka Enberg
     [not found] ` <20120523203506.170219003@linux.com>
2012-05-31 21:15   ` Common 02/22] [slob] No need to zero mapping since it is no longer in use David Rientjes
2012-06-01 13:30     ` JoonSoo Kim
     [not found] ` <20120523203507.324764286@linux.com>
2012-05-31 21:23   ` Common 04/22] [slab] Use page struct fields instead of casting David Rientjes
2012-06-01 13:17     ` JoonSoo Kim
2012-06-01 14:00     ` Christoph Lameter

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=20120523203517.587788826@linux.com \
    --to=cl@linux.com \
    --cc=glommer@parallels.com \
    --cc=js1304@gmail.com \
    --cc=linux-mm@kvack.org \
    --cc=mpm@selenic.com \
    --cc=penberg@kernel.org \
    --cc=rientjes@google.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