All of lore.kernel.org
 help / color / mirror / Atom feed
From: Manfred Spraul <manfred@colorfullife.com>
To: linux-kernel@vger.kernel.org
Subject: [PATCH,RFC] faster kmalloc lookup
Date: Sat, 26 Oct 2002 21:22:12 +0200	[thread overview]
Message-ID: <3DBAEB64.1090109@colorfullife.com> (raw)

[-- Attachment #1: Type: text/plain, Size: 241 bytes --]

kmalloc spends a large part of the total execution time trying to find 
the cache for the passed in size.

What about the attached patch (against 2.5.44-mm5)?
It uses fls jump over the caches that are definitively too small.

--
    Manfred

[-- Attachment #2: patch-fast-kmalloc --]
[-- Type: text/plain, Size: 1219 bytes --]

--- 2.5/mm/slab.c	Sat Oct 26 21:13:33 2002
+++ build-2.5/mm/slab.c	Sat Oct 26 20:40:09 2002
@@ -424,6 +430,7 @@
 	CN("size-131072")
 }; 
 #undef CN
+static struct cache_sizes *malloc_hints[sizeof(size_t)*8];
 
 struct arraycache_init initarray_cache __initdata = { { 0, BOOT_CPUCACHE_ENTRIES, 1, 0} };
 struct arraycache_init initarray_generic __initdata = { { 0, BOOT_CPUCACHE_ENTRIES, 1, 0} };
@@ -587,6 +594,7 @@
 void __init kmem_cache_init(void)
 {
 	size_t left_over;
+	int i;
 
 	init_MUTEX(&cache_chain_sem);
 	INIT_LIST_HEAD(&cache_chain);
@@ -604,6 +612,18 @@
 	 * that initializes ac_data for all new cpus
 	 */
 	register_cpu_notifier(&cpucache_notifier);
+
+	for (i=0;i<sizeof(size_t)*8;i++) {
+		struct cache_sizes *csizep = malloc_sizes;
+		int size = (1<<i)/2+1;
+
+		for ( ; csizep->cs_size; csizep++) {
+			if (size > csizep->cs_size)
+				continue;
+			break;
+		}
+		malloc_hints[i] = csizep;
+	}
 }
 
 
@@ -1796,7 +1816,11 @@
  */
 void * kmalloc (size_t size, int flags)
 {
-	struct cache_sizes *csizep = malloc_sizes;
+	struct cache_sizes *csizep;
+	
+	if(unlikely(size<2))
+		size=2;
+	csizep = malloc_hints[fls((size-1))];
 
 	for (; csizep->cs_size; csizep++) {
 		if (size > csizep->cs_size)

             reply	other threads:[~2002-10-26 19:15 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-10-26 19:22 Manfred Spraul [this message]
2002-10-26 22:30 ` [PATCH,RFC] faster kmalloc lookup Alan Cox
2002-10-27 10:08   ` Manfred Spraul
2002-10-27 13:29     ` Manfred Spraul
2002-10-28 13:05       ` Nikita Danilov
2002-10-28 13:18         ` Marcus Alanen
2002-10-28 13:26           ` Nikita Danilov
2002-10-28 16:03             ` Manfred Spraul
2002-10-31  9:35 ` Arjan van de Ven
2002-10-31 23:03   ` Manfred Spraul
     [not found] <3DBAEB64.1090109@colorfullife.com.suse.lists.linux.kernel>
     [not found] ` <1036056917.2872.0.camel@dhcp59-228.rdu.redhat.com.suse.lists.linux.kernel>
     [not found]   ` <3DC1B6D0.8050202@colorfullife.com.suse.lists.linux.kernel>
2002-11-01  8:03     ` Andi Kleen

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=3DBAEB64.1090109@colorfullife.com \
    --to=manfred@colorfullife.com \
    --cc=linux-kernel@vger.kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.