linux-arch.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Christoph Lameter <clameter@sgi.com>
To: akpm@linux-foundation.org
Cc: linux-arch@vger.kernel.org, linux-kernel@vger.kernel.org,
	David Miller <davem@davemloft.net>,
	Eric Dumazet <dada1@cosmosbay.com>,
	Peter Zijlstra <peterz@infradead.org>
Subject: [patch 03/30] cpu alloc: Remove SLUB fields
Date: Fri, 16 Nov 2007 15:09:23 -0800	[thread overview]
Message-ID: <20071116231102.487589938@sgi.com> (raw)
In-Reply-To: 20071116230920.278761667@sgi.com

[-- Attachment #1: cpu_alloc_remove_slub_fields --]
[-- Type: text/plain, Size: 5883 bytes --]

Remove the fields in kmem_cache_cpu that were used to cache data from
kmem_cache when they were in different cachelines. The cacheline that holds
the per cpu array pointer now also holds these values. We can cut down the
kmem_cache_cpu size to almost half.

The get_freepointer() and set_freepointer() functions that used to be only
intended for the slow path now are also useful for the hot path since access
to the field does not require an additional cacheline anymore. This results
in consistent use of setting the freepointer for objects throughout SLUB.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
---
 include/linux/slub_def.h |    3 --
 mm/slub.c                |   50 +++++++++++++++--------------------------------
 2 files changed, 17 insertions(+), 36 deletions(-)

Index: linux-2.6/include/linux/slub_def.h
===================================================================
--- linux-2.6.orig/include/linux/slub_def.h	2007-11-15 21:25:07.622904866 -0800
+++ linux-2.6/include/linux/slub_def.h	2007-11-15 21:25:10.335154196 -0800
@@ -15,9 +15,6 @@ struct kmem_cache_cpu {
 	void **freelist;
 	struct page *page;
 	int node;
-	unsigned int offset;
-	unsigned int objsize;
-	unsigned int objects;
 };
 
 struct kmem_cache_node {
Index: linux-2.6/mm/slub.c
===================================================================
--- linux-2.6.orig/mm/slub.c	2007-11-15 21:25:07.622904866 -0800
+++ linux-2.6/mm/slub.c	2007-11-15 21:25:10.339154532 -0800
@@ -273,13 +273,6 @@ static inline int check_valid_pointer(st
 	return 1;
 }
 
-/*
- * Slow version of get and set free pointer.
- *
- * This version requires touching the cache lines of kmem_cache which
- * we avoid to do in the fast alloc free paths. There we obtain the offset
- * from the page struct.
- */
 static inline void *get_freepointer(struct kmem_cache *s, void *object)
 {
 	return *(void **)(object + s->offset);
@@ -1438,10 +1431,10 @@ static void deactivate_slab(struct kmem_
 
 		/* Retrieve object from cpu_freelist */
 		object = c->freelist;
-		c->freelist = c->freelist[c->offset];
+		c->freelist = get_freepointer(s, c->freelist);
 
 		/* And put onto the regular freelist */
-		object[c->offset] = page->freelist;
+		set_freepointer(s, object, page->freelist);
 		page->freelist = object;
 		page->inuse--;
 	}
@@ -1573,8 +1566,8 @@ load_freelist:
 		goto debug;
 
 	object = c->page->freelist;
-	c->freelist = object[c->offset];
-	c->page->inuse = c->objects;
+	c->freelist = get_freepointer(s, object);
+	c->page->inuse = s->objects;
 	c->page->freelist = c->page->end;
 	c->node = page_to_nid(c->page);
 unlock_out:
@@ -1602,7 +1595,7 @@ debug:
 		goto another_slab;
 
 	c->page->inuse++;
-	c->page->freelist = object[c->offset];
+	c->page->freelist = get_freepointer(s, object);
 	c->node = -1;
 	goto unlock_out;
 }
@@ -1636,8 +1629,8 @@ static void __always_inline *slab_alloc(
 			}
 			break;
 		}
-	} while (cmpxchg_local(&c->freelist, object, object[c->offset])
-								!= object);
+	} while (cmpxchg_local(&c->freelist, object,
+			get_freepointer(s, object)) != object);
 	preempt_enable();
 #else
 	unsigned long flags;
@@ -1653,13 +1646,13 @@ static void __always_inline *slab_alloc(
 		}
 	} else {
 		object = c->freelist;
-		c->freelist = object[c->offset];
+		c->freelist = get_freepointer(s, object);
 	}
 	local_irq_restore(flags);
 #endif
 
 	if (unlikely((gfpflags & __GFP_ZERO)))
-		memset(object, 0, c->objsize);
+		memset(object, 0, s->objsize);
 out:
 	return object;
 }
@@ -1687,7 +1680,7 @@ EXPORT_SYMBOL(kmem_cache_alloc_node);
  * handling required then we can return immediately.
  */
 static void __slab_free(struct kmem_cache *s, struct page *page,
-				void *x, void *addr, unsigned int offset)
+				void *x, void *addr)
 {
 	void *prior;
 	void **object = (void *)x;
@@ -1703,7 +1696,8 @@ static void __slab_free(struct kmem_cach
 	if (unlikely(state & SLABDEBUG))
 		goto debug;
 checks_ok:
-	prior = object[offset] = page->freelist;
+	prior = page->freelist;
+	set_freepointer(s, object, prior);
 	page->freelist = object;
 	page->inuse--;
 
@@ -1786,10 +1780,10 @@ static void __always_inline slab_free(st
 		 * since the freelist pointers are unique per slab.
 		 */
 		if (unlikely(page != c->page || c->node < 0)) {
-			__slab_free(s, page, x, addr, c->offset);
+			__slab_free(s, page, x, addr);
 			break;
 		}
-		object[c->offset] = freelist;
+		set_freepointer(s, object, freelist);
 	} while (cmpxchg_local(&c->freelist, freelist, object) != freelist);
 	preempt_enable();
 #else
@@ -1799,10 +1793,10 @@ static void __always_inline slab_free(st
 	debug_check_no_locks_freed(object, s->objsize);
 	c = THIS_CPU(s->cpu_slab);
 	if (likely(page == c->page && c->node >= 0)) {
-		object[c->offset] = c->freelist;
+		set_freepointer(s, object, c->freelist);
 		c->freelist = object;
 	} else
-		__slab_free(s, page, x, addr, c->offset);
+		__slab_free(s, page, x, addr);
 
 	local_irq_restore(flags);
 #endif
@@ -1984,9 +1978,6 @@ static void init_kmem_cache_cpu(struct k
 	c->page = NULL;
 	c->freelist = (void *)PAGE_MAPPING_ANON;
 	c->node = 0;
-	c->offset = s->offset / sizeof(void *);
-	c->objsize = s->objsize;
-	c->objects = s->objects;
 }
 
 static void init_kmem_cache_node(struct kmem_cache_node *n)
@@ -2978,21 +2969,14 @@ struct kmem_cache *kmem_cache_create(con
 	down_write(&slub_lock);
 	s = find_mergeable(size, align, flags, name, ctor);
 	if (s) {
-		int cpu;
-
 		s->refcount++;
+
 		/*
 		 * Adjust the object sizes so that we clear
 		 * the complete object on kzalloc.
 		 */
 		s->objsize = max(s->objsize, (int)size);
 
-		/*
-		 * And then we need to update the object size in the
-		 * per cpu structures
-		 */
-		for_each_online_cpu(cpu)
-			CPU_PTR(s->cpu_slab, cpu)->objsize = s->objsize;
 		s->inuse = max_t(int, s->inuse, ALIGN(size, sizeof(void *)));
 		up_write(&slub_lock);
 		if (sysfs_slab_alias(s, name))

-- 

  parent reply	other threads:[~2007-11-16 23:11 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-11-16 23:09 [patch 00/30] cpu alloc v2: Optimize by removing arrays of pointers to per cpu objects Christoph Lameter
2007-11-16 23:09 ` [patch 01/30] cpu alloc: Simple version of the allocator (static allocations) Christoph Lameter
2007-11-16 23:09 ` [patch 02/30] cpu alloc: Use in SLUB Christoph Lameter
2007-11-16 23:09 ` Christoph Lameter [this message]
2007-11-16 23:09 ` [patch 04/30] cpu alloc: page allocator conversion Christoph Lameter
2007-11-16 23:09 ` [patch 05/30] cpu_alloc: Implement dynamically extendable cpu areas Christoph Lameter
2007-11-16 23:09 ` [patch 06/30] cpu alloc: x86 support Christoph Lameter
2007-11-16 23:09 ` [patch 07/30] cpu alloc: IA64 support Christoph Lameter
2007-11-16 23:32   ` Luck, Tony
2007-11-17  0:05     ` Christoph Lameter
2007-11-16 23:09 ` [patch 08/30] cpu_alloc: Sparc64 support Christoph Lameter
2007-11-16 23:09 ` [patch 09/30] cpu alloc: percpu_counter conversion Christoph Lameter
2007-11-16 23:09 ` [patch 10/30] cpu alloc: crash_notes conversion Christoph Lameter
2007-11-16 23:09 ` [patch 11/30] cpu alloc: workqueue conversion Christoph Lameter
2007-11-16 23:09 ` [patch 12/30] cpu alloc: ACPI cstate handling conversion Christoph Lameter
2007-11-16 23:09 ` [patch 13/30] cpu alloc: genhd statistics conversion Christoph Lameter
2007-11-16 23:09 ` [patch 14/30] cpu alloc: blktrace conversion Christoph Lameter
2007-11-16 23:09 ` [patch 15/30] cpu alloc: SRCU Christoph Lameter
2007-11-16 23:09 ` [patch 16/30] cpu alloc: XFS counters Christoph Lameter
2007-11-19 12:58   ` Christoph Hellwig
2007-11-16 23:09 ` [patch 17/30] cpu alloc: NFS statistics Christoph Lameter
2007-11-16 23:09 ` [patch 18/30] cpu alloc: neigbour statistics Christoph Lameter
2007-11-16 23:09 ` [patch 19/30] cpu alloc: tcp statistics Christoph Lameter
2007-11-16 23:09 ` [patch 20/30] cpu alloc: convert scatches Christoph Lameter
2007-11-16 23:09 ` [patch 21/30] cpu alloc: dmaengine conversion Christoph Lameter
2007-11-16 23:09 ` [patch 22/30] cpu alloc: convert loopback statistics Christoph Lameter
2007-11-16 23:09 ` [patch 23/30] cpu alloc: veth conversion Christoph Lameter
2007-11-16 23:09 ` [patch 24/30] cpu alloc: Chelsio statistics conversion Christoph Lameter
2007-11-16 23:09 ` [patch 25/30] cpu alloc: convert mib handling to cpu alloc Christoph Lameter
2007-11-16 23:09 ` [patch 26/30] cpu_alloc: convert network sockets Christoph Lameter
2007-11-16 23:09 ` [patch 27/30] cpu alloc: Explicitly code allocpercpu calls in iucv Christoph Lameter
2007-11-16 23:09 ` [patch 28/30] cpu alloc: Use for infiniband Christoph Lameter
2007-11-16 23:09 ` [patch 29/30] cpu alloc: Use in the crypto subsystem Christoph Lameter
2007-11-16 23:09 ` [patch 30/30] cpu alloc: Remove the allocpercpu functionality 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=20071116231102.487589938@sgi.com \
    --to=clameter@sgi.com \
    --cc=akpm@linux-foundation.org \
    --cc=dada1@cosmosbay.com \
    --cc=davem@davemloft.net \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=peterz@infradead.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).