From: Ravikiran G Thirumalai <kiran@scalex86.org>
To: Andrew Morton <akpm@osdl.org>
Cc: linux-mm@kvack.org, manfred@colorfullife.com,
clameter@engr.sgi.com, Alok Kataria <alokk@calsoftinc.com>
Subject: Re: [patch 3/3] mm: NUMA slab -- minor optimizations
Date: Tue, 29 Nov 2005 00:54:56 -0800 [thread overview]
Message-ID: <20051129085456.GC3573@localhost.localdomain> (raw)
In-Reply-To: <20051129085049.GA3573@localhost.localdomain>
Patch adds some minor optimizations:
1. Keeps on chip interrupts enabled for a bit longer while draining cpu
caches
2. Calls numa_node_id once in cache_reap
Signed-off-by: Alok N Kataria <alokk@calsoftinc.com>
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Shai Fultheim <shai@scalex86.org>
Index: linux-2.6.15-rc1/mm/slab.c
===================================================================
--- linux-2.6.15-rc1.orig/mm/slab.c 2005-11-17 21:32:43.000000000 -0800
+++ linux-2.6.15-rc1/mm/slab.c 2005-11-17 21:32:50.000000000 -0800
@@ -1914,18 +1914,18 @@
smp_call_function_all_cpus(do_drain, cachep);
check_irq_on();
- spin_lock_irq(&cachep->spinlock);
+ spin_lock(&cachep->spinlock);
for_each_online_node(node) {
l3 = cachep->nodelists[node];
if (l3) {
- spin_lock(&l3->list_lock);
+ spin_lock_irq(&l3->list_lock);
drain_array_locked(cachep, l3->shared, 1, node);
- spin_unlock(&l3->list_lock);
+ spin_unlock_irq(&l3->list_lock);
if (l3->alien)
drain_alien_cache(cachep, l3);
}
}
- spin_unlock_irq(&cachep->spinlock);
+ spin_unlock(&cachep->spinlock);
}
static int __node_shrink(kmem_cache_t *cachep, int node)
@@ -3304,7 +3304,7 @@
list_for_each(walk, &cache_chain) {
kmem_cache_t *searchp;
struct list_head* p;
- int tofree;
+ int tofree, nodeid;
struct slab *slabp;
searchp = list_entry(walk, kmem_cache_t, next);
@@ -3314,13 +3314,14 @@
check_irq_on();
- l3 = searchp->nodelists[numa_node_id()];
+ nodeid = numa_node_id();
+ l3 = searchp->nodelists[nodeid];
if (l3->alien)
drain_alien_cache(searchp, l3);
spin_lock_irq(&l3->list_lock);
drain_array_locked(searchp, ac_data(searchp), 0,
- numa_node_id());
+ nodeid);
if (time_after(l3->next_reap, jiffies))
goto next_unlock;
@@ -3329,7 +3330,7 @@
if (l3->shared)
drain_array_locked(searchp, l3->shared, 0,
- numa_node_id());
+ nodeid);
if (l3->free_touched) {
l3->free_touched = 0;
--
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:[~2005-11-29 8:54 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-11-29 8:50 [patch 1/3] mm: NUMA slab -- add alien cache drain statistics Ravikiran G Thirumalai
2005-11-29 8:53 ` [patch 2/3] mm: NUMA slab -- node local memory for off slab slab descriptors Ravikiran G Thirumalai
2005-11-29 17:48 ` Christoph Lameter
2005-11-29 8:54 ` Ravikiran G Thirumalai [this message]
2005-11-29 17:53 ` [patch 3/3] mm: NUMA slab -- minor optimizations Christoph Lameter
2005-12-26 23:42 ` Manfred Spraul
2005-12-27 20:35 ` Christoph Lameter
2005-11-29 9:25 ` [patch 1/3] mm: NUMA slab -- add alien cache drain statistics Andrew Morton
2005-11-29 17:57 ` Christoph Lameter
2005-11-29 18:49 ` Ravikiran G Thirumalai
2006-01-02 15:27 [patch 3/3] mm: NUMA slab -- minor optimizations Alok Kataria
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=20051129085456.GC3573@localhost.localdomain \
--to=kiran@scalex86.org \
--cc=akpm@osdl.org \
--cc=alokk@calsoftinc.com \
--cc=clameter@engr.sgi.com \
--cc=linux-mm@kvack.org \
--cc=manfred@colorfullife.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