From: Ingo Molnar <mingo@elte.hu>
To: Andrew Morton <akpm@osdl.org>
Cc: linux-kernel@vger.kernel.org,
Alexander Viro <viro@parcelfarce.linux.theplanet.co.uk>
Subject: [patch] sched, vfs: fix scheduling latencies in prune_dcache() and select_parent()
Date: Tue, 14 Sep 2004 12:19:04 +0200 [thread overview]
Message-ID: <20040914101904.GD24622@elte.hu> (raw)
In-Reply-To: <20040914100652.GB24622@elte.hu>
[-- Attachment #1: Type: text/plain, Size: 390 bytes --]
the attached patch fixes long scheduling latencies in select_parent()
and prune_dcache(). The prune_dcache() lock-break is easy, but for
select_parent() the only viable solution i found was to break out if
there's a resched necessary - the reordering is not necessary and the
dcache scanning/shrinking will later on do it anyway.
This patch has been in the -VP patchset for weeks.
Ingo
[-- Attachment #2: fix-latency-dcache.patch --]
[-- Type: text/plain, Size: 1249 bytes --]
the attached patch fixes long scheduling latencies in select_parent()
and prune_dcache(). The prune_dcache() lock-break is easy, but for
select_parent() the only viable solution i found was to break out if
there's a resched necessary - the reordering is not necessary and the
dcache scanning/shrinking will later on do it anyway.
This patch has been in the -VP patchset for weeks.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
--- linux/fs/dcache.c.orig
+++ linux/fs/dcache.c
@@ -381,6 +381,8 @@ static void prune_dcache(int count)
struct dentry *dentry;
struct list_head *tmp;
+ cond_resched_lock(&dcache_lock);
+
tmp = dentry_unused.prev;
if (tmp == &dentry_unused)
break;
@@ -553,6 +555,14 @@ resume:
struct dentry *dentry = list_entry(tmp, struct dentry, d_child);
next = tmp->next;
+ /*
+ * select_parent() is a performance optimization, it is
+ * not necessary to complete it. Abort if a reschedule is
+ * pending:
+ */
+ if (need_resched())
+ goto out;
+
if (!list_empty(&dentry->d_lru)) {
dentry_stat.nr_unused--;
list_del_init(&dentry->d_lru);
@@ -590,6 +600,7 @@ this_parent->d_parent->d_name.name, this
#endif
goto resume;
}
+out:
spin_unlock(&dcache_lock);
return found;
}
next prev parent reply other threads:[~2004-09-14 10:17 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-09-14 9:15 [patch] preempt-cleanup.patch, 2.6.9-rc2 Ingo Molnar
2004-09-14 9:34 ` [printk] make console_conditional_schedule() __sched and use cond_resched() William Lee Irwin III
2004-09-14 9:38 ` [patch] preempt-lock-need-resched.patch, 2.6.9-rc2 Ingo Molnar
2004-09-14 9:51 ` [patch] sched: add cond_resched_softirq() Ingo Molnar
2004-09-14 9:57 ` [patch] sched: fix latency in random driver Ingo Molnar
2004-09-14 10:06 ` [patch] sched, ext3: fix scheduling latencies in ext3 Ingo Molnar
2004-09-14 10:13 ` [patch] sched, vfs: fix scheduling latencies in invalidate_inodes() Ingo Molnar
2004-09-14 10:19 ` Ingo Molnar [this message]
2004-09-14 10:25 ` [patch] sched, net: fix scheduling latencies in netstat Ingo Molnar
2004-09-14 10:44 ` [patch] sched, net: fix scheduling latencies in __release_sock Ingo Molnar
2004-09-14 10:50 ` [patch] sched, mm: fix scheduling latencies in copy_page_range() Ingo Molnar
2004-09-14 10:56 ` [patch] sched, mm: fix scheduling latencies in unmap_vmas() Ingo Molnar
2004-09-14 10:59 ` [patch] sched, mm: fix scheduling latencies in get_user_pages() Ingo Molnar
2004-09-14 11:02 ` [patch] sched, mm: fix scheduling latencies in filemap_sync() Ingo Molnar
2004-09-14 11:06 ` [patch] sched, tty: fix scheduling latencies in tty_io.c Ingo Molnar
2004-09-14 10:53 ` Alan Cox
2004-09-14 12:00 ` Ingo Molnar
2004-09-14 11:18 ` Alan Cox
2004-09-14 12:27 ` Ingo Molnar
2004-09-14 12:11 ` Alan Cox
2004-09-14 11:08 ` [patch] sched, pty: fix scheduling latencies in pty.c Ingo Molnar
2004-09-14 11:12 ` [patch] might_sleep() additions to fs-writeback.c Ingo Molnar
2004-09-14 11:25 ` [patch] fix keventd execution dependency Ingo Molnar
2004-09-15 22:18 ` Rusty Russell
2004-09-14 11:28 ` [patch] sched: fix scheduling latencies in mttr.c Ingo Molnar
2004-09-14 11:32 ` [patch] sched: fix scheduling latencies in vgacon.c Ingo Molnar
2004-09-14 11:35 ` [patch] sched: fix scheduling latencies in NTFS mount Ingo Molnar
2004-09-14 13:31 ` Anton Altaparmakov
2004-09-14 11:42 ` [patch] sched: fix scheduling latencies for !PREEMPT kernels Ingo Molnar
2004-09-14 12:55 ` Nick Piggin
2004-09-14 13:22 ` Ingo Molnar
2004-09-14 13:33 ` Nick Piggin
2004-09-14 14:09 ` Andrea Arcangeli
2004-09-14 14:28 ` Nick Piggin
2004-09-14 15:03 ` Andrea Arcangeli
2004-09-14 18:05 ` Robert Love
2004-09-14 18:52 ` William Lee Irwin III
2004-09-14 19:02 ` Robert Love
2004-09-14 19:21 ` William Lee Irwin III
2004-09-14 19:19 ` Alan Cox
2004-09-15 0:22 ` Lee Revell
2004-09-15 1:46 ` William Lee Irwin III
2004-09-15 2:00 ` Lee Revell
2004-09-15 2:36 ` William Lee Irwin III
2004-09-15 2:59 ` Lee Revell
2004-09-15 13:36 ` Hans Reiser
2004-09-15 20:40 ` William Lee Irwin III
2004-09-15 1:18 ` William Lee Irwin III
2004-09-14 19:26 ` Robert Love
2004-09-14 21:06 ` William Lee Irwin III
2004-09-14 19:25 ` Andrea Arcangeli
2004-09-14 19:29 ` Robert Love
2004-09-14 19:34 ` William Lee Irwin III
2004-09-15 1:02 ` Lee Revell
2004-09-15 1:39 ` William Lee Irwin III
2004-09-15 2:11 ` Lee Revell
2004-09-15 11:17 ` Ingo Molnar
2004-09-15 9:56 ` Ingo Molnar
2004-09-15 9:57 ` William Lee Irwin III
2004-09-15 10:12 ` Ingo Molnar
2004-09-14 16:31 ` William Lee Irwin III
2004-09-14 16:39 ` Andrea Arcangeli
2004-09-14 14:54 ` Ingo Molnar
2004-09-14 22:55 ` Nick Piggin
2004-09-15 6:19 ` Ingo Molnar
2004-09-15 8:23 ` Nick Piggin
2004-09-15 8:43 ` Ingo Molnar
2004-09-15 10:09 ` William Lee Irwin III
2004-09-15 10:21 ` Ingo Molnar
2004-09-16 1:03 ` Nick Piggin
2004-09-16 6:14 ` Ingo Molnar
2004-09-15 0:35 ` Lee Revell
2004-09-14 13:25 ` [patch] sched: fix scheduling latencies in mtrr.c Ingo Molnar
2004-09-14 13:15 ` Alan Cox
2004-09-14 15:00 ` Ingo Molnar
2004-09-14 18:22 ` Zwane Mwaikambo
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=20040914101904.GD24622@elte.hu \
--to=mingo@elte.hu \
--cc=akpm@osdl.org \
--cc=linux-kernel@vger.kernel.org \
--cc=viro@parcelfarce.linux.theplanet.co.uk \
/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.