All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eric Dumazet <dada1@cosmosbay.com>
To: Davide Libenzi <davidel@xmailserver.org>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: [PATCH] eventpoll : Suppress a short lived lock from struct file
Date: Mon, 11 Jul 2005 09:18:27 +0200	[thread overview]
Message-ID: <42D21D43.3060300@cosmosbay.com> (raw)
In-Reply-To: <Pine.LNX.4.58.0504172159120.28447@bigblue.dev.mdolabs.com>

[-- Attachment #1: Type: text/plain, Size: 448 bytes --]

Hi Davide

I found in my tests that there is no need to have a f_ep_lock spinlock
attached to each struct file, using 8 bytes on 64bits platforms. The
lock is hold for a very short time period and can be global, with almost
no change in performance for applications using epoll, and a gain for
all others.

Thank you
Eric Dumazet

[PATCH] eventpoll : Suppress a short lived lock from struct file

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>


[-- Attachment #2: patch_eventpoll --]
[-- Type: text/plain, Size: 1856 bytes --]

--- linux-2.6.12/fs/eventpoll.c	2005-06-17 21:48:29.000000000 +0200
+++ linux-2.6.12-ed/fs/eventpoll.c	2005-07-11 08:56:07.000000000 +0200
@@ -179,6 +179,8 @@
 	spinlock_t lock;
 };
 
+static DEFINE_SPINLOCK(f_ep_lock);
+
 /*
  * This structure is stored inside the "private_data" member of the file
  * structure and rapresent the main data sructure for the eventpoll
@@ -426,7 +428,6 @@
 {
 
 	INIT_LIST_HEAD(&file->f_ep_links);
-	spin_lock_init(&file->f_ep_lock);
 }
 
 
@@ -967,9 +968,9 @@
 		goto eexit_2;
 
 	/* Add the current item to the list of active epoll hook for this file */
-	spin_lock(&tfile->f_ep_lock);
+	spin_lock(&f_ep_lock);
 	list_add_tail(&epi->fllink, &tfile->f_ep_links);
-	spin_unlock(&tfile->f_ep_lock);
+	spin_unlock(&f_ep_lock);
 
 	/* We have to drop the new item inside our item list to keep track of it */
 	write_lock_irqsave(&ep->lock, flags);
@@ -1160,7 +1161,6 @@
 {
 	int error;
 	unsigned long flags;
-	struct file *file = epi->ffd.file;
 
 	/*
 	 * Removes poll wait queue hooks. We _have_ to do this without holding
@@ -1173,10 +1173,10 @@
 	ep_unregister_pollwait(ep, epi);
 
 	/* Remove the current item from the list of epoll hooks */
-	spin_lock(&file->f_ep_lock);
+	spin_lock(&f_ep_lock);
 	if (EP_IS_LINKED(&epi->fllink))
 		EP_LIST_DEL(&epi->fllink);
-	spin_unlock(&file->f_ep_lock);
+	spin_unlock(&f_ep_lock);
 
 	/* We need to acquire the write IRQ lock before calling ep_unlink() */
 	write_lock_irqsave(&ep->lock, flags);
--- linux-2.6.12/include/linux/fs.h	2005-06-17 21:48:29.000000000 +0200
+++ linux-2.6.12-ed/include/linux/fs.h	2005-07-11 08:58:02.000000000 +0200
@@ -597,7 +597,6 @@
 #ifdef CONFIG_EPOLL
 	/* Used by fs/eventpoll.c to link all the hooks to this file */
 	struct list_head	f_ep_links;
-	spinlock_t		f_ep_lock;
 #endif /* #ifdef CONFIG_EPOLL */
 	struct address_space	*f_mapping;
 };

  parent reply	other threads:[~2005-07-11  7:18 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-04-18  3:19 [PATCH x86_64] Live Patching Function on 2.6.11.7 Takashi Ikebe
2005-04-18  4:07 ` Chris Wedgwood
2005-04-18  4:19   ` Takashi Ikebe
2005-04-18  4:42     ` Daniel Jacobowitz
2005-04-18  4:55       ` Nicholas Miell
2005-04-18  5:01         ` Davide Libenzi
2005-04-18  5:41           ` Takashi Ikebe
2005-07-11  7:18           ` Eric Dumazet [this message]
2005-07-11  8:34             ` [PATCH] eventpoll : Suppress a short lived lock from struct file Peter Zijlstra
2005-07-11  9:29               ` Eric Dumazet
2005-07-11 14:00                 ` Davide Libenzi
2005-07-11 15:20                   ` Eric Dumazet
2005-04-18  5:00       ` [PATCH x86_64] Live Patching Function on 2.6.11.7 David S. Miller
2005-04-18  6:12     ` Chris Wedgwood
2005-04-18  6:35       ` Chris Friesen
2005-04-18  6:48         ` Chris Wedgwood
2005-04-18 10:03         ` James Courtier-Dutton
2005-04-18  9:10           ` Chris Wedgwood
2005-04-18  7:32       ` Takashi Ikebe
2005-04-18  7:56         ` Chris Wedgwood
2005-04-18  8:37           ` Takashi Ikebe
2005-04-18  8:59             ` Chris Wedgwood
2005-04-18  9:16           ` Paul Jackson
2005-04-18  9:25             ` Chris Wedgwood
2005-04-18 11:30               ` Rik van Riel
2005-04-18 12:52                 ` Takashi Ikebe
2005-04-18 14:06                   ` Rik van Riel
2005-04-19  2:14                     ` Takashi Ikebe
2005-04-19  4:27                       ` Chris Wedgwood
2005-04-19  5:19                         ` Takashi Ikebe
2005-04-19  5:52                           ` Chris Wedgwood
2005-04-20  4:18                             ` Takashi Ikebe
2005-04-20  5:43                               ` Chris Wedgwood
2005-04-20  7:35                                 ` Takashi Ikebe
2005-04-20  7:50                                   ` Chris Wedgwood
2005-04-20  7:57                                     ` Takashi Ikebe
2005-04-20  8:26                                       ` Chris Wedgwood
2005-04-20  8:45                                         ` Takashi Ikebe
2005-04-20  8:51                                           ` Chris Wedgwood
2005-04-20 11:19                                           ` Rik van Riel
2005-04-20 15:06                                             ` Chris Friesen
2005-04-20  8:34                                       ` Miquel van Smoorenburg
2005-04-19  5:57                           ` Takashi Ikebe
2005-04-18 14:28               ` Paul Jackson
2005-04-20 13:10               ` Ralf Baechle
2005-04-20 15:08                 ` Chris Friesen
2005-04-23 16:17 ` Andi Kleen
2005-04-25  2:11   ` Takashi Ikebe
2005-04-25  2:48     ` Kyle Moffett
2005-04-25 10:39       ` Takashi Ikebe
2005-04-25 11:15         ` Kyle Moffett
2005-04-25 15:09         ` Pavel Machek
2005-04-25 15:54         ` Andi Kleen
2005-04-25 16:36         ` Valdis.Kletnieks
2005-04-26  1:34           ` Takashi Ikebe
2005-04-26  2:15             ` Kyle Moffett
2005-04-26  9:36             ` Pavel Machek
2005-04-26 13:05             ` 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=42D21D43.3060300@cosmosbay.com \
    --to=dada1@cosmosbay.com \
    --cc=davidel@xmailserver.org \
    --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.