linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Wu Fengguang <fengguang.wu@intel.com>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Andi Kleen <andi@firstfloor.org>,
	Quentin Barnes <qbarnes+nfs@yahoo-inc.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	linux-fsdevel@vger.kernel.org, Nick Piggin <npiggin@suse.de>,
	Steven Whitehouse <swhiteho@redhat.com>
Subject: [RFC][PATCH 1/2] readahead: replace ra->mmap_miss with ra->flags
Date: Wed, 30 Dec 2009 13:17:47 +0800	[thread overview]
Message-ID: <20091230051747.GA26364@localhost> (raw)
In-Reply-To: <20091230051540.GA16308@localhost>

Introduce a readahead flags field and embed the existing mmap_miss in it
(to save space).

It will be possible to lose the flags in race conditions, however the
impact should be limited.

CC: Nick Piggin <npiggin@suse.de>
CC: Andi Kleen <andi@firstfloor.org>
CC: Steven Whitehouse <swhiteho@redhat.com>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
---
 include/linux/fs.h |   30 +++++++++++++++++++++++++++++-
 mm/filemap.c       |    7 ++-----
 2 files changed, 31 insertions(+), 6 deletions(-)

--- linux.orig/include/linux/fs.h	2009-12-30 11:07:04.000000000 +0800
+++ linux/include/linux/fs.h	2009-12-30 12:59:28.000000000 +0800
@@ -889,10 +889,38 @@ struct file_ra_state {
 					   there are only # of pages ahead */
 
 	unsigned int ra_pages;		/* Maximum readahead window */
-	unsigned int mmap_miss;		/* Cache miss stat for mmap accesses */
+	unsigned int flags;
 	loff_t prev_pos;		/* Cache last read() position */
 };
 
+/* low 16 bits: cache miss stat for mmap accesses */
+#define	RA_FLAG_MMAP_MISS		0x0000ffff
+
+/*
+ * Don't do flags++ directly to avoid possible overflow:
+ * the ra fields can be accessed concurrently in a racy way.
+ */
+static inline unsigned int ra_mmap_miss_inc(struct file_ra_state *ra)
+{
+	unsigned int miss = ra->flags & RA_FLAG_MMAP_MISS;
+
+	if (miss < RA_FLAG_MMAP_MISS) {
+		miss++;
+		ra->flags = miss | (ra->flags &~ RA_FLAG_MMAP_MISS);
+	}
+	return miss;
+}
+
+static inline void ra_mmap_miss_dec(struct file_ra_state *ra)
+{
+	unsigned int miss = ra->flags & RA_FLAG_MMAP_MISS;
+
+	if (miss) {
+		miss--;
+		ra->flags = miss | (ra->flags &~ RA_FLAG_MMAP_MISS);
+	}
+}
+
 /*
  * Check if @index falls in the readahead windows.
  */
--- linux.orig/mm/filemap.c	2009-12-30 11:07:04.000000000 +0800
+++ linux/mm/filemap.c	2009-12-30 11:07:22.000000000 +0800
@@ -1418,14 +1418,12 @@ static void do_sync_mmap_readahead(struc
 		return;
 	}
 
-	if (ra->mmap_miss < INT_MAX)
-		ra->mmap_miss++;
 
 	/*
 	 * Do we miss much more than hit in this file? If so,
 	 * stop bothering with read-ahead. It will only hurt.
 	 */
-	if (ra->mmap_miss > MMAP_LOTSAMISS)
+	if (ra_mmap_miss_inc(ra) > MMAP_LOTSAMISS)
 		return;
 
 	/*
@@ -1455,8 +1453,7 @@ static void do_async_mmap_readahead(stru
 	/* If we don't want any read-ahead, don't bother */
 	if (VM_RandomReadHint(vma))
 		return;
-	if (ra->mmap_miss > 0)
-		ra->mmap_miss--;
+	ra_mmap_miss_dec(ra);
 	if (PageReadahead(page))
 		page_cache_async_readahead(mapping, ra, file,
 					   page, offset, ra->ra_pages);

  reply	other threads:[~2009-12-30  5:17 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20091225000717.GA26949@yahoo-inc.com>
     [not found] ` <87aax18xms.fsf@basil.nowhere.org>
2009-12-30  5:15   ` [RFC][PATCH] Disabling read-ahead makes I/O of large reads small Wu Fengguang
2009-12-30  5:17     ` Wu Fengguang [this message]
2009-12-30  5:24     ` [RFC][PATCH 2/2] readahead: avoid page-by-page reads on POSIX_FADV_RANDOM Wu Fengguang
2009-12-30 18:02       ` Andi Kleen
2009-12-31  1:39         ` Wu Fengguang
2009-12-31  2:53           ` Wu Fengguang
2009-12-31  3:03             ` Wu Fengguang
2009-12-31  4:31         ` [RFC][PATCH v2] readahead: introduce O_RANDOM_READ for POSIX_FADV_RANDOM Wu Fengguang
2010-01-08 13:08           ` Christoph Hellwig
2010-01-09 13:59             ` Wu Fengguang
2010-01-09 14:01               ` Wu Fengguang
2010-01-04  4:50         ` [RFC][PATCH v3] readahead: introduce O_RANDOM " Wu Fengguang
2010-01-04  5:17           ` Stephen Rothwell
2010-01-04  7:33             ` Christoph Hellwig
2010-01-04 12:56               ` [RFC][PATCH v4] " Wu Fengguang
2010-01-05  2:03                 ` Stephen Rothwell
2010-01-05  2:26                   ` Wu Fengguang
2010-01-05  2:28                     ` Stephen Rothwell
2010-01-05  2:45                       ` Wu Fengguang
2010-01-05  5:21                         ` Eric Paris
2010-01-05  3:18                       ` [RFC][PATCH v5] " Wu Fengguang
2010-01-05  3:27                         ` Wu Fengguang
2010-01-04 16:50               ` [RFC][PATCH v3] " Quentin Barnes
2010-01-04 18:57                 ` Andreas Dilger
2010-01-04  5:20           ` Minchan Kim
2010-01-04 12:16             ` Wu Fengguang
2010-01-05  1:46               ` Minchan Kim
2010-01-05  2:16                 ` Wu Fengguang
2010-01-05  3:40                   ` Minchan Kim

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=20091230051747.GA26364@localhost \
    --to=fengguang.wu@intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=andi@firstfloor.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=npiggin@suse.de \
    --cc=qbarnes+nfs@yahoo-inc.com \
    --cc=swhiteho@redhat.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;
as well as URLs for NNTP newsgroup(s).