All of lore.kernel.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@lst.de>
To: linux-fsdevel@vger.kernel.org
Subject: [PATCH 11/11] hfsplus: flush disk caches in sync and fsync
Date: Wed, 17 Nov 2010 23:23:22 +0100	[thread overview]
Message-ID: <20101117222322.GL21700@lst.de> (raw)
In-Reply-To: <20101117222117.GA21700@lst.de>

Flush the disk cache in fsync and sync to make sure data actually is
on disk on completion of these system calls.  There is a nobarrier
mount option to disable this behaviour.  It's slightly misnamed now
that barrier actually are gone, but it matches the name used by all
major filesystems.

Signed-off-by: Christoph Hellwig <hch@tuxera.com>

Index: linux-2.6/fs/hfsplus/inode.c
===================================================================
--- linux-2.6.orig/fs/hfsplus/inode.c	2010-11-17 22:56:01.488273225 +0100
+++ linux-2.6/fs/hfsplus/inode.c	2010-11-17 23:03:35.845003983 +0100
@@ -8,6 +8,7 @@
  * Inode handling routines
  */
 
+#include <linux/blkdev.h>
 #include <linux/mm.h>
 #include <linux/fs.h>
 #include <linux/pagemap.h>
@@ -337,6 +338,9 @@ int hfsplus_file_fsync(struct file *file
 			error = error2;
 	}
 
+	if (!test_bit(HFSPLUS_SB_NOBARRIER, &sbi->flags))
+		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
+
 	return error;
 }
 
Index: linux-2.6/fs/hfsplus/super.c
===================================================================
--- linux-2.6.orig/fs/hfsplus/super.c	2010-11-17 22:56:01.505005590 +0100
+++ linux-2.6/fs/hfsplus/super.c	2010-11-17 23:03:43.735253739 +0100
@@ -10,6 +10,7 @@
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/pagemap.h>
+#include <linux/blkdev.h>
 #include <linux/fs.h>
 #include <linux/slab.h>
 #include <linux/vfs.h>
@@ -212,6 +213,10 @@ int hfsplus_sync_fs(struct super_block *
 out:
 	mutex_unlock(&sbi->alloc_mutex);
 	mutex_unlock(&sbi->vh_mutex);
+
+	if (!test_bit(HFSPLUS_SB_NOBARRIER, &sbi->flags))
+		blkdev_issue_flush(sb->s_bdev, GFP_KERNEL, NULL);
+
 	return error;
 }
 
Index: linux-2.6/fs/hfsplus/hfsplus_fs.h
===================================================================
--- linux-2.6.orig/fs/hfsplus/hfsplus_fs.h	2010-11-17 22:56:01.488273225 +0100
+++ linux-2.6/fs/hfsplus/hfsplus_fs.h	2010-11-17 22:57:48.130255066 +0100
@@ -156,6 +156,7 @@ struct hfsplus_sb_info {
 #define HFSPLUS_SB_FORCE	2
 #define HFSPLUS_SB_HFSX		3
 #define HFSPLUS_SB_CASEFOLD	4
+#define HFSPLUS_SB_NOBARRIER	5
 
 static inline struct hfsplus_sb_info *HFSPLUS_SB(struct super_block *sb)
 {
Index: linux-2.6/fs/hfsplus/options.c
===================================================================
--- linux-2.6.orig/fs/hfsplus/options.c	2010-11-17 22:43:05.314004052 +0100
+++ linux-2.6/fs/hfsplus/options.c	2010-11-17 22:57:48.135254228 +0100
@@ -23,6 +23,7 @@ enum {
 	opt_umask, opt_uid, opt_gid,
 	opt_part, opt_session, opt_nls,
 	opt_nodecompose, opt_decompose,
+	opt_barrier, opt_nobarrier,
 	opt_force, opt_err
 };
 
@@ -37,6 +38,8 @@ static const match_table_t tokens = {
 	{ opt_nls, "nls=%s" },
 	{ opt_decompose, "decompose" },
 	{ opt_nodecompose, "nodecompose" },
+	{ opt_barrier, "barrier" },
+	{ opt_nobarrier, "nobarrier" },
 	{ opt_force, "force" },
 	{ opt_err, NULL }
 };
@@ -174,6 +177,12 @@ int hfsplus_parse_options(char *input, s
 		case opt_nodecompose:
 			set_bit(HFSPLUS_SB_NODECOMPOSE, &sbi->flags);
 			break;
+		case opt_barrier:
+			clear_bit(HFSPLUS_SB_NOBARRIER, &sbi->flags);
+			break;
+		case opt_nobarrier:
+			set_bit(HFSPLUS_SB_NOBARRIER, &sbi->flags);
+			break;
 		case opt_force:
 			set_bit(HFSPLUS_SB_FORCE, &sbi->flags);
 			break;
@@ -212,5 +221,7 @@ int hfsplus_show_options(struct seq_file
 		seq_printf(seq, ",nls=%s", sbi->nls->charset);
 	if (test_bit(HFSPLUS_SB_NODECOMPOSE, &sbi->flags))
 		seq_printf(seq, ",nodecompose");
+	if (test_bit(HFSPLUS_SB_NOBARRIER, &sbi->flags))
+		seq_printf(seq, ",nobarrier");
 	return 0;
 }

      parent reply	other threads:[~2010-11-17 22:23 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-11-17 22:21 hfsplus patch review Christoph Hellwig
2010-11-17 22:21 ` [PATCH 1/11] hfsplus: silence a few debug printks Christoph Hellwig
2010-11-17 22:21 ` [PATCH 2/11] hfsplus: always use hfsplus_sync_fs to write the volume header Christoph Hellwig
2010-11-17 22:22 ` [PATCH 3/11] hfsplus: use raw bio access for the volume headers Christoph Hellwig
2010-11-17 22:22 ` [PATCH 4/11] hfsplus: use raw bio access for partition tables Christoph Hellwig
2010-11-17 22:22 ` [PATCH 5/11] hfsplus: make sure sync writes out all metadata Christoph Hellwig
2010-11-17 22:22 ` [PATCH 6/11] hfsplus: avoid useless work in hfsplus_sync_fs Christoph Hellwig
2010-11-17 22:22 ` [PATCH 7/11] hfsplus: simplify fsync Christoph Hellwig
2010-11-17 22:22 ` [PATCH 8/11] hfsplus: write up fsync for directories Christoph Hellwig
2010-11-17 22:23 ` [PATCH 9/11] hfsplus: split up inode flags Christoph Hellwig
2010-11-17 22:23 ` [PATCH 10/11] hfsplus: optimize fsync Christoph Hellwig
2010-11-18  6:40   ` Nick Piggin
2010-11-18 13:50     ` Christoph Hellwig
2010-11-18 14:13       ` Nick Piggin
2010-11-18 14:16         ` Christoph Hellwig
2010-11-22 13:03           ` Nick Piggin
2010-11-22 13:18             ` Nick Piggin
2010-11-22 13:29               ` Christoph Hellwig
2010-11-22 11:35     ` [PATCH v2] " Christoph Hellwig
2010-11-17 22:23 ` Christoph Hellwig [this message]

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=20101117222322.GL21700@lst.de \
    --to=hch@lst.de \
    --cc=linux-fsdevel@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.