Re: mmotm 2009-01-05-12-50 uploaded (ubifs)

[Andrew Morton]

On Mon, 05 Jan 2009 14:09:25 -0800
Randy Dunlap <randy.dunlap@xxxxxxxxxx> wrote:

> akpm@xxxxxxxxxxxxxxxxxxxx wrote:
> > The mm-of-the-moment snapshot 2009-01-05-12-50 has been uploaded to
> > 
> >    http://userweb.kernel.org/~akpm/mmotm/
> > 
> > and will soon be available at
> > 
> >    git://git.zen-sources.org/zen/mmotm.git
> 
> 
> mmotm-2009-0105-1250/fs/ubifs/super.c:435: error: 'WB_SYNC_HOLD' undeclared (first use in this function)
> 

It's amazing how much shiny new code turns up during the merge window
and screws things up.


commit 304d427cd99eb645b44b08d77e70ce308e6bcd8c
Author:     Artem Bityutskiy <Artem.Bityutskiy@xxxxxxxxx>
AuthorDate: Sun Dec 28 08:04:17 2008 +0200
Commit:     Artem Bityutskiy <Artem.Bityutskiy@xxxxxxxxx>
CommitDate: Wed Dec 31 14:13:24 2008 +0200

    UBIFS: fix file-system synchronization
    
    Argh. The ->sync_fs call is called _before_ all inodes are flushed.
    This means we first sync write buffers and commit, then all
    inodes are synced, and we end up with unflushed write buffers!
    
    Fix this by forcing synching all indoes from 'ubifs_sync_fs()'.
    
    Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@xxxxxxxxx>

diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index 1309783..4713017 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -35,6 +35,7 @@
 #include <linux/seq_file.h>
 #include <linux/mount.h>
 #include <linux/math64.h>
+#include <linux/writeback.h>
 #include "ubifs.h"
 
 /*
@@ -431,6 +432,23 @@ static int ubifs_sync_fs(struct super_block *sb, int wait)
 	struct ubifs_info *c = sb->s_fs_info;
 	int i, ret = 0, err;
 	long long bud_bytes;
+	struct writeback_control wbc = {
+		.sync_mode   = wait ? WB_SYNC_ALL : WB_SYNC_HOLD,
+		.range_start = 0,
+		.range_end   = LLONG_MAX,
+		.nr_to_write = LONG_MAX,
+	};
+
+	/*
+	 * VFS calls '->sync_fs()' before synchronizing all dirty inodes and
+	 * pages, so synchronize them first, then commit the journal. Strictly
+	 * speaking, it is not necessary to commit the journal here,
+	 * synchronizing write-buffers would be enough. But committing makes
+	 * UBIFS free space predictions much more accurate, so we want to let
+	 * the user be able to get more accurate results of 'statfs()' after
+	 * they synchronize the file system.
+	 */
+	generic_sync_sb_inodes(sb, &wbc);
 
 	if (c->jheads) {
 		for (i = 0; i < c->jhead_cnt; i++) {


WB_SYNC_HOLD got removed by
http://userweb.kernel.org/~akpm/mmotm/broken-out/fs-remove-wb_sync_hold.patch

I think I'll just switch that to WB_SYNC_NONE.  The `wait==0' mode is
just an advisory thing to help the fs shove lots of data into the
queues.  If some gets missed then it'll be picked up on the second
->sync_fs call, with wait==1.



--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/