Re: [PATCH 0/8][RFC] IO latency/throughput fixes

[Linus Torvalds <torvalds@linux-foundation.org>]

On Mon, 6 Apr 2009, Linus Torvalds wrote:
> thing that we think people would be happiest with.
> 
> I think "ordered" was a reasonable default, but that was at least partly 
> because _both_ ordered and writeback sucked (partly in different ways).
> 
> I do think we could make it a config option.

A patch _something_ like this.

A few notes:

 - This is UNTESTED (of course)

 - If I did this right, this _only_ overrides the data mode if it's not 
   explicitly specified on disk in the superblock mount options.

IOW, if you have done a 

	tune2fs -o journal_data_ordered

then this will _not_ override that. Only in the absense of any explicit 
flags should this trigger and then make the choice be 'writeback'.

And just to be _extra_ backwards compatible, if you really want the old 
behavior, and don't want to set the ordering flag explicitly, just answer 
'y' to the EXT3_DEFAULTS_TO_ORDERED Kconfig question. 

What do people think? Anybody want to test?

		Linus

---
 fs/ext3/Kconfig |   19 +++++++++++++++++++
 fs/ext3/super.c |    8 +++++++-
 2 files changed, 26 insertions(+), 1 deletions(-)

diff --git a/fs/ext3/Kconfig b/fs/ext3/Kconfig
index 8e0cfe4..fb3c1a2 100644
--- a/fs/ext3/Kconfig
+++ b/fs/ext3/Kconfig
@@ -28,6 +28,25 @@ config EXT3_FS
 	  To compile this file system support as a module, choose M here: the
 	  module will be called ext3.
 
+config EXT3_DEFAULTS_TO_ORDERED
+	bool "Default to 'data=ordered' in ext3 (legacy option)"
+	depends on EXT3_FS
+	help
+	  If a filesystem does not explicitly specify a data ordering
+	  mode, and the journal capability allowed it, ext3 used to
+	  historically default to 'data=ordered'.
+
+	  That was a rather unfortunate choice, because it leads to all
+	  kinds of latency problems, and the 'data=writeback' mode is more
+	  appropriate these days.
+
+	  You should probably always answer 'n' here, and if you really
+	  want to use 'data=ordered' mode, set it in the filesystem itself
+	  with 'tune2fs -o journal_data_ordered'.
+
+	  But if you really want to enable the legacy default, you can do
+	  so by answering 'y' to this question.
+
 config EXT3_FS_XATTR
 	bool "Ext3 extended attributes"
 	depends on EXT3_FS
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 9e5b8e3..599dbfe 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -44,6 +44,12 @@
 #include "acl.h"
 #include "namei.h"
 
+#ifdef CONFIG_EXT3_DEFAULTS_TO_ORDERED
+  #define EXT3_MOUNT_DEFAULT_DATA_MODE EXT3_MOUNT_ORDERED_DATA
+#else
+  #define EXT3_MOUNT_DEFAULT_DATA_MODE EXT3_MOUNT_WRITEBACK_DATA
+#endif
+
 static int ext3_load_journal(struct super_block *, struct ext3_super_block *,
 			     unsigned long journal_devnum);
 static int ext3_create_journal(struct super_block *, struct ext3_super_block *,
@@ -1919,7 +1925,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
                    cope, else JOURNAL_DATA */
 		if (journal_check_available_features
 		    (sbi->s_journal, 0, 0, JFS_FEATURE_INCOMPAT_REVOKE))
-			set_opt(sbi->s_mount_opt, ORDERED_DATA);
+			set_opt(sbi->s_mount_opt, DEFAULT_DATA_MODE);
 		else
 			set_opt(sbi->s_mount_opt, JOURNAL_DATA);
 		break;