linux-kernel - Re: [PATCH 5/6] writeback: don't delay inodes redirtied by a fast dirtier

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20090927020233.GA10318@localhost>
Date:	Sun, 27 Sep 2009 10:02:34 +0800
From:	Wu Fengguang <fengguang.wu@...el.com>
To:	Christoph Hellwig <hch@...radead.org>
Cc:	Andrew Morton <akpm@...ux-foundation.org>,
	Jens Axboe <jens.axboe@...cle.com>, Jan Kara <jack@...e.cz>,
	Theodore Tso <tytso@....edu>,
	Dave Chinner <david@...morbit.com>,
	Chris Mason <chris.mason@...cle.com>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	"linux-fsdevel@...r.kernel.org" <linux-fsdevel@...r.kernel.org>,
	LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 5/6] writeback: don't delay inodes redirtied by a fast
	dirtier

On Sun, Sep 27, 2009 at 03:47:47AM +0800, Christoph Hellwig wrote:
> On Wed, Sep 23, 2009 at 08:33:43PM +0800, Wu Fengguang wrote:
> > So let's distinguish between data redirty and metadata only redirty.
> > The first one is caused by a busy dirtier, while the latter one could
> > happen in XFS, NFS, etc. when they are doing delalloc or updating isize.
> 
> Btw, I'm not sure the existing and preserved behaviour for that case
> is good.  In the worst case the inode writeout gets delayed by another
> 30 seconds, doubling the window until a file is on disk if it was
> extended.

Yes, the preserved behaviour is not optimal for XFS, but safe.

We could try redirty_tail when there are no remaining
dirty pages, and only metadata dirtiness. Like this:

---
 fs/fs-writeback.c |   20 +++++++-------------
 1 file changed, 7 insertions(+), 13 deletions(-)

--- linux.orig/fs/fs-writeback.c	2009-09-27 09:52:15.000000000 +0800
+++ linux/fs/fs-writeback.c	2009-09-27 09:54:23.000000000 +0800
@@ -477,18 +477,7 @@ writeback_single_inode(struct inode *ino
 	spin_lock(&inode_lock);
 	inode->i_state &= ~I_SYNC;
 	if (!(inode->i_state & (I_FREEING | I_CLEAR))) {
-		if ((inode->i_state & I_DIRTY_PAGES) && wbc->for_kupdate) {
-			/*
-			 * More pages get dirtied by a fast dirtier.
-			 */
-			goto select_queue;
-		} else if (inode->i_state & I_DIRTY) {
-			/*
-			 * At least XFS will redirty the inode during the
-			 * writeback (delalloc) and on io completion (isize).
-			 */
-			redirty_tail(inode);
-		} else if (mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) {
+		if (mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) {
 			/*
 			 * We didn't write back all the pages.  nfs_writepages()
 			 * sometimes bales out without doing anything. Redirty
@@ -510,7 +499,6 @@ writeback_single_inode(struct inode *ino
 				 * soon as the queue becomes uncongested.
 				 */
 				inode->i_state |= I_DIRTY_PAGES;
-select_queue:
 				if (wbc->nr_to_write <= 0) {
 					/*
 					 * slice used up: queue for next turn
@@ -533,6 +521,12 @@ select_queue:
 				inode->i_state |= I_DIRTY_PAGES;
 				redirty_tail(inode);
 			}
+		} else if (inode->i_state & I_DIRTY) {
+			/*
+			 * At least XFS will redirty the inode during the
+			 * writeback (delalloc) and on io completion (isize).
+			 */
+			redirty_tail(inode);
 		} else if (atomic_read(&inode->i_count)) {
 			/*
 			 * The inode is clean, inuse
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/