From: Mingming Cao <cmm@us.ibm.com>
To: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: Eric Sandeen <sandeen@redhat.com>, ext4 <linux-ext4@vger.kernel.org>
Subject: Re: BUG with delayed allocation
Date: Wed, 19 Mar 2008 17:46:58 -0700 [thread overview]
Message-ID: <1205974018.3637.9.camel@localhost.localdomain> (raw)
In-Reply-To: <20080319085235.GA6752@skywalker>
On Wed, 2008-03-19 at 14:22 +0530, Aneesh Kumar K.V wrote:
> Hi,
>
> Eric actually observed it yesterday. I am able to reproduce it locally.
> With delayed allocation we are observing wrong value of i_size.
>
The problem is current delalloc does not update on-disk i_size until
writeout time, the in-core i_size is updated though.
Could you try the following patch? It updates the i_disksize at the
write_end time.
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
---
fs/ext4/inode.c | 39 ++++++++++++++++++++++++++++++++++++++-
1 file changed, 38 insertions(+), 1 deletion(-)
Index: linux-2.6.25-rc5/fs/ext4/inode.c
===================================================================
--- linux-2.6.25-rc5.orig/fs/ext4/inode.c 2008-03-19 17:32:44.000000000 -0700
+++ linux-2.6.25-rc5/fs/ext4/inode.c 2008-03-19 17:43:19.000000000 -0700
@@ -1355,6 +1355,43 @@ static int ext4_writeback_write_end(stru
return ret ? ret : copied;
}
+static int ext4_da_write_end(struct file *file,
+ struct address_space *mapping,
+ loff_t pos, unsigned len, unsigned copied,
+ struct page *page, void *fsdata)
+{
+ handle_t *handle;
+ struct inode *inode = file->f_mapping->host;
+ int needed_blocks = ext4_writepage_trans_blocks(inode);
+ int ret = 0, ret2;
+ loff_t new_i_size;
+
+ handle = ext4_journal_start(inode, needed_blocks);
+ if (IS_ERR(handle)) {
+ unlock_page(page);
+ page_cache_release(page);
+ ret = PTR_ERR(handle);
+ return ret;
+ }
+
+ new_i_size = pos + copied;
+ if (new_i_size > EXT4_I(inode)->i_disksize)
+ EXT4_I(inode)->i_disksize = new_i_size;
+
+ copied = ext4_generic_write_end(file, mapping, pos, len, copied,
+ page, fsdata);
+ if (copied < 0)
+ ret = copied;
+
+ ret2 = ext4_journal_stop(handle);
+ if (!ret)
+ ret = ret2;
+ unlock_page(page);
+ page_cache_release(page);
+
+ return ret ? ret : copied;
+}
+
static int ext4_journalled_write_end(struct file *file,
struct address_space *mapping,
loff_t pos, unsigned len, unsigned copied,
@@ -2020,7 +2057,7 @@ static const struct address_space_operat
.writepages = ext4_da_writepages,
.sync_page = block_sync_page,
.write_begin = ext4_da_write_begin,
- .write_end = generic_write_end,
+ .write_end = ext4_da_write_end,
.bmap = ext4_bmap,
.invalidatepage = ext4_da_invalidatepage,
.releasepage = ext4_releasepage,
next prev parent reply other threads:[~2008-03-20 0:47 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-03-19 8:52 BUG with delayed allocation Aneesh Kumar K.V
2008-03-19 14:42 ` Eric Sandeen
2008-03-19 18:32 ` Aneesh Kumar K.V
2008-03-20 4:21 ` Aneesh Kumar K.V
2008-03-20 0:46 ` Mingming Cao [this message]
2008-03-20 5:39 ` Aneesh Kumar K.V
2008-03-20 17:29 ` Mingming Cao
2008-03-20 17:56 ` Aneesh Kumar K.V
2008-03-20 23:55 ` Andreas Dilger
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=1205974018.3637.9.camel@localhost.localdomain \
--to=cmm@us.ibm.com \
--cc=aneesh.kumar@linux.vnet.ibm.com \
--cc=linux-ext4@vger.kernel.org \
--cc=sandeen@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 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.