linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: npiggin@suse.de
To: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-fsdevel@vger.kernel.org,
	Linux Memory Management <linux-mm@kvack.org>
Subject: [patch 06/41] mm: trim more holes
Date: Mon, 14 May 2007 16:06:25 +1000	[thread overview]
Message-ID: <20070514060650.851942000@wotan.suse.de> (raw)
In-Reply-To: 20070514060619.689648000@wotan.suse.de

[-- Attachment #1: mm-trim-more-holes.patch --]
[-- Type: text/plain, Size: 3382 bytes --]


If prepare_write fails with AOP_TRUNCATED_PAGE, or if commit_write fails, then
we may have failed the write operation despite prepare_write having
instantiated blocks past i_size. Fix this, and consolidate the trimming into
one place.

Cc: Linux Memory Management <linux-mm@kvack.org>
Cc: Linux Filesystems <linux-fsdevel@vger.kernel.org>
Signed-off-by: Nick Piggin <npiggin@suse.de>

 mm/filemap.c |   80 +++++++++++++++++++++++++++++------------------------------
 1 file changed, 40 insertions(+), 40 deletions(-)

Index: linux-2.6/mm/filemap.c
===================================================================
--- linux-2.6.orig/mm/filemap.c
+++ linux-2.6/mm/filemap.c
@@ -1957,22 +1957,9 @@ generic_file_buffered_write(struct kiocb
 		}
 
 		status = a_ops->prepare_write(file, page, offset, offset+bytes);
-		if (unlikely(status)) {
-			loff_t isize = i_size_read(inode);
+		if (unlikely(status))
+			goto fs_write_aop_error;
 
-			if (status != AOP_TRUNCATED_PAGE)
-				unlock_page(page);
-			page_cache_release(page);
-			if (status == AOP_TRUNCATED_PAGE)
-				continue;
-			/*
-			 * prepare_write() may have instantiated a few blocks
-			 * outside i_size.  Trim these off again.
-			 */
-			if (pos + bytes > isize)
-				vmtruncate(inode, isize);
-			break;
-		}
 		if (likely(nr_segs == 1))
 			copied = filemap_copy_from_user(page, offset,
 							buf, bytes);
@@ -1981,40 +1968,53 @@ generic_file_buffered_write(struct kiocb
 						cur_iov, iov_offset, bytes);
 		flush_dcache_page(page);
 		status = a_ops->commit_write(file, page, offset, offset+bytes);
-		if (status == AOP_TRUNCATED_PAGE) {
-			page_cache_release(page);
-			continue;
+		if (unlikely(status < 0 || status == AOP_TRUNCATED_PAGE))
+			goto fs_write_aop_error;
+		if (unlikely(copied != bytes)) {
+			status = -EFAULT;
+			goto fs_write_aop_error;
 		}
-		if (likely(copied > 0)) {
-			if (!status)
-				status = copied;
+		if (unlikely(status > 0)) /* filesystem did partial write */
+			copied = status;
 
-			if (status >= 0) {
-				written += status;
-				count -= status;
-				pos += status;
-				buf += status;
-				if (unlikely(nr_segs > 1)) {
-					filemap_set_next_iovec(&cur_iov,
-							&iov_offset, status);
-					if (count)
-						buf = cur_iov->iov_base +
-							iov_offset;
-				} else {
-					iov_offset += status;
-				}
+		if (likely(copied > 0)) {
+			written += copied;
+			count -= copied;
+			pos += copied;
+			buf += copied;
+			if (unlikely(nr_segs > 1)) {
+				filemap_set_next_iovec(&cur_iov,
+						&iov_offset, copied);
+				if (count)
+					buf = cur_iov->iov_base + iov_offset;
+			} else {
+				iov_offset += copied;
 			}
 		}
-		if (unlikely(copied != bytes))
-			if (status >= 0)
-				status = -EFAULT;
 		unlock_page(page);
 		mark_page_accessed(page);
 		page_cache_release(page);
-		if (status < 0)
-			break;
 		balance_dirty_pages_ratelimited(mapping);
 		cond_resched();
+		continue;
+
+fs_write_aop_error:
+		if (status != AOP_TRUNCATED_PAGE)
+			unlock_page(page);
+		page_cache_release(page);
+
+		/*
+		 * prepare_write() may have instantiated a few blocks
+		 * outside i_size.  Trim these off again. Don't need
+		 * i_size_read because we hold i_mutex.
+		 */
+		if (pos + bytes > inode->i_size)
+			vmtruncate(inode, inode->i_size);
+		if (status == AOP_TRUNCATED_PAGE)
+			continue;
+		else
+			break;
+
 	} while (count);
 	*ppos = pos;
 

-- 


  parent reply	other threads:[~2007-05-14  8:23 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-05-14  6:06 [patch 00/41] Buffered write deadlock fix and new aops for 2.6.21-mm2 npiggin
2007-05-14  6:06 ` [patch 01/41] mm: revert KERNEL_DS buffered write optimisation npiggin
2007-05-14  6:06 ` [patch 02/41] Revert 81b0c8713385ce1b1b9058e916edcf9561ad76d6 npiggin
2007-05-14 19:06   ` Dave Jones
2007-05-14 22:45     ` Nick Piggin
2007-05-14  6:06 ` [patch 03/41] Revert 6527c2bdf1f833cc18e8f42bd97973d583e4aa83 npiggin
2007-05-14  6:06 ` [patch 04/41] mm: clean up buffered write code npiggin
2007-05-14  6:06 ` [patch 05/41] mm: debug write deadlocks npiggin
2007-05-14  6:06 ` npiggin [this message]
2007-05-14  6:06 ` [patch 07/41] mm: buffered write cleanup npiggin
2007-05-14  6:06 ` [patch 08/41] mm: write iovec cleanup npiggin
2007-05-14  6:06 ` [patch 09/41] mm: fix pagecache write deadlocks npiggin
2007-05-14  6:06 ` [patch 10/41] mm: buffered write iterator npiggin
2007-05-14  6:06 ` [patch 11/41] fs: fix data-loss on error npiggin
2007-05-14  6:06 ` [patch 12/41] fs: introduce write_begin, write_end, and perform_write aops npiggin
2007-05-14  6:06 ` [patch 13/41] mm: restore KERNEL_DS optimisations npiggin
2007-05-14  6:06 ` [patch 14/41] implement simple fs aops npiggin
2007-05-14  6:06 ` [patch 15/41] block_dev convert to new aops npiggin
2007-05-14  6:06 ` [patch 16/41] rd " npiggin
2007-05-14  6:06 ` [patch 17/41] ext2 " npiggin
2007-05-16 12:45   ` Nick Piggin
2007-05-14  6:06 ` [patch 18/41] ext3 " npiggin
2007-05-14  6:06 ` [patch 19/41] ext4 " npiggin
2007-05-14  6:06 ` [patch 20/41] xfs " npiggin
2007-05-14  6:06 ` [patch 21/41] fs: new cont helpers npiggin
2007-05-14  6:06 ` [patch 22/41] fat convert to new aops npiggin
2007-05-14  6:06 ` [patch 23/41] adfs " npiggin
2007-05-14  6:06 ` [patch 24/41] hfs " npiggin
2007-05-14  6:06 ` [patch 25/41] hfsplus " npiggin
2007-05-14  6:06 ` [patch 26/41] hpfs " npiggin
2007-05-14  6:06 ` [patch 27/41] bfs " npiggin
2007-05-14  6:06 ` [patch 28/41] qnx4 " npiggin
2007-05-14 10:37   ` Anders Larsen
2007-05-14  6:06 ` [patch 29/41] reiserfs use generic write npiggin
2007-05-14  6:06 ` [patch 30/41] reiserfs convert to new aops npiggin
2007-05-14  6:06 ` [patch 31/41] nfs " npiggin
2007-05-14  6:06 ` [patch 32/41] smb " npiggin
2007-05-14  6:06 ` [patch 33/41] GFS2 " npiggin
2007-05-14  6:06 ` [patch 34/41] fuse " npiggin
2007-05-14  6:06 ` [patch 35/41] hostfs " npiggin
2007-05-14  6:06 ` [patch 36/41] jffs2 " npiggin
2007-05-14  6:06 ` [patch 37/41] ufs " npiggin
2007-05-14  6:06 ` [patch 38/41] udf " npiggin
2007-05-14  6:06 ` [patch 39/41] sysv " npiggin
2007-05-14  6:06 ` [patch 40/41] minix " npiggin
2007-05-14  6:07 ` [patch 41/41] jfs " npiggin
     [not found] ` <200705142014.27237.vs@namesys.com>
     [not found]   ` <20070514224042.GD5531@wotan.suse.de>
2007-05-16 17:22     ` [patch 30/41] reiserfs " Vladimir V. Saveliev
  -- strict thread matches above, loose matches on Subject: below --
2007-05-25 12:21 [patch 00/41] Buffered write deadlock fix and new aops for 2.6.22-rc2-mm1 npiggin
2007-05-25 12:21 ` [patch 06/41] mm: trim more holes npiggin

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=20070514060650.851942000@wotan.suse.de \
    --to=npiggin@suse.de \
    --cc=akpm@linux-foundation.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-mm@kvack.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).