public inbox for linux-xfs@vger.kernel.org
 help / color / mirror / Atom feed
From: Brian Foster <bfoster@redhat.com>
To: xfs@oss.sgi.com
Subject: [RFC PATCH 06/11] xfs: use correct transaction reservations in xfs_inactive()
Date: Tue,  3 Sep 2013 14:25:03 -0400	[thread overview]
Message-ID: <1378232708-57156-7-git-send-email-bfoster@redhat.com> (raw)
In-Reply-To: <1378232708-57156-1-git-send-email-bfoster@redhat.com>

The transaction allocated in xfs_inactive() can be passed down into
xfs_inactive_symlink() or xfs_itruncate_extents(), both of which
can commit and reallocate a new transaction. This leads to
reservation issues if the transaction is subsequently passed into
xfs_ifree(), which requires a larger reservation to manage the
finobt.

Reorganize xfs_inactive() to commit any transaction handed back
from symlink or truncate processing and unconditionally allocate
a new transaction for xfs_ifree() with the appropriate reservation.

Signed-off-by: Brian Foster <bfoster@redhat.com>
---
 fs/xfs/xfs_inode.c | 71 ++++++++++++++++++++++++++++--------------------------
 1 file changed, 37 insertions(+), 34 deletions(-)

diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index e3d7538..56cbf63 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -1679,7 +1679,6 @@ xfs_inactive(
 	int			committed;
 	struct xfs_trans	*tp;
 	struct xfs_mount	*mp;
-	struct xfs_trans_res	*resp;
 	int			error;
 	int			truncate = 0;
 
@@ -1724,33 +1723,39 @@ xfs_inactive(
 	if (error)
 		return VN_INACTIVE_CACHE;
 
-	tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE);
-	resp = (truncate || S_ISLNK(ip->i_d.di_mode)) ?
-		&M_RES(mp)->tr_itruncate : &M_RES(mp)->tr_ifree;
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
 
-	error = xfs_trans_reserve(tp, resp, 0, 0);
-	if (error) {
-		ASSERT(XFS_FORCED_SHUTDOWN(mp));
-		xfs_trans_cancel(tp, 0);
-		return VN_INACTIVE_CACHE;
-	}
+	if (truncate || S_ISLNK(ip->i_d.di_mode)) {
+		tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE);
 
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	xfs_trans_ijoin(tp, ip, 0);
+		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0);
+		if (error) {
+			ASSERT(XFS_FORCED_SHUTDOWN(mp));
+			xfs_trans_cancel(tp, 0);
+			return VN_INACTIVE_CACHE;
+		}
 
-	if (S_ISLNK(ip->i_d.di_mode)) {
-		error = xfs_inactive_symlink(ip, &tp);
-		if (error)
-			goto out_cancel;
-	} else if (truncate) {
-		ip->i_d.di_size = 0;
-		xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+		xfs_trans_ijoin(tp, ip, 0);
+
+		if (S_ISLNK(ip->i_d.di_mode)) {
+			error = xfs_inactive_symlink(ip, &tp);
+			if (error)
+				goto out_cancel;
+		} else if (truncate) {
+			ip->i_d.di_size = 0;
+			xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+
+			error = xfs_itruncate_extents(&tp, ip, XFS_DATA_FORK,
+						      0);
+			if (error)
+				goto out_cancel;
 
-		error = xfs_itruncate_extents(&tp, ip, XFS_DATA_FORK, 0);
+			ASSERT(ip->i_d.di_nextents == 0);
+		}
+
+		error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
 		if (error)
 			goto out_cancel;
-
-		ASSERT(ip->i_d.di_nextents == 0);
 	}
 
 	/*
@@ -1762,27 +1767,25 @@ xfs_inactive(
 	if (ip->i_d.di_anextents > 0) {
 		ASSERT(ip->i_d.di_forkoff != 0);
 
-		error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
-		if (error)
-			goto out_unlock;
-
 		xfs_iunlock(ip, XFS_ILOCK_EXCL);
 
 		error = xfs_attr_inactive(ip);
 		if (error)
 			goto out;
 
-		tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE);
-		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ifree, 0, 0);
-		if (error) {
-			xfs_trans_cancel(tp, 0);
-			goto out;
-		}
-
 		xfs_ilock(ip, XFS_ILOCK_EXCL);
-		xfs_trans_ijoin(tp, ip, 0);
 	}
 
+	tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE);
+	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ifree,
+				  XFS_IFREE_SPACE_RES(mp), 0);
+	if (error) {
+		xfs_trans_cancel(tp, 0);
+		goto out_unlock;
+	}
+
+	xfs_trans_ijoin(tp, ip, 0);
+
 	if (ip->i_afp)
 		xfs_idestroy_fork(ip, XFS_ATTR_FORK);
 
-- 
1.8.1.4

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

  parent reply	other threads:[~2013-09-03 18:25 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-09-03 18:24 [RFC PATCH 00/11] xfs: introduce the free inode btree Brian Foster
2013-09-03 18:24 ` [RFC PATCH 01/11] xfs: refactor xfs_ialloc_btree.c to support multiple inobt numbers Brian Foster
2013-09-05  0:36   ` Dave Chinner
2013-09-03 18:24 ` [RFC PATCH 02/11] xfs: reserve v5 superblock read-only compat. feature bit for finobt Brian Foster
2013-09-05  0:39   ` Dave Chinner
2013-09-03 18:25 ` [RFC PATCH 03/11] xfs: support the XFS_BTNUM_FINOBT free inode btree type Brian Foster
2013-09-05  0:54   ` Dave Chinner
2013-09-05 16:17     ` Brian Foster
2013-09-06  0:07       ` Dave Chinner
2013-09-06 11:25         ` Brian Foster
2013-09-06 21:22           ` Dave Chinner
2013-09-03 18:25 ` [RFC PATCH 04/11] xfs: update inode allocation transaction reservations for finobt Brian Foster
2013-09-05  0:59   ` Dave Chinner
2013-09-05 16:17     ` Brian Foster
2013-09-06  0:11       ` Dave Chinner
2013-09-03 18:25 ` [RFC PATCH 05/11] xfs: update ifree " Brian Foster
2013-09-05  1:00   ` Dave Chinner
2013-09-03 18:25 ` Brian Foster [this message]
2013-09-05  1:35   ` [RFC PATCH 06/11] xfs: use correct transaction reservations in xfs_inactive() Dave Chinner
2013-09-05 16:18     ` Brian Foster
2013-09-03 18:25 ` [RFC PATCH 07/11] xfs: retry trans reservation on ENOSPC " Brian Foster
2013-09-05  1:40   ` Dave Chinner
2013-09-05 16:18     ` Brian Foster
2013-09-06  0:17       ` Dave Chinner
2013-09-06 11:30         ` Brian Foster
2013-09-03 18:25 ` [RFC PATCH 08/11] xfs: insert newly allocated inode chunks into the finobt Brian Foster
2013-09-05  2:10   ` Dave Chinner
2013-09-03 18:25 ` [RFC PATCH 09/11] xfs: use and update the finobt on inode allocation Brian Foster
2013-09-05  2:27   ` Dave Chinner
2013-09-05 16:18     ` Brian Foster
2013-09-03 18:25 ` [RFC PATCH 10/11] xfs: update the finobt on inode free Brian Foster
2013-09-05  2:54   ` Dave Chinner
2013-09-05 16:19     ` Brian Foster
2013-09-06  0:28       ` Dave Chinner
2013-09-06 11:39         ` Brian Foster
2013-09-06 21:24           ` Dave Chinner
2013-09-07 12:30             ` Brian Foster
2013-09-08 20:08               ` Michael L. Semon
2013-09-09  2:34               ` Better numbers " Michael L. Semon
2013-09-03 18:25 ` [RFC PATCH 11/11] xfs: add finobt support to growfs Brian Foster
2013-09-05  2:55   ` Dave Chinner
2013-09-05 21:17 ` [RFC PATCH 00/11] xfs: introduce the free inode btree Michael L. Semon
2013-09-06 11:17   ` Brian Foster
2013-09-06 21:35   ` Dave Chinner
2013-09-07 12:31     ` Brian Foster
2013-09-08  1:04       ` Michael L. Semon

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=1378232708-57156-7-git-send-email-bfoster@redhat.com \
    --to=bfoster@redhat.com \
    --cc=xfs@oss.sgi.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox