Message-ID: <4A6D1651.5010102@sandeen.net>
Date: Sun, 26 Jul 2009 21:52:01 -0500
From: Eric Sandeen <sandeen@sandeen.net>
User-Agent: Thunderbird 2.0.0.22 (Macintosh/20090605)
MIME-Version: 1.0
To: Olaf Weber <olaf@sgi.com>
CC: xfs mailing list <xfs@oss.sgi.com>, Felix Blyakher <felixb@sgi.com>
Subject: [PATCH V4] reduce bmv_count in xfs_vn_fiemap
References: <4A5E2F01.7030107@sandeen.net>	<bzybpnkhp62.fsf@fransum.emea.sgi.com>	<4A5F3DBA.9040005@sandeen.net> <4A5F7B7B.6020903@sandeen.net>
In-Reply-To: <4A5F7B7B.6020903@sandeen.net>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

commit 6321e3ed2acf3ee9643cdd403e1c88605d7944ba caused
the full bmv_count's worth of getbmapx structures to get
allocated; telling it to do MAXEXTNUM was a bit insane,
resulting in ENOMEM every time.

Chop it down to something reasonable, the number of slots
in the caller's input buffer.  If this is too large the 
caller may get ENOMEM but the reason should not be a
mystery, and they can try again with something smaller.

We add 1 to the value because in the normal getbmap
world, bmv_count includes the header and xfs_getbmap does:

        nex = bmv->bmv_count - 1;
        if (nex <= 0)
                return XFS_ERROR(EINVAL);

Signed-off-by: Eric Sandeen <sandeen@sandeen.net>
---

V2: set it to the fiemap info's fi_extents_max
V3: bump up by one to accomodate expected bmv header in core
V4: comments

diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index 58973bb..8070b34 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -680,8 +680,8 @@ xfs_vn_fiemap(
 	else
 		bm.bmv_length = BTOBB(length);
 
-	/* our formatter will tell xfs_getbmap when to stop. */
-	bm.bmv_count = MAXEXTNUM;
+	/* We add one because in getbmap world count includes the header */
+	bm.bmv_count = fieinfo->fi_extents_max + 1;
 	bm.bmv_iflags = BMV_IF_PREALLOC;
 	if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR)
 		bm.bmv_iflags |= BMV_IF_ATTRFORK;