From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: with ECARTIS (v1.0.0; list xfs); Wed, 22 Oct 2008 09:29:46 -0700 (PDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.168.28]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m9MGTiYl027562 for ; Wed, 22 Oct 2008 09:29:44 -0700 Received: from bombadil.infradead.org (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 14A8510C97BA for ; Wed, 22 Oct 2008 09:31:29 -0700 (PDT) Received: from bombadil.infradead.org (bombadil.infradead.org [18.85.46.34]) by cuda.sgi.com with ESMTP id aMyFgyB2T1COXOfs for ; Wed, 22 Oct 2008 09:31:29 -0700 (PDT) Received: from hch by bombadil.infradead.org with local (Exim 4.68 #1 (Red Hat Linux)) id 1KsgcL-0000Lu-Dm for xfs@oss.sgi.com; Wed, 22 Oct 2008 16:31:29 +0000 Date: Wed, 22 Oct 2008 12:31:29 -0400 From: Christoph Hellwig Subject: [PATCH 2/2] free partially initialized inodes using destroy_inode Message-ID: <20081022163129.GC25556@infradead.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Sender: xfs-bounce@oss.sgi.com Errors-to: xfs-bounce@oss.sgi.com List-Id: xfs To: xfs@oss.sgi.com To make sure we free the security data inodes need to be freed using the proper VFS helper (which we also need to export for this). We mark these inodes bad so we can skip the flush path for them. Signed-off-by: Christoph Hellwig Index: xfs-2.6/fs/inode.c =================================================================== --- xfs-2.6.orig/fs/inode.c 2008-10-22 17:57:30.000000000 +0200 +++ xfs-2.6/fs/inode.c 2008-10-22 17:58:22.000000000 +0200 @@ -212,6 +212,7 @@ void destroy_inode(struct inode *inode) else kmem_cache_free(inode_cachep, (inode)); } +EXPORT_SYMBOL(destroy_inode); /* Index: xfs-2.6/fs/xfs/xfs_iget.c =================================================================== --- xfs-2.6.orig/fs/xfs/xfs_iget.c 2008-10-22 17:57:30.000000000 +0200 +++ xfs-2.6/fs/xfs/xfs_iget.c 2008-10-22 17:58:22.000000000 +0200 @@ -201,7 +201,7 @@ out_unlock: if (lock_flags) xfs_iunlock(ip, lock_flags); out_destroy: - xfs_idestroy(ip); + xfs_destroy_inode(ip); return error; } Index: xfs-2.6/fs/xfs/xfs_inode.c =================================================================== --- xfs-2.6.orig/fs/xfs/xfs_inode.c 2008-10-22 17:57:30.000000000 +0200 +++ xfs-2.6/fs/xfs/xfs_inode.c 2008-10-22 17:58:22.000000000 +0200 @@ -898,18 +898,14 @@ xfs_iread( * know that this is a new incore inode. */ error = xfs_itobp(mp, tp, ip, &dip, &bp, bno, imap_flags, XFS_BUF_LOCK); - if (error) { - xfs_idestroy(ip); - return error; - } + if (error) + goto out_destroy_inode; /* * If we got something that isn't an inode it means someone * (nfs or dmi) has a stale handle. */ if (be16_to_cpu(dip->di_core.di_magic) != XFS_DINODE_MAGIC) { - xfs_idestroy(ip); - xfs_trans_brelse(tp, bp); #ifdef DEBUG xfs_fs_cmn_err(CE_ALERT, mp, "xfs_iread: " "dip->di_core.di_magic (0x%x) != " @@ -917,7 +913,8 @@ xfs_iread( be16_to_cpu(dip->di_core.di_magic), XFS_DINODE_MAGIC); #endif /* DEBUG */ - return XFS_ERROR(EINVAL); + error = XFS_ERROR(EINVAL); + goto out_brelse; } /* @@ -931,14 +928,12 @@ xfs_iread( xfs_dinode_from_disk(&ip->i_d, &dip->di_core); error = xfs_iformat(ip, dip); if (error) { - xfs_idestroy(ip); - xfs_trans_brelse(tp, bp); #ifdef DEBUG xfs_fs_cmn_err(CE_ALERT, mp, "xfs_iread: " "xfs_iformat() returned error %d", error); #endif /* DEBUG */ - return error; + goto out_brelse; } } else { ip->i_d.di_magic = be16_to_cpu(dip->di_core.di_magic); @@ -1004,6 +999,12 @@ xfs_iread( xfs_trans_brelse(tp, bp); *ipp = ip; return 0; + + out_brelse: + xfs_trans_brelse(tp, bp); + out_destroy_inode: + xfs_destroy_inode(ip); + return error; } /* Index: xfs-2.6/fs/xfs/xfs_inode.h =================================================================== --- xfs-2.6.orig/fs/xfs/xfs_inode.h 2008-10-22 17:57:30.000000000 +0200 +++ xfs-2.6/fs/xfs/xfs_inode.h 2008-10-22 18:00:15.000000000 +0200 @@ -310,6 +310,23 @@ static inline struct inode *VFS_I(struct } /* + * Get rid of a partially initialized inode. + * + * We have to go through destroy_inode to make sure allocations + * from init_inode_always like the security data are undone. + * + * We mark the inode dirty so that it takes the short cut in + * the reclaim path instead of going through the flush path + * which doesn't make sense for an inode that has never seen the + * light of day. + */ +static inline void xfs_destroy_inode(struct xfs_inode *ip) +{ + make_bad_inode(VFS_I(ip)); + return destroy_inode(VFS_I(ip)); +} + +/* * i_flags helper functions */ static inline void