From rdunlop@guralp.com Wed Apr 23 10:22:32 2008 Received: from a.painless.aaisp.net.uk ([81.187.30.51]) by stoneboat.aleph1.co.uk with esmtps (TLS-1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.63) (envelope-from ) id 1JobBA-000427-J8 for yaffs@lists.aleph1.co.uk; Wed, 23 Apr 2008 10:22:31 +0100 Received: from home.pinkhut.co.uk ([81.187.28.34] helo=neddie.localdomain) by a.painless.aaisp.net.uk with esmtp (Exim 4.69) (envelope-from ) id 1JobB9-0004f1-QC for yaffs@lists.aleph1.co.uk; Wed, 23 Apr 2008 10:22:15 +0100 Received: by neddie.localdomain (Postfix, from userid 1000) id 8B05215E441; Wed, 23 Apr 2008 10:22:13 +0100 (BST) Date: Wed, 23 Apr 2008 10:22:13 +0100 From: Bob Dunlop To: yaffs@lists.aleph1.co.uk Message-ID: <20080423092213.GA29474@guralp.com> Mail-Followup-To: yaffs@lists.aleph1.co.uk References: <200803311550.12028.matthias.fuchs@esd-electronics.com> <47F12B4D.50402@googlemail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <47F12B4D.50402@googlemail.com> User-Agent: Mutt/1.5.16 (2007-06-09) X-SA-Exim-Connect-IP: 81.187.30.51 X-SA-Exim-Mail-From: rdunlop@guralp.com X-Spam-Checker-Version: SpamAssassin 3.2.3 (2007-08-08) on stoneboat.aleph1.co.uk X-Spam-Level: X-Spam-Status: No, score=-3.6 required=5.0 tests=BAYES_00,RCVD_IN_DNSWL_LOW autolearn=ham version=3.2.3 X-SA-Exim-Version: 4.2.1 (built Tue, 09 Jan 2007 17:23:22 +0000) X-SA-Exim-Scanned: Yes (on stoneboat.aleph1.co.uk) Subject: Re: [Yaffs] yaffs with kernel 2.6.25: no more iget() X-BeenThere: yaffs@lists.aleph1.co.uk X-Mailman-Version: 2.1.9 Precedence: list List-Id: Discussion of YAFFS NAND flash filesystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 23 Apr 2008 09:22:32 -0000 Hi, On Mon, Mar 31 at 08:19, Dirk Behme wrote: > Matthias Fuchs wrote: >> Hi, >> did anybody spend some time in compiling yaffs for 2.6.25-rcX? >> In February the iget() call got lost. So current yaffs does not compile in >> recent kernels anymore. > > Yes, you are right. With 2.6.25-rc6 I get ... > Looking at the quite good patch comment, fixing this should be easy (?). > > Anybody with a patch? Well I had to put YAFFS2 into the 2.6.25 kernel on Friday and this is the naive patch I came up with based on the example changes. It compiles and works for me but I've only been using YAFFS for a total of 2 days. I've done little more than "it doesn't smoke" testing. --- linux-2.6.git/fs/yaffs2/yaffs_fs.c-orig 2008-04-17 08:49:19.000000000 +0100 +++ linux-2.6.git/fs/yaffs2/yaffs_fs.c 2008-04-17 10:09:40.000000000 +0100 @@ -181,7 +181,7 @@ #else static int yaffs_statfs(struct super_block *sb, struct statfs *buf); #endif -static void yaffs_read_inode(struct inode *inode); +static struct inode *yaffs_iget(struct super_block *sb, unsigned long ino); static void yaffs_put_inode(struct inode *inode); static void yaffs_delete_inode(struct inode *); @@ -284,7 +284,6 @@ static struct super_operations yaffs_super_ops = { .statfs = yaffs_statfs, - .read_inode = yaffs_read_inode, .put_inode = yaffs_put_inode, .put_super = yaffs_put_super, .delete_inode = yaffs_delete_inode, @@ -844,7 +843,9 @@ T(YAFFS_TRACE_OS, (KERN_DEBUG "yaffs_get_inode for object %d\n", obj->objectId)); - inode = iget(sb, obj->objectId); + inode = yaffs_iget(sb, obj->objectId); + if (IS_ERR(inode)) + return NULL; /* NB Side effect: iget calls back to yaffs_read_inode(). */ /* iget also increments the inode's i_count */ @@ -1427,19 +1428,25 @@ } -static void yaffs_read_inode(struct inode *inode) +static struct inode *yaffs_iget(struct super_block *sb, unsigned long ino) { - /* NB This is called as a side effect of other functions, but - * we had to release the lock to prevent deadlocks, so - * need to lock again. - */ - + struct inode *inode; yaffs_Object *obj; - yaffs_Device *dev = yaffs_SuperToDevice(inode->i_sb); + yaffs_Device *dev = yaffs_SuperToDevice(sb); T(YAFFS_TRACE_OS, - (KERN_DEBUG "yaffs_read_inode for %d\n", (int)inode->i_ino)); + (KERN_DEBUG "yaffs_iget for %lu\n", ino)); + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; + + /* NB This is called as a side effect of other functions, but + * we had to release the lock to prevent deadlocks, so + * need to lock again. + */ yaffs_GrossLock(dev); obj = yaffs_FindObjectByNumber(dev, inode->i_ino); @@ -1447,6 +1454,9 @@ yaffs_FillInodeFromObject(inode, obj); yaffs_GrossUnlock(dev); + + unlock_new_inode(inode); + return inode; } static LIST_HEAD(yaffs_dev_list); -- Bob Dunlop Guralp Systems Limited http://www.guralp.com