From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from blu0-omc1-s1.blu0.hotmail.com ([65.55.116.12]) by merlin.infradead.org with esmtp (Exim 4.76 #1 (Red Hat Linux)) id 1ScgpQ-0007L2-JR for linux-mtd@lists.infradead.org; Thu, 07 Jun 2012 17:49:01 +0000 Message-ID: From: Bill Pringlemeir To: linux-mtd@lists.infradead.org Subject: Latest UBI/UBIFS requires DebugFs? Date: Thu, 7 Jun 2012 13:49:02 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , I have pulled the latest UBI/UBIFS from git://git.infradead.org/users/dedekind/ubifs-v2.6.36.git. I think that the same issue may exist on the head. If your root device is ubifs and you do not configure with CONFIG_DEBUG_FS, then several functions will fail and the root will not be mounted. Depending on other configurations, you might not get any error messages. It appears that ubi_debugfs_init_dev, ubi_debugfs_init, dbg_debugfs_init_fs, and dbg_debugfs_init rely on DEBUGFS and return errors when creating DEBUGFS files. I think either we could conditionalize these functions to return 0 when DEBUGFS is not present or use a '#error UBI/UBIFS not supported without DebugFs'. I think maybe this was an over zealous commit to get good bug reports and the DEBUGFS itself is not needed, but just the other printk stuff in the debug.c files? I can submit a patch, but I think the explanation above is probably enough for someone else with a better understanding of the structure. Regards, Bill Pringlemeir.