From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from lithops.sigma-star.at ([195.201.40.130]) by bombadil.infradead.org with esmtps (Exim 4.90_1 #2 (Red Hat Linux)) id 1faNBg-0000NY-8n for linux-mtd@lists.infradead.org; Tue, 03 Jul 2018 15:25:54 +0000 From: Richard Weinberger To: Casey Schaufler Cc: xiaolei li , leqiao.peng@valeo.com, linux-mtd@lists.infradead.org Subject: Re: UBIFS assert when create new sub-dir under a transmute enabled dir with Smack enabled Date: Tue, 03 Jul 2018 17:25:29 +0200 Message-ID: <1779495.8Qfuh5Rb3p@blindfold> In-Reply-To: References: <1529571876.14673.77.camel@mhfsdcap03> <3818822.lxJep7IjOv@blindfold> MIME-Version: 1.0 Content-Transfer-Encoding: 7Bit Content-Type: text/plain; charset="us-ascii" List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Am Dienstag, 3. Juli 2018, 17:22:30 CEST schrieb Casey Schaufler: > >> The root cause is that the inode isn't locked during inode instantiation > >> (from Mr.Schaufler), but we check whether inode is locked in > >> xattr_set. > > There is no need to do this check as all that > gets done with the decision is to print the warning. > The problem is the assumption that the inode is locked, > which is doesn't have to be during instantiation. I agree. Historically ubifs has a lot of (very good) asserts. Some of them turn out to be too paranoid, like this one. Thanks, //richard