From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from 213-239-205-147.clients.your-server.de ([213.239.205.147] helo=mail.tglx.de) by canuck.infradead.org with esmtp (Exim 4.54 #1 (Red Hat Linux)) id 1FBTkj-00005N-6v for linux-mtd@lists.infradead.org; Tue, 21 Feb 2006 04:24:18 -0500 From: Thomas Gleixner To: Robert Liu In-Reply-To: <43FA6A26.8000005@wiscore.com> References: <43FA6A26.8000005@wiscore.com> Content-Type: text/plain Date: Tue, 21 Feb 2006 10:25:14 +0100 Message-Id: <1140513914.2480.861.camel@localhost.localdomain> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Cc: linux-mtd@lists.infradead.org Subject: Re: do R/W tests on JFFS2 and suffer segmentation fault Reply-To: tglx@linutronix.de List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Tue, 2006-02-21 at 09:17 +0800, Robert Liu wrote: > I am doing some read/write tests on JFFS2 partition, and find there may be > a bug in jffs2_remove_from_hash_table(). Which kernel version ? Any patches related to MTD/JFFS2 applied ? > Tests sometimes failed with segmentation fault. > I set the JFFS2 debug level to 1, and add several debug infomation in > jffs2_remove_from_hash_table(). > It seems jffs2 try to remove an entry with prev is 0x200200 and next is > 0x100100. Thats the list_head poison code (see include/linux/list.h LIST_POISON1/2). The entry was removed before. tglx