From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from dell-paw-3.cambridge.redhat.com ([195.224.55.237] helo=passion.cambridge.redhat.com) by pentafluge.infradead.org with esmtp (Exim 3.22 #1 (Red Hat Linux)) id 178ys8-0007fM-00 for ; Sat, 18 May 2002 08:43:24 +0100 From: David Woodhouse In-Reply-To: References: To: Frederic Giasson Cc: linux-mtd@lists.infradead.org Subject: Re: Kernel oops after sync command in jffs2 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 18 May 2002 08:43:23 +0100 Message-ID: <11464.1021707803@redhat.com> Sender: linux-mtd-admin@lists.infradead.org Errors-To: linux-mtd-admin@lists.infradead.org List-Help: List-Post: List-Subscribe: , List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: fgiasson@mediatrix.com said: > When the oops happen, the do_erase_oneblock() function if > cfi_cmdset_0002.c complains that DQ5 has been asserted to "1" while > the flash was still erasing, meaning that a program command has been > attempted. > So I examined the chip driver to see if there were a gap in the mutual > exclusion of the chip access. I found that function > cfi_amdstd_write() send unlock bypass command to the flash without > verifying if the flash is ready first, if cfi->fast_prog is true. I > tryied to disable fast_prog in cfi_probe.c, and the oops did not > happen again. It sound like it were the problem, although I am not > 100% sure yet. Thanks. I've disabled fast programming mode completely, until the code gets fixed. That should have just caused an I/O error though; not an oops. -- dwmw2