From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from [202.37.96.11] (helo=gatekeeper.tait.co.nz) by pentafluge.infradead.org with esmtp (Exim 4.22 #5 (Red Hat Linux)) id 1A9tIc-0003et-Jr for ; Wed, 15 Oct 2003 22:35:19 +0100 Received: from gatekeeper.tait.co.nz (localhost.localdomain [127.0.0.1]) by gatekeeper.tait.co.nz (8.11.2/8.9.3) with ESMTP id h9FLYGS01849 for ; Thu, 16 Oct 2003 10:34:16 +1300 Received: from sunstorm.tait.co.nz (sunstorm.tait.co.nz [172.25.40.9]) by gatekeeper.tait.co.nz (8.11.2/8.9.3) with ESMTP id h9FLYGS01843 for ; Thu, 16 Oct 2003 10:34:16 +1300 Received: from tait.co.nz ([172.25.110.7]) by sunstorm.tait.co.nz (iPlanet Messaging Server 5.1 (built May 7 2001)) with ESMTP id <0HMT00IEHIL26G@sunstorm.tait.co.nz> for linux-mtd@lists.infradead.org; Thu, 16 Oct 2003 10:34:14 +1300 (NZDT) Date: Thu, 16 Oct 2003 10:34:10 +1300 From: Robin Gilks In-reply-to: <1066111497.3172.37.camel@imladris.demon.co.uk> To: mtd list Message-id: <3F8DBD52.1020100@tait.co.nz> MIME-version: 1.0 References: <3F8B814D.3050408@tait.co.nz> <1066111497.3172.37.camel@imladris.demon.co.uk> Content-type: text/plain; charset=us-ascii; format=flowed Content-transfer-encoding: 7BIT Subject: Re: Mount of jffs2 partition fails List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , David Woodhouse wrote: > On Tue, 2003-10-14 at 17:53 +0000, Robin Gilks wrote: > > >>/ # mount -t jffs2 /dev/mtdblock1 /mnt >>Write clean marker to block at 0x000d0000 failed: -30 > > > -30 is also -EROFS but we don't normally see it there so we don't have a > nice helpful message telling you the reason :) > > I have no clue why your flash allows erases but not writes. This is a > flash hardware/driver problem, not JFFS2. Can you do more testing on the > raw /dev/mtd0 device itself using > read/write/ioctl(MEMERASE)/ioctl(MEMUNLOCK) to see what you need to do > to make it work? I've concluded that the ST chip I'm using, although being Intel command set compatible, requires a different approach. That being the case I enabled the ST chip option and on booting I now get the following. Physically mapped flash: Found 1 x16 devices at 0x0 in 16-bit mode kmod: failed to exec /sbin/modprobe -s -k cfi_cmdset_0003, errno = 2 Support for command set 0003 not present gen_probe: No supported Vendor Command Set found kmod: failed to exec /sbin/modprobe -s -k jedec_probe, errno = 2 kmod: failed to exec /sbin/modprobe -s -k map_rom, errno = 2 The items to note here are 1. cfi_cmdset_0003 does not exist, its replaced (as far as I can see) by cfi_cmdset_0020 2. I'm not using any modules so why is the kernel trying to load some!! 3. when I try to use the alternate Jedec probe method, I get a compile error in jedec_probe.c of: jedec_probe.c:1756: error: parse error before "jedec_probe_init" jedec_probe.c:1757: warning: return type defaults to `int' jedec_probe.c:1762: error: parse error before "jedec_probe_exit" jedec_probe.c:1763: warning: return type defaults to `int' jedec_probe.c:1767: warning: type defaults to `int' in declaration of `module_init' jedec_probe.c:1767: warning: parameter names (without types) in function declaration jedec_probe.c:1767: warning: data definition has no type or storage class jedec_probe.c:1768: warning: type defaults to `int' in declaration of `module_exit' jedec_probe.c:1768: warning: parameter names (without types) in function declaration jedec_probe.c:1768: warning: data definition has no type or storage class I take it that the code concerned has only ever been tested as modules? I'll try that in the meantime but that will require major work on my root filesystem :-(( -- Robin Gilks Senior Design Engineer Phone: (+64)(3) 357 1569 Tait Electronics Fax : (+64)(3) 359 4632 PO Box 1645 Christchurch Email : robin.gilks@tait.co.nz New Zealand