From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail.timesys.com ([65.117.135.102] helo=exchange.timesys.com) by pentafluge.infradead.org with esmtp (Exim 4.22 #5 (Red Hat Linux)) id 1AMu2S-0008M0-L8 for ; Thu, 20 Nov 2003 19:00:24 +0000 Date: Thu, 20 Nov 2003 13:58:46 -0500 From: John Whitney To: Message-ID: Mime-version: 1.0 Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit Subject: JFFS2 and 64-bit flash List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Is there anything in latest JFFS2 code that would have problems with 64-bit buswidth flash access? I have an four interleaved 16-bit AMD flash chips on a 64-bit bus. Using the MTD devices, I can write to, erase, and read the flash with no problems. (I had to modify my PowerPC kernel to use a 64-bit floating point register to ensure a 64-bit write, instead of two 32-bit writes. Prior to this, I had some failures). If I copy in a JFFS2 filesystem, I can mount it and use "ls" to get a top level directory listing. However, if I use "ls -l" to get the same listing, I get: Chip not ready after erase suspended: status = 0x2a error -5 reading node at 0x000bc110 in get_inode_nodes() jffs2_get_inode_nodes() for ino 2 returned -5 ls: /mnt/bin: Input/output error Is this most likely a problem with the cfi_cmdset_0002, or something in JFFS2? John Whitney