From mboxrd@z Thu Jan  1 00:00:00 1970
From: =?ISO-8859-1?Q?Eric_B=E9nard?= <eric@eukrea.com>
Date: Fri, 17 Jul 2009 08:06:18 +0200
Subject: [U-Boot] [PATCH v2 2/3] Fix 2k page size NAND for iMX27
In-Reply-To: <20090716195440.GH32758@b07421-ec1.am.freescale.net>
References: <1247671121-5160-1-git-send-email-eric@eukrea.com>	<1247671121-5160-2-git-send-email-eric@eukrea.com>
	<20090716195440.GH32758@b07421-ec1.am.freescale.net>
Message-ID: <4A6014DA.2040400@eukrea.com>
List-Id: <u-boot.lists.denx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: u-boot@lists.denx.de

Scott Wood a ?crit :
> On Wed, Jul 15, 2009 at 05:18:40PM +0200, Eric Benard wrote:
>> +	if (pdata->is2k) {
>> +		host->pagesize_2k = 1;
>> +		NFMS |= (1 << NFMS_BIT);
>> +		this->badblock_pattern = &smallpage_memorybased;
> 
> Why are you using the small-page badblock pattern with large pages?
> 
that's what Freescale is doing in its Linux BSP and it doesn't work 
without, at least on my board (8 bits width NAND):

	if (!this->badblock_pattern) {
		if (mtd->writesize == NAND_PAGESIZE_2KB)
			this->badblock_pattern = &smallpage_memorybased;
		else
			this->badblock_pattern = (mtd->writesize > 512) ?
			    &largepage_memorybased : &smallpage_memorybased;
	}

Eric