From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from pentafluge.infradead.org (pentafluge.infradead.org [213.146.154.40]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by ozlabs.org (Postfix) with ESMTP id 85245DDF20 for ; Tue, 8 Jan 2008 18:59:02 +1100 (EST) Subject: Re: [PATCH 2/4] nand base: Give drivers a chance to do late initialization. From: David Woodhouse To: Scott Wood In-Reply-To: <20071213171551.GA4459@loki.buserror.net> References: <20071213171551.GA4459@loki.buserror.net> Content-Type: text/plain Date: Tue, 08 Jan 2008 07:57:49 +0000 Message-Id: <1199779069.4111.379.camel@shinybook.infradead.org> Mime-Version: 1.0 Cc: linuxppc-dev@ozlabs.org, linux-mtd@lists.infradead.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Thu, 2007-12-13 at 11:15 -0600, Scott Wood wrote: > Some nand controllers, such as the Freescale enhanced local bus controller, > need to do late initialization based on details of the chip that has been > probed, such as chip size, large/small pages, etc. A late_init() callback > method is added to allow this. Do you really need this so late? Can't it be done between nand_scan_ident() and nand_scan_tail()? That's why we split nand_scan() into two in the first place... -- dwmw2