From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from arroyo.ext.ti.com ([192.94.94.40]) by bombadil.infradead.org with esmtps (Exim 4.80.1 #2 (Red Hat Linux)) id 1XQuLV-0004Gb-Gw for linux-mtd@lists.infradead.org; Mon, 08 Sep 2014 08:30:46 +0000 Message-ID: <540D691B.2060202@ti.com> Date: Mon, 8 Sep 2014 11:30:19 +0300 From: Roger Quadros MIME-Version: 1.0 To: Ezequiel Garcia , Brian Norris Subject: Re: [PATCH 0/3] nand: omap2: Two and a half improvements References: <1410033389-32357-1-git-send-email-ezequiel@vanguardiasur.com.ar> In-Reply-To: <1410033389-32357-1-git-send-email-ezequiel@vanguardiasur.com.ar> Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit Cc: Tony Lindgren , linux-mtd@lists.infradead.org, =?ISO-8859-1?Q?Guido_Mart=EDnez?= List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Hi Ezequiel, On 09/06/2014 10:56 PM, Ezequiel Garcia wrote: > Hi Brian, Roger: > > Pekon's attempt to add flash BBT support for this driver made me realise > the addition made sense and there were good reasons for it. The first patch > adds support for enabling a flash BBT either from legacy board files or > from devicetree. > > While testing this, I noticed how the driver relied on a whole bunch of > horrible #ifdefs, which prevented me from loading the driver as a module. > The second patch attempts to fix that. > > The third patch is just a dummy cleanup replacing pr_errs with dev_errs. > This driver is abusing from user messages, but I'm not sure fixing them > worths the trouble. > > Roger, do you think you can test patches 1 and 2 with different ECCs > and configurations? It's an invasive patch and I don't want to see more > regressions with this driver. Yes. I will run them through all the boards that I have and let you know in a day or two. > > And speaking of modules, the driver loads as "modprobe omap2". And it's not > the only one with a clumsy name: "modprobe elm". I guess we cannot fix it now, > but it would be great to be more careful with driver naming in the future. Why can't we fix it now? It seems nobody ever used it as a module before so now is the right time to fix. Unless Tony has any objections. cheers, -roger > > Ezequiel Garcia (3): > nand: omap2: Add support for a flash-based bad block table > nand: omap2: Refactor the code to remove horrible ifdefs > nand: omap2: Replace pr_err with dev_err > > arch/arm/mach-omap2/gpmc.c | 2 + > drivers/mtd/nand/omap2.c | 166 +++++++++++++++------------ > include/linux/platform_data/elm.h | 14 +++ > include/linux/platform_data/mtd-nand-omap2.h | 1 + > 4 files changed, 108 insertions(+), 75 deletions(-) >