From mboxrd@z Thu Jan 1 00:00:00 1970 From: Josh Wu Date: Wed, 15 Aug 2012 16:19:11 +0800 Subject: [U-Boot] [PATCH 0/3] at91: atmel_nand: add PMECC(Programmable Multibit ECC controller) support In-Reply-To: <502B54B3.4050803@gmail.com> References: <1344507709-8003-1-git-send-email-josh.wu@atmel.com> <502B46AD.5030109@atmel.com> <502B54B3.4050803@gmail.com> Message-ID: <502B5B7F.4000606@atmel.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de Hi, Andreas Thank you for the quick reply. On 8/15/2012 3:50 PM, Andreas Bie?mann wrote: > Dear Josh Wu, > > On 15.08.2012 08:50, Josh Wu wrote: >> Hi, Andreas >> >> Sorry, in these patches of current version, it still has a bug, which >> will fail to write with PMECC, I will send out the v2 patch soon which >> include this bug fix. > Oups, sorry I had this patch series not on my list. Will review this or > v2 these days. > On first sight the first two in series are ok, the third one has style > problems: > > ---8<--- > abiessmann at azuregos % ./tools/checkpatch.pl > U-Boot-3-3-at91-atmel_nand-Update-driver-to-support-Programmable-Multibit-ECC-controller.patch > WARNING: line over 80 characters > ... > total: 1 errors, 27 warnings, 787 lines checked > > NOTE: Ignored message types: COMPLEX_MACRO CONSIDER_KSTRTO MINMAX > MULTISTATEMENT_MACRO_USE_DO_WHILE > > U-Boot-3-3-at91-atmel_nand-Update-driver-to-support-Programmable-Multibit-ECC-controller.patch > has style problems, please review. > > If any of these errors are false positives, please report > them to the maintainer, see CHECKPATCH in MAINTAINERS. > --->8--- > > I personally do not really care about the 80 char warning but please > remove these for the sake of peace ;) and please have a look for the > reported error. sure, I will fix that. :) > And the last, you should add at least one user of this feature. Isn't > the lately applied at91sam9x5ek (is in u-boot-atmel/master and > u-boot-arm/master) a perfect user for that? you are right, The PMECC usage is for 9x5ek board. I planed to push the PMECC support patches in 9x5 board in two part: one is this series, which only include nand flash driver changes. I thought those patches should be merged by u-boot mtd maintainer. Another part is some patches for 9x5 board file and config file which enable PMECC in 9x5ek board. Those patches are not sent out yet. Now I think if you can merge all above patches to u-boot mainline, then maybe I can sent those patches together. What do you think about this? > >> On 8/9/2012 6:21 PM, Josh Wu wrote: >>> This patch series will add PMECC support for atmel nand driver in u-boot. >>> >>> The driver patch is ported from kernel patch. The main different from >>> kernel patch is >>> the modification about registers access part. in this version it via a >>> register structure to access. > A question regarding linux driver. You say it is derived from the kernel > driver, but I can not see PMECC stuff in kernel: Sorry, I am not make it clear about this. Actually, The PMECC stuff is not in v3.6-rc1 now, it is queued in kernel MTD maintainer's git tree: http://git.infradead.org/linux-mtd.git/commit/1c7b874d33b463f7150b1ab4617f000af9b327fd Seems the maintainer missed the merge window of v3.6-rc1. I am not sure that can go into v3.6 or not. But I think it is not a bad idea to push those code into u-boot. :) > ---8<--- > abiessmann at azuregos % git describe > v3.6-rc1-355-gddf343f > abiessmann at azuregos % diff -Nrupa drivers/mtd/nand/atmel_nand.c > ../u-boot/drivers/mtd/nand/atmel_nand.c | diffstat > atmel_nand.c | 1231 > ++++++++++++++++++++++++++++++++++------------------------- > 1 file changed, 720 insertions(+), 511 deletions(-) > abiessmann at azuregos % grep -in pmecc drivers/mtd/nand/atmel_nand* > abiessmann at azuregos % > --->8--- > > Which driver/patch do you mean? > >>> It is tested in atmel 9x5 series board. > Unfortunately I can not test this driver cause I do only have older > sam926x devices around. If I send 9x5 board enable patches with nand PMECC patches together, I think maybe other people can test it in 9x5ed board. > > Best regards > > Andreas Bie?mann > Best Regards, Josh Wu