From mboxrd@z Thu Jan 1 00:00:00 1970 From: Txema Lopez Date: Thu, 01 Mar 2007 09:08:33 +0100 Subject: [U-Boot-Users] [PATCH] Support for csb535fs / i.MX21 LiteKit. In-Reply-To: <528646bc0702280855r83d304eo510396e5d87d9c4@mail.gmail.com> References: <45E42BE3.9060305@aotek.es> <528646bc0702270920r2b120aa9s2bf234187153d544@mail.gmail.com> <45E532AA.9030008@aotek.es> <528646bc0702280855r83d304eo510396e5d87d9c4@mail.gmail.com> Message-ID: <45E68A01.3060607@aotek.es> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de Grant Likely wrote: > > > I know that slightly modified duplicate code is common in u-boot, so > this is not an critique on your work, but I'd really like to move away > from this mode of operation. Duplicating the original file and > modifying it is certainly the easiest way to add support for a new And It makes the code more readable. It's the pro. > 10 times and a bug is found in it at a later date; the bug fix needs > to be applied to 10 files, not one. Yes, It's the con. But, how many times does it happen?. > > Unfortunately, this situation is messy because the imx is in > cpu/arm920t and imx32 is in cpu/arm926ejs. It probably requires the > creation of a new directory for the common imx soc bits, but where? > Perhaps under lib_arm/imx? > Ummm, lib_arm/imx?. I think this mixes up concepts. Why not to decouple the arm cores and the SoC code?. For example: cpu/arm/cores/arm926ejs cpu/arm/soc/imx Perhaps, there was a discussion about this in U-Boot and I'm talking nonsense. Anyway, you are the mainteiner so you have the last say. ;-) Best regards, Txema. -------------- next part -------------- A non-text attachment was scrubbed... Name: tlopez.vcf Type: text/x-vcard Size: 324 bytes Desc: not available Url : http://lists.denx.de/pipermail/u-boot/attachments/20070301/4c8fb6c2/attachment.vcf