[U-Boot] [PATCH 3/3 v4] arm: A320: Add support for Faraday A320 evaluation board

[Jean-Christophe PLAGNIOL-VILLARD]

On 23:10 Tue 04 Aug     , Darius Augulis wrote:
> On 08/04/2009 10:48 PM, Jean-Christophe PLAGNIOL-VILLARD wrote:
> >On 09:42 Mon 03 Aug     , Darius Augulis wrote:
> >>On 07/08/2009 02:30 PM, Jean-Christophe PLAGNIOL-VILLARD wrote:
> >>>On 15:14 Fri 03 Jul     , Po-Yu Chuang wrote:
> >>>>This patch adds support for A320 development board from Faraday. This board
> >>>>uses FA526 processor by default and has 512kB and 32MB NOR flash, 64M RAM.
> >>>>FA526 is an ARMv4 processor and uses the ARM920T source in this patch.
> >>>>
> >>>as I understand correctly the faraday and the CS3518 share the same core and
> >>>IP so it will be better to have the same dir
> >>I don't think so. Both A320 and Gemini share the same FA526 core,
> >>but different peripheral. There is no reason to create generic dir
> >>for A320 and Gemini. Also, no reason to create cpu/fa526 because
> >>fa526 is almost arm920t.
> >duplicate code is worse
> >as example on at91 we do support multiple soc of the family in the same
> >generic dir which allow us to factorize common code and ofcourse split
> >soc specific code too
> 
> Gemini and A320 are not from the same family. There isn't any common
> code *yet*. They share only the same core.
> There are more dirs in cpu/arm920t: "imx", "s3c24x0", etc. Perhaps
> you won't suggest to move all these SoC's into single dir?
no as they do share nothing if the code is arm920t specific it will more to
arm/920t if soc arm920t/at91 etc....
> 
> A320 and Gemini *maybe* have the same timer IP, but there are few
> 'small' problems. First: these processors, made in China, have very
> poor documentation. I must do lot of reverse-engineering work to
> create something working. Second, I don't have any A320 hardware to
> test something. It's very difficult to make most optimal release at
> the very beginning. After we have Gemini in main line, somebody
> interested in A320 should feel free to submit patches to optimize
> support for SoC's, based on FA526 core. Now I can't do this job and
> I guess it shouldn't be reason to reject my patches?
you are two dev working to add both soc mainline so please sync together to
not add duplicated code that all I want

Best Regards,
J.