On Monday 03 October 2011 21:40:36 Kukjin Kim wrote: > Yes, absolutely. As I said, the 'mach-exynos' can support upcoming > Samsung EXYNOS SoCs and the way is better than making mach-exynos5. > > My bigger plan is finally to make current directories to 3 mach- dir. > (mach-s3c, mach-s5p, and mach-exynos) and 1 plat-samsung directory. Of > course, the mach- directories can be reduced later. > > Firstly, as you can see, I'm merging plat-s3c24xx and plat-s5p into one > plat-samsung directory and actually the <plat/*.h> files and dev-*.c > files have been moved into plat-samsung. Maybe I can finish it during > 3.2-rc. Ok, sounds good. So for my understanding: is s5p a completely separate line from exynos, or are they more of an evolutionary succession where you draw the line between the last uniprocessor and the first SMP chip? > Secondary, upcoming Samsung EXYNOS SoCs will be supported in mach-exynos > directory and this patch is required to that. For example, the EXYNOS5 > SoCs (Cortex A15) will be supported in that with other EXYNOS4 SoCs soon > and of course it reduces code duplication My feeling is that it would be nicer to just add exynos5 to the mach-exynos4 directory for now, in the same way that omap3/4/5 are in the mach-omap2 directory. We can do a big rename later once the reorganization of code within those directories that you explained above is complete. If you still want to do the big rename right now, I would suggest doing it at the end of the 3.2 merge window, so that all other patches that might need to touch the mach-exynos4 directory can go in first. Arnd -- To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
 |