From: Ian Campbell Date: Fri, 17 Apr 2015 12:06:55 +0100 Subject: [U-Boot] [PATCH 03/10] sunxi: Introduce a hidden ARCH_SUN6I Kconfig bool In-Reply-To: <552F636F.9030904@redhat.com> References: <1429027621-19252-1-git-send-email-hdegoede@redhat.com> <1429027621-19252-3-git-send-email-hdegoede@redhat.com> <552E14DC.60101@redhat.com> <552E1C34.1090401@redhat.com> <1429127245.5660.21.camel@hellion.org.uk> <552F636F.9030904@redhat.com> Message-ID: <1429268815.25195.268.camel@hellion.org.uk> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de On Thu, 2015-04-16 at 09:23 +0200, Hans de Goede wrote: > Hi, > > On 15-04-15 21:47, Ian Campbell wrote: > > On Wed, 2015-04-15 at 10:45 +0200, Michal Suchanek wrote: > >> It is not obvious which MACH_SUN?I are ARCH_SUN6I derived. So if you > >> can come up with a descriptive name for 'a number of things in common, > >> such as having separate ahb reset registers in the ccm' that's fine > >> otherwise this obfuscates the code, not clarifies. > > > > I don't particularly object to the patch but this occurred to me too. I > > suppose the rule is "first sunxi to look this way". > > > > How about we call groups of similar SoCs a "generation", i.e. > > ARCH_SUNXI_GEN2 is what is called ARCH_SUN6I here, meaning GEN1 would be > > SUN4/5/7I. > > I like the GEN idea, but not the numbering as it is a bit too arbitrary > how about: ARCH_SUNXI_GEN_SUN6I or (better I think) just SUNXI_GEN_SUN6I ? Works for me (I agree the second is better too). > > I know that does not solve the fact that MACH_SUN7I is SUNXI_GEN_SUN4I > but we cannot fix that, at least this way it will be explicit. > > Regards, > > Hans >