From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Petazzoni Date: Sat, 29 Aug 2015 10:11:36 +0200 Subject: [Buildroot] [PATCH 1/2] toolchain: add sparc64 architecture support In-Reply-To: <20150829074104.GB8475@waldemar-brodkorb.de> References: <20150811203709.GA9885@waldemar-brodkorb.de> <20150828202224.2f12a93c@free-electrons.com> <20150829074104.GB8475@waldemar-brodkorb.de> Message-ID: <20150829101136.02a50aec@free-electrons.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net Waldemar, On Sat, 29 Aug 2015 09:41:04 +0200, Waldemar Brodkorb wrote: > Yeah, the main reason providing the patch is to discover portability > issues in software packages for this platform. I like the > autobuilder stuff and a greater test coverage is useful in my > opinion. Ok, no problem. > There is already ppc64-pseries, which isn't an embedded system or do > I missed something? :=) Yes, it was contributed by IBM IIRC, so we kind of assumed they would be around to fix issues if any (though in practice we have not yet added a PowerPC 64 toolchain configuration in the autobuilders, so my comment is a bit moot as we have never tested the reactivity of the IBM guys). > > > config BR2_ENDIAN > > > default "BIG" > > > @@ -20,3 +24,4 @@ config BR2_ENDIAN > > > config BR2_GCC_TARGET_CPU > > > default "leon3" if BR2_sparc_leon3 > > > default "v8" if BR2_sparc_v8 > > > + default "ultrasparc" if BR2_sparc_v9 > > > > Why not using v9 here? It's a valid -mcpu value: > > Valid, but not useful for Linux: > http://permalink.gmane.org/gmane.comp.lib.glibc.alpha/10649 So maybe the option should be called v9a, rather than just v9 ? Or just Ultrasparc so it matches the name of the gcc -mcpu option ? Thanks! Thomas -- Thomas Petazzoni, CTO, Free Electrons Embedded Linux, Kernel and Android engineering http://free-electrons.com