On Tue, Apr 17, 2018 at 09:17:12AM +0200, Danny Milosavljevic wrote: > Hi Mark, > > On Mon, 16 Apr 2018 16:05:04 -0400 > Mark H Weaver <m...@netris.org> wrote: > > > If it would be beneficial, you might consider using gcc-7 to compile > > 'rust', by adding 'gcc-7' to native-inputs. We're already using gcc-7 > > to compile a few other packages, including linux-libre on x86_64. > > Yeah, it would be possible. > <snip> > > Also, AFAIK "-march=native" autodetects the CPU in the build machine > and then optimizes for exactly that CPU. Arguably that's not what > we want in general.
This was my original intent with commenting out the '-march=native', it just seemed that it cropped up as a build error more often with arm* processors. As far as aarch64 goes, it doesn't build fully in either case, but the error I experienced with '-march=native' is gone when I add gcc-7 as a native-input. -- Efraim Flashner <efr...@flashner.co.il> אפרים פלשנר GPG key = A28B F40C 3E55 1372 662D 14F7 41AA E7DC CA3D 8351 Confidentiality cannot be guaranteed on emails sent or received unencrypted
signature.asc
Description: PGP signature
