Please stabilize
x86 done
amd64 done all arches done
I am trying to build a binary package server for Raspberry Pi systems, and as a preliminary step I am trying to determine what flags I should be concerned with per https://wiki.gentoo.org/wiki/Binary_package_guide#Handling_*FLAGS_in_detail. On a Raspberry Pi, I'm getting: pizero2 /home/jlpoole # resolve-march-native gcc: error: missing argument to '-march=' ERROR: Command '['gcc', '-S', '-fverbose-asm', '-o', '/tmp/tmp1xqf3pli/march_native.s', '/tmp/tmp1xqf3pli/empty.c', '-march=']' returned non-zero exit status 1. pizero2 /home/jlpoole #man resolve-march-native No manual entry for resolve-march-native pizero2 /home/jlpoole # Should I file a separate bug, or provide here my build log and settings as I might for a bug in a failed build?
John, if you are willing to help with debugging and sharing the output from "resolve-march-native --debug" then a new issue at https://github.com/hartwork/resolve-march-native/issues would be best. '-march=' in your current output already looks off but let's take things to https://github.com/hartwork/resolve-march-native/issues please.
(In reply to Sebastian Pipping from comment #4) > John, if you are willing to help with debugging and sharing the output from > "resolve-march-native --debug" then a new issue at > https://github.com/hartwork/resolve-march-native/issues would be best. > '-march=' in your current output already looks off but let's take things to > https://github.com/hartwork/resolve-march-native/issues please. Thank you. It looks like this tool may be Intel specific. Before seeing your reply, I visited https://github.com/hartwork/resolve-march-native/tree/master and saw in the example using "-march=bonnell". Wikipedia tells me that bonnell is a specific architecture for the Intel Atom series: https://en.wikipedia.org/wiki/List_of_Intel_CPU_microarchitectures. I'll follow up with an issue on the GitHub as requested, if only to clarify that this may be an Intel specific tool and the arm64 might be flagged as incompatible?
(In reply to Sebastian Pipping from comment #4) > John, if you are willing to help with debugging and sharing the output from > "resolve-march-native --debug" then a new issue at > https://github.com/hartwork/resolve-march-native/issues would be best. > '-march=' in your current output already looks off but let's take things to > https://github.com/hartwork/resolve-march-native/issues please. Issue logged: https://github.com/hartwork/resolve-march-native/issues/105