Summary: | dev-lang/R-2.3.0 fails to emerge | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Thomas Veith <th_veith> |
Component: | New packages | Assignee: | Gentoo Science Related Packages <sci> |
Status: | VERIFIED WORKSFORME | ||
Severity: | normal | ||
Priority: | High | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Other | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- |
Description
Thomas Veith
2006-05-15 08:42:09 UTC
It compiles here on amd64 without issue. You are using a masked version of GCC along with -ffast-math -ftracer, and I know -ffast-math can cause issues with GCC 3.4. With 4.1 I am not certain. You could try toning down your CFLAGS. I am currently recompiling my main system with GCC 4.1, and so when it gets there I will see if I can reproduce your error. I have a feeling it has already done R successfully though. Can anyone else reproduce? It looks to me like the -ffast-math has made it fail one of its own tests. I only apply flags like that to specific packages. (In reply to comment #1) > It compiles here on amd64 without issue. You are using a masked version of GCC > along with -ffast-math -ftracer, and I know -ffast-math can cause issues with > GCC 3.4. With 4.1 I am not certain. You could try toning down your CFLAGS. emerging without -ffast-math worked, thanks. nevertheless the bug seems strange... Well GCC 4.1 is very new and so does require more testing, but -ffast-math is known to have issues such as this which is why it is not on by default and why I would not apply it globally. I don't think it failed here on GCC 3.4 - I may check if I get chance but I believe it is arch dependent too. |