| Summary: | sys-devel/gcc-4.7.3-r1 does not compile with gcc 5.4.0: ‘gnu_inline’ attribute present on ‘libc_name_p’ | ||
|---|---|---|---|
| Product: | Gentoo/Alt | Reporter: | Constantin Runge <c.runge> |
| Component: | Prefix Support | Assignee: | Gentoo Prefix <prefix> |
| Status: | RESOLVED FIXED | ||
| Severity: | normal | CC: | samtygier |
| Priority: | Normal | ||
| Version: | unspecified | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
|
Description
Constantin Runge
2017-05-24 12:09:36 UTC
There seems to be a patch for this [1], which apparently is not merged yet [2] if I interpret the State field set to new correctly. [1] https://gcc.gnu.org/ml/gcc-patches/2015-08/msg00375.html [2] https://patchwork.ozlabs.org/patch/504982/ I just ran into this issue as well. I also noticed <sys-devel/gcc-4.9 has been hardmasked since 21 May... It seems the bootstrapper is set to install gcc <4.8, because gcc 4.8+ requires a C++ compiler. Perhaps it's better to check for an installed C++ compiler. bootstrap-prefix.sh now checks for host gcc greater than 4.7 to avoid bootstrapping <gcc-4.8: https://gitweb.gentoo.org/repo/proj/prefix.git/commit/?id=0400d5c9f6345a3e92b54f9591ca5210bcfe2646 Thanks! *** Bug 588092 has been marked as a duplicate of this bug. *** >gcc-4.7, as anything newer provides c++ anyway.
Is that really so? In my distributions, it is possible to uninstall c++.
(In reply to Benda Xu from comment #6) > >gcc-4.7, as anything newer provides c++ anyway. > > Is that really so? In my distributions, it is possible to uninstall c++. Agreed, but how could we work around this situation? I can think of detecting and yelling here, but probably >=gcc-4.8 does this itself already, no? (In reply to Michael Haubenwallner from comment #7) > (In reply to Benda Xu from comment #6) > > >gcc-4.7, as anything newer provides c++ anyway. > > > > Is that really so? In my distributions, it is possible to uninstall c++. > > Agreed, but how could we work around this situation? > > I can think of detecting and yelling here, but probably >=gcc-4.8 does this > itself already, no? Not sure. Does the ebuild yell or the gcc build system yell? I do think it is time to move forward and require c++ to bootstrap Prefix, and probably providing an independent way to bootstrap c++ from c, like what we did for bash bootstrap. |