1 |
R Hill skrev: |
2 |
> Simon Strandman wrote: |
3 |
>> It seems like toolchain.eclass does something wrong when configuring |
4 |
>> gcc 4.1 snapshots. I decided to try gcc 4.1 on my server so I created |
5 |
>> a gcc-4.1.1.20060324 ebuild and defined the SNAPSHOT variable in it |
6 |
>> (current cvs has a lot of bugfixes since the release). This is the |
7 |
>> way I've done it with gcc 4.0 and I never had any problems then. The |
8 |
>> ebuild emerges without problems but it installs files in both |
9 |
>> /usr/lib/gcc/i686-pc-linux-gnu/4.1.1-20060324 and |
10 |
>> /usr/lib/gcc/i686-pc-linux-gnu/4.1.1. So when I try to emerge |
11 |
>> anything it always fails with errors like this: |
12 |
>> |
13 |
>> configure:2239: i686-pc-linux-gnu-gcc -O2 -march=pentium3 -pipe |
14 |
>> -fomit-frame-pointer -D_FORTIFY_SOURCE=1 conftest.c >&5 |
15 |
>> /usr/lib/gcc/i686-pc-linux-gnu/4.1.1/../../../../i686-pc-linux-gnu/bin/ld: |
16 |
>> cannot find -lgcc_s |
17 |
>> collect2: ld returned 1 exit status |
18 |
>> |
19 |
>> Just copying over the files from one dir to the other and then |
20 |
>> symlinking it works around the problems. Any ideas? |
21 |
> |
22 |
> This is caused by changes to the build system in 4.1 and GCC's |
23 |
> BASE-VER not matching portage's ${PV} in snapshot builds. Most of the |
24 |
> system directories are set up by portage during configure using ${PV} |
25 |
> as part of the dirname. (eg. |
26 |
> includedir=/usr/<lib>/gcc/<chost>/${PV}/include). However, libdir and |
27 |
> libexecdir aren't set by portage (because they generate really strange |
28 |
> paths w/ --enable-version-specific-runtime-libs in GCC 3.3/3.4) and |
29 |
> default to /usr/<lib>/gcc/<chost>/BASE-VER/blah. When ${PV} != |
30 |
> BASE-VER, wackiness ensues. |
31 |
> |
32 |
> Try this in your ebuild: |
33 |
> |
34 |
> src_unpack() { |
35 |
> toolchain_src_unpack |
36 |
> |
37 |
> echo ${PV/_/-} > "${S}"/gcc/BASE-VER |
38 |
> echo "" > "${S}"/gcc/DATESTAMP |
39 |
> } |
40 |
> |
41 |
> --de. |
42 |
> |
43 |
Thanks for the help! But now it fails with this error: |
44 |
|
45 |
<built-in>:1: internal compiler error: in define__GNUC__, at c-cppbuiltin.c:296 |
46 |
Please submit a full bug report, |
47 |
|
48 |
|
49 |
This has been reported in gcc bug #19372 and apperantly it has something |
50 |
to do with the version string. :( |
51 |
-- |
52 |
gentoo-dev@g.o mailing list |