1 |
Simon Strandman skrev: |
2 |
> R Hill skrev: |
3 |
>> Simon Strandman wrote: |
4 |
>>> It seems like toolchain.eclass does something wrong when configuring |
5 |
>>> gcc 4.1 snapshots. I decided to try gcc 4.1 on my server so I |
6 |
>>> created a gcc-4.1.1.20060324 ebuild and defined the SNAPSHOT |
7 |
>>> variable in it (current cvs has a lot of bugfixes since the |
8 |
>>> release). This is the way I've done it with gcc 4.0 and I never had |
9 |
>>> any problems then. The ebuild emerges without problems but it |
10 |
>>> installs files in both /usr/lib/gcc/i686-pc-linux-gnu/4.1.1-20060324 |
11 |
>>> and /usr/lib/gcc/i686-pc-linux-gnu/4.1.1. So when I try to emerge |
12 |
>>> anything it always fails with errors like this: |
13 |
>>> |
14 |
>>> configure:2239: i686-pc-linux-gnu-gcc -O2 -march=pentium3 -pipe |
15 |
>>> -fomit-frame-pointer -D_FORTIFY_SOURCE=1 conftest.c >&5 |
16 |
>>> /usr/lib/gcc/i686-pc-linux-gnu/4.1.1/../../../../i686-pc-linux-gnu/bin/ld: |
17 |
>>> cannot find -lgcc_s |
18 |
>>> collect2: ld returned 1 exit status |
19 |
>>> |
20 |
>>> Just copying over the files from one dir to the other and then |
21 |
>>> symlinking it works around the problems. Any ideas? |
22 |
>> |
23 |
>> This is caused by changes to the build system in 4.1 and GCC's |
24 |
>> BASE-VER not matching portage's ${PV} in snapshot builds. Most of |
25 |
>> the system directories are set up by portage during configure using |
26 |
>> ${PV} as part of the dirname. (eg. |
27 |
>> includedir=/usr/<lib>/gcc/<chost>/${PV}/include). However, libdir |
28 |
>> and libexecdir aren't set by portage (because they generate really |
29 |
>> strange paths w/ --enable-version-specific-runtime-libs in GCC |
30 |
>> 3.3/3.4) and default to /usr/<lib>/gcc/<chost>/BASE-VER/blah. When |
31 |
>> ${PV} != BASE-VER, wackiness ensues. |
32 |
>> |
33 |
>> Try this in your ebuild: |
34 |
>> |
35 |
>> src_unpack() { |
36 |
>> toolchain_src_unpack |
37 |
>> |
38 |
>> echo ${PV/_/-} > "${S}"/gcc/BASE-VER |
39 |
>> echo "" > "${S}"/gcc/DATESTAMP |
40 |
>> } |
41 |
>> |
42 |
>> --de. |
43 |
>> |
44 |
> Thanks for the help! But now it fails with this error: |
45 |
> |
46 |
> <built-in>:1: internal compiler error: in define__GNUC__, at |
47 |
> c-cppbuiltin.c:296 |
48 |
> Please submit a full bug report, |
49 |
> |
50 |
> |
51 |
> This has been reported in gcc bug #19372 and apperantly it has |
52 |
> something to do with the version string. :( |
53 |
I opened a bug about it: |
54 |
http://bugs.gentoo.org/show_bug.cgi?id=127724 |
55 |
-- |
56 |
gentoo-dev@g.o mailing list |