1 |
Hi, |
2 |
|
3 |
On Sat, 10 Sep 2005 20:09:29 +0200 |
4 |
Frank Schafer <frank.schafer@×××××××××.cz> wrote: |
5 |
|
6 |
> When I returned home from work I found in the logs, that ``emerge |
7 |
> --emptytree system'' failed at package 28 of 186 |
8 |
> |
9 |
> python-fcksum-1.7.1 |
10 |
> i386-pc-linux-gnu-gcc ....bla...bla |
11 |
> ^ |
12 |
> | |
13 |
> +- !!!!!!!!!!!!!!!!!!!!! |
14 |
> |
15 |
> gcc-config error: |
16 |
> could not run/locate "i386-pc-linux-gnu-gcc" |
17 |
> |
18 |
> My architecture is i686 and it seems that 27 packages before |
19 |
> python-fchksum found the i686(that's SIX-eight-six)-pc-linux-gnu-gcc. |
20 |
|
21 |
Hm, when editing /etc/make.conf, did you change the CHOST setting? That |
22 |
could cause such behaviour... I think that may be amongst the reasons |
23 |
why /etc/make.conf.example reads like: |
24 |
|
25 |
---snip--- |
26 |
# Host Setting |
27 |
# ============ |
28 |
# |
29 |
# DO NOT CHANGE THIS SETTING UNLESS YOU ARE USING STAGE1! |
30 |
---snip--- |
31 |
|
32 |
...I guess you aren't using a stage1? |
33 |
|
34 |
|
35 |
As a side note: When building a minimal system, using |
36 |
USE=-* |
37 |
in /etc/make.conf should be considered. You can then fine tune each and |
38 |
every package in /etc/portage/package.use. That's quite like my setup, |
39 |
my global USE is set to "-* nptl ssl nls pam". All the other stuff is |
40 |
set per package. I have one build host machine that distributes the |
41 |
binary packages and portage tree snapshots to my other computers. |
42 |
|
43 |
-hwh |
44 |
-- |
45 |
gentoo-user@g.o mailing list |