1 |
Paul Stear posted on Mon, 27 Sep 2010 13:33:22 +0100 as excerpted: |
2 |
|
3 |
> Makefile:72: warning: overriding commands for target |
4 |
> `/var/tmp/portage/sys-libs/glibc-2.12.1-r1/work/build-x86-x86_64-pc- |
5 |
linux-gnu-nptl/login/pt_chown' |
6 |
> ../Rules:120: warning: ignoring old commands for target |
7 |
> `/var/tmp/portage/sys-libs/glibc-2.12.1-r1/work/build-x86-x86_64-pc- |
8 |
linux-gnu-nptl/login/pt_chown' |
9 |
> mkdir /var/tmp/portage/sys-libs/glibc-2.12.1-r1/image/sbin |
10 |
> unexpected reloc type in static binary |
11 |
> make[1]: *** [install-symbolic-link] Error 139 |
12 |
> make: *** [install] Error 2 |
13 |
> emake failed |
14 |
> |
15 |
|
16 |
> ~/ > emerge --info |
17 |
> Portage 2.1.9.11 (default/linux/amd64/10.0/desktop, gcc-4.4.4, |
18 |
> glibc-2.11.2-r0, 2.6.34-gentoo-r2 x86_64) |
19 |
|
20 |
> ccache version 2.4 [enabled] |
21 |
|
22 |
> ACCEPT_KEYWORDS="amd64 ~amd64" |
23 |
|
24 |
> CFLAGS="-Os -march=k8 -mno-tls-direct-seg-refs -mmmx -msse3 -pipe |
25 |
> -fomit-frame-pointer" |
26 |
|
27 |
> FEATURES="assume-digests binpkg-logs ccache distlocks fixlafiles |
28 |
> fixpackages metadata-transfer news parallel-fetch protect-owned sandbox |
29 |
> sfperms strict unknown-features-warn unmerge-logs unmerge-orphans |
30 |
> userfetch" |
31 |
|
32 |
> MAKEOPTS="-j5 -s" |
33 |
|
34 |
No specific help on your error, but... |
35 |
|
36 |
One troubleshooting tip, when there are build issues, is to switch to |
37 |
MAKEOPTS=-j1, forcing it to serialize so the error won't be mixed in with |
38 |
a bunch of extraneous "noise" from other make jobs. The -s may or may not |
39 |
matter, but might as well eliminate it for the test as well, just in case. |
40 |
|
41 |
IOW, I'm not sure if those warnings are related or not, and with multiple |
42 |
jobs, it's quite possible the error is actually some way above the bit |
43 |
that you posted. -j1 solves that sort of problem as unlike with multiple |
44 |
make jobs, there's normally very little else after the error, with it. |
45 |
|
46 |
Meanwhile, AFAIK entirely unrelated to your problem but I'm curious as I |
47 |
always like to see what other people have in their CFLAGS when I have a |
48 |
chance, why are you using -mno-tls-direct-seg-refs ? I don't see anything |
49 |
in the gcc manpage info on that machine/arch-flag to suggest why I might |
50 |
wish to disable it, and it seems a rather odd arch-flag to simply throw in |
51 |
at random, so I'm wondering what the reasoning is. I'm definitely NOT |
52 |
saying it's wrong. Quite the contrary; with a bit more info, I might find |
53 |
it useful here... or not, but I don't know at this point. So got a link |
54 |
or something I could read? |
55 |
|
56 |
-- |
57 |
Duncan - List replies preferred. No HTML msgs. |
58 |
"Every nonfree program has a lord, a master -- |
59 |
and if you use the program, he is your master." Richard Stallman |