| Summary: | emerging atk-1.4.0 with USE=+doc fails | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | John Whitlock <John-Whitlock> |
| Component: | [OLD] Library | Assignee: | Gentoo Linux Gnome Desktop Team <gnome> |
| Status: | RESOLVED INVALID | ||
| Severity: | normal | CC: | ph, richard |
| Priority: | High | ||
| Version: | unspecified | ||
| Hardware: | x86 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Attachments: |
Output of emerge attempt
Output of emerge attempt Output of emerge attempt Output of emerge attempt Output of emerge attempt |
||
|
Description
John Whitlock
2003-09-16 11:43:13 UTC
*** Bug 28947 has been marked as a duplicate of this bug. *** could you please attach the full build log? thanks. *** Bug 28932 has been marked as a duplicate of this bug. *** Sorry, I'm not going to be much help - the system than generated the bug is toast. I added doc to make.conf after running for a few weeks. I thought this might be the problem (previous packages built without doc), so I did the following to rebuild my system, as sugested by the Portage manual: # emerge --update world --emptytree The emerge failed on gcc. On reboot, lots of things fail, including: emerge: line 9, ImportError: No Module named time depscan.sh: /lib/rcscripts/filefuncs.so: undefined symbol: CONVFMTidx vi: Segmentation fault So, I have a screwed system. If I get it working again, I'll post the atk build log (assuming it's still working) OT: That bug is gcc 3.3.1-r2 and rebuilding. If possible revert to gcc 3.3.1-r1 and rebuild python and it should work Created attachment 17909 [details]
Output of emerge attempt
I've installed a binary gcc-3.2.3-r1 and python-2.2.3, and I can now emerge atk
with the same results as previously. build.all was created with:
bash# emerge atk &> build.all
Created attachment 17910 [details]
Output of emerge attempt
I've installed a binary gcc-3.2.3-r1 and python-2.2.3, and I can now emerge atk
with the same results as previously. build.all was created with:
bash# emerge atk &> build.all
Created attachment 17911 [details]
Output of emerge attempt
I've installed a binary gcc-3.2.3-r1 and python-2.2.3, and I can now emerge atk
with the same results as previously. build.all was created with:
bash# emerge atk &> build.all
Created attachment 17912 [details]
Output of emerge attempt
I've installed a binary gcc-3.2.3-r1 and python-2.2.3, and I can now emerge atk
with the same results as previously. build.all was created with:
bash# emerge atk &> build.all
Created attachment 17913 [details]
Output of emerge attempt
I've installed a binary gcc-3.2.3-r1 and python-2.2.3, and I can now emerge atk
with the same results as previously. build.all was created with:
bash# emerge atk &> build.all
Comment on attachment 17913 [details]
Output of emerge attempt
I've installed a binary gcc-3.2.3-r1 and python-2.2.3, and I can now emerge atk
with the same results as previously. Captured with:
bash# emerge atk &> build.all
Sorry about the spam. Bugzilla was giving me errors, and I tried resubmitting rather than seeing if the attachment worked. can you try using some different CFLAGS ? like remove -mcpu= and -Os ? Sorry, the system is dead again. I tried:
> emerge -e gcc
Which builds... just about everything. The build died on glibc-2.3.2-r1, with 'multiple definition of `_errno`, but other things that were built are now in trouble (emerge, python, etc.) Plus, the disk is getting too full (92%, 500 MB left).
I'm thinking of going back to a stage 3 or stage 1 system, with known good versions of all the tools. I'm headed out of town for the weekend, so I don't mind a three-day build.
resolved invalid as it looks like system trouble to me, reopen if needed after a clean install After clean install (only 10 days!), atk-1.4.0 builds cleanly with same compiler options. NOT using ~x86 anymore, so I had to specify version 1.4.0, but this bug has gone away for me. I agree this was system trouble, probably related to gcc 3.3.1-r2 and invalid library builds. |
