Summary: | [4.1.1/build] stage2/xgcc not found | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Robin Johnson <robbat2> |
Component: | [OLD] Core system | Assignee: | Gentoo Toolchain Maintainers <toolchain> |
Status: | RESOLVED DUPLICATE | ||
Severity: | critical | ||
Priority: | Highest | ||
Version: | 2006.0 | ||
Hardware: | PPC | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: |
Failure log from gcc-4.1.1 compile
Failure log with MAKEOPTS=-j1 FEATURES=-confcache emerge gcc Failure log with MAKEOPTS=-j1 FEATURES=-confcache emerge gcc |
Description
Robin Johnson
2006-05-29 15:22:42 UTC
Created attachment 87830 [details]
Failure log from gcc-4.1.1 compile
gzipped log, as it was too big to upload directly.
Is this reproducible? I've had no problems on PPC. Yup, this is still reproducable for me. Going MAKEOPTS=-j1 makes it work more often, but no gaurentee (at least one failure with -j1 iirc). Not 100% reproducable: root@buck-int:/var/log/portage # pwd /var/log/portage root@buck-int:/var/log/portage # find -name '*-gcc-4*' ./4658-gcc-4.1.1.log ./4000/4468-gcc-4.1.1.log ./4000/4468-gcc-4.1.1.log.1 ./4000/4469-gcc-4.1.1.log ./4000/4594-gcc-4.1.1.log.0 ./4000/4525-gcc-4.1.1.log ./4000/4521-gcc-4.1.1.log ./4000/4594-gcc-4.1.1.log root@buck-int:/var/log/portage # find -name '*-gcc-4*' | xargs grep 'xgcc.*not found' -i ./4658-gcc-4.1.1.log:make[5]: stage2/xgcc: Command not found ./4000/4468-gcc-4.1.1.log.1:make[5]: stage2/xgcc: Command not found ./4000/4521-gcc-4.1.1.log:make[5]: stage2/xgcc: Command not found ./4000/4594-gcc-4.1.1.log:make[5]: stage2/xgcc: Command not found Just reproduced it again. FEATURES=-confcache MAKEOPTS=-j1 emerge '=gcc-4*' I'll attach the logfile in a sec. Created attachment 88571 [details]
Failure log with MAKEOPTS=-j1 FEATURES=-confcache emerge gcc
Created attachment 88572 [details]
Failure log with MAKEOPTS=-j1 FEATURES=-confcache emerge gcc
attachment is a gzipped log again. Ok, I'm seeing some other weird behavior with make due to timestamps and the fact my box is very fast, give me some more time to look at it. stop using tmpfs and things will fix itself *** This bug has been marked as a duplicate of 114722 *** Sorry Spanky. You lose. I can reproduce this outside of tmpfs as well. I do however have another lead on the source of the problem. still not a bug in gcc *** This bug has been marked as a duplicate of 114722 *** |