| Summary: | app-text/ghostscript-gpl-9.15-r1 fails to build with "unpack: failure unpacking ghostscript-gpl-9.12-patchset-1.tar.bz2" | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Alex Orange <crazycasta> |
| Component: | Current packages | Assignee: | Printing Team <printing> |
| Status: | RESOLVED FIXED | ||
| Severity: | normal | CC: | Tanktalus |
| Priority: | Normal | ||
| Version: | unspecified | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Attachments: |
emerge --info '=app-text/ghostscript-gpl-9.15-r1::gentoo'
emerge -pqv '=app-text/ghostscript-gpl-9.15-r1::gentoo' environment build.log |
||
|
Description
Alex Orange
2015-08-04 17:50:23 UTC
Created attachment 408300 [details]
emerge --info '=app-text/ghostscript-gpl-9.15-r1::gentoo'
Created attachment 408302 [details]
emerge -pqv '=app-text/ghostscript-gpl-9.15-r1::gentoo'
Created attachment 408304 [details]
environment
Created attachment 408306 [details]
build.log
P.S. I masked this version and previous package works fine. I can not reproduce this bug on my system. Ok, thank you Jonas Stein. Turns out my tar was broken. Upon further investigation, tar does seem to work with -O3. Should I file a separate bug for that or rename this one or what? # tar tvjf /usr/portage/distfiles/ghostscript-gpl-9.12-patchset-1.tar.bz2 tar: This does not look like a tar archive tar: Skipping to next header tar: Exiting with failure status due to previous errors I'm getting the same problem, and I'm not sure how to fix this. (Using stable tar: 1.27-r2) Upgrading tar to 1.28-r1 seems to solve the problem. Perhaps tar 1.28-r1 should be stabilised instead? Oops, just realized that I screwed up what I wrote. Meant to put that tar DOESN'T work with -O3. I would be hesitant to say 1.28-r1 necessarily "fixes" the problem unless there's something in their patch notes about it. It's an optimization problem so it could very well be that it's magically fixed for certain arches and not others. TL;DR: unless upstream claims they fixed an optimization related bug I think tar should force -O2 instead of -O3. I have the same issue with the following flags:
CFLAGS="-march=native -O2 -pipe -ftree-vectorize -fno-stack-protector"
CXXFLAGS="${CFLAGS}"
Rebuilding tar without the -ftree-vectorize flag fixes the problem.
(In reply to Tanktalus from comment #9) > Upgrading tar to 1.28-r1 seems to solve the problem. Perhaps tar 1.28-r1 > should be stabilised instead? 1.29-r1 is stable now, so this should be fixed. |