Summary: | merge of broken sed 4.1.4 causes segmentation faults on ppc | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Nicolas Vilz <niv> |
Component: | [OLD] Core system | Assignee: | The Gentoo Linux Hardened Team <hardened> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | dberkholz, nixnut |
Priority: | High | ||
Version: | unspecified | ||
Hardware: | PPC | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Bug Depends on: | 86857 | ||
Bug Blocks: |
Description
Nicolas Vilz
2005-03-06 10:25:19 UTC
if you emerge it with CFLAGS="-pipe" does it work ? sounds good... i merged sed with CFLAGS="-pipe" and tried to merge pam... looks very good... i'm creating libtool now... if i have any problems with pam, I will tell you :) looks good :) thx for that hint... no problems yet. merging now all other packages (thats alot) So, this bug could be considered as fixed. I don't understand how it can be considered fixed when no CFLAGS are being stripped in the ebuild, yet using pretty reasonable CFLAGS breaks it on ppc. I just emerged this and now I'm kinda stuck, because the sed configure script uses sed, so I can't downgrade without grabbing a binary package. CFLAGS="-O2 -mcpu=7450 -fno-strict-aliasing -pipe -fomit-frame-pointer -fstack-protector" (In reply to comment #5) > I don't understand how it can be considered fixed when no CFLAGS are being > stripped in the ebuild, yet using pretty reasonable CFLAGS breaks it on ppc. > > I just emerged this and now I'm kinda stuck, because the sed configure script > uses sed, so I can't downgrade without grabbing a binary package. > > CFLAGS="-O2 -mcpu=7450 -fno-strict-aliasing -pipe -fomit-frame-pointer > -fstack-protector" Do you have hardened in your use flags perhaps? Solar's patch works for me: https://bugs.gentoo.org/show_bug.cgi?id=86857#c6 What is the status of this bug? well, i don't have that selinux-ppc installation on my powerbook at the moment, i have some backup somewhere but don't have the time now to reroll it. I'd guess a "won't fix" or "can't fix", or quite "fixed with a work-around" would be the status... (In reply to comment #7) > What is the status of this bug? 1. Status FIXED in bug #83549, but that's probably because the true character of the problem was not recognised there. 2. A workaround here: https://bugs.gentoo.org/show_bug.cgi?id=86857#c6 3. And a comment that says "Don't do that" in comment #9 of the same bug. The choice seems to be between applying solars patch or not using pie. I've applied solars patch and haven't had any problems with it so far. This is a hardened bug, re-assigning to them. Is this still even remotely relevant? Haven't heard back and pretty sure this is fixed, so closing as such. Re-open if it isn't. |