Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 238975 - cruft files in stage 3 hardened x86/i686
Summary: cruft files in stage 3 hardened x86/i686
Status: RESOLVED INVALID
Alias: None
Product: Gentoo Release Media
Classification: Unclassified
Component: Stages (show other bugs)
Hardware: All Linux
: High trivial (vote)
Assignee: Gentoo Release Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2008-09-28 20:15 UTC by djinnZ
Modified: 2008-09-29 12:08 UTC (History)
0 users

See Also:
Package list:
Runtime testing required: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description djinnZ 2008-09-28 20:15:17 UTC
In stage 3 archives downloaded at
ftp://ftp.unina.it/pub/linux/distributions/gentoo/releases/x86/current/stages/hardened/
there are an unexpected /lib/libgcc_s.so.1 file (not owned by any package, if requested for any reason must be owned by gcc in order to update it properly I think).
In stage 3 for i686 I find /usr/i486-pc-linux-gnu directory and some related files also.

Reproducible: Always

Steps to Reproduce:
Comment 1 Andrew Gaffney (RETIRED) gentoo-dev 2008-09-28 20:19:58 UTC
I believe the first file is created by gcc-config, so it wouldn't be owned by any package.

The i486-pc-linux-gnu stuff was probably created in pkg_postinst of the gcc ebuild in the original stage1 build. That means that portage won't remove them. There's nothing we can do about it.

Are any of these files actually causing you a problem, or are you just complaining for the sake of complaining? :)
Comment 2 djinnZ 2008-09-29 12:08:21 UTC
Yes, the first file is created by gcc-config, but because is not owned by gcc it will be not updated with gcc, expecially if it must be rebuild for any reason.
In the past I had problems in replacing and recompile a damaged copy (gcc compiled with eccessive optimization flags or on fault cpu) of the gcc or at version update (3/4 years in the past).
_Nothing urgent or sure cause of bug_ (limited to backward compatibility), but is possible cause of errors, better remove it.
In fact call gcc-config to refresh the state of the current gcc in a post_rm function can be a solution. As I have reported another minor problem 

The bug is just posted as trivial (Defined for "cosmetical issue" exactly what is not clear in them? ;) ) in relation of the presence of the files from the 486 optimization; also possible cause of mismatch in revdep-rebuild, fixed only with later versions, and especially cause of confusion in problems diagnosis (one of the tipical error of newbies is to install the x86 stage in front of the i686).

If you will find the time to fix is better, if is to complicate or will take too much time leave it as is. I was expect the bug closed immediately as wontfix or at least remind, in fact.

...and the time I spend to write this answer is too much for a minor issue as this. :(