Gentoo Archives: gentoo-dev

From: Michael Orlitzky <mjo@g.o>
To: gentoo-dev@l.g.o
Subject: Re: [gentoo-dev] Re: portage reliance on GNU objcopy ownership perseverance behavior in strip
Date: Wed, 10 Feb 2021 02:44:53
Message-Id: b04ab01d5b6dfdb1f12a593aed1eb443dda9ba2f.camel@gentoo.org
In Reply to: Re: [gentoo-dev] Re: portage reliance on GNU objcopy ownership perseverance behavior in strip by Manoj Gupta
1 On Tue, 2021-02-09 at 18:25 -0800, Manoj Gupta wrote:
2 > >
3 > > Root is the owner but often there is also a group that has access to the
4 > files.
5 > After stripping with llvm-strip, new ownership is root:root instead of
6 > root:<group>.
7 > Therefore, the members of the group lose access to the files post stripping.
8 >
9 > We found this issue in Chrome OS when we tried to switch the defaults to
10 > llvm's objcopy/strip.
11
12 Thanks, I still agree that some consistent behavior is needed. The only
13 reason I asked is because the fact that you *noticed* the problem is a
14 red flag to me. A suid group is a valid use case, but a few of these
15 examples don't set any special group permissions on the executables
16 whose group they change. For example...
17
18 > ./net-analyzer/netselect/netselect-9999.ebuild: fowners root:wheel
19 > /usr/bin/netselect
20
21 This ebuild does,
22
23 fowners root:wheel /usr/bin/netselect
24 fperms 4711 /usr/bin/netselect
25
26 which makes you wonder why it changed the group in the first place. The
27 ebuilds for mail-filter/procmail and games-arcade/xboing are also
28 suspect.

Replies