Gentoo Archives: gentoo-dev

From: Alexis Ballier <aballier@g.o>
To: gentoo-dev@l.g.o
Subject: Re: [gentoo-dev] Developers, please work on underlinking issues!
Date: Fri, 19 Aug 2016 12:47:06
Message-Id: 20160819144648.217a5462@portable
In Reply to: Re: [gentoo-dev] Developers, please work on underlinking issues! by "Michał Górny"
1 On Fri, 19 Aug 2016 06:58:44 +0200
2 Michał Górny <mgorny@g.o> wrote:
3
4 > On Thu, 18 Aug 2016 15:21:16 +0200
5 > Alexis Ballier <aballier@g.o> wrote:
6 >
7 > > On Thu, 18 Aug 2016 08:13:14 -0400
8 > > Rich Freeman <rich0@g.o> wrote:
9 > >
10 > > > If you just check your packages occassionally to make sure they
11 > > > build with gold it completely achieves the goal, and it will
12 > > > actually result in fewer bugs using the non-gold linker as
13 > > > well.
14 > >
15 > > That's what a tinderbox is for. The only QA problem I see here is
16 > > that QA doesn't automate that kind of checks anymore since Diego
17 > > left. Maybe QA should ask Toralf to run a ld.gold tinderbox and
18 > > avoid asking people to randomly test random packages ?
19 >
20 > Yes, tinderboxing makes a lot of sense if the bugs are afterwards
21 > ignored by package maintainers.
22
23 "ignored" is rather strong here; considering people join gentoo and
24 basically work for free, I'd rather say "not considered important
25 enough" which yields basically the same result but without accusing
26 people of refusing to fix a bug. With that in mind, everybody is free
27 to submit patches to bugzie, and if it is a QA goal, QA is free to
28 apply it after some time. Nothing is blocked here, expect when people
29 prefer to hit fellow developers "with a cluebat" rather than getting
30 things done.
31
32 > Or in the best case, the maintainer
33 > tells reporter (Toralf) to file the bug upstream.
34
35 It's not because Toralf reported it that he is the only one that has
36 the right to report it upstream. Obtaining fixes from those that know
37 the package best is good, isn't it ? I thought we learned from the past
38 and tried to avoid, e.g., fixing valgrind warnings in libssl by our
39 own :)
40
41 Alexis.