Gentoo Archives: gentoo-dev

From: Fabian Groffen <grobian@g.o>
To: gentoo-dev@l.g.o
Subject: Re: [gentoo-dev] Changing policy about -Werror
Date: Thu, 13 Sep 2018 16:03:57
Message-Id: 20180913160347.GC26329@gentoo.org
In Reply to: Re: [gentoo-dev] Changing policy about -Werror by Richard Yao
1 On 13-09-2018 07:36:09 -0400, Richard Yao wrote:
2 >
3 >
4 > > On Sep 12, 2018, at 6:55 PM, Thomas Deutschmann <whissi@g.o> wrote:
5 > >
6 > >> On 2018-09-12 16:50, Rich Freeman wrote:
7 > >> There is also the case where we want these warnings to block
8 > >> installation, because the risk of there being a problem is too great.
9 > >
10 > > I really disagree with that. So many devs have already said multiple
11 > > times in this thread that "-Werror" is only turning existing warnings
12 > > into fatal errors but "-Werror" itself doesn't add any new checks and
13 > > more often requires "-O3" to be useful.
14 > The way that compilers work is that the warnings are generated in the front end while the optimization level affects the backend. That means that -O3 has no effect on the code that does error generation. This remark about -O3 being needed to make -Werror useful is just plain wrong.
15
16 Huh? -O3 enables more checks, which can generate more warnings. -O3
17 isn't "needed", but if upstream is so interested in clean and correct
18 code, they should've fixed all warnings in the first place and thus
19 enabled all of them. In fact, I expect every sane upstream to use "-O3
20 -Wall -Werror" in one of their automated builds. Not that this catches
21 anything useful on x86{,_64} when there is for instance use of signed
22 and unsigned char types, so it isn't conclusive.
23
24 The whole point in here is that -Werror doesn't add much if you care.
25 The whole point why it is not desired in Gentoo is that users don't
26 necessarily are developers, or even interested in fixing warnings --
27 regardless whether they point to real problems or not.
28
29 If there are real problems in a package (exposed by a compiler or not)
30 then this should ideally stand out during ~arch testing, or even before
31 when the Gentoo maintainer examines the build (might even use -Werror
32 for his own purposes). If such code ends up in stable arch we just made
33 a stabilisation mistake, or got royally messed up by upstream, depending
34 how you look at it.
35
36 Fabian
37
38 --
39 Fabian Groffen
40 Gentoo on a different level

Attachments

File name MIME type
signature.asc application/pgp-signature

Replies

Subject Author
Re: [gentoo-dev] Changing policy about -Werror Richard Yao <ryao@g.o>
Re: [gentoo-dev] Changing policy about -Werror Richard Yao <ryao@g.o>