Re: compiler warnings, -Werror, etc.

[Colin Walters <walters verbum org>]

Hi Benjamin,

Thanks for responding!  The reason I structured my original email as a
question/proposal because I was interested in (ideally constructive)
individual maintainer feedback.

On Sat, 2012-07-28 at 12:58 +0000, Benjamin Otte wrote:

> I think the general consensus is that code that triggers these warnings is bad
> code. The job of source code is not just to ensure that the compiler generates
> correct output, but also to ensure that somebody who modifies that source code
> later will still make the compiler generate correct output.

No argument for me that fixing -Wmaybe-uninitialized is a valuable
activity.  However, I think it ranks lower in priority than
-Wformat-security or -Wmissing-prototypes.

> Second, I don't think you are the right person to determine which warnings are
> or are not important. Apparently the GCC developers think otherwise and I
> frankly trust them more than I trust you.

What does that mean, concretely?  You think nothing should change from
the status quo?
 
> I work in a module where people do not want to use -Werror. This leads to
> regular commits with warnings and nobody fixes them but me. 

Are they using the same version of GCC and such?  There are various
mechanisms to address this - if we have working continuous integration,
we can say "Ok, this commit introduced 3 new warnings on x86_64-opt".

>From what I can see in "git log" in gtk+, it looks like you fixed
several in March (awesome!), but only 2-3 commits since. Has it really
been a continuing burden recently?

> So I assume that 
> in projects where I am not a member these warnings often go unfixed for a long
> time.

Also note that what I'm suggesting isn't a regression from your
perspective.

> I think what we should do is restrict the list of supported compilers - in
> particular for git checkouts, but I'd argue we should require gcc >= 4.4 for
> GNOME compilation. We do the same thing with all the other stuff we depend on
> (autoconf, make, libraries, you name it) so why are we so nice on the 
> compiler front?

That does help some, but as someone who is very interested in keeping
everything building, -Werror=deprecated-declarations for example is
really problematic.