Re: Suspicion of regression in uninitialized value detection

[David Brown <david at westcontrol dot com>]

On 06/12/2011 20:33, Jeff Law wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On 12/06/11 12:21, Ian Lance Taylor wrote:
> > While using the optimizers to improve the quality of uninitialized
> > warnings does have some benefits, those benefits are outweighed by
> > the drawbacks.  We need to completely reimplement this warning,
> > either in the C/C++ common frontend, or before any optimization
> > passes run.
> Run it just after going into SSA; mark everything that has an
> uninitialized use.  For those who want the larger, much more stable
> set of warnings this is going to meet their needs much better.
>
> That's what my patch from a few years ago did.  IIRC the only
> outstanding issue was what should the default be.

Would it be possible then to have switches for different levels, such as 
is done with the strict aliasing warnings?

> In theory we could go ahead and translate into SSA when not optimizing
> which would remove the dependency on -O, at the expense of
> compile-time performance.

Do people often intentionally run gcc without any optimisations these 
days?  The compiler speed on -O1, relative to the speed of host 
computers, means there are rarely compile speed reasons to turn off 
optimisations.  And for debugging, -O1 is often much easier to work with 
than -O0, as the generated assembly makes more sense (more use of 
registers and less blind use of the stack).  If that is the case, then 
it may be legitimate to always enable the basic analysis passes that are 
needed for uninitialised warnings.

> > Then we can rename the existing warning to
> > -Wunpredictable-uninitialized and take it out of -Wall.
> I'd tend to think this would be a mistake.  We'd be changing -Wall in
> a very very visible way and I suspect the number of developers
> complaining would be huge.
>
> Ideally we'd just nuke -Wall and have everyone use the -W options they
> care about, but sadly that's just a pipe dream.

In my opinion, we should nuke -Wall - these warning options should all 
be enabled unless explicitly /disabled/ with the -W options.  And 
-Werror should be enabled by default too.  Just think how many bugs 
would be found in existing software if developers were pushed into using 
their tools effectively!



> jeff