Re: How to understand gcc source code?

["Richard Guenther" <richard dot guenther at gmail dot com>]

On Sat, Mar 22, 2008 at 7:16 PM, Denys Vlasenko
<vda.linux@googlemail.com> wrote:
> On Saturday 22 March 2008 11:14, Basile STARYNKEVITCH wrote:
>  > Actually you ask an interesting & difficult question. Some thoughts from
>  > somebody who works on GCC for more than a year but still considers
>  > himself a newbie:
>  >
>
> > * my impression is that nobody understands fully the GCC compiler. Maybe
>  > there are some few exceptions, but I tend to believe that most of GCC
>  > contributors (in particular myself) do not understand all of it, and had
>  > to give up the idea of grasping it all! As usual, the way to go into it
>  > is abstraction: you agree to not understand fully some stuff, and you
>  > abstract it.
>
>  Cons of such an approach: the project as a whole becomes somewhat unmanaged
>  (unmanageable?) and sometimes accumulates unneeded layers of abstraction
>  and other random cruft without anyone noticing/caring about it.
>
>  Your example of MS Windows is a prime one of overgrown project
>  out of human control.
>
>
>  > * on the positive side, GCC is still doing well and alive
>
>  Then why LLVM (at least is said to) surpass gcc in generated code's quality?
>  Why Intel and MS compilers are surpassing it?

Are they?

Honestly, as a volunteer driven (and admittedly "evolved") project it
is obviously
more "sexy" to spend time on new features than to improve maintainability by
doing (often non-trivial) cleanup work.  At some point starting over is the more
appealing solution from an engineering perspective (and LLVM certainly has
gained momentum from its "freshness"), but it is reality that this is
not an option
for GCC.

And of course you can't easily have both, reduced cruft and better code quality
with the same amount of development ressources.  Everyone set his priorities.

>  Sorry for "negative" comment, but burying your head in the sand
>  and pretending that everything is ok does not seem like
>  a good strategy to me.

I don't think anyone does this.  But non-constructive critism is not
appreciated.

Thanks,
Richard.