Re: Unreachable code warning conflicts with optomisation

["David Brown" <david at westcontrol dot com>]

> Your test2() function finds a real bug, but there's already
> a record in bugzilla about it; I found it by searching for
> Wunreachable-code.  Here's the record:
> http://gcc.gnu.org/PR8828
> It says it's fixed in gcc-3.2 and later; sure enough,
> in gcc-3.2.3 on my Debian system, it's fixed.
> I don't see any regression tests in gcc's testsuite for this yet,
> so I whipped one up by simplifying your testcase to the bone,
> and posted it.  It's at
>
> http://gcc.gnu.org/ml/gcc-patches/2003-09/msg01880.html
>
> and perhaps it will make it into the gcc tree.
> - Dan
>

Many thanks.  The older record has some interesting explanations of how
switch code is re-organized for efficiency.  It's more important that the
generated code is good - I can cope with superfluous warnings, and one day
(when I have time :-) I will start using a lint program properly.  But if
the gcc wizards can get good code and good warnings, then that's the best of
both worlds.

David




------
Want more information?  See the CrossGCC FAQ, http://www.objsw.com/CrossGCC/
Want to unsubscribe? Send a note to crossgcc-unsubscribe@sources.redhat.com