Re: gcc 4.0.0 on amd64: problem with -O2?

[Andrew Pinski <pinskia at physics dot uc dot edu>]

On May 8, 2005, at 2:21 AM, Harald Dunkel wrote:

> Andrew Pinski wrote:
> > Wrong, try again.  Violating aliasing rules cause undefined behavior
> > so seg faulting is an okay thing to do.
>
> But producing a warning message and bad code is not OK. Either
> using a "type-punned pointer" should be treated as a fatal
> error, because gcc would create bad code, or gcc should create
> working code.
>
> If gcc knows that the code is bad, why does it continue?

Undefined code is required to compile.  There is a C defect report about
undefined and erroring out.  The code is undefined at runtime and not at
compile time.

So GCC is not producing bad code at all, your code/X11R6's code is bad.

-- Pinski