This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: type based aliasing - summary
- To: pcg at opengroup dot org
- Subject: Re: type based aliasing - summary
- From: Richard Stallman <rms at gnu dot org>
- Date: Sun, 19 Sep 1999 14:53:01 -0600 (MDT)
- CC: gcc at gcc dot gnu dot org
- Reply-to: rms at gnu dot org
Personally I think, the first alternative is plain silly, the arguments to
support it also support switching off any other optimization.
If one considers the situation based only on its qualitative features,
then many of the same arguments that apply to this situation might
apply to many others. The same kinds of advantages and disadvantages
occur, more or less.
But the weight of each advantage or disadvantage varies from case to
case, depending on the circumstances. So in general each decision is
different.