Re: [tree-ssa] dead const/pure/alloca call removal

[Andreas Jaeger <aj at suse dot de>]

Gabriel Dos Reis <gdr@integrable-solutions.net> writes:

> Daniel Jacobowitz <drow@mvista.com> writes:
>
> [...]
>
> | The implementation provides malloc; GCC is merely a part of the
> | implementation.  We already assume a conforming library for many
> | optimizations; glibc doesn't provide strlen either.
>
> Sure, but isn't a tracing malloc a conforming implementation?

We do have -fno-builtins and e.g. -fno-builtin-sin and there could be
also a -fno-builtin-malloc for such cases.  

> | I disagree that the C standard's definition is too weak to allow
> | eliding of calls, as-if.  What problems do you see?
>
> The as-fi would apply *if* GCC is the malloc implementation provider.
> Currently , it is not,

Then we shouldn't do any other builtins at all.  GCC does not provide
sin either but optimizes it in some cases...

Andreas
-- 
 Andreas Jaeger, aj@suse.de, http://www.suse.de/~aj
  SuSE Linux AG, Deutschherrnstr. 15-19, 90429 Nürnberg, Germany
   GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126

Attachment: pgp00000.pgp
Description: PGP signature