PowerPC Floating Point compare bug?

Geoff Keating geoffk@geoffk.org
Wed May 16 19:37:00 GMT 2001


David Edelsohn <dje@watson.ibm.com> writes:

> 	The difference is the use of the PowerPC Floating-Point Select
> instruction in combination with float/double.

Yes.  Specifically, it's the code sequence fsel/fneg/fsel that gets
emitted for a comparison; GCC is emitting it wrongly.

-- 
- Geoffrey Keating <geoffk@geoffk.org>



More information about the Gcc mailing list