|Low-order uncertainty in 87 math processing email@example.com (1997-09-24)|
|Re: Low-order uncertainty in 87 math processing firstname.lastname@example.org (David L Moore) (1997-09-24)|
|Re: Low-order uncertainty in 87 math processing email@example.com (William D Clinger) (1997-09-27)|
|From:||William D Clinger <firstname.lastname@example.org>|
|Date:||27 Sep 1997 01:10:53 -0400|
> [code not changing x or y]
> [ why is q not zero]
David L Moore's suggestion (that the compiler's choice between
memory and floating point registers can affect precision) would
explain why q is not always zero, and John's suggestion (that
the precision control flags are being changed) would then explain
why the problem sometimes goes away until reboot.
Return to the
Search the comp.compilers archives again.