[RFC] BZ #877 Math library function "logb" and "nextafter" inconsistent
Ryan S. Arnold
rsa@us.ibm.com
Fri Oct 6 15:35:00 GMT 2006
Jakub,
Regarding bugzilla 877 - Math library function "logb" and "nextafter"
inconsistent:
http://sources.redhat.com/bugzilla/show_bug.cgi?id=887#c9
Since you originally questioned the validity of this bug could you
please comment on my analysis.
I've reviewed the ISO-C, and IEEE754(r) specifications and I can find no
indication that the math functions are intended to return a consistent
value across rounding mode changes.
The currently observed behavior is consistent with the processor
specification. I'd like to reject this bug as INVALID.
Thanks,
Ryan S. Arnold
IBM Linux Technology Center
More information about the Libc-alpha
mailing list