This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: What is acceptable for -ffast-math? (Was: associative law in combine)
- To: dewar at gnat dot com, torvalds at transmeta dot com
- Subject: Re: What is acceptable for -ffast-math? (Was: associative law in combine)
- From: dewar at gnat dot com
- Date: Tue, 31 Jul 2001 22:10:30 -0400 (EDT)
- Cc: fjh at cs dot mu dot oz dot au, gcc at gcc dot gnu dot org, gdr at codesourcery dot com,moshier at moshier dot ne dot mediaone dot net, tprince at computer dot org
by the way, I find that flush to zero is a perfectly reasonable semantics
for a C compiler, especially on machines that do this naturally. For example,
on an Alpha, I would do this by default, and only do real denormals in -mieee
mode, because otherwise the cost is indeed high, same think for the R10K.
There is certainly nothing in the C or C++ standard requiring denormals.