On Fri, Jun 11, 2010 at 7:33 PM, Kevin D. Kissell <kevink@paralogos.com> wrote:
>> An optimized, assembly-language soft-float library implementation is *much*
>> faster than the kernel emulator, but I benchmarked it once upon a time
>> against a portable gnu soft-float library in C, and the difference wasn't
>> nearly as dramatic.
> The in-kernel emulator always works. The float conformance test app Ralf
> pointed out a few weeks ago doesn't run correctly when built with a recent
> softfloat gcc with any optimization higher than O0 (tested with 4.4.4, 4.3.4).
> I'd take correctness over speed any day of the week...
Thank you all for your comments. I have compiled a new crosstoolchain( gcc
4.3.3 binutils 2.19.1 and glibc 2.9) and library with soft float enabled. I
could see considerable performance improvement with my application.
Coming back to correctnes , I ran paranoia.c [1] for verification and I got
following result
"No failures, defects nor flaws have been discovered.
Rounding appears to conform to the proposed IEEE standard P754.
The arithmetic diagnosed appears to be Excellent!"
Shall I beleive the results and go ahead with soft float ?. Is there any other
test to verify reliability of soft float ?.
Thanks and Regards
Jabir
[1] http://www.netlib.org/paranoia/paranoia.c
|