ecraven / r7rs-benchmarks

Benchmarks for various Scheme implementations. Taken with kind permission from the Larceny project, based on the Gabriel and Gambit benchmarks.
271 stars 32 forks source link

Fix code. #14

Closed michaellenaghan closed 8 years ago

michaellenaghan commented 8 years ago

Btw, you can add ?w=1 to the GiHub URL to hide whitespace differences.

michaellenaghan commented 8 years ago

Btw, the benchmarks should run much faster now; the bugs in fft.scm and mbrotZ.scm created endless loops, which caused those particular benchmarks to always run to CPU_LIMIT for all implementations.

ecraven commented 8 years ago

Thanks for catching these, great work!