bump to v2.0.21 - Githubissues

HigherOrderCO / HVM

A massively parallel, optimal functional runtime in Rust

Apache License 2.0

10.52k stars 400 forks source link

Perf run for 95237ae:

compiled
========

file            runtime         main            (local)       
==============================================================
sort_bitonic    c                       13.10s          12.58s
                cuda                     0.14s           0.14s
--------------------------------------------------------------
sum_rec         c                        2.90s           2.86s
                cuda                     0.05s           0.05s
--------------------------------------------------------------
sum_tree        c                        1.47s           0.26s
                cuda                     0.02s           0.02s
--------------------------------------------------------------
tuples          c                       11.57s          10.59s
                cuda                   timeout         timeout
--------------------------------------------------------------

interpreted
===========

file            runtime         main            (local)       
==============================================================
sort_bitonic    c                       15.21s          11.46s
                cuda                     0.14s           0.14s
                rust                   timeout         timeout
--------------------------------------------------------------
sum_rec         c                        3.88s           4.15s
                cuda                     0.05s           0.05s
                rust                   timeout         timeout
--------------------------------------------------------------
sum_tree        c                        0.49s           0.41s
                cuda                     0.01s           0.01s
                rust                     1.96s           2.07s
--------------------------------------------------------------
tuples          c                        5.75s           7.81s
                cuda                   timeout         timeout
                rust                     9.94s           8.67s
--------------------------------------------------------------

compiled ======== file runtime main (local) ============================================================== sort_bitonic c 10.09s 11.03s cuda 0.14s 0.14s -------------------------------------------------------------- sum_rec c 2.37s 2.41s cuda 0.05s 0.05s -------------------------------------------------------------- sum_tree c 0.29s 0.23s cuda 0.02s 0.02s -------------------------------------------------------------- tuples c 8.79s 10.94s cuda timeout timeout -------------------------------------------------------------- interpreted =========== file runtime main (local) ============================================================== sort_bitonic c 10.05s timeout cuda 0.14s 0.14s rust timeout timeout -------------------------------------------------------------- sum_rec c 5.40s 7.59s cuda 0.05s 0.05s rust timeout timeout -------------------------------------------------------------- sum_tree c 0.70s 0.41s cuda 0.01s 0.01s rust 2.09s 1.96s -------------------------------------------------------------- tuples c 7.29s 7.61s cuda timeout timeout rust 8.76s 8.91s --------------------------------------------------------------

HigherOrderCO / HVM

bump to v2.0.21 #402