benchmarks game

54 views
Skip to first unread message

Raoul Schorer

unread,
Oct 25, 2020, 6:47:39 AM10/25/20
to Racket Users

Hi,

I got interested into The Computer Language Benchmarks Game.

The binary-trees benchmark is different between the BC (parallel) and CS (single threaded) versions. On trying to execute the parallel version on RacketCS with the expected commands (`raco make binarytrees`, and then `racket binarytrees 21`), I get nothing printed on stdout while everything is working as expected with the same commands for the single-threaded version. Why is that?

As an aside, doing the same without compiling first yields the expected behaviour in both programs and the parallel version seems faster in RacketCS, so it would seem like a worthy change to the benchmarks.

Cheers,
Raoul

Matthew Flatt

unread,
Oct 25, 2020, 9:21:10 PM10/25/20
to Raoul Schorer, Racket Users
Hi,

Can you say more about which version of Racket you're running and on
what platform?

Thanks!
Matthew
> --
> You received this message because you are subscribed to the Google Groups
> "Racket Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to racket-users...@googlegroups.com.
> To view this discussion on the web visit

Raoul Schorer

unread,
Oct 26, 2020, 6:17:17 AM10/26/20
to Racket Users
Hi Matthew,

I tried on:

- Windows 8.1 + RacketCS 7.8
- Linux Debian 10 + Racket 7.9 (current master branch)
- Linux Debian 10 + RacketBC 7.8

All have the same behaviour: while the sequential program prints to stdout as expected in all cases (with or without `raco make` prior to running, and running either the .rkt or the .zo), the parallel version indeed prints to stdout when doing `racket bt_parallel.rkt 21` but running the bytecode directly as `racket bt_parallel.zo 21` returns without printing.

Perhaps I'm missing something obvious since it apparently is working correctly on the benchmark website?

Thanks!
Reply all
Reply to author
Forward
0 new messages