Recently, on a MIPS box a couple of ms away ...
[root@tbird rc5]# ./rc564 -benchmark
RC5-64 v2.6401 client - a project of distributed.net
Copyright distributed.net 1997
Visit http://www.distributed.net/ for more information. 'rc564 HELP' for usage
Benchmarking with 10000000 tests:
.....10%.....20%.....30%.....40%.....50%.....60%.....70%.....80%.....90%....
Complete in 88.86 seconds. [112532.37 keys/sec]
[root@tbird rc5]# uname -a
Linux rm200.mips-man.org 2.1.56 #204 Wed Nov 26 23:21:24 PST 1997 mips unknown
[root@tbird rc5]# cat /proc/cpuinfo
cpu : MIPS
cpu model : R4600 V2.0
system type : SNI RM200 PCI
BogoMIPS : 133.12
byteorder : little endian
unaligned accesses : 48
wait instruction : yes
microsecond timers : yes
extra interrupt vector : no
hardware watchpoint : no
[root@tbird rc5]#
How about joining the RC5-64 effort as "Team MIPS"? I'll build binary
packages and will make them available via the usual ftp sites (ftp.fnet.fr,
ftp.linux.sgi.com) as well as via www.distributed.net.
I didn't take myself the time for comparing the benchmarks with other
machine but I think the numbers leave whishes about higher performance
open.
Btw, the RC5 client for MIPS was delayed alot by the recently reported
libg++ showstopper bug. I haven't fixed the problem yet for real but
once again it smells like a dynamic linker bug. I was able to work
around by a static (yuck) executable.
Ok, and now it's up to you people to burn your cycles.
Ralf (... gimme all your CPU ...)
|