> From:   "Andreas Raab" < andreas.raab at g... >
> Date:   Sat Jul 19, 2003  8:14 pm
> Subject:   RE: [VM] HashBits, a lazy way
>
>
>
> Hi John,
>
> > Not quite the same, however in running tinybenchmarks I noted
> >
> > june 30thVM 40,738,383 bytecode/sec 972,076 sends/sec
> > 3.6.1bxVM 40,429,564 bytecodes/sec 984,169 sends/sec
>
> Interesting. I actually +10% in both, sends and bytecodes per second  
> based
> purely on the code generator enhancements which (consistently)  
> translate
> into +5% in macro benchmarks. I wonder if that's because of Processor
> differences (you say "Pentium" and by that I presume you mean PI) or  
> if it
> might be an effect of the dual processor system or if it could be a
> difference in the compiler (I use 2.95.2) or optimization flags used.

To clarify in the startup log it says
2 of these...
Pentium II/Pentium II Xeon/Celeron (350.80-Mhz 686-class CPU)
Origin = "genuineIntel", Id = 0x652 stepping =2
Features=0x183fbff

It's not a Celeron, comes in the paperback sized black things that fit  
in a slot...

gcc version is 2.95.2
-g -O2 -fomit-frame-pointer


We are comparing AHC changes versus your localization changes. Versus  
say a interp.c that has 20+ t1,t2,... in it... Don't know if the 10%  
you talk about is AHC(sp) CGeneratorEnhancements-ajh.1.cs versus yours?  
or to a VM that didn't have the change...

>
> > Similar performance, but the 3.6.1bxVM has better
> > sends/second. Due to the CodeGenEnh I think...
> > Certainly one is not extraordinary faster than the other.
> >
> > However when I looked at allocation rates for 1 asFloat,
> >2 at 3... , Array new: 5 I saw
> >
> > june30thVM 37,654, 40,198, 31,037
> >
> > 3.6.1bxVM 75,952, 82,524, 39,325
>
> Now _that's_ interesting. BTW, I don't think you've ever posted the
> benchmark - can you please post it, so that we can compare the results
> appropriately?
>
> Cheers,
> - Andreas

Sure declare JMMWhy float as a global, set to zero, then inspect
this below. Of course you need something else to periodically get the  
time, and counter and
do math etc to compare. I'll note the +1 versus oh say +5, but we're  
more interested in the
differences between the VM, changing to +5 makes the numbers bigger but  
doesn't affect compare results.

^100000000 timesRepeat: [
1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat.
JMMwhy _ JMMwhy + 1.
1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat.
JMMwhy _ JMMwhy + 1.
1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat.
JMMwhy _ JMMwhy + 1.
1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat.
JMMwhy _ JMMwhy + 1.
1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat.
JMMwhy _ JMMwhy + 1.
1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat.
JMMwhy _ JMMwhy + 1.
1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat.
JMMwhy _ JMMwhy + 1.
1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat.
JMMwhy _ JMMwhy + 1.
1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat.
JMMwhy _ JMMwhy + 1.
1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat. 1 asFloat.
JMMwhy _ JMMwhy + 1.
	Processor yield
]] fork

--
======================================================================== 
===
John M. McIntosh <johnmci at smalltalkconsulting.com> 1-800-477-2659
Corporate Smalltalk Consulting Ltd.  http://www.smalltalkconsulting.com
======================================================================== 
===