Discussion:
[Pearpc-devel] cpu_generic vs cpu_jitc_x86
Carlos Boneti
2006-01-03 17:44:25 UTC
Permalink
Hi everyone,

This summer I made some quick study (<2 weeks) involving the PearPC. I
came up with some comparative study between the jitc-x86 and the
cpu_generic implementation. Probably things that some [many] you already
know.

You can have a version of the report (2 pages) at the link below.
http://www.boneti.com/pearpc/report.pdf


My CVS version was from 6/2005 and, as a clue, the version of my
cpu_generic/ppc_cpu.cc was the same actually on the server (1.10), the
same for the cpu_jitc_x86/ppc_cpu.cc (1.10).

Well, I hope you find it interesting. If you think that we can take
something out of this report (like some documentation or something), or
just have critics, please let me know.

Thank you,

Carlos
--
Carlos Boneti
PhD Student
Dep. Arquitectura de Computadors - DAC
Univ. Politecnica de Catalunya
+34 635390755
***@boneti.com


La Ley, en su magnífica ecuanimidad, prohibe, tanto al rico como al
pobre, dormir bajo los puentes, mendigar por las calles y robar pan.
-- Anatole France. (1844-1924) Escritor francés.
Daniel Foesch
2006-01-04 17:54:48 UTC
Permalink
Well, this certainly answers some of the questions I had about
switching to a hybrid model.

I'll likely look into the information more, and such to see if there's
anything deep I can really get out of this study to help with what I'm
doing with PearPC. Thanks for the research though :)
Post by Carlos Boneti
Hi everyone,
This summer I made some quick study (<2 weeks) involving the PearPC. I
came up with some comparative study between the jitc-x86 and the
cpu_generic implementation. Probably things that some [many] you already
know.
You can have a version of the report (2 pages) at the link below.
http://www.boneti.com/pearpc/report.pdf
My CVS version was from 6/2005 and, as a clue, the version of my
cpu_generic/ppc_cpu.cc was the same actually on the server (1.10), the
same for the cpu_jitc_x86/ppc_cpu.cc (1.10).
Well, I hope you find it interesting. If you think that we can take
something out of this report (like some documentation or something), or
just have critics, please let me know.
--
Daniel Foesch
Carlos Boneti
2006-01-04 18:15:59 UTC
Permalink
Well, tell me if you want to see the data I found and more or less what.
It's a little mess so I could organize it better for you (even worse
because it's a 6months old work). If you need any workforce for the
task, just tell me. :)

Carlos
Post by Daniel Foesch
Well, this certainly answers some of the questions I had about
switching to a hybrid model.
I'll likely look into the information more, and such to see if there's
anything deep I can really get out of this study to help with what I'm
doing with PearPC. Thanks for the research though :)
Post by Carlos Boneti
Hi everyone,
This summer I made some quick study (<2 weeks) involving the PearPC. I
came up with some comparative study between the jitc-x86 and the
cpu_generic implementation. Probably things that some [many] you already
know.
You can have a version of the report (2 pages) at the link below.
http://www.boneti.com/pearpc/report.pdf
My CVS version was from 6/2005 and, as a clue, the version of my
cpu_generic/ppc_cpu.cc was the same actually on the server (1.10), the
same for the cpu_jitc_x86/ppc_cpu.cc (1.10).
Well, I hope you find it interesting. If you think that we can take
something out of this report (like some documentation or something), or
just have critics, please let me know.
--
Daniel Foesch
-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems? Stop! Download the new AJAX search engine that makes
searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_idv37&alloc_id865&op=click
_______________________________________________
Pearpc-devel mailing list
https://lists.sourceforge.net/lists/listinfo/pearpc-devel
--
Carlos Boneti
PhD Student
Dep. Arquitectura de Computadors - DAC
Univ. Politecnica de Catalunya
+34 635390755
***@boneti.com

La Ley, en su magnífica ecuanimidad, prohibe, tanto al rico como al
pobre, dormir bajo los puentes, mendigar por las calles y robar pan.
-- Anatole France. (1844-1924) Escritor francés.
Sebastian Biallas
2006-01-10 02:28:59 UTC
Permalink
Post by Carlos Boneti
Well, tell me if you want to see the data I found and more or less what.
It's a little mess so I could organize it better for you (even worse
because it's a 6months old work). If you need any workforce for the
task, just tell me. :)
If it's fun for you benchmarking those things, please go ahead. Not
everything there is really optimized..
Post by Carlos Boneti
Carlos
Sebastian

Loading...