Hi,

I accidentally sent a message to Igor off-list. Here's a summary of the brief 
conversation that ensued.

Am 24.12.2010 um 15:25 schrieb Igor Stasenko:
> Interesting. The code it emitting using only three registers: eax, ecx and 
> esp.
> And without any heavy optimization and any kind of register allocation
> whatsoever, it is 40% of speed of optimized C code.

To this, I replied that optimising was "not too hard" given the intermediate 
representation were ASTs of the same shape as the input, and that this would 
perhaps be even more true for the code generation chain described in the S3 
paper, where Ian even suggested some peep hole optimisations that look quite 
straightforward. I also added that details about these optimisations could be 
found in the S3 paper, Sec. 4, and that the paper is available from 
http://www.vpri.org/pdf/tr2010003_PEG.pdf.

Best,

Michael

-- 
Dr.-Ing. Michael Haupt                [email protected]
Software Architecture Group           Phone:  ++49 (0) 331-5509-542
Hasso Plattner Institute for          Fax:    ++49 (0) 331-5509-229
Software Systems Engineering          http://www.hpi.uni-potsdam.de/swa/
Prof.-Dr.-Helmert-Str. 2-3, D-14482 Potsdam, Germany 

Hasso-Plattner-Institut für Softwaresystemtechnik GmbH, Potsdam
Amtsgericht Potsdam, HRB 12184
Geschäftsführung: Prof. Dr. Christoph Meinel






_______________________________________________
fonc mailing list
[email protected]
http://vpri.org/mailman/listinfo/fonc

Reply via email to