Carl Friedrich Bolz <[EMAIL PROTECTED]> wrote on 07/08/2006 15:56:55:
> Hi Ben! > > [EMAIL PROTECTED] wrote: > > Here's a patch that counts the number of times the different bytecodes > are > > used. I haven't tried doing translation yet as I can't use the work > > machine cycles, but it can't be far off. I've put the counts on the > object > > space as this is accessable from most places. > > > > I've also implemented something that should resort the opcodes in the > > switch statement of the bytecode dispatcher according to the statistics I > > found. This may have a slight speed improvement. > > Looks quite good. The only thing I don't really like is the fact that > the logbytecodes config options is in the global option name space. > Should probably be config.objspace.logbytecodes or even introduce an new > interpreter name space and be config.interpreter.logbytecodes > Thanks! I think it should have gone in objspace. It originally was, but then I moved the count onto ExecutionContext for a while, so I put it in global. Feel free to move it. The other bit I'm not sure about it the reporting of the results. At the momement it's just a hack in py.py, but it should go in objspace.finish. However, I couldn't work out how to print from there. Cheers, Ben > Cheers, > > Carl Friedrich > > _______________________________________________ [email protected] http://codespeak.net/mailman/listinfo/pypy-dev
