Author pitrou
Recipients collinwinter, jyasskin, pitrou
Date 2009-01-13.23:46:52
SpamBayes Score 0.179059
Marked as misclassified No
Message-id <1231890431.7109.15.camel@localhost>
In-reply-to <43aa6ff70901131529n34730cbbk74cc48455fbde86c@mail.gmail.com>
Content
> http://code.google.com/p/spitfire/. It's a template system designed
> for performance that I have some experience with.

6% faster on a template system simply by optimizing conditional jumps is
quite neat.
(the spitfire homepage itself is rather intriguing, they claim to be
faster than plain string joining...)

> I was mostly curious whether the POP_OR_JUMP and JUMP_OR_POP opcodes
> had a noticeable performance impact, ie, do they make things fast
> enough to warrant their inclusion over the old JUMP_IF_FALSE
> implementations.

Well, I don't remember really, although this is only a few weeks old.
POP_OR_JUMP / JUMP_OR_POP will be used when the "and" and "or" keywords
are involved. That's basically the only situation where you don't want
to pop the top of stack after a conditional jump.
History
Date User Action Args
2009-01-13 23:46:53pitrousetrecipients: + pitrou, collinwinter, jyasskin
2009-01-13 23:46:53pitroulinkissue4715 messages
2009-01-13 23:46:53pitroucreate