On Apr 21, 2004, at 4:05 AM, Leopold Toetsch wrote:
Jeff Clites <[EMAIL PROTECTED]> wrote:
Here's another tiny patch, to let us fast-fail string_equal if we have cached hashval's which don't match.
Tested and applied now. I've also adoped JIT/i386 to use string_equal for C<eq> and C<ne> string ops. This speeds up these ops considerably *and* in the case what your test is showing, numbers indicate:
No stored hashval, 1000000 lookups each same: 0.018617 sec equal: 0.487958 sec not equal: 0.517992 sec With stored hashval, 1000000 lookups each same: 0.019674 sec equal: 0.487740 sec not equal: 0.038465 sec
... a factor ~14 performance increase for the "not equal" case.
Ah, great! (And the "not equal" case is the only one which should be showing a speed up--the "same" and "equal" cases are expected to be unaffected.)
Just to make sure... we're making sure the strings are always properly decomposed before comparing, right?
--
Dan
--------------------------------------"it's like this"-------------------
Dan Sugalski even samurai
[EMAIL PROTECTED] have teddy bears and even
teddy bears get drunk
