-findirect-dispatch performance

Tue Sep 30 18:58:00 GMT 2003

Bryce McKinlay writes:
 > On Tuesday, Sep 30, 2003, at 04:47 Pacific/Auckland, Andrew Haley wrote:
 > 
 > > The difference between these figures is less than the measurement
 > > noise. So, I'd have to put my machine into single user mode to
 > > determine which was fastest.
 > 
 > oprofile could be useful here. It can measure the actual cycles used by 
 > the app, as report by the CPU's performance counters. I'm not sure if 
 > its 100% accurate, but should be much more so than using timers.
I've been looking for an excuse to investigate oprofile...
 > > Now, the embedded CaffeineMark is not the best Java benchmark program,
 > > so if anyone can suggest anything better I'm interested. However, on
 > > the basis of these timings I think we can agree that this is good
 > > enough.
 > 
 > Yes. I guess it depends how many cross-class field accesses the 
 > benchmark is doing, though.
 > 
 > Are you planning to check this in soon?
Tomorrow. I have a few changes.
One issue is that inlining is more-or-less useless with the current
patch. I will fix that.
Andrew.