any impact on performance from javac's debug option ?
John.Rose at Sun.COM
Mon Aug 4 13:17:11 PDT 2008
We've seen cases where performance went down after an assert was
added, solely due to an inlining heuristic that flipped. There will
always be cases where a semantically null change will cause the
heuristics to detune. But as Tom says there are things we want to do
to make it happen less often, such as adjusting the sizing metrics to
not charge for unused basic blocks.
On Aug 4, 2008, at 1:09 PM, Tom Rodriguez wrote:
> Unfortunately yes. There's a bug filed about it though I don't
> remember the id offhand. The compilers should use sizing metrics
> which take asserts into account, along with having some bytecode
> specific metrics, but we haven't done any work on this yet. Hiding
> the existence of asserts is slighly tricky since it requires at
> least som control flow analysis.
> On Aug 4, 2008, at 12:53 PM, Charles Oliver Nutter wrote:
>> On a similar note, what about asserts? Do they add to bytecode
>> size and thereby affect inlining decisions?
More information about the hotspot-dev