RFR (S): 8059241: Incremental inlining is too hot when compiling Nashorn/Octane
vladimir.x.ivanov at oracle.com
Wed Apr 29 18:11:08 UTC 2015
Sigh, I missed an important case.
Incremental inlining fails miserably when a call site dies while waiting
for being inlined. It happens when previously inlined call causes some
branches to be eliminated, but the info hasn't been propagated yet.
I tried to enhance dead code detection logic, but failed. So, I reverted
the following part of original fix:
"(1) Reduce PhaseRemoveUseless frequency: inline in larger chunks
until IR size LiveNodeCountInliningCutoff, then eliminate dead nodes."
On 4/27/15 11:14 AM, Vladimir Ivanov wrote:
> Thanks for reviews, John, Roland, and Aleksey.
> Best regards,
> Vladimir Ivanov
> On 4/27/15 10:39 AM, Roland Westrelin wrote:
>> That looks good to me.
More information about the hotspot-compiler-dev