Description[turbofan] greedy: heuristic for memory operands
When we have a memory operand (HasSpillOperand() == true), and it
doesn't need a register immediately, split in an optimal position, which
is outside the outermost possible loop - just like Linear does.
This results in some modest improvements in perf, when compared
to baseline greedy. In particular Jetstream zlib x64: 4.66%, Life
(Emscripten x64) 11%; largest regression is in AreWeFastYet x64: 8%
and Corrections (Emsccripten x32) 10%
BUG=
Committed: https://crrev.com/8937bfc1d165ff6d72dede1b0ce6f7c1ab9fb260
Cr-Commit-Position: refs/heads/master@{#30498}
Patch Set 1 : #Patch Set 2 : #
Messages
Total messages: 9 (3 generated)
|