Reviewers: Jakob,
Message:
PTAL.
Guidance:
- I removed most of the flag gates around smi-only array code, only leaving
them
around the central bottleneck that makes Arrays and Objects have
FAST_SMI_ONLY_ELEMENTS.
- I added a special map tweak in the ArrayLiteral boilerplate code that is
tuned
specifically to activate FAST_SMI_ONLY_ELEMENTS on the cases that are in the
Kraken benchmark.
- A implemented a few micro optimizations for regressions that showed up in
other benchmarks
Description:
Activate smi-only optimizations for large array literals.
BUG=none
TEST=none
Please review this at http://codereview.chromium.org/8177005/
SVN Base: https://v8.googlecode.com/svn/branches/bleeding_edge
Affected files:
M src/arm/full-codegen-arm.cc
M src/arm/ic-arm.cc
M src/arm/stub-cache-arm.cc
M src/builtins.cc
M src/hydrogen.cc
M src/ia32/full-codegen-ia32.cc
M src/ia32/ic-ia32.cc
M src/ia32/stub-cache-ia32.cc
M src/objects-inl.h
M src/objects.cc
M src/runtime.cc
M src/x64/full-codegen-x64.cc
M src/x64/ic-x64.cc
M src/x64/stub-cache-x64.cc
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
