Reviewers: Toon Verwaest,
Message:
should not change perf
http://108.170.219.0:8080/Comparison#targetA%3Dv8%3BmachineTypeA%3Dlinux-ia32%3BrevisionA%3D14777%3BpatchA%3Dolivf-fix_compar_nil_ic%3BtargetB%3Dv8%3BmachineTypeB%3Dlinux-ia32%3BrevisionB%3D14777%3BpatchB%3DNone
Description:
fix the compare nil ic
The compare nil ic introduces a mechanism called lightweight miss. The
idea is to insert a direct call to the miss handler instead of going
through the deopt routine. This can be use for uninitialized stubs, to
directly jump to the runtime.
But apparently the lightweight miss code is generated, when the stub is
marked a
miss.
This is exactly the opposite of what should be happening, since a miss
means,
we now have type info and can insert a real stub.
But this does not trigger a bug, since the compare nil stub is the only
code using the mechanism and it was consistetly marking miss situations
as non-miss, and non-miss situations as miss.
BUG=
Please review this at https://chromiumcodereview.appspot.com/15806005/
SVN Base: https://v8.googlecode.com/svn/branches/bleeding_edge
Affected files:
M src/arm/code-stubs-arm.cc
M src/code-stubs-hydrogen.cc
M src/code-stubs.h
M src/code-stubs.cc
M src/ia32/code-stubs-ia32.cc
M src/ic.cc
M src/mips/code-stubs-mips.cc
M src/x64/code-stubs-x64.cc
M test/cctest/test-compare-nil-ic-stub.cc
--
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
---
You received this message because you are subscribed to the Google Groups "v8-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.
