Weiliang is helping me for upstreaming X32 port. This CL will eliminate all
the
__q annotation from https://codereview.chromium.org/18014003/. I have added
some
explanation in the codes for your review.
LGTM with a nit. Could we change the title to "Use
kRegisterSize/kPCOnStackSize/kFPOnStackSize to make the semantics accurate
for
X64" or a better sentence?
I could help commit this CL after it gets lgtm from reviewers.
https://codereview.chromium.org/20262004/diff/1/src/x64/code-stubs-x64.cc
File src/x64/code-stubs-x64.cc (right):
https://codereview.chromium.org/20262004/diff/1/src/x64/code-stubs-x64.cc#newcode2767
src/x64/code-stubs-x64.cc:2767: __ movq(Operand(rsp,
(argument_slots_on_stack - 1) * kRegisterSize),
X64 and X32 both use register and 8-byte stack slots in their calling
conversion. kRegisterSize makes those codes apply for both.
https://codereview.chromium.org/20262004/diff/1/src/x64/deoptimizer-x64.cc
File src/x64/deoptimizer-x64.cc (right):
https://codereview.chromium.org/20262004/diff/1/src/x64/deoptimizer-x64.cc#newcode410
src/x64/deoptimizer-x64.cc:410: const int kSavedRegistersAreaSize =
kNumberOfRegisters * kRegisterSize +
For X32, we used 8-byte push for bailout id and register saving. Using
kRegister here make those code apply for both.
https://codereview.chromium.org/20262004/diff/1/src/x64/macro-assembler-x64.cc
File src/x64/macro-assembler-x64.cc (right):
https://codereview.chromium.org/20262004/diff/1/src/x64/macro-assembler-x64.cc#newcode3510
src/x64/macro-assembler-x64.cc:3510: arg_stack_space * kRegisterSize;
Explained in the code-stubs-x64.cc for calling conversion.
https://codereview.chromium.org/20262004/
--
--
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.
