LGTM, however I am wondering if it wouldn't be better to collect what we
need
instead of the deltas where we generate to little?
AddToMinRelocSizeForDeoptimization
min_reloc_size_for_deoptimization_
The call AddToMinRelocSizeForDeoptimization at all places where deopt will
have
to write new reloc info.
That will make it more clear what is happening, and we don't need to do
padding
if other reloc info provides the space we need as we end up wiht a min size
for
the reloc info required for deopt.
http://codereview.chromium.org/6541053/diff/2002/src/assembler.h
File src/assembler.h (right):
http://codereview.chromium.org/6541053/diff/2002/src/assembler.h#newcode187
src/assembler.h:187: // The size of a comment is equal to tree bytes for
the extra tagged pc +
tree -> three
http://codereview.chromium.org/6541053/diff/2002/src/ia32/lithium-codegen-ia32.cc
File src/ia32/lithium-codegen-ia32.cc (right):
http://codereview.chromium.org/6541053/diff/2002/src/ia32/lithium-codegen-ia32.cc#newcode396
src/ia32/lithium-codegen-ia32.cc:396:
RecordPosition(pointers->position());
Move this comment and code to after the '__ call'?
http://codereview.chromium.org/6541053/diff/2002/src/ia32/lithium-codegen-ia32.h
File src/ia32/lithium-codegen-ia32.h (right):
http://codereview.chromium.org/6541053/diff/2002/src/ia32/lithium-codegen-ia32.h#newcode106
src/ia32/lithium-codegen-ia32.h:106: void AddRelocPadding(int count) {
reloc_padding_count_ += count; }
AddRelocPadding -> AddRelocPaddingForDeoptimization
http://codereview.chromium.org/6541053/diff/2002/src/ia32/lithium-codegen-ia32.h#newcode259
src/ia32/lithium-codegen-ia32.h:259: int reloc_padding_count_;
reloc_padding_count_ -> reloc_padding_count_for_deoptimization_
http://codereview.chromium.org/6541053/
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
