LGTM with nits.
I must say I'm surprised that there's not even a single for-loop in any of
the
test cases. Don't most interesting array accesses happen inside loops?
https://chromiumcodereview.appspot.com/11486007/diff/1/src/hydrogen.cc
File src/hydrogen.cc (right):
https://chromiumcodereview.appspot.com/11486007/diff/1/src/hydrogen.cc#newcode3542
src/hydrogen.cc:3542: return add->context();
nit: {}
https://chromiumcodereview.appspot.com/11486007/diff/1/src/hydrogen.cc#newcode3544
src/hydrogen.cc:3544: return
HBinaryOperation::cast(check->index())->context();
nit: {}
https://chromiumcodereview.appspot.com/11486007/diff/1/src/hydrogen.cc#newcode3649
src/hydrogen.cc:3649: !data->CoverCheck(check, offset)) {
nit: please align:
} else if (data->BasicBlock() != bb ||
!data->CoverCheck(check, offset)) {
https://chromiumcodereview.appspot.com/11486007/
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
