Thanks for speedy review! Could please double check the conditions for
clearing
the inline code? I don't want to hold the old global map forever.
-- Vitaly
http://codereview.chromium.org/3537003/diff/1/3
File src/ia32/ic-ia32.cc (right):
http://codereview.chromium.org/3537003/diff/1/3#newcode1671
src/ia32/ic-ia32.cc:1671: // cells. Generated code doesn't have the hole
check.
On 2010/09/29 14:22:07, Søren Gjesse wrote:
Maybe extend the comment here with the actual instruction sequence
generated in
the two cases.
Done.
http://codereview.chromium.org/3537003/diff/1/7
File test/cctest/test-api.cc (right):
http://codereview.chromium.org/3537003/diff/1/7#newcode11461
test/cctest/test-api.cc:11461: ExpectString("readCell()", "first");
On 2010/09/29 14:22:07, Søren Gjesse wrote:
Please extend the test with actually deleting the property using the
ForceDelete
API.
Maybe also add a test where the don't delete property is added through
the API
using the v8::DontDelete property attribute.
Both done.
http://codereview.chromium.org/3537003/show
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
