Another small comment: Kevin made me aware that Page has an is_valid() method that can be used instead of explicit checking Page pointers against NULL. Also, it seems that in most other code, we use Page::FromAddress(NULL) instead of using NULL explicitly in initializations.
Yeah, not using is_valid() is my fault because I was aware of it's existence. I will prepare small a cleanup CL including: - using is_valid() instead of NULL-checks, using Page::FromAddress(NULL) instead of initialization with NULL - using CreateFillerArrayAt instead of current dummy-free-list-node travesty [I was unaware of CreateFillerArrayAt before yesterday] - Win64 build fix [probably I will need to use someone's Windows box to test this build issues if it is possible] -- Vyacheslav Egorov On Fri, Apr 23, 2010 at 8:48 AM, <[email protected]> wrote: > LGTM > > > http://codereview.chromium.org/1700005/diff/6001/7004 > File src/mark-compact.cc (right): > > http://codereview.chromium.org/1700005/diff/6001/7004#newcode1396 > src/mark-compact.cc:1396: ASSERT(((first_empty_page == NULL) && > Now that you have an #ifdef DEBUG here in any case it would be more > readable with ifs and multiple ASSERTS. :) > > Another small comment: Kevin made me aware that Page has an is_valid() > method that can be used instead of explicit checking Page pointers > against NULL. Also, it seems that in most other code, we use > Page::FromAddress(NULL) instead of using NULL explicitly in > initializations. > > > http://codereview.chromium.org/1700005/show > -- v8-dev mailing list [email protected] http://groups.google.com/group/v8-dev
