OK, not much to say about all the other files.
http://codereview.chromium.org/13127/diff/209/17 File src/codegen-ia32.cc (right): http://codereview.chromium.org/13127/diff/209/17#newcode250 Line 250: CheckStack(); Were the only changes necessary to make CheckStack work without a SpillAll the implementation of real merge on the jump targets? There was already a SpillAll on the slow case of CheckStack? And will a return from the debugger put all the values that should be in registers, in registers? That is pretty amazing. http://codereview.chromium.org/13127 --~--~---------~--~----~------------~-------~--~----~ v8-dev mailing list [email protected] http://groups.google.com/group/v8-dev -~----------~----~----~----~------~----~------~--~---
