LGTM. One question though.
https://codereview.chromium.org/11818025/diff/2001/src/isolate.cc File src/isolate.cc (right): https://codereview.chromium.org/11818025/diff/2001/src/isolate.cc#newcode543 src/isolate.cc:543: // TODO(dcarney): convert buffer to utf8? I don't think correct utf8 output is a priority here. What we foremost want to do here is to spill the java script stack trace string onto the stack so that it makes into the minidump on crash. https://codereview.chromium.org/11818025/diff/2001/src/objects-inl.h File src/objects-inl.h (right): https://codereview.chromium.org/11818025/diff/2001/src/objects-inl.h#newcode2731 src/objects-inl.h:2731: return reinterpret_cast<const uint8_t*>(resource()->data()); I guess this cast would be unnecessary once we change the API for external string to be latin1? https://codereview.chromium.org/11818025/diff/2001/src/objects.h File src/objects.h (right): https://codereview.chromium.org/11818025/diff/2001/src/objects.h#newcode7387 src/objects.h:7387: NonAsciiStart(reinterpret_cast<const char*>(chars), length) >= length; Could we also change NonAsciiStart to take const uint8_t* as first parameter? https://codereview.chromium.org/11818025/diff/2001/src/string-search.h File src/string-search.h (right): https://codereview.chromium.org/11818025/diff/2001/src/string-search.h#newcode158 src/string-search.h:158: return false; Don't we still need to gate this with ENABLE_LATIN_1? https://codereview.chromium.org/11818025/ -- v8-dev mailing list [email protected] http://groups.google.com/group/v8-dev
