On 2012/02/28 14:31:38, kalmard wrote:
rebased on r10858.
The existing check is not for invalid, it is for whatever value undefined
produces, which is NaN. Apparently casting NaN to int differs on MIPS, ARM and
Intel.

I think that the need for this CL disappears after r10867, which changes the
test to explicitly test for the casted NaN, but please verify that it works on
MIPS.

http://codereview.chromium.org/9447015/

--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev

Reply via email to