Fixed constant folding in HMod.

Fixed constant folding in HMod.

We have to check for overflow before attempting to do a modulo operation,
otherwise Crankshaft itself segfaults on some platforms, e.g. ia32. Added tests
even for division, where the problem doesn't show up, just to be sure...

R=mvstanton@chromium.org

Committed: http://code.google.com/p/v8/source/detail?r=14629

[mvstanton, 2013-05-13 07:10:08]

lgtm, but one question in the test.

https://codereview.chromium.org/14617014/diff/2001/test/mjsunit/constant-fold...
File test/mjsunit/constant-folding-2.js (right):

https://codereview.chromium.org/14617014/diff/2001/test/mjsunit/constant-fold...
test/mjsunit/constant-folding-2.js:266: assertEquals(2147483648, -2147483648 *
(-1));
Why is it * instead of /?

[Sven Panne, 2013-05-13 07:22:34]

https://codereview.chromium.org/14617014/diff/2001/test/mjsunit/constant-fold...
File test/mjsunit/constant-folding-2.js (right):

https://codereview.chromium.org/14617014/diff/2001/test/mjsunit/constant-fold...
test/mjsunit/constant-folding-2.js:266: assertEquals(2147483648, -2147483648 *
(-1));
On 2013/05/13 07:10:08, mvstanton wrote:
> Why is it * instead of /?

Ooops, it's just a typo which survived after tons of experiments. I'll change
that to match the function name. :-}

[Sven Panne, 2013-05-13 07:32:44]

Committed patchset #3 manually as r14629 (presubmit successful).