Status: New
Owner: ----
New issue 3151 by [email protected]: (MinInt / 2^n) | 0 gives
result with wrong sign
http://code.google.com/p/v8/issues/detail?id=3151
Simple test case:
===
function div_by_two(x) {
return (x / 2) | 0;
}
div_by_two(42);
%OptimizeFunctionOnNextCall(div_by_two);
div_by_two(42);
assertEquals(((-2147483648) / 2) | 0, div_by_two(-2147483648));
===
This gives +1073741824 on at least arm and ia32, when it should be negative.
I think it's caused by lithium codegen negating -2^31, resulting in -2^31,
then using arithmetic shift right to do the power-of-two division. This
result is negated again to give the positive, wrong answer.
--
You received this message because this project is configured to send all
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings
--
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
---
You received this message because you are subscribed to the Google Groups "v8-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.
