Richard Biener writes:
> On Thu, Dec 11, 2014 at 1:26 PM, Richard Sandiford
> wrote:
>> As pointed out in PR 64182, wide-int rounded division gets the
>> ties-away-from-zero case wrong for odd-numbered dividends, while
>> double_int gets the unsigned case wrong by unconditionally treating
>> a di
On Thu, Dec 11, 2014 at 1:26 PM, Richard Sandiford
wrote:
> As pointed out in PR 64182, wide-int rounded division gets the
> ties-away-from-zero case wrong for odd-numbered dividends, while
> double_int gets the unsigned case wrong by unconditionally treating
> a dividend or remainder with the top
As pointed out in PR 64182, wide-int rounded division gets the
ties-away-from-zero case wrong for odd-numbered dividends, while
double_int gets the unsigned case wrong by unconditionally treating
a dividend or remainder with the top bit set as negative. As Jakub
says, the test used in double_int m
