Re: [Qemu-devel] [PATCH v1 2/2] fpu/softfloat: raise float_invalid for NaN in float_to_int

["Alex Bennée" <alex.bennee@linaro.org>]

Peter Maydell <peter.maydell@linaro.org> writes:

> On 12 April 2018 at 12:58, Alex Bennée <alex.bennee@linaro.org> wrote:
>> Fixes https://bugs.launchpad.net/qemu/+bug/1759264
>>
>> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
>> Cc: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
>> ---
>>  fpu/softfloat.c | 1 +
>>  1 file changed, 1 insertion(+)
>>
>> diff --git a/fpu/softfloat.c b/fpu/softfloat.c
>> index 9b99aa6ec8..ddc77c273c 100644
>> --- a/fpu/softfloat.c
>> +++ b/fpu/softfloat.c
>> @@ -1344,6 +1344,7 @@ static int64_t round_to_int_and_pack(FloatParts in, int rmode,
>>      case float_class_qnan:
>>      case float_class_dnan:
>>      case float_class_msnan:
>> +        s->float_exception_flags = orig_flags | float_flag_invalid;
>>          return max;
>>      case float_class_inf:
>>          return p.sign ? min : max;
>
> Don't we also need to raise the Invalid flag for float_class_inf ?
>
> In both cases, this is fixing a regression introduced in
> commit ab52f973a50, I think.

And I guess the round_to_uint_and_pack as well which doesn't handle the
input inf case.

>
> thanks
> -- PMM


--
Alex Bennée