On 5/18/21 6:20 AM, Peter Maydell wrote:
This code change looks OK as far as it goes, but there are a bunch
of other places in softfloat.c which switch on the float rounding mode.
If this rounding mode is only supported for a particular subset of
operations we should at least document that in the
On Sat, 17 Apr 2021 at 01:02, Richard Henderson
wrote:
>
> For Arm BFDOT and BFMMLA, we need a version of round-to-odd
> that overflows to infinity, instead of the max normal number.
>
> Signed-off-by: Richard Henderson
> ---
> include/fpu/softfloat-types.h | 4 +++-
> fpu/softfloat.c
For Arm BFDOT and BFMMLA, we need a version of round-to-odd
that overflows to infinity, instead of the max normal number.
Signed-off-by: Richard Henderson
---
include/fpu/softfloat-types.h | 4 +++-
fpu/softfloat.c | 8 ++--
2 files changed, 9 insertions(+), 3 deletions(-)
dif
