All the current lowerings produce their result using a floating-point
multiply or add, so denorms should already be flushed (e.g.
nir_op_frcp), or they never produce a denorm (e.g. nir_op_ftrunc), so
I don't think this is necessary.
On Wed, Dec 5, 2018 at 4:56 PM Samuel Iglesias Gonsálvez
<sigles...@igalia.com> wrote:
>
> Signed-off-by: Samuel Iglesias Gonsálvez <sigles...@igalia.com>
> ---
> src/compiler/nir/nir_lower_double_ops.c | 12 ++++++++++++
> 1 file changed, 12 insertions(+)
>
> diff --git a/src/compiler/nir/nir_lower_double_ops.c
> b/src/compiler/nir/nir_lower_double_ops.c
> index b3543bc6963..97b825d2fdb 100644
> --- a/src/compiler/nir/nir_lower_double_ops.c
> +++ b/src/compiler/nir/nir_lower_double_ops.c
> @@ -558,6 +558,18 @@ lower_doubles_instr(nir_alu_instr *instr,
> nir_lower_doubles_options options)
> unreachable("unhandled opcode");
> }
>
> + bool denorm_flush_to_zero =
> + bld.shader->info.shader_float_controls_execution_mode &
> SHADER_DENORM_FLUSH_TO_ZERO_FP64;
> + if (denorm_flush_to_zero) {
> + /* TODO: add support for flushing negative denorms to -0.0 */
> + /* Flush to zero if the result value is a denorm */
> + result = nir_bcsel(&bld,
> + nir_flt(&bld, nir_fabs(&bld, result),
> + nir_imm_double(&bld,
> 2.22507385850720138309023271733e-308)),
> + nir_imm_double(&bld, 0.0),
> + result);
> + }
> +
> nir_ssa_def_rewrite_uses(&instr->dest.dest.ssa, nir_src_for_ssa(result));
> nir_instr_remove(&instr->instr);
> return true;
> --
> 2.19.1
>
> _______________________________________________
> mesa-dev mailing list
> mesa-dev@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/mesa-dev
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
