softfloat: add APIs to handle alternative sNaN propagation for fmax/fmin

For "fmax/fmin ft0, ft1, ft2" and if one of the inputs is sNaN,

  The original logic:
    Return NaN and set invalid flag if ft1 == sNaN || ft2 == sNan.

  The alternative path:
    Set invalid flag if ft1 == sNaN || ft2 == sNaN.
    Return NaN only if ft1 == NaN && ft2 == NaN.

The IEEE 754 spec allows both implementation and some architecture such
as riscv choose different defintions in two spec versions.
(riscv-spec-v2.2 use original version, riscv-spec-20191213 changes to
 alternative)

Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
Signed-off-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20211021160847.2748577-2-frank.chang@sifive.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

1 files changed, 23 insertions, 2 deletions

diff --git a/fpu/softfloat-parts.c.inc b/fpu/softfloat-parts.c.inc
index dddee92d6e..41d4b17e41 100644
--- a/fpu/softfloat-parts.c.inc
+++ b/fpu/softfloat-parts.c.inc
@@ -1219,14 +1219,35 @@ static FloatPartsN *partsN(minmax)(FloatPartsN *a, FloatPartsN *b,
 
     if (unlikely(ab_mask & float_cmask_anynan)) {
         /*
-         * For minnum/maxnum, if one operand is a QNaN, and the other
+         * For minNum/maxNum (IEEE 754-2008)
+         * or minimumNumber/maximumNumber (IEEE 754-2019),
+         * if one operand is a QNaN, and the other
          * operand is numerical, then return numerical argument.
          */
-        if ((flags & minmax_isnum)
+        if ((flags & (minmax_isnum | minmax_isnumber))
             && !(ab_mask & float_cmask_snan)
             && (ab_mask & ~float_cmask_qnan)) {
             return is_nan(a->cls) ? b : a;
         }
+
+        /*
+         * In IEEE 754-2019, minNum, maxNum, minNumMag and maxNumMag
+         * are removed and replaced with minimum, minimumNumber, maximum
+         * and maximumNumber.
+         * minimumNumber/maximumNumber behavior for SNaN is changed to:
+         *   If both operands are NaNs, a QNaN is returned.
+         *   If either operand is a SNaN,
+         *   an invalid operation exception is signaled,
+         *   but unless both operands are NaNs,
+         *   the SNaN is otherwise ignored and not converted to a QNaN.
+         */
+        if ((flags & minmax_isnumber)
+            && (ab_mask & float_cmask_snan)
+            && (ab_mask & ~float_cmask_anynan)) {
+            float_raise(float_flag_invalid, s);
+            return is_nan(a->cls) ? b : a;
+        }
+
         return parts_pick_nan(a, b, s);
     }