add sv.madded sv.subfe to divmnu64.c

diff --git a/openpower/sv/biginteger/divmnu64.c b/openpower/sv/biginteger/divmnu64.c
index 93f106b30355c896d8eb2765e0231dd85aaeec32..eb157139d6c6907c384f150ef5bde81e340f0efd 100644 (file)
--- a/openpower/sv/biginteger/divmnu64.c
+++ b/openpower/sv/biginteger/divmnu64.c
@@ -135,7 +135,8 @@ int divmnu(unsigned q[], unsigned r[], const unsigned u[], const unsigned v[],
             if (rhat < b)
                 goto again;
         }
-#define MUL_RSUB_CARRY_2_STAGE2
+// don't define here, allowing easily testing all options by passing -D...
+// #define MUL_RSUB_CARRY_2_STAGE2
 #ifdef ORIGINAL
         // Multiply and subtract.
         k = 0;
@@ -288,6 +289,32 @@ int divmnu(unsigned q[], unsigned r[], const unsigned u[], const unsigned v[],
             un[i + j] = (uint32_t)result;
         }
         bool need_fixup = carry != 0;
+#elif defined(MADDED_SUBFE)
+        (void)p; // shut up unused variable warning
+
+        // Multiply and subtract.
+        uint32_t carry = 0;
+        uint32_t product[n + 1];
+        // VL = n + 1
+        // sv.madded product.v, vn.v, qhat.s, carry.s
+        for (int i = 0; i <= n; i++)
+        {
+            uint32_t vn_v = i < n ? vn[i] : 0;
+            uint64_t value = (uint64_t)vn_v * (uint64_t)qhat + carry;
+            carry = (uint32_t)(value >> 32);
+            product[i] = (uint32_t)value;
+        }
+        bool ca = true;
+        uint32_t *un_j = &un[j];
+        // VL = n + 1
+        // sv.subfe un_j.v, product.v, un_j.v
+        for (int i = 0; i <= n; i++)
+        {
+            uint64_t value = (uint64_t)~product[i] + (uint64_t)un_j[i] + ca;
+            ca = value >> 32 != 0;
+            un_j[i] = value;
+        }
+        bool need_fixup = !ca;
 #else
 #error need to choose one of the algorithm options; e.g. -DORIGINAL
 #endif