target/arm: Implement MVE VRMLALDAVH, VRMLSLDAVH

Implement the MVE VRMLALDAVH and VRMLSLDAVH insns, which accumulate
the results of a rounded multiply of pairs of elements into a 72-bit
accumulator, returning the top 64 bits in a pair of general purpose
registers.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20210617121628.20116-22-peter.maydell@linaro.org

4 files changed, 76 insertions, 0 deletions

diff --git a/target/arm/helper-mve.h b/target/arm/helper-mve.h
index 7356385d60..f9d4b242be 100644
--- a/target/arm/helper-mve.h
+++ b/target/arm/helper-mve.h
@@ -157,3 +157,11 @@ DEF_HELPER_FLAGS_4(mve_vmlsldavsh, TCG_CALL_NO_WG, i64, env, ptr, ptr, i64)
 DEF_HELPER_FLAGS_4(mve_vmlsldavsw, TCG_CALL_NO_WG, i64, env, ptr, ptr, i64)
 DEF_HELPER_FLAGS_4(mve_vmlsldavxsh, TCG_CALL_NO_WG, i64, env, ptr, ptr, i64)
 DEF_HELPER_FLAGS_4(mve_vmlsldavxsw, TCG_CALL_NO_WG, i64, env, ptr, ptr, i64)
+
+DEF_HELPER_FLAGS_4(mve_vrmlaldavhsw, TCG_CALL_NO_WG, i64, env, ptr, ptr, i64)
+DEF_HELPER_FLAGS_4(mve_vrmlaldavhxsw, TCG_CALL_NO_WG, i64, env, ptr, ptr, i64)
+
+DEF_HELPER_FLAGS_4(mve_vrmlaldavhuw, TCG_CALL_NO_WG, i64, env, ptr, ptr, i64)
+
+DEF_HELPER_FLAGS_4(mve_vrmlsldavhsw, TCG_CALL_NO_WG, i64, env, ptr, ptr, i64)
+DEF_HELPER_FLAGS_4(mve_vrmlsldavhxsw, TCG_CALL_NO_WG, i64, env, ptr, ptr, i64)
diff --git a/target/arm/mve.decode b/target/arm/mve.decode
index 1be2d6b270..ac68f072bb 100644
--- a/target/arm/mve.decode
+++ b/target/arm/mve.decode
@@ -143,7 +143,14 @@ VDUP             1110 1110 1 0 10 ... 0 .... 1011 . 0 0 1 0000 @vdup size=2
 
 @vmlaldav        .... .... . ... ... . ... . .... .... qm:3 . \
                  qn=%qn rdahi=%rdahi rdalo=%rdalo size=%size_16 &vmlaldav
+@vmlaldav_nosz   .... .... . ... ... . ... . .... .... qm:3 . \
+                 qn=%qn rdahi=%rdahi rdalo=%rdalo size=0 &vmlaldav
 VMLALDAV_S       1110 1110 1 ... ... . ... x:1 1110 . 0 a:1 0 ... 0 @vmlaldav
 VMLALDAV_U       1111 1110 1 ... ... . ... x:1 1110 . 0 a:1 0 ... 0 @vmlaldav
 
 VMLSLDAV         1110 1110 1 ... ... . ... x:1 1110 . 0 a:1 0 ... 1 @vmlaldav
+
+VRMLALDAVH_S     1110 1110 1 ... ... 0 ... x:1 1111 . 0 a:1 0 ... 0 @vmlaldav_nosz
+VRMLALDAVH_U     1111 1110 1 ... ... 0 ... x:1 1111 . 0 a:1 0 ... 0 @vmlaldav_nosz
+
+VRMLSLDAVH       1111 1110 1 ... ... 0 ... x:1 1110 . 0 a:1 0 ... 1 @vmlaldav_nosz
diff --git a/target/arm/mve_helper.c b/target/arm/mve_helper.c
index b496e1f4d3..c1427b0f02 100644
--- a/target/arm/mve_helper.c
+++ b/target/arm/mve_helper.c
@@ -18,6 +18,7 @@
  */
 
 #include "qemu/osdep.h"
+#include "qemu/int128.h"
 #include "cpu.h"
 #include "internals.h"
 #include "vec_internal.h"
@@ -527,3 +528,39 @@ DO_LDAV(vmlsldavsh, 2, int16_t, false, +=, -=)
 DO_LDAV(vmlsldavxsh, 2, int16_t, true, +=, -=)
 DO_LDAV(vmlsldavsw, 4, int32_t, false, +=, -=)
 DO_LDAV(vmlsldavxsw, 4, int32_t, true, +=, -=)
+
+/*
+ * Rounding multiply add long dual accumulate high: we must keep
+ * a 72-bit internal accumulator value and return the top 64 bits.
+ */
+#define DO_LDAVH(OP, ESIZE, TYPE, XCHG, EVENACC, ODDACC, TO128)         \
+    uint64_t HELPER(glue(mve_, OP))(CPUARMState *env, void *vn,         \
+                                    void *vm, uint64_t a)               \
+    {                                                                   \
+        uint16_t mask = mve_element_mask(env);                          \
+        unsigned e;                                                     \
+        TYPE *n = vn, *m = vm;                                          \
+        Int128 acc = int128_lshift(TO128(a), 8);                        \
+        for (e = 0; e < 16 / ESIZE; e++, mask >>= ESIZE) {              \
+            if (mask & 1) {                                             \
+                if (e & 1) {                                            \
+                    acc = ODDACC(acc, TO128(n[H##ESIZE(e - 1 * XCHG)] * \
+                                            m[H##ESIZE(e)]));           \
+                } else {                                                \
+                    acc = EVENACC(acc, TO128(n[H##ESIZE(e + 1 * XCHG)] * \
+                                             m[H##ESIZE(e)]));          \
+                }                                                       \
+                acc = int128_add(acc, int128_make64(1 << 7));           \
+            }                                                           \
+        }                                                               \
+        mve_advance_vpt(env);                                           \
+        return int128_getlo(int128_rshift(acc, 8));                     \
+    }
+
+DO_LDAVH(vrmlaldavhsw, 4, int32_t, false, int128_add, int128_add, int128_makes64)
+DO_LDAVH(vrmlaldavhxsw, 4, int32_t, true, int128_add, int128_add, int128_makes64)
+
+DO_LDAVH(vrmlaldavhuw, 4, uint32_t, false, int128_add, int128_add, int128_make64)
+
+DO_LDAVH(vrmlsldavhsw, 4, int32_t, false, int128_add, int128_sub, int128_makes64)
+DO_LDAVH(vrmlsldavhxsw, 4, int32_t, true, int128_add, int128_sub, int128_makes64)
diff --git a/target/arm/translate-mve.c b/target/arm/translate-mve.c
index 77b461c218..a0c4f10a93 100644
--- a/target/arm/translate-mve.c
+++ b/target/arm/translate-mve.c
@@ -472,3 +472,27 @@ static bool trans_VMLSLDAV(DisasContext *s, arg_vmlaldav *a)
     };
     return do_long_dual_acc(s, a, fns[a->size][a->x]);
 }
+
+static bool trans_VRMLALDAVH_S(DisasContext *s, arg_vmlaldav *a)
+{
+    static MVEGenDualAccOpFn * const fns[] = {
+        gen_helper_mve_vrmlaldavhsw, gen_helper_mve_vrmlaldavhxsw,
+    };
+    return do_long_dual_acc(s, a, fns[a->x]);
+}
+
+static bool trans_VRMLALDAVH_U(DisasContext *s, arg_vmlaldav *a)
+{
+    static MVEGenDualAccOpFn * const fns[] = {
+        gen_helper_mve_vrmlaldavhuw, NULL,
+    };
+    return do_long_dual_acc(s, a, fns[a->x]);
+}
+
+static bool trans_VRMLSLDAVH(DisasContext *s, arg_vmlaldav *a)
+{
+    static MVEGenDualAccOpFn * const fns[] = {
+        gen_helper_mve_vrmlsldavhsw, gen_helper_mve_vrmlsldavhxsw,
+    };
+    return do_long_dual_acc(s, a, fns[a->x]);
+}