target/arm: Implement MVE VADC, VSBC

Implement the MVE VADC and VSBC insns.  These perform an
add-with-carry or subtract-with-carry of the 32-bit elements in each
lane of the input vectors, where the carry-out of each add is the
carry-in of the next.  The initial carry input is either 1 or is from
FPSCR.C; the carry out at the end is written back to FPSCR.C.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20210617121628.20116-41-peter.maydell@linaro.org

target/arm/helper-mve.h

@@ -246,6 +246,11 @@ DEF_HELPER_FLAGS_4(mve_vrhaddub, TCG_CALL_NO_WG, void, env, ptr, ptr, ptr)
 DEF_HELPER_FLAGS_4(mve_vrhadduh, TCG_CALL_NO_WG, void, env, ptr, ptr, ptr)
 DEF_HELPER_FLAGS_4(mve_vrhadduw, TCG_CALL_NO_WG, void, env, ptr, ptr, ptr)
 
+DEF_HELPER_FLAGS_4(mve_vadc, TCG_CALL_NO_WG, void, env, ptr, ptr, ptr)
+DEF_HELPER_FLAGS_4(mve_vadci, TCG_CALL_NO_WG, void, env, ptr, ptr, ptr)
+DEF_HELPER_FLAGS_4(mve_vsbc, TCG_CALL_NO_WG, void, env, ptr, ptr, ptr)
+DEF_HELPER_FLAGS_4(mve_vsbci, TCG_CALL_NO_WG, void, env, ptr, ptr, ptr)
+
 DEF_HELPER_FLAGS_4(mve_vadd_scalarb, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
 DEF_HELPER_FLAGS_4(mve_vadd_scalarh, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)
 DEF_HELPER_FLAGS_4(mve_vadd_scalarw, TCG_CALL_NO_WG, void, env, ptr, ptr, i32)

target/arm/mve.decode

@@ -160,6 +160,11 @@ VQDMULLT         111 . 1110 0 . 11 ... 0 ... 1 1111 . 0 . 0 ... 1 @2op_sz28
 VRHADD_S         111 0 1111 0 . .. ... 0 ... 0 0001 . 1 . 0 ... 0 @2op
 VRHADD_U         111 1 1111 0 . .. ... 0 ... 0 0001 . 1 . 0 ... 0 @2op
 
+VADC             1110 1110 0 . 11 ... 0 ... 0 1111 . 0 . 0 ... 0 @2op_nosz
+VSBC             1111 1110 0 . 11 ... 0 ... 0 1111 . 0 . 0 ... 0 @2op_nosz
+VADCI            1110 1110 0 . 11 ... 0 ... 1 1111 . 0 . 0 ... 0 @2op_nosz
+VSBCI            1111 1110 0 . 11 ... 0 ... 1 1111 . 0 . 0 ... 0 @2op_nosz
+
 # Vector miscellaneous
 
 VCLS             1111 1111 1 . 11 .. 00 ... 0 0100 01 . 0 ... 0 @1op

target/arm/mve_helper.c

@@ -537,6 +537,58 @@ DO_2OP_U(vrshlu, DO_VRSHLU)
 DO_2OP_S(vrhadds, DO_RHADD_S)
 DO_2OP_U(vrhaddu, DO_RHADD_U)
 
+static void do_vadc(CPUARMState *env, uint32_t *d, uint32_t *n, uint32_t *m,
+                    uint32_t inv, uint32_t carry_in, bool update_flags)
+{
+    uint16_t mask = mve_element_mask(env);
+    unsigned e;
+
+    /* If any additions trigger, we will update flags. */
+    if (mask & 0x1111) {
+        update_flags = true;
+    }
+
+    for (e = 0; e < 16 / 4; e++, mask >>= 4) {
+        uint64_t r = carry_in;
+        r += n[H4(e)];
+        r += m[H4(e)] ^ inv;
+        if (mask & 1) {
+            carry_in = r >> 32;
+        }
+        mergemask(&d[H4(e)], r, mask);
+    }
+
+    if (update_flags) {
+        /* Store C, clear NZV. */
+        env->vfp.xregs[ARM_VFP_FPSCR] &= ~FPCR_NZCV_MASK;
+        env->vfp.xregs[ARM_VFP_FPSCR] |= carry_in * FPCR_C;
+    }
+    mve_advance_vpt(env);
+}
+
+void HELPER(mve_vadc)(CPUARMState *env, void *vd, void *vn, void *vm)
+{
+    bool carry_in = env->vfp.xregs[ARM_VFP_FPSCR] & FPCR_C;
+    do_vadc(env, vd, vn, vm, 0, carry_in, false);
+}
+
+void HELPER(mve_vsbc)(CPUARMState *env, void *vd, void *vn, void *vm)
+{
+    bool carry_in = env->vfp.xregs[ARM_VFP_FPSCR] & FPCR_C;
+    do_vadc(env, vd, vn, vm, -1, carry_in, false);
+}
+
+
+void HELPER(mve_vadci)(CPUARMState *env, void *vd, void *vn, void *vm)
+{
+    do_vadc(env, vd, vn, vm, 0, 0, true);
+}
+
+void HELPER(mve_vsbci)(CPUARMState *env, void *vd, void *vn, void *vm)
+{
+    do_vadc(env, vd, vn, vm, -1, 1, true);
+}
+
 static inline int32_t do_sat_bhw(int64_t val, int64_t min, int64_t max, bool *s)
 {
     if (val > max) {

target/arm/translate-mve.c

@@ -451,6 +451,43 @@ static bool trans_VQDMULLT(DisasContext *s, arg_2op *a)
     return do_2op(s, a, fns[a->size]);
 }
 
+/*
+ * VADC and VSBC: these perform an add-with-carry or subtract-with-carry
+ * of the 32-bit elements in each lane of the input vectors, where the
+ * carry-out of each add is the carry-in of the next.  The initial carry
+ * input is either fixed (0 for VADCI, 1 for VSBCI) or is from FPSCR.C
+ * (for VADC and VSBC); the carry out at the end is written back to FPSCR.C.
+ * These insns are subject to beat-wise execution.  Partial execution
+ * of an I=1 (initial carry input fixed) insn which does not
+ * execute the first beat must start with the current FPSCR.NZCV
+ * value, not the fixed constant input.
+ */
+static bool trans_VADC(DisasContext *s, arg_2op *a)
+{
+    return do_2op(s, a, gen_helper_mve_vadc);
+}
+
+static bool trans_VADCI(DisasContext *s, arg_2op *a)
+{
+    if (mve_skip_first_beat(s)) {
+        return trans_VADC(s, a);
+    }
+    return do_2op(s, a, gen_helper_mve_vadci);
+}
+
+static bool trans_VSBC(DisasContext *s, arg_2op *a)
+{
+    return do_2op(s, a, gen_helper_mve_vsbc);
+}
+
+static bool trans_VSBCI(DisasContext *s, arg_2op *a)
+{
+    if (mve_skip_first_beat(s)) {
+        return trans_VSBC(s, a);
+    }
+    return do_2op(s, a, gen_helper_mve_vsbci);
+}
+
 static bool do_2op_scalar(DisasContext *s, arg_2scalar *a,
                           MVEGenTwoOpScalarFn fn)
 {