target/arm: Convert Neon 3-reg-diff VABAL, VABDL to decodetree

Convert the Neon 3-reg-diff insns VABAL and VABDL to decodetree.
Like almost all the remaining insns in this group, these are
a combination of a two-input operation which returns a double width
result and then a possible accumulation of that double width
result into the destination.

Backports commit f5b28401200ec95ba89552df3ecdcdc342f6b90b from qemu

qemu/target/arm/neon-dp.decode

@@ -442,7 +442,13 @@ Vimm_1r          1111 001 . 1 . 000 ... .... cmode:4 0 . op:1 1 .... @1reg_imm
     VADDHN_3d    1111 001 0 1 . .. .... .... 0100 . 0 . 0 .... @3diff
     VRADDHN_3d   1111 001 1 1 . .. .... .... 0100 . 0 . 0 .... @3diff
 
+    VABAL_S_3d   1111 001 0 1 . .. .... .... 0101 . 0 . 0 .... @3diff
+    VABAL_U_3d   1111 001 1 1 . .. .... .... 0101 . 0 . 0 .... @3diff
+
     VSUBHN_3d    1111 001 0 1 . .. .... .... 0110 . 0 . 0 .... @3diff
     VRSUBHN_3d   1111 001 1 1 . .. .... .... 0110 . 0 . 0 .... @3diff
+
+    VABDL_S_3d   1111 001 0 1 . .. .... .... 0111 . 0 . 0 .... @3diff
+    VABDL_U_3d   1111 001 1 1 . .. .... .... 0111 . 0 . 0 .... @3diff
   ]
 }

qemu/target/arm/translate-neon.inc.c

@@ -2045,3 +2045,136 @@ DO_NARROW_3D(VADDHN, add, narrow, tcg_gen_extrh_i64_i32)
 DO_NARROW_3D(VSUBHN, sub, narrow, tcg_gen_extrh_i64_i32)
 DO_NARROW_3D(VRADDHN, add, narrow_round, gen_narrow_round_high_u32)
 DO_NARROW_3D(VRSUBHN, sub, narrow_round, gen_narrow_round_high_u32)
+
+static bool do_long_3d(DisasContext *s, arg_3diff *a,
+                       NeonGenTwoOpWidenFn *opfn,
+                       NeonGenTwo64OpFn *accfn)
+{
+    /*
+     * 3-regs different lengths, long operations.
+     * These perform an operation on two inputs that returns a double-width
+     * result, and then possibly perform an accumulation operation of
+     * that result into the double-width destination.
+     */
+    TCGContext *tcg_ctx = s->uc->tcg_ctx;
+    TCGv_i64 rd0, rd1, tmp;
+    TCGv_i32 rn, rm;
+
+    if (!arm_dc_feature(s, ARM_FEATURE_NEON)) {
+        return false;
+    }
+
+    /* UNDEF accesses to D16-D31 if they don't exist. */
+    if (!dc_isar_feature(aa32_simd_r32, s) &&
+        ((a->vd | a->vn | a->vm) & 0x10)) {
+        return false;
+    }
+
+    if (!opfn) {
+        /* size == 3 case, which is an entirely different insn group */
+        return false;
+    }
+
+    if (a->vd & 1) {
+        return false;
+    }
+
+    if (!vfp_access_check(s)) {
+        return true;
+    }
+
+    rd0 = tcg_temp_new_i64(tcg_ctx);
+    rd1 = tcg_temp_new_i64(tcg_ctx);
+
+    rn = neon_load_reg(s, a->vn, 0);
+    rm = neon_load_reg(s, a->vm, 0);
+    opfn(tcg_ctx, rd0, rn, rm);
+    tcg_temp_free_i32(tcg_ctx, rn);
+    tcg_temp_free_i32(tcg_ctx, rm);
+
+    rn = neon_load_reg(s, a->vn, 1);
+    rm = neon_load_reg(s, a->vm, 1);
+    opfn(tcg_ctx, rd1, rn, rm);
+    tcg_temp_free_i32(tcg_ctx, rn);
+    tcg_temp_free_i32(tcg_ctx, rm);
+
+    /* Don't store results until after all loads: they might overlap */
+    if (accfn) {
+        tmp = tcg_temp_new_i64(tcg_ctx);
+        neon_load_reg64(s, tmp, a->vd);
+        accfn(tcg_ctx, tmp, tmp, rd0);
+        neon_store_reg64(s, tmp, a->vd);
+        neon_load_reg64(s, tmp, a->vd + 1);
+        accfn(tcg_ctx, tmp, tmp, rd1);
+        neon_store_reg64(s, tmp, a->vd + 1);
+        tcg_temp_free_i64(tcg_ctx, tmp);
+    } else {
+        neon_store_reg64(s, rd0, a->vd);
+        neon_store_reg64(s, rd1, a->vd + 1);
+    }
+
+    tcg_temp_free_i64(tcg_ctx, rd0);
+    tcg_temp_free_i64(tcg_ctx, rd1);
+
+    return true;
+}
+
+static bool trans_VABDL_S_3d(DisasContext *s, arg_3diff *a)
+{
+    static NeonGenTwoOpWidenFn * const opfn[] = {
+        gen_helper_neon_abdl_s16,
+        gen_helper_neon_abdl_s32,
+        gen_helper_neon_abdl_s64,
+        NULL,
+    };
+
+    return do_long_3d(s, a, opfn[a->size], NULL);
+}
+
+static bool trans_VABDL_U_3d(DisasContext *s, arg_3diff *a)
+{
+    static NeonGenTwoOpWidenFn * const opfn[] = {
+        gen_helper_neon_abdl_u16,
+        gen_helper_neon_abdl_u32,
+        gen_helper_neon_abdl_u64,
+        NULL,
+    };
+
+    return do_long_3d(s, a, opfn[a->size], NULL);
+}
+
+static bool trans_VABAL_S_3d(DisasContext *s, arg_3diff *a)
+{
+    static NeonGenTwoOpWidenFn * const opfn[] = {
+        gen_helper_neon_abdl_s16,
+        gen_helper_neon_abdl_s32,
+        gen_helper_neon_abdl_s64,
+        NULL,
+    };
+    static NeonGenTwo64OpFn * const addfn[] = {
+        gen_helper_neon_addl_u16,
+        gen_helper_neon_addl_u32,
+        tcg_gen_add_i64,
+        NULL,
+    };
+
+    return do_long_3d(s, a, opfn[a->size], addfn[a->size]);
+}
+
+static bool trans_VABAL_U_3d(DisasContext *s, arg_3diff *a)
+{
+    static NeonGenTwoOpWidenFn * const opfn[] = {
+        gen_helper_neon_abdl_u16,
+        gen_helper_neon_abdl_u32,
+        gen_helper_neon_abdl_u64,
+        NULL,
+    };
+    static NeonGenTwo64OpFn * const addfn[] = {
+        gen_helper_neon_addl_u16,
+        gen_helper_neon_addl_u32,
+        tcg_gen_add_i64,
+        NULL,
+    };
+
+    return do_long_3d(s, a, opfn[a->size], addfn[a->size]);
+}

qemu/target/arm/translate.c

@@ -5367,9 +5367,9 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
                     {0, 0, 0, 7}, /* VSUBL: handled by decodetree */
                     {0, 0, 0, 7}, /* VSUBW: handled by decodetree */
                     {0, 0, 0, 7}, /* VADDHN: handled by decodetree */
-                    {0, 0, 0, 0}, /* VABAL */
+                    {0, 0, 0, 7}, /* VABAL */
                     {0, 0, 0, 7}, /* VSUBHN: handled by decodetree */
-                    {0, 0, 0, 0}, /* VABDL */
+                    {0, 0, 0, 7}, /* VABDL */
                     {0, 0, 0, 0}, /* VMLAL */
                     {0, 0, 0, 9}, /* VQDMLAL */
                     {0, 0, 0, 0}, /* VMLSL */
@@ -5430,31 +5430,6 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
                         tmp2 = neon_load_reg(s, rm, pass);
                     }
                     switch (op) {
-                    case 5: case 7: /* VABAL, VABDL */
-                        switch ((size << 1) | u) {
-                        case 0:
-                            gen_helper_neon_abdl_s16(tcg_ctx, s->V0, tmp, tmp2);
-                            break;
-                        case 1:
-                            gen_helper_neon_abdl_u16(tcg_ctx, s->V0, tmp, tmp2);
-                            break;
-                        case 2:
-                            gen_helper_neon_abdl_s32(tcg_ctx, s->V0, tmp, tmp2);
-                            break;
-                        case 3:
-                            gen_helper_neon_abdl_u32(tcg_ctx, s->V0, tmp, tmp2);
-                            break;
-                        case 4:
-                            gen_helper_neon_abdl_s64(tcg_ctx, s->V0, tmp, tmp2);
-                            break;
-                        case 5:
-                            gen_helper_neon_abdl_u64(tcg_ctx, s->V0, tmp, tmp2);
-                            break;
-                        default: abort();
-                        }
-                        tcg_temp_free_i32(tcg_ctx, tmp2);
-                        tcg_temp_free_i32(tcg_ctx, tmp);
-                        break;
                     case 8: case 9: case 10: case 11: case 12: case 13:
                         /* VMLAL, VQDMLAL, VMLSL, VQDMLSL, VMULL, VQDMULL */
                         gen_neon_mull(s, s->V0, tmp, tmp2, size, u);
@@ -5473,7 +5448,7 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
                         case 10: /* VMLSL */
                             gen_neon_negl(s, s->V0, size);
                             /* Fall through */
-                        case 5: case 8: /* VABAL, VMLAL */
+                        case 8: /* VABAL, VMLAL */
                             gen_neon_addl(s, size);
                             break;
                         case 9: case 11: /* VQDMLAL, VQDMLSL */

qemu/target/arm/translate.h

@@ -378,6 +378,7 @@ typedef void NeonGenTwo64OpEnvFn(TCGContext *t, TCGv_i64, TCGv_ptr, TCGv_i64, TC
 typedef void NeonGenNarrowFn(TCGContext *t, TCGv_i32, TCGv_i64);
 typedef void NeonGenNarrowEnvFn(TCGContext *t, TCGv_i32, TCGv_ptr, TCGv_i64);
 typedef void NeonGenWidenFn(TCGContext *t, TCGv_i64, TCGv_i32);
+typedef void NeonGenTwoOpWidenFn(TCGContext *t, TCGv_i64, TCGv_i32, TCGv_i32);
 typedef void NeonGenTwoSingleOPFn(TCGContext *t, TCGv_i32, TCGv_i32, TCGv_i32, TCGv_ptr);
 typedef void NeonGenTwoDoubleOPFn(TCGContext *t, TCGv_i64, TCGv_i64, TCGv_i64, TCGv_ptr);
 typedef void NeonGenOneOpFn(TCGContext *t, TCGv_i64, TCGv_i64);