target/arm: Convert Neon 3-reg-diff narrowing ops to decodetree

Convert the narrow-to-high-half insns VADDHN, VSUBHN, VRADDHN,
VRSUBHN in the Neon 3-registers-different-lengths group to
decodetree.

Backports commit 0fa1ab0302badabc3581aefcbb2f189ef52c4985 from qemu

qemu/target/arm/neon-dp.decode

@@ -438,5 +438,11 @@ Vimm_1r          1111 001 . 1 . 000 ... .... cmode:4 0 . op:1 1 .... @1reg_imm
 
     VSUBW_S_3d   1111 001 0 1 . .. .... .... 0011 . 0 . 0 .... @3diff
     VSUBW_U_3d   1111 001 1 1 . .. .... .... 0011 . 0 . 0 .... @3diff
+
+    VADDHN_3d    1111 001 0 1 . .. .... .... 0100 . 0 . 0 .... @3diff
+    VRADDHN_3d   1111 001 1 1 . .. .... .... 0100 . 0 . 0 .... @3diff
+
+    VSUBHN_3d    1111 001 0 1 . .. .... .... 0110 . 0 . 0 .... @3diff
+    VRSUBHN_3d   1111 001 1 1 . .. .... .... 0110 . 0 . 0 .... @3diff
   ]
 }

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

@@ -1859,6 +1859,7 @@ static bool do_prewiden_3d(DisasContext *s, arg_3diff *a,
                            bool src1_wide)
 {
     /* 3-regs different lengths, prewidening case (VADDL/VSUBL/VAADW/VSUBW) */
+    TCGContext *tcg_ctx = s->uc->tcg_ctx;
     TCGv_i64 rn0_64, rn1_64, rm_64;
     TCGv_i32 rm;
 
@@ -1956,3 +1957,91 @@ DO_PREWIDEN(VADDW_S, s, ext, add, true)
 DO_PREWIDEN(VADDW_U, u, extu, add, true)
 DO_PREWIDEN(VSUBW_S, s, ext, sub, true)
 DO_PREWIDEN(VSUBW_U, u, extu, sub, true)
+
+static bool do_narrow_3d(DisasContext *s, arg_3diff *a,
+                         NeonGenTwo64OpFn *opfn, NeonGenNarrowFn *narrowfn)
+{
+    /* 3-regs different lengths, narrowing (VADDHN/VSUBHN/VRADDHN/VRSUBHN) */
+    TCGContext *tcg_ctx = s->uc->tcg_ctx;
+    TCGv_i64 rn_64, rm_64;
+    TCGv_i32 rd0, rd1;
+
+    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 || !narrowfn) {
+        /* size == 3 case, which is an entirely different insn group */
+        return false;
+    }
+
+    if ((a->vn | a->vm) & 1) {
+        return false;
+    }
+
+    if (!vfp_access_check(s)) {
+        return true;
+    }
+
+    rn_64 = tcg_temp_new_i64(tcg_ctx);
+    rm_64 = tcg_temp_new_i64(tcg_ctx);
+    rd0 = tcg_temp_new_i32(tcg_ctx);
+    rd1 = tcg_temp_new_i32(tcg_ctx);
+
+    neon_load_reg64(s, rn_64, a->vn);
+    neon_load_reg64(s, rm_64, a->vm);
+
+    opfn(tcg_ctx, rn_64, rn_64, rm_64);
+
+    narrowfn(tcg_ctx, rd0, rn_64);
+
+    neon_load_reg64(s, rn_64, a->vn + 1);
+    neon_load_reg64(s, rm_64, a->vm + 1);
+
+    opfn(tcg_ctx, rn_64, rn_64, rm_64);
+
+    narrowfn(tcg_ctx, rd1, rn_64);
+
+    neon_store_reg(s, a->vd, 0, rd0);
+    neon_store_reg(s, a->vd, 1, rd1);
+
+    tcg_temp_free_i64(tcg_ctx, rn_64);
+    tcg_temp_free_i64(tcg_ctx, rm_64);
+
+    return true;
+}
+
+#define DO_NARROW_3D(INSN, OP, NARROWTYPE, EXTOP)                       \
+    static bool trans_##INSN##_3d(DisasContext *s, arg_3diff *a)        \
+    {                                                                   \
+        static NeonGenTwo64OpFn * const addfn[] = {                     \
+            gen_helper_neon_##OP##l_u16,                                \
+            gen_helper_neon_##OP##l_u32,                                \
+            tcg_gen_##OP##_i64,                                         \
+            NULL,                                                       \
+        };                                                              \
+        static NeonGenNarrowFn * const narrowfn[] = {                   \
+            gen_helper_neon_##NARROWTYPE##_high_u8,                     \
+            gen_helper_neon_##NARROWTYPE##_high_u16,                    \
+            EXTOP,                                                      \
+            NULL,                                                       \
+        };                                                              \
+        return do_narrow_3d(s, a, addfn[a->size], narrowfn[a->size]);   \
+    }
+
+static void gen_narrow_round_high_u32(TCGContext *s, TCGv_i32 rd, TCGv_i64 rn)
+{
+    tcg_gen_addi_i64(s, rn, rn, 1u << 31);
+    tcg_gen_extrh_i64_i32(s, rd, rn);
+}
+
+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)

qemu/target/arm/translate.c

@@ -3353,17 +3353,6 @@ static inline void gen_neon_addl(DisasContext *s, int size)
     }
 }
 
-static inline void gen_neon_subl(DisasContext *s, int size)
-{
-    TCGContext *tcg_ctx = s->uc->tcg_ctx;
-    switch (size) {
-    case 0: gen_helper_neon_subl_u16(tcg_ctx, CPU_V001); break;
-    case 1: gen_helper_neon_subl_u32(tcg_ctx, CPU_V001); break;
-    case 2: tcg_gen_sub_i64(tcg_ctx, CPU_V001); break;
-    default: abort();
-    }
-}
-
 static inline void gen_neon_negl(DisasContext *s, TCGv_i64 var, int size)
 {
     TCGContext *tcg_ctx = s->uc->tcg_ctx;
@@ -5364,8 +5353,6 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
             op = (insn >> 8) & 0xf;
             if ((insn & (1 << 6)) == 0) {
                 /* Three registers of different lengths.  */
-                int src1_wide;
-                int src2_wide;
                 /* undefreq: bit 0 : UNDEF if size == 0
                  *           bit 1 : UNDEF if size == 1
                  *           bit 2 : UNDEF if size == 2
@@ -5379,9 +5366,9 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
                     {0, 0, 0, 7}, /* VADDW: handled by decodetree */
                     {0, 0, 0, 7}, /* VSUBL: handled by decodetree */
                     {0, 0, 0, 7}, /* VSUBW: handled by decodetree */
-                    {0, 1, 1, 0}, /* VADDHN */
+                    {0, 0, 0, 7}, /* VADDHN: handled by decodetree */
                     {0, 0, 0, 0}, /* VABAL */
-                    {0, 1, 1, 0}, /* VSUBHN */
+                    {0, 0, 0, 7}, /* VSUBHN: handled by decodetree */
                     {0, 0, 0, 0}, /* VABDL */
                     {0, 0, 0, 0}, /* VMLAL */
                     {0, 0, 0, 9}, /* VQDMLAL */
@@ -5393,17 +5380,13 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
                     {0, 0, 0, 7}, /* Reserved: always UNDEF */
                 };
 
-                src1_wide = neon_3reg_wide[op][1];
-                src2_wide = neon_3reg_wide[op][2];
                 undefreq = neon_3reg_wide[op][3];
 
                 if ((undefreq & (1 << size)) ||
                     ((undefreq & 8) && u)) {
                     return 1;
                 }
-                if ((src1_wide && (rn & 1)) ||
+                if (rd & 1) {
-                    (src2_wide && (rm & 1)) ||
-                    (!src2_wide && (rd & 1))) {
                     return 1;
                 }
 
@@ -5427,42 +5410,26 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
                 /* Avoid overlapping operands.  Wide source operands are
                    always aligned so will never overlap with wide
                    destinations in problematic ways.  */
-                if (rd == rm && !src2_wide) {
+                if (rd == rm) {
                     tmp = neon_load_reg(s, rm, 1);
                     neon_store_scratch(s, 2, tmp);
-                } else if (rd == rn && !src1_wide) {
+                } else if (rd == rn) {
                     tmp = neon_load_reg(s, rn, 1);
                     neon_store_scratch(s, 2, tmp);
                 }
                 tmp3 = NULL;
                 for (pass = 0; pass < 2; pass++) {
-                    if (src1_wide) {
+                    if (pass == 1 && rd == rn) {
-                        neon_load_reg64(s, s->V0, rn + pass);
+                        tmp = neon_load_scratch(s, 2);
-                        tmp = NULL;
                     } else {
-                        if (pass == 1 && rd == rn) {
+                        tmp = neon_load_reg(s, rn, pass);
-                            tmp = neon_load_scratch(s, 2);
-                        } else {
-                            tmp = neon_load_reg(s, rn, pass);
-                        }
                     }
-                    if (src2_wide) {
+                    if (pass == 1 && rd == rm) {
-                        neon_load_reg64(s, s->V1, rm + pass);
+                        tmp2 = neon_load_scratch(s, 2);
-                        tmp2 = NULL;
                     } else {
-                        if (pass == 1 && rd == rm) {
+                        tmp2 = neon_load_reg(s, rm, pass);
-                            tmp2 = neon_load_scratch(s, 2);
-                        } else {
-                            tmp2 = neon_load_reg(s, rm, pass);
-                        }
                     }
                     switch (op) {
-                    case 0: case 1: case 4: /* VADDL, VADDW, VADDHN, VRADDHN */
-                        gen_neon_addl(s, size);
-                        break;
-                    case 2: case 3: case 6: /* VSUBL, VSUBW, VSUBHN, VRSUBHN */
-                        gen_neon_subl(s, size);
-                        break;
                     case 5: case 7: /* VABAL, VABDL */
                         switch ((size << 1) | u) {
                         case 0:
@@ -5520,43 +5487,6 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
                             abort();
                         }
                         neon_store_reg64(s, s->V0, rd + pass);
-                    } else if (op == 4 || op == 6) {
-                        /* Narrowing operation.  */
-                        tmp = tcg_temp_new_i32(tcg_ctx);
-                        if (!u) {
-                            switch (size) {
-                            case 0:
-                                gen_helper_neon_narrow_high_u8(tcg_ctx, tmp, s->V0);
-                                break;
-                            case 1:
-                                gen_helper_neon_narrow_high_u16(tcg_ctx, tmp, s->V0);
-                                break;
-                            case 2:
-                                tcg_gen_extrh_i64_i32(tcg_ctx, tmp, s->V0);
-                                break;
-                            default: abort();
-                            }
-                        } else {
-                            switch (size) {
-                            case 0:
-                                gen_helper_neon_narrow_round_high_u8(tcg_ctx, tmp, s->V0);
-                                break;
-                            case 1:
-                                gen_helper_neon_narrow_round_high_u16(tcg_ctx, tmp, s->V0);
-                                break;
-                            case 2:
-                                tcg_gen_addi_i64(tcg_ctx, s->V0, s->V0, 1u << 31);
-                                tcg_gen_extrh_i64_i32(tcg_ctx, tmp, s->V0);
-                                break;
-                            default: abort();
-                            }
-                        }
-                        if (pass == 0) {
-                            tmp3 = tmp;
-                        } else {
-                            neon_store_reg(s, rd, 0, tmp3);
-                            neon_store_reg(s, rd, 1, tmp);
-                        }
                     } else {
                         /* Write back the result.  */
                         neon_store_reg64(s, s->V0, rd + pass);