target-arm: convert NEON VZIP/VUZP/VTRN helper functions to pure TCG

The neon_trn_u8, neon_trn_u16, neon_unzip_u8, neon_zip_u8 and neon_zip_u16
helpers used fixed registers to return values. This patch replaces that with
TCG code, so T0/T1 is no longer directly used by the helper functions.

Bugs in the gen_neon_unzip register load code were also fixed.

Signed-off-by: Filip Navara <filip.navara@gmail.com>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>

3 files changed, 142 insertions, 73 deletions

diff --git a/target-arm/helpers.h b/target-arm/helpers.h
index 01175e157a..f298efff42 100644
--- a/target-arm/helpers.h
+++ b/target-arm/helpers.h
@@ -338,12 +338,6 @@ DEF_HELPER_2(neon_qneg_s8, i32, env, i32)
 DEF_HELPER_2(neon_qneg_s16, i32, env, i32)
 DEF_HELPER_2(neon_qneg_s32, i32, env, i32)
 
-DEF_HELPER_0(neon_trn_u8, void)
-DEF_HELPER_0(neon_trn_u16, void)
-DEF_HELPER_0(neon_unzip_u8, void)
-DEF_HELPER_0(neon_zip_u8, void)
-DEF_HELPER_0(neon_zip_u16, void)
-
 DEF_HELPER_2(neon_min_f32, i32, i32, i32)
 DEF_HELPER_2(neon_max_f32, i32, i32, i32)
 DEF_HELPER_2(neon_abd_f32, i32, i32, i32)
diff --git a/target-arm/op_helper.c b/target-arm/op_helper.c
index d4ae4ae7e3..5ac631df0e 100644
--- a/target-arm/op_helper.c
+++ b/target-arm/op_helper.c
@@ -495,61 +495,3 @@ uint64_t HELPER(neon_sub_saturate_u64)(uint64_t src1, uint64_t src2)
     }
     return res;
 }
-
-/* These need to return a pair of value, so still use T0/T1.  */
-/* Transpose.  Argument order is rather strange to avoid special casing
-   the tranlation code.
-   On input T0 = rm, T1 = rd.  On output T0 = rd, T1 = rm  */
-void HELPER(neon_trn_u8)(void)
-{
-    uint32_t rd;
-    uint32_t rm;
-    rd = ((T0 & 0x00ff00ff) << 8) | (T1 & 0x00ff00ff);
-    rm = ((T1 & 0xff00ff00) >> 8) | (T0 & 0xff00ff00);
-    T0 = rd;
-    T1 = rm;
-}
-
-void HELPER(neon_trn_u16)(void)
-{
-    uint32_t rd;
-    uint32_t rm;
-    rd = (T0 << 16) | (T1 & 0xffff);
-    rm = (T1 >> 16) | (T0 & 0xffff0000);
-    T0 = rd;
-    T1 = rm;
-}
-
-/* Worker routines for zip and unzip.  */
-void HELPER(neon_unzip_u8)(void)
-{
-    uint32_t rd;
-    uint32_t rm;
-    rd = (T0 & 0xff) | ((T0 >> 8) & 0xff00)
-         | ((T1 << 16) & 0xff0000) | ((T1 << 8) & 0xff000000);
-    rm = ((T0 >> 8) & 0xff) | ((T0 >> 16) & 0xff00)
-         | ((T1 << 8) & 0xff0000) | (T1 & 0xff000000);
-    T0 = rd;
-    T1 = rm;
-}
-
-void HELPER(neon_zip_u8)(void)
-{
-    uint32_t rd;
-    uint32_t rm;
-    rd = (T0 & 0xff) | ((T1 << 8) & 0xff00)
-         | ((T0 << 16) & 0xff0000) | ((T1 << 24) & 0xff000000);
-    rm = ((T0 >> 16) & 0xff) | ((T1 >> 8) & 0xff00)
-         | ((T0 >> 8) & 0xff0000) | (T1 & 0xff000000);
-    T0 = rd;
-    T1 = rm;
-}
-
-void HELPER(neon_zip_u16)(void)
-{
-    uint32_t tmp;
-
-    tmp = (T0 & 0xffff) | (T1 << 16);
-    T1 = (T1 & 0xffff0000) | (T0 >> 16);
-    T0 = tmp;
-}
diff --git a/target-arm/translate.c b/target-arm/translate.c
index 8e5380cc6e..8c5afb7234 100644
--- a/target-arm/translate.c
+++ b/target-arm/translate.c
@@ -3627,24 +3627,157 @@ static inline void gen_neon_get_scalar(int size, int reg)
     }
 }
 
+static void gen_neon_unzip_u8(TCGv t0, TCGv t1)
+{
+    TCGv rd, rm, tmp;
+
+    rd = new_tmp();
+    rm = new_tmp();
+    tmp = new_tmp();
+
+    tcg_gen_andi_i32(rd, t0, 0xff);
+    tcg_gen_shri_i32(tmp, t0, 8);
+    tcg_gen_andi_i32(tmp, tmp, 0xff00);
+    tcg_gen_or_i32(rd, rd, tmp);
+    tcg_gen_shli_i32(tmp, t1, 16);
+    tcg_gen_andi_i32(tmp, tmp, 0xff0000);
+    tcg_gen_or_i32(rd, rd, tmp);
+    tcg_gen_shli_i32(tmp, t1, 8);
+    tcg_gen_andi_i32(tmp, tmp, 0xff000000);
+    tcg_gen_or_i32(rd, rd, tmp);
+
+    tcg_gen_shri_i32(rm, t0, 8);
+    tcg_gen_andi_i32(rm, rm, 0xff);
+    tcg_gen_shri_i32(tmp, t0, 16);
+    tcg_gen_andi_i32(tmp, tmp, 0xff00);
+    tcg_gen_or_i32(rm, rm, tmp);
+    tcg_gen_shli_i32(tmp, t1, 8);
+    tcg_gen_andi_i32(tmp, tmp, 0xff0000);
+    tcg_gen_or_i32(rm, rm, tmp);
+    tcg_gen_andi_i32(tmp, t1, 0xff000000);
+    tcg_gen_or_i32(t1, rm, tmp);
+    tcg_gen_mov_i32(t0, rd);
+
+    dead_tmp(tmp);
+    dead_tmp(rm);
+    dead_tmp(rd);
+}
+
+static void gen_neon_zip_u8(TCGv t0, TCGv t1)
+{
+    TCGv rd, rm, tmp;
+
+    rd = new_tmp();
+    rm = new_tmp();
+    tmp = new_tmp();
+
+    tcg_gen_andi_i32(rd, t0, 0xff);
+    tcg_gen_shli_i32(tmp, t1, 8);
+    tcg_gen_andi_i32(tmp, tmp, 0xff00);
+    tcg_gen_or_i32(rd, rd, tmp);
+    tcg_gen_shli_i32(tmp, t0, 16);
+    tcg_gen_andi_i32(tmp, tmp, 0xff0000);
+    tcg_gen_or_i32(rd, rd, tmp);
+    tcg_gen_shli_i32(tmp, t1, 24);
+    tcg_gen_andi_i32(tmp, tmp, 0xff000000);
+    tcg_gen_or_i32(rd, rd, tmp);
+
+    tcg_gen_andi_i32(rm, t1, 0xff000000);
+    tcg_gen_shri_i32(tmp, t0, 8);
+    tcg_gen_andi_i32(tmp, tmp, 0xff0000);
+    tcg_gen_or_i32(rm, rm, tmp);
+    tcg_gen_shri_i32(tmp, t1, 8);
+    tcg_gen_andi_i32(tmp, tmp, 0xff00);
+    tcg_gen_or_i32(rm, rm, tmp);
+    tcg_gen_shri_i32(tmp, t0, 16);
+    tcg_gen_andi_i32(tmp, tmp, 0xff);
+    tcg_gen_or_i32(t1, rm, tmp);
+    tcg_gen_mov_i32(t0, rd);
+
+    dead_tmp(tmp);
+    dead_tmp(rm);
+    dead_tmp(rd);
+}
+
+static void gen_neon_zip_u16(TCGv t0, TCGv t1)
+{
+    TCGv tmp, tmp2;
+
+    tmp = new_tmp();
+    tmp2 = new_tmp();
+
+    tcg_gen_andi_i32(tmp, t0, 0xffff);
+    tcg_gen_shli_i32(tmp2, t1, 16);
+    tcg_gen_or_i32(tmp, tmp, tmp2);
+    tcg_gen_andi_i32(t1, t1, 0xffff0000);
+    tcg_gen_shri_i32(tmp2, t0, 16);
+    tcg_gen_or_i32(t1, t1, tmp2);
+    tcg_gen_mov_i32(t0, tmp);
+
+    dead_tmp(tmp2);
+    dead_tmp(tmp);
+}
+
 static void gen_neon_unzip(int reg, int q, int tmp, int size)
 {
     int n;
 
     for (n = 0; n < q + 1; n += 2) {
         NEON_GET_REG(T0, reg, n);
-        NEON_GET_REG(T0, reg, n + n);
+        NEON_GET_REG(T1, reg, n + 1);
         switch (size) {
-        case 0: gen_helper_neon_unzip_u8(); break;
-        case 1: gen_helper_neon_zip_u16(); break; /* zip and unzip are the same.  */
+        case 0: gen_neon_unzip_u8(cpu_T[0], cpu_T[1]); break;
+        case 1: gen_neon_zip_u16(cpu_T[0], cpu_T[1]); break; /* zip and unzip are the same.  */
         case 2: /* no-op */; break;
         default: abort();
         }
-        gen_neon_movl_scratch_T0(tmp + n);
-        gen_neon_movl_scratch_T1(tmp + n + 1);
+        gen_neon_movl_T0_scratch(tmp + n);
+        gen_neon_movl_T1_scratch(tmp + n + 1);
     }
 }
 
+static void gen_neon_trn_u8(TCGv t0, TCGv t1)
+{
+    TCGv rd, tmp;
+
+    rd = new_tmp();
+    tmp = new_tmp();
+
+    tcg_gen_shli_i32(rd, t0, 8);
+    tcg_gen_andi_i32(rd, rd, 0xff00ff00);
+    tcg_gen_andi_i32(tmp, t1, 0x00ff00ff);
+    tcg_gen_or_i32(rd, rd, tmp);
+
+    tcg_gen_shri_i32(t1, t1, 8);
+    tcg_gen_andi_i32(t1, t1, 0x00ff00ff);
+    tcg_gen_andi_i32(tmp, t0, 0xff00ff00);
+    tcg_gen_or_i32(t1, t1, tmp);
+    tcg_gen_mov_i32(t0, rd);
+
+    dead_tmp(tmp);
+    dead_tmp(rd);
+}
+
+static void gen_neon_trn_u16(TCGv t0, TCGv t1)
+{
+    TCGv rd, tmp;
+
+    rd = new_tmp();
+    tmp = new_tmp();
+
+    tcg_gen_shli_i32(rd, t0, 16);
+    tcg_gen_andi_i32(tmp, t1, 0xffff);
+    tcg_gen_or_i32(rd, rd, tmp);
+    tcg_gen_shri_i32(t1, t1, 16);
+    tcg_gen_andi_i32(tmp, t0, 0xffff0000);
+    tcg_gen_or_i32(t1, t1, tmp);
+    tcg_gen_mov_i32(t0, rd);
+
+    dead_tmp(tmp);
+    dead_tmp(rd);
+}
+
+
 static struct {
     int nregs;
     int interleave;
@@ -5256,8 +5389,8 @@ static int disas_neon_data_insn(CPUState * env, DisasContext *s, uint32_t insn)
                         NEON_GET_REG(T0, rd, n);
                         NEON_GET_REG(T1, rd, n);
                         switch (size) {
-                        case 0: gen_helper_neon_zip_u8(); break;
-                        case 1: gen_helper_neon_zip_u16(); break;
+                        case 0: gen_neon_zip_u8(cpu_T[0], cpu_T[1]); break;
+                        case 1: gen_neon_zip_u16(cpu_T[0], cpu_T[1]); break;
                         case 2: /* no-op */; break;
                         default: abort();
                         }
@@ -5442,8 +5575,8 @@ static int disas_neon_data_insn(CPUState * env, DisasContext *s, uint32_t insn)
                         case 33: /* VTRN */
                             NEON_GET_REG(T1, rd, pass);
                             switch (size) {
-                            case 0: gen_helper_neon_trn_u8(); break;
-                            case 1: gen_helper_neon_trn_u16(); break;
+                            case 0: gen_neon_trn_u8(cpu_T[0], cpu_T[1]); break;
+                            case 1: gen_neon_trn_u16(cpu_T[0], cpu_T[1]); break;
                             case 2: abort();
                             default: return 1;
                             }