target-sh4: implement addv and subv using TCG

addv and subv helpers implementation is directly copied from the SH4
manual and looks quite complex. It is however possible to explain it
without branches, and is therefore possible to implement it with TCG.
Signed-off-by: NAurelien Jarno <aurelien@aurel32.net>

target-sh4/helper.h

@@ -13,8 +13,6 @@ DEF_HELPER_3(movcal, void, env, i32, i32)
 DEF_HELPER_1(discard_movcal_backup, void, env)
 DEF_HELPER_2(ocbi, void, env, i32)
 
-DEF_HELPER_3(addv, i32, env, i32, i32)
-DEF_HELPER_3(subv, i32, env, i32, i32)
 DEF_HELPER_3(div1, i32, env, i32, i32)
 DEF_HELPER_3(macl, void, env, i32, i32)
 DEF_HELPER_3(macw, void, env, i32, i32)

target-sh4/op_helper.c

@@ -177,35 +177,6 @@ void helper_ocbi(CPUSH4State *env, uint32_t address)
     }
 }
 
-uint32_t helper_addv(CPUSH4State *env, uint32_t arg0, uint32_t arg1)
-{
-    uint32_t dest, src, ans;
-
-    if ((int32_t) arg1 >= 0)
-	dest = 0;
-    else
-	dest = 1;
-    if ((int32_t) arg0 >= 0)
-	src = 0;
-    else
-	src = 1;
-    src += dest;
-    arg1 += arg0;
-    if ((int32_t) arg1 >= 0)
-	ans = 0;
-    else
-	ans = 1;
-    ans += dest;
-    if (src == 0 || src == 2) {
-	if (ans == 1)
-	    env->sr |= SR_T;
-	else
-	    env->sr &= ~SR_T;
-    } else
-	env->sr &= ~SR_T;
-    return arg1;
-}
-
 #define T (env->sr & SR_T)
 #define Q (env->sr & SR_Q ? 1 : 0)
 #define M (env->sr & SR_M ? 1 : 0)
@@ -359,35 +330,6 @@ void helper_macw(CPUSH4State *env, uint32_t arg0, uint32_t arg1)
     }
 }
 
-uint32_t helper_subv(CPUSH4State *env, uint32_t arg0, uint32_t arg1)
-{
-    int32_t dest, src, ans;
-
-    if ((int32_t) arg1 >= 0)
-	dest = 0;
-    else
-	dest = 1;
-    if ((int32_t) arg0 >= 0)
-	src = 0;
-    else
-	src = 1;
-    src += dest;
-    arg1 -= arg0;
-    if ((int32_t) arg1 >= 0)
-	ans = 0;
-    else
-	ans = 1;
-    ans += dest;
-    if (src == 1) {
-	if (ans == 1)
-	    env->sr |= SR_T;
-	else
-	    env->sr &= ~SR_T;
-    } else
-	env->sr &= ~SR_T;
-    return arg1;
-}
-
 static inline void set_t(CPUSH4State *env)
 {
     env->sr |= SR_T;

target-sh4/translate.c

@@ -781,7 +781,23 @@ static void _decode_opc(DisasContext * ctx)
         }
 	return;
     case 0x300f:		/* addv Rm,Rn */
-        gen_helper_addv(REG(B11_8), cpu_env, REG(B7_4), REG(B11_8));
+        {
+            TCGv t0, t1, t2;
+            t0 = tcg_temp_new();
+            tcg_gen_add_i32(t0, REG(B7_4), REG(B11_8));
+            t1 = tcg_temp_new();
+            tcg_gen_xor_i32(t1, t0, REG(B11_8));
+            t2 = tcg_temp_new();
+            tcg_gen_xor_i32(t2, REG(B7_4), REG(B11_8));
+            tcg_gen_andc_i32(t1, t1, t2);
+            tcg_temp_free(t2);
+            tcg_gen_shri_i32(t1, t1, 31);
+            tcg_gen_andi_i32(cpu_sr, cpu_sr, ~SR_T);
+            tcg_gen_or_i32(cpu_sr, cpu_sr, t1);
+            tcg_temp_free(t1);
+            tcg_gen_mov_i32(REG(B7_4), t0);
+            tcg_temp_free(t0);
+        }
 	return;
     case 0x2009:		/* and Rm,Rn */
 	tcg_gen_and_i32(REG(B11_8), REG(B11_8), REG(B7_4));
@@ -1050,7 +1066,23 @@ static void _decode_opc(DisasContext * ctx)
         }
 	return;
     case 0x300b:		/* subv Rm,Rn */
-        gen_helper_subv(REG(B11_8), cpu_env, REG(B7_4), REG(B11_8));
+        {
+            TCGv t0, t1, t2;
+            t0 = tcg_temp_new();
+            tcg_gen_sub_i32(t0, REG(B11_8), REG(B7_4));
+            t1 = tcg_temp_new();
+            tcg_gen_xor_i32(t1, t0, REG(B7_4));
+            t2 = tcg_temp_new();
+            tcg_gen_xor_i32(t2, REG(B11_8), REG(B7_4));
+            tcg_gen_and_i32(t1, t1, t2);
+            tcg_temp_free(t2);
+            tcg_gen_shri_i32(t1, t1, 31);
+            tcg_gen_andi_i32(cpu_sr, cpu_sr, ~SR_T);
+            tcg_gen_or_i32(cpu_sr, cpu_sr, t1);
+            tcg_temp_free(t1);
+            tcg_gen_mov_i32(REG(B11_8), t0);
+            tcg_temp_free(t0);
+        }
 	return;
     case 0x2008:		/* tst Rm,Rn */
 	{