target-tricore: Add instructions of RRR opcode format

Add microcode generator function gen_cond_sub.

Add helper functions:
    * ixmax/ixmin: search for the max/min value and its related index in a
                   vector of 16-bit values.
    * pack: dack two data registers into an IEEE-754 single precision floating
            point format number.
    * dvadj: divide-adjust the result after dvstep instructions.
    * dvstep: divide a reg by a divisor, producing 8-bits of quotient at a time.

OPCM_32_RRR_FLOAT -> OPCM_32_RRR_DIVIDE

Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
Reviewed-by: Richard Henderson <rth@twiddle.net>

4 files changed, 319 insertions, 1 deletions

diff --git a/target-tricore/helper.h b/target-tricore/helper.h
index 068dc7b..7405fee 100644
--- a/target-tricore/helper.h
+++ b/target-tricore/helper.h
@@ -60,10 +60,14 @@ DEF_HELPER_FLAGS_2(max_b, TCG_CALL_NO_RWG_SE, i32, i32, i32)
 DEF_HELPER_FLAGS_2(max_bu, TCG_CALL_NO_RWG_SE, i32, i32, i32)
 DEF_HELPER_FLAGS_2(max_h, TCG_CALL_NO_RWG_SE, i32, i32, i32)
 DEF_HELPER_FLAGS_2(max_hu, TCG_CALL_NO_RWG_SE, i32, i32, i32)
+DEF_HELPER_FLAGS_2(ixmax, TCG_CALL_NO_RWG_SE, i64, i64, i32)
+DEF_HELPER_FLAGS_2(ixmax_u, TCG_CALL_NO_RWG_SE, i64, i64, i32)
 DEF_HELPER_FLAGS_2(min_b, TCG_CALL_NO_RWG_SE, i32, i32, i32)
 DEF_HELPER_FLAGS_2(min_bu, TCG_CALL_NO_RWG_SE, i32, i32, i32)
 DEF_HELPER_FLAGS_2(min_h, TCG_CALL_NO_RWG_SE, i32, i32, i32)
 DEF_HELPER_FLAGS_2(min_hu, TCG_CALL_NO_RWG_SE, i32, i32, i32)
+DEF_HELPER_FLAGS_2(ixmin, TCG_CALL_NO_RWG_SE, i64, i64, i32)
+DEF_HELPER_FLAGS_2(ixmin_u, TCG_CALL_NO_RWG_SE, i64, i64, i32)
 /* count leading ... */
 DEF_HELPER_FLAGS_1(clo, TCG_CALL_NO_RWG_SE, i32, i32)
 DEF_HELPER_FLAGS_1(clo_h, TCG_CALL_NO_RWG_SE, i32, i32)
@@ -81,12 +85,16 @@ DEF_HELPER_FLAGS_2(bmerge, TCG_CALL_NO_RWG_SE, i32, i32, i32)
 DEF_HELPER_FLAGS_1(bsplit, TCG_CALL_NO_RWG_SE, i64, i32)
 DEF_HELPER_FLAGS_1(parity, TCG_CALL_NO_RWG_SE, i32, i32)
 /* float */
+DEF_HELPER_FLAGS_4(pack, TCG_CALL_NO_RWG_SE, i32, i32, i32, i32, i32)
 DEF_HELPER_1(unpack, i64, i32)
 /* dvinit */
 DEF_HELPER_3(dvinit_b_13, i64, env, i32, i32)
 DEF_HELPER_3(dvinit_b_131, i64, env, i32, i32)
 DEF_HELPER_3(dvinit_h_13, i64, env, i32, i32)
 DEF_HELPER_3(dvinit_h_131, i64, env, i32, i32)
+DEF_HELPER_FLAGS_2(dvadj, TCG_CALL_NO_RWG_SE, i64, i64, i32)
+DEF_HELPER_FLAGS_2(dvstep, TCG_CALL_NO_RWG_SE, i64, i64, i32)
+DEF_HELPER_FLAGS_2(dvstep_u, TCG_CALL_NO_RWG_SE, i64, i64, i32)
 /* mulh */
 DEF_HELPER_FLAGS_5(mul_h, TCG_CALL_NO_RWG_SE, i64, i32, i32, i32, i32, i32)
 DEF_HELPER_FLAGS_5(mulm_h, TCG_CALL_NO_RWG_SE, i64, i32, i32, i32, i32, i32)
diff --git a/target-tricore/op_helper.c b/target-tricore/op_helper.c
index 254135e..ed26b30 100644
--- a/target-tricore/op_helper.c
+++ b/target-tricore/op_helper.c
@@ -883,6 +883,50 @@ uint32_t helper_##name ##_hu(target_ulong r1, target_ulong r2)\
                                                               \
     return ret;                                               \
 }                                                             \
+                                                              \
+uint64_t helper_ix##name(uint64_t r1, uint32_t r2)            \
+{                                                             \
+    int64_t r2l, r2h, r1hl;                                   \
+    uint64_t ret = 0;                                         \
+                                                              \
+    ret = ((r1 + 2) & 0xffff);                                \
+    r2l = sextract64(r2, 0, 16);                              \
+    r2h = sextract64(r2, 16, 16);                             \
+    r1hl = sextract64(r1, 32, 16);                            \
+                                                              \
+    if ((r2l op ## = r2h) && (r2l op r1hl)) {                 \
+        ret |= (r2l & 0xffff) << 32;                          \
+        ret |= extract64(r1, 0, 16) << 16;                    \
+    } else if ((r2h op r2l) && (r2h op r1hl)) {               \
+        ret |= extract64(r2, 16, 16) << 32;                   \
+        ret |= extract64(r1 + 1, 0, 16) << 16;                \
+    } else {                                                  \
+        ret |= r1 & 0xffffffff0000ull;                        \
+    }                                                         \
+    return ret;                                               \
+}                                                             \
+                                                              \
+uint64_t helper_ix##name ##_u(uint64_t r1, uint32_t r2)       \
+{                                                             \
+    int64_t r2l, r2h, r1hl;                                   \
+    uint64_t ret = 0;                                         \
+                                                              \
+    ret = ((r1 + 2) & 0xffff);                                \
+    r2l = extract64(r2, 0, 16);                               \
+    r2h = extract64(r2, 16, 16);                              \
+    r1hl = extract64(r1, 32, 16);                             \
+                                                              \
+    if ((r2l op ## = r2h) && (r2l op r1hl)) {                 \
+        ret |= (r2l & 0xffff) << 32;                          \
+        ret |= extract64(r1, 0, 16) << 16;                    \
+    } else if ((r2h op r2l) && (r2h op r1hl)) {               \
+        ret |= extract64(r2, 16, 16) << 32;                   \
+        ret |= extract64(r1 + 1, 0, 16) << 16;                \
+    } else {                                                  \
+        ret |= r1 & 0xffffffff0000ull;                        \
+    }                                                         \
+    return ret;                                               \
+}
 
 EXTREMA_H_B(max, >)
 EXTREMA_H_B(min, <)
@@ -1116,6 +1160,48 @@ uint32_t helper_parity(target_ulong r1)
     return ret;
 }
 
+uint32_t helper_pack(uint32_t carry, uint32_t r1_low, uint32_t r1_high,
+                     target_ulong r2)
+{
+    uint32_t ret;
+    int32_t fp_exp, fp_frac, temp_exp, fp_exp_frac;
+    int32_t int_exp  = r1_high;
+    int32_t int_mant = r1_low;
+    uint32_t flag_rnd = (int_mant & (1 << 7)) && (
+                        (int_mant & (1 << 8)) ||
+                        (int_mant & 0x7f)     ||
+                        (carry != 0));
+    if (((int_mant & (1<<31)) == 0) && (int_exp == 255)) {
+        fp_exp = 255;
+        fp_frac = extract32(int_mant, 8, 23);
+    } else if ((int_mant & (1<<31)) && (int_exp >= 127)) {
+        fp_exp  = 255;
+        fp_frac = 0;
+    } else if ((int_mant & (1<<31)) && (int_exp <= -128)) {
+        fp_exp  = 0;
+        fp_frac = 0;
+    } else if (int_mant == 0) {
+        fp_exp  = 0;
+        fp_frac = 0;
+    } else {
+        if (((int_mant & (1 << 31)) == 0)) {
+            temp_exp = 0;
+        } else {
+            temp_exp = int_exp + 128;
+        }
+        fp_exp_frac = (((temp_exp & 0xff) << 23) |
+                      extract32(int_mant, 8, 23))
+                      + flag_rnd;
+        fp_exp  = extract32(fp_exp_frac, 23, 8);
+        fp_frac = extract32(fp_exp_frac, 0, 23);
+    }
+    ret = r2 & (1 << 31);
+    ret = ret + (fp_exp << 23);
+    ret = ret + (fp_frac & 0x7fffff);
+
+    return ret;
+}
+
 uint64_t helper_unpack(target_ulong arg1)
 {
     int32_t fp_exp  = extract32(arg1, 23, 8);
@@ -1244,6 +1330,80 @@ uint64_t helper_dvinit_h_131(CPUTriCoreState *env, uint32_t r1, uint32_t r2)
     return ret;
 }
 
+uint64_t helper_dvadj(uint64_t r1, uint32_t r2)
+{
+    int32_t x_sign = (r1 >> 63);
+    int32_t q_sign = x_sign ^ (r2 >> 31);
+    int32_t eq_pos = x_sign & ((r1 >> 32) == r2);
+    int32_t eq_neg = x_sign & ((r1 >> 32) == -r2);
+    uint32_t quotient;
+    uint64_t ret, remainder;
+
+    if ((q_sign & ~eq_neg) | eq_pos) {
+        quotient = (r1 + 1) & 0xffffffff;
+    } else {
+        quotient = r1 & 0xffffffff;
+    }
+
+    if (eq_pos | eq_neg) {
+        remainder = 0;
+    } else {
+        remainder = (r1 & 0xffffffff00000000ull);
+    }
+    ret = remainder|quotient;
+    return ret;
+}
+
+uint64_t helper_dvstep(uint64_t r1, uint32_t r2)
+{
+    int32_t dividend_sign = extract64(r1, 63, 1);
+    int32_t divisor_sign = extract32(r2, 31, 1);
+    int32_t quotient_sign = (dividend_sign != divisor_sign);
+    int32_t addend, dividend_quotient, remainder;
+    int32_t i, temp;
+
+    if (quotient_sign) {
+        addend = r2;
+    } else {
+        addend = -r2;
+    }
+    dividend_quotient = (int32_t)r1;
+    remainder = (int32_t)(r1 >> 32);
+
+    for (i = 0; i < 8; i++) {
+        remainder = (remainder << 1) | extract32(dividend_quotient, 31, 1);
+        dividend_quotient <<= 1;
+        temp = remainder + addend;
+        if ((temp < 0) == dividend_sign) {
+            remainder = temp;
+        }
+        if (((temp < 0) == dividend_sign)) {
+            dividend_quotient = dividend_quotient | !quotient_sign;
+        } else {
+            dividend_quotient = dividend_quotient | quotient_sign;
+        }
+    }
+    return ((uint64_t)remainder << 32) | (uint32_t)dividend_quotient;
+}
+
+uint64_t helper_dvstep_u(uint64_t r1, uint32_t r2)
+{
+    int32_t dividend_quotient = extract64(r1, 0, 32);
+    int64_t remainder = extract64(r1, 32, 32);
+    int32_t i;
+    int64_t temp;
+    for (i = 0; i < 8; i++) {
+        remainder = (remainder << 1) | extract32(dividend_quotient, 31, 1);
+        dividend_quotient <<= 1;
+        temp = (remainder & 0xffffffff) - r2;
+        if (temp >= 0) {
+            remainder = temp;
+        }
+        dividend_quotient = dividend_quotient | !(temp < 0);
+    }
+    return ((uint64_t)remainder << 32) | (uint32_t)dividend_quotient;
+}
+
 uint64_t helper_mul_h(uint32_t arg00, uint32_t arg01,
                       uint32_t arg10, uint32_t arg11, uint32_t n)
 {
diff --git a/target-tricore/translate.c b/target-tricore/translate.c
index 8943a39..a73b700 100644
--- a/target-tricore/translate.c
+++ b/target-tricore/translate.c
@@ -182,6 +182,18 @@ void tricore_cpu_dump_state(CPUState *cs, FILE *f,
     tcg_temp_free(arg11);                                    \
 } while (0)
 
+#define GEN_HELPER_RRR(name, rl, rh, al1, ah1, arg2) do {    \
+    TCGv_i64 ret = tcg_temp_new_i64();                       \
+    TCGv_i64 arg1 = tcg_temp_new_i64();                      \
+                                                             \
+    tcg_gen_concat_i32_i64(arg1, al1, ah1);                  \
+    gen_helper_##name(ret, arg1, arg2);                      \
+    tcg_gen_extr_i64_i32(rl, rh, ret);                       \
+                                                             \
+    tcg_temp_free_i64(ret);                                  \
+    tcg_temp_free_i64(arg1);                                 \
+} while (0)
+
 #define EA_ABS_FORMAT(con) (((con & 0x3C000) << 14) + (con & 0x3FFF))
 #define EA_B_ABSOLUT(con) (((offset & 0xf00000) << 8) | \
                            ((offset & 0x0fffff) << 1))
@@ -820,6 +832,45 @@ static inline void gen_subc_CC(TCGv ret, TCGv r1, TCGv r2)
     tcg_temp_free(temp);
 }
 
+static inline void gen_cond_sub(TCGCond cond, TCGv r1, TCGv r2, TCGv r3,
+                                TCGv r4)
+{
+    TCGv temp = tcg_temp_new();
+    TCGv temp2 = tcg_temp_new();
+    TCGv result = tcg_temp_new();
+    TCGv mask = tcg_temp_new();
+    TCGv t0 = tcg_const_i32(0);
+
+    /* create mask for sticky bits */
+    tcg_gen_setcond_tl(cond, mask, r4, t0);
+    tcg_gen_shli_tl(mask, mask, 31);
+
+    tcg_gen_sub_tl(result, r1, r2);
+    /* Calc PSW_V */
+    tcg_gen_xor_tl(temp, result, r1);
+    tcg_gen_xor_tl(temp2, r1, r2);
+    tcg_gen_and_tl(temp, temp, temp2);
+    tcg_gen_movcond_tl(cond, cpu_PSW_V, r4, t0, temp, cpu_PSW_V);
+    /* Set PSW_SV */
+    tcg_gen_and_tl(temp, temp, mask);
+    tcg_gen_or_tl(cpu_PSW_SV, temp, cpu_PSW_SV);
+    /* calc AV bit */
+    tcg_gen_add_tl(temp, result, result);
+    tcg_gen_xor_tl(temp, temp, result);
+    tcg_gen_movcond_tl(cond, cpu_PSW_AV, r4, t0, temp, cpu_PSW_AV);
+    /* calc SAV bit */
+    tcg_gen_and_tl(temp, temp, mask);
+    tcg_gen_or_tl(cpu_PSW_SAV, temp, cpu_PSW_SAV);
+    /* write back result */
+    tcg_gen_movcond_tl(cond, r3, r4, t0, result, r1);
+
+    tcg_temp_free(t0);
+    tcg_temp_free(temp);
+    tcg_temp_free(temp2);
+    tcg_temp_free(result);
+    tcg_temp_free(mask);
+}
+
 static inline void gen_abs(TCGv ret, TCGv r1)
 {
     TCGv temp = tcg_temp_new();
@@ -5042,6 +5093,99 @@ static void decode_rrpw_extract_insert(CPUTriCoreState *env, DisasContext *ctx)
     }
 }
 
+/* RRR format */
+static void decode_rrr_cond_select(CPUTriCoreState *env, DisasContext *ctx)
+{
+    uint32_t op2;
+    int r1, r2, r3, r4;
+    TCGv temp;
+
+    op2 = MASK_OP_RRR_OP2(ctx->opcode);
+    r1  = MASK_OP_RRR_S1(ctx->opcode);
+    r2  = MASK_OP_RRR_S2(ctx->opcode);
+    r3  = MASK_OP_RRR_S3(ctx->opcode);
+    r4  = MASK_OP_RRR_D(ctx->opcode);
+
+    switch (op2) {
+    case OPC2_32_RRR_CADD:
+        gen_cond_add(TCG_COND_NE, cpu_gpr_d[r1], cpu_gpr_d[r2],
+                     cpu_gpr_d[r4], cpu_gpr_d[r3]);
+        break;
+    case OPC2_32_RRR_CADDN:
+        gen_cond_add(TCG_COND_EQ, cpu_gpr_d[r1], cpu_gpr_d[r2], cpu_gpr_d[r4],
+                     cpu_gpr_d[r3]);
+        break;
+    case OPC2_32_RRR_CSUB:
+        gen_cond_sub(TCG_COND_NE, cpu_gpr_d[r1], cpu_gpr_d[r2], cpu_gpr_d[r4],
+                     cpu_gpr_d[r3]);
+        break;
+    case OPC2_32_RRR_CSUBN:
+        gen_cond_sub(TCG_COND_EQ, cpu_gpr_d[r1], cpu_gpr_d[r2], cpu_gpr_d[r4],
+                     cpu_gpr_d[r3]);
+        break;
+    case OPC2_32_RRR_SEL:
+        temp = tcg_const_i32(0);
+        tcg_gen_movcond_tl(TCG_COND_NE, cpu_gpr_d[r4], cpu_gpr_d[r3], temp,
+                           cpu_gpr_d[r1], cpu_gpr_d[r2]);
+        tcg_temp_free(temp);
+        break;
+    case OPC2_32_RRR_SELN:
+        temp = tcg_const_i32(0);
+        tcg_gen_movcond_tl(TCG_COND_EQ, cpu_gpr_d[r4], cpu_gpr_d[r3], temp,
+                           cpu_gpr_d[r1], cpu_gpr_d[r2]);
+        tcg_temp_free(temp);
+        break;
+    }
+}
+
+static void decode_rrr_divide(CPUTriCoreState *env, DisasContext *ctx)
+{
+    uint32_t op2;
+
+    int r1, r2, r3, r4;
+
+    op2 = MASK_OP_RRR_OP2(ctx->opcode);
+    r1 = MASK_OP_RRR_S1(ctx->opcode);
+    r2 = MASK_OP_RRR_S2(ctx->opcode);
+    r3 = MASK_OP_RRR_S3(ctx->opcode);
+    r4 = MASK_OP_RRR_D(ctx->opcode);
+
+    switch (op2) {
+    case OPC2_32_RRR_DVADJ:
+        GEN_HELPER_RRR(dvadj, cpu_gpr_d[r4], cpu_gpr_d[r4+1], cpu_gpr_d[r3],
+                       cpu_gpr_d[r3+1], cpu_gpr_d[r2]);
+        break;
+    case OPC2_32_RRR_DVSTEP:
+        GEN_HELPER_RRR(dvstep, cpu_gpr_d[r4], cpu_gpr_d[r4+1], cpu_gpr_d[r3],
+                       cpu_gpr_d[r3+1], cpu_gpr_d[r2]);
+        break;
+    case OPC2_32_RRR_DVSTEP_U:
+        GEN_HELPER_RRR(dvstep_u, cpu_gpr_d[r4], cpu_gpr_d[r4+1], cpu_gpr_d[r3],
+                       cpu_gpr_d[r3+1], cpu_gpr_d[r2]);
+        break;
+    case OPC2_32_RRR_IXMAX:
+        GEN_HELPER_RRR(ixmax, cpu_gpr_d[r4], cpu_gpr_d[r4+1], cpu_gpr_d[r3],
+                       cpu_gpr_d[r3+1], cpu_gpr_d[r2]);
+        break;
+    case OPC2_32_RRR_IXMAX_U:
+        GEN_HELPER_RRR(ixmax_u, cpu_gpr_d[r4], cpu_gpr_d[r4+1], cpu_gpr_d[r3],
+                       cpu_gpr_d[r3+1], cpu_gpr_d[r2]);
+        break;
+    case OPC2_32_RRR_IXMIN:
+        GEN_HELPER_RRR(ixmin, cpu_gpr_d[r4], cpu_gpr_d[r4+1], cpu_gpr_d[r3],
+                       cpu_gpr_d[r3+1], cpu_gpr_d[r2]);
+        break;
+    case OPC2_32_RRR_IXMIN_U:
+        GEN_HELPER_RRR(ixmin_u, cpu_gpr_d[r4], cpu_gpr_d[r4+1], cpu_gpr_d[r3],
+                       cpu_gpr_d[r3+1], cpu_gpr_d[r2]);
+        break;
+    case OPC2_32_RRR_PACK:
+        gen_helper_pack(cpu_gpr_d[r4], cpu_PSW_C, cpu_gpr_d[r3],
+                        cpu_gpr_d[r3+1], cpu_gpr_d[r1]);
+        break;
+    }
+}
+
 static void decode_32Bit_opc(CPUTriCoreState *env, DisasContext *ctx)
 {
     int op1;
@@ -5325,6 +5469,12 @@ static void decode_32Bit_opc(CPUTriCoreState *env, DisasContext *ctx)
             tcg_temp_free(temp);
         }
         break;
+/* RRR Format */
+    case OPCM_32_RRR_COND_SELECT:
+        decode_rrr_cond_select(env, ctx);
+        break;
+    case OPCM_32_RRR_DIVIDE:
+        decode_rrr_divide(env, ctx);
     }
 }
 
diff --git a/target-tricore/tricore-opcodes.h b/target-tricore/tricore-opcodes.h
index 82bd161..baf537f 100644
--- a/target-tricore/tricore-opcodes.h
+++ b/target-tricore/tricore-opcodes.h
@@ -516,7 +516,7 @@ enum {
     OPC1_32_RRPW_DEXTR                               = 0x77,
 /* RRR Format */
     OPCM_32_RRR_COND_SELECT                          = 0x2b,
-    OPCM_32_RRR_FLOAT                                = 0x6b,
+    OPCM_32_RRR_DIVIDE                               = 0x6b,
 /* RRR1 Format */
     OPCM_32_RRR1_MADD                                = 0x83,
     OPCM_32_RRR1_MADDQ_H                             = 0x43,