This patch adds REQUIRE_FP/FP_SP/FP_DP to check CPUCFG2.FP/FP_SP/FP_DP.

Signed-off-by: Song Gao <gaosong@loongson.cn>
---
 target/loongarch/cpu.h                        |   6 +
 .../loongarch/insn_trans/trans_farith.c.inc   | 132 ++++++++++++------
 target/loongarch/insn_trans/trans_fcmp.c.inc  |   4 +
 target/loongarch/insn_trans/trans_fcnv.c.inc  |  56 ++++----
 .../loongarch/insn_trans/trans_fmemory.c.inc  | 104 ++++++++++----
 target/loongarch/insn_trans/trans_fmov.c.inc  |  47 +++++--
 6 files changed, 247 insertions(+), 102 deletions(-)

diff --git a/target/loongarch/cpu.h b/target/loongarch/cpu.h
index 9f550793ca..5594d83011 100644
--- a/target/loongarch/cpu.h
+++ b/target/loongarch/cpu.h
@@ -459,6 +459,9 @@ static inline void set_pc(CPULoongArchState *env, uint64_t value)
 #define HW_FLAGS_EUEN_FPE   0x04
 #define HW_FLAGS_EUEN_SXE   0x08
 #define HW_FLAGS_VA32       0x20
+#define HW_FLAGS_FP         0x40
+#define HW_FLAGS_FP_SP      0x80
+#define HW_FLAGS_FP_DP      0x100
 
 static inline void cpu_get_tb_cpu_state(CPULoongArchState *env, vaddr *pc,
                                         uint64_t *cs_base, uint32_t *flags)
@@ -469,6 +472,9 @@ static inline void cpu_get_tb_cpu_state(CPULoongArchState *env, vaddr *pc,
     *flags |= FIELD_EX64(env->CSR_EUEN, CSR_EUEN, FPE) * HW_FLAGS_EUEN_FPE;
     *flags |= FIELD_EX64(env->CSR_EUEN, CSR_EUEN, SXE) * HW_FLAGS_EUEN_SXE;
     *flags |= is_va32(env) * HW_FLAGS_VA32;
+    *flags |= FIELD_EX32(env->cpucfg[2], CPUCFG2, FP) * HW_FLAGS_FP;
+    *flags |= FIELD_EX32(env->cpucfg[2], CPUCFG2, FP_SP) * HW_FLAGS_FP_SP;
+    *flags |= FIELD_EX32(env->cpucfg[2], CPUCFG2, FP_DP) * HW_FLAGS_FP_DP;
 }
 
 void loongarch_cpu_list(void);
diff --git a/target/loongarch/insn_trans/trans_farith.c.inc b/target/loongarch/insn_trans/trans_farith.c.inc
index 21ea47308b..f8ff5b7bb6 100644
--- a/target/loongarch/insn_trans/trans_farith.c.inc
+++ b/target/loongarch/insn_trans/trans_farith.c.inc
@@ -10,17 +10,45 @@
         return true; \
     } \
 } while (0)
+
+#define REQUIRE_FP do { \
+    if ((ctx->base.tb->flags & HW_FLAGS_FP) == 0) { \
+        return false; \
+    } \
+} while (0)
+
+#define REQUIRE_FP_SP do { \
+    if ((ctx->base.tb->flags & HW_FLAGS_FP_SP) == 0) { \
+        return false; \
+    } \
+} while (0)
+
+#define REQUIRE_FP_DP do { \
+    if ((ctx->base.tb->flags & HW_FLAGS_FP_DP) == 0) { \
+        return false; \
+    } \
+} while (0)
 #else
 #define CHECK_FPE
+#define REQUIRE_FP
+#define REQUIRE_FP_SP
+#define REQUIRE_FP_DP
 #endif
 
-static bool gen_fff(DisasContext *ctx, arg_fff *a,
+static bool gen_fff(DisasContext *ctx, arg_fff *a, int mode,
                     void (*func)(TCGv, TCGv_env, TCGv, TCGv))
 {
     TCGv dest = get_fpr(ctx, a->fd);
     TCGv src1 = get_fpr(ctx, a->fj);
     TCGv src2 = get_fpr(ctx, a->fk);
 
+    REQUIRE_FP;
+    if (mode == 0) {
+        REQUIRE_FP_SP;
+    } else {
+        REQUIRE_FP_DP;
+    }
+
     CHECK_FPE;
 
     func(dest, cpu_env, src1, src2);
@@ -29,12 +57,19 @@ static bool gen_fff(DisasContext *ctx, arg_fff *a,
     return true;
 }
 
-static bool gen_ff(DisasContext *ctx, arg_ff *a,
+static bool gen_ff(DisasContext *ctx, arg_ff *a, int mode,
                    void (*func)(TCGv, TCGv_env, TCGv))
 {
     TCGv dest = get_fpr(ctx, a->fd);
     TCGv src = get_fpr(ctx, a->fj);
 
+    REQUIRE_FP;
+    if (mode == 0) {
+        REQUIRE_FP_SP;
+    } else {
+        REQUIRE_FP_DP;
+    }
+
     CHECK_FPE;
 
     func(dest, cpu_env, src);
@@ -43,7 +78,7 @@ static bool gen_ff(DisasContext *ctx, arg_ff *a,
     return true;
 }
 
-static bool gen_muladd(DisasContext *ctx, arg_ffff *a,
+static bool gen_muladd(DisasContext *ctx, arg_ffff *a, int mode,
                        void (*func)(TCGv, TCGv_env, TCGv, TCGv, TCGv, TCGv_i32),
                        int flag)
 {
@@ -53,6 +88,13 @@ static bool gen_muladd(DisasContext *ctx, arg_ffff *a,
     TCGv src2 = get_fpr(ctx, a->fk);
     TCGv src3 = get_fpr(ctx, a->fa);
 
+    REQUIRE_FP;
+    if (mode == 0) {
+        REQUIRE_FP_SP;
+    } else {
+        REQUIRE_FP_DP;
+    }
+
     CHECK_FPE;
 
     func(dest, cpu_env, src1, src2, src3, tflag);
@@ -67,6 +109,8 @@ static bool trans_fcopysign_s(DisasContext *ctx, arg_fcopysign_s *a)
     TCGv src1 = get_fpr(ctx, a->fk);
     TCGv src2 = get_fpr(ctx, a->fj);
 
+    REQUIRE_FP;
+    REQUIRE_FP_SP;
     CHECK_FPE;
 
     tcg_gen_deposit_i64(dest, src1, src2, 0, 31);
@@ -81,6 +125,8 @@ static bool trans_fcopysign_d(DisasContext *ctx, arg_fcopysign_d *a)
     TCGv src1 = get_fpr(ctx, a->fk);
     TCGv src2 = get_fpr(ctx, a->fj);
 
+    REQUIRE_FP;
+    REQUIRE_FP_DP;
     CHECK_FPE;
 
     tcg_gen_deposit_i64(dest, src1, src2, 0, 63);
@@ -94,6 +140,8 @@ static bool trans_fabs_s(DisasContext *ctx, arg_fabs_s *a)
     TCGv dest = get_fpr(ctx, a->fd);
     TCGv src = get_fpr(ctx, a->fj);
 
+    REQUIRE_FP;
+    REQUIRE_FP_SP;
     CHECK_FPE;
 
     tcg_gen_andi_i64(dest, src, MAKE_64BIT_MASK(0, 31));
@@ -108,6 +156,8 @@ static bool trans_fabs_d(DisasContext *ctx, arg_fabs_d *a)
     TCGv dest = get_fpr(ctx, a->fd);
     TCGv src = get_fpr(ctx, a->fj);
 
+    REQUIRE_FP;
+    REQUIRE_FP_DP;
     CHECK_FPE;
 
     tcg_gen_andi_i64(dest, src, MAKE_64BIT_MASK(0, 63));
@@ -121,6 +171,8 @@ static bool trans_fneg_s(DisasContext *ctx, arg_fneg_s *a)
     TCGv dest = get_fpr(ctx, a->fd);
     TCGv src = get_fpr(ctx, a->fj);
 
+    REQUIRE_FP;
+    REQUIRE_FP_SP;
     CHECK_FPE;
 
     tcg_gen_xori_i64(dest, src, 0x80000000);
@@ -135,6 +187,8 @@ static bool trans_fneg_d(DisasContext *ctx, arg_fneg_d *a)
     TCGv dest = get_fpr(ctx, a->fd);
     TCGv src = get_fpr(ctx, a->fj);
 
+    REQUIRE_FP;
+    REQUIRE_FP_DP;
     CHECK_FPE;
 
     tcg_gen_xori_i64(dest, src, 0x8000000000000000LL);
@@ -143,41 +197,41 @@ static bool trans_fneg_d(DisasContext *ctx, arg_fneg_d *a)
     return true;
 }
 
-TRANS(fadd_s, gen_fff, gen_helper_fadd_s)
-TRANS(fadd_d, gen_fff, gen_helper_fadd_d)
-TRANS(fsub_s, gen_fff, gen_helper_fsub_s)
-TRANS(fsub_d, gen_fff, gen_helper_fsub_d)
-TRANS(fmul_s, gen_fff, gen_helper_fmul_s)
-TRANS(fmul_d, gen_fff, gen_helper_fmul_d)
-TRANS(fdiv_s, gen_fff, gen_helper_fdiv_s)
-TRANS(fdiv_d, gen_fff, gen_helper_fdiv_d)
-TRANS(fmax_s, gen_fff, gen_helper_fmax_s)
-TRANS(fmax_d, gen_fff, gen_helper_fmax_d)
-TRANS(fmin_s, gen_fff, gen_helper_fmin_s)
-TRANS(fmin_d, gen_fff, gen_helper_fmin_d)
-TRANS(fmaxa_s, gen_fff, gen_helper_fmaxa_s)
-TRANS(fmaxa_d, gen_fff, gen_helper_fmaxa_d)
-TRANS(fmina_s, gen_fff, gen_helper_fmina_s)
-TRANS(fmina_d, gen_fff, gen_helper_fmina_d)
-TRANS(fscaleb_s, gen_fff, gen_helper_fscaleb_s)
-TRANS(fscaleb_d, gen_fff, gen_helper_fscaleb_d)
-TRANS(fsqrt_s, gen_ff, gen_helper_fsqrt_s)
-TRANS(fsqrt_d, gen_ff, gen_helper_fsqrt_d)
-TRANS(frecip_s, gen_ff, gen_helper_frecip_s)
-TRANS(frecip_d, gen_ff, gen_helper_frecip_d)
-TRANS(frsqrt_s, gen_ff, gen_helper_frsqrt_s)
-TRANS(frsqrt_d, gen_ff, gen_helper_frsqrt_d)
-TRANS(flogb_s, gen_ff, gen_helper_flogb_s)
-TRANS(flogb_d, gen_ff, gen_helper_flogb_d)
-TRANS(fclass_s, gen_ff, gen_helper_fclass_s)
-TRANS(fclass_d, gen_ff, gen_helper_fclass_d)
-TRANS(fmadd_s, gen_muladd, gen_helper_fmuladd_s, 0)
-TRANS(fmadd_d, gen_muladd, gen_helper_fmuladd_d, 0)
-TRANS(fmsub_s, gen_muladd, gen_helper_fmuladd_s, float_muladd_negate_c)
-TRANS(fmsub_d, gen_muladd, gen_helper_fmuladd_d, float_muladd_negate_c)
-TRANS(fnmadd_s, gen_muladd, gen_helper_fmuladd_s, float_muladd_negate_result)
-TRANS(fnmadd_d, gen_muladd, gen_helper_fmuladd_d, float_muladd_negate_result)
-TRANS(fnmsub_s, gen_muladd, gen_helper_fmuladd_s,
+TRANS(fadd_s, gen_fff, 0, gen_helper_fadd_s)
+TRANS(fadd_d, gen_fff, 1, gen_helper_fadd_d)
+TRANS(fsub_s, gen_fff, 0, gen_helper_fsub_s)
+TRANS(fsub_d, gen_fff, 1, gen_helper_fsub_d)
+TRANS(fmul_s, gen_fff, 0, gen_helper_fmul_s)
+TRANS(fmul_d, gen_fff, 1, gen_helper_fmul_d)
+TRANS(fdiv_s, gen_fff, 0, gen_helper_fdiv_s)
+TRANS(fdiv_d, gen_fff, 1, gen_helper_fdiv_d)
+TRANS(fmax_s, gen_fff, 0, gen_helper_fmax_s)
+TRANS(fmax_d, gen_fff, 1, gen_helper_fmax_d)
+TRANS(fmin_s, gen_fff, 0, gen_helper_fmin_s)
+TRANS(fmin_d, gen_fff, 1, gen_helper_fmin_d)
+TRANS(fmaxa_s, gen_fff, 0, gen_helper_fmaxa_s)
+TRANS(fmaxa_d, gen_fff, 1, gen_helper_fmaxa_d)
+TRANS(fmina_s, gen_fff, 0, gen_helper_fmina_s)
+TRANS(fmina_d, gen_fff, 1, gen_helper_fmina_d)
+TRANS(fscaleb_s, gen_fff, 0, gen_helper_fscaleb_s)
+TRANS(fscaleb_d, gen_fff, 1, gen_helper_fscaleb_d)
+TRANS(fsqrt_s, gen_ff, 0, gen_helper_fsqrt_s)
+TRANS(fsqrt_d, gen_ff, 1, gen_helper_fsqrt_d)
+TRANS(frecip_s, gen_ff, 0, gen_helper_frecip_s)
+TRANS(frecip_d, gen_ff, 1, gen_helper_frecip_d)
+TRANS(frsqrt_s, gen_ff, 0, gen_helper_frsqrt_s)
+TRANS(frsqrt_d, gen_ff, 1, gen_helper_frsqrt_d)
+TRANS(flogb_s, gen_ff, 0, gen_helper_flogb_s)
+TRANS(flogb_d, gen_ff, 1, gen_helper_flogb_d)
+TRANS(fclass_s, gen_ff, 0, gen_helper_fclass_s)
+TRANS(fclass_d, gen_ff, 1, gen_helper_fclass_d)
+TRANS(fmadd_s, gen_muladd, 0, gen_helper_fmuladd_s, 0)
+TRANS(fmadd_d, gen_muladd, 1, gen_helper_fmuladd_d, 0)
+TRANS(fmsub_s, gen_muladd, 0, gen_helper_fmuladd_s, float_muladd_negate_c)
+TRANS(fmsub_d, gen_muladd, 1, gen_helper_fmuladd_d, float_muladd_negate_c)
+TRANS(fnmadd_s, gen_muladd, 0, gen_helper_fmuladd_s, float_muladd_negate_result)
+TRANS(fnmadd_d, gen_muladd, 1, gen_helper_fmuladd_d, float_muladd_negate_result)
+TRANS(fnmsub_s, gen_muladd, 0, gen_helper_fmuladd_s,
       float_muladd_negate_c | float_muladd_negate_result)
-TRANS(fnmsub_d, gen_muladd, gen_helper_fmuladd_d,
+TRANS(fnmsub_d, gen_muladd, 1, gen_helper_fmuladd_d,
       float_muladd_negate_c | float_muladd_negate_result)
diff --git a/target/loongarch/insn_trans/trans_fcmp.c.inc b/target/loongarch/insn_trans/trans_fcmp.c.inc
index a78868dbc4..17ee28a1a2 100644
--- a/target/loongarch/insn_trans/trans_fcmp.c.inc
+++ b/target/loongarch/insn_trans/trans_fcmp.c.inc
@@ -29,6 +29,8 @@ static bool trans_fcmp_cond_s(DisasContext *ctx, arg_fcmp_cond_s *a)
     uint32_t flags;
     void (*fn)(TCGv, TCGv_env, TCGv, TCGv, TCGv_i32);
 
+    REQUIRE_FP;
+    REQUIRE_FP_SP;
     CHECK_FPE;
 
     var = tcg_temp_new();
@@ -49,6 +51,8 @@ static bool trans_fcmp_cond_d(DisasContext *ctx, arg_fcmp_cond_d *a)
     uint32_t flags;
     void (*fn)(TCGv, TCGv_env, TCGv, TCGv, TCGv_i32);
 
+    REQUIRE_FP;
+    REQUIRE_FP_DP;
     CHECK_FPE;
 
     var = tcg_temp_new();
diff --git a/target/loongarch/insn_trans/trans_fcnv.c.inc b/target/loongarch/insn_trans/trans_fcnv.c.inc
index c1c6918ad1..db33abcc32 100644
--- a/target/loongarch/insn_trans/trans_fcnv.c.inc
+++ b/target/loongarch/insn_trans/trans_fcnv.c.inc
@@ -3,31 +3,31 @@
  * Copyright (c) 2021 Loongson Technology Corporation Limited
  */
 
-TRANS(fcvt_s_d, gen_ff, gen_helper_fcvt_s_d)
-TRANS(fcvt_d_s, gen_ff, gen_helper_fcvt_d_s)
-TRANS(ftintrm_w_s, gen_ff, gen_helper_ftintrm_w_s)
-TRANS(ftintrm_w_d, gen_ff, gen_helper_ftintrm_w_d)
-TRANS(ftintrm_l_s, gen_ff, gen_helper_ftintrm_l_s)
-TRANS(ftintrm_l_d, gen_ff, gen_helper_ftintrm_l_d)
-TRANS(ftintrp_w_s, gen_ff, gen_helper_ftintrp_w_s)
-TRANS(ftintrp_w_d, gen_ff, gen_helper_ftintrp_w_d)
-TRANS(ftintrp_l_s, gen_ff, gen_helper_ftintrp_l_s)
-TRANS(ftintrp_l_d, gen_ff, gen_helper_ftintrp_l_d)
-TRANS(ftintrz_w_s, gen_ff, gen_helper_ftintrz_w_s)
-TRANS(ftintrz_w_d, gen_ff, gen_helper_ftintrz_w_d)
-TRANS(ftintrz_l_s, gen_ff, gen_helper_ftintrz_l_s)
-TRANS(ftintrz_l_d, gen_ff, gen_helper_ftintrz_l_d)
-TRANS(ftintrne_w_s, gen_ff, gen_helper_ftintrne_w_s)
-TRANS(ftintrne_w_d, gen_ff, gen_helper_ftintrne_w_d)
-TRANS(ftintrne_l_s, gen_ff, gen_helper_ftintrne_l_s)
-TRANS(ftintrne_l_d, gen_ff, gen_helper_ftintrne_l_d)
-TRANS(ftint_w_s, gen_ff, gen_helper_ftint_w_s)
-TRANS(ftint_w_d, gen_ff, gen_helper_ftint_w_d)
-TRANS(ftint_l_s, gen_ff, gen_helper_ftint_l_s)
-TRANS(ftint_l_d, gen_ff, gen_helper_ftint_l_d)
-TRANS(ffint_s_w, gen_ff, gen_helper_ffint_s_w)
-TRANS(ffint_s_l, gen_ff, gen_helper_ffint_s_l)
-TRANS(ffint_d_w, gen_ff, gen_helper_ffint_d_w)
-TRANS(ffint_d_l, gen_ff, gen_helper_ffint_d_l)
-TRANS(frint_s, gen_ff, gen_helper_frint_s)
-TRANS(frint_d, gen_ff, gen_helper_frint_d)
+TRANS(fcvt_s_d, gen_ff, 0, gen_helper_fcvt_s_d)
+TRANS(fcvt_d_s, gen_ff, 1, gen_helper_fcvt_d_s)
+TRANS(ftintrm_w_s, gen_ff, 0, gen_helper_ftintrm_w_s)
+TRANS(ftintrm_w_d, gen_ff, 1, gen_helper_ftintrm_w_d)
+TRANS(ftintrm_l_s, gen_ff, 0, gen_helper_ftintrm_l_s)
+TRANS(ftintrm_l_d, gen_ff, 1, gen_helper_ftintrm_l_d)
+TRANS(ftintrp_w_s, gen_ff, 0, gen_helper_ftintrp_w_s)
+TRANS(ftintrp_w_d, gen_ff, 1, gen_helper_ftintrp_w_d)
+TRANS(ftintrp_l_s, gen_ff, 0, gen_helper_ftintrp_l_s)
+TRANS(ftintrp_l_d, gen_ff, 1, gen_helper_ftintrp_l_d)
+TRANS(ftintrz_w_s, gen_ff, 0, gen_helper_ftintrz_w_s)
+TRANS(ftintrz_w_d, gen_ff, 1, gen_helper_ftintrz_w_d)
+TRANS(ftintrz_l_s, gen_ff, 0, gen_helper_ftintrz_l_s)
+TRANS(ftintrz_l_d, gen_ff, 1, gen_helper_ftintrz_l_d)
+TRANS(ftintrne_w_s, gen_ff, 0, gen_helper_ftintrne_w_s)
+TRANS(ftintrne_w_d, gen_ff, 1, gen_helper_ftintrne_w_d)
+TRANS(ftintrne_l_s, gen_ff, 0, gen_helper_ftintrne_l_s)
+TRANS(ftintrne_l_d, gen_ff, 1, gen_helper_ftintrne_l_d)
+TRANS(ftint_w_s, gen_ff, 0, gen_helper_ftint_w_s)
+TRANS(ftint_w_d, gen_ff, 1, gen_helper_ftint_w_d)
+TRANS(ftint_l_s, gen_ff, 0, gen_helper_ftint_l_s)
+TRANS(ftint_l_d, gen_ff, 1, gen_helper_ftint_l_d)
+TRANS(ffint_s_w, gen_ff, 0, gen_helper_ffint_s_w)
+TRANS(ffint_s_l, gen_ff, 1, gen_helper_ffint_s_l)
+TRANS(ffint_d_w, gen_ff, 0, gen_helper_ffint_d_w)
+TRANS(ffint_d_l, gen_ff, 1, gen_helper_ffint_d_l)
+TRANS(frint_s, gen_ff, 0, gen_helper_frint_s)
+TRANS(frint_d, gen_ff, 1, gen_helper_frint_d)
diff --git a/target/loongarch/insn_trans/trans_fmemory.c.inc b/target/loongarch/insn_trans/trans_fmemory.c.inc
index bd3aba2c49..b8cf3f63b2 100644
--- a/target/loongarch/insn_trans/trans_fmemory.c.inc
+++ b/target/loongarch/insn_trans/trans_fmemory.c.inc
@@ -10,11 +10,18 @@ static void maybe_nanbox_load(TCGv freg, MemOp mop)
     }
 }
 
-static bool gen_fload_i(DisasContext *ctx, arg_fr_i *a, MemOp mop)
+static bool gen_fload_i(DisasContext *ctx, arg_fr_i *a, int mode, MemOp mop)
 {
     TCGv addr = gpr_src(ctx, a->rj, EXT_NONE);
     TCGv dest = get_fpr(ctx, a->fd);
 
+    REQUIRE_FP;
+    if (mode == 0) {
+        REQUIRE_FP_SP;
+    } else {
+        REQUIRE_FP_DP;
+    }
+
     CHECK_FPE;
 
     addr = make_address_i(ctx, addr, a->imm);
@@ -26,11 +33,18 @@ static bool gen_fload_i(DisasContext *ctx, arg_fr_i *a, MemOp mop)
     return true;
 }
 
-static bool gen_fstore_i(DisasContext *ctx, arg_fr_i *a, MemOp mop)
+static bool gen_fstore_i(DisasContext *ctx, arg_fr_i *a, int mode,  MemOp mop)
 {
     TCGv addr = gpr_src(ctx, a->rj, EXT_NONE);
     TCGv src = get_fpr(ctx, a->fd);
 
+    REQUIRE_FP;
+    if (mode == 0) {
+        REQUIRE_FP_SP;
+    } else {
+        REQUIRE_FP_DP;
+    }
+
     CHECK_FPE;
 
     addr = make_address_i(ctx, addr, a->imm);
@@ -40,13 +54,20 @@ static bool gen_fstore_i(DisasContext *ctx, arg_fr_i *a, MemOp mop)
     return true;
 }
 
-static bool gen_floadx(DisasContext *ctx, arg_frr *a, MemOp mop)
+static bool gen_floadx(DisasContext *ctx, arg_frr *a, int mode, MemOp mop)
 {
     TCGv src1 = gpr_src(ctx, a->rj, EXT_NONE);
     TCGv src2 = gpr_src(ctx, a->rk, EXT_NONE);
     TCGv dest = get_fpr(ctx, a->fd);
     TCGv addr;
 
+    REQUIRE_FP;
+    if (mode == 0) {
+        REQUIRE_FP_SP;
+    } else {
+        REQUIRE_FP_DP;
+    }
+
     CHECK_FPE;
 
     addr = make_address_x(ctx, src1, src2);
@@ -57,13 +78,20 @@ static bool gen_floadx(DisasContext *ctx, arg_frr *a, MemOp mop)
     return true;
 }
 
-static bool gen_fstorex(DisasContext *ctx, arg_frr *a, MemOp mop)
+static bool gen_fstorex(DisasContext *ctx, arg_frr *a, int mode, MemOp mop)
 {
     TCGv src1 = gpr_src(ctx, a->rj, EXT_NONE);
     TCGv src2 = gpr_src(ctx, a->rk, EXT_NONE);
     TCGv src3 = get_fpr(ctx, a->fd);
     TCGv addr;
 
+    REQUIRE_FP;
+    if (mode == 0) {
+        REQUIRE_FP_SP;
+    } else {
+        REQUIRE_FP_DP;
+    }
+
     CHECK_FPE;
 
     addr = make_address_x(ctx, src1, src2);
@@ -72,13 +100,20 @@ static bool gen_fstorex(DisasContext *ctx, arg_frr *a, MemOp mop)
     return true;
 }
 
-static bool gen_fload_gt(DisasContext *ctx, arg_frr *a, MemOp mop)
+static bool gen_fload_gt(DisasContext *ctx, arg_frr *a, int mode, MemOp mop)
 {
     TCGv src1 = gpr_src(ctx, a->rj, EXT_NONE);
     TCGv src2 = gpr_src(ctx, a->rk, EXT_NONE);
     TCGv dest = get_fpr(ctx, a->fd);
     TCGv addr;
 
+    REQUIRE_FP;
+    if (mode == 0) {
+        REQUIRE_FP_SP;
+    } else {
+        REQUIRE_FP_DP;
+    }
+
     CHECK_FPE;
 
     gen_helper_asrtgt_d(cpu_env, src1, src2);
@@ -90,13 +125,20 @@ static bool gen_fload_gt(DisasContext *ctx, arg_frr *a, MemOp mop)
     return true;
 }
 
-static bool gen_fstore_gt(DisasContext *ctx, arg_frr *a, MemOp mop)
+static bool gen_fstore_gt(DisasContext *ctx, arg_frr *a, int mode, MemOp mop)
 {
     TCGv src1 = gpr_src(ctx, a->rj, EXT_NONE);
     TCGv src2 = gpr_src(ctx, a->rk, EXT_NONE);
     TCGv src3 = get_fpr(ctx, a->fd);
     TCGv addr;
 
+    REQUIRE_FP;
+    if (mode == 0) {
+        REQUIRE_FP_SP;
+    } else {
+        REQUIRE_FP_DP;
+    }
+
     CHECK_FPE;
 
     gen_helper_asrtgt_d(cpu_env, src1, src2);
@@ -106,13 +148,20 @@ static bool gen_fstore_gt(DisasContext *ctx, arg_frr *a, MemOp mop)
     return true;
 }
 
-static bool gen_fload_le(DisasContext *ctx, arg_frr *a, MemOp mop)
+static bool gen_fload_le(DisasContext *ctx, arg_frr *a, int mode, MemOp mop)
 {
     TCGv src1 = gpr_src(ctx, a->rj, EXT_NONE);
     TCGv src2 = gpr_src(ctx, a->rk, EXT_NONE);
     TCGv dest = get_fpr(ctx, a->fd);
     TCGv addr;
 
+    REQUIRE_FP;
+    if (mode == 0) {
+        REQUIRE_FP_SP;
+    } else {
+        REQUIRE_FP_DP;
+    }
+
     CHECK_FPE;
 
     gen_helper_asrtle_d(cpu_env, src1, src2);
@@ -124,13 +173,20 @@ static bool gen_fload_le(DisasContext *ctx, arg_frr *a, MemOp mop)
     return true;
 }
 
-static bool gen_fstore_le(DisasContext *ctx, arg_frr *a, MemOp mop)
+static bool gen_fstore_le(DisasContext *ctx, arg_frr *a, int mode, MemOp mop)
 {
     TCGv src1 = gpr_src(ctx, a->rj, EXT_NONE);
     TCGv src2 = gpr_src(ctx, a->rk, EXT_NONE);
     TCGv src3 = get_fpr(ctx, a->fd);
     TCGv addr;
 
+    REQUIRE_FP;
+    if (mode == 0) {
+        REQUIRE_FP_SP;
+    } else {
+        REQUIRE_FP_DP;
+    }
+
     CHECK_FPE;
 
     gen_helper_asrtle_d(cpu_env, src1, src2);
@@ -140,19 +196,19 @@ static bool gen_fstore_le(DisasContext *ctx, arg_frr *a, MemOp mop)
     return true;
 }
 
-TRANS(fld_s, gen_fload_i, MO_TEUL)
-TRANS(fst_s, gen_fstore_i, MO_TEUL)
-TRANS(fld_d, gen_fload_i, MO_TEUQ)
-TRANS(fst_d, gen_fstore_i, MO_TEUQ)
-TRANS(fldx_s, gen_floadx, MO_TEUL)
-TRANS(fldx_d, gen_floadx, MO_TEUQ)
-TRANS(fstx_s, gen_fstorex, MO_TEUL)
-TRANS(fstx_d, gen_fstorex, MO_TEUQ)
-TRANS(fldgt_s, gen_fload_gt, MO_TEUL)
-TRANS(fldgt_d, gen_fload_gt, MO_TEUQ)
-TRANS(fldle_s, gen_fload_le, MO_TEUL)
-TRANS(fldle_d, gen_fload_le, MO_TEUQ)
-TRANS(fstgt_s, gen_fstore_gt, MO_TEUL)
-TRANS(fstgt_d, gen_fstore_gt, MO_TEUQ)
-TRANS(fstle_s, gen_fstore_le, MO_TEUL)
-TRANS(fstle_d, gen_fstore_le, MO_TEUQ)
+TRANS(fld_s, gen_fload_i, 0, MO_TEUL)
+TRANS(fst_s, gen_fstore_i, 0, MO_TEUL)
+TRANS(fld_d, gen_fload_i, 1, MO_TEUQ)
+TRANS(fst_d, gen_fstore_i, 1, MO_TEUQ)
+TRANS(fldx_s, gen_floadx, 0, MO_TEUL)
+TRANS(fldx_d, gen_floadx, 1, MO_TEUQ)
+TRANS(fstx_s, gen_fstorex, 0, MO_TEUL)
+TRANS(fstx_d, gen_fstorex, 1, MO_TEUQ)
+TRANS(fldgt_s, gen_fload_gt, 0, MO_TEUL)
+TRANS(fldgt_d, gen_fload_gt, 1, MO_TEUQ)
+TRANS(fldle_s, gen_fload_le, 0, MO_TEUL)
+TRANS(fldle_d, gen_fload_le, 1, MO_TEUQ)
+TRANS(fstgt_s, gen_fstore_gt, 0, MO_TEUL)
+TRANS(fstgt_d, gen_fstore_gt, 1, MO_TEUQ)
+TRANS(fstle_s, gen_fstore_le, 0, MO_TEUL)
+TRANS(fstle_d, gen_fstore_le, 1, MO_TEUQ)
diff --git a/target/loongarch/insn_trans/trans_fmov.c.inc b/target/loongarch/insn_trans/trans_fmov.c.inc
index c58c5c6534..09a7b5bc60 100644
--- a/target/loongarch/insn_trans/trans_fmov.c.inc
+++ b/target/loongarch/insn_trans/trans_fmov.c.inc
@@ -15,6 +15,7 @@ static bool trans_fsel(DisasContext *ctx, arg_fsel *a)
     TCGv src2 = get_fpr(ctx, a->fk);
     TCGv cond;
 
+    REQUIRE_FP;
     CHECK_FPE;
 
     cond = tcg_temp_new();
@@ -25,12 +26,20 @@ static bool trans_fsel(DisasContext *ctx, arg_fsel *a)
     return true;
 }
 
-static bool gen_f2f(DisasContext *ctx, arg_ff *a,
+static bool gen_f2f(DisasContext *ctx, arg_ff *a, int mode,
                     void (*func)(TCGv, TCGv), bool nanbox)
 {
     TCGv dest = get_fpr(ctx, a->fd);
     TCGv src = get_fpr(ctx, a->fj);
 
+
+    REQUIRE_FP;
+    if (mode == 0) {
+        REQUIRE_FP_SP;
+    } else {
+        REQUIRE_FP_DP;
+    }
+
     CHECK_FPE;
 
     func(dest, src);
@@ -42,12 +51,17 @@ static bool gen_f2f(DisasContext *ctx, arg_ff *a,
     return true;
 }
 
-static bool gen_r2f(DisasContext *ctx, arg_fr *a,
+static bool gen_r2f(DisasContext *ctx, arg_fr *a, int mode,
                     void (*func)(TCGv, TCGv))
 {
     TCGv src = gpr_src(ctx, a->rj, EXT_NONE);
     TCGv dest = get_fpr(ctx, a->fd);
 
+    REQUIRE_FP;
+    if (mode == 1) {
+        REQUIRE_FP_DP;
+    }
+
     CHECK_FPE;
 
     func(dest, src);
@@ -56,12 +70,17 @@ static bool gen_r2f(DisasContext *ctx, arg_fr *a,
     return true;
 }
 
-static bool gen_f2r(DisasContext *ctx, arg_rf *a,
+static bool gen_f2r(DisasContext *ctx, arg_rf *a, int mode,
                     void (*func)(TCGv, TCGv))
 {
     TCGv dest = gpr_dst(ctx, a->rd, EXT_NONE);
     TCGv src = get_fpr(ctx, a->fj);
 
+    REQUIRE_FP;
+    if (mode == 1) {
+        REQUIRE_FP_DP;
+    }
+
     CHECK_FPE;
 
     func(dest, src);
@@ -75,6 +94,7 @@ static bool trans_movgr2fcsr(DisasContext *ctx, arg_movgr2fcsr *a)
     uint32_t mask = fcsr_mask[a->fcsrd];
     TCGv Rj = gpr_src(ctx, a->rj, EXT_NONE);
 
+    REQUIRE_FP;
     CHECK_FPE;
 
     if (mask == UINT32_MAX) {
@@ -105,6 +125,7 @@ static bool trans_movfcsr2gr(DisasContext *ctx, arg_movfcsr2gr *a)
 {
     TCGv dest = gpr_dst(ctx, a->rd, EXT_NONE);
 
+    REQUIRE_FP;
     CHECK_FPE;
 
     tcg_gen_ld32u_i64(dest, cpu_env, offsetof(CPULoongArchState, fcsr0));
@@ -134,6 +155,7 @@ static bool trans_movfr2cf(DisasContext *ctx, arg_movfr2cf *a)
     TCGv t0;
     TCGv src = get_fpr(ctx, a->fj);
 
+    REQUIRE_FP;
     CHECK_FPE;
 
     t0 = tcg_temp_new();
@@ -147,6 +169,7 @@ static bool trans_movcf2fr(DisasContext *ctx, arg_movcf2fr *a)
 {
     TCGv dest = get_fpr(ctx, a->fd);
 
+    REQUIRE_FP;
     CHECK_FPE;
 
     tcg_gen_ld8u_tl(dest, cpu_env,
@@ -160,6 +183,7 @@ static bool trans_movgr2cf(DisasContext *ctx, arg_movgr2cf *a)
 {
     TCGv t0;
 
+    REQUIRE_FP;
     CHECK_FPE;
 
     t0 = tcg_temp_new();
@@ -171,6 +195,7 @@ static bool trans_movgr2cf(DisasContext *ctx, arg_movgr2cf *a)
 
 static bool trans_movcf2gr(DisasContext *ctx, arg_movcf2gr *a)
 {
+    REQUIRE_FP;
     CHECK_FPE;
 
     tcg_gen_ld8u_tl(gpr_dst(ctx, a->rd, EXT_NONE), cpu_env,
@@ -178,11 +203,11 @@ static bool trans_movcf2gr(DisasContext *ctx, arg_movcf2gr *a)
     return true;
 }
 
-TRANS(fmov_s, gen_f2f, tcg_gen_mov_tl, true)
-TRANS(fmov_d, gen_f2f, tcg_gen_mov_tl, false)
-TRANS(movgr2fr_w, gen_r2f, gen_movgr2fr_w)
-TRANS_64(movgr2fr_d, gen_r2f, tcg_gen_mov_tl)
-TRANS(movgr2frh_w, gen_r2f, gen_movgr2frh_w)
-TRANS(movfr2gr_s, gen_f2r, tcg_gen_ext32s_tl)
-TRANS_64(movfr2gr_d, gen_f2r, tcg_gen_mov_tl)
-TRANS(movfrh2gr_s, gen_f2r, gen_movfrh2gr_s)
+TRANS(fmov_s, gen_f2f, 0, tcg_gen_mov_tl, true)
+TRANS(fmov_d, gen_f2f, 1, tcg_gen_mov_tl, false)
+TRANS(movgr2fr_w, gen_r2f, 0, gen_movgr2fr_w)
+TRANS_64(movgr2fr_d, gen_r2f, 1, tcg_gen_mov_tl)
+TRANS(movgr2frh_w, gen_r2f, 0, gen_movgr2frh_w)
+TRANS(movfr2gr_s, gen_f2r, 0, tcg_gen_ext32s_tl)
+TRANS_64(movfr2gr_d, gen_f2r, 1, tcg_gen_mov_tl)
+TRANS(movfrh2gr_s, gen_f2r, 0, gen_movfrh2gr_s)
-- 
2.39.1

On 8/10/23 05:41, Song Gao wrote:
> This patch adds REQUIRE_FP/FP_SP/FP_DP to check CPUCFG2.FP/FP_SP/FP_DP.
> 
> Signed-off-by: Song Gao <gaosong@loongson.cn>
> ---
>   target/loongarch/cpu.h                        |   6 +
>   .../loongarch/insn_trans/trans_farith.c.inc   | 132 ++++++++++++------
>   target/loongarch/insn_trans/trans_fcmp.c.inc  |   4 +
>   target/loongarch/insn_trans/trans_fcnv.c.inc  |  56 ++++----
>   .../loongarch/insn_trans/trans_fmemory.c.inc  | 104 ++++++++++----
>   target/loongarch/insn_trans/trans_fmov.c.inc  |  47 +++++--
>   6 files changed, 247 insertions(+), 102 deletions(-)
> 
> diff --git a/target/loongarch/cpu.h b/target/loongarch/cpu.h
> index 9f550793ca..5594d83011 100644
> --- a/target/loongarch/cpu.h
> +++ b/target/loongarch/cpu.h
> @@ -459,6 +459,9 @@ static inline void set_pc(CPULoongArchState *env, uint64_t value)
>   #define HW_FLAGS_EUEN_FPE   0x04
>   #define HW_FLAGS_EUEN_SXE   0x08
>   #define HW_FLAGS_VA32       0x20
> +#define HW_FLAGS_FP         0x40
> +#define HW_FLAGS_FP_SP      0x80
> +#define HW_FLAGS_FP_DP      0x100
>   
>   static inline void cpu_get_tb_cpu_state(CPULoongArchState *env, vaddr *pc,
>                                           uint64_t *cs_base, uint32_t *flags)
> @@ -469,6 +472,9 @@ static inline void cpu_get_tb_cpu_state(CPULoongArchState *env, vaddr *pc,
>       *flags |= FIELD_EX64(env->CSR_EUEN, CSR_EUEN, FPE) * HW_FLAGS_EUEN_FPE;
>       *flags |= FIELD_EX64(env->CSR_EUEN, CSR_EUEN, SXE) * HW_FLAGS_EUEN_SXE;
>       *flags |= is_va32(env) * HW_FLAGS_VA32;
> +    *flags |= FIELD_EX32(env->cpucfg[2], CPUCFG2, FP) * HW_FLAGS_FP;
> +    *flags |= FIELD_EX32(env->cpucfg[2], CPUCFG2, FP_SP) * HW_FLAGS_FP_SP;
> +    *flags |= FIELD_EX32(env->cpucfg[2], CPUCFG2, FP_DP) * HW_FLAGS_FP_DP;

You do not need to put any of these in HW_FLAGS, because CPUCFG space never changes for 
the lifetime of the cpu.

You can extract these into DisasContext in loongarch_tr_init_disas_context.

> +#define REQUIRE_FP do { \
> +    if ((ctx->base.tb->flags & HW_FLAGS_FP) == 0) { \
> +        return false; \
> +    } \
> +} while (0)
> +
> +#define REQUIRE_FP_SP do { \
> +    if ((ctx->base.tb->flags & HW_FLAGS_FP_SP) == 0) { \
> +        return false; \
> +    } \
> +} while (0)
> +
> +#define REQUIRE_FP_DP do { \
> +    if ((ctx->base.tb->flags & HW_FLAGS_FP_DP) == 0) { \
> +        return false; \
> +    } \
> +} while (0)

It would be much better to not create so many of these REQUIRE macros.

> +TRANS(fadd_s, gen_fff, 0, gen_helper_fadd_s)
> +TRANS(fadd_d, gen_fff, 1, gen_helper_fadd_d)

0 vs 1 is very opaque.

Better is something like Jiajie Chen's TRANS_64,

> +/* for LoongArch64-only instructions */
> +#define TRANS_64(NAME, FUNC, ...) \
> +    static bool trans_##NAME(DisasContext *ctx, arg_##NAME * a) \
> +    { \
> +        return ctx->la64 && FUNC(ctx, a, __VA_ARGS__); \
> +    }

But as we now know, we would need at least 7 of these.

Even better would be to generalize this so that every instruction records the condition 
under which it is valid.

Perhaps

typedef struct DisasContext {
     ...
     uint32_t cpucfg1;
     uint32_t cpucfg2;
};

static void loongarch_tr_init_disas_context(...)
{
     ...
     ctx->cpucfg1 = env->cpucfg[1];
     ctx->cpucfg2 = env->cpucfg[2];
}

#define avail_ALL(C)  true
#define avail_64(C)   FIELD_EX32((C)->cpucfg1, CPUCFG1, ARCH) == CPUCFG1_ARCH_LA64
#define avail_FP(C)   FIELD_EX32((C)->cpucfg2, CPUCFG2, FP)
etc


#define TRANS(NAME, AVAIL, FUNC, ...) \
     static bool trans_##NAME(DisasContext *ctx, arg_##NAME *a)  \
     { return avail_##AVAIL(ctx) && FUNC(ctx, a, __VA_ARGS__); }

so that the above becomes

TRANS(fadd_s, FP_SP, gen_fff, gen_helper_fadd_s)
TRANS(fadd_d, FP_DP, gen_fff, gen_helper_fadd_d)

and even simple instructions get

TRANS(add_w, ALL, gen_rrr, EXT_NONE, EXT_NONE, EXT_SIGN, tcg_gen_add_tl)
TRANS(add_d,  64, gen_rrr, EXT_NONE, EXT_NONE, EXT_SIGN, tcg_gen_add_tl)


r~

在 2023/8/10 下午11:03, Richard Henderson 写道:
> On 8/10/23 05:41, Song Gao wrote:
>> This patch adds REQUIRE_FP/FP_SP/FP_DP to check CPUCFG2.FP/FP_SP/FP_DP.
>>
>> Signed-off-by: Song Gao <gaosong@loongson.cn>
>> ---
>>   target/loongarch/cpu.h                        |   6 +
>>   .../loongarch/insn_trans/trans_farith.c.inc   | 132 ++++++++++++------
>>   target/loongarch/insn_trans/trans_fcmp.c.inc  |   4 +
>>   target/loongarch/insn_trans/trans_fcnv.c.inc  |  56 ++++----
>>   .../loongarch/insn_trans/trans_fmemory.c.inc  | 104 ++++++++++----
>>   target/loongarch/insn_trans/trans_fmov.c.inc  |  47 +++++--
>>   6 files changed, 247 insertions(+), 102 deletions(-)
>>
>> diff --git a/target/loongarch/cpu.h b/target/loongarch/cpu.h
>> index 9f550793ca..5594d83011 100644
>> --- a/target/loongarch/cpu.h
>> +++ b/target/loongarch/cpu.h
>> @@ -459,6 +459,9 @@ static inline void set_pc(CPULoongArchState *env, 
>> uint64_t value)
>>   #define HW_FLAGS_EUEN_FPE   0x04
>>   #define HW_FLAGS_EUEN_SXE   0x08
>>   #define HW_FLAGS_VA32       0x20
>> +#define HW_FLAGS_FP         0x40
>> +#define HW_FLAGS_FP_SP      0x80
>> +#define HW_FLAGS_FP_DP      0x100
>>   static inline void cpu_get_tb_cpu_state(CPULoongArchState *env, 
>> vaddr *pc,
>>                                           uint64_t *cs_base, uint32_t 
>> *flags)
>> @@ -469,6 +472,9 @@ static inline void 
>> cpu_get_tb_cpu_state(CPULoongArchState *env, vaddr *pc,
>>       *flags |= FIELD_EX64(env->CSR_EUEN, CSR_EUEN, FPE) * 
>> HW_FLAGS_EUEN_FPE;
>>       *flags |= FIELD_EX64(env->CSR_EUEN, CSR_EUEN, SXE) * 
>> HW_FLAGS_EUEN_SXE;
>>       *flags |= is_va32(env) * HW_FLAGS_VA32;
>> +    *flags |= FIELD_EX32(env->cpucfg[2], CPUCFG2, FP) * HW_FLAGS_FP;
>> +    *flags |= FIELD_EX32(env->cpucfg[2], CPUCFG2, FP_SP) * 
>> HW_FLAGS_FP_SP;
>> +    *flags |= FIELD_EX32(env->cpucfg[2], CPUCFG2, FP_DP) * 
>> HW_FLAGS_FP_DP;
> 
> You do not need to put any of these in HW_FLAGS, because CPUCFG space 
> never changes for the lifetime of the cpu.
> 
> You can extract these into DisasContext in loongarch_tr_init_disas_context.
> 
>> +#define REQUIRE_FP do { \
>> +    if ((ctx->base.tb->flags & HW_FLAGS_FP) == 0) { \
>> +        return false; \
>> +    } \
>> +} while (0)
>> +
>> +#define REQUIRE_FP_SP do { \
>> +    if ((ctx->base.tb->flags & HW_FLAGS_FP_SP) == 0) { \
>> +        return false; \
>> +    } \
>> +} while (0)
>> +
>> +#define REQUIRE_FP_DP do { \
>> +    if ((ctx->base.tb->flags & HW_FLAGS_FP_DP) == 0) { \
>> +        return false; \
>> +    } \
>> +} while (0)
> 
> It would be much better to not create so many of these REQUIRE macros.
> 
>> +TRANS(fadd_s, gen_fff, 0, gen_helper_fadd_s)
>> +TRANS(fadd_d, gen_fff, 1, gen_helper_fadd_d)
> 
> 0 vs 1 is very opaque.
> 
> Better is something like Jiajie Chen's TRANS_64,
> 
>> +/* for LoongArch64-only instructions */
>> +#define TRANS_64(NAME, FUNC, ...) \
>> +    static bool trans_##NAME(DisasContext *ctx, arg_##NAME * a) \
>> +    { \
>> +        return ctx->la64 && FUNC(ctx, a, __VA_ARGS__); \
>> +    }
> 
> But as we now know, we would need at least 7 of these.
> 
> Even better would be to generalize this so that every instruction 
> records the condition under which it is valid.
> 
> Perhaps
> 
> typedef struct DisasContext {
>      ...
>      uint32_t cpucfg1;
>      uint32_t cpucfg2;
> };
> 
> static void loongarch_tr_init_disas_context(...)
> {
>      ...
>      ctx->cpucfg1 = env->cpucfg[1];
>      ctx->cpucfg2 = env->cpucfg[2];
> }
> 
> #define avail_ALL(C)  true
> #define avail_64(C)   FIELD_EX32((C)->cpucfg1, CPUCFG1, ARCH) == 
> CPUCFG1_ARCH_LA64
> #define avail_FP(C)   FIELD_EX32((C)->cpucfg2, CPUCFG2, FP)
> etc
> 
> 
> #define TRANS(NAME, AVAIL, FUNC, ...) \
>      static bool trans_##NAME(DisasContext *ctx, arg_##NAME *a)  \
>      { return avail_##AVAIL(ctx) && FUNC(ctx, a, __VA_ARGS__); }
> 
> so that the above becomes
> 
> TRANS(fadd_s, FP_SP, gen_fff, gen_helper_fadd_s)
> TRANS(fadd_d, FP_DP, gen_fff, gen_helper_fadd_d)
> 
> and even simple instructions get
> 
> TRANS(add_w, ALL, gen_rrr, EXT_NONE, EXT_NONE, EXT_SIGN, tcg_gen_add_tl)
> TRANS(add_d,  64, gen_rrr, EXT_NONE, EXT_NONE, EXT_SIGN, tcg_gen_add_tl)
> 
Thanks for your suggestions, I will send v2 as soon as possblie.

Thanks.
Song Gao