Usually, eliminate an operation from the translator by combining
a shift with an extract.
In the case of gen_set_NZ64, we don't need a boolean value for cpu_ZF,
merely a non-zero value. Given that we can extract both halves of a
64-bit input in one call, this simplifies the code.
Signed-off-by: Richard Henderson <r...@twiddle.net>
---
target-arm/translate-a64.c | 34 +++++++++-------------------------
1 file changed, 9 insertions(+), 25 deletions(-)
diff --git a/target-arm/translate-a64.c b/target-arm/translate-a64.c
index 815ec7d..1c10448 100644
--- a/target-arm/translate-a64.c
+++ b/target-arm/translate-a64.c
@@ -530,13 +530,8 @@ static TCGv_ptr get_fpstatus_ptr(void)
*/
static inline void gen_set_NZ64(TCGv_i64 result)
{
- TCGv_i64 flag = tcg_temp_new_i64();
-
- tcg_gen_setcondi_i64(TCG_COND_NE, flag, result, 0);
- tcg_gen_extrl_i64_i32(cpu_ZF, flag);
- tcg_gen_shri_i64(flag, result, 32);
- tcg_gen_extrl_i64_i32(cpu_NF, flag);
- tcg_temp_free_i64(flag);
+ tcg_gen_extr_i64_i32(cpu_ZF, cpu_NF, result);
+ tcg_gen_or_i32(cpu_ZF, cpu_ZF, cpu_NF);
}
/* Set NZCV as for a logical operation: NZ as per result, CV cleared. */
@@ -546,7 +541,7 @@ static inline void gen_logic_CC(int sf, TCGv_i64 result)
gen_set_NZ64(result);
} else {
tcg_gen_extrl_i64_i32(cpu_ZF, result);
- tcg_gen_extrl_i64_i32(cpu_NF, result);
+ tcg_gen_mov_i32(cpu_NF, cpu_ZF);
}
tcg_gen_movi_i32(cpu_CF, 0);
tcg_gen_movi_i32(cpu_VF, 0);
@@ -572,8 +567,7 @@ static void gen_add_CC(int sf, TCGv_i64 dest, TCGv_i64 t0,
TCGv_i64 t1)
tcg_gen_xor_i64(tmp, t0, t1);
tcg_gen_andc_i64(flag, flag, tmp);
tcg_temp_free_i64(tmp);
- tcg_gen_shri_i64(flag, flag, 32);
- tcg_gen_extrl_i64_i32(cpu_VF, flag);
+ tcg_gen_extrh_i64_i32(cpu_VF, flag);
tcg_gen_mov_i64(dest, result);
tcg_temp_free_i64(result);
@@ -621,8 +615,7 @@ static void gen_sub_CC(int sf, TCGv_i64 dest, TCGv_i64 t0,
TCGv_i64 t1)
tcg_gen_xor_i64(tmp, t0, t1);
tcg_gen_and_i64(flag, flag, tmp);
tcg_temp_free_i64(tmp);
- tcg_gen_shri_i64(flag, flag, 32);
- tcg_gen_extrl_i64_i32(cpu_VF, flag);
+ tcg_gen_extrh_i64_i32(cpu_VF, flag);
tcg_gen_mov_i64(dest, result);
tcg_temp_free_i64(flag);
tcg_temp_free_i64(result);
@@ -681,8 +674,7 @@ static void gen_adc_CC(int sf, TCGv_i64 dest, TCGv_i64 t0,
TCGv_i64 t1)
tcg_gen_xor_i64(vf_64, result, t0);
tcg_gen_xor_i64(tmp, t0, t1);
tcg_gen_andc_i64(vf_64, vf_64, tmp);
- tcg_gen_shri_i64(vf_64, vf_64, 32);
- tcg_gen_extrl_i64_i32(cpu_VF, vf_64);
+ tcg_gen_extrh_i64_i32(cpu_VF, vf_64);
tcg_gen_mov_i64(dest, result);
@@ -7743,10 +7735,8 @@ static void handle_2misc_narrow(DisasContext *s, bool
scalar,
} else {
TCGv_i32 tcg_lo = tcg_temp_new_i32();
TCGv_i32 tcg_hi = tcg_temp_new_i32();
- tcg_gen_extrl_i64_i32(tcg_lo, tcg_op);
+ tcg_gen_extr_i64_i32(tcg_lo, tcg_hi, tcg_op);
gen_helper_vfp_fcvt_f32_to_f16(tcg_lo, tcg_lo, cpu_env);
- tcg_gen_shri_i64(tcg_op, tcg_op, 32);
- tcg_gen_extrl_i64_i32(tcg_hi, tcg_op);
gen_helper_vfp_fcvt_f32_to_f16(tcg_hi, tcg_hi, cpu_env);
tcg_gen_deposit_i32(tcg_res[pass], tcg_lo, tcg_hi, 16, 16);
tcg_temp_free_i32(tcg_lo);
@@ -8652,16 +8642,10 @@ static void handle_3rd_wide(DisasContext *s, int is_q,
int is_u, int size,
}
}
-static void do_narrow_high_u32(TCGv_i32 res, TCGv_i64 in)
-{
- tcg_gen_shri_i64(in, in, 32);
- tcg_gen_extrl_i64_i32(res, in);
-}
-
static void do_narrow_round_high_u32(TCGv_i32 res, TCGv_i64 in)
{
tcg_gen_addi_i64(in, in, 1U << 31);
- do_narrow_high_u32(res, in);
+ tcg_gen_extrh_i64_i32(res, in);
}
static void handle_3rd_narrowing(DisasContext *s, int is_q, int is_u, int size,
@@ -8680,7 +8664,7 @@ static void handle_3rd_narrowing(DisasContext *s, int
is_q, int is_u, int size,
gen_helper_neon_narrow_round_high_u8 },
{ gen_helper_neon_narrow_high_u16,
gen_helper_neon_narrow_round_high_u16 },
- { do_narrow_high_u32, do_narrow_round_high_u32 },
+ { tcg_gen_extrh_i64_i32, do_narrow_round_high_u32 },
};
NeonGenNarrowFn *gennarrow = narrowfns[size][is_u];
--
2.4.3
