[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [PATCH v3 07/69] target/arm: Simplify UMAAL
From: |
Richard Henderson |
Subject: |
[Qemu-devel] [PATCH v3 07/69] target/arm: Simplify UMAAL |
Date: |
Wed, 28 Aug 2019 12:03:54 -0700 |
Since all of the inputs and outputs are i32, dispense with
the intermediate promotion to i64 and use tcg_gen_mulu2_i32
and tcg_gen_add2_i32.
Reviewed-by: Peter Maydell <address@hidden>
Signed-off-by: Richard Henderson <address@hidden>
---
target/arm/translate.c | 34 ++++++++++++----------------------
1 file changed, 12 insertions(+), 22 deletions(-)
diff --git a/target/arm/translate.c b/target/arm/translate.c
index b7845d825b..17659627b1 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -7320,21 +7320,6 @@ static void gen_storeq_reg(DisasContext *s, int rlow,
int rhigh, TCGv_i64 val)
store_reg(s, rhigh, tmp);
}
-/* load a 32-bit value from a register and perform a 64-bit accumulate. */
-static void gen_addq_lo(DisasContext *s, TCGv_i64 val, int rlow)
-{
- TCGv_i64 tmp;
- TCGv_i32 tmp2;
-
- /* Load value and extend to 64 bits. */
- tmp = tcg_temp_new_i64();
- tmp2 = load_reg(s, rlow);
- tcg_gen_extu_i32_i64(tmp, tmp2);
- tcg_temp_free_i32(tmp2);
- tcg_gen_add_i64(val, val, tmp);
- tcg_temp_free_i64(tmp);
-}
-
/* load and add a 64-bit value from a register pair. */
static void gen_addq(DisasContext *s, TCGv_i64 val, int rlow, int rhigh)
{
@@ -8093,8 +8078,7 @@ static bool trans_SMLAL(DisasContext *s, arg_SMLAL *a)
static bool trans_UMAAL(DisasContext *s, arg_UMAAL *a)
{
- TCGv_i32 t0, t1;
- TCGv_i64 t64;
+ TCGv_i32 t0, t1, t2, zero;
if (s->thumb
? !arm_dc_feature(s, ARM_FEATURE_THUMB_DSP)
@@ -8104,11 +8088,17 @@ static bool trans_UMAAL(DisasContext *s, arg_UMAAL *a)
t0 = load_reg(s, a->rm);
t1 = load_reg(s, a->rn);
- t64 = gen_mulu_i64_i32(t0, t1);
- gen_addq_lo(s, t64, a->ra);
- gen_addq_lo(s, t64, a->rd);
- gen_storeq_reg(s, a->ra, a->rd, t64);
- tcg_temp_free_i64(t64);
+ tcg_gen_mulu2_i32(t0, t1, t0, t1);
+ zero = tcg_const_i32(0);
+ t2 = load_reg(s, a->ra);
+ tcg_gen_add2_i32(t0, t1, t0, t1, t2, zero);
+ tcg_temp_free_i32(t2);
+ t2 = load_reg(s, a->rd);
+ tcg_gen_add2_i32(t0, t1, t0, t1, t2, zero);
+ tcg_temp_free_i32(t2);
+ tcg_temp_free_i32(zero);
+ store_reg(s, a->ra, t0);
+ store_reg(s, a->rd, t1);
return true;
}
--
2.17.1
- [Qemu-devel] [PATCH v3 00/69] target/arm: Convert aa32 base isa to decodetree, Richard Henderson, 2019/08/28
- [Qemu-devel] [PATCH v3 01/69] target/arm: Use store_reg_from_load in thumb2 code, Richard Henderson, 2019/08/28
- [Qemu-devel] [PATCH v3 02/69] target/arm: Add stubs for aa32 decodetree, Richard Henderson, 2019/08/28
- [Qemu-devel] [PATCH v3 03/69] target/arm: Convert Data Processing (register), Richard Henderson, 2019/08/28
- [Qemu-devel] [PATCH v3 07/69] target/arm: Simplify UMAAL,
Richard Henderson <=
- [Qemu-devel] [PATCH v3 04/69] target/arm: Convert Data Processing (reg-shifted-reg), Richard Henderson, 2019/08/28
- [Qemu-devel] [PATCH v3 08/69] target/arm: Convert Saturating addition and subtraction, Richard Henderson, 2019/08/28
- [Qemu-devel] [PATCH v3 05/69] target/arm: Convert Data Processing (immediate), Richard Henderson, 2019/08/28
- [Qemu-devel] [PATCH v3 06/69] target/arm: Convert multiply and multiply accumulate, Richard Henderson, 2019/08/28
- [Qemu-devel] [PATCH v3 09/69] target/arm: Convert Halfword multiply and multiply accumulate, Richard Henderson, 2019/08/28
- [Qemu-devel] [PATCH v3 12/69] target/arm: Convert MSR (immediate) and hints, Richard Henderson, 2019/08/28
- [Qemu-devel] [PATCH v3 10/69] target/arm: Simplify op_smlaxxx for SMLAL*, Richard Henderson, 2019/08/28
- [Qemu-devel] [PATCH v3 15/69] target/arm: Convert BX, BXJ, BLX (register), Richard Henderson, 2019/08/28
- [Qemu-devel] [PATCH v3 11/69] target/arm: Simplify op_smlawx for SMLAW*, Richard Henderson, 2019/08/28
- [Qemu-devel] [PATCH v3 16/69] target/arm: Convert CLZ, Richard Henderson, 2019/08/28