diff mbox series

[v2,03/12] target/ppc: Move V(ADD|SUB)CUW to decodetree and use gvec

Message ID 20221010191356.83659-4-lucas.araujo@eldorado.org.br (mailing list archive)
State New, archived
Headers show
Series VMX/VSX instructions with gvec | expand

Commit Message

Lucas Mateus Martins Araujo e Castro Oct. 10, 2022, 7:13 p.m. UTC 
From: "Lucas Mateus Castro (alqotel)" <lucas.araujo@eldorado.org.br>

This patch moves VADDCUW and VSUBCUW to decodtree with gvec using an
implementation based on the helper, with the main difference being
changing the -1 (aka all bits set to 1) result returned by cmp when
true to +1. It also implemented a .fni4 version of those instructions
and dropped the helper.

vaddcuw:
rept    loop    master             patch
8       12500   0,01008200         0,00612400 (-39.3%)
25      4000    0,01091500         0,00471600 (-56.8%)
100     1000    0,01332500         0,00593700 (-55.4%)
500     200     0,01998500         0,01275700 (-36.2%)
2500    40      0,04704300         0,04364300 (-7.2%)
8000    12      0,10748200         0,11241000 (+4.6%)

vsubcuw:
rept    loop    master             patch
8       12500   0,01226200         0,00571600 (-53.4%)
25      4000    0,01493500         0,00462100 (-69.1%)
100     1000    0,01522700         0,00455100 (-70.1%)
500     200     0,02384600         0,01133500 (-52.5%)
2500    40      0,04935200         0,03178100 (-35.6%)
8000    12      0,09039900         0,09440600 (+4.4%)

Overall there was a gain in performance, but the TCGop code was still
slightly bigger in the new version (it went from 4 to 5).

Signed-off-by: Lucas Mateus Castro (alqotel) <lucas.araujo@eldorado.org.br>
---
 target/ppc/helper.h                 |  2 -
 target/ppc/insn32.decode            |  2 +
 target/ppc/int_helper.c             | 18 ---------
 target/ppc/translate/vmx-impl.c.inc | 61 +++++++++++++++++++++++++++--
 target/ppc/translate/vmx-ops.c.inc  |  3 +-
 5 files changed, 60 insertions(+), 26 deletions(-)

Comments

Richard Henderson Oct. 10, 2022, 7:22 p.m. UTC | #1 
On 10/10/22 12:13, Lucas Mateus Castro(alqotel) wrote:
> From: "Lucas Mateus Castro (alqotel)"<lucas.araujo@eldorado.org.br>
> 
> This patch moves VADDCUW and VSUBCUW to decodtree with gvec using an
> implementation based on the helper, with the main difference being
> changing the -1 (aka all bits set to 1) result returned by cmp when
> true to +1. It also implemented a .fni4 version of those instructions
> and dropped the helper.
> 
> vaddcuw:
> rept    loop    master             patch
> 8       12500   0,01008200         0,00612400 (-39.3%)
> 25      4000    0,01091500         0,00471600 (-56.8%)
> 100     1000    0,01332500         0,00593700 (-55.4%)
> 500     200     0,01998500         0,01275700 (-36.2%)
> 2500    40      0,04704300         0,04364300 (-7.2%)
> 8000    12      0,10748200         0,11241000 (+4.6%)
> 
> vsubcuw:
> rept    loop    master             patch
> 8       12500   0,01226200         0,00571600 (-53.4%)
> 25      4000    0,01493500         0,00462100 (-69.1%)
> 100     1000    0,01522700         0,00455100 (-70.1%)
> 500     200     0,02384600         0,01133500 (-52.5%)
> 2500    40      0,04935200         0,03178100 (-35.6%)
> 8000    12      0,09039900         0,09440600 (+4.4%)
> 
> Overall there was a gain in performance, but the TCGop code was still
> slightly bigger in the new version (it went from 4 to 5).
> 
> Signed-off-by: Lucas Mateus Castro (alqotel)<lucas.araujo@eldorado.org.br>
> ---
>   target/ppc/helper.h                 |  2 -
>   target/ppc/insn32.decode            |  2 +
>   target/ppc/int_helper.c             | 18 ---------
>   target/ppc/translate/vmx-impl.c.inc | 61 +++++++++++++++++++++++++++--
>   target/ppc/translate/vmx-ops.c.inc  |  3 +-
>   5 files changed, 60 insertions(+), 26 deletions(-)

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>

r~
diff mbox series

Patch

diff --git a/target/ppc/helper.h b/target/ppc/helper.h
index f02a9497b7..f7047ed2aa 100644
--- a/target/ppc/helper.h
+++ b/target/ppc/helper.h
@@ -193,11 +193,9 @@  DEF_HELPER_FLAGS_3(vslo, TCG_CALL_NO_RWG, void, avr, avr, avr)
 DEF_HELPER_FLAGS_3(vsro, TCG_CALL_NO_RWG, void, avr, avr, avr)
 DEF_HELPER_FLAGS_3(vsrv, TCG_CALL_NO_RWG, void, avr, avr, avr)
 DEF_HELPER_FLAGS_3(vslv, TCG_CALL_NO_RWG, void, avr, avr, avr)
-DEF_HELPER_FLAGS_3(vaddcuw, TCG_CALL_NO_RWG, void, avr, avr, avr)
 DEF_HELPER_FLAGS_2(vprtybw, TCG_CALL_NO_RWG, void, avr, avr)
 DEF_HELPER_FLAGS_2(vprtybd, TCG_CALL_NO_RWG, void, avr, avr)
 DEF_HELPER_FLAGS_2(vprtybq, TCG_CALL_NO_RWG, void, avr, avr)
-DEF_HELPER_FLAGS_3(vsubcuw, TCG_CALL_NO_RWG, void, avr, avr, avr)
 DEF_HELPER_FLAGS_5(vaddsbs, TCG_CALL_NO_RWG, void, avr, avr, avr, avr, i32)
 DEF_HELPER_FLAGS_5(vaddshs, TCG_CALL_NO_RWG, void, avr, avr, avr, avr, i32)
 DEF_HELPER_FLAGS_5(vaddsws, TCG_CALL_NO_RWG, void, avr, avr, avr, avr, i32)
diff --git a/target/ppc/insn32.decode b/target/ppc/insn32.decode
index 9a509e84df..aebc7b73c8 100644
--- a/target/ppc/insn32.decode
+++ b/target/ppc/insn32.decode
@@ -608,12 +608,14 @@  VRLQNM          000100 ..... ..... ..... 00101000101    @VX
 
 ## Vector Integer Arithmetic Instructions
 
+VADDCUW         000100 ..... ..... ..... 00110000000    @VX
 VADDCUQ         000100 ..... ..... ..... 00101000000    @VX
 VADDUQM         000100 ..... ..... ..... 00100000000    @VX
 
 VADDEUQM        000100 ..... ..... ..... ..... 111100   @VA
 VADDECUQ        000100 ..... ..... ..... ..... 111101   @VA
 
+VSUBCUW         000100 ..... ..... ..... 10110000000    @VX
 VSUBCUQ         000100 ..... ..... ..... 10101000000    @VX
 VSUBUQM         000100 ..... ..... ..... 10100000000    @VX
 
diff --git a/target/ppc/int_helper.c b/target/ppc/int_helper.c
index ae1ba8084d..f8dd12e8ae 100644
--- a/target/ppc/int_helper.c
+++ b/target/ppc/int_helper.c
@@ -492,15 +492,6 @@  static inline void set_vscr_sat(CPUPPCState *env)
     env->vscr_sat.u32[0] = 1;
 }
 
-void helper_vaddcuw(ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b)
-{
-    int i;
-
-    for (i = 0; i < ARRAY_SIZE(r->u32); i++) {
-        r->u32[i] = ~a->u32[i] < b->u32[i];
-    }
-}
-
 /* vprtybw */
 void helper_vprtybw(ppc_avr_t *r, ppc_avr_t *b)
 {
@@ -1962,15 +1953,6 @@  void helper_vsro(ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b)
 #endif
 }
 
-void helper_vsubcuw(ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b)
-{
-    int i;
-
-    for (i = 0; i < ARRAY_SIZE(r->u32); i++) {
-        r->u32[i] = a->u32[i] >= b->u32[i];
-    }
-}
-
 void helper_vsumsws(CPUPPCState *env, ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b)
 {
     int64_t t;
diff --git a/target/ppc/translate/vmx-impl.c.inc b/target/ppc/translate/vmx-impl.c.inc
index 3acd585a2f..f52485a5f1 100644
--- a/target/ppc/translate/vmx-impl.c.inc
+++ b/target/ppc/translate/vmx-impl.c.inc
@@ -803,8 +803,6 @@  GEN_VXFORM(vsrv, 2, 28);
 GEN_VXFORM(vslv, 2, 29);
 GEN_VXFORM(vslo, 6, 16);
 GEN_VXFORM(vsro, 6, 17);
-GEN_VXFORM(vaddcuw, 0, 6);
-GEN_VXFORM(vsubcuw, 0, 22);
 
 static bool do_vector_gvec3_VX(DisasContext *ctx, arg_VX *a, int vece,
                                void (*gen_gvec)(unsigned, uint32_t, uint32_t,
@@ -2847,8 +2845,6 @@  static void gen_xpnd04_2(DisasContext *ctx)
 }
 
 
-GEN_VXFORM_DUAL(vsubcuw, PPC_ALTIVEC, PPC_NONE, \
-                xpnd04_1, PPC_NONE, PPC2_ISA300)
 GEN_VXFORM_DUAL(vsubsws, PPC_ALTIVEC, PPC_NONE, \
                 xpnd04_2, PPC_NONE, PPC2_ISA300)
 
@@ -3110,6 +3106,63 @@  TRANS_FLAGS2(ALTIVEC_207, VPMSUMD, do_vx_helper, gen_helper_VPMSUMD)
 TRANS_FLAGS2(ALTIVEC_207, VSUBCUQ, do_vx_helper, gen_helper_VSUBCUQ)
 TRANS_FLAGS2(ALTIVEC_207, VSUBUQM, do_vx_helper, gen_helper_VSUBUQM)
 
+static void gen_VADDCUW_vec(unsigned vece, TCGv_vec t, TCGv_vec a, TCGv_vec b)
+{
+    tcg_gen_not_vec(vece, a, a);
+    tcg_gen_cmp_vec(TCG_COND_LTU, vece, t, a, b);
+    tcg_gen_and_vec(vece, t, t, tcg_constant_vec_matching(t, vece, 1));
+}
+
+static void gen_VADDCUW_i32(TCGv_i32 t, TCGv_i32 a, TCGv_i32 b)
+{
+    tcg_gen_not_i32(a, a);
+    tcg_gen_setcond_i32(TCG_COND_LTU, t, a, b);
+}
+
+static void gen_VSUBCUW_vec(unsigned vece, TCGv_vec t, TCGv_vec a, TCGv_vec b)
+{
+    tcg_gen_cmp_vec(TCG_COND_GEU, vece, t, a, b);
+    tcg_gen_and_vec(vece, t, t, tcg_constant_vec_matching(t, vece, 1));
+}
+
+static void gen_VSUBCUW_i32(TCGv_i32 t, TCGv_i32 a, TCGv_i32 b)
+{
+    tcg_gen_setcond_i32(TCG_COND_GEU, t, a, b);
+}
+
+static bool do_vx_vaddsubcuw(DisasContext *ctx, arg_VX *a, int add)
+{
+    static const TCGOpcode vecop_list[] = {
+        INDEX_op_cmp_vec, 0
+    };
+
+    static const GVecGen3 op[] = {
+        {
+            .fniv = gen_VSUBCUW_vec,
+            .fni4 = gen_VSUBCUW_i32,
+            .opt_opc = vecop_list,
+            .vece = MO_32
+        },
+        {
+            .fniv = gen_VADDCUW_vec,
+            .fni4 = gen_VADDCUW_i32,
+            .opt_opc = vecop_list,
+            .vece = MO_32
+        },
+    };
+
+    REQUIRE_INSNS_FLAGS(ctx, ALTIVEC);
+    REQUIRE_VECTOR(ctx);
+
+    tcg_gen_gvec_3(avr_full_offset(a->vrt), avr_full_offset(a->vra),
+                   avr_full_offset(a->vrb), 16, 16, &op[add]);
+
+    return true;
+}
+
+TRANS(VSUBCUW, do_vx_vaddsubcuw, 0)
+TRANS(VADDCUW, do_vx_vaddsubcuw, 1)
+
 static bool do_vx_vmuleo(DisasContext *ctx, arg_VX *a, bool even,
                          void (*gen_mul)(TCGv_i64, TCGv_i64, TCGv_i64, TCGv_i64))
 {
diff --git a/target/ppc/translate/vmx-ops.c.inc b/target/ppc/translate/vmx-ops.c.inc
index 7cd9d40e06..ded0234123 100644
--- a/target/ppc/translate/vmx-ops.c.inc
+++ b/target/ppc/translate/vmx-ops.c.inc
@@ -106,12 +106,11 @@  GEN_VXFORM_300(vsrv, 2, 28),
 GEN_VXFORM_300(vslv, 2, 29),
 GEN_VXFORM(vslo, 6, 16),
 GEN_VXFORM(vsro, 6, 17),
-GEN_VXFORM(vaddcuw, 0, 6),
 GEN_HANDLER_E_2(vprtybw, 0x4, 0x1, 0x18, 8, 0, PPC_NONE, PPC2_ISA300),
 GEN_HANDLER_E_2(vprtybd, 0x4, 0x1, 0x18, 9, 0, PPC_NONE, PPC2_ISA300),
 GEN_HANDLER_E_2(vprtybq, 0x4, 0x1, 0x18, 10, 0, PPC_NONE, PPC2_ISA300),
 
-GEN_VXFORM_DUAL(vsubcuw, xpnd04_1, 0, 22, PPC_ALTIVEC, PPC_NONE),
+GEN_VXFORM(xpnd04_1, 0, 22),
 GEN_VXFORM_300(bcdsr, 0, 23),
 GEN_VXFORM_300(bcdsr, 0, 31),
 GEN_VXFORM_DUAL(vaddubs, vmul10uq, 0, 8, PPC_ALTIVEC, PPC_NONE),