* [PATCH] RISC-V: Add test cases for the RVV mask insn shortcut.
@ 2023-04-14 2:32 pan2.li
2023-04-14 2:35 ` juzhe.zhong
` (2 more replies)
0 siblings, 3 replies; 12+ messages in thread
From: pan2.li @ 2023-04-14 2:32 UTC (permalink / raw)
To: gcc-patches; +Cc: juzhe.zhong, kito.cheng, yanzhang.wang, pan2.li
From: Pan Li <pan2.li@intel.com>
There are sorts of shortcut codegen for the RVV mask insn. For
example.
vmxor vd, va, va => vmclr vd.
We would like to add more optimization like this but first of all
we must add the tests for the existing shortcut optimization, to
ensure we don't break existing optimization from underlying shortcut
optimization.
gcc/testsuite/ChangeLog:
* gcc.target/riscv/rvv/base/mask_insn_shortcut.c: New test.
---
.../riscv/rvv/base/mask_insn_shortcut.c | 237 ++++++++++++++++++
1 file changed, 237 insertions(+)
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
new file mode 100644
index 00000000000..8310aabaf59
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
@@ -0,0 +1,237 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv -mabi=lp64 -O3" } */
+
+#include "riscv_vector.h"
+
+vbool1_t test_shortcut_for_riscv_vmand_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmand_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmand_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmand_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmand_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmand_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmand_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmand_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmand_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmand_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmand_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmand_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmand_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmand_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmnand_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmnand_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmnand_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmnand_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmnand_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmnand_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmnand_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmandn_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmandn_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmandn_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmandn_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmandn_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmandn_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmandn_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmxor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmxor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmxor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmxor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmxor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmxor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmxor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmnor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmnor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmnor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmnor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmnor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmnor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmnor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmorn_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmorn_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmorn_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmorn_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmorn_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmorn_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmorn_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmxnor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmxnor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmxnor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmxnor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmxnor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmxnor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmxnor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b64(v1, v1, vl);
+}
+
+/* { dg-final { scan-assembler-not {vmand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnandn\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmxor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-times {vmorn\.mm\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 7 } } */
+/* { dg-final { scan-assembler-not {vmxnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
--
2.34.1
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] RISC-V: Add test cases for the RVV mask insn shortcut.
2023-04-14 2:32 [PATCH] RISC-V: Add test cases for the RVV mask insn shortcut pan2.li
@ 2023-04-14 2:35 ` juzhe.zhong
2023-04-14 2:37 ` Li, Pan2
2023-04-14 2:45 ` [PATCH v2] " pan2.li
2023-04-14 3:25 ` [PATCH v3] " pan2.li
2 siblings, 1 reply; 12+ messages in thread
From: juzhe.zhong @ 2023-04-14 2:35 UTC (permalink / raw)
To: pan2.li, gcc-patches; +Cc: Kito.cheng, yanzhang.wang, pan2.li
[-- Attachment #1: Type: text/plain, Size: 9750 bytes --]
+/* { dg-final { scan-assembler-not {vmand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnandn\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmxor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-times {vmorn\.mm\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 7 } } */
+/* { dg-final { scan-assembler-not {vmxnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
It's better add more assembler check
check how many vmclr.m or vmset.m should be.
juzhe.zhong@rivai.ai
From: pan2.li
Date: 2023-04-14 10:32
To: gcc-patches
CC: juzhe.zhong; kito.cheng; yanzhang.wang; pan2.li
Subject: [PATCH] RISC-V: Add test cases for the RVV mask insn shortcut.
From: Pan Li <pan2.li@intel.com>
There are sorts of shortcut codegen for the RVV mask insn. For
example.
vmxor vd, va, va => vmclr vd.
We would like to add more optimization like this but first of all
we must add the tests for the existing shortcut optimization, to
ensure we don't break existing optimization from underlying shortcut
optimization.
gcc/testsuite/ChangeLog:
* gcc.target/riscv/rvv/base/mask_insn_shortcut.c: New test.
---
.../riscv/rvv/base/mask_insn_shortcut.c | 237 ++++++++++++++++++
1 file changed, 237 insertions(+)
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
new file mode 100644
index 00000000000..8310aabaf59
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
@@ -0,0 +1,237 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv -mabi=lp64 -O3" } */
+
+#include "riscv_vector.h"
+
+vbool1_t test_shortcut_for_riscv_vmand_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmand_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmand_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmand_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmand_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmand_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmand_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmand_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmand_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmand_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmand_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmand_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmand_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmand_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmnand_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmnand_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmnand_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmnand_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmnand_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmnand_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmnand_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmandn_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmandn_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmandn_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmandn_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmandn_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmandn_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmandn_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmxor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmxor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmxor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmxor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmxor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmxor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmxor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmnor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmnor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmnor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmnor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmnor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmnor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmnor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmorn_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmorn_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmorn_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmorn_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmorn_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmorn_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmorn_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmxnor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmxnor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmxnor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmxnor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmxnor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmxnor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmxnor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b64(v1, v1, vl);
+}
+
+/* { dg-final { scan-assembler-not {vmand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnandn\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmxor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-times {vmorn\.mm\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 7 } } */
+/* { dg-final { scan-assembler-not {vmxnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
--
2.34.1
^ permalink raw reply [flat|nested] 12+ messages in thread
* RE: [PATCH] RISC-V: Add test cases for the RVV mask insn shortcut.
2023-04-14 2:35 ` juzhe.zhong
@ 2023-04-14 2:37 ` Li, Pan2
0 siblings, 0 replies; 12+ messages in thread
From: Li, Pan2 @ 2023-04-14 2:37 UTC (permalink / raw)
To: juzhe.zhong, gcc-patches; +Cc: Kito.cheng, Wang, Yanzhang
[-- Attachment #1: Type: text/plain, Size: 10399 bytes --]
Sure thing, let me update it ASAP.
Pan
From: juzhe.zhong@rivai.ai <juzhe.zhong@rivai.ai>
Sent: Friday, April 14, 2023 10:35 AM
To: Li, Pan2 <pan2.li@intel.com>; gcc-patches <gcc-patches@gcc.gnu.org>
Cc: Kito.cheng <kito.cheng@sifive.com>; Wang, Yanzhang <yanzhang.wang@intel.com>; Li, Pan2 <pan2.li@intel.com>
Subject: Re: [PATCH] RISC-V: Add test cases for the RVV mask insn shortcut.
+/* { dg-final { scan-assembler-not {vmand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnandn\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmxor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-times {vmorn\.mm\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 7 } } */
+/* { dg-final { scan-assembler-not {vmxnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
It's better add more assembler check
check how many vmclr.m or vmset.m should be.
________________________________
juzhe.zhong@rivai.ai<mailto:juzhe.zhong@rivai.ai>
From: pan2.li<mailto:pan2.li@intel.com>
Date: 2023-04-14 10:32
To: gcc-patches<mailto:gcc-patches@gcc.gnu.org>
CC: juzhe.zhong<mailto:juzhe.zhong@rivai.ai>; kito.cheng<mailto:kito.cheng@sifive.com>; yanzhang.wang<mailto:yanzhang.wang@intel.com>; pan2.li<mailto:pan2.li@intel.com>
Subject: [PATCH] RISC-V: Add test cases for the RVV mask insn shortcut.
From: Pan Li <pan2.li@intel.com<mailto:pan2.li@intel.com>>
There are sorts of shortcut codegen for the RVV mask insn. For
example.
vmxor vd, va, va => vmclr vd.
We would like to add more optimization like this but first of all
we must add the tests for the existing shortcut optimization, to
ensure we don't break existing optimization from underlying shortcut
optimization.
gcc/testsuite/ChangeLog:
* gcc.target/riscv/rvv/base/mask_insn_shortcut.c: New test.
---
.../riscv/rvv/base/mask_insn_shortcut.c | 237 ++++++++++++++++++
1 file changed, 237 insertions(+)
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
new file mode 100644
index 00000000000..8310aabaf59
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
@@ -0,0 +1,237 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv -mabi=lp64 -O3" } */
+
+#include "riscv_vector.h"
+
+vbool1_t test_shortcut_for_riscv_vmand_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmand_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmand_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmand_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmand_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmand_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmand_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmand_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmand_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmand_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmand_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmand_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmand_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmand_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmnand_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmnand_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmnand_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmnand_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmnand_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmnand_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmnand_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmandn_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmandn_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmandn_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmandn_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmandn_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmandn_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmandn_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmxor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmxor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmxor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmxor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmxor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmxor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmxor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmnor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmnor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmnor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmnor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmnor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmnor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmnor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmorn_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmorn_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmorn_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmorn_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmorn_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmorn_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmorn_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmxnor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmxnor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmxnor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmxnor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmxnor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmxnor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmxnor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b64(v1, v1, vl);
+}
+
+/* { dg-final { scan-assembler-not {vmand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnandn\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmxor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-times {vmorn\.mm\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 7 } } */
+/* { dg-final { scan-assembler-not {vmxnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
--
2.34.1
^ permalink raw reply [flat|nested] 12+ messages in thread
* [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
2023-04-14 2:32 [PATCH] RISC-V: Add test cases for the RVV mask insn shortcut pan2.li
2023-04-14 2:35 ` juzhe.zhong
@ 2023-04-14 2:45 ` pan2.li
2023-04-14 2:46 ` juzhe.zhong
2023-04-14 3:25 ` [PATCH v3] " pan2.li
2 siblings, 1 reply; 12+ messages in thread
From: pan2.li @ 2023-04-14 2:45 UTC (permalink / raw)
To: gcc-patches; +Cc: juzhe.zhong, kito.cheng, yanzhang.wang, pan2.li
From: Pan Li <pan2.li@intel.com>
There are sorts of shortcut codegen for the RVV mask insn. For
example.
vmxor vd, va, va => vmclr vd.
We would like to add more optimization like this but first of all
we must add the tests for the existing shortcut optimization, to
ensure we don't break existing optimization from underlying shortcut
optimization.
gcc/testsuite/ChangeLog:
* gcc.target/riscv/rvv/base/mask_insn_shortcut.c: New test.
Signed-off-by: Pan Li <pan2.li@intel.com>
---
.../riscv/rvv/base/mask_insn_shortcut.c | 239 ++++++++++++++++++
1 file changed, 239 insertions(+)
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
new file mode 100644
index 00000000000..efc3af39fc3
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
@@ -0,0 +1,239 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv -mabi=lp64 -O3" } */
+
+#include "riscv_vector.h"
+
+vbool1_t test_shortcut_for_riscv_vmand_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmand_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmand_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmand_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmand_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmand_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmand_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmand_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmand_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmand_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmand_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmand_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmand_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmand_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmnand_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmnand_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmnand_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmnand_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmnand_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmnand_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmnand_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmandn_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmandn_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmandn_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmandn_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmandn_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmandn_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmandn_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmxor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmxor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmxor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmxor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmxor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmxor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmxor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmnor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmnor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmnor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmnor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmnor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmnor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmnor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmorn_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmorn_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmorn_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmorn_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmorn_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmorn_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmorn_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmxnor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmxnor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmxnor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmxnor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmxnor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmxnor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmxnor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b64(v1, v1, vl);
+}
+
+/* { dg-final { scan-assembler-not {vmand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnandn\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmxor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-times {vmorn\.mm\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 7 } } */
+/* { dg-final { scan-assembler-not {vmxnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-times {vmclr\.m\s+v[0-9]+} 14 } } */
+/* { dg-final { scan-assembler-times {vmset\.m\s+v[0-9]+} 7 } } */
--
2.34.1
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
2023-04-14 2:45 ` [PATCH v2] " pan2.li
@ 2023-04-14 2:46 ` juzhe.zhong
2023-04-14 3:26 ` Li, Pan2
0 siblings, 1 reply; 12+ messages in thread
From: juzhe.zhong @ 2023-04-14 2:46 UTC (permalink / raw)
To: pan2.li, gcc-patches; +Cc: Kito.cheng, yanzhang.wang, pan2.li
[-- Attachment #1: Type: text/plain, Size: 9254 bytes --]
LGTM. Wait for Kito more comments.
juzhe.zhong@rivai.ai
From: pan2.li
Date: 2023-04-14 10:45
To: gcc-patches
CC: juzhe.zhong; kito.cheng; yanzhang.wang; pan2.li
Subject: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
From: Pan Li <pan2.li@intel.com>
There are sorts of shortcut codegen for the RVV mask insn. For
example.
vmxor vd, va, va => vmclr vd.
We would like to add more optimization like this but first of all
we must add the tests for the existing shortcut optimization, to
ensure we don't break existing optimization from underlying shortcut
optimization.
gcc/testsuite/ChangeLog:
* gcc.target/riscv/rvv/base/mask_insn_shortcut.c: New test.
Signed-off-by: Pan Li <pan2.li@intel.com>
---
.../riscv/rvv/base/mask_insn_shortcut.c | 239 ++++++++++++++++++
1 file changed, 239 insertions(+)
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
new file mode 100644
index 00000000000..efc3af39fc3
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
@@ -0,0 +1,239 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv -mabi=lp64 -O3" } */
+
+#include "riscv_vector.h"
+
+vbool1_t test_shortcut_for_riscv_vmand_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmand_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmand_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmand_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmand_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmand_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmand_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmand_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmand_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmand_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmand_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmand_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmand_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmand_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmnand_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmnand_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmnand_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmnand_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmnand_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmnand_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmnand_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmandn_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmandn_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmandn_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmandn_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmandn_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmandn_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmandn_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmxor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmxor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmxor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmxor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmxor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmxor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmxor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmnor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmnor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmnor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmnor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmnor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmnor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmnor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmorn_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmorn_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmorn_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmorn_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmorn_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmorn_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmorn_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmxnor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmxnor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmxnor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmxnor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmxnor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmxnor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmxnor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b64(v1, v1, vl);
+}
+
+/* { dg-final { scan-assembler-not {vmand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnandn\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmxor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-times {vmorn\.mm\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 7 } } */
+/* { dg-final { scan-assembler-not {vmxnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-times {vmclr\.m\s+v[0-9]+} 14 } } */
+/* { dg-final { scan-assembler-times {vmset\.m\s+v[0-9]+} 7 } } */
--
2.34.1
^ permalink raw reply [flat|nested] 12+ messages in thread
* [PATCH v3] RISC-V: Add test cases for the RVV mask insn shortcut.
2023-04-14 2:32 [PATCH] RISC-V: Add test cases for the RVV mask insn shortcut pan2.li
2023-04-14 2:35 ` juzhe.zhong
2023-04-14 2:45 ` [PATCH v2] " pan2.li
@ 2023-04-14 3:25 ` pan2.li
2 siblings, 0 replies; 12+ messages in thread
From: pan2.li @ 2023-04-14 3:25 UTC (permalink / raw)
To: gcc-patches; +Cc: juzhe.zhong, kito.cheng, yanzhang.wang, pan2.li
From: Pan Li <pan2.li@intel.com>
There are sorts of shortcut codegen for the RVV mask insn. For
example.
vmxor vd, va, va => vmclr vd.
We would like to add more optimization like this but first of all
we must add the tests for the existing shortcut optimization, to
ensure we don't break existing optimization from underlying shortcut
optimization.
gcc/testsuite/ChangeLog:
* gcc.target/riscv/rvv/base/mask_insn_shortcut.c: New test.
Signed-off-by: Pan Li <pan2.li@intel.com>
---
.../riscv/rvv/base/mask_insn_shortcut.c | 241 ++++++++++++++++++
1 file changed, 241 insertions(+)
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
new file mode 100644
index 00000000000..83cc4a1b5a5
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
@@ -0,0 +1,241 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv -mabi=lp64 -O3" } */
+
+#include "riscv_vector.h"
+
+vbool1_t test_shortcut_for_riscv_vmand_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmand_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmand_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmand_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmand_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmand_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmand_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmand_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmand_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmand_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmand_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmand_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmand_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmand_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmnand_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmnand_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmnand_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmnand_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmnand_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmnand_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmnand_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmandn_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmandn_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmandn_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmandn_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmandn_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmandn_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmandn_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmxor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmxor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmxor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmxor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmxor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmxor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmxor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmnor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmnor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmnor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmnor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmnor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmnor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmnor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmorn_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmorn_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmorn_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmorn_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmorn_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmorn_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmorn_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmxnor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmxnor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmxnor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmxnor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmxnor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmxnor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmxnor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b64(v1, v1, vl);
+}
+
+/* { dg-final { scan-assembler-not {vmand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnandn\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmxor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-times {vmorn\.mm\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 7 } } */
+/* { dg-final { scan-assembler-not {vmxnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-times {vmclr\.m\s+v[0-9]+} 14 } } */
+/* { dg-final { scan-assembler-times {vmset\.m\s+v[0-9]+} 7 } } */
+/* { dg-final { scan-assembler-times {vmmv\.m\s+v[0-9]+,\s*v[0-9]+} 14 } } */
+/* { dg-final { scan-assembler-times {vmnot\.m\s+v[0-9]+,\s*v[0-9]+} 14 } } */
--
2.34.1
^ permalink raw reply [flat|nested] 12+ messages in thread
* RE: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
2023-04-14 2:46 ` juzhe.zhong
@ 2023-04-14 3:26 ` Li, Pan2
2023-04-14 6:41 ` Kito Cheng
0 siblings, 1 reply; 12+ messages in thread
From: Li, Pan2 @ 2023-04-14 3:26 UTC (permalink / raw)
To: juzhe.zhong, gcc-patches; +Cc: Kito.cheng, Wang, Yanzhang
[-- Attachment #1: Type: text/plain, Size: 9962 bytes --]
Thanks juzhe, update new version [PATCH v3] for even more checks.
Pan
From: juzhe.zhong@rivai.ai <juzhe.zhong@rivai.ai>
Sent: Friday, April 14, 2023 10:46 AM
To: Li, Pan2 <pan2.li@intel.com>; gcc-patches <gcc-patches@gcc.gnu.org>
Cc: Kito.cheng <kito.cheng@sifive.com>; Wang, Yanzhang <yanzhang.wang@intel.com>; Li, Pan2 <pan2.li@intel.com>
Subject: Re: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
LGTM. Wait for Kito more comments.
________________________________
juzhe.zhong@rivai.ai<mailto:juzhe.zhong@rivai.ai>
From: pan2.li<mailto:pan2.li@intel.com>
Date: 2023-04-14 10:45
To: gcc-patches<mailto:gcc-patches@gcc.gnu.org>
CC: juzhe.zhong<mailto:juzhe.zhong@rivai.ai>; kito.cheng<mailto:kito.cheng@sifive.com>; yanzhang.wang<mailto:yanzhang.wang@intel.com>; pan2.li<mailto:pan2.li@intel.com>
Subject: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
From: Pan Li <pan2.li@intel.com<mailto:pan2.li@intel.com>>
There are sorts of shortcut codegen for the RVV mask insn. For
example.
vmxor vd, va, va => vmclr vd.
We would like to add more optimization like this but first of all
we must add the tests for the existing shortcut optimization, to
ensure we don't break existing optimization from underlying shortcut
optimization.
gcc/testsuite/ChangeLog:
* gcc.target/riscv/rvv/base/mask_insn_shortcut.c: New test.
Signed-off-by: Pan Li <pan2.li@intel.com<mailto:pan2.li@intel.com>>
---
.../riscv/rvv/base/mask_insn_shortcut.c | 239 ++++++++++++++++++
1 file changed, 239 insertions(+)
create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
new file mode 100644
index 00000000000..efc3af39fc3
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
@@ -0,0 +1,239 @@
+/* { dg-do compile } */
+/* { dg-options "-march=rv64gcv -mabi=lp64 -O3" } */
+
+#include "riscv_vector.h"
+
+vbool1_t test_shortcut_for_riscv_vmand_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmand_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmand_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmand_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmand_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmand_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmand_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmand_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmand_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmand_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmand_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmand_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmand_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmand_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmnand_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmnand_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmnand_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmnand_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmnand_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmnand_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmnand_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmnand_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmandn_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmandn_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmandn_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmandn_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmandn_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmandn_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmandn_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmandn_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmxor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmxor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmxor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmxor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmxor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmxor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmxor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmxor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmnor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmnor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmnor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmnor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmnor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmnor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmnor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmnor_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmorn_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmorn_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmorn_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmorn_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmorn_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmorn_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmorn_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmorn_mm_b64(v1, v1, vl);
+}
+
+vbool1_t test_shortcut_for_riscv_vmxnor_case_0(vbool1_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b1(v1, v1, vl);
+}
+
+vbool2_t test_shortcut_for_riscv_vmxnor_case_1(vbool2_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b2(v1, v1, vl);
+}
+
+vbool4_t test_shortcut_for_riscv_vmxnor_case_2(vbool4_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b4(v1, v1, vl);
+}
+
+vbool8_t test_shortcut_for_riscv_vmxnor_case_3(vbool8_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b8(v1, v1, vl);
+}
+
+vbool16_t test_shortcut_for_riscv_vmxnor_case_4(vbool16_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b16(v1, v1, vl);
+}
+
+vbool32_t test_shortcut_for_riscv_vmxnor_case_5(vbool32_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b32(v1, v1, vl);
+}
+
+vbool64_t test_shortcut_for_riscv_vmxnor_case_6(vbool64_t v1, size_t vl) {
+ return __riscv_vmxnor_mm_b64(v1, v1, vl);
+}
+
+/* { dg-final { scan-assembler-not {vmand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnandn\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmxor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-not {vmnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-times {vmorn\.mm\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 7 } } */
+/* { dg-final { scan-assembler-not {vmxnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
+/* { dg-final { scan-assembler-times {vmclr\.m\s+v[0-9]+} 14 } } */
+/* { dg-final { scan-assembler-times {vmset\.m\s+v[0-9]+} 7 } } */
--
2.34.1
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
2023-04-14 3:26 ` Li, Pan2
@ 2023-04-14 6:41 ` Kito Cheng
2023-04-14 6:47 ` Li, Pan2
0 siblings, 1 reply; 12+ messages in thread
From: Kito Cheng @ 2023-04-14 6:41 UTC (permalink / raw)
To: Li, Pan2; +Cc: juzhe.zhong, gcc-patches, Kito.cheng, Wang, Yanzhang
OK, thanks for the patch :)
On Fri, Apr 14, 2023 at 11:27 AM Li, Pan2 via Gcc-patches
<gcc-patches@gcc.gnu.org> wrote:
>
> Thanks juzhe, update new version [PATCH v3] for even more checks.
>
> Pan
>
> From: juzhe.zhong@rivai.ai <juzhe.zhong@rivai.ai>
> Sent: Friday, April 14, 2023 10:46 AM
> To: Li, Pan2 <pan2.li@intel.com>; gcc-patches <gcc-patches@gcc.gnu.org>
> Cc: Kito.cheng <kito.cheng@sifive.com>; Wang, Yanzhang <yanzhang.wang@intel.com>; Li, Pan2 <pan2.li@intel.com>
> Subject: Re: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
>
> LGTM. Wait for Kito more comments.
>
> ________________________________
> juzhe.zhong@rivai.ai<mailto:juzhe.zhong@rivai.ai>
>
> From: pan2.li<mailto:pan2.li@intel.com>
> Date: 2023-04-14 10:45
> To: gcc-patches<mailto:gcc-patches@gcc.gnu.org>
> CC: juzhe.zhong<mailto:juzhe.zhong@rivai.ai>; kito.cheng<mailto:kito.cheng@sifive.com>; yanzhang.wang<mailto:yanzhang.wang@intel.com>; pan2.li<mailto:pan2.li@intel.com>
> Subject: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
> From: Pan Li <pan2.li@intel.com<mailto:pan2.li@intel.com>>
>
> There are sorts of shortcut codegen for the RVV mask insn. For
> example.
>
> vmxor vd, va, va => vmclr vd.
>
> We would like to add more optimization like this but first of all
> we must add the tests for the existing shortcut optimization, to
> ensure we don't break existing optimization from underlying shortcut
> optimization.
>
> gcc/testsuite/ChangeLog:
>
> * gcc.target/riscv/rvv/base/mask_insn_shortcut.c: New test.
>
> Signed-off-by: Pan Li <pan2.li@intel.com<mailto:pan2.li@intel.com>>
> ---
> .../riscv/rvv/base/mask_insn_shortcut.c | 239 ++++++++++++++++++
> 1 file changed, 239 insertions(+)
> create mode 100644 gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
>
> diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> new file mode 100644
> index 00000000000..efc3af39fc3
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> @@ -0,0 +1,239 @@
> +/* { dg-do compile } */
> +/* { dg-options "-march=rv64gcv -mabi=lp64 -O3" } */
> +
> +#include "riscv_vector.h"
> +
> +vbool1_t test_shortcut_for_riscv_vmand_case_0(vbool1_t v1, size_t vl) {
> + return __riscv_vmand_mm_b1(v1, v1, vl);
> +}
> +
> +vbool2_t test_shortcut_for_riscv_vmand_case_1(vbool2_t v1, size_t vl) {
> + return __riscv_vmand_mm_b2(v1, v1, vl);
> +}
> +
> +vbool4_t test_shortcut_for_riscv_vmand_case_2(vbool4_t v1, size_t vl) {
> + return __riscv_vmand_mm_b4(v1, v1, vl);
> +}
> +
> +vbool8_t test_shortcut_for_riscv_vmand_case_3(vbool8_t v1, size_t vl) {
> + return __riscv_vmand_mm_b8(v1, v1, vl);
> +}
> +
> +vbool16_t test_shortcut_for_riscv_vmand_case_4(vbool16_t v1, size_t vl) {
> + return __riscv_vmand_mm_b16(v1, v1, vl);
> +}
> +
> +vbool32_t test_shortcut_for_riscv_vmand_case_5(vbool32_t v1, size_t vl) {
> + return __riscv_vmand_mm_b32(v1, v1, vl);
> +}
> +
> +vbool64_t test_shortcut_for_riscv_vmand_case_6(vbool64_t v1, size_t vl) {
> + return __riscv_vmand_mm_b64(v1, v1, vl);
> +}
> +
> +vbool1_t test_shortcut_for_riscv_vmnand_case_0(vbool1_t v1, size_t vl) {
> + return __riscv_vmnand_mm_b1(v1, v1, vl);
> +}
> +
> +vbool2_t test_shortcut_for_riscv_vmnand_case_1(vbool2_t v1, size_t vl) {
> + return __riscv_vmnand_mm_b2(v1, v1, vl);
> +}
> +
> +vbool4_t test_shortcut_for_riscv_vmnand_case_2(vbool4_t v1, size_t vl) {
> + return __riscv_vmnand_mm_b4(v1, v1, vl);
> +}
> +
> +vbool8_t test_shortcut_for_riscv_vmnand_case_3(vbool8_t v1, size_t vl) {
> + return __riscv_vmnand_mm_b8(v1, v1, vl);
> +}
> +
> +vbool16_t test_shortcut_for_riscv_vmnand_case_4(vbool16_t v1, size_t vl) {
> + return __riscv_vmnand_mm_b16(v1, v1, vl);
> +}
> +
> +vbool32_t test_shortcut_for_riscv_vmnand_case_5(vbool32_t v1, size_t vl) {
> + return __riscv_vmnand_mm_b32(v1, v1, vl);
> +}
> +
> +vbool64_t test_shortcut_for_riscv_vmnand_case_6(vbool64_t v1, size_t vl) {
> + return __riscv_vmnand_mm_b64(v1, v1, vl);
> +}
> +
> +vbool1_t test_shortcut_for_riscv_vmandn_case_0(vbool1_t v1, size_t vl) {
> + return __riscv_vmandn_mm_b1(v1, v1, vl);
> +}
> +
> +vbool2_t test_shortcut_for_riscv_vmandn_case_1(vbool2_t v1, size_t vl) {
> + return __riscv_vmandn_mm_b2(v1, v1, vl);
> +}
> +
> +vbool4_t test_shortcut_for_riscv_vmandn_case_2(vbool4_t v1, size_t vl) {
> + return __riscv_vmandn_mm_b4(v1, v1, vl);
> +}
> +
> +vbool8_t test_shortcut_for_riscv_vmandn_case_3(vbool8_t v1, size_t vl) {
> + return __riscv_vmandn_mm_b8(v1, v1, vl);
> +}
> +
> +vbool16_t test_shortcut_for_riscv_vmandn_case_4(vbool16_t v1, size_t vl) {
> + return __riscv_vmandn_mm_b16(v1, v1, vl);
> +}
> +
> +vbool32_t test_shortcut_for_riscv_vmandn_case_5(vbool32_t v1, size_t vl) {
> + return __riscv_vmandn_mm_b32(v1, v1, vl);
> +}
> +
> +vbool64_t test_shortcut_for_riscv_vmandn_case_6(vbool64_t v1, size_t vl) {
> + return __riscv_vmandn_mm_b64(v1, v1, vl);
> +}
> +
> +vbool1_t test_shortcut_for_riscv_vmxor_case_0(vbool1_t v1, size_t vl) {
> + return __riscv_vmxor_mm_b1(v1, v1, vl);
> +}
> +
> +vbool2_t test_shortcut_for_riscv_vmxor_case_1(vbool2_t v1, size_t vl) {
> + return __riscv_vmxor_mm_b2(v1, v1, vl);
> +}
> +
> +vbool4_t test_shortcut_for_riscv_vmxor_case_2(vbool4_t v1, size_t vl) {
> + return __riscv_vmxor_mm_b4(v1, v1, vl);
> +}
> +
> +vbool8_t test_shortcut_for_riscv_vmxor_case_3(vbool8_t v1, size_t vl) {
> + return __riscv_vmxor_mm_b8(v1, v1, vl);
> +}
> +
> +vbool16_t test_shortcut_for_riscv_vmxor_case_4(vbool16_t v1, size_t vl) {
> + return __riscv_vmxor_mm_b16(v1, v1, vl);
> +}
> +
> +vbool32_t test_shortcut_for_riscv_vmxor_case_5(vbool32_t v1, size_t vl) {
> + return __riscv_vmxor_mm_b32(v1, v1, vl);
> +}
> +
> +vbool64_t test_shortcut_for_riscv_vmxor_case_6(vbool64_t v1, size_t vl) {
> + return __riscv_vmxor_mm_b64(v1, v1, vl);
> +}
> +
> +vbool1_t test_shortcut_for_riscv_vmor_case_0(vbool1_t v1, size_t vl) {
> + return __riscv_vmor_mm_b1(v1, v1, vl);
> +}
> +
> +vbool2_t test_shortcut_for_riscv_vmor_case_1(vbool2_t v1, size_t vl) {
> + return __riscv_vmor_mm_b2(v1, v1, vl);
> +}
> +
> +vbool4_t test_shortcut_for_riscv_vmor_case_2(vbool4_t v1, size_t vl) {
> + return __riscv_vmor_mm_b4(v1, v1, vl);
> +}
> +
> +vbool8_t test_shortcut_for_riscv_vmor_case_3(vbool8_t v1, size_t vl) {
> + return __riscv_vmor_mm_b8(v1, v1, vl);
> +}
> +
> +vbool16_t test_shortcut_for_riscv_vmor_case_4(vbool16_t v1, size_t vl) {
> + return __riscv_vmor_mm_b16(v1, v1, vl);
> +}
> +
> +vbool32_t test_shortcut_for_riscv_vmor_case_5(vbool32_t v1, size_t vl) {
> + return __riscv_vmor_mm_b32(v1, v1, vl);
> +}
> +
> +vbool64_t test_shortcut_for_riscv_vmor_case_6(vbool64_t v1, size_t vl) {
> + return __riscv_vmor_mm_b64(v1, v1, vl);
> +}
> +
> +vbool1_t test_shortcut_for_riscv_vmnor_case_0(vbool1_t v1, size_t vl) {
> + return __riscv_vmnor_mm_b1(v1, v1, vl);
> +}
> +
> +vbool2_t test_shortcut_for_riscv_vmnor_case_1(vbool2_t v1, size_t vl) {
> + return __riscv_vmnor_mm_b2(v1, v1, vl);
> +}
> +
> +vbool4_t test_shortcut_for_riscv_vmnor_case_2(vbool4_t v1, size_t vl) {
> + return __riscv_vmnor_mm_b4(v1, v1, vl);
> +}
> +
> +vbool8_t test_shortcut_for_riscv_vmnor_case_3(vbool8_t v1, size_t vl) {
> + return __riscv_vmnor_mm_b8(v1, v1, vl);
> +}
> +
> +vbool16_t test_shortcut_for_riscv_vmnor_case_4(vbool16_t v1, size_t vl) {
> + return __riscv_vmnor_mm_b16(v1, v1, vl);
> +}
> +
> +vbool32_t test_shortcut_for_riscv_vmnor_case_5(vbool32_t v1, size_t vl) {
> + return __riscv_vmnor_mm_b32(v1, v1, vl);
> +}
> +
> +vbool64_t test_shortcut_for_riscv_vmnor_case_6(vbool64_t v1, size_t vl) {
> + return __riscv_vmnor_mm_b64(v1, v1, vl);
> +}
> +
> +vbool1_t test_shortcut_for_riscv_vmorn_case_0(vbool1_t v1, size_t vl) {
> + return __riscv_vmorn_mm_b1(v1, v1, vl);
> +}
> +
> +vbool2_t test_shortcut_for_riscv_vmorn_case_1(vbool2_t v1, size_t vl) {
> + return __riscv_vmorn_mm_b2(v1, v1, vl);
> +}
> +
> +vbool4_t test_shortcut_for_riscv_vmorn_case_2(vbool4_t v1, size_t vl) {
> + return __riscv_vmorn_mm_b4(v1, v1, vl);
> +}
> +
> +vbool8_t test_shortcut_for_riscv_vmorn_case_3(vbool8_t v1, size_t vl) {
> + return __riscv_vmorn_mm_b8(v1, v1, vl);
> +}
> +
> +vbool16_t test_shortcut_for_riscv_vmorn_case_4(vbool16_t v1, size_t vl) {
> + return __riscv_vmorn_mm_b16(v1, v1, vl);
> +}
> +
> +vbool32_t test_shortcut_for_riscv_vmorn_case_5(vbool32_t v1, size_t vl) {
> + return __riscv_vmorn_mm_b32(v1, v1, vl);
> +}
> +
> +vbool64_t test_shortcut_for_riscv_vmorn_case_6(vbool64_t v1, size_t vl) {
> + return __riscv_vmorn_mm_b64(v1, v1, vl);
> +}
> +
> +vbool1_t test_shortcut_for_riscv_vmxnor_case_0(vbool1_t v1, size_t vl) {
> + return __riscv_vmxnor_mm_b1(v1, v1, vl);
> +}
> +
> +vbool2_t test_shortcut_for_riscv_vmxnor_case_1(vbool2_t v1, size_t vl) {
> + return __riscv_vmxnor_mm_b2(v1, v1, vl);
> +}
> +
> +vbool4_t test_shortcut_for_riscv_vmxnor_case_2(vbool4_t v1, size_t vl) {
> + return __riscv_vmxnor_mm_b4(v1, v1, vl);
> +}
> +
> +vbool8_t test_shortcut_for_riscv_vmxnor_case_3(vbool8_t v1, size_t vl) {
> + return __riscv_vmxnor_mm_b8(v1, v1, vl);
> +}
> +
> +vbool16_t test_shortcut_for_riscv_vmxnor_case_4(vbool16_t v1, size_t vl) {
> + return __riscv_vmxnor_mm_b16(v1, v1, vl);
> +}
> +
> +vbool32_t test_shortcut_for_riscv_vmxnor_case_5(vbool32_t v1, size_t vl) {
> + return __riscv_vmxnor_mm_b32(v1, v1, vl);
> +}
> +
> +vbool64_t test_shortcut_for_riscv_vmxnor_case_6(vbool64_t v1, size_t vl) {
> + return __riscv_vmxnor_mm_b64(v1, v1, vl);
> +}
> +
> +/* { dg-final { scan-assembler-not {vmand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
> +/* { dg-final { scan-assembler-not {vmnand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
> +/* { dg-final { scan-assembler-not {vmnandn\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
> +/* { dg-final { scan-assembler-not {vmxor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
> +/* { dg-final { scan-assembler-not {vmor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
> +/* { dg-final { scan-assembler-not {vmnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
> +/* { dg-final { scan-assembler-times {vmorn\.mm\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 7 } } */
> +/* { dg-final { scan-assembler-not {vmxnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
> +/* { dg-final { scan-assembler-times {vmclr\.m\s+v[0-9]+} 14 } } */
> +/* { dg-final { scan-assembler-times {vmset\.m\s+v[0-9]+} 7 } } */
> --
> 2.34.1
>
>
^ permalink raw reply [flat|nested] 12+ messages in thread
* RE: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
2023-04-14 6:41 ` Kito Cheng
@ 2023-04-14 6:47 ` Li, Pan2
2023-04-17 1:46 ` Li, Pan2
0 siblings, 1 reply; 12+ messages in thread
From: Li, Pan2 @ 2023-04-14 6:47 UTC (permalink / raw)
To: Kito Cheng; +Cc: juzhe.zhong, gcc-patches, Kito.cheng, Wang, Yanzhang
You're very welcome!
Looks vmorn(v,v) doesn't perform any shortcut, while vmandn(v, v) will be converted to vmclr in upstream. As I understand, there should be no difference between vmORn and vmANDn excepts the operator, will take a look from RTL CSE for more details, 😊!
Pan
-----Original Message-----
From: Kito Cheng <kito.cheng@gmail.com>
Sent: Friday, April 14, 2023 2:42 PM
To: Li, Pan2 <pan2.li@intel.com>
Cc: juzhe.zhong@rivai.ai; gcc-patches <gcc-patches@gcc.gnu.org>; Kito.cheng <kito.cheng@sifive.com>; Wang, Yanzhang <yanzhang.wang@intel.com>
Subject: Re: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
OK, thanks for the patch :)
On Fri, Apr 14, 2023 at 11:27 AM Li, Pan2 via Gcc-patches <gcc-patches@gcc.gnu.org> wrote:
>
> Thanks juzhe, update new version [PATCH v3] for even more checks.
>
> Pan
>
> From: juzhe.zhong@rivai.ai <juzhe.zhong@rivai.ai>
> Sent: Friday, April 14, 2023 10:46 AM
> To: Li, Pan2 <pan2.li@intel.com>; gcc-patches
> <gcc-patches@gcc.gnu.org>
> Cc: Kito.cheng <kito.cheng@sifive.com>; Wang, Yanzhang
> <yanzhang.wang@intel.com>; Li, Pan2 <pan2.li@intel.com>
> Subject: Re: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
>
> LGTM. Wait for Kito more comments.
>
> ________________________________
> juzhe.zhong@rivai.ai<mailto:juzhe.zhong@rivai.ai>
>
> From: pan2.li<mailto:pan2.li@intel.com>
> Date: 2023-04-14 10:45
> To: gcc-patches<mailto:gcc-patches@gcc.gnu.org>
> CC: juzhe.zhong<mailto:juzhe.zhong@rivai.ai>;
> kito.cheng<mailto:kito.cheng@sifive.com>;
> yanzhang.wang<mailto:yanzhang.wang@intel.com>;
> pan2.li<mailto:pan2.li@intel.com>
> Subject: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
> From: Pan Li <pan2.li@intel.com<mailto:pan2.li@intel.com>>
>
> There are sorts of shortcut codegen for the RVV mask insn. For
> example.
>
> vmxor vd, va, va => vmclr vd.
>
> We would like to add more optimization like this but first of all we
> must add the tests for the existing shortcut optimization, to ensure
> we don't break existing optimization from underlying shortcut
> optimization.
>
> gcc/testsuite/ChangeLog:
>
> * gcc.target/riscv/rvv/base/mask_insn_shortcut.c: New test.
>
> Signed-off-by: Pan Li <pan2.li@intel.com<mailto:pan2.li@intel.com>>
> ---
> .../riscv/rvv/base/mask_insn_shortcut.c | 239 ++++++++++++++++++
> 1 file changed, 239 insertions(+)
> create mode 100644
> gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
>
> diff --git
> a/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> new file mode 100644
> index 00000000000..efc3af39fc3
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> @@ -0,0 +1,239 @@
> +/* { dg-do compile } */
> +/* { dg-options "-march=rv64gcv -mabi=lp64 -O3" } */
> +
> +#include "riscv_vector.h"
> +
> +vbool1_t test_shortcut_for_riscv_vmand_case_0(vbool1_t v1, size_t vl)
> +{
> + return __riscv_vmand_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmand_case_1(vbool2_t v1, size_t vl)
> +{
> + return __riscv_vmand_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmand_case_2(vbool4_t v1, size_t vl)
> +{
> + return __riscv_vmand_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmand_case_3(vbool8_t v1, size_t vl)
> +{
> + return __riscv_vmand_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmand_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmand_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmand_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmand_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmand_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmand_mm_b64(v1, v1, vl); }
> +
> +vbool1_t test_shortcut_for_riscv_vmnand_case_0(vbool1_t v1, size_t
> +vl) {
> + return __riscv_vmnand_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmnand_case_1(vbool2_t v1, size_t
> +vl) {
> + return __riscv_vmnand_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmnand_case_2(vbool4_t v1, size_t
> +vl) {
> + return __riscv_vmnand_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmnand_case_3(vbool8_t v1, size_t
> +vl) {
> + return __riscv_vmnand_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmnand_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmnand_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmnand_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmnand_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmnand_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmnand_mm_b64(v1, v1, vl); }
> +
> +vbool1_t test_shortcut_for_riscv_vmandn_case_0(vbool1_t v1, size_t
> +vl) {
> + return __riscv_vmandn_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmandn_case_1(vbool2_t v1, size_t
> +vl) {
> + return __riscv_vmandn_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmandn_case_2(vbool4_t v1, size_t
> +vl) {
> + return __riscv_vmandn_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmandn_case_3(vbool8_t v1, size_t
> +vl) {
> + return __riscv_vmandn_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmandn_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmandn_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmandn_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmandn_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmandn_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmandn_mm_b64(v1, v1, vl); }
> +
> +vbool1_t test_shortcut_for_riscv_vmxor_case_0(vbool1_t v1, size_t vl)
> +{
> + return __riscv_vmxor_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmxor_case_1(vbool2_t v1, size_t vl)
> +{
> + return __riscv_vmxor_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmxor_case_2(vbool4_t v1, size_t vl)
> +{
> + return __riscv_vmxor_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmxor_case_3(vbool8_t v1, size_t vl)
> +{
> + return __riscv_vmxor_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmxor_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmxor_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmxor_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmxor_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmxor_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmxor_mm_b64(v1, v1, vl); }
> +
> +vbool1_t test_shortcut_for_riscv_vmor_case_0(vbool1_t v1, size_t vl)
> +{
> + return __riscv_vmor_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmor_case_1(vbool2_t v1, size_t vl)
> +{
> + return __riscv_vmor_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmor_case_2(vbool4_t v1, size_t vl)
> +{
> + return __riscv_vmor_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmor_case_3(vbool8_t v1, size_t vl)
> +{
> + return __riscv_vmor_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmor_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmor_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmor_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmor_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmor_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmor_mm_b64(v1, v1, vl); }
> +
> +vbool1_t test_shortcut_for_riscv_vmnor_case_0(vbool1_t v1, size_t vl)
> +{
> + return __riscv_vmnor_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmnor_case_1(vbool2_t v1, size_t vl)
> +{
> + return __riscv_vmnor_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmnor_case_2(vbool4_t v1, size_t vl)
> +{
> + return __riscv_vmnor_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmnor_case_3(vbool8_t v1, size_t vl)
> +{
> + return __riscv_vmnor_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmnor_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmnor_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmnor_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmnor_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmnor_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmnor_mm_b64(v1, v1, vl); }
> +
> +vbool1_t test_shortcut_for_riscv_vmorn_case_0(vbool1_t v1, size_t vl)
> +{
> + return __riscv_vmorn_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmorn_case_1(vbool2_t v1, size_t vl)
> +{
> + return __riscv_vmorn_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmorn_case_2(vbool4_t v1, size_t vl)
> +{
> + return __riscv_vmorn_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmorn_case_3(vbool8_t v1, size_t vl)
> +{
> + return __riscv_vmorn_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmorn_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmorn_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmorn_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmorn_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmorn_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmorn_mm_b64(v1, v1, vl); }
> +
> +vbool1_t test_shortcut_for_riscv_vmxnor_case_0(vbool1_t v1, size_t
> +vl) {
> + return __riscv_vmxnor_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmxnor_case_1(vbool2_t v1, size_t
> +vl) {
> + return __riscv_vmxnor_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmxnor_case_2(vbool4_t v1, size_t
> +vl) {
> + return __riscv_vmxnor_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmxnor_case_3(vbool8_t v1, size_t
> +vl) {
> + return __riscv_vmxnor_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmxnor_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmxnor_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmxnor_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmxnor_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmxnor_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmxnor_mm_b64(v1, v1, vl); }
> +
> +/* { dg-final { scan-assembler-not {vmand\.mm\s+v[0-9]+,\s*v[0-9]+} }
> +} */
> +/* { dg-final { scan-assembler-not {vmnand\.mm\s+v[0-9]+,\s*v[0-9]+}
> +} } */
> +/* { dg-final { scan-assembler-not {vmnandn\.mm\s+v[0-9]+,\s*v[0-9]+}
> +} } */
> +/* { dg-final { scan-assembler-not {vmxor\.mm\s+v[0-9]+,\s*v[0-9]+} }
> +} */
> +/* { dg-final { scan-assembler-not {vmor\.mm\s+v[0-9]+,\s*v[0-9]+} }
> +} */
> +/* { dg-final { scan-assembler-not {vmnor\.mm\s+v[0-9]+,\s*v[0-9]+} }
> +} */
> +/* { dg-final { scan-assembler-times
> +{vmorn\.mm\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 7 } } */
> +/* { dg-final { scan-assembler-not {vmxnor\.mm\s+v[0-9]+,\s*v[0-9]+}
> +} } */
> +/* { dg-final { scan-assembler-times {vmclr\.m\s+v[0-9]+} 14 } } */
> +/* { dg-final { scan-assembler-times {vmset\.m\s+v[0-9]+} 7 } } */
> --
> 2.34.1
>
>
^ permalink raw reply [flat|nested] 12+ messages in thread
* RE: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
2023-04-14 6:47 ` Li, Pan2
@ 2023-04-17 1:46 ` Li, Pan2
2023-04-17 1:52 ` Kito Cheng
0 siblings, 1 reply; 12+ messages in thread
From: Li, Pan2 @ 2023-04-17 1:46 UTC (permalink / raw)
To: Kito Cheng; +Cc: juzhe.zhong, gcc-patches, Kito.cheng, Wang, Yanzhang
BTW, this patch will be in GCC 13 or master? The underlying patches may depend on this tests for ensuring correctness.
Pan
-----Original Message-----
From: Li, Pan2
Sent: Friday, April 14, 2023 2:47 PM
To: Kito Cheng <kito.cheng@gmail.com>
Cc: juzhe.zhong@rivai.ai; gcc-patches <gcc-patches@gcc.gnu.org>; Kito.cheng <kito.cheng@sifive.com>; Wang, Yanzhang <yanzhang.wang@intel.com>
Subject: RE: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
You're very welcome!
Looks vmorn(v,v) doesn't perform any shortcut, while vmandn(v, v) will be converted to vmclr in upstream. As I understand, there should be no difference between vmORn and vmANDn excepts the operator, will take a look from RTL CSE for more details, 😊!
Pan
-----Original Message-----
From: Kito Cheng <kito.cheng@gmail.com>
Sent: Friday, April 14, 2023 2:42 PM
To: Li, Pan2 <pan2.li@intel.com>
Cc: juzhe.zhong@rivai.ai; gcc-patches <gcc-patches@gcc.gnu.org>; Kito.cheng <kito.cheng@sifive.com>; Wang, Yanzhang <yanzhang.wang@intel.com>
Subject: Re: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
OK, thanks for the patch :)
On Fri, Apr 14, 2023 at 11:27 AM Li, Pan2 via Gcc-patches <gcc-patches@gcc.gnu.org> wrote:
>
> Thanks juzhe, update new version [PATCH v3] for even more checks.
>
> Pan
>
> From: juzhe.zhong@rivai.ai <juzhe.zhong@rivai.ai>
> Sent: Friday, April 14, 2023 10:46 AM
> To: Li, Pan2 <pan2.li@intel.com>; gcc-patches
> <gcc-patches@gcc.gnu.org>
> Cc: Kito.cheng <kito.cheng@sifive.com>; Wang, Yanzhang
> <yanzhang.wang@intel.com>; Li, Pan2 <pan2.li@intel.com>
> Subject: Re: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
>
> LGTM. Wait for Kito more comments.
>
> ________________________________
> juzhe.zhong@rivai.ai<mailto:juzhe.zhong@rivai.ai>
>
> From: pan2.li<mailto:pan2.li@intel.com>
> Date: 2023-04-14 10:45
> To: gcc-patches<mailto:gcc-patches@gcc.gnu.org>
> CC: juzhe.zhong<mailto:juzhe.zhong@rivai.ai>;
> kito.cheng<mailto:kito.cheng@sifive.com>;
> yanzhang.wang<mailto:yanzhang.wang@intel.com>;
> pan2.li<mailto:pan2.li@intel.com>
> Subject: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
> From: Pan Li <pan2.li@intel.com<mailto:pan2.li@intel.com>>
>
> There are sorts of shortcut codegen for the RVV mask insn. For
> example.
>
> vmxor vd, va, va => vmclr vd.
>
> We would like to add more optimization like this but first of all we
> must add the tests for the existing shortcut optimization, to ensure
> we don't break existing optimization from underlying shortcut
> optimization.
>
> gcc/testsuite/ChangeLog:
>
> * gcc.target/riscv/rvv/base/mask_insn_shortcut.c: New test.
>
> Signed-off-by: Pan Li <pan2.li@intel.com<mailto:pan2.li@intel.com>>
> ---
> .../riscv/rvv/base/mask_insn_shortcut.c | 239 ++++++++++++++++++
> 1 file changed, 239 insertions(+)
> create mode 100644
> gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
>
> diff --git
> a/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> new file mode 100644
> index 00000000000..efc3af39fc3
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> @@ -0,0 +1,239 @@
> +/* { dg-do compile } */
> +/* { dg-options "-march=rv64gcv -mabi=lp64 -O3" } */
> +
> +#include "riscv_vector.h"
> +
> +vbool1_t test_shortcut_for_riscv_vmand_case_0(vbool1_t v1, size_t vl)
> +{
> + return __riscv_vmand_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmand_case_1(vbool2_t v1, size_t vl)
> +{
> + return __riscv_vmand_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmand_case_2(vbool4_t v1, size_t vl)
> +{
> + return __riscv_vmand_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmand_case_3(vbool8_t v1, size_t vl)
> +{
> + return __riscv_vmand_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmand_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmand_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmand_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmand_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmand_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmand_mm_b64(v1, v1, vl); }
> +
> +vbool1_t test_shortcut_for_riscv_vmnand_case_0(vbool1_t v1, size_t
> +vl) {
> + return __riscv_vmnand_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmnand_case_1(vbool2_t v1, size_t
> +vl) {
> + return __riscv_vmnand_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmnand_case_2(vbool4_t v1, size_t
> +vl) {
> + return __riscv_vmnand_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmnand_case_3(vbool8_t v1, size_t
> +vl) {
> + return __riscv_vmnand_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmnand_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmnand_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmnand_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmnand_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmnand_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmnand_mm_b64(v1, v1, vl); }
> +
> +vbool1_t test_shortcut_for_riscv_vmandn_case_0(vbool1_t v1, size_t
> +vl) {
> + return __riscv_vmandn_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmandn_case_1(vbool2_t v1, size_t
> +vl) {
> + return __riscv_vmandn_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmandn_case_2(vbool4_t v1, size_t
> +vl) {
> + return __riscv_vmandn_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmandn_case_3(vbool8_t v1, size_t
> +vl) {
> + return __riscv_vmandn_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmandn_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmandn_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmandn_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmandn_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmandn_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmandn_mm_b64(v1, v1, vl); }
> +
> +vbool1_t test_shortcut_for_riscv_vmxor_case_0(vbool1_t v1, size_t vl)
> +{
> + return __riscv_vmxor_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmxor_case_1(vbool2_t v1, size_t vl)
> +{
> + return __riscv_vmxor_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmxor_case_2(vbool4_t v1, size_t vl)
> +{
> + return __riscv_vmxor_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmxor_case_3(vbool8_t v1, size_t vl)
> +{
> + return __riscv_vmxor_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmxor_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmxor_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmxor_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmxor_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmxor_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmxor_mm_b64(v1, v1, vl); }
> +
> +vbool1_t test_shortcut_for_riscv_vmor_case_0(vbool1_t v1, size_t vl)
> +{
> + return __riscv_vmor_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmor_case_1(vbool2_t v1, size_t vl)
> +{
> + return __riscv_vmor_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmor_case_2(vbool4_t v1, size_t vl)
> +{
> + return __riscv_vmor_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmor_case_3(vbool8_t v1, size_t vl)
> +{
> + return __riscv_vmor_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmor_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmor_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmor_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmor_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmor_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmor_mm_b64(v1, v1, vl); }
> +
> +vbool1_t test_shortcut_for_riscv_vmnor_case_0(vbool1_t v1, size_t vl)
> +{
> + return __riscv_vmnor_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmnor_case_1(vbool2_t v1, size_t vl)
> +{
> + return __riscv_vmnor_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmnor_case_2(vbool4_t v1, size_t vl)
> +{
> + return __riscv_vmnor_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmnor_case_3(vbool8_t v1, size_t vl)
> +{
> + return __riscv_vmnor_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmnor_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmnor_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmnor_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmnor_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmnor_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmnor_mm_b64(v1, v1, vl); }
> +
> +vbool1_t test_shortcut_for_riscv_vmorn_case_0(vbool1_t v1, size_t vl)
> +{
> + return __riscv_vmorn_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmorn_case_1(vbool2_t v1, size_t vl)
> +{
> + return __riscv_vmorn_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmorn_case_2(vbool4_t v1, size_t vl)
> +{
> + return __riscv_vmorn_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmorn_case_3(vbool8_t v1, size_t vl)
> +{
> + return __riscv_vmorn_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmorn_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmorn_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmorn_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmorn_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmorn_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmorn_mm_b64(v1, v1, vl); }
> +
> +vbool1_t test_shortcut_for_riscv_vmxnor_case_0(vbool1_t v1, size_t
> +vl) {
> + return __riscv_vmxnor_mm_b1(v1, v1, vl); }
> +
> +vbool2_t test_shortcut_for_riscv_vmxnor_case_1(vbool2_t v1, size_t
> +vl) {
> + return __riscv_vmxnor_mm_b2(v1, v1, vl); }
> +
> +vbool4_t test_shortcut_for_riscv_vmxnor_case_2(vbool4_t v1, size_t
> +vl) {
> + return __riscv_vmxnor_mm_b4(v1, v1, vl); }
> +
> +vbool8_t test_shortcut_for_riscv_vmxnor_case_3(vbool8_t v1, size_t
> +vl) {
> + return __riscv_vmxnor_mm_b8(v1, v1, vl); }
> +
> +vbool16_t test_shortcut_for_riscv_vmxnor_case_4(vbool16_t v1, size_t
> +vl) {
> + return __riscv_vmxnor_mm_b16(v1, v1, vl); }
> +
> +vbool32_t test_shortcut_for_riscv_vmxnor_case_5(vbool32_t v1, size_t
> +vl) {
> + return __riscv_vmxnor_mm_b32(v1, v1, vl); }
> +
> +vbool64_t test_shortcut_for_riscv_vmxnor_case_6(vbool64_t v1, size_t
> +vl) {
> + return __riscv_vmxnor_mm_b64(v1, v1, vl); }
> +
> +/* { dg-final { scan-assembler-not {vmand\.mm\s+v[0-9]+,\s*v[0-9]+} }
> +} */
> +/* { dg-final { scan-assembler-not {vmnand\.mm\s+v[0-9]+,\s*v[0-9]+}
> +} } */
> +/* { dg-final { scan-assembler-not {vmnandn\.mm\s+v[0-9]+,\s*v[0-9]+}
> +} } */
> +/* { dg-final { scan-assembler-not {vmxor\.mm\s+v[0-9]+,\s*v[0-9]+} }
> +} */
> +/* { dg-final { scan-assembler-not {vmor\.mm\s+v[0-9]+,\s*v[0-9]+} }
> +} */
> +/* { dg-final { scan-assembler-not {vmnor\.mm\s+v[0-9]+,\s*v[0-9]+} }
> +} */
> +/* { dg-final { scan-assembler-times
> +{vmorn\.mm\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 7 } } */
> +/* { dg-final { scan-assembler-not {vmxnor\.mm\s+v[0-9]+,\s*v[0-9]+}
> +} } */
> +/* { dg-final { scan-assembler-times {vmclr\.m\s+v[0-9]+} 14 } } */
> +/* { dg-final { scan-assembler-times {vmset\.m\s+v[0-9]+} 7 } } */
> --
> 2.34.1
>
>
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
2023-04-17 1:46 ` Li, Pan2
@ 2023-04-17 1:52 ` Kito Cheng
2023-04-17 1:55 ` Li, Pan2
0 siblings, 1 reply; 12+ messages in thread
From: Kito Cheng @ 2023-04-17 1:52 UTC (permalink / raw)
To: Li, Pan2; +Cc: juzhe.zhong, gcc-patches, Kito.cheng, Wang, Yanzhang
Pushed to trunk :)
On Mon, Apr 17, 2023 at 9:47 AM Li, Pan2 via Gcc-patches
<gcc-patches@gcc.gnu.org> wrote:
>
> BTW, this patch will be in GCC 13 or master? The underlying patches may depend on this tests for ensuring correctness.
>
> Pan
>
> -----Original Message-----
> From: Li, Pan2
> Sent: Friday, April 14, 2023 2:47 PM
> To: Kito Cheng <kito.cheng@gmail.com>
> Cc: juzhe.zhong@rivai.ai; gcc-patches <gcc-patches@gcc.gnu.org>; Kito.cheng <kito.cheng@sifive.com>; Wang, Yanzhang <yanzhang.wang@intel.com>
> Subject: RE: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
>
> You're very welcome!
>
> Looks vmorn(v,v) doesn't perform any shortcut, while vmandn(v, v) will be converted to vmclr in upstream. As I understand, there should be no difference between vmORn and vmANDn excepts the operator, will take a look from RTL CSE for more details, 😊!
>
> Pan
>
> -----Original Message-----
> From: Kito Cheng <kito.cheng@gmail.com>
> Sent: Friday, April 14, 2023 2:42 PM
> To: Li, Pan2 <pan2.li@intel.com>
> Cc: juzhe.zhong@rivai.ai; gcc-patches <gcc-patches@gcc.gnu.org>; Kito.cheng <kito.cheng@sifive.com>; Wang, Yanzhang <yanzhang.wang@intel.com>
> Subject: Re: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
>
> OK, thanks for the patch :)
>
> On Fri, Apr 14, 2023 at 11:27 AM Li, Pan2 via Gcc-patches <gcc-patches@gcc.gnu.org> wrote:
> >
> > Thanks juzhe, update new version [PATCH v3] for even more checks.
> >
> > Pan
> >
> > From: juzhe.zhong@rivai.ai <juzhe.zhong@rivai.ai>
> > Sent: Friday, April 14, 2023 10:46 AM
> > To: Li, Pan2 <pan2.li@intel.com>; gcc-patches
> > <gcc-patches@gcc.gnu.org>
> > Cc: Kito.cheng <kito.cheng@sifive.com>; Wang, Yanzhang
> > <yanzhang.wang@intel.com>; Li, Pan2 <pan2.li@intel.com>
> > Subject: Re: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
> >
> > LGTM. Wait for Kito more comments.
> >
> > ________________________________
> > juzhe.zhong@rivai.ai<mailto:juzhe.zhong@rivai.ai>
> >
> > From: pan2.li<mailto:pan2.li@intel.com>
> > Date: 2023-04-14 10:45
> > To: gcc-patches<mailto:gcc-patches@gcc.gnu.org>
> > CC: juzhe.zhong<mailto:juzhe.zhong@rivai.ai>;
> > kito.cheng<mailto:kito.cheng@sifive.com>;
> > yanzhang.wang<mailto:yanzhang.wang@intel.com>;
> > pan2.li<mailto:pan2.li@intel.com>
> > Subject: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
> > From: Pan Li <pan2.li@intel.com<mailto:pan2.li@intel.com>>
> >
> > There are sorts of shortcut codegen for the RVV mask insn. For
> > example.
> >
> > vmxor vd, va, va => vmclr vd.
> >
> > We would like to add more optimization like this but first of all we
> > must add the tests for the existing shortcut optimization, to ensure
> > we don't break existing optimization from underlying shortcut
> > optimization.
> >
> > gcc/testsuite/ChangeLog:
> >
> > * gcc.target/riscv/rvv/base/mask_insn_shortcut.c: New test.
> >
> > Signed-off-by: Pan Li <pan2.li@intel.com<mailto:pan2.li@intel.com>>
> > ---
> > .../riscv/rvv/base/mask_insn_shortcut.c | 239 ++++++++++++++++++
> > 1 file changed, 239 insertions(+)
> > create mode 100644
> > gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> >
> > diff --git
> > a/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> > b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> > new file mode 100644
> > index 00000000000..efc3af39fc3
> > --- /dev/null
> > +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> > @@ -0,0 +1,239 @@
> > +/* { dg-do compile } */
> > +/* { dg-options "-march=rv64gcv -mabi=lp64 -O3" } */
> > +
> > +#include "riscv_vector.h"
> > +
> > +vbool1_t test_shortcut_for_riscv_vmand_case_0(vbool1_t v1, size_t vl)
> > +{
> > + return __riscv_vmand_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmand_case_1(vbool2_t v1, size_t vl)
> > +{
> > + return __riscv_vmand_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmand_case_2(vbool4_t v1, size_t vl)
> > +{
> > + return __riscv_vmand_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmand_case_3(vbool8_t v1, size_t vl)
> > +{
> > + return __riscv_vmand_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmand_case_4(vbool16_t v1, size_t
> > +vl) {
> > + return __riscv_vmand_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmand_case_5(vbool32_t v1, size_t
> > +vl) {
> > + return __riscv_vmand_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmand_case_6(vbool64_t v1, size_t
> > +vl) {
> > + return __riscv_vmand_mm_b64(v1, v1, vl); }
> > +
> > +vbool1_t test_shortcut_for_riscv_vmnand_case_0(vbool1_t v1, size_t
> > +vl) {
> > + return __riscv_vmnand_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmnand_case_1(vbool2_t v1, size_t
> > +vl) {
> > + return __riscv_vmnand_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmnand_case_2(vbool4_t v1, size_t
> > +vl) {
> > + return __riscv_vmnand_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmnand_case_3(vbool8_t v1, size_t
> > +vl) {
> > + return __riscv_vmnand_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmnand_case_4(vbool16_t v1, size_t
> > +vl) {
> > + return __riscv_vmnand_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmnand_case_5(vbool32_t v1, size_t
> > +vl) {
> > + return __riscv_vmnand_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmnand_case_6(vbool64_t v1, size_t
> > +vl) {
> > + return __riscv_vmnand_mm_b64(v1, v1, vl); }
> > +
> > +vbool1_t test_shortcut_for_riscv_vmandn_case_0(vbool1_t v1, size_t
> > +vl) {
> > + return __riscv_vmandn_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmandn_case_1(vbool2_t v1, size_t
> > +vl) {
> > + return __riscv_vmandn_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmandn_case_2(vbool4_t v1, size_t
> > +vl) {
> > + return __riscv_vmandn_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmandn_case_3(vbool8_t v1, size_t
> > +vl) {
> > + return __riscv_vmandn_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmandn_case_4(vbool16_t v1, size_t
> > +vl) {
> > + return __riscv_vmandn_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmandn_case_5(vbool32_t v1, size_t
> > +vl) {
> > + return __riscv_vmandn_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmandn_case_6(vbool64_t v1, size_t
> > +vl) {
> > + return __riscv_vmandn_mm_b64(v1, v1, vl); }
> > +
> > +vbool1_t test_shortcut_for_riscv_vmxor_case_0(vbool1_t v1, size_t vl)
> > +{
> > + return __riscv_vmxor_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmxor_case_1(vbool2_t v1, size_t vl)
> > +{
> > + return __riscv_vmxor_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmxor_case_2(vbool4_t v1, size_t vl)
> > +{
> > + return __riscv_vmxor_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmxor_case_3(vbool8_t v1, size_t vl)
> > +{
> > + return __riscv_vmxor_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmxor_case_4(vbool16_t v1, size_t
> > +vl) {
> > + return __riscv_vmxor_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmxor_case_5(vbool32_t v1, size_t
> > +vl) {
> > + return __riscv_vmxor_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmxor_case_6(vbool64_t v1, size_t
> > +vl) {
> > + return __riscv_vmxor_mm_b64(v1, v1, vl); }
> > +
> > +vbool1_t test_shortcut_for_riscv_vmor_case_0(vbool1_t v1, size_t vl)
> > +{
> > + return __riscv_vmor_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmor_case_1(vbool2_t v1, size_t vl)
> > +{
> > + return __riscv_vmor_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmor_case_2(vbool4_t v1, size_t vl)
> > +{
> > + return __riscv_vmor_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmor_case_3(vbool8_t v1, size_t vl)
> > +{
> > + return __riscv_vmor_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmor_case_4(vbool16_t v1, size_t
> > +vl) {
> > + return __riscv_vmor_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmor_case_5(vbool32_t v1, size_t
> > +vl) {
> > + return __riscv_vmor_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmor_case_6(vbool64_t v1, size_t
> > +vl) {
> > + return __riscv_vmor_mm_b64(v1, v1, vl); }
> > +
> > +vbool1_t test_shortcut_for_riscv_vmnor_case_0(vbool1_t v1, size_t vl)
> > +{
> > + return __riscv_vmnor_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmnor_case_1(vbool2_t v1, size_t vl)
> > +{
> > + return __riscv_vmnor_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmnor_case_2(vbool4_t v1, size_t vl)
> > +{
> > + return __riscv_vmnor_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmnor_case_3(vbool8_t v1, size_t vl)
> > +{
> > + return __riscv_vmnor_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmnor_case_4(vbool16_t v1, size_t
> > +vl) {
> > + return __riscv_vmnor_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmnor_case_5(vbool32_t v1, size_t
> > +vl) {
> > + return __riscv_vmnor_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmnor_case_6(vbool64_t v1, size_t
> > +vl) {
> > + return __riscv_vmnor_mm_b64(v1, v1, vl); }
> > +
> > +vbool1_t test_shortcut_for_riscv_vmorn_case_0(vbool1_t v1, size_t vl)
> > +{
> > + return __riscv_vmorn_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmorn_case_1(vbool2_t v1, size_t vl)
> > +{
> > + return __riscv_vmorn_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmorn_case_2(vbool4_t v1, size_t vl)
> > +{
> > + return __riscv_vmorn_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmorn_case_3(vbool8_t v1, size_t vl)
> > +{
> > + return __riscv_vmorn_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmorn_case_4(vbool16_t v1, size_t
> > +vl) {
> > + return __riscv_vmorn_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmorn_case_5(vbool32_t v1, size_t
> > +vl) {
> > + return __riscv_vmorn_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmorn_case_6(vbool64_t v1, size_t
> > +vl) {
> > + return __riscv_vmorn_mm_b64(v1, v1, vl); }
> > +
> > +vbool1_t test_shortcut_for_riscv_vmxnor_case_0(vbool1_t v1, size_t
> > +vl) {
> > + return __riscv_vmxnor_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmxnor_case_1(vbool2_t v1, size_t
> > +vl) {
> > + return __riscv_vmxnor_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmxnor_case_2(vbool4_t v1, size_t
> > +vl) {
> > + return __riscv_vmxnor_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmxnor_case_3(vbool8_t v1, size_t
> > +vl) {
> > + return __riscv_vmxnor_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmxnor_case_4(vbool16_t v1, size_t
> > +vl) {
> > + return __riscv_vmxnor_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmxnor_case_5(vbool32_t v1, size_t
> > +vl) {
> > + return __riscv_vmxnor_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmxnor_case_6(vbool64_t v1, size_t
> > +vl) {
> > + return __riscv_vmxnor_mm_b64(v1, v1, vl); }
> > +
> > +/* { dg-final { scan-assembler-not {vmand\.mm\s+v[0-9]+,\s*v[0-9]+} }
> > +} */
> > +/* { dg-final { scan-assembler-not {vmnand\.mm\s+v[0-9]+,\s*v[0-9]+}
> > +} } */
> > +/* { dg-final { scan-assembler-not {vmnandn\.mm\s+v[0-9]+,\s*v[0-9]+}
> > +} } */
> > +/* { dg-final { scan-assembler-not {vmxor\.mm\s+v[0-9]+,\s*v[0-9]+} }
> > +} */
> > +/* { dg-final { scan-assembler-not {vmor\.mm\s+v[0-9]+,\s*v[0-9]+} }
> > +} */
> > +/* { dg-final { scan-assembler-not {vmnor\.mm\s+v[0-9]+,\s*v[0-9]+} }
> > +} */
> > +/* { dg-final { scan-assembler-times
> > +{vmorn\.mm\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 7 } } */
> > +/* { dg-final { scan-assembler-not {vmxnor\.mm\s+v[0-9]+,\s*v[0-9]+}
> > +} } */
> > +/* { dg-final { scan-assembler-times {vmclr\.m\s+v[0-9]+} 14 } } */
> > +/* { dg-final { scan-assembler-times {vmset\.m\s+v[0-9]+} 7 } } */
> > --
> > 2.34.1
> >
> >
^ permalink raw reply [flat|nested] 12+ messages in thread
* RE: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
2023-04-17 1:52 ` Kito Cheng
@ 2023-04-17 1:55 ` Li, Pan2
0 siblings, 0 replies; 12+ messages in thread
From: Li, Pan2 @ 2023-04-17 1:55 UTC (permalink / raw)
To: Kito Cheng; +Cc: juzhe.zhong, gcc-patches, Kito.cheng, Wang, Yanzhang
Cool, thank you!
Pan
-----Original Message-----
From: Kito Cheng <kito.cheng@gmail.com>
Sent: Monday, April 17, 2023 9:52 AM
To: Li, Pan2 <pan2.li@intel.com>
Cc: juzhe.zhong@rivai.ai; gcc-patches <gcc-patches@gcc.gnu.org>; Kito.cheng <kito.cheng@sifive.com>; Wang, Yanzhang <yanzhang.wang@intel.com>
Subject: Re: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
Pushed to trunk :)
On Mon, Apr 17, 2023 at 9:47 AM Li, Pan2 via Gcc-patches <gcc-patches@gcc.gnu.org> wrote:
>
> BTW, this patch will be in GCC 13 or master? The underlying patches may depend on this tests for ensuring correctness.
>
> Pan
>
> -----Original Message-----
> From: Li, Pan2
> Sent: Friday, April 14, 2023 2:47 PM
> To: Kito Cheng <kito.cheng@gmail.com>
> Cc: juzhe.zhong@rivai.ai; gcc-patches <gcc-patches@gcc.gnu.org>;
> Kito.cheng <kito.cheng@sifive.com>; Wang, Yanzhang
> <yanzhang.wang@intel.com>
> Subject: RE: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
>
> You're very welcome!
>
> Looks vmorn(v,v) doesn't perform any shortcut, while vmandn(v, v) will be converted to vmclr in upstream. As I understand, there should be no difference between vmORn and vmANDn excepts the operator, will take a look from RTL CSE for more details, 😊!
>
> Pan
>
> -----Original Message-----
> From: Kito Cheng <kito.cheng@gmail.com>
> Sent: Friday, April 14, 2023 2:42 PM
> To: Li, Pan2 <pan2.li@intel.com>
> Cc: juzhe.zhong@rivai.ai; gcc-patches <gcc-patches@gcc.gnu.org>;
> Kito.cheng <kito.cheng@sifive.com>; Wang, Yanzhang
> <yanzhang.wang@intel.com>
> Subject: Re: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
>
> OK, thanks for the patch :)
>
> On Fri, Apr 14, 2023 at 11:27 AM Li, Pan2 via Gcc-patches <gcc-patches@gcc.gnu.org> wrote:
> >
> > Thanks juzhe, update new version [PATCH v3] for even more checks.
> >
> > Pan
> >
> > From: juzhe.zhong@rivai.ai <juzhe.zhong@rivai.ai>
> > Sent: Friday, April 14, 2023 10:46 AM
> > To: Li, Pan2 <pan2.li@intel.com>; gcc-patches
> > <gcc-patches@gcc.gnu.org>
> > Cc: Kito.cheng <kito.cheng@sifive.com>; Wang, Yanzhang
> > <yanzhang.wang@intel.com>; Li, Pan2 <pan2.li@intel.com>
> > Subject: Re: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
> >
> > LGTM. Wait for Kito more comments.
> >
> > ________________________________
> > juzhe.zhong@rivai.ai<mailto:juzhe.zhong@rivai.ai>
> >
> > From: pan2.li<mailto:pan2.li@intel.com>
> > Date: 2023-04-14 10:45
> > To: gcc-patches<mailto:gcc-patches@gcc.gnu.org>
> > CC: juzhe.zhong<mailto:juzhe.zhong@rivai.ai>;
> > kito.cheng<mailto:kito.cheng@sifive.com>;
> > yanzhang.wang<mailto:yanzhang.wang@intel.com>;
> > pan2.li<mailto:pan2.li@intel.com>
> > Subject: [PATCH v2] RISC-V: Add test cases for the RVV mask insn shortcut.
> > From: Pan Li <pan2.li@intel.com<mailto:pan2.li@intel.com>>
> >
> > There are sorts of shortcut codegen for the RVV mask insn. For
> > example.
> >
> > vmxor vd, va, va => vmclr vd.
> >
> > We would like to add more optimization like this but first of all we
> > must add the tests for the existing shortcut optimization, to ensure
> > we don't break existing optimization from underlying shortcut
> > optimization.
> >
> > gcc/testsuite/ChangeLog:
> >
> > * gcc.target/riscv/rvv/base/mask_insn_shortcut.c: New test.
> >
> > Signed-off-by: Pan Li <pan2.li@intel.com<mailto:pan2.li@intel.com>>
> > ---
> > .../riscv/rvv/base/mask_insn_shortcut.c | 239 ++++++++++++++++++
> > 1 file changed, 239 insertions(+)
> > create mode 100644
> > gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> >
> > diff --git
> > a/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> > b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> > new file mode 100644
> > index 00000000000..efc3af39fc3
> > --- /dev/null
> > +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/mask_insn_shortcut.c
> > @@ -0,0 +1,239 @@
> > +/* { dg-do compile } */
> > +/* { dg-options "-march=rv64gcv -mabi=lp64 -O3" } */
> > +
> > +#include "riscv_vector.h"
> > +
> > +vbool1_t test_shortcut_for_riscv_vmand_case_0(vbool1_t v1, size_t
> > +vl) {
> > + return __riscv_vmand_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmand_case_1(vbool2_t v1, size_t
> > +vl) {
> > + return __riscv_vmand_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmand_case_2(vbool4_t v1, size_t
> > +vl) {
> > + return __riscv_vmand_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmand_case_3(vbool8_t v1, size_t
> > +vl) {
> > + return __riscv_vmand_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmand_case_4(vbool16_t v1, size_t
> > +vl) {
> > + return __riscv_vmand_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmand_case_5(vbool32_t v1, size_t
> > +vl) {
> > + return __riscv_vmand_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmand_case_6(vbool64_t v1, size_t
> > +vl) {
> > + return __riscv_vmand_mm_b64(v1, v1, vl); }
> > +
> > +vbool1_t test_shortcut_for_riscv_vmnand_case_0(vbool1_t v1, size_t
> > +vl) {
> > + return __riscv_vmnand_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmnand_case_1(vbool2_t v1, size_t
> > +vl) {
> > + return __riscv_vmnand_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmnand_case_2(vbool4_t v1, size_t
> > +vl) {
> > + return __riscv_vmnand_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmnand_case_3(vbool8_t v1, size_t
> > +vl) {
> > + return __riscv_vmnand_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmnand_case_4(vbool16_t v1,
> > +size_t
> > +vl) {
> > + return __riscv_vmnand_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmnand_case_5(vbool32_t v1,
> > +size_t
> > +vl) {
> > + return __riscv_vmnand_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmnand_case_6(vbool64_t v1,
> > +size_t
> > +vl) {
> > + return __riscv_vmnand_mm_b64(v1, v1, vl); }
> > +
> > +vbool1_t test_shortcut_for_riscv_vmandn_case_0(vbool1_t v1, size_t
> > +vl) {
> > + return __riscv_vmandn_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmandn_case_1(vbool2_t v1, size_t
> > +vl) {
> > + return __riscv_vmandn_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmandn_case_2(vbool4_t v1, size_t
> > +vl) {
> > + return __riscv_vmandn_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmandn_case_3(vbool8_t v1, size_t
> > +vl) {
> > + return __riscv_vmandn_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmandn_case_4(vbool16_t v1,
> > +size_t
> > +vl) {
> > + return __riscv_vmandn_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmandn_case_5(vbool32_t v1,
> > +size_t
> > +vl) {
> > + return __riscv_vmandn_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmandn_case_6(vbool64_t v1,
> > +size_t
> > +vl) {
> > + return __riscv_vmandn_mm_b64(v1, v1, vl); }
> > +
> > +vbool1_t test_shortcut_for_riscv_vmxor_case_0(vbool1_t v1, size_t
> > +vl) {
> > + return __riscv_vmxor_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmxor_case_1(vbool2_t v1, size_t
> > +vl) {
> > + return __riscv_vmxor_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmxor_case_2(vbool4_t v1, size_t
> > +vl) {
> > + return __riscv_vmxor_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmxor_case_3(vbool8_t v1, size_t
> > +vl) {
> > + return __riscv_vmxor_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmxor_case_4(vbool16_t v1, size_t
> > +vl) {
> > + return __riscv_vmxor_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmxor_case_5(vbool32_t v1, size_t
> > +vl) {
> > + return __riscv_vmxor_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmxor_case_6(vbool64_t v1, size_t
> > +vl) {
> > + return __riscv_vmxor_mm_b64(v1, v1, vl); }
> > +
> > +vbool1_t test_shortcut_for_riscv_vmor_case_0(vbool1_t v1, size_t
> > +vl) {
> > + return __riscv_vmor_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmor_case_1(vbool2_t v1, size_t
> > +vl) {
> > + return __riscv_vmor_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmor_case_2(vbool4_t v1, size_t
> > +vl) {
> > + return __riscv_vmor_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmor_case_3(vbool8_t v1, size_t
> > +vl) {
> > + return __riscv_vmor_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmor_case_4(vbool16_t v1, size_t
> > +vl) {
> > + return __riscv_vmor_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmor_case_5(vbool32_t v1, size_t
> > +vl) {
> > + return __riscv_vmor_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmor_case_6(vbool64_t v1, size_t
> > +vl) {
> > + return __riscv_vmor_mm_b64(v1, v1, vl); }
> > +
> > +vbool1_t test_shortcut_for_riscv_vmnor_case_0(vbool1_t v1, size_t
> > +vl) {
> > + return __riscv_vmnor_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmnor_case_1(vbool2_t v1, size_t
> > +vl) {
> > + return __riscv_vmnor_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmnor_case_2(vbool4_t v1, size_t
> > +vl) {
> > + return __riscv_vmnor_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmnor_case_3(vbool8_t v1, size_t
> > +vl) {
> > + return __riscv_vmnor_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmnor_case_4(vbool16_t v1, size_t
> > +vl) {
> > + return __riscv_vmnor_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmnor_case_5(vbool32_t v1, size_t
> > +vl) {
> > + return __riscv_vmnor_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmnor_case_6(vbool64_t v1, size_t
> > +vl) {
> > + return __riscv_vmnor_mm_b64(v1, v1, vl); }
> > +
> > +vbool1_t test_shortcut_for_riscv_vmorn_case_0(vbool1_t v1, size_t
> > +vl) {
> > + return __riscv_vmorn_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmorn_case_1(vbool2_t v1, size_t
> > +vl) {
> > + return __riscv_vmorn_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmorn_case_2(vbool4_t v1, size_t
> > +vl) {
> > + return __riscv_vmorn_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmorn_case_3(vbool8_t v1, size_t
> > +vl) {
> > + return __riscv_vmorn_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmorn_case_4(vbool16_t v1, size_t
> > +vl) {
> > + return __riscv_vmorn_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmorn_case_5(vbool32_t v1, size_t
> > +vl) {
> > + return __riscv_vmorn_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmorn_case_6(vbool64_t v1, size_t
> > +vl) {
> > + return __riscv_vmorn_mm_b64(v1, v1, vl); }
> > +
> > +vbool1_t test_shortcut_for_riscv_vmxnor_case_0(vbool1_t v1, size_t
> > +vl) {
> > + return __riscv_vmxnor_mm_b1(v1, v1, vl); }
> > +
> > +vbool2_t test_shortcut_for_riscv_vmxnor_case_1(vbool2_t v1, size_t
> > +vl) {
> > + return __riscv_vmxnor_mm_b2(v1, v1, vl); }
> > +
> > +vbool4_t test_shortcut_for_riscv_vmxnor_case_2(vbool4_t v1, size_t
> > +vl) {
> > + return __riscv_vmxnor_mm_b4(v1, v1, vl); }
> > +
> > +vbool8_t test_shortcut_for_riscv_vmxnor_case_3(vbool8_t v1, size_t
> > +vl) {
> > + return __riscv_vmxnor_mm_b8(v1, v1, vl); }
> > +
> > +vbool16_t test_shortcut_for_riscv_vmxnor_case_4(vbool16_t v1,
> > +size_t
> > +vl) {
> > + return __riscv_vmxnor_mm_b16(v1, v1, vl); }
> > +
> > +vbool32_t test_shortcut_for_riscv_vmxnor_case_5(vbool32_t v1,
> > +size_t
> > +vl) {
> > + return __riscv_vmxnor_mm_b32(v1, v1, vl); }
> > +
> > +vbool64_t test_shortcut_for_riscv_vmxnor_case_6(vbool64_t v1,
> > +size_t
> > +vl) {
> > + return __riscv_vmxnor_mm_b64(v1, v1, vl); }
> > +
> > +/* { dg-final { scan-assembler-not {vmand\.mm\s+v[0-9]+,\s*v[0-9]+}
> > +} } */
> > +/* { dg-final { scan-assembler-not
> > +{vmnand\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
> > +/* { dg-final { scan-assembler-not
> > +{vmnandn\.mm\s+v[0-9]+,\s*v[0-9]+}
> > +} } */
> > +/* { dg-final { scan-assembler-not {vmxor\.mm\s+v[0-9]+,\s*v[0-9]+}
> > +} } */
> > +/* { dg-final { scan-assembler-not {vmor\.mm\s+v[0-9]+,\s*v[0-9]+}
> > +} } */
> > +/* { dg-final { scan-assembler-not {vmnor\.mm\s+v[0-9]+,\s*v[0-9]+}
> > +} } */
> > +/* { dg-final { scan-assembler-times
> > +{vmorn\.mm\s+v[0-9]+,\s*v[0-9]+,\s*v[0-9]+} 7 } } */
> > +/* { dg-final { scan-assembler-not
> > +{vmxnor\.mm\s+v[0-9]+,\s*v[0-9]+} } } */
> > +/* { dg-final { scan-assembler-times {vmclr\.m\s+v[0-9]+} 14 } } */
> > +/* { dg-final { scan-assembler-times {vmset\.m\s+v[0-9]+} 7 } } */
> > --
> > 2.34.1
> >
> >
^ permalink raw reply [flat|nested] 12+ messages in thread
end of thread, other threads:[~2023-04-17 1:55 UTC | newest]
Thread overview: 12+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-04-14 2:32 [PATCH] RISC-V: Add test cases for the RVV mask insn shortcut pan2.li
2023-04-14 2:35 ` juzhe.zhong
2023-04-14 2:37 ` Li, Pan2
2023-04-14 2:45 ` [PATCH v2] " pan2.li
2023-04-14 2:46 ` juzhe.zhong
2023-04-14 3:26 ` Li, Pan2
2023-04-14 6:41 ` Kito Cheng
2023-04-14 6:47 ` Li, Pan2
2023-04-17 1:46 ` Li, Pan2
2023-04-17 1:52 ` Kito Cheng
2023-04-17 1:55 ` Li, Pan2
2023-04-14 3:25 ` [PATCH v3] " pan2.li
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).