arm: vst1_types_x3 ACLE intrinsics

This patch is part of a series of patches implementing the _xN
variants of the vst1 intrinsic for the arm port. This patch adds the
_x3 variants of the vst1 intrinsic.

ACLE documents:
https://developer.arm.com/documentation/ihi0053/latest/

ISA documents:
https://developer.arm.com/documentation/ddi0487/latest/

gcc/ChangeLog:
	* config/arm/arm_neon.h
	(vst1_u8_x3, vst1_u16_x3, vst1_u32_x3, vst1_u64_x3): New.
	(vst1_s8_x3, vst1_s16_x3, vst1_s32_x3, vst1_s64_x3): New.
	(vst1_f16_x3, vst1_f32_x3): New.
	(vst1_p8_x3, vst1_p16_x3, vst1_p64_x3): New.
	(vst1_bf16_x3): New.
	* config/arm/arm_neon_builtins.def (vst1_x3): New entries.
	* config/arm/neon.md (vst1_x3<mode>): New.

gcc/testsuite/ChangeLog:
	* gcc.target/arm/simd/vst1_base_xN_1.c: Add new test.
	* gcc.target/arm/simd/vst1_bf16_xN_1.c: Add new test.
	* gcc.target/arm/simd/vst1_fp16_xN_1.c: Add new test.
	* gcc.target/arm/simd/vst1_p64_xN_1.c: Add new test.

4 files changed, 77 insertions, 7 deletions

diff --git a/gcc/testsuite/gcc.target/arm/simd/vst1_base_xN_1.c b/gcc/testsuite/gcc.target/arm/simd/vst1_base_xN_1.c
index 575897f..5f820a6 100644
--- a/gcc/testsuite/gcc.target/arm/simd/vst1_base_xN_1.c
+++ b/gcc/testsuite/gcc.target/arm/simd/vst1_base_xN_1.c
@@ -60,8 +60,63 @@ void test_vst1_p16_x2 (poly16_t * ptr, poly16x4x2_t val)
     vst1_p16_x2 (ptr, val);
 }
 
+void test_vst1_u8_x3 (uint8_t * ptr, uint8x8x3_t val)
+{
+    vst1_u8_x3 (ptr, val);
+}
+
+void test_vst1_u16_x3 (uint16_t * ptr, uint16x4x3_t val)
+{
+    vst1_u16_x3 (ptr, val);
+}
+
+void test_vst1_u32_x3 (uint32_t * ptr, uint32x2x3_t val)
+{
+    vst1_u32_x3 (ptr, val);
+}
+
+void test_vst1_u64_x3 (uint64_t * ptr, uint64x1x3_t val)
+{
+    vst1_u64_x3 (ptr, val);
+}
+
+void test_vst1_s8_x3 (int8_t * ptr, int8x8x3_t val)
+{
+    vst1_s8_x3 (ptr, val);
+}
+
+void test_vst1_s16_x3 (int16_t * ptr, int16x4x3_t val)
+{
+    vst1_s16_x3 (ptr, val);
+}
+
+void test_vst1_s32_x3 (int32_t * ptr, int32x2x3_t val)
+{
+    vst1_s32_x3 (ptr, val);
+}
+
+void test_vst1_s64_x3 (int64_t * ptr, int64x1x3_t val)
+{
+    vst1_s64_x3 (ptr, val);
+}
+
+void test_vst1_f32_x3 (float32_t * ptr, float32x2x3_t val)
+{
+    vst1_f32_x3 (ptr, val);
+}
+
+void test_vst1_p8_x3 (poly8_t * ptr, poly8x8x3_t val)
+{
+    vst1_p8_x3 (ptr, val);
+}
+
+void test_vst1_p16_x3 (poly16_t * ptr, poly16x4x3_t val)
+{
+    vst1_p16_x3 (ptr, val);
+}
+
 
-/* { dg-final { scan-assembler-times {vst1.8\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 3 } }  */
-/* { dg-final { scan-assembler-times {vst1.16\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 3 } }  */
-/* { dg-final { scan-assembler-times {vst1.32\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 3 } }  */
-/* { dg-final { scan-assembler-times {vst1.64\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+:64\]\n} 2 } }  */
+/* { dg-final { scan-assembler-times {vst1.8\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 6 } }  */
+/* { dg-final { scan-assembler-times {vst1.16\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 6 } }  */
+/* { dg-final { scan-assembler-times {vst1.32\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 6 } }  */
+/* { dg-final { scan-assembler-times {vst1.64\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+:64\]\n} 4 } }  */
diff --git a/gcc/testsuite/gcc.target/arm/simd/vst1_bf16_xN_1.c b/gcc/testsuite/gcc.target/arm/simd/vst1_bf16_xN_1.c
index 213fd20..a3a00ea 100644
--- a/gcc/testsuite/gcc.target/arm/simd/vst1_bf16_xN_1.c
+++ b/gcc/testsuite/gcc.target/arm/simd/vst1_bf16_xN_1.c
@@ -10,4 +10,9 @@ void test_vst1_bf16_x2 (bfloat16_t * ptr, bfloat16x4x2_t val)
     vst1_bf16_x2 (ptr, val);
 }
 
-/* { dg-final { scan-assembler-times {vst1.16\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 1 } }  */
+void test_vst1_bf16_x3 (bfloat16_t * ptr, bfloat16x4x3_t val)
+{
+    vst1_bf16_x3 (ptr, val);
+}
+
+/* { dg-final { scan-assembler-times {vst1.16\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 2 } }  */
diff --git a/gcc/testsuite/gcc.target/arm/simd/vst1_fp16_xN_1.c b/gcc/testsuite/gcc.target/arm/simd/vst1_fp16_xN_1.c
index 523aec9..0a6863e 100644
--- a/gcc/testsuite/gcc.target/arm/simd/vst1_fp16_xN_1.c
+++ b/gcc/testsuite/gcc.target/arm/simd/vst1_fp16_xN_1.c
@@ -10,4 +10,9 @@ void test_vst1_f16_x2 (float16_t * ptr, float16x4x2_t val)
     vst1_f16_x2 (ptr, val);
 }
 
-/* { dg-final { scan-assembler-times {vst1.16\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 1 } }  */
+void test_vst1_f16_x3 (float16_t * ptr, float16x4x3_t val)
+{
+    vst1_f16_x3 (ptr, val);
+}
+
+/* { dg-final { scan-assembler-times {vst1.16\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 2 } }  */
diff --git a/gcc/testsuite/gcc.target/arm/simd/vst1_p64_xN_1.c b/gcc/testsuite/gcc.target/arm/simd/vst1_p64_xN_1.c
index f590ebd..5dbd604 100644
--- a/gcc/testsuite/gcc.target/arm/simd/vst1_p64_xN_1.c
+++ b/gcc/testsuite/gcc.target/arm/simd/vst1_p64_xN_1.c
@@ -10,4 +10,9 @@ void test_vst1_p64_x2 (poly64_t * ptr, poly64x1x2_t val)
     vst1_p64_x2 (ptr, val);
 }
 
-/* { dg-final { scan-assembler-times {vst1.64\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+:64\]\n} 1 } }  */
\ No newline at end of file
+void test_vst1_p64_x3 (poly64_t * ptr, poly64x1x3_t val)
+{
+    vst1_p64_x3 (ptr, val);
+}
+
+/* { dg-final { scan-assembler-times {vst1.64\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+:64\]\n} 2 } }  */
\ No newline at end of file