[PATCH v3 08/12] [GCC] arm: vst1q_types_x3 ACLE intrinsics

[<Ezra.Sitorus@arm.com>]

From: Ezra Sitorus <ezra.sitorus@arm.com>

This patch is part of a series of patches implementing the _xN
variants of the vst1q intrinsic for the arm port. This patch adds the
_x3 variants of the vst1q 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
	(vst1q_u8_x3, vst1q_u16_x3, vst1q_u32_x3, vst1q_u64_x3): New.
	(vst1q_s8_x3, vst1q_s16_x3, vst1q_s32_x3, vst1q_s64_x3): New.
	(vst1q_f16_x3, vst1q_f32_x3): New.
	(vst1q_p8_x3, vst1q_p16_x3, vst1q_p64_x3): New.
	(vst1q_bf16_x3): New.
	* config/arm/arm_neon_builtins.def (vst1q_x3): New entries.
	* config/arm/neon.md
	(neon_vst1q_x3<mode>): New.
	(neon_vld1x3qa<mode>, neon_vst1x3qb<mode>): New.
	* config/arm/unspecs.md
	(UNSPEC_VST1X3A, UNSPEC_VST1X3B): New.

gcc/testsuite/ChangeLog:
	* gcc.target/arm/simd/vst1q_base_xN_1.c: Add new tests.
	* gcc.target/arm/simd/vst1q_bf16_xN_1.c: Add new tests.
	* gcc.target/arm/simd/vst1q_fp16_xN_1.c: Add new tests.
	* gcc.target/arm/simd/vst1q_p64_xN_1.c: Add new tests.
---
 gcc/config/arm/arm_neon.h                     | 114 ++++++++++++++++++
 gcc/config/arm/arm_neon_builtins.def          |   1 +
 gcc/config/arm/neon.md                        |  47 ++++++++
 gcc/config/arm/unspecs.md                     |   2 +
 .../gcc.target/arm/simd/vst1q_base_xN_1.c     |  68 ++++++++++-
 .../gcc.target/arm/simd/vst1q_bf16_xN_1.c     |   8 +-
 .../gcc.target/arm/simd/vst1q_fp16_xN_1.c     |   8 +-
 .../gcc.target/arm/simd/vst1q_p64_xN_1.c      |   8 +-
 8 files changed, 249 insertions(+), 7 deletions(-)

diff --git a/gcc/config/arm/arm_neon.h b/gcc/config/arm/arm_neon.h
index 1c447b6d42f..5cec7dd876f 100644
--- a/gcc/config/arm/arm_neon.h
+++ b/gcc/config/arm/arm_neon.h
@@ -11359,6 +11359,38 @@ vst1q_s64_x2 (int64_t * __a, int64x2x2_t __b)
   __builtin_neon_vst1q_x2v2di ((__builtin_neon_di *) __a, __bu.__o);
 }
 
+__extension__ extern __inline void
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vst1q_s8_x3 (int8_t * __a, int8x16x3_t __b)
+{
+  union { int8x16x3_t __i; __builtin_neon_ci __o; } __bu = { __b };
+  __builtin_neon_vst1q_x3v16qi ((__builtin_neon_qi *) __a, __bu.__o);
+}
+
+__extension__ extern __inline void
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vst1q_s16_x3 (int16_t * __a, int16x8x3_t __b)
+{
+  union { int16x8x3_t __i; __builtin_neon_ci __o; } __bu = { __b };
+  __builtin_neon_vst1q_x3v8hi ((__builtin_neon_hi *) __a, __bu.__o);
+}
+
+__extension__ extern __inline void
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vst1q_s32_x3 (int32_t * __a, int32x4x3_t __b)
+{
+  union { int32x4x3_t __i; __builtin_neon_ci __o; } __bu = { __b };
+  __builtin_neon_vst1q_x3v4si ((__builtin_neon_si *) __a, __bu.__o);
+}
+
+__extension__ extern __inline void
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vst1q_s64_x3 (int64_t * __a, int64x2x3_t __b)
+{
+  union { int64x2x3_t __i; __builtin_neon_ci __o; } __bu = { __b };
+  __builtin_neon_vst1q_x3v2di ((__builtin_neon_di *) __a, __bu.__o);
+}
+
 __extension__ extern __inline void
 __attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_s8_x3 (int8_t * __a, int8x8x3_t __b)
@@ -11696,6 +11728,14 @@ vst1q_p64_x2 (poly64_t * __a, poly64x2x2_t __b)
   __builtin_neon_vst1q_x2v2di ((__builtin_neon_di *) __a, __bu.__o);
 }
 
+__extension__ extern __inline void
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vst1q_p64_x3 (poly64_t * __a, poly64x2x3_t __b)
+{
+  union { poly64x2x3_t __i; __builtin_neon_ci __o; } __bu = { __b };
+  __builtin_neon_vst1q_x3v2di ((__builtin_neon_di *) __a, __bu.__o);
+}
+
 #pragma GCC pop_options
 __extension__ extern __inline void
 __attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
@@ -11759,6 +11799,24 @@ vst1q_f32_x2 (float32_t * __a, float32x4x2_t __b)
   __builtin_neon_vst1q_x2v4sf (__a, __bu.__o);
 }
 
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+__extension__ extern __inline void
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vst1q_f16_x3 (float16_t * __a, float16x8x3_t __b)
+{
+  union { float16x8x3_t __i; __builtin_neon_ci __o; } __bu = { __b };
+  __builtin_neon_vst1q_x3v8hf (__a, __bu.__o);
+}
+#endif
+
+__extension__ extern __inline void
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vst1q_f32_x3 (float32_t * __a, float32x4x3_t __b)
+{
+  union { float32x4x3_t __i; __builtin_neon_ci __o; } __bu = { __b };
+  __builtin_neon_vst1q_x3v4sf (__a, __bu.__o);
+}
+
 __extension__ extern __inline void
 __attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_u8 (uint8_t * __a, uint8x16_t __b)
@@ -11819,6 +11877,38 @@ vst1q_u64_x2 (uint64_t * __a, uint64x2x2_t __b)
   __builtin_neon_vst1q_x2v2di ((__builtin_neon_di *) __a, __bu.__o);
 }
 
+__extension__ extern __inline void
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vst1q_u8_x3 (uint8_t * __a, uint8x16x3_t __b)
+{
+  union { uint8x16x3_t __i; __builtin_neon_ci __o; } __bu = { __b };
+  __builtin_neon_vst1q_x3v16qi ((__builtin_neon_qi *) __a, __bu.__o);
+}
+
+__extension__ extern __inline void
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vst1q_u16_x3 (uint16_t * __a, uint16x8x3_t __b)
+{
+  union { uint16x8x3_t __i; __builtin_neon_ci __o; } __bu = { __b };
+  __builtin_neon_vst1q_x3v8hi ((__builtin_neon_hi *) __a, __bu.__o);
+}
+
+__extension__ extern __inline void
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vst1q_u32_x3 (uint32_t * __a, uint32x4x3_t __b)
+{
+  union { uint32x4x3_t __i; __builtin_neon_ci __o; } __bu = { __b };
+  __builtin_neon_vst1q_x3v4si ((__builtin_neon_si *) __a, __bu.__o);
+}
+
+__extension__ extern __inline void
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vst1q_u64_x3 (uint64_t * __a, uint64x2x3_t __b)
+{
+  union { uint64x2x3_t __i; __builtin_neon_ci __o; } __bu = { __b };
+  __builtin_neon_vst1q_x3v2di ((__builtin_neon_di *) __a, __bu.__o);
+}
+
 __extension__ extern __inline void
 __attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_p8 (poly8_t * __a, poly8x16_t __b)
@@ -11849,6 +11939,22 @@ vst1q_p16_x2 (poly16_t * __a, poly16x8x2_t __b)
   __builtin_neon_vst1q_x2v8hi ((__builtin_neon_hi *) __a, __bu.__o);
 }
 
+__extension__ extern __inline void
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vst1q_p8_x3 (poly8_t * __a, poly8x16x3_t __b)
+{
+  union { poly8x16x3_t __i; __builtin_neon_ci __o; } __bu = { __b };
+  __builtin_neon_vst1q_x3v16qi ((__builtin_neon_qi *) __a, __bu.__o);
+}
+
+__extension__ extern __inline void
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vst1q_p16_x3 (poly16_t * __a, poly16x8x3_t __b)
+{
+  union { poly16x8x3_t __i; __builtin_neon_ci __o; } __bu = { __b };
+  __builtin_neon_vst1q_x3v8hi ((__builtin_neon_hi *) __a, __bu.__o);
+}
+
 __extension__ extern __inline void
 __attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_lane_s8 (int8_t * __a, int8x8_t __b, const int __c)
@@ -20533,6 +20639,14 @@ vst1q_bf16_x2 (bfloat16_t * __a, bfloat16x8x2_t __b)
   __builtin_neon_vst1q_x2v8bf (__a, __bu.__o);
 }
 
+__extension__ extern __inline void
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vst1q_bf16_x3 (bfloat16_t * __a, bfloat16x8x3_t __b)
+{
+  union { bfloat16x8x3_t __i; __builtin_neon_ci __o; } __bu = { __b };
+  __builtin_neon_vst1q_x3v8bf (__a, __bu.__o);
+}
+
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_bf16 (bfloat16_t * __ptr, bfloat16x4x2_t __val)
diff --git a/gcc/config/arm/arm_neon_builtins.def b/gcc/config/arm/arm_neon_builtins.def
index d44abb80139..f8ff87c0c7a 100644
--- a/gcc/config/arm/arm_neon_builtins.def
+++ b/gcc/config/arm/arm_neon_builtins.def
@@ -314,6 +314,7 @@ VAR14 (STORE1, vst1,
 VAR7 (STORE1, vst1_x2, v8qi, v4hi, v2si, di, v4hf, v2sf, v4bf)
 VAR7 (STORE1, vst1q_x2, v16qi, v8hi, v4si, v2di, v8hf, v4sf, v8bf)
 VAR7 (STORE1, vst1_x3, v8qi, v4hi, v2si, di, v4hf, v2sf, v4bf)
+VAR7 (STORE1, vst1q_x3, v16qi, v8hi, v4si, v2di, v8hf, v4sf, v8bf)
 VAR7 (STORE1, vst1_x4, v8qi, v4hi, v2si, di, v4hf, v2sf, v4bf)
 VAR14 (STORE1LANE, vst1_lane,
        v8qi, v4hi, v4hf, v2si, v2sf, di, v16qi, v8hi, v8hf, v4si, v4sf, v2di, v4bf, v8bf)
diff --git a/gcc/config/arm/neon.md b/gcc/config/arm/neon.md
index 270130c4086..a311ee0c997 100644
--- a/gcc/config/arm/neon.md
+++ b/gcc/config/arm/neon.md
@@ -5192,6 +5192,53 @@ if (BYTES_BIG_ENDIAN)
   [(set_attr "type" "neon_store1_3reg<q>")]
 )
 
+(define_expand "neon_vst1q_x3<mode>"
+  [(match_operand:CI 0 "neon_struct_operand")
+   (match_operand:CI 1 "s_register_operand")
+   (unspec:VDQX [(const_int 0)] UNSPEC_VSTRUCTDUMMY)]
+  "TARGET_NEON"
+{
+  rtx mem = adjust_address (operands[0], EImode, 0);
+  emit_insn (gen_neon_vst1x3qa<mode> (mem, operands[1]));
+  mem = adjust_address (mem, EImode, GET_MODE_SIZE (EImode));
+  emit_insn (gen_neon_vst1x3qb<mode> (mem, operands[1]));
+  DONE;
+})
+
+(define_insn "neon_vst1x3qa<mode>"
+  [(set (match_operand:EI 0 "neon_struct_operand" "=Um")
+        (unspec:EI [(match_operand:CI 1 "s_register_operand" "w")
+                    (unspec:VDQX [(const_int 0)] UNSPEC_VSTRUCTDUMMY)]
+                   UNSPEC_VST1X3A))]
+  "TARGET_NEON"
+{
+  rtx ops[2];
+  ops[0] = operands[0];
+  ops[1] = gen_rtx_REG (EImode, REGNO (operands[1]));
+
+  output_asm_insn ("vst1.<V_sz_elem>\t%h1, %A0", ops);
+  return "";
+}
+  [(set_attr "type" "neon_store1_3reg<q>")]
+)
+
+(define_insn "neon_vst1x3qb<mode>"
+  [(set (match_operand:EI 0 "neon_struct_operand" "=Um")
+        (unspec:EI [(match_operand:CI 1 "s_register_operand" "w")
+                    (unspec:VDQX [(const_int 0)] UNSPEC_VSTRUCTDUMMY)]
+                   UNSPEC_VST1X3B))]
+  "TARGET_NEON"
+{
+  rtx ops[2];
+  ops[0] = operands[0];
+  ops[1] = gen_rtx_REG (EImode, REGNO (operands[1]) + 6);
+
+  output_asm_insn ("vst1.<V_sz_elem>\t%h1, %A0", ops);
+  return "";
+}
+  [(set_attr "type" "neon_store1_3reg<q>")]
+)
+
 (define_insn "neon_vst1_x4<mode>"
   [(set (match_operand:OI 0 "neon_struct_operand" "=Um")
         (unspec:OI [(match_operand:OI 1 "s_register_operand" "w")
diff --git a/gcc/config/arm/unspecs.md b/gcc/config/arm/unspecs.md
index 4753968d52f..573bdf00944 100644
--- a/gcc/config/arm/unspecs.md
+++ b/gcc/config/arm/unspecs.md
@@ -463,6 +463,8 @@
   UNSPEC_VRSRA_U_N
   UNSPEC_VSRI
   UNSPEC_VST1
+  UNSPEC_VST1X3A
+  UNSPEC_VST1X3B
   UNSPEC_VST1_LANE
   UNSPEC_VST2
   UNSPEC_VST2_LANE
diff --git a/gcc/testsuite/gcc.target/arm/simd/vst1q_base_xN_1.c b/gcc/testsuite/gcc.target/arm/simd/vst1q_base_xN_1.c
index 232feafade0..e9b7d67f06a 100644
--- a/gcc/testsuite/gcc.target/arm/simd/vst1q_base_xN_1.c
+++ b/gcc/testsuite/gcc.target/arm/simd/vst1q_base_xN_1.c
@@ -61,10 +61,70 @@ void test_vst1q_p16_x2 (poly16_t * ptr, poly16x8x2_t val)
     vst1q_p16_x2 (ptr, val);
 }
 
-/* { dg-final { scan-assembler-times {vst1.8\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 3 } }  */
+void test_vst1q_u8_x3 (uint8_t * ptr, uint8x16x3_t val)
+{
+    vst1q_u8_x3 (ptr, val);
+}
+
+void test_vst1q_u16_x3 (uint16_t * ptr, uint16x8x3_t val)
+{
+    vst1q_u16_x3 (ptr, val);
+}
+
+void test_vst1q_u32_x3 (uint32_t * ptr, uint32x4x3_t val)
+{
+    vst1q_u32_x3 (ptr, val);
+}
+
+void test_vst1q_u64_x3 (uint64_t * ptr, uint64x2x3_t val)
+{
+    vst1q_u64_x3 (ptr, val);
+}
+
+void test_vst1q_s8_x3 (int8_t * ptr, int8x16x3_t val)
+{
+    vst1q_s8_x3 (ptr, val);
+}
+
+void test_vst1q_s16_x3 (int16_t * ptr, int16x8x3_t val)
+{
+    vst1q_s16_x3 (ptr, val);
+}
+
+void test_vst1q_s32_x3 (int32_t * ptr, int32x4x3_t val)
+{
+    vst1q_s32_x3 (ptr, val);
+}
+
+void test_vst1q_s64_x3 (int64_t * ptr, int64x2x3_t val)
+{
+    vst1q_s64_x3 (ptr, val);
+}
+
+void test_vst1q_f32_x3 (float32_t * ptr, float32x4x3_t val)
+{
+    vst1q_f32_x3 (ptr, val);
+}
+
+void test_vst1q_p8_x3 (poly8_t * ptr, poly8x16x3_t val)
+{
+    vst1q_p8_x3 (ptr, val);
+}
+
+void test_vst1q_p16_x3 (poly16_t * ptr, poly16x8x3_t val)
+{
+    vst1q_p16_x3 (ptr, val);
+}
+
+
+/* { dg-final { scan-assembler-times {vst1.8\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 6 } }  */
+/* { dg-final { scan-assembler-times {vst1.8\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 {vst1.16\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} 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 {vst1.32\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} 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 {vst1.64\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+:64\]\n} 4 } }  */
+/* { dg-final { scan-assembler-times {vst1.64\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+:64\]!\n} 2 } }  */
diff --git a/gcc/testsuite/gcc.target/arm/simd/vst1q_bf16_xN_1.c b/gcc/testsuite/gcc.target/arm/simd/vst1q_bf16_xN_1.c
index 2a4579f0aae..44d69522dff 100644
--- a/gcc/testsuite/gcc.target/arm/simd/vst1q_bf16_xN_1.c
+++ b/gcc/testsuite/gcc.target/arm/simd/vst1q_bf16_xN_1.c
@@ -10,4 +10,10 @@ void test_vst1q_bf16_x2 (bfloat16_t * ptr, bfloat16x8x2_t val)
     vst1q_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_vst1q_bf16_x3 (bfloat16_t * ptr, bfloat16x8x3_t val)
+{
+    vst1q_bf16_x3 (ptr, val);
+}
+
+/* { dg-final { scan-assembler-times {vst1.16\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 2 } }  */
+/* { dg-final { scan-assembler-times {vst1.16\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]!\n} 1 } }  */
\ No newline at end of file
diff --git a/gcc/testsuite/gcc.target/arm/simd/vst1q_fp16_xN_1.c b/gcc/testsuite/gcc.target/arm/simd/vst1q_fp16_xN_1.c
index 61a7e558c48..f0ac2d35c1c 100644
--- a/gcc/testsuite/gcc.target/arm/simd/vst1q_fp16_xN_1.c
+++ b/gcc/testsuite/gcc.target/arm/simd/vst1q_fp16_xN_1.c
@@ -10,4 +10,10 @@ void test_vst1q_f16_x2 (float16_t * ptr, float16x8x2_t val)
     vst1q_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_vst1q_f16_x3 (float16_t * ptr, float16x8x3_t val)
+{
+    vst1q_f16_x3 (ptr, val);
+}
+
+/* { dg-final { scan-assembler-times {vst1.16\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 2 } }  */
+/* { dg-final { scan-assembler-times {vst1.16\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]!\n} 1 } }  */
diff --git a/gcc/testsuite/gcc.target/arm/simd/vst1q_p64_xN_1.c b/gcc/testsuite/gcc.target/arm/simd/vst1q_p64_xN_1.c
index 82f3dad293c..24fdf3c0c1b 100644
--- a/gcc/testsuite/gcc.target/arm/simd/vst1q_p64_xN_1.c
+++ b/gcc/testsuite/gcc.target/arm/simd/vst1q_p64_xN_1.c
@@ -10,4 +10,10 @@ void test_vst1q_p64_x2 (poly64_t * ptr, poly64x2x2_t val)
     vst1q_p64_x2 (ptr, val);
 }
 
-/* { dg-final { scan-assembler-times {vst1.64\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+:64\]\n} 1 } }  */
+void test_vst1q_p64_x3 (poly64_t * ptr, poly64x2x3_t val)
+{
+    vst1q_p64_x3 (ptr, val);
+}
+
+/* { 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 {vst1.64\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+:64\]!\n} 1 } }  */
-- 
2.25.1