arm: vld1q_types_x2 ACLE intrinsics

This patch is part of a series of patches implementing the _xN
variants of the vld1q intrinsic for the arm port. This patch adds the
_x2 variants of the vld1q 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
	(vld1q_u8_x2, vld1q_u16_x2, vld1q_u32_x2, vld1q_u64_x2): New.
	(vld1q_s8_x2, vld1q_s16_x2, vld1q_s32_x2, vld1q_s64_x2): New.
	(vld1q_f16_x2, vld1q_f32_x2): New.
	(vld1q_p8_x2, vld1q_p16_x2, vld1q_p64_x2): New.
	(vld1q_bf16_x2): New.
	* config/arm/arm_neon_builtins.def (vld1_x2): New entries.
	* config/arm/neon.md (vld1_x2<mode>): New.

gcc/testsuite/ChangeLog:
	* gcc.target/arm/simd/vld1q_base_xN_1.c: Add new test.
	* gcc.target/arm/simd/vld1q_bf16_xN_1.c: Add new test.
	* gcc.target/arm/simd/vld1q_fp16_xN_1.c: Add new test.
	* gcc.target/arm/simd/vld1q_p64_xN_1.c: Add new test.

7 files changed, 247 insertions, 0 deletions

diff --git a/gcc/config/arm/arm_neon.h b/gcc/config/arm/arm_neon.h
index cdfdb44..3eb41c6 100644
--- a/gcc/config/arm/arm_neon.h
+++ b/gcc/config/arm/arm_neon.h
@@ -10403,6 +10403,15 @@ vld1q_p64 (const poly64_t * __a)
   return (poly64x2_t)__builtin_neon_vld1v2di ((const __builtin_neon_di *) __a);
 }
 
+__extension__ extern __inline poly64x2x2_t
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vld1q_p64_x2 (const poly64_t * __a)
+{
+  union { poly64x2x2_t __i; __builtin_neon_oi __o; } __rv;
+  __rv.__o = __builtin_neon_vld1_x2v2di ((const __builtin_neon_di *) __a);
+  return __rv.__i;
+}
+
 #pragma GCC pop_options
 __extension__ extern __inline int8x16_t
 __attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
@@ -10432,6 +10441,42 @@ vld1q_s64 (const int64_t * __a)
   return (int64x2_t)__builtin_neon_vld1v2di ((const __builtin_neon_di *) __a);
 }
 
+__extension__ extern __inline int8x16x2_t
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vld1q_s8_x2 (const int8_t * __a)
+{
+  union { int8x16x2_t __i; __builtin_neon_oi __o; } __rv;
+  __rv.__o = __builtin_neon_vld1_x2v16qi ((const __builtin_neon_qi *) __a);
+  return __rv.__i;
+}
+
+__extension__ extern __inline int16x8x2_t
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vld1q_s16_x2 (const int16_t * __a)
+{
+  union { int16x8x2_t __i; __builtin_neon_oi __o; } __rv;
+  __rv.__o = __builtin_neon_vld1_x2v8hi ((const __builtin_neon_hi *) __a);
+  return __rv.__i;
+}
+
+__extension__ extern __inline int32x4x2_t
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vld1q_s32_x2 (const int32_t * __a)
+{
+  union { int32x4x2_t __i; __builtin_neon_oi __o; } __rv;
+  __rv.__o = __builtin_neon_vld1_x2v4si ((const __builtin_neon_si *) __a);
+  return __rv.__i;
+}
+
+__extension__ extern __inline int64x2x2_t
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vld1q_s64_x2 (const int64_t * __a)
+{
+  union { int64x2x2_t __i; __builtin_neon_oi __o; } __rv;
+  __rv.__o = __builtin_neon_vld1_x2v2di ((const __builtin_neon_di *) __a);
+  return __rv.__i;
+}
+
 #if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
 __extension__ extern __inline float16x8_t
 __attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
@@ -10448,6 +10493,26 @@ vld1q_f32 (const float32_t * __a)
   return (float32x4_t)__builtin_neon_vld1v4sf ((const __builtin_neon_sf *) __a);
 }
 
+#if defined (__ARM_FP16_FORMAT_IEEE) || defined (__ARM_FP16_FORMAT_ALTERNATIVE)
+__extension__ extern __inline float16x8x2_t
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vld1q_f16_x2 (const float16_t * __a)
+{
+  union { float16x8x2_t __i; __builtin_neon_oi __o; } __rv;
+  __rv.__o = __builtin_neon_vld1_x2v8hf (__a);
+  return __rv.__i;
+}
+#endif
+
+__extension__ extern __inline float32x4x2_t
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vld1q_f32_x2 (const float32_t * __a)
+{
+  union { float32x4x2_t __i; __builtin_neon_oi __o; } __rv;
+  __rv.__o = __builtin_neon_vld1_x2v4sf ((const __builtin_neon_sf *) __a);
+  return __rv.__i;
+}
+
 __extension__ extern __inline uint8x16_t
 __attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u8 (const uint8_t * __a)
@@ -10476,6 +10541,42 @@ vld1q_u64 (const uint64_t * __a)
   return (uint64x2_t)__builtin_neon_vld1v2di ((const __builtin_neon_di *) __a);
 }
 
+__extension__ extern __inline uint8x16x2_t
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vld1q_u8_x2 (const uint8_t * __a)
+{
+  union { uint8x16x2_t __i; __builtin_neon_oi __o; } __rv;
+  __rv.__o = __builtin_neon_vld1_x2v16qi ((const __builtin_neon_qi *) __a);
+  return __rv.__i;
+}
+
+__extension__ extern __inline uint16x8x2_t
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vld1q_u16_x2 (const uint16_t * __a)
+{
+  union { uint16x8x2_t __i; __builtin_neon_oi __o; } __rv;
+  __rv.__o = __builtin_neon_vld1_x2v8hi ((const __builtin_neon_hi *) __a);
+  return __rv.__i;
+}
+
+__extension__ extern __inline uint32x4x2_t
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vld1q_u32_x2 (const uint32_t * __a)
+{
+  union { uint32x4x2_t __i; __builtin_neon_oi __o; } __rv;
+  __rv.__o = __builtin_neon_vld1_x2v4si ((const __builtin_neon_si *) __a);
+  return __rv.__i;
+}
+
+__extension__ extern __inline uint64x2x2_t
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vld1q_u64_x2 (const uint64_t * __a)
+{
+  union { uint64x2x2_t __i; __builtin_neon_oi __o; } __rv;
+  __rv.__o = __builtin_neon_vld1_x2v2di ((const __builtin_neon_di *) __a);
+  return __rv.__i;
+}
+
 __extension__ extern __inline poly8x16_t
 __attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_p8 (const poly8_t * __a)
@@ -10490,6 +10591,24 @@ vld1q_p16 (const poly16_t * __a)
   return (poly16x8_t)__builtin_neon_vld1v8hi ((const __builtin_neon_hi *) __a);
 }
 
+__extension__ extern __inline poly8x16x2_t
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vld1q_p8_x2 (const poly8_t * __a)
+{
+  union { poly8x16x2_t __i; __builtin_neon_oi __o; } __rv;
+  __rv.__o = __builtin_neon_vld1_x2v16qi ((const __builtin_neon_qi *) __a);
+  return __rv.__i;
+}
+
+__extension__ extern __inline poly16x8x2_t
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vld1q_p16_x2 (const poly16_t * __a)
+{
+  union { poly16x8x2_t __i; __builtin_neon_oi __o; } __rv;
+  __rv.__o = __builtin_neon_vld1_x2v8hi ((const __builtin_neon_hi *) __a);
+  return __rv.__i;
+}
+
 __extension__ extern __inline int8x8_t
 __attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_lane_s8 (const int8_t * __a, int8x8_t __b, const int __c)
@@ -19782,6 +19901,15 @@ vld1q_bf16 (const bfloat16_t * __ptr)
   return __builtin_neon_vld1v8bf (__ptr);
 }
 
+__extension__ extern __inline bfloat16x8x2_t
+__attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
+vld1q_bf16_x2 (const bfloat16_t * __ptr)
+{
+  union { bfloat16x8x2_t __i; __builtin_neon_oi __o; } __rv;
+  __rv.__o = __builtin_neon_vld1_x2v8bf ((const __builtin_neon_bf *) __ptr);
+  return __rv.__i;
+}
+
 __extension__ extern __inline bfloat16x4x2_t
 __attribute__  ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_bf16 (bfloat16_t const * __ptr)
diff --git a/gcc/config/arm/arm_neon_builtins.def b/gcc/config/arm/arm_neon_builtins.def
index 94b1523..6a8f0cb 100644
--- a/gcc/config/arm/arm_neon_builtins.def
+++ b/gcc/config/arm/arm_neon_builtins.def
@@ -301,6 +301,7 @@ VAR1 (TERNOP, vtbx4, v8qi)
 VAR13 (LOAD1, vld1,
         v8qi, v4hi, v4hf, v2si, v2sf, v16qi, v8hi, v8hf, v4si, v4sf, v2di,
         v4bf, v8bf)
+VAR7 (LOAD1, vld1_x2, v16qi, v8hi, v4si, v2di, v8hf, v4sf, v8bf)
 VAR12 (LOAD1LANE, vld1_lane,
 	v8qi, v4hi, v2si, v2sf, di, v16qi, v8hi, v4si, v4sf, v2di, v4bf, v8bf)
 VAR10 (LOAD1, vld1_dup,
diff --git a/gcc/config/arm/neon.md b/gcc/config/arm/neon.md
index d213369..55049ea 100644
--- a/gcc/config/arm/neon.md
+++ b/gcc/config/arm/neon.md
@@ -4957,6 +4957,16 @@ if (BYTES_BIG_ENDIAN)
   [(set_attr "type" "neon_load1_1reg<q>")]
 )
 
+(define_insn "neon_vld1_x2<mode>"
+  [(set (match_operand:OI 0 "s_register_operand" "=w")
+        (unspec:OI [(match_operand:OI 1 "neon_struct_operand" "Um")
+                    (unspec:VQXBF [(const_int 0)] UNSPEC_VSTRUCTDUMMY)]
+                   UNSPEC_VLD1))]
+  "TARGET_NEON"
+  "vld1.<V_sz_elem>\t%h0, %A1"
+  [(set_attr "type" "neon_load1_2reg<q>")]
+)
+
 ;; The lane numbers in the RTL are in GCC lane order, having been flipped
 ;; in arm_expand_neon_args. The lane numbers are restored to architectural
 ;; lane order here.
diff --git a/gcc/testsuite/gcc.target/arm/simd/vld1q_base_xN_1.c b/gcc/testsuite/gcc.target/arm/simd/vld1q_base_xN_1.c
new file mode 100644
index 0000000..1d31777
--- /dev/null
+++ b/gcc/testsuite/gcc.target/arm/simd/vld1q_base_xN_1.c
@@ -0,0 +1,67 @@
+/* { dg-do assemble } */
+/* { dg-require-effective-target arm_neon_ok } */
+/* { dg-options "-save-temps -O2" } */
+/* { dg-add-options arm_neon } */
+
+#include "arm_neon.h"
+
+uint8x16x2_t test_vld1q_u8_x2 (uint8_t * a)
+{
+    return vld1q_u8_x2 (a);
+}
+
+uint16x8x2_t test_vld1q_u16_x2 (uint16_t * a)
+{
+    return vld1q_u16_x2 (a);
+}
+
+uint32x4x2_t test_vld1q_u32_x2 (uint32_t * a)
+{
+    return vld1q_u32_x2 (a);
+}
+
+uint64x2x2_t test_vld1q_u64_x2 (uint64_t * a)
+{
+    return vld1q_u64_x2 (a);
+}
+
+int8x16x2_t test_vld1q_s8_x2 (int8_t * a)
+{
+    return vld1q_s8_x2 (a);
+}
+
+int16x8x2_t test_vld1q_s16_x2 (int16_t * a)
+{
+    return vld1q_s16_x2 (a);
+}
+
+int32x4x2_t test_vld1q_s32_x2 (int32_t * a)
+{
+    return vld1q_s32_x2 (a);
+}
+
+int64x2x2_t test_vld1q_s64_x2 (int64_t * a)
+{
+    return vld1q_s64_x2 (a);
+}
+
+float32x4x2_t test_vld1q_f32_x2 (float32_t * a)
+{
+    return vld1q_f32_x2 (a);
+}
+
+poly8x16x2_t test_vld1q_p8_x2 (poly8_t * a)
+{
+    return vld1q_p8_x2 (a);
+}
+
+poly16x8x2_t test_vld1q_p16_x2 (poly16_t * a)
+{
+    return vld1q_p16_x2 (a);
+}
+
+/* { dg-final { scan-assembler-times {vld1.8\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 3 } }  */
+/* { dg-final { scan-assembler-times {vld1.16\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 3 } }  */
+/* { dg-final { scan-assembler-times {vld1.32\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 3 } }  */
+/* { dg-final { scan-assembler-times {vld1.64\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+:64\]\n} 2 } }  */
+
diff --git a/gcc/testsuite/gcc.target/arm/simd/vld1q_bf16_xN_1.c b/gcc/testsuite/gcc.target/arm/simd/vld1q_bf16_xN_1.c
new file mode 100644
index 0000000..5f6fc98
--- /dev/null
+++ b/gcc/testsuite/gcc.target/arm/simd/vld1q_bf16_xN_1.c
@@ -0,0 +1,13 @@
+/* { dg-do assemble } */
+/* { dg-require-effective-target arm_v8_2a_bf16_neon_ok } */
+/* { dg-options "-save-temps -O2" } */
+/* { dg-add-options arm_v8_2a_bf16_neon } */
+
+#include "arm_neon.h"
+
+bfloat16x8x2_t test_vld1q_bf16_x2 (bfloat16_t * a)
+{
+    return vld1q_bf16_x2 (a);
+}
+
+/* { dg-final { scan-assembler-times {vld1.16\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 1 } }  */
diff --git a/gcc/testsuite/gcc.target/arm/simd/vld1q_fp16_xN_1.c b/gcc/testsuite/gcc.target/arm/simd/vld1q_fp16_xN_1.c
new file mode 100644
index 0000000..aecf491
--- /dev/null
+++ b/gcc/testsuite/gcc.target/arm/simd/vld1q_fp16_xN_1.c
@@ -0,0 +1,14 @@
+/* { dg-do assemble } */
+/* { dg-require-effective-target arm_neon_fp16_ok } */
+/* { dg-options "-save-temps -O2" } */
+/* { dg-add-options arm_neon_fp16 } */
+
+#include "arm_neon.h"
+
+float16x8x2_t test_vld1q_f16_x2 (float16_t * a)
+{
+    return vld1q_f16_x2 (a);
+}
+
+/* { dg-final { scan-assembler-times {vld1.16\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+\]\n} 1 } }  */
+
diff --git a/gcc/testsuite/gcc.target/arm/simd/vld1q_p64_xN_1.c b/gcc/testsuite/gcc.target/arm/simd/vld1q_p64_xN_1.c
new file mode 100644
index 0000000..04ceb5e
--- /dev/null
+++ b/gcc/testsuite/gcc.target/arm/simd/vld1q_p64_xN_1.c
@@ -0,0 +1,14 @@
+/* { dg-do assemble } */
+/* { dg-require-effective-target arm_crypto_ok } */
+/* { dg-options "-save-temps -O2" } */
+/* { dg-add-options arm_crypto } */
+
+#include "arm_neon.h"
+
+poly64x2x2_t test_vld1q_p64_x2 (poly64_t * a)
+{
+    return vld1q_p64_x2 (a);
+}
+
+/* { dg-final { scan-assembler-times {vld1.64\t\{d[0-9]+-d[0-9]+\}, \[r[0-9]+:64\]\n} 1 } }  */
+