aarch64: Use type-qualified builtins for [R]SUBHN[2] Neon intrinsics

Declare unsigned type-qualified builtins and use them to implement
(rounding) halving-narrowing-subtract Neon intrinsics. This removes
the need for many casts in arm_neon.h.

gcc/ChangeLog:

2021-11-09  Jonathan Wright  <jonathan.wright@arm.com>

	* config/aarch64/aarch64-simd-builtins.def: Declare unsigned
	builtins for [r]subhn[2].
	* config/aarch64/arm_neon.h (vsubhn_s16): Remove unnecessary
	cast.
	(vsubhn_s32): Likewise.
	(vsubhn_s64): Likewise.
	(vsubhn_u16): Use type-qualified builtin and remove casts.
	(vsubhn_u32): Likewise.
	(vsubhn_u64): Likewise.
	(vrsubhn_s16): Remove unnecessary cast.
	(vrsubhn_s32): Likewise.
	(vrsubhn_s64): Likewise.
	(vrsubhn_u16): Use type-qualified builtin and remove casts.
	(vrsubhn_u32): Likewise.
	(vrsubhn_u64): Likewise.
	(vrsubhn_high_s16): Remove unnecessary cast.
	(vrsubhn_high_s32): Likewise.
	(vrsubhn_high_s64): Likewise.
	(vrsubhn_high_u16): Use type-qualified builtin and remove
	casts.
	(vrsubhn_high_u32): Likewise.
	(vrsubhn_high_u64): Likewise.
	(vsubhn_high_s16): Remove unnecessary cast.
	(vsubhn_high_s32): Likewise.
	(vsubhn_high_s64): Likewise.
	(vsubhn_high_u16): Use type-qualified builtin and remove
	casts.
	(vsubhn_high_u32): Likewise.
	(vsubhn_high_u64): Likewise.

2 files changed, 28 insertions, 42 deletions

diff --git a/gcc/config/aarch64/aarch64-simd-builtins.def b/gcc/config/aarch64/aarch64-simd-builtins.def
index e57dee9..6f632f6 100644
--- a/gcc/config/aarch64/aarch64-simd-builtins.def
+++ b/gcc/config/aarch64/aarch64-simd-builtins.def
@@ -222,16 +222,20 @@
   BUILTIN_VQN (BINOP, addhn, 0, NONE)
   BUILTIN_VQN (BINOPU, addhn, 0, NONE)
   BUILTIN_VQN (BINOP, subhn, 0, NONE)
+  BUILTIN_VQN (BINOPU, subhn, 0, NONE)
   BUILTIN_VQN (BINOP, raddhn, 0, NONE)
   BUILTIN_VQN (BINOPU, raddhn, 0, NONE)
   BUILTIN_VQN (BINOP, rsubhn, 0, NONE)
+  BUILTIN_VQN (BINOPU, rsubhn, 0, NONE)
   /* Implemented by aarch64_<sur><addsub>hn2<mode>.  */
   BUILTIN_VQN (TERNOP, addhn2, 0, NONE)
   BUILTIN_VQN (TERNOPU, addhn2, 0, NONE)
   BUILTIN_VQN (TERNOP, subhn2, 0, NONE)
+  BUILTIN_VQN (TERNOPU, subhn2, 0, NONE)
   BUILTIN_VQN (TERNOP, raddhn2, 0, NONE)
   BUILTIN_VQN (TERNOPU, raddhn2, 0, NONE)
   BUILTIN_VQN (TERNOP, rsubhn2, 0, NONE)
+  BUILTIN_VQN (TERNOPU, rsubhn2, 0, NONE)
 
   /* Implemented by aarch64_<us>xtl<mode>.  */
   BUILTIN_VQN (UNOP, sxtl, 0, NONE)
diff --git a/gcc/config/aarch64/arm_neon.h b/gcc/config/aarch64/arm_neon.h
index fa404e0..288f017 100644
--- a/gcc/config/aarch64/arm_neon.h
+++ b/gcc/config/aarch64/arm_neon.h
@@ -2022,186 +2022,168 @@ __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_s16 (int16x8_t __a, int16x8_t __b)
 {
-  return (int8x8_t) __builtin_aarch64_subhnv8hi (__a, __b);
+  return __builtin_aarch64_subhnv8hi (__a, __b);
 }
 
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_s32 (int32x4_t __a, int32x4_t __b)
 {
-  return (int16x4_t) __builtin_aarch64_subhnv4si (__a, __b);
+  return __builtin_aarch64_subhnv4si (__a, __b);
 }
 
 __extension__ extern __inline int32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_s64 (int64x2_t __a, int64x2_t __b)
 {
-  return (int32x2_t) __builtin_aarch64_subhnv2di (__a, __b);
+  return __builtin_aarch64_subhnv2di (__a, __b);
 }
 
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_u16 (uint16x8_t __a, uint16x8_t __b)
 {
-  return (uint8x8_t) __builtin_aarch64_subhnv8hi ((int16x8_t) __a,
-						  (int16x8_t) __b);
+  return __builtin_aarch64_subhnv8hi_uuu (__a, __b);
 }
 
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_u32 (uint32x4_t __a, uint32x4_t __b)
 {
-  return (uint16x4_t) __builtin_aarch64_subhnv4si ((int32x4_t) __a,
-						   (int32x4_t) __b);
+  return __builtin_aarch64_subhnv4si_uuu (__a, __b);
 }
 
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_u64 (uint64x2_t __a, uint64x2_t __b)
 {
-  return (uint32x2_t) __builtin_aarch64_subhnv2di ((int64x2_t) __a,
-						   (int64x2_t) __b);
+  return __builtin_aarch64_subhnv2di_uuu (__a, __b);
 }
 
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_s16 (int16x8_t __a, int16x8_t __b)
 {
-  return (int8x8_t) __builtin_aarch64_rsubhnv8hi (__a, __b);
+  return __builtin_aarch64_rsubhnv8hi (__a, __b);
 }
 
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_s32 (int32x4_t __a, int32x4_t __b)
 {
-  return (int16x4_t) __builtin_aarch64_rsubhnv4si (__a, __b);
+  return __builtin_aarch64_rsubhnv4si (__a, __b);
 }
 
 __extension__ extern __inline int32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_s64 (int64x2_t __a, int64x2_t __b)
 {
-  return (int32x2_t) __builtin_aarch64_rsubhnv2di (__a, __b);
+  return __builtin_aarch64_rsubhnv2di (__a, __b);
 }
 
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_u16 (uint16x8_t __a, uint16x8_t __b)
 {
-  return (uint8x8_t) __builtin_aarch64_rsubhnv8hi ((int16x8_t) __a,
-						   (int16x8_t) __b);
+  return __builtin_aarch64_rsubhnv8hi_uuu (__a, __b);
 }
 
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_u32 (uint32x4_t __a, uint32x4_t __b)
 {
-  return (uint16x4_t) __builtin_aarch64_rsubhnv4si ((int32x4_t) __a,
-						    (int32x4_t) __b);
+  return __builtin_aarch64_rsubhnv4si_uuu (__a, __b);
 }
 
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_u64 (uint64x2_t __a, uint64x2_t __b)
 {
-  return (uint32x2_t) __builtin_aarch64_rsubhnv2di ((int64x2_t) __a,
-						    (int64x2_t) __b);
+  return __builtin_aarch64_rsubhnv2di_uuu (__a, __b);
 }
 
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_high_s16 (int8x8_t __a, int16x8_t __b, int16x8_t __c)
 {
-  return (int8x16_t) __builtin_aarch64_rsubhn2v8hi (__a, __b, __c);
+  return __builtin_aarch64_rsubhn2v8hi (__a, __b, __c);
 }
 
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_high_s32 (int16x4_t __a, int32x4_t __b, int32x4_t __c)
 {
-  return (int16x8_t) __builtin_aarch64_rsubhn2v4si (__a, __b, __c);
+  return __builtin_aarch64_rsubhn2v4si (__a, __b, __c);
 }
 
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_high_s64 (int32x2_t __a, int64x2_t __b, int64x2_t __c)
 {
-  return (int32x4_t) __builtin_aarch64_rsubhn2v2di (__a, __b, __c);
+  return __builtin_aarch64_rsubhn2v2di (__a, __b, __c);
 }
 
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_high_u16 (uint8x8_t __a, uint16x8_t __b, uint16x8_t __c)
 {
-  return (uint8x16_t) __builtin_aarch64_rsubhn2v8hi ((int8x8_t) __a,
-						     (int16x8_t) __b,
-						     (int16x8_t) __c);
+  return __builtin_aarch64_rsubhn2v8hi_uuuu (__a, __b, __c);
 }
 
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_high_u32 (uint16x4_t __a, uint32x4_t __b, uint32x4_t __c)
 {
-  return (uint16x8_t) __builtin_aarch64_rsubhn2v4si ((int16x4_t) __a,
-						     (int32x4_t) __b,
-						     (int32x4_t) __c);
+  return __builtin_aarch64_rsubhn2v4si_uuuu (__a, __b, __c);
 }
 
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_high_u64 (uint32x2_t __a, uint64x2_t __b, uint64x2_t __c)
 {
-  return (uint32x4_t) __builtin_aarch64_rsubhn2v2di ((int32x2_t) __a,
-						     (int64x2_t) __b,
-						     (int64x2_t) __c);
+  return __builtin_aarch64_rsubhn2v2di_uuuu (__a, __b, __c);
 }
 
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_high_s16 (int8x8_t __a, int16x8_t __b, int16x8_t __c)
 {
-  return (int8x16_t) __builtin_aarch64_subhn2v8hi (__a, __b, __c);
+  return __builtin_aarch64_subhn2v8hi (__a, __b, __c);
 }
 
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_high_s32 (int16x4_t __a, int32x4_t __b, int32x4_t __c)
 {
-  return (int16x8_t) __builtin_aarch64_subhn2v4si (__a, __b, __c);;
+  return __builtin_aarch64_subhn2v4si (__a, __b, __c);;
 }
 
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_high_s64 (int32x2_t __a, int64x2_t __b, int64x2_t __c)
 {
-  return (int32x4_t) __builtin_aarch64_subhn2v2di (__a, __b, __c);
+  return __builtin_aarch64_subhn2v2di (__a, __b, __c);
 }
 
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_high_u16 (uint8x8_t __a, uint16x8_t __b, uint16x8_t __c)
 {
-  return (uint8x16_t) __builtin_aarch64_subhn2v8hi ((int8x8_t) __a,
-						    (int16x8_t) __b,
-						    (int16x8_t) __c);
+  return __builtin_aarch64_subhn2v8hi_uuuu (__a, __b, __c);
 }
 
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_high_u32 (uint16x4_t __a, uint32x4_t __b, uint32x4_t __c)
 {
-  return (uint16x8_t) __builtin_aarch64_subhn2v4si ((int16x4_t) __a,
-						    (int32x4_t) __b,
-						    (int32x4_t) __c);
+  return __builtin_aarch64_subhn2v4si_uuuu (__a, __b, __c);
 }
 
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_high_u64 (uint32x2_t __a, uint64x2_t __b, uint64x2_t __c)
 {
-  return (uint32x4_t) __builtin_aarch64_subhn2v2di ((int32x2_t) __a,
-						    (int64x2_t) __b,
-						    (int64x2_t) __c);
+  return __builtin_aarch64_subhn2v2di_uuuu (__a, __b, __c);
 }
 
 __extension__ extern __inline uint16x4_t