#define vstrwq_p_f32(__addr, __value, __p) __arm_vstrwq_p_f32(__addr, __value, __p)
#define vstrwq_p_s32(__addr, __value, __p) __arm_vstrwq_p_s32(__addr, __value, __p)
#define vstrwq_p_u32(__addr, __value, __p) __arm_vstrwq_p_u32(__addr, __value, __p)
+#define vstrdq_scatter_base_p_s64(__addr, __offset, __value, __p) __arm_vstrdq_scatter_base_p_s64(__addr, __offset, __value, __p)
+#define vstrdq_scatter_base_p_u64(__addr, __offset, __value, __p) __arm_vstrdq_scatter_base_p_u64(__addr, __offset, __value, __p)
+#define vstrdq_scatter_base_s64(__addr, __offset, __value) __arm_vstrdq_scatter_base_s64(__addr, __offset, __value)
+#define vstrdq_scatter_base_u64(__addr, __offset, __value) __arm_vstrdq_scatter_base_u64(__addr, __offset, __value)
+#define vstrdq_scatter_offset_p_s64(__base, __offset, __value, __p) __arm_vstrdq_scatter_offset_p_s64(__base, __offset, __value, __p)
+#define vstrdq_scatter_offset_p_u64(__base, __offset, __value, __p) __arm_vstrdq_scatter_offset_p_u64(__base, __offset, __value, __p)
+#define vstrdq_scatter_offset_s64(__base, __offset, __value) __arm_vstrdq_scatter_offset_s64(__base, __offset, __value)
+#define vstrdq_scatter_offset_u64(__base, __offset, __value) __arm_vstrdq_scatter_offset_u64(__base, __offset, __value)
+#define vstrdq_scatter_shifted_offset_p_s64(__base, __offset, __value, __p) __arm_vstrdq_scatter_shifted_offset_p_s64(__base, __offset, __value, __p)
+#define vstrdq_scatter_shifted_offset_p_u64(__base, __offset, __value, __p) __arm_vstrdq_scatter_shifted_offset_p_u64(__base, __offset, __value, __p)
+#define vstrdq_scatter_shifted_offset_s64(__base, __offset, __value) __arm_vstrdq_scatter_shifted_offset_s64(__base, __offset, __value)
+#define vstrdq_scatter_shifted_offset_u64(__base, __offset, __value) __arm_vstrdq_scatter_shifted_offset_u64(__base, __offset, __value)
+#define vstrhq_scatter_offset_f16(__base, __offset, __value) __arm_vstrhq_scatter_offset_f16(__base, __offset, __value)
+#define vstrhq_scatter_offset_p_f16(__base, __offset, __value, __p) __arm_vstrhq_scatter_offset_p_f16(__base, __offset, __value, __p)
+#define vstrhq_scatter_shifted_offset_f16(__base, __offset, __value) __arm_vstrhq_scatter_shifted_offset_f16(__base, __offset, __value)
+#define vstrhq_scatter_shifted_offset_p_f16(__base, __offset, __value, __p) __arm_vstrhq_scatter_shifted_offset_p_f16(__base, __offset, __value, __p)
+#define vstrwq_scatter_base_f32(__addr, __offset, __value) __arm_vstrwq_scatter_base_f32(__addr, __offset, __value)
+#define vstrwq_scatter_base_p_f32(__addr, __offset, __value, __p) __arm_vstrwq_scatter_base_p_f32(__addr, __offset, __value, __p)
+#define vstrwq_scatter_offset_f32(__base, __offset, __value) __arm_vstrwq_scatter_offset_f32(__base, __offset, __value)
+#define vstrwq_scatter_offset_p_f32(__base, __offset, __value, __p) __arm_vstrwq_scatter_offset_p_f32(__base, __offset, __value, __p)
+#define vstrwq_scatter_offset_p_s32(__base, __offset, __value, __p) __arm_vstrwq_scatter_offset_p_s32(__base, __offset, __value, __p)
+#define vstrwq_scatter_offset_p_u32(__base, __offset, __value, __p) __arm_vstrwq_scatter_offset_p_u32(__base, __offset, __value, __p)
+#define vstrwq_scatter_offset_s32(__base, __offset, __value) __arm_vstrwq_scatter_offset_s32(__base, __offset, __value)
+#define vstrwq_scatter_offset_u32(__base, __offset, __value) __arm_vstrwq_scatter_offset_u32(__base, __offset, __value)
+#define vstrwq_scatter_shifted_offset_f32(__base, __offset, __value) __arm_vstrwq_scatter_shifted_offset_f32(__base, __offset, __value)
+#define vstrwq_scatter_shifted_offset_p_f32(__base, __offset, __value, __p) __arm_vstrwq_scatter_shifted_offset_p_f32(__base, __offset, __value, __p)
+#define vstrwq_scatter_shifted_offset_p_s32(__base, __offset, __value, __p) __arm_vstrwq_scatter_shifted_offset_p_s32(__base, __offset, __value, __p)
+#define vstrwq_scatter_shifted_offset_p_u32(__base, __offset, __value, __p) __arm_vstrwq_scatter_shifted_offset_p_u32(__base, __offset, __value, __p)
+#define vstrwq_scatter_shifted_offset_s32(__base, __offset, __value) __arm_vstrwq_scatter_shifted_offset_s32(__base, __offset, __value)
+#define vstrwq_scatter_shifted_offset_u32(__base, __offset, __value) __arm_vstrwq_scatter_shifted_offset_u32(__base, __offset, __value)
#endif
__extension__ extern __inline void
__builtin_mve_vstrwq_p_uv4si ((__builtin_neon_si *) __addr, __value, __p);
}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrdq_scatter_base_p_s64 (uint64x2_t __addr, const int __offset, int64x2_t __value, mve_pred16_t __p)
+{
+ __builtin_mve_vstrdq_scatter_base_p_sv2di (__addr, __offset, __value, __p);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrdq_scatter_base_p_u64 (uint64x2_t __addr, const int __offset, uint64x2_t __value, mve_pred16_t __p)
+{
+ __builtin_mve_vstrdq_scatter_base_p_uv2di (__addr, __offset, __value, __p);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrdq_scatter_base_s64 (uint64x2_t __addr, const int __offset, int64x2_t __value)
+{
+ __builtin_mve_vstrdq_scatter_base_sv2di (__addr, __offset, __value);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrdq_scatter_base_u64 (uint64x2_t __addr, const int __offset, uint64x2_t __value)
+{
+ __builtin_mve_vstrdq_scatter_base_uv2di (__addr, __offset, __value);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrdq_scatter_offset_p_s64 (int64_t * __base, uint64x2_t __offset, int64x2_t __value, mve_pred16_t __p)
+{
+ __builtin_mve_vstrdq_scatter_offset_p_sv2di (__base, __offset, __value, __p);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrdq_scatter_offset_p_u64 (uint64_t * __base, uint64x2_t __offset, uint64x2_t __value, mve_pred16_t __p)
+{
+ __builtin_mve_vstrdq_scatter_offset_p_uv2di (__base, __offset, __value, __p);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrdq_scatter_offset_s64 (int64_t * __base, uint64x2_t __offset, int64x2_t __value)
+{
+ __builtin_mve_vstrdq_scatter_offset_sv2di (__base, __offset, __value);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrdq_scatter_offset_u64 (uint64_t * __base, uint64x2_t __offset, uint64x2_t __value)
+{
+ __builtin_mve_vstrdq_scatter_offset_uv2di (__base, __offset, __value);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrdq_scatter_shifted_offset_p_s64 (int64_t * __base, uint64x2_t __offset, int64x2_t __value, mve_pred16_t __p)
+{
+ __builtin_mve_vstrdq_scatter_shifted_offset_p_sv2di (__base, __offset, __value, __p);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrdq_scatter_shifted_offset_p_u64 (uint64_t * __base, uint64x2_t __offset, uint64x2_t __value, mve_pred16_t __p)
+{
+ __builtin_mve_vstrdq_scatter_shifted_offset_p_uv2di (__base, __offset, __value, __p);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrdq_scatter_shifted_offset_s64 (int64_t * __base, uint64x2_t __offset, int64x2_t __value)
+{
+ __builtin_mve_vstrdq_scatter_shifted_offset_sv2di (__base, __offset, __value);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrdq_scatter_shifted_offset_u64 (uint64_t * __base, uint64x2_t __offset, uint64x2_t __value)
+{
+ __builtin_mve_vstrdq_scatter_shifted_offset_uv2di (__base, __offset, __value);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrwq_scatter_offset_p_s32 (int32_t * __base, uint32x4_t __offset, int32x4_t __value, mve_pred16_t __p)
+{
+ __builtin_mve_vstrwq_scatter_offset_p_sv4si ((__builtin_neon_si *) __base, __offset, __value, __p);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrwq_scatter_offset_p_u32 (uint32_t * __base, uint32x4_t __offset, uint32x4_t __value, mve_pred16_t __p)
+{
+ __builtin_mve_vstrwq_scatter_offset_p_uv4si ((__builtin_neon_si *) __base, __offset, __value, __p);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrwq_scatter_offset_s32 (int32_t * __base, uint32x4_t __offset, int32x4_t __value)
+{
+ __builtin_mve_vstrwq_scatter_offset_sv4si ((__builtin_neon_si *) __base, __offset, __value);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrwq_scatter_offset_u32 (uint32_t * __base, uint32x4_t __offset, uint32x4_t __value)
+{
+ __builtin_mve_vstrwq_scatter_offset_uv4si ((__builtin_neon_si *) __base, __offset, __value);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrwq_scatter_shifted_offset_p_s32 (int32_t * __base, uint32x4_t __offset, int32x4_t __value, mve_pred16_t __p)
+{
+ __builtin_mve_vstrwq_scatter_shifted_offset_p_sv4si ((__builtin_neon_si *) __base, __offset, __value, __p);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrwq_scatter_shifted_offset_p_u32 (uint32_t * __base, uint32x4_t __offset, uint32x4_t __value, mve_pred16_t __p)
+{
+ __builtin_mve_vstrwq_scatter_shifted_offset_p_uv4si ((__builtin_neon_si *) __base, __offset, __value, __p);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrwq_scatter_shifted_offset_s32 (int32_t * __base, uint32x4_t __offset, int32x4_t __value)
+{
+ __builtin_mve_vstrwq_scatter_shifted_offset_sv4si ((__builtin_neon_si *) __base, __offset, __value);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrwq_scatter_shifted_offset_u32 (uint32_t * __base, uint32x4_t __offset, uint32x4_t __value)
+{
+ __builtin_mve_vstrwq_scatter_shifted_offset_uv4si ((__builtin_neon_si *) __base, __offset, __value);
+}
+
#if (__ARM_FEATURE_MVE & 2) /* MVE Floating point. */
__extension__ extern __inline void
__builtin_mve_vstrhq_p_fv8hf (__addr, __value, __p);
}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrhq_scatter_offset_f16 (float16_t * __base, uint16x8_t __offset, float16x8_t __value)
+{
+ __builtin_mve_vstrhq_scatter_offset_fv8hf (__base, __offset, __value);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrhq_scatter_offset_p_f16 (float16_t * __base, uint16x8_t __offset, float16x8_t __value, mve_pred16_t __p)
+{
+ __builtin_mve_vstrhq_scatter_offset_p_fv8hf (__base, __offset, __value, __p);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrhq_scatter_shifted_offset_f16 (float16_t * __base, uint16x8_t __offset, float16x8_t __value)
+{
+ __builtin_mve_vstrhq_scatter_shifted_offset_fv8hf (__base, __offset, __value);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrhq_scatter_shifted_offset_p_f16 (float16_t * __base, uint16x8_t __offset, float16x8_t __value, mve_pred16_t __p)
+{
+ __builtin_mve_vstrhq_scatter_shifted_offset_p_fv8hf (__base, __offset, __value, __p);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrwq_scatter_base_f32 (uint32x4_t __addr, const int __offset, float32x4_t __value)
+{
+ __builtin_mve_vstrwq_scatter_base_fv4sf (__addr, __offset, __value);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrwq_scatter_base_p_f32 (uint32x4_t __addr, const int __offset, float32x4_t __value, mve_pred16_t __p)
+{
+ __builtin_mve_vstrwq_scatter_base_p_fv4sf (__addr, __offset, __value, __p);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrwq_scatter_offset_f32 (float32_t * __base, uint32x4_t __offset, float32x4_t __value)
+{
+ __builtin_mve_vstrwq_scatter_offset_fv4sf (__base, __offset, __value);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrwq_scatter_offset_p_f32 (float32_t * __base, uint32x4_t __offset, float32x4_t __value, mve_pred16_t __p)
+{
+ __builtin_mve_vstrwq_scatter_offset_p_fv4sf (__base, __offset, __value, __p);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrwq_scatter_shifted_offset_f32 (float32_t * __base, uint32x4_t __offset, float32x4_t __value)
+{
+ __builtin_mve_vstrwq_scatter_shifted_offset_fv4sf (__base, __offset, __value);
+}
+
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_vstrwq_scatter_shifted_offset_p_f32 (float32_t * __base, uint32x4_t __offset, float32x4_t __value, mve_pred16_t __p)
+{
+ __builtin_mve_vstrwq_scatter_shifted_offset_p_fv4sf (__base, __offset, __value, __p);
+}
+
#endif
enum {
int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_u32 (__ARM_mve_coerce(__p0, uint32_t *), __ARM_mve_coerce(__p1, uint32x4_t)), \
int (*)[__ARM_mve_type_float32_t_ptr][__ARM_mve_type_float32x4_t]: __arm_vstrwq_f32 (__ARM_mve_coerce(__p0, float32_t *), __ARM_mve_coerce(__p1, float32x4_t)));})
+#define vstrhq_scatter_offset(p0,p1,p2) __arm_vstrhq_scatter_offset(p0,p1,p2)
+#define __arm_vstrhq_scatter_offset(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_int16x8_t]: __arm_vstrhq_scatter_offset_s16 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, int16x8_t)), \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_int32x4_t]: __arm_vstrhq_scatter_offset_s32 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, int32x4_t)), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_uint16x8_t]: __arm_vstrhq_scatter_offset_u16 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, uint16x8_t)), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_uint32x4_t]: __arm_vstrhq_scatter_offset_u32 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, uint32x4_t)), \
+ int (*)[__ARM_mve_type_float16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_float16x8_t]: __arm_vstrhq_scatter_offset_f16 (__ARM_mve_coerce(__p0, float16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, float16x8_t)));})
+
+#define vstrhq_scatter_offset_p(p0,p1,p2,p3) __arm_vstrhq_scatter_offset_p(p0,p1,p2,p3)
+#define __arm_vstrhq_scatter_offset_p(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_int16x8_t]: __arm_vstrhq_scatter_offset_p_s16 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, int16x8_t), p3), \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_int32x4_t]: __arm_vstrhq_scatter_offset_p_s32 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, int32x4_t), p3), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_uint16x8_t]: __arm_vstrhq_scatter_offset_p_u16 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, uint16x8_t), p3), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_uint32x4_t]: __arm_vstrhq_scatter_offset_p_u32 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, uint32x4_t), p3), \
+ int (*)[__ARM_mve_type_float16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_float16x8_t]: __arm_vstrhq_scatter_offset_p_f16 (__ARM_mve_coerce(__p0, float16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, float16x8_t), p3));})
+
+#define vstrhq_scatter_shifted_offset(p0,p1,p2) __arm_vstrhq_scatter_shifted_offset(p0,p1,p2)
+#define __arm_vstrhq_scatter_shifted_offset(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_int16x8_t]: __arm_vstrhq_scatter_shifted_offset_s16 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, int16x8_t)), \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_int32x4_t]: __arm_vstrhq_scatter_shifted_offset_s32 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, int32x4_t)), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_uint16x8_t]: __arm_vstrhq_scatter_shifted_offset_u16 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, uint16x8_t)), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_uint32x4_t]: __arm_vstrhq_scatter_shifted_offset_u32 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, uint32x4_t)), \
+ int (*)[__ARM_mve_type_float16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_float16x8_t]: __arm_vstrhq_scatter_shifted_offset_f16 (__ARM_mve_coerce(__p0, float16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, float16x8_t)));})
+
+#define vstrhq_scatter_shifted_offset_p(p0,p1,p2,p3) __arm_vstrhq_scatter_shifted_offset_p(p0,p1,p2,p3)
+#define __arm_vstrhq_scatter_shifted_offset_p(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_int16x8_t]: __arm_vstrhq_scatter_shifted_offset_p_s16 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, int16x8_t), p3), \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_int32x4_t]: __arm_vstrhq_scatter_shifted_offset_p_s32 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, int32x4_t), p3), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_uint16x8_t]: __arm_vstrhq_scatter_shifted_offset_p_u16 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, uint16x8_t), p3), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_uint32x4_t]: __arm_vstrhq_scatter_shifted_offset_p_u32 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, uint32x4_t), p3), \
+ int (*)[__ARM_mve_type_float16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_float16x8_t]: __arm_vstrhq_scatter_shifted_offset_p_f16 (__ARM_mve_coerce(__p0, float16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, float16x8_t), p3));})
+
+#define vstrwq_scatter_base(p0,p1,p2) __arm_vstrwq_scatter_base(p0,p1,p2)
+#define __arm_vstrwq_scatter_base(p0,p1,p2) ({ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_base_s32 (p0, p1, __ARM_mve_coerce(__p2, int32x4_t)), \
+ int (*)[__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_base_u32 (p0, p1, __ARM_mve_coerce(__p2, uint32x4_t)), \
+ int (*)[__ARM_mve_type_float32x4_t]: __arm_vstrwq_scatter_base_f32 (p0, p1, __ARM_mve_coerce(__p2, float32x4_t)));})
+
+#define vstrwq_scatter_base_p(p0,p1,p2,p3) __arm_vstrwq_scatter_base_p(p0,p1,p2,p3)
+#define __arm_vstrwq_scatter_base_p(p0,p1,p2,p3) ({ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_base_p_s32(p0, p1, __ARM_mve_coerce(__p2, int32x4_t), p3), \
+ int (*)[__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_base_p_u32(p0, p1, __ARM_mve_coerce(__p2, uint32x4_t), p3), \
+ int (*)[__ARM_mve_type_float32x4_t]: __arm_vstrwq_scatter_base_p_f32(p0, p1, __ARM_mve_coerce(__p2, float32x4_t), p3));})
+
+#define vstrwq_scatter_offset(p0,p1,p2) __arm_vstrwq_scatter_offset(p0,p1,p2)
+#define __arm_vstrwq_scatter_offset(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32_t_ptr][__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_offset_s32 (__ARM_mve_coerce(__p0, int32_t *), __p1, __ARM_mve_coerce(__p2, int32x4_t)), \
+ int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_offset_u32 (__ARM_mve_coerce(__p0, uint32_t *), __p1, __ARM_mve_coerce(__p2, uint32x4_t)), \
+ int (*)[__ARM_mve_type_float32_t_ptr][__ARM_mve_type_float32x4_t]: __arm_vstrwq_scatter_offset_f32 (__ARM_mve_coerce(__p0, float32_t *), __p1, __ARM_mve_coerce(__p2, float32x4_t)));})
+
+#define vstrwq_scatter_offset_p(p0,p1,p2,p3) __arm_vstrwq_scatter_offset_p(p0,p1,p2,p3)
+#define __arm_vstrwq_scatter_offset_p(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32_t_ptr][__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_offset_p_s32 (__ARM_mve_coerce(__p0, int32_t *), __p1, __ARM_mve_coerce(__p2, int32x4_t), p3), \
+ int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_offset_p_u32 (__ARM_mve_coerce(__p0, uint32_t *), __p1, __ARM_mve_coerce(__p2, uint32x4_t), p3), \
+ int (*)[__ARM_mve_type_float32_t_ptr][__ARM_mve_type_float32x4_t]: __arm_vstrwq_scatter_offset_p_f32 (__ARM_mve_coerce(__p0, float32_t *), __p1, __ARM_mve_coerce(__p2, float32x4_t), p3));})
+
+#define vstrwq_scatter_offset_p(p0,p1,p2,p3) __arm_vstrwq_scatter_offset_p(p0,p1,p2,p3)
+#define __arm_vstrwq_scatter_offset_p(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32_t_ptr][__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_offset_p_s32 (__ARM_mve_coerce(__p0, int32_t *), __p1, __ARM_mve_coerce(__p2, int32x4_t), p3), \
+ int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_offset_p_u32 (__ARM_mve_coerce(__p0, uint32_t *), __p1, __ARM_mve_coerce(__p2, uint32x4_t), p3), \
+ int (*)[__ARM_mve_type_float32_t_ptr][__ARM_mve_type_float32x4_t]: __arm_vstrwq_scatter_offset_p_f32 (__ARM_mve_coerce(__p0, float32_t *), __p1, __ARM_mve_coerce(__p2, float32x4_t), p3));})
+
+#define vstrwq_scatter_offset(p0,p1,p2) __arm_vstrwq_scatter_offset(p0,p1,p2)
+#define __arm_vstrwq_scatter_offset(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32_t_ptr][__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_offset_s32 (__ARM_mve_coerce(__p0, int32_t *), __p1, __ARM_mve_coerce(__p2, int32x4_t)), \
+ int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_offset_u32 (__ARM_mve_coerce(__p0, uint32_t *), __p1, __ARM_mve_coerce(__p2, uint32x4_t)), \
+ int (*)[__ARM_mve_type_float32_t_ptr][__ARM_mve_type_float32x4_t]: __arm_vstrwq_scatter_offset_f32 (__ARM_mve_coerce(__p0, float32_t *), __p1, __ARM_mve_coerce(__p2, float32x4_t)));})
+
+#define vstrwq_scatter_shifted_offset(p0,p1,p2) __arm_vstrwq_scatter_shifted_offset(p0,p1,p2)
+#define __arm_vstrwq_scatter_shifted_offset(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32_t_ptr][__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_shifted_offset_s32 (__ARM_mve_coerce(__p0, int32_t *), __p1, __ARM_mve_coerce(__p2, int32x4_t)), \
+ int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_shifted_offset_u32 (__ARM_mve_coerce(__p0, uint32_t *), __p1, __ARM_mve_coerce(__p2, uint32x4_t)), \
+ int (*)[__ARM_mve_type_float32_t_ptr][__ARM_mve_type_float32x4_t]: __arm_vstrwq_scatter_shifted_offset_f32 (__ARM_mve_coerce(__p0, float32_t *), __p1, __ARM_mve_coerce(__p2, float32x4_t)));})
+
+#define vstrwq_scatter_shifted_offset_p(p0,p1,p2,p3) __arm_vstrwq_scatter_shifted_offset_p(p0,p1,p2,p3)
+#define __arm_vstrwq_scatter_shifted_offset_p(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32_t_ptr][__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_shifted_offset_p_s32 (__ARM_mve_coerce(__p0, int32_t *), __p1, __ARM_mve_coerce(__p2, int32x4_t), p3), \
+ int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_shifted_offset_p_u32 (__ARM_mve_coerce(__p0, uint32_t *), __p1, __ARM_mve_coerce(__p2, uint32x4_t), p3), \
+ int (*)[__ARM_mve_type_float32_t_ptr][__ARM_mve_type_float32x4_t]: __arm_vstrwq_scatter_shifted_offset_p_f32 (__ARM_mve_coerce(__p0, float32_t *), __p1, __ARM_mve_coerce(__p2, float32x4_t), p3));})
+
+#define vstrwq_scatter_shifted_offset_p(p0,p1,p2,p3) __arm_vstrwq_scatter_shifted_offset_p(p0,p1,p2,p3)
+#define __arm_vstrwq_scatter_shifted_offset_p(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32_t_ptr][__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_shifted_offset_p_s32 (__ARM_mve_coerce(__p0, int32_t *), __p1, __ARM_mve_coerce(__p2, int32x4_t), p3), \
+ int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_shifted_offset_p_u32 (__ARM_mve_coerce(__p0, uint32_t *), __p1, __ARM_mve_coerce(__p2, uint32x4_t), p3), \
+ int (*)[__ARM_mve_type_float32_t_ptr][__ARM_mve_type_float32x4_t]: __arm_vstrwq_scatter_shifted_offset_p_f32 (__ARM_mve_coerce(__p0, float32_t *), __p1, __ARM_mve_coerce(__p2, float32x4_t), p3));})
+
+#define vstrwq_scatter_shifted_offset(p0,p1,p2) __arm_vstrwq_scatter_shifted_offset(p0,p1,p2)
+#define __arm_vstrwq_scatter_shifted_offset(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32_t_ptr][__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_shifted_offset_s32 (__ARM_mve_coerce(__p0, int32_t *), __p1, __ARM_mve_coerce(__p2, int32x4_t)), \
+ int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_shifted_offset_u32 (__ARM_mve_coerce(__p0, uint32_t *), __p1, __ARM_mve_coerce(__p2, uint32x4_t)), \
+ int (*)[__ARM_mve_type_float32_t_ptr][__ARM_mve_type_float32x4_t]: __arm_vstrwq_scatter_shifted_offset_f32 (__ARM_mve_coerce(__p0, float32_t *), __p1, __ARM_mve_coerce(__p2, float32x4_t)));})
+
#else /* MVE Integer. */
#define vst4q(p0,p1) __arm_vst4q(p0,p1)
int (*)[__ARM_mve_type_int32_t_ptr][__ARM_mve_type_int32x4_t]: __arm_vstrwq_p_s32 (__ARM_mve_coerce(__p0, int32_t *), __ARM_mve_coerce(__p1, int32x4_t), p2), \
int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_p_u32 (__ARM_mve_coerce(__p0, uint32_t *), __ARM_mve_coerce(__p1, uint32x4_t), p2));})
+#define vstrdq_scatter_base_p(p0,p1,p2,p3) __arm_vstrdq_scatter_base_p(p0,p1,p2,p3)
+#define __arm_vstrdq_scatter_base_p(p0,p1,p2,p3) ({ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int64x2_t]: __arm_vstrdq_scatter_base_p_s64 (p0, p1, __ARM_mve_coerce(__p2, int64x2_t), p3), \
+ int (*)[__ARM_mve_type_uint64x2_t]: __arm_vstrdq_scatter_base_p_u64 (p0, p1, __ARM_mve_coerce(__p2, uint64x2_t), p3));})
+
+#define vstrdq_scatter_base(p0,p1,p2) __arm_vstrdq_scatter_base(p0,p1,p2)
+#define __arm_vstrdq_scatter_base(p0,p1,p2) ({ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int64x2_t]: __arm_vstrdq_scatter_base_s64 (p0, p1, __ARM_mve_coerce(__p2, int64x2_t)), \
+ int (*)[__ARM_mve_type_uint64x2_t]: __arm_vstrdq_scatter_base_u64 (p0, p1, __ARM_mve_coerce(__p2, uint64x2_t)));})
+
+#define vstrdq_scatter_offset_p(p0,p1,p2,p3) __arm_vstrdq_scatter_offset_p(p0,p1,p2,p3)
+#define __arm_vstrdq_scatter_offset_p(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int64_t_ptr][__ARM_mve_type_int64x2_t]: __arm_vstrdq_scatter_offset_p_s64 (__ARM_mve_coerce(__p0, int64_t *), __p1, __ARM_mve_coerce(__p2, int64x2_t), p3), \
+ int (*)[__ARM_mve_type_uint64_t_ptr][__ARM_mve_type_uint64x2_t]: __arm_vstrdq_scatter_offset_p_u64 (__ARM_mve_coerce(__p0, uint64_t *), __p1, __ARM_mve_coerce(__p2, uint64x2_t), p3));})
+
+#define vstrdq_scatter_offset(p0,p1,p2) __arm_vstrdq_scatter_offset(p0,p1,p2)
+#define __arm_vstrdq_scatter_offset(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int64_t_ptr][__ARM_mve_type_int64x2_t]: __arm_vstrdq_scatter_offset_s64 (__ARM_mve_coerce(__p0, int64_t *), __p1, __ARM_mve_coerce(__p2, int64x2_t)), \
+ int (*)[__ARM_mve_type_uint64_t_ptr][__ARM_mve_type_uint64x2_t]: __arm_vstrdq_scatter_offset_u64 (__ARM_mve_coerce(__p0, uint64_t *), __p1, __ARM_mve_coerce(__p2, uint64x2_t)));})
+
+#define vstrdq_scatter_shifted_offset_p(p0,p1,p2,p3) __arm_vstrdq_scatter_shifted_offset_p(p0,p1,p2,p3)
+#define __arm_vstrdq_scatter_shifted_offset_p(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int64_t_ptr][__ARM_mve_type_int64x2_t]: __arm_vstrdq_scatter_shifted_offset_p_s64 (__ARM_mve_coerce(__p0, int64_t *), __p1, __ARM_mve_coerce(__p2, int64x2_t), p3), \
+ int (*)[__ARM_mve_type_uint64_t_ptr][__ARM_mve_type_uint64x2_t]: __arm_vstrdq_scatter_shifted_offset_p_u64 (__ARM_mve_coerce(__p0, uint64_t *), __p1, __ARM_mve_coerce(__p2, uint64x2_t), p3));})
+
+#define vstrdq_scatter_shifted_offset(p0,p1,p2) __arm_vstrdq_scatter_shifted_offset(p0,p1,p2)
+#define __arm_vstrdq_scatter_shifted_offset(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int64_t_ptr][__ARM_mve_type_int64x2_t]: __arm_vstrdq_scatter_shifted_offset_s64 (__ARM_mve_coerce(__p0, int64_t *), __p1, __ARM_mve_coerce(__p2, int64x2_t)), \
+ int (*)[__ARM_mve_type_uint64_t_ptr][__ARM_mve_type_uint64x2_t]: __arm_vstrdq_scatter_shifted_offset_u64 (__ARM_mve_coerce(__p0, uint64_t *), __p1, __ARM_mve_coerce(__p2, uint64x2_t)));})
+
+#define vstrhq_scatter_offset(p0,p1,p2) __arm_vstrhq_scatter_offset(p0,p1,p2)
+#define __arm_vstrhq_scatter_offset(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_int16x8_t]: __arm_vstrhq_scatter_offset_s16 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, int16x8_t)), \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_int32x4_t]: __arm_vstrhq_scatter_offset_s32 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, int32x4_t)), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_uint16x8_t]: __arm_vstrhq_scatter_offset_u16 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, uint16x8_t)), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_uint32x4_t]: __arm_vstrhq_scatter_offset_u32 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, uint32x4_t)));})
+
+#define vstrhq_scatter_offset_p(p0,p1,p2,p3) __arm_vstrhq_scatter_offset_p(p0,p1,p2,p3)
+#define __arm_vstrhq_scatter_offset_p(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_int16x8_t]: __arm_vstrhq_scatter_offset_p_s16 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, int16x8_t), p3), \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_int32x4_t]: __arm_vstrhq_scatter_offset_p_s32 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, int32x4_t), p3), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_uint16x8_t]: __arm_vstrhq_scatter_offset_p_u16 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, uint16x8_t), p3), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_uint32x4_t]: __arm_vstrhq_scatter_offset_p_u32 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, uint32x4_t), p3));})
+
+#define vstrhq_scatter_shifted_offset(p0,p1,p2) __arm_vstrhq_scatter_shifted_offset(p0,p1,p2)
+#define __arm_vstrhq_scatter_shifted_offset(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_int16x8_t]: __arm_vstrhq_scatter_shifted_offset_s16 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, int16x8_t)), \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_int32x4_t]: __arm_vstrhq_scatter_shifted_offset_s32 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, int32x4_t)), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_uint16x8_t]: __arm_vstrhq_scatter_shifted_offset_u16 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, uint16x8_t)), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_uint32x4_t]: __arm_vstrhq_scatter_shifted_offset_u32 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, uint32x4_t)));})
+
+#define vstrhq_scatter_shifted_offset_p(p0,p1,p2,p3) __arm_vstrhq_scatter_shifted_offset_p(p0,p1,p2,p3)
+#define __arm_vstrhq_scatter_shifted_offset_p(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p1)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_int16x8_t]: __arm_vstrhq_scatter_shifted_offset_p_s16 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, int16x8_t), p3), \
+ int (*)[__ARM_mve_type_int16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_int32x4_t]: __arm_vstrhq_scatter_shifted_offset_p_s32 (__ARM_mve_coerce(__p0, int16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, int32x4_t), p3), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint16x8_t][__ARM_mve_type_uint16x8_t]: __arm_vstrhq_scatter_shifted_offset_p_u16 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint16x8_t), __ARM_mve_coerce(__p2, uint16x8_t), p3), \
+ int (*)[__ARM_mve_type_uint16_t_ptr][__ARM_mve_type_uint32x4_t][__ARM_mve_type_uint32x4_t]: __arm_vstrhq_scatter_shifted_offset_p_u32 (__ARM_mve_coerce(__p0, uint16_t *), __ARM_mve_coerce(__p1, uint32x4_t), __ARM_mve_coerce(__p2, uint32x4_t), p3));})
+
+#define vstrwq_scatter_offset(p0,p1,p2) __arm_vstrwq_scatter_offset(p0,p1,p2)
+#define __arm_vstrwq_scatter_offset(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32_t_ptr][__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_offset_s32 (__ARM_mve_coerce(__p0, int32_t *), __p1, __ARM_mve_coerce(__p2, int32x4_t)), \
+ int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_offset_u32 (__ARM_mve_coerce(__p0, uint32_t *), __p1, __ARM_mve_coerce(__p2, uint32x4_t)));})
+
+#define vstrwq_scatter_offset_p(p0,p1,p2,p3) __arm_vstrwq_scatter_offset_p(p0,p1,p2,p3)
+#define __arm_vstrwq_scatter_offset_p(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32_t_ptr][__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_offset_p_s32 (__ARM_mve_coerce(__p0, int32_t *), __p1, __ARM_mve_coerce(__p2, int32x4_t), p3), \
+ int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_offset_p_u32 (__ARM_mve_coerce(__p0, uint32_t *), __p1, __ARM_mve_coerce(__p2, uint32x4_t)));})
+
+#define vstrwq_scatter_offset_p(p0,p1,p2,p3) __arm_vstrwq_scatter_offset_p(p0,p1,p2,p3)
+#define __arm_vstrwq_scatter_offset_p(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32_t_ptr][__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_offset_p_s32 (__ARM_mve_coerce(__p0, int32_t *), __p1, __ARM_mve_coerce(__p2, int32x4_t), p3), \
+ int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_offset_p_u32 (__ARM_mve_coerce(__p0, uint32_t *), __p1, __ARM_mve_coerce(__p2, uint32x4_t), p3));})
+
+#define vstrwq_scatter_offset(p0,p1,p2) __arm_vstrwq_scatter_offset(p0,p1,p2)
+#define __arm_vstrwq_scatter_offset(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32_t_ptr][__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_offset_s32 (__ARM_mve_coerce(__p0, int32_t *), __p1, __ARM_mve_coerce(__p2, int32x4_t)), \
+ int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_offset_u32 (__ARM_mve_coerce(__p0, uint32_t *), __p1, __ARM_mve_coerce(__p2, uint32x4_t)));})
+
+#define vstrwq_scatter_shifted_offset(p0,p1,p2) __arm_vstrwq_scatter_shifted_offset(p0,p1,p2)
+#define __arm_vstrwq_scatter_shifted_offset(p0,p1,p2) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32_t_ptr][__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_shifted_offset_s32 (__ARM_mve_coerce(__p0, int32_t *), __p1, __ARM_mve_coerce(__p2, int32x4_t)), \
+ int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_shifted_offset_u32 (__ARM_mve_coerce(__p0, uint32_t *), __p1, __ARM_mve_coerce(__p2, uint32x4_t)));})
+
+#define vstrwq_scatter_shifted_offset_p(p0,p1,p2,p3) __arm_vstrwq_scatter_shifted_offset_p(p0,p1,p2,p3)
+#define __arm_vstrwq_scatter_shifted_offset_p(p0,p1,p2,p3) ({ __typeof(p0) __p0 = (p0); \
+ __typeof(p1) __p1 = (p1); \
+ __typeof(p2) __p2 = (p2); \
+ _Generic( (int (*)[__ARM_mve_typeid(__p0)][__ARM_mve_typeid(__p2)])0, \
+ int (*)[__ARM_mve_type_int32_t_ptr][__ARM_mve_type_int32x4_t]: __arm_vstrwq_scatter_shifted_offset_p_s32 (__ARM_mve_coerce(__p0, int32_t *), __p1, __ARM_mve_coerce(__p2, int32x4_t), p3), \
+ int (*)[__ARM_mve_type_uint32_t_ptr][__ARM_mve_type_uint32x4_t]: __arm_vstrwq_scatter_shifted_offset_p_u32 (__ARM_mve_coerce(__p0, uint32_t *), __p1, __ARM_mve_coerce(__p2, uint32x4_t), p3));})
+
#endif /* MVE Integer. */
#define vldrdq_gather_offset(p0,p1) __arm_vldrdq_gather_offset(p0,p1)
VLDRWQGO_S VLDRWQGO_U VLDRWQGSO_F VLDRWQGSO_S
VLDRWQGSO_U VSTRHQ_F VST1Q_S VST1Q_U VSTRHQSO_S
VSTRHQSO_U VSTRHQSSO_S VSTRHQSSO_U VSTRHQ_S
- VSTRHQ_U VSTRWQ_S VSTRWQ_U VSTRWQ_F VST1Q_F])
+ VSTRHQ_U VSTRWQ_S VSTRWQ_U VSTRWQ_F VST1Q_F VSTRDQSB_S
+ VSTRDQSB_U VSTRDQSO_S VSTRDQSO_U VSTRDQSSO_S
+ VSTRDQSSO_U VSTRWQSO_S VSTRWQSO_U VSTRWQSSO_S
+ VSTRWQSSO_U VSTRHQSO_F VSTRHQSSO_F VSTRWQSB_F
+ VSTRWQSO_F VSTRWQSSO_F])
(define_mode_attr MVE_CNVT [(V8HI "V8HF") (V4SI "V4SF") (V8HF "V8HI")
(V4SF "V4SI")])
(VLDRWQGSO_S "s") (VLDRWQGSO_U "u") (VST1Q_S "s")
(VST1Q_U "u") (VSTRHQSO_S "s") (VSTRHQSO_U "u")
(VSTRHQSSO_S "s") (VSTRHQSSO_U "u") (VSTRHQ_S "s")
- (VSTRHQ_U "u") (VSTRWQ_S "s") (VSTRWQ_U "u")])
+ (VSTRHQ_U "u") (VSTRWQ_S "s") (VSTRWQ_U "u")
+ (VSTRDQSB_S "s") (VSTRDQSB_U "u") (VSTRDQSO_S "s")
+ (VSTRDQSO_U "u") (VSTRDQSSO_S "s") (VSTRDQSSO_U "u")
+ (VSTRWQSO_U "u") (VSTRWQSO_S "s") (VSTRWQSSO_U "u")
+ (VSTRWQSSO_S "s")])
(define_int_attr mode1 [(VCTP8Q "8") (VCTP16Q "16") (VCTP32Q "32")
(VCTP64Q "64") (VCTP8Q_M "8") (VCTP16Q_M "16")
(define_int_iterator VSTRHSSOQ [VSTRHQSSO_S VSTRHQSSO_U])
(define_int_iterator VSTRHQ [VSTRHQ_S VSTRHQ_U])
(define_int_iterator VSTRWQ [VSTRWQ_S VSTRWQ_U])
+(define_int_iterator VSTRDSBQ [VSTRDQSB_S VSTRDQSB_U])
+(define_int_iterator VSTRDSOQ [VSTRDQSO_S VSTRDQSO_U])
+(define_int_iterator VSTRDSSOQ [VSTRDQSSO_S VSTRDQSSO_U])
+(define_int_iterator VSTRWSOQ [VSTRWQSO_S VSTRWQSO_U])
+(define_int_iterator VSTRWSSOQ [VSTRWQSSO_S VSTRWQSSO_U])
(define_insn "*mve_mov<mode>"
[(set (match_operand:MVE_types 0 "nonimmediate_operand" "=w,w,r,w,w,r,w,Us")
emit_insn (gen_mve_vstr<V_sz_elem1>q_<supf><mode>(operands[0],operands[1]));
DONE;
})
+
+;;
+;; [vstrdq_scatter_base_p_s vstrdq_scatter_base_p_u]
+;;
+(define_insn "mve_vstrdq_scatter_base_p_<supf>v2di"
+ [(set (mem:BLK (scratch))
+ (unspec:BLK
+ [(match_operand:V2DI 0 "s_register_operand" "w")
+ (match_operand:SI 1 "mve_vldrd_immediate" "Ri")
+ (match_operand:V2DI 2 "s_register_operand" "w")
+ (match_operand:HI 3 "vpr_register_operand" "Up")]
+ VSTRDSBQ))
+ ]
+ "TARGET_HAVE_MVE"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vpst\;\tvstrdt.u64\t%q2, [%q0, %1]",ops);
+ return "";
+}
+ [(set_attr "length" "8")])
+
+;;
+;; [vstrdq_scatter_base_s vstrdq_scatter_base_u]
+;;
+(define_insn "mve_vstrdq_scatter_base_<supf>v2di"
+ [(set (mem:BLK (scratch))
+ (unspec:BLK
+ [(match_operand:V2DI 0 "s_register_operand" "=w")
+ (match_operand:SI 1 "mve_vldrd_immediate" "Ri")
+ (match_operand:V2DI 2 "s_register_operand" "w")]
+ VSTRDSBQ))
+ ]
+ "TARGET_HAVE_MVE"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vstrd.u64\t%q2, [%q0, %1]",ops);
+ return "";
+}
+ [(set_attr "length" "4")])
+
+;;
+;; [vstrdq_scatter_offset_p_s vstrdq_scatter_offset_p_u]
+;;
+(define_insn "mve_vstrdq_scatter_offset_p_<supf>v2di"
+ [(set (match_operand:V2DI 0 "memory_operand" "=Us")
+ (unspec:V2DI
+ [(match_operand:V2DI 1 "s_register_operand" "w")
+ (match_operand:V2DI 2 "s_register_operand" "w")
+ (match_operand:HI 3 "vpr_register_operand" "Up")]
+ VSTRDSOQ))
+ ]
+ "TARGET_HAVE_MVE"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vpst\;\tvstrdt.64\t%q2, [%m0, %q1]",ops);
+ return "";
+}
+ [(set_attr "length" "8")])
+
+;;
+;; [vstrdq_scatter_offset_s vstrdq_scatter_offset_u]
+;;
+(define_insn "mve_vstrdq_scatter_offset_<supf>v2di"
+ [(set (match_operand:V2DI 0 "memory_operand" "=Us")
+ (unspec:V2DI
+ [(match_operand:V2DI 1 "s_register_operand" "w")
+ (match_operand:V2DI 2 "s_register_operand" "w")]
+ VSTRDSOQ))
+ ]
+ "TARGET_HAVE_MVE"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vstrd.64\t%q2, [%m0, %q1]",ops);
+ return "";
+}
+ [(set_attr "length" "4")])
+
+;;
+;; [vstrdq_scatter_shifted_offset_p_s vstrdq_scatter_shifted_offset_p_u]
+;;
+(define_insn "mve_vstrdq_scatter_shifted_offset_p_<supf>v2di"
+ [(set (match_operand:V2DI 0 "memory_operand" "=Us")
+ (unspec:V2DI
+ [(match_operand:V2DI 1 "s_register_operand" "w")
+ (match_operand:V2DI 2 "s_register_operand" "w")
+ (match_operand:HI 3 "vpr_register_operand" "Up")]
+ VSTRDSSOQ))
+ ]
+ "TARGET_HAVE_MVE"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vpst\;\tvstrdt.64\t%q2, [%m0, %q1, UXTW #3]",ops);
+ return "";
+}
+ [(set_attr "length" "8")])
+
+;;
+;; [vstrdq_scatter_shifted_offset_s vstrdq_scatter_shifted_offset_u]
+;;
+(define_insn "mve_vstrdq_scatter_shifted_offset_<supf>v2di"
+ [(set (match_operand:V2DI 0 "memory_operand" "=Us")
+ (unspec:V2DI
+ [(match_operand:V2DI 1 "s_register_operand" "w")
+ (match_operand:V2DI 2 "s_register_operand" "w")]
+ VSTRDSSOQ))
+ ]
+ "TARGET_HAVE_MVE"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vstrd.64\t%q2, [%m0, %q1, UXTW #3]",ops);
+ return "";
+}
+ [(set_attr "length" "4")])
+
+;;
+;; [vstrhq_scatter_offset_f]
+;;
+(define_insn "mve_vstrhq_scatter_offset_fv8hf"
+ [(set (match_operand:V8HI 0 "memory_operand" "=Us")
+ (unspec:V8HI
+ [(match_operand:V8HI 1 "s_register_operand" "w")
+ (match_operand:V8HF 2 "s_register_operand" "w")]
+ VSTRHQSO_F))
+ ]
+ "TARGET_HAVE_MVE && TARGET_HAVE_MVE_FLOAT"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vstrh.16\t%q2, [%m0, %q1]",ops);
+ return "";
+}
+ [(set_attr "length" "4")])
+
+;;
+;; [vstrhq_scatter_offset_p_f]
+;;
+(define_insn "mve_vstrhq_scatter_offset_p_fv8hf"
+ [(set (match_operand:V8HI 0 "memory_operand" "=Us")
+ (unspec:V8HI
+ [(match_operand:V8HI 1 "s_register_operand" "w")
+ (match_operand:V8HF 2 "s_register_operand" "w")
+ (match_operand:HI 3 "vpr_register_operand" "Up")]
+ VSTRHQSO_F))
+ ]
+ "TARGET_HAVE_MVE && TARGET_HAVE_MVE_FLOAT"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vpst\n\tvstrht.16\t%q2, [%m0, %q1]",ops);
+ return "";
+}
+ [(set_attr "length" "8")])
+
+;;
+;; [vstrhq_scatter_shifted_offset_f]
+;;
+(define_insn "mve_vstrhq_scatter_shifted_offset_fv8hf"
+ [(set (match_operand:V8HI 0 "memory_operand" "=Us")
+ (unspec:V8HI
+ [(match_operand:V8HI 1 "s_register_operand" "w")
+ (match_operand:V8HF 2 "s_register_operand" "w")]
+ VSTRHQSSO_F))
+ ]
+ "TARGET_HAVE_MVE && TARGET_HAVE_MVE_FLOAT"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vstrh.16\t%q2, [%m0, %q1, uxtw #1]",ops);
+ return "";
+}
+ [(set_attr "length" "4")])
+
+;;
+;; [vstrhq_scatter_shifted_offset_p_f]
+;;
+(define_insn "mve_vstrhq_scatter_shifted_offset_p_fv8hf"
+ [(set (match_operand:V8HI 0 "memory_operand" "=Us")
+ (unspec:V8HI
+ [(match_operand:V8HI 1 "s_register_operand" "w")
+ (match_operand:V8HF 2 "s_register_operand" "w")
+ (match_operand:HI 3 "vpr_register_operand" "Up")]
+ VSTRHQSSO_F))
+ ]
+ "TARGET_HAVE_MVE && TARGET_HAVE_MVE_FLOAT"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vpst\n\tvstrht.16\t%q2, [%m0, %q1, uxtw #1]",ops);
+ return "";
+}
+ [(set_attr "length" "8")])
+
+;;
+;; [vstrwq_scatter_base_f]
+;;
+(define_insn "mve_vstrwq_scatter_base_fv4sf"
+ [(set (mem:BLK (scratch))
+ (unspec:BLK
+ [(match_operand:V4SI 0 "s_register_operand" "w")
+ (match_operand:SI 1 "immediate_operand" "i")
+ (match_operand:V4SF 2 "s_register_operand" "w")]
+ VSTRWQSB_F))
+ ]
+ "TARGET_HAVE_MVE && TARGET_HAVE_MVE_FLOAT"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vstrw.u32\t%q2, [%q0, %1]",ops);
+ return "";
+}
+ [(set_attr "length" "4")])
+
+;;
+;; [vstrwq_scatter_base_p_f]
+;;
+(define_insn "mve_vstrwq_scatter_base_p_fv4sf"
+ [(set (mem:BLK (scratch))
+ (unspec:BLK
+ [(match_operand:V4SI 0 "s_register_operand" "w")
+ (match_operand:SI 1 "immediate_operand" "i")
+ (match_operand:V4SF 2 "s_register_operand" "w")
+ (match_operand:HI 3 "vpr_register_operand" "Up")]
+ VSTRWQSB_F))
+ ]
+ "TARGET_HAVE_MVE && TARGET_HAVE_MVE_FLOAT"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vpst\n\tvstrwt.u32\t%q2, [%q0, %1]",ops);
+ return "";
+}
+ [(set_attr "length" "8")])
+
+;;
+;; [vstrwq_scatter_offset_f]
+;;
+(define_insn "mve_vstrwq_scatter_offset_fv4sf"
+ [(set (match_operand:V4SI 0 "memory_operand" "=Us")
+ (unspec:V4SI
+ [(match_operand:V4SI 1 "s_register_operand" "w")
+ (match_operand:V4SF 2 "s_register_operand" "w")]
+ VSTRWQSO_F))
+ ]
+ "TARGET_HAVE_MVE && TARGET_HAVE_MVE_FLOAT"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vstrw.32\t%q2, [%m0, %q1]",ops);
+ return "";
+}
+ [(set_attr "length" "4")])
+
+;;
+;; [vstrwq_scatter_offset_p_f]
+;;
+(define_insn "mve_vstrwq_scatter_offset_p_fv4sf"
+ [(set (match_operand:V4SI 0 "memory_operand" "=Us")
+ (unspec:V4SI
+ [(match_operand:V4SI 1 "s_register_operand" "w")
+ (match_operand:V4SF 2 "s_register_operand" "w")
+ (match_operand:HI 3 "vpr_register_operand" "Up")]
+ VSTRWQSO_F))
+ ]
+ "TARGET_HAVE_MVE && TARGET_HAVE_MVE_FLOAT"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vpst\n\tvstrwt.32\t%q2, [%m0, %q1]",ops);
+ return "";
+}
+ [(set_attr "length" "8")])
+
+;;
+;; [vstrwq_scatter_offset_p_s vstrwq_scatter_offset_p_u]
+;;
+(define_insn "mve_vstrwq_scatter_offset_p_<supf>v4si"
+ [(set (match_operand:V4SI 0 "memory_operand" "=Us")
+ (unspec:V4SI
+ [(match_operand:V4SI 1 "s_register_operand" "w")
+ (match_operand:V4SI 2 "s_register_operand" "w")
+ (match_operand:HI 3 "vpr_register_operand" "Up")]
+ VSTRWSOQ))
+ ]
+ "TARGET_HAVE_MVE"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vpst\n\tvstrwt.32\t%q2, [%m0, %q1]",ops);
+ return "";
+}
+ [(set_attr "length" "8")])
+
+;;
+;; [vstrwq_scatter_offset_s vstrwq_scatter_offset_u]
+;;
+(define_insn "mve_vstrwq_scatter_offset_<supf>v4si"
+ [(set (match_operand:V4SI 0 "memory_operand" "=Us")
+ (unspec:V4SI
+ [(match_operand:V4SI 1 "s_register_operand" "w")
+ (match_operand:V4SI 2 "s_register_operand" "w")]
+ VSTRWSOQ))
+ ]
+ "TARGET_HAVE_MVE"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vstrw.32\t%q2, [%m0, %q1]",ops);
+ return "";
+}
+ [(set_attr "length" "4")])
+
+;;
+;; [vstrwq_scatter_shifted_offset_f]
+;;
+(define_insn "mve_vstrwq_scatter_shifted_offset_fv4sf"
+ [(set (match_operand:V4SI 0 "memory_operand" "=Us")
+ (unspec:V4SI
+ [(match_operand:V4SI 1 "s_register_operand" "w")
+ (match_operand:V4SF 2 "s_register_operand" "w")]
+ VSTRWQSSO_F))
+ ]
+ "TARGET_HAVE_MVE && TARGET_HAVE_MVE_FLOAT"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vstrw.32\t%q2, [%m0, %q1, uxtw #2]",ops);
+ return "";
+}
+ [(set_attr "length" "4")])
+
+;;
+;; [vstrwq_scatter_shifted_offset_p_f]
+;;
+(define_insn "mve_vstrwq_scatter_shifted_offset_p_fv4sf"
+ [(set (match_operand:V4SI 0 "memory_operand" "=Us")
+ (unspec:V4SI
+ [(match_operand:V4SI 1 "s_register_operand" "w")
+ (match_operand:V4SF 2 "s_register_operand" "w")
+ (match_operand:HI 3 "vpr_register_operand" "Up")]
+ VSTRWQSSO_F))
+ ]
+ "TARGET_HAVE_MVE && TARGET_HAVE_MVE_FLOAT"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vpst\;\tvstrwt.32\t%q2, [%m0, %q1, uxtw #2]",ops);
+ return "";
+}
+ [(set_attr "length" "8")])
+
+;;
+;; [vstrwq_scatter_shifted_offset_p_s vstrwq_scatter_shifted_offset_p_u]
+;;
+(define_insn "mve_vstrwq_scatter_shifted_offset_p_<supf>v4si"
+ [(set (match_operand:V4SI 0 "memory_operand" "=Us")
+ (unspec:V4SI
+ [(match_operand:V4SI 1 "s_register_operand" "w")
+ (match_operand:V4SI 2 "s_register_operand" "w")
+ (match_operand:HI 3 "vpr_register_operand" "Up")]
+ VSTRWSSOQ))
+ ]
+ "TARGET_HAVE_MVE"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vpst\;\tvstrwt.32\t%q2, [%m0, %q1, uxtw #2]",ops);
+ return "";
+}
+ [(set_attr "length" "8")])
+
+;;
+;; [vstrwq_scatter_shifted_offset_s vstrwq_scatter_shifted_offset_u]
+;;
+(define_insn "mve_vstrwq_scatter_shifted_offset_<supf>v4si"
+ [(set (match_operand:V4SI 0 "memory_operand" "=Us")
+ (unspec:V4SI
+ [(match_operand:V4SI 1 "s_register_operand" "w")
+ (match_operand:V4SI 2 "s_register_operand" "w")]
+ VSTRWSSOQ))
+ ]
+ "TARGET_HAVE_MVE"
+{
+ rtx ops[3];
+ ops[0] = operands[0];
+ ops[1] = operands[1];
+ ops[2] = operands[2];
+ output_asm_insn ("vstrw.32\t%q2, [%m0, %q1, uxtw #2]",ops);
+ return "";
+}
+ [(set_attr "length" "4")])
projects / gcc.git / commitdiff