/* SPDX-License-Identifier: MIT
 *
 * Permission is hereby granted, free of charge, to any person
 * obtaining a copy of this software and associated documentation
 * files (the "Software"), to deal in the Software without
 * restriction, including without limitation the rights to use, copy,
 * modify, merge, publish, distribute, sublicense, and/or sell copies
 * of the Software, and to permit persons to whom the Software is
 * furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be
 * included in all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
 * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
 * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
 * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 * SOFTWARE.
 *
 * Copyright:
 *   2020      Evan Nemerson <evan@nemerson.com>
 *   2020      Christopher Moore <moore@free.fr>
 *   2023      Yi-Yen Chung <eric681@andestech.com> (Copyright owned by Andes Technology)
 */

#if !defined(SIMDE_ARM_NEON_CGT_H)
#define SIMDE_ARM_NEON_CGT_H

#include "combine.h"
#include "get_low.h"
#include "types.h"

HEDLEY_DIAGNOSTIC_PUSH
SIMDE_DISABLE_UNWANTED_DIAGNOSTICS
SIMDE_BEGIN_DECLS_

SIMDE_FUNCTION_ATTRIBUTES
uint64_t
simde_vcgtd_f64(simde_float64_t a, simde_float64_t b) {
  #if defined(SIMDE_ARM_NEON_A64V8_NATIVE)
    return HEDLEY_STATIC_CAST(uint64_t, vcgtd_f64(a, b));
  #else
    return (a > b) ? UINT64_MAX : 0;
  #endif
}
#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES)
  #undef vcgtd_f64
  #define vcgtd_f64(a, b) simde_vcgtd_f64((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
uint64_t
simde_vcgtd_s64(int64_t a, int64_t b) {
  #if defined(SIMDE_ARM_NEON_A64V8_NATIVE)
    return HEDLEY_STATIC_CAST(uint64_t, vcgtd_s64(a, b));
  #else
    return (a > b) ? UINT64_MAX : 0;
  #endif
}
#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES)
  #undef vcgtd_s64
  #define vcgtd_s64(a, b) simde_vcgtd_s64((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
uint64_t
simde_vcgtd_u64(uint64_t a, uint64_t b) {
  #if defined(SIMDE_ARM_NEON_A64V8_NATIVE)
    return HEDLEY_STATIC_CAST(uint64_t, vcgtd_u64(a, b));
  #else
    return (a > b) ? UINT64_MAX : 0;
  #endif
}
#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES)
  #undef vcgtd_u64
  #define vcgtd_u64(a, b) simde_vcgtd_u64((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
uint16_t
simde_vcgth_f16(simde_float16_t a, simde_float16_t b) {
  #if defined(SIMDE_ARM_NEON_A64V8_NATIVE) && defined(SIMDE_ARM_NEON_FP16)
    return HEDLEY_STATIC_CAST(uint16_t, vcgth_f16(a, b));
  #else
    simde_float32_t a_ = simde_float16_to_float32(a);
    simde_float32_t b_ = simde_float16_to_float32(b);

    return (a_ > b_) ? UINT16_MAX : 0;
  #endif
}
#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES)
  #undef vcgth_f16
  #define vcgth_f16(a, b) simde_vcgth_f16((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
uint32_t
simde_vcgts_f32(simde_float32_t a, simde_float32_t b) {
  #if defined(SIMDE_ARM_NEON_A64V8_NATIVE)
    return HEDLEY_STATIC_CAST(uint32_t, vcgts_f32(a, b));
  #else
    return (a > b) ? UINT32_MAX : 0;
  #endif
}
#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES)
  #undef vcgts_f32
  #define vcgts_f32(a, b) simde_vcgts_f32((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint16x8_t
simde_vcgtq_f16(simde_float16x8_t a, simde_float16x8_t b) {
  #if defined(SIMDE_ARM_NEON_A32V8_NATIVE) && defined(SIMDE_ARM_NEON_FP16)
    return vcgtq_f16(a, b);
  #else
    simde_float16x8_private
      a_ = simde_float16x8_to_private(a),
      b_ = simde_float16x8_to_private(b);
    simde_uint16x8_private r_;

    SIMDE_VECTORIZE
    for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
      r_.values[i] = simde_vcgth_f16(a_.values[i], b_.values[i]);
    }

    return simde_uint16x8_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V8_ENABLE_NATIVE_ALIASES)
  #undef vcgtq_f16
  #define vcgtq_f16(a, b) simde_vcgtq_f16((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint32x4_t
simde_vcgtq_f32(simde_float32x4_t a, simde_float32x4_t b) {
  #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
    return vcgtq_f32(a, b);
  #elif defined(SIMDE_POWER_ALTIVEC_P6_NATIVE)
    return HEDLEY_REINTERPRET_CAST(SIMDE_POWER_ALTIVEC_VECTOR(unsigned int), vec_cmpgt(a, b));
  #else
    simde_float32x4_private
      a_ = simde_float32x4_to_private(a),
      b_ = simde_float32x4_to_private(b);
    simde_uint32x4_private r_;

    #if defined(SIMDE_X86_SSE2_NATIVE)
      r_.m128i = _mm_castps_si128(_mm_cmpgt_ps(a_.m128, b_.m128));
    #elif defined(SIMDE_WASM_SIMD128_NATIVE)
      r_.v128 = wasm_f32x4_gt(a_.v128, b_.v128);
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = simde_vcgts_f32(a_.values[i], b_.values[i]);
      }
    #endif

    return simde_uint32x4_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
  #undef vcgtq_f32
  #define vcgtq_f32(a, b) simde_vcgtq_f32((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint64x2_t
simde_vcgtq_f64(simde_float64x2_t a, simde_float64x2_t b) {
  #if defined(SIMDE_ARM_NEON_A64V8_NATIVE)
    return vcgtq_f64(a, b);
  #elif defined(SIMDE_POWER_ALTIVEC_P7_NATIVE)
    return HEDLEY_REINTERPRET_CAST(SIMDE_POWER_ALTIVEC_VECTOR(unsigned long long), vec_cmpgt(a, b));
  #else
    simde_float64x2_private
      a_ = simde_float64x2_to_private(a),
      b_ = simde_float64x2_to_private(b);
    simde_uint64x2_private r_;

    #if defined(SIMDE_X86_SSE2_NATIVE)
      r_.m128i = _mm_castpd_si128(_mm_cmpgt_pd(a_.m128d, b_.m128d));
    #elif defined(SIMDE_WASM_SIMD128_NATIVE)
      r_.v128 = wasm_f64x2_gt(a_.v128, b_.v128);
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = simde_vcgtd_f64(a_.values[i], b_.values[i]);
      }
    #endif

    return simde_uint64x2_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES)
  #undef vcgtq_f64
  #define vcgtq_f64(a, b) simde_vcgtq_f64((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint8x16_t
simde_vcgtq_s8(simde_int8x16_t a, simde_int8x16_t b) {
  #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
    return vcgtq_s8(a, b);
  #elif defined(SIMDE_POWER_ALTIVEC_P6_NATIVE)
    return HEDLEY_REINTERPRET_CAST(SIMDE_POWER_ALTIVEC_VECTOR(unsigned char), vec_cmpgt(a, b));
  #else
    simde_int8x16_private
      a_ = simde_int8x16_to_private(a),
      b_ = simde_int8x16_to_private(b);
    simde_uint8x16_private r_;

    #if defined(SIMDE_X86_SSE2_NATIVE)
      r_.m128i = _mm_cmpgt_epi8(a_.m128i, b_.m128i);
    #elif defined(SIMDE_WASM_SIMD128_NATIVE)
      r_.v128 = wasm_i8x16_gt(a_.v128, b_.v128);
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = (a_.values[i] > b_.values[i]) ? UINT8_MAX : 0;
      }
    #endif

    return simde_uint8x16_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
  #undef vcgtq_s8
  #define vcgtq_s8(a, b) simde_vcgtq_s8((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint16x8_t
simde_vcgtq_s16(simde_int16x8_t a, simde_int16x8_t b) {
  #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
    return vcgtq_s16(a, b);
  #elif defined(SIMDE_POWER_ALTIVEC_P6_NATIVE)
    return HEDLEY_REINTERPRET_CAST(SIMDE_POWER_ALTIVEC_VECTOR(unsigned short), vec_cmpgt(a, b));
  #else
    simde_int16x8_private
      a_ = simde_int16x8_to_private(a),
      b_ = simde_int16x8_to_private(b);
    simde_uint16x8_private r_;

    #if defined(SIMDE_X86_SSE2_NATIVE)
      r_.m128i = _mm_cmpgt_epi16(a_.m128i, b_.m128i);
    #elif defined(SIMDE_WASM_SIMD128_NATIVE)
      r_.v128 = wasm_i16x8_gt(a_.v128, b_.v128);
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = (a_.values[i] > b_.values[i]) ? UINT16_MAX : 0;
      }
    #endif

    return simde_uint16x8_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
  #undef vcgtq_s16
  #define vcgtq_s16(a, b) simde_vcgtq_s16((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint32x4_t
simde_vcgtq_s32(simde_int32x4_t a, simde_int32x4_t b) {
  #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
    return vcgtq_s32(a, b);
  #elif defined(SIMDE_POWER_ALTIVEC_P6_NATIVE)
    return HEDLEY_REINTERPRET_CAST(SIMDE_POWER_ALTIVEC_VECTOR(unsigned int), vec_cmpgt(a, b));
  #else
    simde_int32x4_private
      a_ = simde_int32x4_to_private(a),
      b_ = simde_int32x4_to_private(b);
    simde_uint32x4_private r_;

    #if defined(SIMDE_X86_SSE2_NATIVE)
      r_.m128i = _mm_cmpgt_epi32(a_.m128i, b_.m128i);
    #elif defined(SIMDE_WASM_SIMD128_NATIVE)
      r_.v128 = wasm_i32x4_gt(a_.v128, b_.v128);
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = (a_.values[i] > b_.values[i]) ? UINT32_MAX : 0;
      }
    #endif

    return simde_uint32x4_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
  #undef vcgtq_s32
  #define vcgtq_s32(a, b) simde_vcgtq_s32((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint64x2_t
simde_vcgtq_s64(simde_int64x2_t a, simde_int64x2_t b) {
  #if defined(SIMDE_ARM_NEON_A64V8_NATIVE)
    return vcgtq_s64(a, b);
  #elif defined(SIMDE_ARM_NEON_A32V7_NATIVE)
    return vreinterpretq_u64_s64(vshrq_n_s64(vqsubq_s64(b, a), 63));
  #elif defined(SIMDE_POWER_ALTIVEC_P8_NATIVE)
    return HEDLEY_REINTERPRET_CAST(SIMDE_POWER_ALTIVEC_VECTOR(unsigned long long), vec_cmpgt(a, b));
  #else
    simde_int64x2_private
      a_ = simde_int64x2_to_private(a),
      b_ = simde_int64x2_to_private(b);
    simde_uint64x2_private r_;

    #if defined(SIMDE_X86_SSE4_2_NATIVE)
      r_.m128i = _mm_cmpgt_epi64(a_.m128i, b_.m128i);
    #elif defined(SIMDE_X86_SSE2_NATIVE)
      /* https://stackoverflow.com/a/65175746/501126 */
      __m128i r = _mm_and_si128(_mm_cmpeq_epi32(a_.m128i, b_.m128i), _mm_sub_epi64(b_.m128i, a_.m128i));
      r = _mm_or_si128(r, _mm_cmpgt_epi32(a_.m128i, b_.m128i));
      r_.m128i = _mm_shuffle_epi32(r, _MM_SHUFFLE(3,3,1,1));
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = simde_vcgtd_s64(a_.values[i], b_.values[i]);
      }
    #endif

    return simde_uint64x2_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES)
  #undef vcgtq_s64
  #define vcgtq_s64(a, b) simde_vcgtq_s64((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint8x16_t
simde_vcgtq_u8(simde_uint8x16_t a, simde_uint8x16_t b) {
  #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
    return vcgtq_u8(a, b);
  #elif defined(SIMDE_POWER_ALTIVEC_P6_NATIVE)
    return HEDLEY_REINTERPRET_CAST(SIMDE_POWER_ALTIVEC_VECTOR(unsigned char), vec_cmpgt(a, b));
  #else
    simde_uint8x16_private
      r_,
      a_ = simde_uint8x16_to_private(a),
      b_ = simde_uint8x16_to_private(b);

    #if defined(SIMDE_X86_SSE2_NATIVE)
      __m128i tmp = _mm_subs_epu8(a_.m128i, b_.m128i);
      r_.m128i = _mm_adds_epu8(tmp, _mm_sub_epi8(_mm_setzero_si128(), tmp));
    #elif defined(SIMDE_WASM_SIMD128_NATIVE)
      r_.v128 = wasm_u8x16_gt(a_.v128, b_.v128);
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = (a_.values[i] > b_.values[i]) ? UINT8_MAX : 0;
      }
    #endif

    return simde_uint8x16_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
  #undef vcgtq_u8
  #define vcgtq_u8(a, b) simde_vcgtq_u8((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint16x8_t
simde_vcgtq_u16(simde_uint16x8_t a, simde_uint16x8_t b) {
  #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
    return vcgtq_u16(a, b);
  #elif defined(SIMDE_POWER_ALTIVEC_P6_NATIVE)
    return HEDLEY_REINTERPRET_CAST(SIMDE_POWER_ALTIVEC_VECTOR(unsigned short), vec_cmpgt(a, b));
  #else
    simde_uint16x8_private
      r_,
      a_ = simde_uint16x8_to_private(a),
      b_ = simde_uint16x8_to_private(b);

    #if defined(SIMDE_X86_SSE2_NATIVE)
      __m128i tmp = _mm_subs_epu16(a_.m128i, b_.m128i);
      r_.m128i = _mm_adds_epu16(tmp, _mm_sub_epi16(_mm_setzero_si128(), tmp));
    #elif defined(SIMDE_WASM_SIMD128_NATIVE)
      r_.v128 = wasm_u16x8_gt(a_.v128, b_.v128);
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = (a_.values[i] > b_.values[i]) ? UINT16_MAX : 0;
      }
    #endif

    return simde_uint16x8_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
  #undef vcgtq_u16
  #define vcgtq_u16(a, b) simde_vcgtq_u16((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint32x4_t
simde_vcgtq_u32(simde_uint32x4_t a, simde_uint32x4_t b) {
  #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
    return vcgtq_u32(a, b);
  #elif defined(SIMDE_POWER_ALTIVEC_P6_NATIVE)
    return HEDLEY_REINTERPRET_CAST(SIMDE_POWER_ALTIVEC_VECTOR(unsigned int), vec_cmpgt(a, b));
  #else
    simde_uint32x4_private
      r_,
      a_ = simde_uint32x4_to_private(a),
      b_ = simde_uint32x4_to_private(b);

    #if defined(SIMDE_X86_SSE2_NATIVE)
      r_.m128i =
        _mm_xor_si128(
          _mm_cmpgt_epi32(a_.m128i, b_.m128i),
          _mm_srai_epi32(_mm_xor_si128(a_.m128i, b_.m128i), 31)
        );
    #elif defined(SIMDE_WASM_SIMD128_NATIVE)
      r_.v128 = wasm_u32x4_gt(a_.v128, b_.v128);
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = (a_.values[i] > b_.values[i]) ? UINT32_MAX : 0;
      }
    #endif

    return simde_uint32x4_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
  #undef vcgtq_u32
  #define vcgtq_u32(a, b) simde_vcgtq_u32((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint64x2_t
simde_vcgtq_u64(simde_uint64x2_t a, simde_uint64x2_t b) {
  #if defined(SIMDE_ARM_NEON_A64V8_NATIVE)
    return vcgtq_u64(a, b);
  #elif defined(SIMDE_POWER_ALTIVEC_P8_NATIVE)
    return HEDLEY_REINTERPRET_CAST(SIMDE_POWER_ALTIVEC_VECTOR(unsigned long long), vec_cmpgt(a, b));
  #else
    simde_uint64x2_private
      r_,
      a_ = simde_uint64x2_to_private(a),
      b_ = simde_uint64x2_to_private(b);

    #if defined(SIMDE_X86_SSE4_2_NATIVE)
      __m128i sign_bit = _mm_set1_epi64x(INT64_MIN);
      r_.m128i = _mm_cmpgt_epi64(_mm_xor_si128(a_.m128i, sign_bit), _mm_xor_si128(b_.m128i, sign_bit));
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = simde_vcgtd_u64(a_.values[i], b_.values[i]);
      }
    #endif

    return simde_uint64x2_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES)
  #undef vcgtq_u64
  #define vcgtq_u64(a, b) simde_vcgtq_u64((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint16x4_t
simde_vcgt_f16(simde_float16x4_t a, simde_float16x4_t b) {
  #if defined(SIMDE_ARM_NEON_A32V8_NATIVE) && defined(SIMDE_ARM_NEON_FP16)
    return vcgt_f16(a, b);
  #else
    simde_float16x4_private
      a_ = simde_float16x4_to_private(a),
      b_ = simde_float16x4_to_private(b);
    simde_uint16x4_private r_;

    SIMDE_VECTORIZE
    for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
      r_.values[i] = simde_vcgth_f16(a_.values[i], b_.values[i]);
    }

    return simde_uint16x4_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V8_ENABLE_NATIVE_ALIASES)
  #undef vcgt_f16
  #define vcgt_f16(a, b) simde_vcgt_f16((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint32x2_t
simde_vcgt_f32(simde_float32x2_t a, simde_float32x2_t b) {
  #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
    return vcgt_f32(a, b);
  #else
    simde_float32x2_private
      a_ = simde_float32x2_to_private(a),
      b_ = simde_float32x2_to_private(b);
    simde_uint32x2_private r_;

    #if defined(SIMDE_VECTOR_SUBSCRIPT_OPS) && !defined(SIMDE_BUG_GCC_100762)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = simde_vcgts_f32(a_.values[i], b_.values[i]);
      }
    #endif

    return simde_uint32x2_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
  #undef vcgt_f32
  #define vcgt_f32(a, b) simde_vcgt_f32((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint64x1_t
simde_vcgt_f64(simde_float64x1_t a, simde_float64x1_t b) {
  #if defined(SIMDE_ARM_NEON_A64V8_NATIVE)
    return vcgt_f64(a, b);
  #else
    simde_float64x1_private
      a_ = simde_float64x1_to_private(a),
      b_ = simde_float64x1_to_private(b);
    simde_uint64x1_private r_;

    #if defined(SIMDE_VECTOR_SUBSCRIPT_OPS)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = simde_vcgtd_f64(a_.values[i], b_.values[i]);
      }
    #endif

    return simde_uint64x1_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES)
  #undef vcgt_f64
  #define vcgt_f64(a, b) simde_vcgt_f64((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint8x8_t
simde_vcgt_s8(simde_int8x8_t a, simde_int8x8_t b) {
  #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
    return vcgt_s8(a, b);
  #else
    simde_int8x8_private
      a_ = simde_int8x8_to_private(a),
      b_ = simde_int8x8_to_private(b);
    simde_uint8x8_private r_;

    #if defined(SIMDE_X86_MMX_NATIVE)
      r_.m64 = _mm_cmpgt_pi8(a_.m64, b_.m64);
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS) && !defined(SIMDE_BUG_GCC_100762)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = (a_.values[i] > b_.values[i]) ? UINT8_MAX : 0;
      }
    #endif

    return simde_uint8x8_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
  #undef vcgt_s8
  #define vcgt_s8(a, b) simde_vcgt_s8((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint16x4_t
simde_vcgt_s16(simde_int16x4_t a, simde_int16x4_t b) {
  #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
    return vcgt_s16(a, b);
  #else
    simde_int16x4_private
      a_ = simde_int16x4_to_private(a),
      b_ = simde_int16x4_to_private(b);
    simde_uint16x4_private r_;

    #if defined(SIMDE_X86_MMX_NATIVE)
      r_.m64 = _mm_cmpgt_pi16(a_.m64, b_.m64);
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS) && !defined(SIMDE_BUG_GCC_100762)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = (a_.values[i] > b_.values[i]) ? UINT16_MAX : 0;
      }
    #endif

    return simde_uint16x4_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
  #undef vcgt_s16
  #define vcgt_s16(a, b) simde_vcgt_s16((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint32x2_t
simde_vcgt_s32(simde_int32x2_t a, simde_int32x2_t b) {
  #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
    return vcgt_s32(a, b);
  #else
    simde_int32x2_private
      a_ = simde_int32x2_to_private(a),
      b_ = simde_int32x2_to_private(b);
    simde_uint32x2_private r_;

    #if defined(SIMDE_X86_MMX_NATIVE)
      r_.m64 = _mm_cmpgt_pi32(a_.m64, b_.m64);
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS) && !defined(SIMDE_BUG_GCC_100762)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = (a_.values[i] > b_.values[i]) ? UINT32_MAX : 0;
      }
    #endif

    return simde_uint32x2_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
  #undef vcgt_s32
  #define vcgt_s32(a, b) simde_vcgt_s32((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint64x1_t
simde_vcgt_s64(simde_int64x1_t a, simde_int64x1_t b) {
  #if defined(SIMDE_ARM_NEON_A64V8_NATIVE)
    return vcgt_s64(a, b);
  #else
    simde_int64x1_private
      a_ = simde_int64x1_to_private(a),
      b_ = simde_int64x1_to_private(b);
    simde_uint64x1_private r_;

    #if defined(SIMDE_VECTOR_SUBSCRIPT_OPS)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = simde_vcgtd_s64(a_.values[i], b_.values[i]);
      }
    #endif

    return simde_uint64x1_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES)
  #undef vcgt_s64
  #define vcgt_s64(a, b) simde_vcgt_s64((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint8x8_t
simde_vcgt_u8(simde_uint8x8_t a, simde_uint8x8_t b) {
  #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
    return vcgt_u8(a, b);
  #else
    simde_uint8x8_private
      r_,
      a_ = simde_uint8x8_to_private(a),
      b_ = simde_uint8x8_to_private(b);

    #if defined(SIMDE_X86_MMX_NATIVE)
      __m64 sign_bit = _mm_set1_pi8(INT8_MIN);
      r_.m64 = _mm_cmpgt_pi8(_mm_xor_si64(a_.m64, sign_bit), _mm_xor_si64(b_.m64, sign_bit));
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS) && !defined(SIMDE_BUG_GCC_100762)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = (a_.values[i] > b_.values[i]) ? UINT8_MAX : 0;
      }
    #endif

    return simde_uint8x8_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
  #undef vcgt_u8
  #define vcgt_u8(a, b) simde_vcgt_u8((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint16x4_t
simde_vcgt_u16(simde_uint16x4_t a, simde_uint16x4_t b) {
  #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
    return vcgt_u16(a, b);
  #else
    simde_uint16x4_private
      r_,
      a_ = simde_uint16x4_to_private(a),
      b_ = simde_uint16x4_to_private(b);

    #if defined(SIMDE_X86_MMX_NATIVE)
      __m64 sign_bit = _mm_set1_pi16(INT16_MIN);
      r_.m64 = _mm_cmpgt_pi16(_mm_xor_si64(a_.m64, sign_bit), _mm_xor_si64(b_.m64, sign_bit));
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS) && !defined(SIMDE_BUG_GCC_100762)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = (a_.values[i] > b_.values[i]) ? UINT16_MAX : 0;
      }
    #endif

    return simde_uint16x4_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
  #undef vcgt_u16
  #define vcgt_u16(a, b) simde_vcgt_u16((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint32x2_t
simde_vcgt_u32(simde_uint32x2_t a, simde_uint32x2_t b) {
  #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
    return vcgt_u32(a, b);
  #else
    simde_uint32x2_private
      r_,
      a_ = simde_uint32x2_to_private(a),
      b_ = simde_uint32x2_to_private(b);

    #if defined(SIMDE_X86_MMX_NATIVE)
      __m64 sign_bit = _mm_set1_pi32(INT32_MIN);
      r_.m64 = _mm_cmpgt_pi32(_mm_xor_si64(a_.m64, sign_bit), _mm_xor_si64(b_.m64, sign_bit));
    #elif defined(SIMDE_VECTOR_SUBSCRIPT_OPS) && !defined(SIMDE_BUG_GCC_100762)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = (a_.values[i] > b_.values[i]) ? UINT32_MAX : 0;
      }
    #endif

    return simde_uint32x2_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
  #undef vcgt_u32
  #define vcgt_u32(a, b) simde_vcgt_u32((a), (b))
#endif

SIMDE_FUNCTION_ATTRIBUTES
simde_uint64x1_t
simde_vcgt_u64(simde_uint64x1_t a, simde_uint64x1_t b) {
  #if defined(SIMDE_ARM_NEON_A64V8_NATIVE)
    return vcgt_u64(a, b);
  #else
    simde_uint64x1_private
      r_,
      a_ = simde_uint64x1_to_private(a),
      b_ = simde_uint64x1_to_private(b);

    #if defined(SIMDE_VECTOR_SUBSCRIPT_OPS)
      r_.values = HEDLEY_REINTERPRET_CAST(__typeof__(r_.values), a_.values > b_.values);
    #else
      SIMDE_VECTORIZE
      for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
        r_.values[i] = simde_vcgtd_u64(a_.values[i], b_.values[i]);
      }
    #endif

    return simde_uint64x1_from_private(r_);
  #endif
}
#if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES)
  #undef vcgt_u64
  #define vcgt_u64(a, b) simde_vcgt_u64((a), (b))
#endif

SIMDE_END_DECLS_
HEDLEY_DIAGNOSTIC_POP

#endif /* !defined(SIMDE_ARM_NEON_CGT_H) */