2 * Copyright (c) 2017 ARM Limited.
4 * SPDX-License-Identifier: MIT
6 * Permission is hereby granted, free of charge, to any person obtaining a copy
7 * of this software and associated documentation files (the "Software"), to
8 * deal in the Software without restriction, including without limitation the
9 * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
10 * sell copies of the Software, and to permit persons to whom the Software is
11 * furnished to do so, subject to the following conditions:
13 * The above copyright notice and this permission notice shall be included in all
14 * copies or substantial portions of the Software.
16 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
17 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
18 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
19 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
20 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
21 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
28 #define DATA_TYPE_MIN 0x0
29 #endif /* DATA_TYPE_MIN */
32 #define DATA_TYPE_MAX 0xFF
33 #endif /* DATA_TYPE_MAX */
35 inline int FloatFlip(float val)
43 return (u_val.int_val >= 0) ? u_val.int_val : u_val.int_val ^ 0x7FFFFFFF;
46 __constant VEC_DATA_TYPE(DATA_TYPE, 16) type_min = (VEC_DATA_TYPE(DATA_TYPE, 16))(DATA_TYPE_MIN);
47 __constant VEC_DATA_TYPE(DATA_TYPE, 16) type_max = (VEC_DATA_TYPE(DATA_TYPE, 16))(DATA_TYPE_MAX);
48 __constant int16 idx16 = (int16)(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
50 /** This function identifies the min and maximum value of an input image.
52 * @note Input image data type must be passed as a preprocessor argument using -DDATA_TYPE.
53 * Moreover, the minimum and maximum value of the given data type must be provided using -DDATA_TYPE_MIN and -DDATA_TYPE_MAX respectively.
54 * @note In case image width is not a multiple of 16 then -DNON_MULTIPLE_OF_16 must be passed.
56 * @param[in] src_ptr Pointer to the source image. Supported data types: U8
57 * @param[in] src_stride_x Stride of the source image in X dimension (in bytes)
58 * @param[in] src_step_x src_stride_x * number of elements along X processed per workitem(in bytes)
59 * @param[in] src_stride_y Stride of the source image in Y dimension (in bytes)
60 * @param[in] src_step_y src_stride_y * number of elements along Y processed per workitem(in bytes)
61 * @param[in] src_offset_first_element_in_bytes The offset of the first element in the source image
62 * @param[out] min_max Pointer to buffer with minimum value in position 0 and maximum value in position 1
63 * @param[in] width Input image width
66 IMAGE_DECLARATION(src),
67 __global int *min_max,
70 Image src = CONVERT_TO_IMAGE_STRUCT(src);
72 // Initialize local minimum and local maximum
73 VEC_DATA_TYPE(DATA_TYPE, 16)
75 VEC_DATA_TYPE(DATA_TYPE, 16)
78 // Calculate min/max of row
80 for(; i + 16 <= width; i += 16)
82 VEC_DATA_TYPE(DATA_TYPE, 16)
83 data = vload16(0, (__global DATA_TYPE *)offset(&src, i, 0));
84 local_min = min(data, local_min);
85 local_max = max(data, local_max);
88 #ifdef NON_MULTIPLE_OF_16
89 // Handle non multiple of 16
90 VEC_DATA_TYPE(DATA_TYPE, 16)
91 data = vload16(0, (__global DATA_TYPE *)offset(&src, i, 0));
92 #ifdef IS_DATA_TYPE_FLOAT
93 int16 valid_indices = (i + idx16) < width;
94 #else /* IS_DATA_TYPE_FLOAT */
95 VEC_DATA_TYPE(DATA_TYPE, 16)
96 valid_indices = CONVERT((i + idx16) < width, VEC_DATA_TYPE(DATA_TYPE, 16));
97 #endif /* IS_DATA_TYPE_FLOAT */
98 local_max = max(local_max, select(type_min, data, valid_indices));
99 local_min = min(local_min, select(type_max, data, valid_indices));
100 #endif /* NON_MULTIPLE_OF_16 */
102 // Perform min/max reduction
103 local_min.s01234567 = min(local_min.s01234567, local_min.s89ABCDEF);
104 local_max.s01234567 = max(local_max.s01234567, local_max.s89ABCDEF);
106 local_min.s0123 = min(local_min.s0123, local_min.s4567);
107 local_max.s0123 = max(local_max.s0123, local_max.s4567);
109 local_min.s01 = min(local_min.s01, local_min.s23);
110 local_max.s01 = max(local_max.s01, local_max.s23);
112 local_min.s0 = min(local_min.s0, local_min.s1);
113 local_max.s0 = max(local_max.s0, local_max.s1);
115 // Update global min/max
116 #ifdef IS_DATA_TYPE_FLOAT
117 atomic_min(&min_max[0], FloatFlip(local_min.s0));
118 atomic_max(&min_max[1], FloatFlip(local_max.s0));
119 #else /* IS_DATA_TYPE_FLOAT */
120 atomic_min(&min_max[0], local_min.s0);
121 atomic_max(&min_max[1], local_max.s0);
122 #endif /* IS_DATA_TYPE_FLOAT */
125 /** This function counts the min and max occurrences in an image and tags their position.
127 * @note -DCOUNT_MIN_MAX should be specified if we want to count the occurrences of the minimum and maximum values.
128 * @note -DLOCATE_MIN and/or -DLOCATE_MAX should be specified if we want to store the position of each occurrence on the given array.
130 * @param[in] src_ptr Pointer to the source image. Supported data types: U8
131 * @param[in] src_stride_x Stride of the source image in X dimension (in bytes)
132 * @param[in] src_step_x src_stride_x * number of elements along X processed per workitem(in bytes)
133 * @param[in] src_stride_y Stride of the source image in Y dimension (in bytes)
134 * @param[in] src_step_y src_stride_y * number of elements along Y processed per workitem(in bytes)
135 * @param[in] src_offset_first_element_in_bytes The offset of the first element in the source image
136 * @param[in] min_max Pointer to buffer with minimum value in position 0 and maximum value in position 1
137 * @param[out] min_max_count Pointer to buffer with minimum value occurrences in position 0 and maximum value occurrences in position 1
138 * @param[out] min_loc Array that holds the location of the minimum value occurrences
139 * @param[in] max_min_loc_count The maximum number of min value occurrences coordinates the array can hold
140 * @param[out] max_loc Array that holds the location of the maximum value occurrences
141 * @param[in] max_max_loc_count The maximum number of max value occurrences coordinates the array can hold
143 __kernel void minmaxloc(
144 IMAGE_DECLARATION(src),
145 __global int *min_max,
146 __global uint *min_max_count
149 __global Coordinates2D *min_loc, uint max_min_loc_count
150 #endif /* LOCATE_MIN */
153 __global Coordinates2D *max_loc, uint max_max_loc_count
154 #endif /* LOCATE_MAX */
157 Image src = CONVERT_TO_IMAGE_STRUCT(src);
159 #ifdef IS_DATA_TYPE_FLOAT
160 __global float *min_max_ptr = (__global float *)min_max;
161 float min_value = min_max_ptr[0];
162 float max_value = min_max_ptr[1];
163 #else /* IS_DATA_TYPE_FLOAT */
164 int min_value = min_max[0];
165 int max_value = min_max[1];
166 #endif /* IS_DATA_TYPE_FLOAT */
168 DATA_TYPE value = *((__global DATA_TYPE *)src.ptr);
170 if(value == min_value)
172 uint idx = atomic_inc(&min_max_count[0]);
174 if(idx < max_min_loc_count)
176 min_loc[idx].x = get_global_id(0);
177 min_loc[idx].y = get_global_id(1);
179 #endif /* LOCATE_MIN */
181 if(value == max_value)
183 uint idx = atomic_inc(&min_max_count[1]);
185 if(idx < max_max_loc_count)
187 max_loc[idx].x = get_global_id(0);
188 max_loc[idx].y = get_global_id(1);
190 #endif /* LOCATE_MAX */
192 #endif /* COUNT_MIN_MAX */