/dports/emulators/mess/mame-mame0226/3rdparty/bgfx/examples/15-shadowmaps-simple/ |
H A D | fs_sms_shadow.sh | 30 float hardShadow(Sampler _sampler, vec4 _shadowCoord, float _bias) 34 return step(texCoord.z-_bias, unpackRgbaToFloat(texture2D(_sampler, texCoord.xy) ) ); 36 return shadow2D(_sampler, vec3(texCoord.xy, texCoord.z-_bias) ); 40 float PCF(Sampler _sampler, vec4 _shadowCoord, float _bias, vec2 _texelSize) 56 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, -1.5) * offset, 0.0, 0.0), _bias); 57 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, -0.5) * offset, 0.0, 0.0), _bias); 58 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, 0.5) * offset, 0.0, 0.0), _bias); 59 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, 1.5) * offset, 0.0, 0.0), _bias); 66 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, -1.5) * offset, 0.0, 0.0), _bias); 67 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, -0.5) * offset, 0.0, 0.0), _bias); [all …]
|
/dports/emulators/mame/mame-mame0226/3rdparty/bgfx/examples/15-shadowmaps-simple/ |
H A D | fs_sms_shadow.sh | 30 float hardShadow(Sampler _sampler, vec4 _shadowCoord, float _bias) 34 return step(texCoord.z-_bias, unpackRgbaToFloat(texture2D(_sampler, texCoord.xy) ) ); 36 return shadow2D(_sampler, vec3(texCoord.xy, texCoord.z-_bias) ); 40 float PCF(Sampler _sampler, vec4 _shadowCoord, float _bias, vec2 _texelSize) 56 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, -1.5) * offset, 0.0, 0.0), _bias); 57 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, -0.5) * offset, 0.0, 0.0), _bias); 58 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, 0.5) * offset, 0.0, 0.0), _bias); 59 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, 1.5) * offset, 0.0, 0.0), _bias); 66 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, -1.5) * offset, 0.0, 0.0), _bias); 67 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, -0.5) * offset, 0.0, 0.0), _bias); [all …]
|
/dports/emulators/mess/mame-mame0226/3rdparty/bgfx/examples/31-rsm/ |
H A D | fs_rsm_combine.sc | 23 float hardShadow(sampler2DShadow _sampler, vec4 _shadowCoord, float _bias) 26 return shadow2D(_sampler, vec3(texCoord.xy, _shadowCoord.z-_bias) ); 29 float PCF(sampler2DShadow _sampler, vec4 _shadowCoord, float _bias, vec2 _texelSize) 45 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, -1.5) * offset, 0.0, 0.0), _bias); 46 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, -0.5) * offset, 0.0, 0.0), _bias); 47 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, 0.5) * offset, 0.0, 0.0), _bias); 48 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, 1.5) * offset, 0.0, 0.0), _bias); 50 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-0.5, -1.5) * offset, 0.0, 0.0), _bias); 55 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, -1.5) * offset, 0.0, 0.0), _bias); 56 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, -0.5) * offset, 0.0, 0.0), _bias); [all …]
|
/dports/emulators/mame/mame-mame0226/3rdparty/bgfx/examples/31-rsm/ |
H A D | fs_rsm_combine.sc | 23 float hardShadow(sampler2DShadow _sampler, vec4 _shadowCoord, float _bias) 26 return shadow2D(_sampler, vec3(texCoord.xy, _shadowCoord.z-_bias) ); 29 float PCF(sampler2DShadow _sampler, vec4 _shadowCoord, float _bias, vec2 _texelSize) 45 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, -1.5) * offset, 0.0, 0.0), _bias); 46 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, -0.5) * offset, 0.0, 0.0), _bias); 47 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, 0.5) * offset, 0.0, 0.0), _bias); 48 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, 1.5) * offset, 0.0, 0.0), _bias); 50 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-0.5, -1.5) * offset, 0.0, 0.0), _bias); 55 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, -1.5) * offset, 0.0, 0.0), _bias); 56 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, -0.5) * offset, 0.0, 0.0), _bias); [all …]
|
/dports/emulators/mess/mame-mame0226/3rdparty/bgfx/examples/16-shadowmaps/ |
H A D | common.sh | 80 float hardShadow(sampler2D _sampler, vec4 _shadowCoord, float _bias) 93 float receiver = (_shadowCoord.z-_bias)/_shadowCoord.w; 100 float PCF(sampler2D _sampler, vec4 _shadowCoord, float _bias, vec4 _pcfParams, vec2 _texelSize) 105 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, -1.5) * offset, 0.0, 0.0), _bias); 115 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, -1.5) * offset, 0.0, 0.0), _bias); 116 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, -0.5) * offset, 0.0, 0.0), _bias); 117 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, 0.5) * offset, 0.0, 0.0), _bias); 118 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, 1.5) * offset, 0.0, 0.0), _bias); 141 float receiver = (_shadowCoord.z-_bias)/_shadowCoord.w * _depthMultiplier; 158 float ESM(sampler2D _sampler, vec4 _shadowCoord, float _bias, float _depthMultiplier) [all …]
|
/dports/emulators/mame/mame-mame0226/3rdparty/bgfx/examples/16-shadowmaps/ |
H A D | common.sh | 80 float hardShadow(sampler2D _sampler, vec4 _shadowCoord, float _bias) 93 float receiver = (_shadowCoord.z-_bias)/_shadowCoord.w; 100 float PCF(sampler2D _sampler, vec4 _shadowCoord, float _bias, vec4 _pcfParams, vec2 _texelSize) 105 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(-1.5, -1.5) * offset, 0.0, 0.0), _bias); 115 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, -1.5) * offset, 0.0, 0.0), _bias); 116 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, -0.5) * offset, 0.0, 0.0), _bias); 117 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, 0.5) * offset, 0.0, 0.0), _bias); 118 result += hardShadow(_sampler, _shadowCoord + vec4(vec2(0.5, 1.5) * offset, 0.0, 0.0), _bias); 141 float receiver = (_shadowCoord.z-_bias)/_shadowCoord.w * _depthMultiplier; 158 float ESM(sampler2D _sampler, vec4 _shadowCoord, float _bias, float _depthMultiplier) [all …]
|
/dports/graphics/vapoursynth-waifu2x-ncnn-vulkan/vapoursynth-waifu2x-ncnn-vulkan-r4/deps/ncnn/src/layer/arm/ |
H A D | dequantize_arm.cpp | 97 _v = vfmaq_f32(_bias, _v, _scale); in forward() 99 _v = vmlaq_f32(_bias, _v, _scale); in forward() 115 _v = vfmaq_f32(_bias, _v, _scale); in forward() 117 _v = vmlaq_f32(_bias, _v, _scale); in forward() 152 _v = vfmaq_f32(_bias, _v, _scale); in forward() 780 float32x4_t _bias = vdupq_n_f32(bias); in forward() local 785 _v = vfmaq_f32(_bias, _v, _scale); in forward() 787 _v = vmlaq_f32(_bias, _v, _scale); in forward() 869 float32x4_t _bias = vdupq_n_f32(bias); in forward() local 891 _v = vfmaq_f32(_bias, _v, _scale); in forward() [all …]
|
/dports/graphics/waifu2x-ncnn-vulkan/waifu2x-ncnn-vulkan-20210521/src/ncnn/src/layer/arm/ |
H A D | dequantize_arm.cpp | 97 _v = vfmaq_f32(_bias, _v, _scale); in forward() 99 _v = vmlaq_f32(_bias, _v, _scale); in forward() 115 _v = vfmaq_f32(_bias, _v, _scale); in forward() 117 _v = vmlaq_f32(_bias, _v, _scale); in forward() 152 _v = vfmaq_f32(_bias, _v, _scale); in forward() 780 float32x4_t _bias = vdupq_n_f32(bias); in forward() local 785 _v = vfmaq_f32(_bias, _v, _scale); in forward() 787 _v = vmlaq_f32(_bias, _v, _scale); in forward() 869 float32x4_t _bias = vdupq_n_f32(bias); in forward() local 891 _v = vfmaq_f32(_bias, _v, _scale); in forward() [all …]
|
/dports/benchmarks/vkpeak/vkpeak-20210430/ncnn/src/layer/arm/ |
H A D | dequantize_arm.cpp | 97 _v = vfmaq_f32(_bias, _v, _scale); in forward() 99 _v = vmlaq_f32(_bias, _v, _scale); in forward() 115 _v = vfmaq_f32(_bias, _v, _scale); in forward() 117 _v = vmlaq_f32(_bias, _v, _scale); in forward() 152 _v = vfmaq_f32(_bias, _v, _scale); in forward() 780 float32x4_t _bias = vdupq_n_f32(bias); in forward() local 785 _v = vfmaq_f32(_bias, _v, _scale); in forward() 787 _v = vmlaq_f32(_bias, _v, _scale); in forward() 869 float32x4_t _bias = vdupq_n_f32(bias); in forward() local 891 _v = vfmaq_f32(_bias, _v, _scale); in forward() [all …]
|
/dports/misc/ncnn/ncnn-20211208/src/layer/arm/ |
H A D | dequantize_arm.cpp | 101 _v = vfmaq_f32(_bias, _v, _scale); in forward() 103 _v = vmlaq_f32(_bias, _v, _scale); in forward() 119 _v = vfmaq_f32(_bias, _v, _scale); in forward() 121 _v = vmlaq_f32(_bias, _v, _scale); in forward() 156 _v = vfmaq_f32(_bias, _v, _scale); in forward() 784 float32x4_t _bias = vdupq_n_f32(bias); in forward() local 789 _v = vfmaq_f32(_bias, _v, _scale); in forward() 791 _v = vmlaq_f32(_bias, _v, _scale); in forward() 873 float32x4_t _bias = vdupq_n_f32(bias); in forward() local 895 _v = vfmaq_f32(_bias, _v, _scale); in forward() [all …]
|
/dports/graphics/waifu2x-ncnn-vulkan/waifu2x-ncnn-vulkan-20210521/src/ncnn/src/layer/x86/ |
H A D | scale_x86.cpp | 60 __m256 _bias = _mm256_loadu_ps(bias + i * 8); in forward_inplace() local 61 _p = _mm256_fmadd_ps(_p, _s, _bias); in forward_inplace() 97 _p = _mm256_fmadd_ps(_p, _s, _bias); in forward_inplace() 143 _p = _mm256_fmadd_ps(_p, _s, _bias); in forward_inplace() 191 __m128 _bias = _mm_loadu_ps(bias + i * 4); in forward_inplace() local 192 _p = _mm_add_ps(_mm_mul_ps(_p, _s), _bias); in forward_inplace() 223 __m128 _bias = _mm_loadu_ps((const float*)bias_data + i * 4); in forward_inplace() local 228 _p = _mm_add_ps(_mm_mul_ps(_p, _s), _bias); in forward_inplace() 274 _p = _mm_add_ps(_mm_mul_ps(_p, _s), _bias); in forward_inplace() 334 __m256 _bias = _mm256_set1_ps(bias); in forward_inplace() local [all …]
|
H A D | dequantize_x86.cpp | 69 __m256 _bias = _mm256_set1_ps(bias_data[0]); in forward() local 78 _v = _mm256_fmadd_ps(_v, _scale, _bias); in forward() 92 _v = _mm256_fmadd_ps(_v, _scale, _bias); in forward() 115 __m256 _bias = _mm256_set1_ps(bias_data[0]); in forward() local 125 _v = _mm256_fmadd_ps(_v, _scale, _bias); in forward() 290 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 336 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 539 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 585 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 856 __m128 _bias = _mm_set1_ps(bias); in forward() local [all …]
|
H A D | convolution_7x7.h | 15 …sse(const Mat& bottom_blob, Mat& top_blob, const Mat& _kernel, const Mat& _bias, const Option& opt) in conv7x7s1_sse() argument 23 …conv_im2col_sgemm_sse(bottom_blob, top_blob, _kernel, _bias, kernel_w, kernel_h, stride_w, stride_… in conv7x7s1_sse() 26 …sse(const Mat& bottom_blob, Mat& top_blob, const Mat& _kernel, const Mat& _bias, const Option& opt) in conv7x7s2_sse() argument 34 …conv_im2col_sgemm_sse(bottom_blob, top_blob, _kernel, _bias, kernel_w, kernel_h, stride_w, stride_… in conv7x7s2_sse()
|
/dports/graphics/vapoursynth-waifu2x-ncnn-vulkan/vapoursynth-waifu2x-ncnn-vulkan-r4/deps/ncnn/src/layer/x86/ |
H A D | scale_x86.cpp | 60 __m256 _bias = _mm256_loadu_ps(bias + i * 8); in forward_inplace() local 61 _p = _mm256_fmadd_ps(_p, _s, _bias); in forward_inplace() 97 _p = _mm256_fmadd_ps(_p, _s, _bias); in forward_inplace() 143 _p = _mm256_fmadd_ps(_p, _s, _bias); in forward_inplace() 191 __m128 _bias = _mm_loadu_ps(bias + i * 4); in forward_inplace() local 192 _p = _mm_add_ps(_mm_mul_ps(_p, _s), _bias); in forward_inplace() 223 __m128 _bias = _mm_loadu_ps((const float*)bias_data + i * 4); in forward_inplace() local 228 _p = _mm_add_ps(_mm_mul_ps(_p, _s), _bias); in forward_inplace() 274 _p = _mm_add_ps(_mm_mul_ps(_p, _s), _bias); in forward_inplace() 334 __m256 _bias = _mm256_set1_ps(bias); in forward_inplace() local [all …]
|
H A D | dequantize_x86.cpp | 69 __m256 _bias = _mm256_set1_ps(bias_data[0]); in forward() local 78 _v = _mm256_fmadd_ps(_v, _scale, _bias); in forward() 92 _v = _mm256_fmadd_ps(_v, _scale, _bias); in forward() 115 __m256 _bias = _mm256_set1_ps(bias_data[0]); in forward() local 125 _v = _mm256_fmadd_ps(_v, _scale, _bias); in forward() 290 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 336 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 539 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 585 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 856 __m128 _bias = _mm_set1_ps(bias); in forward() local [all …]
|
H A D | convolution_7x7.h | 15 …sse(const Mat& bottom_blob, Mat& top_blob, const Mat& _kernel, const Mat& _bias, const Option& opt) in conv7x7s1_sse() argument 23 …conv_im2col_sgemm_sse(bottom_blob, top_blob, _kernel, _bias, kernel_w, kernel_h, stride_w, stride_… in conv7x7s1_sse() 26 …sse(const Mat& bottom_blob, Mat& top_blob, const Mat& _kernel, const Mat& _bias, const Option& opt) in conv7x7s2_sse() argument 34 …conv_im2col_sgemm_sse(bottom_blob, top_blob, _kernel, _bias, kernel_w, kernel_h, stride_w, stride_… in conv7x7s2_sse()
|
/dports/benchmarks/vkpeak/vkpeak-20210430/ncnn/src/layer/x86/ |
H A D | scale_x86.cpp | 60 __m256 _bias = _mm256_loadu_ps(bias + i * 8); in forward_inplace() local 61 _p = _mm256_fmadd_ps(_p, _s, _bias); in forward_inplace() 97 _p = _mm256_fmadd_ps(_p, _s, _bias); in forward_inplace() 143 _p = _mm256_fmadd_ps(_p, _s, _bias); in forward_inplace() 191 __m128 _bias = _mm_loadu_ps(bias + i * 4); in forward_inplace() local 192 _p = _mm_add_ps(_mm_mul_ps(_p, _s), _bias); in forward_inplace() 223 __m128 _bias = _mm_loadu_ps((const float*)bias_data + i * 4); in forward_inplace() local 228 _p = _mm_add_ps(_mm_mul_ps(_p, _s), _bias); in forward_inplace() 274 _p = _mm_add_ps(_mm_mul_ps(_p, _s), _bias); in forward_inplace() 334 __m256 _bias = _mm256_set1_ps(bias); in forward_inplace() local [all …]
|
H A D | dequantize_x86.cpp | 69 __m256 _bias = _mm256_set1_ps(bias_data[0]); in forward() local 78 _v = _mm256_fmadd_ps(_v, _scale, _bias); in forward() 92 _v = _mm256_fmadd_ps(_v, _scale, _bias); in forward() 115 __m256 _bias = _mm256_set1_ps(bias_data[0]); in forward() local 125 _v = _mm256_fmadd_ps(_v, _scale, _bias); in forward() 290 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 336 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 539 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 585 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 856 __m128 _bias = _mm_set1_ps(bias); in forward() local [all …]
|
H A D | convolution_7x7.h | 15 …sse(const Mat& bottom_blob, Mat& top_blob, const Mat& _kernel, const Mat& _bias, const Option& opt) in conv7x7s1_sse() argument 23 …conv_im2col_sgemm_sse(bottom_blob, top_blob, _kernel, _bias, kernel_w, kernel_h, stride_w, stride_… in conv7x7s1_sse() 26 …sse(const Mat& bottom_blob, Mat& top_blob, const Mat& _kernel, const Mat& _bias, const Option& opt) in conv7x7s2_sse() argument 34 …conv_im2col_sgemm_sse(bottom_blob, top_blob, _kernel, _bias, kernel_w, kernel_h, stride_w, stride_… in conv7x7s2_sse()
|
/dports/misc/ncnn/ncnn-20211208/src/layer/x86/ |
H A D | scale_x86.cpp | 60 __m256 _bias = _mm256_loadu_ps(bias + i * 8); in forward_inplace() local 61 _p = _mm256_comp_fmadd_ps(_p, _s, _bias); in forward_inplace() 97 _p = _mm256_comp_fmadd_ps(_p, _s, _bias); in forward_inplace() 143 _p = _mm256_comp_fmadd_ps(_p, _s, _bias); in forward_inplace() 191 __m128 _bias = _mm_loadu_ps(bias + i * 4); in forward_inplace() local 192 _p = _mm_add_ps(_mm_mul_ps(_p, _s), _bias); in forward_inplace() 223 __m128 _bias = _mm_loadu_ps((const float*)bias_data + i * 4); in forward_inplace() local 228 _p = _mm_add_ps(_mm_mul_ps(_p, _s), _bias); in forward_inplace() 274 _p = _mm_add_ps(_mm_mul_ps(_p, _s), _bias); in forward_inplace() 334 __m256 _bias = _mm256_set1_ps(bias); in forward_inplace() local [all …]
|
H A D | dequantize_x86.cpp | 71 __m256 _bias = _mm256_set1_ps(bias_data[0]); in forward() local 80 _v = _mm256_comp_fmadd_ps(_v, _scale, _bias); in forward() 94 _v = _mm256_comp_fmadd_ps(_v, _scale, _bias); in forward() 117 __m256 _bias = _mm256_set1_ps(bias_data[0]); in forward() local 127 _v = _mm256_comp_fmadd_ps(_v, _scale, _bias); in forward() 292 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 338 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 541 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 587 __m128 _bias = _mm_set1_ps(bias_data[0]); in forward() local 858 __m128 _bias = _mm_set1_ps(bias); in forward() local [all …]
|
H A D | convolution_7x7.h | 15 …sse(const Mat& bottom_blob, Mat& top_blob, const Mat& _kernel, const Mat& _bias, const Option& opt) in conv7x7s1_sse() argument 23 …conv_im2col_sgemm_sse(bottom_blob, top_blob, _kernel, _bias, kernel_w, kernel_h, stride_w, stride_… in conv7x7s1_sse() 26 …sse(const Mat& bottom_blob, Mat& top_blob, const Mat& _kernel, const Mat& _bias, const Option& opt) in conv7x7s2_sse() argument 34 …conv_im2col_sgemm_sse(bottom_blob, top_blob, _kernel, _bias, kernel_w, kernel_h, stride_w, stride_… in conv7x7s2_sse()
|
/dports/graphics/realsr-ncnn-vulkan/realsr-ncnn-vulkan-20210210/src/ncnn/src/layer/x86/ |
H A D | scale_x86.cpp | 60 __m256 _bias = _mm256_loadu_ps(bias + i * 8); in forward_inplace() local 61 _p = _mm256_fmadd_ps(_p, _s, _bias); in forward_inplace() 97 _p = _mm256_fmadd_ps(_p, _s, _bias); in forward_inplace() 143 _p = _mm256_fmadd_ps(_p, _s, _bias); in forward_inplace() 191 __m128 _bias = _mm_loadu_ps(bias + i * 4); in forward_inplace() local 192 _p = _mm_add_ps(_mm_mul_ps(_p, _s), _bias); in forward_inplace() 223 __m128 _bias = _mm_loadu_ps((const float*)bias_data + i * 4); in forward_inplace() local 228 _p = _mm_add_ps(_mm_mul_ps(_p, _s), _bias); in forward_inplace() 274 _p = _mm_add_ps(_mm_mul_ps(_p, _s), _bias); in forward_inplace() 334 __m256 _bias = _mm256_set1_ps(bias); in forward_inplace() local [all …]
|
H A D | convolution_7x7.h | 15 …sse(const Mat& bottom_blob, Mat& top_blob, const Mat& _kernel, const Mat& _bias, const Option& opt) in conv7x7s1_sse() argument 23 …conv_im2col_sgemm_sse(bottom_blob, top_blob, _kernel, _bias, kernel_w, kernel_h, stride_w, stride_… in conv7x7s1_sse() 26 …sse(const Mat& bottom_blob, Mat& top_blob, const Mat& _kernel, const Mat& _bias, const Option& opt) in conv7x7s2_sse() argument 34 …conv_im2col_sgemm_sse(bottom_blob, top_blob, _kernel, _bias, kernel_w, kernel_h, stride_w, stride_… in conv7x7s2_sse()
|
/dports/misc/ncnn/ncnn-20211208/src/layer/mips/ |
H A D | convolution_sgemm_pack4to1.h | 25 const float* bias = _bias; in im2col_sgemm_pack4to1_msa() 258 v4i32 _bias = __msa_ld_w(biasptr, 0); in im2col_sgemm_pack4to1_msa() local 259 v4f32 _sum0 = (v4f32)__msa_splati_w(_bias, 0); in im2col_sgemm_pack4to1_msa() 260 v4f32 _sum1 = (v4f32)__msa_splati_w(_bias, 0); in im2col_sgemm_pack4to1_msa() 261 v4f32 _sum2 = (v4f32)__msa_splati_w(_bias, 0); in im2col_sgemm_pack4to1_msa() 262 v4f32 _sum3 = (v4f32)__msa_splati_w(_bias, 1); in im2col_sgemm_pack4to1_msa() 263 v4f32 _sum4 = (v4f32)__msa_splati_w(_bias, 1); in im2col_sgemm_pack4to1_msa() 264 v4f32 _sum5 = (v4f32)__msa_splati_w(_bias, 1); in im2col_sgemm_pack4to1_msa() 265 v4f32 _sum6 = (v4f32)__msa_splati_w(_bias, 2); in im2col_sgemm_pack4to1_msa() 322 v4i32 _bias = __msa_ld_w(biasptr, 0); in im2col_sgemm_pack4to1_msa() local [all …]
|