/external/XNNPACK/src/qs8-gemm/gen/ |
D | 3x8c8-minmax-neon-mlal-padal.c | 163 int16x8_t vprod2x6 = vmull_s8(vb6x0, va2x0); in xnn_qs8_gemm_minmax_ukernel_3x8c8__neon_mlal_padal() local 235 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_gemm_minmax_ukernel_3x8c8__neon_mlal_padal() local
|
D | 4x8c8-minmax-neon-mlal-padal.c | 197 int16x8_t vprod2x6 = vmull_s8(vb6x0, va2x0); in xnn_qs8_gemm_minmax_ukernel_4x8c8__neon_mlal_padal() local 288 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_gemm_minmax_ukernel_4x8c8__neon_mlal_padal() local
|
D | 3x8c16-minmax-neon-mlal-padal.c | 153 int16x8_t vprod2x6 = vmull_s8(vget_low_s8(vb6), vget_low_s8(va2)); in xnn_qs8_gemm_minmax_ukernel_3x8c16__neon_mlal_padal() local
|
D | 3x8c8-minmax-neon-mull-padal.c | 134 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_gemm_minmax_ukernel_3x8c8__neon_mull_padal() local
|
D | 3x16c8-minmax-neon-mlal-padal.c | 195 int16x8_t vprod2x6 = vmull_s8(vb6x0, va2x0); in xnn_qs8_gemm_minmax_ukernel_3x16c8__neon_mlal_padal() local 347 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_gemm_minmax_ukernel_3x16c8__neon_mlal_padal() local
|
D | 4x8c8-minmax-neon-mull-padal.c | 161 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_gemm_minmax_ukernel_4x8c8__neon_mull_padal() local
|
D | 4x8c16-minmax-neon-mlal-padal.c | 186 int16x8_t vprod2x6 = vmull_s8(vget_low_s8(vb6), vget_low_s8(va2)); in xnn_qs8_gemm_minmax_ukernel_4x8c16__neon_mlal_padal() local
|
D | 4x16c8-minmax-neon-mlal-padal.c | 237 int16x8_t vprod2x6 = vmull_s8(vb6x0, va2x0); in xnn_qs8_gemm_minmax_ukernel_4x16c8__neon_mlal_padal() local 432 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_gemm_minmax_ukernel_4x16c8__neon_mlal_padal() local
|
D | 3x16c8-minmax-neon-mull-padal.c | 158 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_gemm_minmax_ukernel_3x16c8__neon_mull_padal() local
|
D | 3x16c16-minmax-neon-mlal-padal.c | 185 int16x8_t vprod2x6 = vmull_s8(vget_low_s8(vb6), vget_low_s8(va2)); in xnn_qs8_gemm_minmax_ukernel_3x16c16__neon_mlal_padal() local
|
D | 4x16c16-minmax-neon-mlal-padal.c | 226 int16x8_t vprod2x6 = vmull_s8(vget_low_s8(vb6), vget_low_s8(va2)); in xnn_qs8_gemm_minmax_ukernel_4x16c16__neon_mlal_padal() local
|
D | 4x16c8-minmax-neon-mull-padal.c | 193 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_gemm_minmax_ukernel_4x16c8__neon_mull_padal() local
|
/external/XNNPACK/src/qs8-igemm/gen/ |
D | 3x8c8-minmax-neon-mlal-padal.c | 178 int16x8_t vprod2x6 = vmull_s8(vb6x0, va2x0); in xnn_qs8_igemm_minmax_ukernel_3x8c8__neon_mlal_padal() local 250 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_igemm_minmax_ukernel_3x8c8__neon_mlal_padal() local
|
D | 4x8c8-minmax-neon-mlal-padal.c | 214 int16x8_t vprod2x6 = vmull_s8(vb6x0, va2x0); in xnn_qs8_igemm_minmax_ukernel_4x8c8__neon_mlal_padal() local 305 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_igemm_minmax_ukernel_4x8c8__neon_mlal_padal() local
|
D | 3x8c8-minmax-neon-mull-padal.c | 149 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_igemm_minmax_ukernel_3x8c8__neon_mull_padal() local
|
D | 3x8c16-minmax-neon-mlal-padal.c | 168 int16x8_t vprod2x6 = vmull_s8(vget_low_s8(vb6), vget_low_s8(va2)); in xnn_qs8_igemm_minmax_ukernel_3x8c16__neon_mlal_padal() local
|
D | 3x16c8-minmax-neon-mlal-padal.c | 210 int16x8_t vprod2x6 = vmull_s8(vb6x0, va2x0); in xnn_qs8_igemm_minmax_ukernel_3x16c8__neon_mlal_padal() local 362 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_igemm_minmax_ukernel_3x16c8__neon_mlal_padal() local
|
D | 4x8c8-minmax-neon-mull-padal.c | 178 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_igemm_minmax_ukernel_4x8c8__neon_mull_padal() local
|
D | 4x8c16-minmax-neon-mlal-padal.c | 203 int16x8_t vprod2x6 = vmull_s8(vget_low_s8(vb6), vget_low_s8(va2)); in xnn_qs8_igemm_minmax_ukernel_4x8c16__neon_mlal_padal() local
|
D | 4x16c8-minmax-neon-mlal-padal.c | 254 int16x8_t vprod2x6 = vmull_s8(vb6x0, va2x0); in xnn_qs8_igemm_minmax_ukernel_4x16c8__neon_mlal_padal() local 449 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_igemm_minmax_ukernel_4x16c8__neon_mlal_padal() local
|
D | 3x16c8-minmax-neon-mull-padal.c | 173 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_igemm_minmax_ukernel_3x16c8__neon_mull_padal() local
|
D | 3x16c16-minmax-neon-mlal-padal.c | 200 int16x8_t vprod2x6 = vmull_s8(vget_low_s8(vb6), vget_low_s8(va2)); in xnn_qs8_igemm_minmax_ukernel_3x16c16__neon_mlal_padal() local
|
D | 4x16c8-minmax-neon-mull-padal.c | 210 const int16x8_t vprod2x6 = vmull_s8(vb6, va2); in xnn_qs8_igemm_minmax_ukernel_4x16c8__neon_mull_padal() local
|
D | 4x16c16-minmax-neon-mlal-padal.c | 243 int16x8_t vprod2x6 = vmull_s8(vget_low_s8(vb6), vget_low_s8(va2)); in xnn_qs8_igemm_minmax_ukernel_4x16c16__neon_mlal_padal() local
|