/external/XNNPACK/src/qs8-gemm/gen/ |
D | 3x8c8-minmax-rndnu-neon-mlal.c | 123 int16x8_t vprod2x2 = vmull_s8(vb2x0, va2x0); in xnn_qs8_gemm_minmax_rndnu_ukernel_3x8c8__neon_mlal() local 207 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_gemm_minmax_rndnu_ukernel_3x8c8__neon_mlal() local
|
D | 4x8c8-minmax-rndnu-neon-mlal.c | 145 int16x8_t vprod2x2 = vmull_s8(vb2x0, va2x0); in xnn_qs8_gemm_minmax_rndnu_ukernel_4x8c8__neon_mlal() local 252 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_gemm_minmax_rndnu_ukernel_4x8c8__neon_mlal() local
|
D | 3x8c8-minmax-rndnu-neon-mull.c | 106 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_gemm_minmax_rndnu_ukernel_3x8c8__neon_mull() local
|
D | 3x8c16-minmax-rndnu-neon-mlal.c | 117 int16x8_t vprod2x2 = vmull_s8(vget_low_s8(vb2), vget_low_s8(va2)); in xnn_qs8_gemm_minmax_rndnu_ukernel_3x8c16__neon_mlal() local
|
D | 3x16c8-minmax-rndnu-neon-mlal.c | 155 int16x8_t vprod2x2 = vmull_s8(vb2x0, va2x0); in xnn_qs8_gemm_minmax_rndnu_ukernel_3x16c8__neon_mlal() local 319 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_gemm_minmax_rndnu_ukernel_3x16c8__neon_mlal() local
|
D | 4x8c8-minmax-rndnu-neon-mull.c | 125 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_gemm_minmax_rndnu_ukernel_4x8c8__neon_mull() local
|
D | 4x8c16-minmax-rndnu-neon-mlal.c | 138 int16x8_t vprod2x2 = vmull_s8(vget_low_s8(vb2), vget_low_s8(va2)); in xnn_qs8_gemm_minmax_rndnu_ukernel_4x8c16__neon_mlal() local
|
D | 4x16c8-minmax-rndnu-neon-mlal.c | 185 int16x8_t vprod2x2 = vmull_s8(vb2x0, va2x0); in xnn_qs8_gemm_minmax_rndnu_ukernel_4x16c8__neon_mlal() local 396 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_gemm_minmax_rndnu_ukernel_4x16c8__neon_mlal() local
|
D | 3x16c8-minmax-rndnu-neon-mull.c | 130 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_gemm_minmax_rndnu_ukernel_3x16c8__neon_mull() local
|
D | 3x16c16-minmax-rndnu-neon-mlal.c | 149 int16x8_t vprod2x2 = vmull_s8(vget_low_s8(vb2), vget_low_s8(va2)); in xnn_qs8_gemm_minmax_rndnu_ukernel_3x16c16__neon_mlal() local
|
D | 4x16c8-minmax-rndnu-neon-mull.c | 157 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_gemm_minmax_rndnu_ukernel_4x16c8__neon_mull() local
|
D | 4x16c16-minmax-rndnu-neon-mlal.c | 178 int16x8_t vprod2x2 = vmull_s8(vget_low_s8(vb2), vget_low_s8(va2)); in xnn_qs8_gemm_minmax_rndnu_ukernel_4x16c16__neon_mlal() local
|
/external/XNNPACK/src/qs8-igemm/gen/ |
D | 3x8c8-minmax-rndnu-neon-mlal.c | 138 int16x8_t vprod2x2 = vmull_s8(vb2x0, va2x0); in xnn_qs8_igemm_minmax_rndnu_ukernel_3x8c8__neon_mlal() local 222 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_igemm_minmax_rndnu_ukernel_3x8c8__neon_mlal() local
|
D | 4x8c8-minmax-rndnu-neon-mlal.c | 162 int16x8_t vprod2x2 = vmull_s8(vb2x0, va2x0); in xnn_qs8_igemm_minmax_rndnu_ukernel_4x8c8__neon_mlal() local 269 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_igemm_minmax_rndnu_ukernel_4x8c8__neon_mlal() local
|
D | 3x8c16-minmax-rndnu-neon-mlal.c | 132 int16x8_t vprod2x2 = vmull_s8(vget_low_s8(vb2), vget_low_s8(va2)); in xnn_qs8_igemm_minmax_rndnu_ukernel_3x8c16__neon_mlal() local
|
D | 3x8c8-minmax-rndnu-neon-mull.c | 121 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_igemm_minmax_rndnu_ukernel_3x8c8__neon_mull() local
|
D | 3x16c8-minmax-rndnu-neon-mlal.c | 170 int16x8_t vprod2x2 = vmull_s8(vb2x0, va2x0); in xnn_qs8_igemm_minmax_rndnu_ukernel_3x16c8__neon_mlal() local 334 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_igemm_minmax_rndnu_ukernel_3x16c8__neon_mlal() local
|
D | 4x8c16-minmax-rndnu-neon-mlal.c | 155 int16x8_t vprod2x2 = vmull_s8(vget_low_s8(vb2), vget_low_s8(va2)); in xnn_qs8_igemm_minmax_rndnu_ukernel_4x8c16__neon_mlal() local
|
D | 4x8c8-minmax-rndnu-neon-mull.c | 142 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_igemm_minmax_rndnu_ukernel_4x8c8__neon_mull() local
|
D | 4x16c8-minmax-rndnu-neon-mlal.c | 202 int16x8_t vprod2x2 = vmull_s8(vb2x0, va2x0); in xnn_qs8_igemm_minmax_rndnu_ukernel_4x16c8__neon_mlal() local 413 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_igemm_minmax_rndnu_ukernel_4x16c8__neon_mlal() local
|
D | 3x16c8-minmax-rndnu-neon-mull.c | 145 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_igemm_minmax_rndnu_ukernel_3x16c8__neon_mull() local
|
D | 3x16c16-minmax-rndnu-neon-mlal.c | 164 int16x8_t vprod2x2 = vmull_s8(vget_low_s8(vb2), vget_low_s8(va2)); in xnn_qs8_igemm_minmax_rndnu_ukernel_3x16c16__neon_mlal() local
|
D | 4x16c16-minmax-rndnu-neon-mlal.c | 195 int16x8_t vprod2x2 = vmull_s8(vget_low_s8(vb2), vget_low_s8(va2)); in xnn_qs8_igemm_minmax_rndnu_ukernel_4x16c16__neon_mlal() local
|
D | 4x16c8-minmax-rndnu-neon-mull.c | 174 const int16x8_t vprod2x2 = vmull_s8(vb2, va2); in xnn_qs8_igemm_minmax_rndnu_ukernel_4x16c8__neon_mull() local
|