/external/XNNPACK/src/f32-dwconv2d-chw/gen/ |
D | 5x5s2p2-minmax-neon-1x4.c | 168 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE9BDF.val[0], vi3xGIKMHJLN.val[0], 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neon_1x4() local 264 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE, vzero, 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neon_1x4() local
|
D | 5x5s2p2-minmax-neonfma-1x4-acc3.c | 168 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE9BDF.val[0], vi3xGIKMHJLN.val[0], 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neonfma_1x4_acc3() local 266 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE, vzero, 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neonfma_1x4_acc3() local
|
D | 5x5s2p2-minmax-neon-1x4-acc2.c | 168 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE9BDF.val[0], vi3xGIKMHJLN.val[0], 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neon_1x4_acc2() local 265 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE, vzero, 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neon_1x4_acc2() local
|
D | 5x5s2p2-minmax-neon-1x4-acc3.c | 168 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE9BDF.val[0], vi3xGIKMHJLN.val[0], 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neon_1x4_acc3() local 266 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE, vzero, 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neon_1x4_acc3() local
|
D | 5x5s2p2-minmax-neonfma-1x4.c | 168 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE9BDF.val[0], vi3xGIKMHJLN.val[0], 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neonfma_1x4() local 264 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE, vzero, 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neonfma_1x4() local
|
D | 5x5s2p2-minmax-neonfma-1x4-acc2.c | 168 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE9BDF.val[0], vi3xGIKMHJLN.val[0], 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neonfma_1x4_acc2() local 265 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE, vzero, 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neonfma_1x4_acc2() local
|
D | 5x5s2p2-minmax-neon-1x4-acc5.c | 168 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE9BDF.val[0], vi3xGIKMHJLN.val[0], 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neon_1x4_acc5() local 268 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE, vzero, 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neon_1x4_acc5() local
|
D | 5x5s2p2-minmax-neon-1x4-acc4.c | 168 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE9BDF.val[0], vi3xGIKMHJLN.val[0], 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neon_1x4_acc4() local 267 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE, vzero, 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neon_1x4_acc4() local
|
D | 5x5s2p2-minmax-neonfma-1x4-acc5.c | 168 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE9BDF.val[0], vi3xGIKMHJLN.val[0], 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neonfma_1x4_acc5() local 268 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE, vzero, 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neonfma_1x4_acc5() local
|
D | 5x5s2p2-minmax-neonfma-1x4-acc4.c | 168 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE9BDF.val[0], vi3xGIKMHJLN.val[0], 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neonfma_1x4_acc4() local 267 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE, vzero, 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neonfma_1x4_acc4() local
|
D | 5x5s2p2-minmax-wasmsimd-x86-splat-1x4-acc2.c | 216 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vi3xGIKM, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_x86_splat_1x4_acc2() local 313 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vzero, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_x86_splat_1x4_acc2() local
|
D | 5x5s2p2-minmax-wasmsimd-arm-splat-1x4.c | 216 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vi3xGIKM, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_arm_splat_1x4() local 312 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vzero, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_arm_splat_1x4() local
|
D | 5x5s2p2-minmax-wasmsimd-arm-splat-1x4-acc2.c | 216 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vi3xGIKM, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_arm_splat_1x4_acc2() local 313 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vzero, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_arm_splat_1x4_acc2() local
|
D | 5x5s2p2-minmax-wasmsimd-x86-splat-1x4.c | 216 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vi3xGIKM, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_x86_splat_1x4() local 312 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vzero, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_x86_splat_1x4() local
|
D | 5x5s2p2-minmax-wasmsimd-arm-splat-1x4-acc5.c | 216 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vi3xGIKM, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_arm_splat_1x4_acc5() local 316 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vzero, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_arm_splat_1x4_acc5() local
|
D | 5x5s2p2-minmax-wasmsimd-arm-splat-1x4-acc3.c | 216 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vi3xGIKM, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_arm_splat_1x4_acc3() local 314 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vzero, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_arm_splat_1x4_acc3() local
|
D | 5x5s2p2-minmax-wasmsimd-x86-splat-1x4-acc3.c | 216 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vi3xGIKM, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_x86_splat_1x4_acc3() local 314 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vzero, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_x86_splat_1x4_acc3() local
|
D | 5x5s2p2-minmax-wasmsimd-x86-splat-1x4-acc5.c | 216 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vi3xGIKM, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_x86_splat_1x4_acc5() local 316 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vzero, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_x86_splat_1x4_acc5() local
|
D | 5x5s2p2-minmax-wasmsimd-arm-splat-1x4-acc4.c | 216 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vi3xGIKM, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_arm_splat_1x4_acc4() local 315 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vzero, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_arm_splat_1x4_acc4() local
|
D | 5x5s2p2-minmax-wasmsimd-x86-splat-1x4-acc4.c | 216 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vi3xGIKM, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_x86_splat_1x4_acc4() local 315 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vzero, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_x86_splat_1x4_acc4() local
|
D | 5x5s2p2-minmax-neon-2x4-acc2.c | 216 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE9BDF.val[0], vi3xGIKMHJLN.val[0], 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neon_2x4_acc2() local 355 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE, vzero, 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neon_2x4_acc2() local
|
D | 5x5s2p2-minmax-neon-2x4.c | 216 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE9BDF.val[0], vi3xGIKMHJLN.val[0], 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neon_2x4() local 353 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE, vzero, 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neon_2x4() local
|
D | 5x5s2p2-minmax-neonfma-2x4-acc2.c | 216 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE9BDF.val[0], vi3xGIKMHJLN.val[0], 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neonfma_2x4_acc2() local 355 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE, vzero, 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neonfma_2x4_acc2() local
|
D | 5x5s2p2-minmax-neonfma-2x4.c | 216 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE9BDF.val[0], vi3xGIKMHJLN.val[0], 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neonfma_2x4() local 353 const float32x4_t vi3xACEG = vextq_f32(vi3x8ACE, vzero, 1); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__neonfma_2x4() local
|
D | 5x5s2p2-minmax-wasmsimd-arm-loadsplat-1x4-acc2.c | 236 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vi3xGIKM, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_arm_loadsplat_1x4_acc2() local 333 const v128_t vi3xACEG = wasm_v32x4_shuffle(vi3x8ACE, vzero, 1, 2, 3, 4); in xnn_f32_dwconv2d_chw_ukernel_5x5s2p2__wasmsimd_arm_loadsplat_1x4_acc2() local
|