/dports/misc/tvm/incubator-tvm-0.6.1/tests/python/unittest/ |
H A D | test_schedule_tensor_core.py | 143 thread_z = tvm.thread_axis('threadIdx.z') 161 s[C].bind(j, thread_z) 179 s[AS].bind(ty, thread_z) 189 s[BS].bind(ty, thread_z) 311 thread_z = tvm.thread_axis('threadIdx.z') 324 s[Conv].bind(oc, thread_z) 343 s[AS].bind(ty, thread_z) 352 s[WS].bind(ty, thread_z)
|
/dports/misc/mxnet/incubator-mxnet-1.9.0/3rdparty/tvm/tests/python/unittest/ |
H A D | test_te_schedule_tensor_core.py | 203 thread_z = te.thread_axis("threadIdx.z") 221 s[C].bind(j, thread_z) 239 s[AS].bind(ty, thread_z) 249 s[BS].bind(ty, thread_z) 383 thread_z = te.thread_axis("threadIdx.z") 396 s[Conv].bind(oc, thread_z) 415 s[AS].bind(ty, thread_z) 424 s[WS].bind(ty, thread_z)
|
/dports/misc/py-tvm/incubator-tvm-0.6.1/tests/python/unittest/ |
H A D | test_schedule_tensor_core.py | 143 thread_z = tvm.thread_axis('threadIdx.z') 161 s[C].bind(j, thread_z) 179 s[AS].bind(ty, thread_z) 189 s[BS].bind(ty, thread_z) 311 thread_z = tvm.thread_axis('threadIdx.z') 324 s[Conv].bind(oc, thread_z) 343 s[AS].bind(ty, thread_z) 352 s[WS].bind(ty, thread_z)
|
/dports/misc/mxnet/incubator-mxnet-1.9.0/3rdparty/tvm/python/tvm/topi/intel_graphics/ |
H A D | conv2d.py | 136 thread_z = te.thread_axis((0, z_factor), "threadIdx.z") 140 s[tensor].bind(zi, thread_z) 145 return xi, thread_z, thread_y, thread_x 377 thread_z = te.thread_axis((0, z_factor), "threadIdx.z") 389 s[conv].bind(oohi, thread_z) 417 s[temp_W].bind(zi, thread_z) 597 thread_z = te.thread_axis((0, z_factor), "threadIdx.z") 609 s[conv].bind(oohi, thread_z) 636 s[temp_W].bind(zi, thread_z)
|
/dports/misc/tvm/incubator-tvm-0.6.1/topi/python/topi/intel_graphics/ |
H A D | conv2d.py | 147 thread_z = tvm.thread_axis((0, z_factor), "threadIdx.z") 151 s[tensor].bind(zi, thread_z) 156 return xi, thread_z, thread_y, thread_x 459 thread_z = tvm.thread_axis((0, z_factor), "threadIdx.z") 471 s[conv].bind(oohi, thread_z) 499 s[temp_W].bind(zi, thread_z) 702 thread_z = tvm.thread_axis((0, z_factor), "threadIdx.z") 714 s[conv].bind(oohi, thread_z) 741 s[temp_W].bind(zi, thread_z)
|
/dports/misc/py-tvm/incubator-tvm-0.6.1/topi/python/topi/intel_graphics/ |
H A D | conv2d.py | 147 thread_z = tvm.thread_axis((0, z_factor), "threadIdx.z") 151 s[tensor].bind(zi, thread_z) 156 return xi, thread_z, thread_y, thread_x 459 thread_z = tvm.thread_axis((0, z_factor), "threadIdx.z") 471 s[conv].bind(oohi, thread_z) 499 s[temp_W].bind(zi, thread_z) 702 thread_z = tvm.thread_axis((0, z_factor), "threadIdx.z") 714 s[conv].bind(oohi, thread_z) 741 s[temp_W].bind(zi, thread_z)
|
/dports/misc/mxnet/incubator-mxnet-1.9.0/3rdparty/tvm/python/tvm/topi/cuda/ |
H A D | conv2d_nhwc_tensorcore.py | 173 thread_z = te.thread_axis("threadIdx.z") 205 s[output].bind(tz, thread_z) 223 s[OL].bind(oc, thread_z) 262 s[AS].bind(tz, thread_z) 275 s[WS].bind(tz, thread_z)
|
H A D | conv2d_hwnc_tensorcore.py | 196 thread_z = te.thread_axis("threadIdx.z") 304 s[output].bind(oc, thread_z) 315 s[OL].bind(oc, thread_z) 352 s[AS].bind(ty, thread_z) 373 s[WS].bind(ty, thread_z)
|
H A D | conv3d_ndhwc_tensorcore.py | 182 thread_z = te.thread_axis("threadIdx.z") 214 s[output].bind(tz, thread_z) 232 s[OL].bind(oc, thread_z) 271 s[AS].bind(tz, thread_z) 284 s[WS].bind(tz, thread_z)
|
H A D | dense_tensorcore.py | 172 thread_z = te.thread_axis("threadIdx.z") 188 s[C].bind(tz, thread_z) 238 s[stage].bind(tz, thread_z)
|
/dports/misc/tvm/incubator-tvm-0.6.1/tutorials/optimize/ |
H A D | opt_conv_tensorcore.py | 264 thread_z = tvm.thread_axis('threadIdx.z') variable 277 s[Conv].bind(oc, thread_z) 297 s[AS].bind(ty, thread_z) 308 s[WS].bind(ty, thread_z)
|
/dports/misc/mxnet/incubator-mxnet-1.9.0/3rdparty/tvm/tutorials/optimize/ |
H A D | opt_conv_tensorcore.py | 327 thread_z = te.thread_axis("threadIdx.z") variable 340 s[Conv].bind(oc, thread_z) 360 s[AS].bind(ty, thread_z) 371 s[WS].bind(ty, thread_z)
|
/dports/misc/py-tvm/incubator-tvm-0.6.1/tutorials/optimize/ |
H A D | opt_conv_tensorcore.py | 264 thread_z = tvm.thread_axis('threadIdx.z') variable 277 s[Conv].bind(oc, thread_z) 297 s[AS].bind(ty, thread_z) 308 s[WS].bind(ty, thread_z)
|
/dports/lang/clover/mesa-21.3.6/src/amd/common/ |
H A D | ac_sqtt.h | 329 uint32_t thread_z; member
|
/dports/graphics/libosmesa-gallium/mesa-21.3.6/src/amd/common/ |
H A D | ac_sqtt.h | 329 uint32_t thread_z; member
|
/dports/graphics/libosmesa/mesa-21.3.6/src/amd/common/ |
H A D | ac_sqtt.h | 329 uint32_t thread_z; member
|
/dports/graphics/mesa-gallium-xa/mesa-21.3.6/src/amd/common/ |
H A D | ac_sqtt.h | 329 uint32_t thread_z; member
|
/dports/graphics/mesa-libs/mesa-21.3.6/src/amd/common/ |
H A D | ac_sqtt.h | 329 uint32_t thread_z; member
|
/dports/graphics/mesa-gallium-va/mesa-21.3.6/src/amd/common/ |
H A D | ac_sqtt.h | 329 uint32_t thread_z; member
|
/dports/graphics/mesa-dri-gallium/mesa-21.3.6/src/amd/common/ |
H A D | ac_sqtt.h | 329 uint32_t thread_z; member
|
/dports/graphics/mesa-gallium-vdpau/mesa-21.3.6/src/amd/common/ |
H A D | ac_sqtt.h | 329 uint32_t thread_z; member
|
/dports/graphics/mesa-dri/mesa-21.3.6/src/amd/common/ |
H A D | ac_sqtt.h | 329 uint32_t thread_z; member
|
/dports/graphics/mesa-devel/mesa-22.0-branchpoint-2059-ge8a63cf61ec/src/amd/common/ |
H A D | ac_sqtt.h | 334 uint32_t thread_z; member
|
/dports/misc/mxnet/incubator-mxnet-1.9.0/3rdparty/mkldnn/src/cpu/x64/gemm/ |
H A D | gemm_driver.cpp | 1199 = [](dim_t size_z, dim_t &thread_z, int &nthr_z, dim_t block_z_init, in set_thread_opts_pack() 1201 thread_z = utils::div_up(size_z, nthr_z); in set_thread_opts_pack() 1202 auto num_blk = utils::div_up(thread_z, block_z_init); in set_thread_opts_pack() 1203 block_z = utils::div_up(thread_z, num_blk); in set_thread_opts_pack() 1205 thread_z = num_blk * block_z; in set_thread_opts_pack() 1206 if (thread_z * nthr_z > size_z) in set_thread_opts_pack() 1207 nthr_z = utils::div_up(size_z, thread_z); in set_thread_opts_pack()
|
/dports/math/onednn/oneDNN-2.5.1/src/cpu/x64/gemm/ |
H A D | gemm_driver.cpp | 1199 = [](dim_t size_z, dim_t &thread_z, int &nthr_z, dim_t block_z_init, in set_thread_opts_pack() 1201 thread_z = utils::div_up(size_z, nthr_z); in set_thread_opts_pack() 1202 auto num_blk = utils::div_up(thread_z, block_z_init); in set_thread_opts_pack() 1203 block_z = utils::div_up(thread_z, num_blk); in set_thread_opts_pack() 1205 thread_z = num_blk * block_z; in set_thread_opts_pack() 1206 if (thread_z * nthr_z > size_z) in set_thread_opts_pack() 1207 nthr_z = utils::div_up(size_z, thread_z); in set_thread_opts_pack()
|