This is an automated email from the ASF dual-hosted git repository.
vega pushed a change to branch ci-docker-staging
in repository https://gitbox.apache.org/repos/asf/tvm.git.
omit 30b5bc9 update variables
omit 2d9f741 update
omit 85203fe fix offload
omit 5a8fb26 update jenkins file
omit bdc20f9 test this case
omit e960f29 test this
omit 1c126d9 update tests
omit c6bf006 add files
add a97fc32 fix docker image when installing rust (#7004)
add 93758ca [TVMC] use target_host when it is set (#6855)
add 7950ea1 Dynamic Batch Support for TRT (#6955)
add 0421efb [RELAY,TOPI] Add scatter_nd op (#6854)
add 73a1a9a [TOPI] deformable_conv2d in NHWC (#6999)
add fe4c66b Fix call mkl gemm in mkldnn.py (#7007)
add 0778afd Use channels from attrs if possible (#7011)
add 636739a [µTVM] Minor fixes to the Reference VM tutorial (#7012)
add e212f96 [Backend][Verilator] regression tests (#7000)
add 399e703 [µTVM] Modify reference VMs to support new µTVM demo (#7001)
add a7bf979 [AutoScheduler] Support layout rewrite for whole networks
(#6987)
add 22a0877 Fix trt Test (#7016)
add 3afde62 [AutoScheduler] Add a tutorial on auto-scheduling a network
for x86 CPU (#7019)
add 965a67e [auto_scheduler] metal default hardware params (#7022)
add 42583d6 [µTVM] Fix paths in the reference VM tutorial and add vbguest
recommendation (#7015)
add 8daa97e [Diagnostics] Add environment variable for controlling
top-level printing and fix issue with pretty printing/parsing roundtrip. (#6874)
add c1f7820 [RPC] Prefer IPv4 between IPv4 and IPv6 (#7013)
add b06b64d [CI] Hotfix CI (see #7010) (#7025)
add e6c1baf [AutoScheduler] Misc update to hardware parameter and task
scheduler (#7020)
add f4c6517 [Topi] Fix GPU Dynamic Topk by Improving Dynamic Strided
Slice in Topi (#7018)
add 91c905d [Relay][Pass] Clean up DCE tests in preparation for
refactoring. (#7029)
add 75afcd7 [AutoScheduler] Refactor task interface for tuning single
operators (#7028)
add f278c42 Save PyTorch frontend state in object (#7023)
add c8397bf [Relay][Frontend][Onnx] Add support for Size op in Onnx
frontend. (#7031)
add 9554e64 [Frontend] Prevent tflite frontend from producing int64
shape/parameters (#7030)
add 91dafd9 Add version 11.1 in finding CUDA libdevice (#7033)
add 19459e1 [CI] Update docs style dependency. (#7034)
add a78c695 [GraphRuntime] remove print from GetInputIndex (#7027)
add 54cd235 [TOPI][OP] cuda for argwhere (#6868)
add 0d46cf7 Implement Keras Conv1D (#7035)
add fdfc7eb [TVMSCRIPT] Attach span information to tir nodes in tvmscript
(#6910)
add fd5ce64 [AutoScheduler] Improve CPU matmul tutorial (#7037)
add 3d9ae3e [Relay][Topi] Fix GPU NMS when return_indices is True (#7005)
add 878a0a9 [AutoScheduler] Remove `max_registers_per_block` in
HardwareParams (#7040)
add 76b4ad0 [FIX] disable cuda test for argwhere (#7042)
add 750e7f4 [AutoScheduler] Add tips on resuming the search from a log
file (#7039)
add 3ec739c [ROCm][Auto scheduler] Support Auto scheduler and NHWC
convolution on ROCm (#7038)
add 2a2081e [TOPI] GPU scatter_add using atomic (#7044)
add 5e68e6a [DOCS] Document cloudpickle dependency in tutorials (#7049)
add a867bcb [Auto Scheduler] Add target host to measure record (#7046)
add 7a0d10c [Relay][Frontend][Onnx] MaxUnpool Operator (#7036)
add 3144cec [LLVM] Support atomic for GPU backend (NVPTX, ROCm) (#7051)
add 8ac40fa fix missing ffi binding of relay.attrs.DequantizeAttrs (#7054)
add 0095b21 [BYOC][TRT] Support batch norm for all ranks <=5, and all
axes (#7026)
add 465cd14 [TOPI] GPU scatter 1D via sorting based approach (#7056)
add c31e338 fix nvcc compile option to be compatible with older cuda
(#7065)
add e848af1 [FIX] Improve error messages and docs (#7064)
add db0215e Incremental type inference (#6900)
add f606637 [µTVM] Allow for platform-specific malloc in runtime (#6948)
add a72bdd3 [Auto Scheduler][Auto TVM] Fix infer tile size for NHWC
winograd (#7068)
add 94b2e44 [AutoSchedule] Compatibility improvement with XGBoost v1.3.0
(#7069)
add db8edc1 [TFLite] added scalar axis value handling in reduce (#6970)
add 1ccf2c2 [AutoScheduler] Delete deprecated file auto_schedule.py
(#7071)
add ec60a50 [FIX] Remove debugging print statement (#7072)
add fcead9f [Relay] Support deformable Conv2D NHWC (#7075)
add ffb6029 Fix QNN type inference (#7074)
add f7e13d7 [AutoTVM] Compatibility improvement with XGBoost v1.3.0
(#7076)
add 7a3278a [Relay][Strategy] Allow cuda cross compilation without
physical device. (#7063)
add 653f697 [VTA] update 3rdparty submodule (#7081)
add 19e2631 #7058 [Tutorial] Import errors in deploy_detection.py and
deploy_classification.py (#7059)
add 1e6e202 [VTA][OpenCL] add device_annot support in graphpack (#6125)
add df5ba51 Handle case where ListConstruct makes a python list which is
output of whole model (#7088)
add ac25b4b Add softplus operator conversion to Onnx. (#7089)
add 5535d48 Add test for MergeComposite on a QNN graph (#7080)
new 6af25b2 add verilator to ci
This update added new revisions after undoing existing revisions.
That is to say, some revisions that were in the old version of the
branch are not in the new version. This situation occurs
when a user --force pushes a change and generates a repository
containing something like this:
* -- * -- B -- O -- O -- O (30b5bc9)
\
N -- N -- N refs/heads/ci-docker-staging (6af25b2)
You should already have received notification emails for all of the O
revisions, and so the following emails describe only the N revisions
from the common base, B.
Any revisions marked "omit" are not gone; other references still
refer to them. Any revisions marked "discard" are gone forever.
The 1 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails. The revisions
listed as "add" were already present in the repository and have only
been added to this reference.
Summary of changes:
3rdparty/vta-hw | 2 +-
Jenkinsfile | 4 +-
apps/bundle_deploy/Makefile | 15 +-
apps/bundle_deploy/bundle.c | 19 +-
apps/bundle_deploy/bundle_static.c | 19 +-
apps/microtvm/reference-vm/base-box-tool.py | 33 +-
apps/microtvm/reference-vm/zephyr/Vagrantfile | 4 +
.../microtvm/reference-vm/zephyr/base-box/setup.sh | 11 +-
.../reference-vm/zephyr/base-box/test-config.json | 4 +
apps/microtvm/reference-vm/zephyr/pyproject.toml | 5 +-
apps/microtvm/reference-vm/zephyr/rebuild-tvm.sh | 1 +
apps/microtvm/reference-vm/zephyr/setup.sh | 9 +-
cmake/modules/CUDA.cmake | 1 +
cmake/modules/StandaloneCrt.cmake | 3 +-
docker/install/ubuntu_install_rust.sh | 2 +-
docs/conf.py | 1 +
docs/dev/convert_layout.rst | 1 +
docs/install/from_source.rst | 2 +-
include/tvm/auto_scheduler/compute_dag.h | 4 +-
include/tvm/auto_scheduler/search_task.h | 28 +-
include/tvm/ir/expr.h | 14 +-
include/tvm/ir/transform.h | 7 +
include/tvm/relay/attrs/nn.h | 1 +
include/tvm/relay/attrs/transform.h | 22 +
include/tvm/relay/transform.h | 14 +
include/tvm/runtime/crt/crt.h | 8 +-
include/tvm/runtime/crt/error_codes.h | 1 +
include/tvm/runtime/crt/graph_runtime.h | 11 +-
include/tvm/runtime/crt/memory.h | 56 +-
include/tvm/runtime/crt/platform.h | 22 +
include/tvm/runtime/crt/utvm_rpc_server.h | 11 +-
include/tvm/runtime/packed_func.h | 4 +-
include/tvm/tir/buffer.h | 1 +
include/tvm/tir/builtin.h | 5 +
include/tvm/tir/expr.h | 23 +-
include/tvm/tir/op.h | 443 +++-
include/tvm/tir/stmt.h | 22 +-
include/tvm/tir/var.h | 2 +
include/tvm/topi/detail/constant_utils.h | 15 +
include/tvm/topi/detail/tensor_utils.h | 37 +
include/tvm/topi/nn.h | 13 +-
include/tvm/topi/reduction.h | 24 +-
include/tvm/topi/transform.h | 111 +-
python/tvm/__init__.py | 21 +-
python/tvm/auto_scheduler/__init__.py | 14 +-
python/tvm/auto_scheduler/auto_schedule.py | 187 --
python/tvm/auto_scheduler/compute_dag.py | 40 +-
python/tvm/auto_scheduler/cost_model/xgb_model.py | 6 +-
python/tvm/auto_scheduler/measure.py | 24 +-
python/tvm/auto_scheduler/measure_record.py | 2 +-
python/tvm/auto_scheduler/relay_integration.py | 124 +-
python/tvm/auto_scheduler/search_policy.py | 4 +-
python/tvm/auto_scheduler/search_task.py | 312 ++-
python/tvm/auto_scheduler/task_scheduler.py | 23 +-
python/tvm/autotvm/tuner/xgboost_cost_model.py | 6 +-
python/tvm/contrib/mkldnn.py | 2 +-
python/tvm/contrib/nvcc.py | 44 +-
python/tvm/driver/build_module.py | 5 +-
python/tvm/driver/tvmc/compiler.py | 6 +-
python/tvm/ir/expr.py | 16 +-
python/tvm/relay/backend/compile_engine.py | 5 +-
python/tvm/relay/build_module.py | 22 +-
python/tvm/relay/frontend/common.py | 11 +
python/tvm/relay/frontend/keras.py | 85 +-
python/tvm/relay/frontend/onnx.py | 70 +
python/tvm/relay/frontend/pytorch.py | 2316 ++++++++------------
python/tvm/relay/frontend/tflite.py | 36 +-
python/tvm/relay/op/_tensor.py | 4 +
python/tvm/relay/op/_tensor_grad.py | 7 +
python/tvm/relay/op/_transform.py | 27 +-
python/tvm/relay/op/contrib/arm_compute_lib.py | 2 +-
python/tvm/relay/op/contrib/tensorrt.py | 123 +-
python/tvm/relay/op/op_attrs.py | 5 +
python/tvm/relay/op/strategy/cuda.py | 57 +-
python/tvm/relay/op/strategy/generic.py | 110 +-
python/tvm/relay/op/strategy/rocm.py | 52 +-
python/tvm/relay/op/strategy/x86.py | 33 +-
python/tvm/relay/op/transform.py | 24 +
python/tvm/relay/testing/__init__.py | 14 +-
python/tvm/rpc/base.py | 3 +
python/tvm/runtime/object_generic.py | 25 +-
python/tvm/script/intrin.py | 118 +-
python/tvm/script/parser.py | 79 +-
python/tvm/script/scope_handler.py | 74 +-
python/tvm/script/special_stmt.py | 20 +-
python/tvm/script/utils.py | 18 +-
python/tvm/te/operation.py | 37 +-
python/tvm/te/tensor.py | 2 +-
python/tvm/tir/buffer.py | 7 +-
python/tvm/tir/expr.py | 84 +-
python/tvm/tir/generic.py | 36 +-
python/tvm/tir/op.py | 222 +-
python/tvm/tir/stmt.py | 2 +-
python/tvm/topi/argwhere.py | 2 +
python/tvm/topi/cuda/__init__.py | 1 +
python/tvm/topi/cuda/argwhere.py | 654 ++++++
python/tvm/topi/cuda/conv2d_alter_op.py | 4 +-
python/tvm/topi/cuda/conv2d_winograd.py | 8 +-
python/tvm/topi/cuda/nms.py | 113 +-
python/tvm/topi/cuda/scatter.py | 292 ++-
python/tvm/topi/cuda/sort.py | 83 +-
python/tvm/topi/cuda/ssd/multibox.py | 2 +-
python/tvm/topi/generic/nn.py | 18 +
python/tvm/topi/math.py | 7 +-
python/tvm/topi/nn/conv2d.py | 41 +-
python/tvm/topi/nn/deformable_conv2d.py | 110 +-
python/tvm/topi/scatter.py | 120 +-
python/tvm/topi/testing/__init__.py | 3 +-
python/tvm/topi/testing/common.py | 31 +
..._nchw_python.py => deformable_conv2d_python.py} | 49 +
python/tvm/topi/vision/nms.py | 24 +-
python/tvm/topi/x86/__init__.py | 1 +
python/tvm/topi/x86/scatter.py | 109 +
src/auto_scheduler/compute_dag.cc | 24 +-
src/auto_scheduler/feature.cc | 5 +-
src/auto_scheduler/measure_record.cc | 16 +-
src/auto_scheduler/search_task.cc | 62 +-
src/auto_scheduler/utils.h | 2 +-
src/ir/expr.cc | 10 +-
src/ir/transform.cc | 54 +-
src/parser/parser.cc | 91 +-
src/relay/analysis/type_solver.cc | 9 +-
src/relay/backend/build_module.cc | 17 +
src/relay/backend/compile_engine.cc | 26 +-
src/relay/backend/compile_engine.h | 9 +
src/relay/backend/utils.h | 12 +-
src/relay/op/annotation/annotation.cc | 7 +-
src/relay/op/make_op.h | 2 +
src/relay/op/nn/convolution.h | 71 +-
src/relay/op/tensor/transform.cc | 137 +-
src/relay/qnn/op/concatenate.cc | 36 +-
src/relay/qnn/op/convolution.cc | 13 +-
src/relay/qnn/op/convolution_transpose.cc | 11 +-
src/relay/qnn/op/dense.cc | 15 +-
src/relay/qnn/op/op_common.h | 13 +
src/relay/qnn/op/requantize.cc | 7 +
.../transforms/auto_scheduler_layout_rewrite.cc | 160 ++
...pr_subst.cc => auto_scheduler_layout_rewrite.h} | 38 +-
src/runtime/contrib/tensorrt/tensorrt_ops.cc | 51 +-
src/runtime/contrib/tensorrt/tensorrt_runtime.cc | 40 +-
src/runtime/contrib/thrust/thrust.cu | 73 +
src/runtime/crt/Makefile | 8 +-
src/runtime/crt/common/crt_backend_api.c | 19 +-
src/runtime/crt/common/crt_runtime_api.c | 57 +-
src/runtime/crt/common/ndarray.c | 81 +-
src/runtime/crt/graph_runtime/graph_runtime.c | 477 +++-
src/runtime/crt/graph_runtime/load_json.c | 198 +-
.../graph_runtime_module/graph_runtime_module.c | 8 +-
src/runtime/crt/host/main.cc | 20 +-
.../tvm/runtime/crt/internal/common/ndarray.h | 12 +-
.../runtime/crt/internal/graph_runtime/load_json.h | 14 +-
.../crt/internal/{common => memory}/memory.h | 56 +-
src/runtime/crt/{common => memory}/memory.c | 164 +-
src/runtime/crt/utvm_rpc_server/rpc_server.cc | 27 +-
src/runtime/graph/graph_runtime.cc | 1 -
src/target/llvm/codegen_amdgpu.cc | 19 +
src/target/llvm/codegen_llvm.cc | 4 +
src/target/llvm/codegen_nvptx.cc | 14 +
src/target/llvm/llvm_module.cc | 3 +-
src/te/operation/compute_op.cc | 3 +-
src/te/schedule/schedule_dataflow_rewrite.cc | 13 +-
src/tir/ir/buffer.cc | 11 +-
src/tir/ir/expr.cc | 28 +-
src/tir/op/builtin.cc | 3 +
src/tir/op/op.cc | 504 +++--
src/tir/transforms/arg_binder.cc | 8 +-
src/tir/transforms/inject_double_buffer.cc | 6 +-
src/tir/transforms/inject_virtual_thread.cc | 6 +-
src/tir/transforms/storage_rewrite.cc | 12 +-
src/topi/schedule.cc | 4 +
tests/crt/memory_test.cc | 69 +-
tests/micro/qemu/zephyr-runtime/src/main.c | 20 +-
tests/python/contrib/test_sort.py | 34 +
tests/python/contrib/test_tensorrt.py | 194 ++
tests/python/driver/tvmc/conftest.py | 13 +
tests/python/driver/tvmc/test_compiler.py | 18 +
tests/python/frontend/keras/test_forward.py | 15 +
tests/python/frontend/onnx/test_forward.py | 126 ++
tests/python/frontend/pytorch/qnn_test.py | 63 +
tests/python/frontend/pytorch/test_forward.py | 10 +-
tests/python/frontend/tflite/test_forward.py | 18 +-
tests/python/relay/dyn/test_dynamic_op_level6.py | 4 +-
tests/python/relay/test_any.py | 123 +-
.../relay/test_auto_scheduler_layout_rewrite.py | 121 +
tests/python/relay/test_ir_parser.py | 14 +
tests/python/relay/test_json_compact.py | 27 +-
tests/python/relay/test_op_grad_level2.py | 51 +-
tests/python/relay/test_op_grad_level3.py | 9 +
tests/python/relay/test_op_level3.py | 19 +-
tests/python/relay/test_op_level5.py | 51 +-
.../relay/test_pass_dead_code_elimination.py | 267 ++-
tests/python/topi/python/test_topi_argwhere.py | 89 +
.../topi/python/test_topi_deformable_conv2d.py | 95 +-
tests/python/topi/python/test_topi_scatter.py | 67 +
tests/python/topi/python/test_topi_vision.py | 10 +-
.../unittest/test_auto_scheduler_compute_dag.py | 8 +-
.../unittest/test_auto_scheduler_cost_model.py | 2 +-
.../test_auto_scheduler_evolutionary_search.py | 10 +-
.../python/unittest/test_auto_scheduler_feature.py | 17 +-
.../unittest/test_auto_scheduler_layout_rewrite.py | 22 +-
.../python/unittest/test_auto_scheduler_measure.py | 41 +-
.../unittest/test_auto_scheduler_search_policy.py | 8 +-
.../test_auto_scheduler_sketch_generation.py | 2 +-
.../unittest/test_auto_scheduler_task_scheduler.py | 12 +-
tests/python/unittest/test_crt.py | 1 +
tests/python/unittest/test_link_params.py | 2 +-
tests/python/unittest/test_node_reflection.py | 4 +-
tests/python/unittest/test_target_codegen_llvm.py | 68 +-
tests/python/unittest/test_tvmscript_spans.py | 74 +
tests/scripts/task_ci_python_setup.sh | 2 +-
tests/scripts/task_clean.sh | 2 +-
tutorials/auto_scheduler/ci_logs/matmul.json | 2 +-
.../ci_logs/resnet-18-NHWC-B1-cuda.json | 26 +
.../auto_scheduler/ci_logs/resnet-18-NHWC-B1.json | 26 -
.../ci_logs/resnet-50-NHWC-B1-llvm.json | 31 +
tutorials/auto_scheduler/tune_conv2d_layer_cuda.py | 36 +-
tutorials/auto_scheduler/tune_matmul_x86.py | 51 +-
tutorials/auto_scheduler/tune_network_cuda.py | 30 +-
.../{tune_network_cuda.py => tune_network_x86.py} | 115 +-
tutorials/autotvm/tune_conv2d_cuda.py | 2 +-
tutorials/autotvm/tune_relay_arm.py | 2 +-
tutorials/autotvm/tune_relay_cuda.py | 2 +-
tutorials/autotvm/tune_relay_mobile_gpu.py | 2 +-
tutorials/autotvm/tune_simple_template.py | 2 +-
tutorials/micro/micro_reference_vm.py | 8 +-
tutorials/micro/micro_tflite.py | 3 +
vta/python/vta/top/graphpack.py | 132 +-
vta/tutorials/autotvm/tune_relay_vta.py | 2 +-
vta/tutorials/frontend/legacy/deploy_detection.py | 6 +-
229 files changed, 8841 insertions(+), 3503 deletions(-)
create mode 100644 apps/microtvm/reference-vm/zephyr/base-box/test-config.json
delete mode 100644 python/tvm/auto_scheduler/auto_schedule.py
create mode 100644 python/tvm/topi/cuda/argwhere.py
rename python/tvm/topi/testing/{deformable_conv2d_nchw_python.py =>
deformable_conv2d_python.py} (74%)
create mode 100644 python/tvm/topi/x86/scatter.py
create mode 100644 src/relay/transforms/auto_scheduler_layout_rewrite.cc
copy src/relay/transforms/{expr_subst.cc => auto_scheduler_layout_rewrite.h}
(56%)
rename src/runtime/crt/include/tvm/runtime/crt/internal/{common =>
memory}/memory.h (57%)
rename src/runtime/crt/{common => memory}/memory.c (70%)
create mode 100644 tests/python/relay/test_auto_scheduler_layout_rewrite.py
create mode 100644 tests/python/topi/python/test_topi_argwhere.py
create mode 100644 tests/python/topi/python/test_topi_scatter.py
create mode 100644 tests/python/unittest/test_tvmscript_spans.py
create mode 100644 tutorials/auto_scheduler/ci_logs/resnet-18-NHWC-B1-cuda.json
delete mode 100644 tutorials/auto_scheduler/ci_logs/resnet-18-NHWC-B1.json
create mode 100644 tutorials/auto_scheduler/ci_logs/resnet-50-NHWC-B1-llvm.json
copy tutorials/auto_scheduler/{tune_network_cuda.py => tune_network_x86.py}
(74%)
