This is an automated email from the ASF dual-hosted git repository.
github-bot pushed a change to branch
dependabot/pip/apps/microtvm/ethosu/tflite-2.10.0
in repository https://gitbox.apache.org/repos/asf/tvm.git
discard 54b3e74823 Bump tflite from 2.4.0 to 2.10.0 in /apps/microtvm/ethosu
add e662970647 [microTVM] enable building microTVM components by default
(#13073)
add 1c67768483 [MetaSchedule][Minor] Unify Cuda-TensorCore Naming in
Schedule Rule Kind (#13473)
add a41e192dc8 [MetaSchedule] TorchBench tuning script: add task
extraction mode (#13452)
add e2fc4d7e98 [TVMScript] Improvements tvm.script.highlight (#13438)
add 8cccc253da [ci] Enable CRT tests for CPU minimal build (#13471)
add b2058f4dd2 [CI][Docker] Store GTest sources in GPU docker image
(#13468)
add ca5bc958d1 [tir]delete useless param in driver_api.cc (#13474)
add dcea36e76d Add python venvs to demo_ images that build (#13435)
add 3680b3cb49 [microTVM][Zephyr] Add 'serial_number' option (#13377)
add fae4c5f1b3 [TIR] Fix an error when the result of compute_at has unit
loop (#13481)
add 2105b937cd [OpenCL][Adreno] Remove PrimFunc parameters annotation
(#13483)
add f5a102c83c [CMSIS-NN] Support for int16 in fully connected layer
(#13484)
add 101e3a4ade [TIR][Transform] Optional data-flow analysis in RemoveNoOp
(#13217)
add fc606c09b2 [TIR][TVMScript] Cleaner printing of And/Or chains (#13432)
add 9098b497bb [TIR] Correct type annotation for `rfactor` (#13485)
add 61a4f21412 [RUNTIME] Correctly handling export_module when exporting
modules of different type (#13489)
add ae4fd7df7d [Relay][Pattern] Enable rewrite_once in
class:DFPatternRewrite (#13490)
add b8d7cd7fe0 [DOCKER][ADRENO] we don't need microtvm being built for
android cross… (#13486)
add c38a0c50a7 [Adreno] Add documentation for Adreno deployment (#13393)
add 5d8fc204a5 [ACL] Enable int8 data type in QNN DENSE (#13487)
add b6151bcaa2 Fix typo in golang sample (#13476)
add fc59b6dbdf Bump pillow from 8.3.2 to 9.3.0 in /apps/microtvm/ethosu
(#13464)
add a92a3dd84f [ACL] Enable int8 data type in QNN CONV2D (#13496)
add 61144f9d87 [ACL] Enable int8 data type in CONCATENATE (#13497)
add 25e98dd5e4 [ACL] Enable int8 data type in pooling operators (#13488)
add 1e5fc25649 [Hexagon]Call Acquire/Release resources API in Hexagon
Launcher durin… (#13495)
add 449d674e8d [TIR][Schedule] Add condition to check buffer type (#13429)
add 2a812f90f3 Bump pillow from 8.3.2 to 9.3.0 in /apps/microtvm/cmsisnn
(#13465)
add 36d18e905b [TIR] Fix buffer shape and IndexMap indices dtype mismatch
(#13463)
add 3252362d94 [MetaSchedule] Enhance Database Validation Script (#13459)
add 5b1a1e3d39 [skip ci][ci][wasm] Add package-lock.json to git (#13505)
add 435df5081a [microTVM] Use `serial_number` in Zephyr tutorials (#13479)
add 40d0ec7515 [ci][docker] Update Docker image tags (#13504)
add 95d2e9fa35 [ARM] Add dynamic matvec support (#13502)
add 57de9e7f3d Revert "[microTVM] enable building microTVM components by
default" (#13503)
add c0ba8a1951 [docs][Adreno] Remove unnecessary compilation flag (#13509)
add f6f7feafb2 [CMSIS-NN] Support int16 handling for pooling functions
(#13498)
add b9f89a2eb9 [Docs][Bug] Fix broken link to tvmc python (#13499)
add b587e33027 [microTVM][Tutorial] Fix micro_aot and micro_autotune
tutorials (#13513)
add 694d4bf5ea [tir] Add copy on write to all nodes (#13512)
add e47eed13d9 Add methods to get the size of VTCM on device as well as
the allocated size of the HexagonVtcmPool (#13511)
add 8dc8d248a1 [LLVM] Switch to using New Pass Manager (NPM) with LLVM 16+
(#13515)
add c2dd53d531 [TE][TIR] Improved naming when converting TE to schedulable
TIR (#13431)
add ab3f54d3f7 [CI] Add `set -x` for demo scripts (#13523)
add 6782a35018 [TIR] Disable RewriteSimplifier extensions in RemoveNoOp
(#13524)
add c7d7164c42 [QNN] support zero points as variable scalar for
QnnBatchMatMul op (#13469)
add bf16b42edb [Hexagon] Add HVX quant conv2d implementation (#13256)
add afbfb7aa7e [TIR][Analysis][Hexagon] Add vtcm memory capacity
verification for Hexagon target (#13349)
add a923ed3c89 [TIR] Introduce ReduceBranchingThroughOvercompute (#13299)
add 72c3399207 [ci][docker] Update NRF command line tools URL (#13541)
add d31a1fb0db [ci] Dis-allow any non-S3 URLs in CI (#13283)
add 7bc41ecca2 [Logging] Mark LogFatal::~LogFatal as [[noreturn]] (#13542)
add 3a81aef40b [Fix] Use proper target in VerifyGPUCode (#13548)
add e7160d569a Add recursive on loop with marked kUnrolled (#13536)
add 012551ffda [microNPU] Fix cascade scheduling stability (#13428)
add 2b110367d1 [microTVM][Arduino]Add `serial_number` to project options
and tests (#13518)
add ec09bbfbd0 Bump tflite from 2.4.0 to 2.10.0 in /apps/microtvm/ethosu
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 (54b3e74823)
\
N -- N -- N
refs/heads/dependabot/pip/apps/microtvm/ethosu/tflite-2.10.0 (ec09bbfbd0)
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.
No new revisions were added by this update.
Summary of changes:
Jenkinsfile | 24 +-
apps/dso_plugin_module/plugin_module.cc | 1 -
apps/hexagon_launcher/launcher_hexagon.cc | 6 +
.../template_project/microtvm_api_server.py | 51 +-
apps/microtvm/cmsisnn/requirements.txt | 116 +-
apps/microtvm/cmsisnn/run_demo.sh | 1 +
apps/microtvm/ethosu/requirements.txt | 116 +-
apps/microtvm/ethosu/run_demo.sh | 1 +
apps/microtvm/poetry.lock | 556 +-
apps/microtvm/pyproject.toml | 1 +
.../template_project/CMakeLists.txt.template | 4 +-
.../zephyr/template_project/microtvm_api_server.py | 192 +-
ci/jenkins/Build.groovy.j2 | 3 +-
ci/jenkins/Jenkinsfile.j2 | 18 +-
ci/jenkins/Test.groovy.j2 | 1 +
cmake/modules/Hexagon.cmake | 9 +
docker/Dockerfile.ci_gpu | 2 +-
docker/Dockerfile.demo_android | 7 +
docker/Dockerfile.demo_rocm | 8 +-
docker/Dockerfile.demo_vitis_ai | 4 +
docker/install/ubuntu_install_googletest.sh | 21 +-
docker/install/ubuntu_install_nrfjprog.sh | 2 +-
docs/how_to/deploy/adreno.rst | 336 +
docs/how_to/deploy/index.rst | 1 +
.../how_to/deploy_models/deploy_model_on_adreno.py | 351 +
gallery/how_to/work_with_microtvm/micro_aot.py | 8 +-
.../how_to/work_with_microtvm/micro_autotune.py | 9 +-
gallery/how_to/work_with_microtvm/micro_tflite.py | 3 +-
gallery/tutorial/tvmc_command_line_driver.py | 8 +-
golang/sample/complex.go | 2 +-
include/tvm/ir/expr.h | 2 +
include/tvm/runtime/data_type.h | 1 -
include/tvm/runtime/device_api.h | 1 -
include/tvm/runtime/logging.h | 6 +-
include/tvm/runtime/packed_func.h | 1 -
include/tvm/tir/analysis.h | 16 +
include/tvm/tir/expr.h | 30 +
include/tvm/tir/expr_functor.h | 1 -
include/tvm/tir/op.h | 1 -
include/tvm/tir/op_attr_types.h | 1 -
include/tvm/tir/stmt.h | 13 +-
include/tvm/tir/stmt_functor.h | 1 -
python/tvm/autotvm/measure/measure_methods.py | 33 +-
python/tvm/ir/module.py | 16 +-
python/tvm/meta_schedule/testing/torchbench/run.py | 28 +-
.../tvm/meta_schedule/testing/validate_database.py | 781 ++-
python/tvm/micro/testing/pytest_plugin.py | 13 +
python/tvm/relay/op/contrib/arm_compute_lib.py | 21 +-
python/tvm/relay/op/contrib/cmsisnn.py | 45 +-
python/tvm/relay/op/strategy/arm_cpu.py | 18 +-
python/tvm/relay/op/strategy/x86.py | 9 +-
python/tvm/runtime/module.py | 17 +-
python/tvm/script/highlight.py | 264 +-
python/tvm/target/target.py | 8 +
python/tvm/tir/analysis/analysis.py | 16 +
python/tvm/tir/function.py | 16 +-
python/tvm/tir/schedule/schedule.py | 6 +-
python/tvm/tir/schedule/trace.py | 14 +-
python/tvm/tir/transform/transform.py | 22 +
python/tvm/topi/utils.py | 7 +-
src/arith/bound_deducer.cc | 1 -
src/arith/iter_affine_map.cc | 2 -
src/arith/narrow_predicate_expression.cc | 1 -
src/arith/rewrite_simplify.cc | 26 +-
src/arith/transitive_comparison_analyzer.cc | 3 -
src/auto_scheduler/feature.cc | 7 +
src/auto_scheduler/search_policy/utils.h | 7 +-
src/auto_scheduler/utils.h | 1 -
src/contrib/ethosu/cascader/pareto.cc | 6 +
src/driver/driver_api.cc | 19 +-
src/ir/function.cc | 1 -
src/ir/module.cc | 1 -
.../measure_callback/remove_build_artifact.cc | 2 +-
src/meta_schedule/module_equality.cc | 1 -
src/meta_schedule/postproc/verify_gpu_code.cc | 11 +-
.../space_generator/space_generator.cc | 4 +-
src/parser/meta_ref.cc | 1 -
src/parser/parser.cc | 1 -
src/parser/token.h | 2 -
src/printer/tvmscript_printer.cc | 9 +-
src/relay/backend/aot/aot_lower_main.cc | 1 -
src/relay/backend/aot_executor_codegen.cc | 1 -
src/relay/backend/contrib/cmsisnn/relay_to_tir.cc | 64 +-
.../backend/contrib/cmsisnn/tir_to_runtime.cc | 6 +-
src/relay/backend/contrib/codegen_c/codegen.cc | 1 -
.../backend/contrib/codegen_json/codegen_json.h | 1 -
src/relay/backend/contrib/cutlass/codegen.cc | 2 -
src/relay/backend/contrib/dnnl/codegen.cc | 3 -
src/relay/backend/graph_executor_codegen.cc | 8 -
src/relay/backend/interpreter.cc | 6 -
src/relay/backend/te_compiler_cache.cc | 6 -
src/relay/backend/vm/compiler.cc | 1 -
src/relay/op/nn/sparse.cc | 2 -
src/relay/qnn/op/batch_matmul.cc | 78 +-
src/relay/qnn/utils.h | 2 -
src/relay/transforms/annotate_texture_storage.cc | 12 -
src/relay/transforms/defunctionalization.cc | 1 -
src/relay/transforms/fold_constant.cc | 1 -
src/relay/transforms/fold_scale_axis.cc | 2 -
src/relay/transforms/infer_layout_utils.cc | 1 -
src/relay/transforms/simplify_expr.h | 4 +-
src/relay/transforms/to_mixed_precision.cc | 1 -
src/runtime/contrib/arm_compute_lib/acl_utils.cc | 1 -
src/runtime/contrib/bnns/bnns_json_runtime.cc | 1 -
src/runtime/contrib/clml/clml_runtime.cc | 2 -
src/runtime/contrib/cublas/cublas_utils.h | 1 -
src/runtime/contrib/dnnl/dnnl_tensor_requisite.h | 1 -
src/runtime/contrib/tflite/tflite_runtime.cc | 1 -
.../graph_executor/debug/graph_executor_debug.cc | 1 -
src/runtime/hexagon/hexagon_buffer.cc | 1 -
src/runtime/hexagon/hexagon_device_api.cc | 6 +
src/runtime/hexagon/hexagon_module.cc | 1 -
src/runtime/hexagon/hexagon_vtcm_pool.cc | 15 +-
src/runtime/hexagon/hexagon_vtcm_pool.h | 14 +-
{include/tvm => src}/runtime/hexagon/ops/conv2d.h | 145 +-
src/runtime/hexagon/ops/conv2d_fp16_hvx.cc | 57 +-
src/runtime/hexagon/ops/conv2d_quant_hvx.cc | 319 +
src/runtime/hexagon/ops/conv_utils.cc | 170 +-
src/runtime/logging.cc | 1 -
src/runtime/module.cc | 2 -
src/runtime/opencl/opencl_common.h | 1 -
src/runtime/opencl/opencl_device_api.cc | 1 -
src/runtime/pack_args.h | 1 -
src/runtime/pipeline/pipeline_executor.cc | 1 -
src/runtime/pipeline/pipeline_struct.h | 3 -
src/runtime/rpc/rpc_event_impl.cc | 6 +-
src/runtime/rpc/rpc_module.cc | 1 -
src/runtime/rpc/rpc_session.cc | 1 -
src/runtime/stackvm/stackvm.cc | 1 -
src/runtime/stackvm/stackvm.h | 4 -
src/runtime/thread_storage_scope.h | 2 -
src/runtime/vm/executable.cc | 2 -
src/runtime/vm/vm.cc | 1 -
src/runtime/vulkan/vulkan_device.cc | 1 -
src/support/base64.h | 1 -
src/support/scalars.cc | 2 -
src/support/socket.h | 1 -
src/target/llvm/codegen_amdgpu.cc | 4 +-
src/target/llvm/codegen_cpu.cc | 2 -
src/target/llvm/codegen_llvm.cc | 83 +-
src/target/llvm/codegen_llvm.h | 3 +
src/target/llvm/codegen_nvptx.cc | 4 +-
src/target/source/ptx.cc | 2 -
src/target/spirv/codegen_spirv.cc | 2 -
src/target/target.cc | 1 -
src/target/target_kind.cc | 1 +
src/te/autodiff/jacobian.cc | 1 -
src/te/operation/create_primfunc.cc | 14 +-
src/te/schedule/schedule_dataflow_rewrite.cc | 1 -
src/tir/analysis/calculate_allocated_memory.cc | 117 +
src/tir/analysis/control_flow_graph.cc | 117 +-
src/tir/analysis/control_flow_graph.h | 12 +-
src/tir/analysis/stmt_finding.cc | 1 -
src/tir/ir/expr_functor.cc | 1 -
src/tir/ir/stmt_functor.cc | 3 -
src/tir/op/op.cc | 6 -
src/tir/schedule/primitive/cache_index.cc | 1 -
src/tir/schedule/primitive/cache_read_write.cc | 3 -
src/tir/schedule/primitive/compute_at.cc | 2 +-
src/tir/schedule/primitive/compute_inline.cc | 2 -
.../schedule/primitive/layout_transformation.cc | 35 +-
src/tir/transforms/bf16_legalize.cc | 2 -
src/tir/transforms/bound_checker.cc | 2 -
src/tir/transforms/inject_double_buffer.cc | 2 -
src/tir/transforms/inject_virtual_thread.cc | 4 -
src/tir/transforms/ir_utils.cc | 2 -
src/tir/transforms/lift_attr_scope.cc | 1 -
src/tir/transforms/loop_partition.cc | 3 +-
src/tir/transforms/lower_custom_datatypes.cc | 2 -
src/tir/transforms/lower_thread_allreduce.cc | 2 -
src/tir/transforms/lower_warp_memory.cc | 2 -
.../merge_dynamic_shared_memory_allocations.cc | 2 -
src/tir/transforms/narrow_datatype.cc | 2 -
.../reduce_branching_through_overcompute.cc | 178 +
src/tir/transforms/remove_no_op.cc | 232 +-
src/tir/transforms/remove_no_op.h | 60 +
src/tir/transforms/renew_defs.cc | 2 -
src/tir/transforms/rewrite_unsafe_select.cc | 1 -
src/tir/transforms/simplify.cc | 1 -
.../script_complete.h => transforms/simplify.h} | 21 +-
src/tir/transforms/split_host_device.cc | 2 -
src/tir/transforms/storage_flatten.cc | 5 -
src/tir/transforms/storage_rewrite.cc | 4 -
src/tir/transforms/thread_storage_sync.cc | 2 -
src/tir/transforms/unroll_loop.cc | 1 -
src/tir/transforms/update_pointer_storage_scope.cc | 2 -
src/tir/transforms/vectorize_loop.cc | 6 -
src/tir/usmp/algo/hill_climb.cc | 1 -
.../cpp-runtime/hexagon/hexagon_conv_utils_test.h | 102 +
.../hexagon/hexagon_fp16_utils_tests.cc | 96 +-
...utils_tests.cc => hexagon_quant_utils_tests.cc} | 175 +-
.../cpp-runtime/hexagon/hexagon_vtcm_pool_tests.cc | 9 +-
tests/micro/arduino/README.md | 7 +
.../micro/arduino/test_arduino_error_detection.py | 7 +-
tests/micro/arduino/test_arduino_rpc_server.py | 91 +-
tests/micro/arduino/test_arduino_workflow.py | 3 +-
tests/micro/arduino/test_utils.py | 3 +-
.../test_hexagon => micro/project_api}/__init__.py | 2 +-
tests/micro/project_api/test_project_api.py | 34 +-
.../test_zephyr_microtvm_api_server.py} | 19 +-
.../project_api/{test_project_api.py => utils.py} | 38 +-
tests/micro/zephyr/README.md | 6 +
.../test_hexagon => micro/zephyr}/__init__.py | 2 +-
tests/micro/zephyr/conftest.py | 8 -
tests/micro/zephyr/test_zephyr.py | 123 +-
tests/micro/zephyr/test_zephyr_aot_exec.py | 34 +-
.../zephyr/test_zephyr_aot_exec_standalone.py | 31 +-
tests/micro/zephyr/test_zephyr_armv7m.py | 30 +-
tests/micro/zephyr/{test_utils.py => utils.py} | 16 +-
.../contrib/test_arm_compute_lib/infrastructure.py | 18 +
.../test_arm_compute_lib/test_concatenate.py | 72 +-
.../contrib/test_arm_compute_lib/test_conv2d.py | 25 +-
.../contrib/test_arm_compute_lib/test_dense.py | 29 +-
.../contrib/test_arm_compute_lib/test_pooling.py | 414 +-
.../contrib/test_cmsisnn/test_fully_connected.py | 25 +-
tests/python/contrib/test_cmsisnn/test_pooling.py | 12 +-
.../python/contrib/test_hexagon/infrastructure.py | 6 +-
tests/python/contrib/test_hexagon/test_vtcm.py | 55 +-
.../test_hexagon/topi/test_conv2d_quant_intrin.py | 261 +
tests/python/frontend/darknet/test_forward.py | 200 +-
.../opencl_texture/test_conv2d_nchw_texture.py | 14 +-
.../python/unittest/test_arith_rewrite_simplify.py | 9 +
.../test_meta_schedule_schedule_rule_mlt.py | 10 +-
.../python/unittest/test_runtime_module_export.py | 6 +-
...test_tir_analysis_calculate_allocated_memory.py | 101 +
.../unittest/test_tir_schedule_compute_at.py | 52 +
.../unittest/test_tir_schedule_transform_layout.py | 48 +-
.../test_tir_transform_inject_software_pipeline.py | 4 +-
.../unittest/test_tir_transform_loop_partition.py | 69 +
...ansform_reduce_branching_through_overcompute.py | 219 +
.../unittest/test_tir_transform_remove_no_op.py | 521 ++
.../python/unittest/test_tir_transform_simplify.py | 1 +
tests/python/unittest/test_tvmscript_roundtrip.py | 32 +
tests/scripts/request_hook/request_hook.py | 161 +-
tests/scripts/task_build_adreno_bins.sh | 1 +
tests/scripts/task_config_build_minimal.sh | 1 +
vta/runtime/runtime.cc | 1 -
web/.gitignore | 1 -
web/emcc/webgpu_runtime.cc | 16 +-
web/package-lock.json | 6724 ++++++++++++++++++++
web/package.json | 3 +-
241 files changed, 13299 insertions(+), 1959 deletions(-)
create mode 100644 docs/how_to/deploy/adreno.rst
create mode 100644 gallery/how_to/deploy_models/deploy_model_on_adreno.py
rename {include/tvm => src}/runtime/hexagon/ops/conv2d.h (54%)
create mode 100644 src/runtime/hexagon/ops/conv2d_quant_hvx.cc
create mode 100644 src/tir/analysis/calculate_allocated_memory.cc
create mode 100644 src/tir/transforms/reduce_branching_through_overcompute.cc
create mode 100644 src/tir/transforms/remove_no_op.h
copy src/tir/{ir/script/script_complete.h => transforms/simplify.h} (66%)
create mode 100644 tests/cpp-runtime/hexagon/hexagon_conv_utils_test.h
copy tests/cpp-runtime/hexagon/{hexagon_fp16_utils_tests.cc =>
hexagon_quant_utils_tests.cc} (58%)
copy tests/{python/contrib/test_hexagon => micro/project_api}/__init__.py (92%)
copy tests/{python/unittest/test_tir_transform_decorate_device_scope.py =>
micro/project_api/test_zephyr_microtvm_api_server.py} (70%)
copy tests/micro/project_api/{test_project_api.py => utils.py} (66%)
copy tests/{python/contrib/test_hexagon => micro/zephyr}/__init__.py (93%)
rename tests/micro/zephyr/{test_utils.py => utils.py} (96%)
create mode 100644
tests/python/contrib/test_hexagon/topi/test_conv2d_quant_intrin.py
create mode 100644
tests/python/unittest/test_tir_analysis_calculate_allocated_memory.py
create mode 100644
tests/python/unittest/test_tir_transform_reduce_branching_through_overcompute.py
create mode 100644 web/package-lock.json
