This is an automated email from the ASF dual-hosted git repository.
driazati pushed a change to branch separate-scripts
in repository https://gitbox.apache.org/repos/asf/tvm.git
discard a4b0489fe9 Add GITHUB_SCRIPT_ROOT and JENKINS_SCRIPT_ROOT
omit 889dced86e [ci] Split the ci/scripts directory into ci/scripts/github
and ci/scripts/jenkins
add ea01e3ffb4 [TIR] Preserve loop annotations in inject_software_pipeline
pass (#12937)
add 2379917985 [MetaSchedule] Add Script for TorchBench Model Tuning &
Benchmarking (#12914)
add 595f0b3975 [HEXAGON][QHL] Clippling the inputs of HVX version of QHL
Sigmoid operation (#12919)
add 25a54fb791 [TIR] Remove unused iters from the result of reindex
(#12946)
add 77c8b6e163 [Support] Add fallback definition of ccache in libinfo
(#12945)
add 4e4089edda [MetaSchedule] Fix XGBoost Import Issue (#12936)
add e9eb0bc660 [LLVM] Change CHECK_NE(x, nullptr) to CHECK(x != nullptr),
NFC (#12943)
add dedf6393f1 [Hexagon] Change NULL to nullptr, NFC (#12944)
add d4bf9ecf55 [Target] Add target_device_type attribute to override
default device_type (#12509)
add bf5637dc32 [DOCS][COMMUNITY] Elaborate Independence Principle for
Project Participation (#12962)
add c3357f6820 [Relay][Op] Register some forgotten op in Python side
(#12963)
add fa17da22c7 [Hexagon] Support template-free meta schedule tuning
(#12854)
add f121e5e355 [Hexagon] [runtime] VTCM Allocator (#12947)
add f3d3ecebe1 [Hexagon] vrmpy tensorization for e2e compilation of int8
models (#12911)
add 4e260d183f [BugFix][Pattern] Fixed a bug in PatternGrouper (#12901)
add de6d806775 [CMSIS-NN] Support for int16 conv2d (#12950)
add 1ea1a0bc88 [Hexagon] 3-stage pipeline; multi queue async DMA for cache
read / write (#12954)
add 61a7632b55 [microTVM] Use default Project Options in template projects
and add Makefile for Arduino template project (#12818)
add 3a125375a3 [Hexagon] Don't print simulator protocol messages (#12958)
add a661864f85 [skip ci] Modify issue templates to align with Issue
Tracking RFC (#12898)
add af01526ae2 [skip ci] Edits to the Bug & Flaky test Issue templates to
reduce word count (#12985)
add a997c23e94 [CODEGEN][OPENCL] Sampler definition should be at outermost
scope (#12951)
add 9618e6a457 [TIR] Use IndexMap to transform NDArray (#12949)
add 393d5f7fd2 [FIX,PROFILER] Fix PAPI test and test in CI (#12953)
add 2e257f0376 [Hexagon] Do not use `target` test fixture in Hexagon tests
(#12981)
add 2860a50ff9 [Hexagon] Fix RPC session close by adding shutdown
PackedFunc (#12960)
add d023ef47fc [TIR, MetaSchedule] Preserve unit block iters for
auto-tensorization (#12974)
add 59b945742d [FIX,AUTOSCHEDULER,METASCHEDULE] Handle negative extents in
featurization (#12990)
add 3d5e440fff [ci] Disable flaky ethosu + roofline tests (#12956)
add 2d50979606 [TVMScript] Allow T.bool type annotations (#12975)
add 7fc35da3b9 [TEST] CPU feature detection for x86 and ARM dot product
instructions (#12980)
add 1b9e20a807 [ETHOSN] Transpose fully connected weights (#12970)
add 6147c31b36 [ETHOSN] Remove backwards copy in the runtime (#12968)
add 5e24aa2b78 [Hexagon] [runtime] Query for total and available VTCM
(#12992)
add 87c466cdee [Textures] Improve error reporting (#12986)
add e375c311da [Arith][IndexMap] Correct MapShape result for small
vectorized dims (#12927)
add 7ff42a1d4e [TIR] Require dtype.is_float() inside FloatImm (#12928)
add 50df4abf3b [Relay] optimize dumpir time (#12792)
add 78df9c1d79 [LOGGING] Optionally print backtrace on segfault (#12959)
add bb2cec1941 [Fix,MetaSchedule] Correct log usage in arithmetic
intensity feature (#12806)
add 6780c9f87d [MetaSchedule] Tuning API cleanup & ergonomics (#12895)
add 7804a9886c [Hexagon] disable cache_write schedule type from sw
pipeline test (#13004)
add fc333f90e8 [TIR][Arith] Prove conditionals by transitively applying
knowns (#12863)
add d9e01475af [UnitTest][TIR] Support IRModule comparisons in
CompareBeforeAfter (#12920)
add 20aa0cf2f7 [ci] Re-enable roofline test (#13007)
add 189338c919 [MetaSchedule] Support RewriteLayout postproc on
AllocateConst (#12991)
add d92d47ad7f [docs] fixed codebase_walkthrough document bug (#13008)
add f21b5ca4ae [TIR][Schedule] Add cache_inplace primitive to cache opaque
buffer (#12939)
add aedbe45b6a [Node] Expose StructuralEqual/Hash handler implemenation to
header (#13001)
add f8666a9469 [ci] Template build steps (#12983)
add 84c50ef045 [ci][docs] Add external network resources to S3 (#12989)
add fcbcd156c7 [microTVM] Add Cortex-M DSP schedules for optimal conv2d
layouts (#12969)
add 0b034d77bd [Doc] Fix the typo in the debugging doc (#13032)
add cf96072a04 [COMMUNITY] Add andrewzhaoluo key for release (#13025)
add afeab6e8a3 [TVMScript] AST, Source and diagnostics for Parser (#12978)
add cd077575f3 Update to 0.11.dev0 (#13036)
add df0ffc3b50 [FIX,LOWERING] Add attrs from Relay Functions to PrimFuncs
in Metaschedule Lowering (#13034)
add 05d30f2502 [Docs] Update Release Process Docs -- Add Ver. Bump Step
(#13039)
add 4f975c7506 [Hexagon] [runtime] Support VTCM alignments of 128 or 2k
(#12999)
add cc235f88a4 [TIR] Update ir_comparator message to be more clear about
what is being compared (#13038)
add a752b74956 [Conda] Specify python version in build-environment (#13047)
add 48829756e1 [TIR] Fix the stride calculation in InjectRollingBuffer
(#13016)
add b8cfc4ca82 [CI] Update Compute Library to v22.08 (#13035)
add 256075566b [MetaSchedule] Allow skipping exact NDArray rewrite in
RemoveWeightLayoutRewriteBlock (#13052)
add b18f6425f9 [Hexagon] [runtime] Manage RPC and runtime buffers
separately (#13028)
add 90c666f860 [Relay][Pass] ConcretizeCastLikeRewrite for SimplifyExpr
(#12923)
add 61c9742ea7 [Hexagon] Enable multi input Async DMA; same queue / stage
(#13037)
add 46c16eed83 [microTVM][Windows] Enable building TVM on Windows with
USE_MICRO (#12851)
add f06896f38f [Relay] Fix simplifying consecutive casts when intermediate
type has fewer bits (#13056)
add 29a8f06066 [Arith] Optional rewriting and simplification into AND of
ORs (#12972)
add b389d4dac4 [Torch] Fix torch contrib issues (#13061)
add f232272e75 [TIR] Refactor NarrowDataType into DataTypeLegalizer
(#13049)
add 493458e552 [TE] Raise error for non-bijective transformation (#12926)
add 605876e638 [TEST] Fix the broken VNNI MetaSchedule test (#13067)
add 44c35dcd96 [TVMScript] Fix parsing int64 loop with optional loop start
(#13068)
add 342ffb91d6 [Hexagon]Register fast softmax schedule with default
schedule (#13083)
add 5eab64885a [ROOFLINE] Add support for different dtypes (#13003)
add 5ed94eefad [Node] Fix structural equal path tracing pointer usage
(#13082)
add 71f32ca4e8 [MetaSchedule][UX] Support Interactive Performance Table
Printing in Notebook (#13006)
add ec5c692148 Add include directory for OpenBLAS on RedHat (#13087)
add 5e862d4e41 [Frontend][PyTorch]Fix keywords to canonicalize scale and
zero point access for FX-quantized graphs (#13071)
add 9299a29fd3 [TVMScript] Evaluator, core parser, var table (#13088)
add cbca28da00 [MetaSchedule] Consolidate module hashing and equality
testing (#13050)
add e8ba1dc4cc [CI] Update Docker Image tag to 20221013-060115-61c9742ea
(#13078)
add 3f0d3f24b7 [AOT] Sanitize input/output name in runtime (#13046)
add b1c8c90384 [skip ci] Added label tags links to the wiki page in issue
templates to align with Issue Tracking RFC (#12988)
add 8d2dba1bbf [ci] Split the ci/scripts directory into ci/scripts/github
and ci/scripts/jenkins
add 923dcb21b6 Add GITHUB_SCRIPT_ROOT and JENKINS_SCRIPT_ROOT
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 (a4b0489fe9)
\
N -- N -- N refs/heads/separate-scripts (923dcb21b6)
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:
.github/ISSUE_TEMPLATE/bug-report.md | 11 +-
.github/ISSUE_TEMPLATE/ci-image.md | 29 -
.github/ISSUE_TEMPLATE/ci-problem.md | 13 +-
.github/ISSUE_TEMPLATE/documentation.md | 11 +-
.github/ISSUE_TEMPLATE/feature-tracking.md | 11 +-
.github/ISSUE_TEMPLATE/flaky-test.md | 10 +-
CMakeLists.txt | 10 +-
Jenkinsfile | 274 ++++---
KEYS | 59 ++
.../arduino/template_project/Makefile.template | 64 ++
.../template_project/microtvm_api_server.py | 271 ++++---
.../reference-vm/base-box/base_box_test.sh | 6 +-
.../zephyr/template_project/microtvm_api_server.py | 80 +--
apps/pt_tvmdsoop/tests/test_as_torch.py | 20 +-
ci/jenkins/Build.groovy.j2 | 328 ++++-----
ci/jenkins/Jenkinsfile.j2 | 20 +-
ci/jenkins/macros.j2 | 47 +-
cmake/config.cmake | 4 +
cmake/modules/Arduino.cmake | 1 +
cmake/modules/LibInfo.cmake | 1 +
cmake/modules/Logging.cmake | 5 +
cmake/modules/StandaloneCrt.cmake | 30 +-
cmake/modules/contrib/BLAS.cmake | 5 +
cmake/modules/contrib/CMSISNN.cmake | 4 +-
cmake/modules/contrib/EthosN.cmake | 3 +-
conda/build-environment.yaml | 1 +
conda/recipe/meta.yaml | 2 +-
.../ubuntu_download_arm_compute_lib_binaries.sh | 2 +-
docs/conf.py | 2 +-
docs/contribute/committer_guide.rst | 13 +
docs/contribute/release_process.rst | 7 +
docs/dev/how_to/debugging_tvm.rst | 2 +-
docs/dev/tutorial/codebase_walkthrough.rst | 5 +-
gallery/how_to/work_with_microtvm/micro_aot.py | 2 +-
.../how_to/work_with_microtvm/micro_autotune.py | 6 +-
.../work_with_microtvm/micro_reference_vm.py | 4 +-
gallery/how_to/work_with_microtvm/micro_tflite.py | 2 +-
gallery/how_to/work_with_microtvm/micro_train.py | 2 +-
include/tvm/arith/analyzer.h | 122 +++-
include/tvm/meta_schedule/database.h | 59 +-
include/tvm/meta_schedule/measure_callback.h | 7 +-
include/tvm/meta_schedule/mutator.h | 15 +-
include/tvm/meta_schedule/postproc.h | 9 +
include/tvm/meta_schedule/schedule_rule.h | 18 +-
include/tvm/meta_schedule/search_strategy.h | 32 +-
include/tvm/meta_schedule/space_generator.h | 53 +-
include/tvm/meta_schedule/task_scheduler.h | 282 ++++----
include/tvm/meta_schedule/tune_context.h | 87 +--
include/tvm/node/structural_equal.h | 42 ++
include/tvm/node/structural_hash.h | 38 +
include/tvm/runtime/c_runtime_api.h | 2 +-
include/tvm/runtime/crt/logging.h | 4 +
include/tvm/runtime/crt/platform.h | 4 +
include/tvm/runtime/crt/rpc_common/session.h | 13 +
include/tvm/runtime/crt/rpc_common/write_stream.h | 2 +
.../write_stream.h => name_transforms.h} | 33 +-
include/tvm/support/random_engine.h | 53 +-
include/tvm/target/compilation_config.h | 2 +-
include/tvm/target/target.h | 6 +-
include/tvm/target/target_kind.h | 15 +-
include/tvm/target/virtual_device.h | 10 +-
include/tvm/tir/index_map.h | 8 +
include/tvm/tir/schedule/schedule.h | 10 +
include/tvm/tir/stmt_functor.h | 51 ++
include/tvm/tir/transform.h | 10 +-
python/tvm/_ffi/libinfo.py | 2 +-
python/tvm/auto_scheduler/feature.py | 8 +-
python/tvm/contrib/hexagon/session.py | 2 +
python/tvm/contrib/torch/as_torch.py | 74 +-
python/tvm/contrib/torch/optimize_torch.py | 183 +++--
python/tvm/ir/expr.py | 7 +-
python/tvm/meta_schedule/__init__.py | 20 +-
python/tvm/meta_schedule/builder/builder.py | 44 +-
python/tvm/meta_schedule/builder/local_builder.py | 8 +-
python/tvm/meta_schedule/cost_model/cost_model.py | 46 +-
python/tvm/meta_schedule/cost_model/mlp_model.py | 6 +-
.../tvm/meta_schedule/cost_model/random_model.py | 3 +-
python/tvm/meta_schedule/cost_model/xgb_model.py | 362 +++++-----
python/tvm/meta_schedule/database/database.py | 89 ++-
python/tvm/meta_schedule/database/json_database.py | 8 +-
.../tvm/meta_schedule/database/memory_database.py | 18 +-
.../meta_schedule/database/schedule_fn_database.py | 15 +-
python/tvm/meta_schedule/default_config.py | 401 -----------
.../feature_extractor/feature_extractor.py | 26 +-
.../feature_extractor/random_feature_extractor.py | 8 +-
python/tvm/meta_schedule/logging.py | 263 +++++++
.../tvm/meta_schedule/measure_callback/__init__.py | 7 +-
.../measure_callback/echo_statistics.py | 30 -
.../measure_callback/measure_callback.py | 16 +-
python/tvm/meta_schedule/mutator/mutator.py | 44 +-
python/tvm/meta_schedule/postproc/postproc.py | 36 +-
python/tvm/meta_schedule/profiler.py | 4 -
python/tvm/meta_schedule/relay_integration.py | 374 ++++++++--
python/tvm/meta_schedule/runner/local_runner.py | 4 +-
python/tvm/meta_schedule/runner/rpc_runner.py | 4 +-
python/tvm/meta_schedule/runner/runner.py | 37 +-
.../meta_schedule/schedule_rule/schedule_rule.py | 35 +
.../search_strategy/evolutionary_search.py | 10 -
.../meta_schedule/search_strategy/replay_func.py | 11 +-
.../meta_schedule/search_strategy/replay_trace.py | 12 +-
.../search_strategy/search_strategy.py | 86 ++-
.../space_generator/post_order_apply.py | 26 +-
.../meta_schedule/space_generator/schedule_fn.py | 20 +-
.../space_generator/space_generator.py | 130 +++-
.../space_generator/space_generator_union.py | 20 +-
.../meta_schedule/task_scheduler/gradient_based.py | 49 +-
.../meta_schedule/task_scheduler/round_robin.py | 77 +-
.../meta_schedule/task_scheduler/task_scheduler.py | 276 +++----
.../meta_schedule/testing/dataset_extract_tasks.py | 4 +-
.../testing/dataset_sample_candidates.py | 17 +-
python/tvm/meta_schedule/testing/relay_workload.py | 52 +-
python/tvm/meta_schedule/testing/schedule_rule.py | 36 -
.../tvm/meta_schedule/testing/space_generation.py | 49 +-
python/tvm/meta_schedule/testing/tlcbench.py | 5 +-
.../testing/torchbench}/__init__.py | 0
python/tvm/meta_schedule/testing/torchbench/run.py | 584 +++++++++++++++
.../tvm/meta_schedule/testing/torchbench/utils.py | 103 +++
python/tvm/meta_schedule/testing/tune_onnx.py | 52 +-
python/tvm/meta_schedule/testing/tune_relay.py | 44 +-
python/tvm/meta_schedule/testing/tune_te.py | 41 +-
python/tvm/meta_schedule/tir_integration.py | 155 ++++
python/tvm/meta_schedule/tune.py | 660 ++---------------
python/tvm/meta_schedule/tune_context.py | 201 ++----
python/tvm/meta_schedule/utils.py | 175 ++---
python/tvm/micro/model_library_format.py | 32 +-
python/tvm/micro/project_api/server.py | 84 +++
python/tvm/micro/testing/evaluation.py | 4 +-
python/tvm/relay/backend/name_transforms.py | 11 -
python/tvm/relay/build_module.py | 4 +-
python/tvm/relay/collage/collage.py | 2 +-
python/tvm/relay/frontend/qnn_torch.py | 4 +-
python/tvm/relay/frontend/tflite.py | 2 +-
python/tvm/relay/op/contrib/cmsisnn.py | 47 +-
python/tvm/relay/op/op_attrs.py | 60 ++
python/tvm/relay/op/strategy/arm_cpu.py | 67 +-
python/tvm/relay/op/strategy/hexagon.py | 53 +-
.../__init__.py => runtime/name_transforms.py} | 20 +-
.../tvm/script/_parser/__init__.py | 11 +-
.../tvm/script/_parser/_core.py | 12 +-
.../tvm/script/_parser/core/__init__.py | 9 +-
python/tvm/script/_parser/core/diagnostics.py | 241 +++++++
python/tvm/script/_parser/core/dispatch.py | 156 ++++
python/tvm/script/_parser/core/doc.py | 462 ++++++++++++
.../script/{printer => _parser/core}/doc_core.py | 0
.../dense.py => script/_parser/core/entry.py} | 34 +-
python/tvm/script/_parser/core/evaluator.py | 509 +++++++++++++
python/tvm/script/_parser/core/parser.py | 647 +++++++++++++++++
python/tvm/script/_parser/core/utils.py | 61 ++
python/tvm/script/tir/__init__.py | 1 +
python/tvm/script/tir/scope_handler.py | 42 +-
python/tvm/script/tir/ty.py | 12 +-
python/tvm/target/target.py | 4 +
python/tvm/testing/utils.py | 172 +++--
python/tvm/tir/function.py | 16 +
python/tvm/tir/schedule/schedule.py | 89 +++
python/tvm/tir/tensor_intrin/cuda.py | 7 +-
python/tvm/tir/transform/transform.py | 16 +-
python/tvm/topi/arm_cpu/conv2d.py | 16 +
python/tvm/topi/arm_cpu/depthwise_conv2d.py | 19 +-
.../arm_cpu/mprofile/dsp/micro_kernel/tensordot.py | 155 ++++
.../topi/arm_cpu/mprofile/dsp/tensordot_conv2ds.py | 271 +++++++
python/tvm/topi/generic/conv2d.py | 11 +-
python/tvm/topi/hexagon/__init__.py | 2 +
python/tvm/topi/hexagon/conv2d.py | 49 +-
python/tvm/topi/hexagon/conv2d_alter_op.py | 111 +++
python/tvm/topi/hexagon/dense.py | 73 +-
python/tvm/topi/hexagon/dense_alter_op.py | 147 ++++
python/tvm/topi/hexagon/injective.py | 3 +-
python/tvm/topi/hexagon/tensor_intrin.py | 86 +++
python/tvm/topi/utils.py | 30 +-
python/tvm/utils/roofline/__init__.py | 64 +-
python/tvm/utils/roofline/cuda.py | 167 ++++-
python/tvm/utils/roofline/registry.py | 46 +-
python/tvm/utils/roofline/x86.py | 261 ++++---
src/arith/analyzer.cc | 3 +
src/arith/canonical_simplify.cc | 10 +-
src/arith/conjunctive_normal_form.cc | 430 +++++++++++
.../arith/conjunctive_normal_form.h | 39 +-
src/arith/iter_affine_map.cc | 4 +-
src/arith/rewrite_simplify.cc | 90 ++-
src/arith/rewrite_simplify.h | 32 +-
src/arith/transitive_comparison_analyzer.cc | 791 +++++++++++++++++++++
src/auto_scheduler/feature.cc | 7 +-
src/auto_scheduler/search_policy/utils.h | 15 +-
src/auto_scheduler/search_task.cc | 4 +-
src/driver/driver_api.cc | 9 +-
src/ir/expr.cc | 3 +
src/meta_schedule/arg_info.cc | 3 +-
src/meta_schedule/database/database.cc | 20 +-
src/meta_schedule/database/json_database.cc | 22 +-
src/meta_schedule/database/memory_database.cc | 13 +-
src/meta_schedule/database/schedule_fn_database.cc | 7 +-
.../feature_extractor/per_store_feature.cc | 27 +-
.../measure_callback/add_to_database.cc | 6 +-
.../measure_callback/echo_statistics.cc | 134 ----
.../measure_callback/measure_callback.cc | 10 +
.../measure_callback/update_cost_model.cc | 11 +-
.../meta_schedule/module_equality.cc | 41 +-
src/meta_schedule/module_equality.h | 73 ++
src/meta_schedule/mutator/mutator.cc | 30 +
src/meta_schedule/postproc/postproc.cc | 47 ++
.../postproc/rewrite_cooperative_fetch.cc | 2 +-
src/meta_schedule/postproc/rewrite_layout.cc | 56 +-
.../schedule_rule/cross_thread_reduction.cc | 4 +-
.../schedule_rule/multi_level_tiling.cc | 4 +-
.../schedule_rule/multi_level_tiling.h | 2 +-
.../multi_level_tiling_tensor_core.cc | 28 +-
.../multi_level_tiling_wide_vector.cc | 6 +
.../multi_level_tiling_with_intrin.cc | 6 +-
src/meta_schedule/schedule_rule/schedule_rule.cc | 154 ++++
.../search_strategy/evolutionary_search.cc | 174 ++---
src/meta_schedule/search_strategy/replay_func.cc | 98 +--
src/meta_schedule/search_strategy/replay_trace.cc | 97 +--
.../search_strategy/search_strategy.cc | 5 +-
.../space_generator/post_order_apply.cc | 35 +-
src/meta_schedule/space_generator/schedule_fn.cc | 11 +-
.../space_generator/space_generator.cc | 96 +++
.../space_generator/space_generator_union.cc | 16 +-
src/meta_schedule/task_scheduler/gradient_based.cc | 219 ++----
src/meta_schedule/task_scheduler/round_robin.cc | 23 +-
src/meta_schedule/task_scheduler/task_scheduler.cc | 408 +++++++----
src/meta_schedule/tune_context.cc | 180 +----
src/meta_schedule/utils.h | 79 +-
src/node/structural_equal.cc | 105 ++-
src/node/structural_hash.cc | 64 +-
src/printer/relay_text_printer.cc | 9 +-
src/relay/backend/aot/aot_lower_main.cc | 5 +-
src/relay/backend/aot_executor_codegen.cc | 3 +-
src/relay/backend/build_module.cc | 4 +-
src/relay/backend/contrib/cmsisnn/buffer_size.cc | 80 ++-
src/relay/backend/contrib/cmsisnn/buffer_size.h | 36 +-
.../backend/contrib/cmsisnn/generate_constants.cc | 20 +-
src/relay/backend/contrib/cmsisnn/relay_to_tir.cc | 52 +-
.../contrib/cmsisnn/scalar_to_tensor_constant.cc | 6 +-
.../backend/contrib/cmsisnn/tir_to_runtime.cc | 4 +-
src/relay/backend/contrib/constant_transforms.cc | 58 ++
src/relay/backend/contrib/constant_transforms.h | 59 ++
src/relay/backend/contrib/ethosn/codegen.cc | 4 +-
.../backend/contrib/ethosn/convert_equivalent.cc | 1 +
src/relay/backend/contrib/ethosn/ethosn_api.cc | 49 +-
src/relay/backend/contrib/ethosn/ethosn_api.h | 13 +-
src/relay/backend/contrib/uma/targets.cc | 2 +-
src/relay/backend/interpreter.cc | 6 +-
src/relay/backend/name_transforms.cc | 12 +-
src/relay/backend/name_transforms.h | 7 -
src/relay/backend/task_extraction.cc | 34 +-
src/relay/backend/te_compiler.cc | 2 +-
src/relay/backend/te_compiler_cache.cc | 167 ++++-
src/relay/backend/utils.h | 7 +
src/relay/backend/vm/compiler.cc | 6 +-
src/relay/ir/dataflow_matcher.cc | 5 +
.../transforms/meta_schedule_layout_rewrite.cc | 11 +-
src/relay/transforms/partition_graph.cc | 5 +-
src/relay/transforms/simplify_expr.cc | 33 +-
src/runtime/aot_executor/aot_executor.cc | 11 +-
src/runtime/contrib/ethosn/ethosn_device.cc | 64 +-
src/runtime/hexagon/hexagon_buffer.cc | 47 +-
src/runtime/hexagon/hexagon_buffer_manager.h | 13 +
src/runtime/hexagon/hexagon_device_api.cc | 55 +-
src/runtime/hexagon/hexagon_device_api.h | 55 +-
src/runtime/hexagon/hexagon_user_dma.cc | 24 +-
src/runtime/hexagon/hexagon_user_dma.h | 15 +-
src/runtime/hexagon/hexagon_vtcm_pool.cc | 160 +++++
src/runtime/hexagon/hexagon_vtcm_pool.h | 100 +++
src/runtime/hexagon/ring_buffer.h | 41 +-
src/runtime/hexagon/rpc/android/session.cc | 2 +-
src/runtime/hexagon/rpc/hexagon/rpc_server.cc | 32 +-
src/runtime/hexagon/rpc/simulator/session.cc | 1 -
src/runtime/logging.cc | 29 +
src/runtime/micro/micro_session.cc | 9 +
.../runtime/name_transforms.cc | 38 +-
src/runtime/rpc/rpc_endpoint.cc | 5 +
src/runtime/rpc/rpc_endpoint.h | 18 +-
src/runtime/rpc/rpc_module.cc | 4 +
src/runtime/rpc/rpc_session.h | 5 +
src/runtime/vulkan/vulkan_device.h | 2 +-
src/support/libinfo.cc | 5 +
src/target/compilation_config.cc | 23 +-
src/target/llvm/codegen_llvm.cc | 2 +-
src/target/llvm/intrin_rule_hexagon.cc | 10 +-
src/target/source/codegen_opencl.cc | 13 +-
src/target/source/codegen_opencl.h | 1 +
src/target/source/interface_c.cc | 1 +
src/target/source/source_module.cc | 9 +-
src/target/spirv/spirv_support.cc | 2 +-
src/target/target.cc | 14 +-
src/target/virtual_device.cc | 8 +-
src/te/operation/create_primfunc.cc | 24 +-
src/tir/analysis/verify_memory.cc | 2 +-
src/tir/ir/data_type_rewriter.cc | 195 +++++
src/tir/ir/index_map.cc | 135 +++-
src/tir/ir/stmt.cc | 2 +-
src/tir/op/op.cc | 2 +-
src/tir/schedule/analysis.h | 12 +
src/tir/schedule/analysis/analysis.cc | 9 +
src/tir/schedule/concrete_schedule.cc | 15 +-
src/tir/schedule/concrete_schedule.h | 2 +
src/tir/schedule/ir_comparator.cc | 8 +-
src/tir/schedule/primitive.h | 14 +-
src/tir/schedule/primitive/cache_read_write.cc | 282 +++++++-
.../schedule/primitive/layout_transformation.cc | 48 +-
src/tir/schedule/schedule.cc | 2 +
src/tir/schedule/traced_schedule.cc | 16 +
src/tir/schedule/traced_schedule.h | 2 +
src/tir/schedule/transform.cc | 25 +-
src/tir/schedule/transform.h | 9 +-
src/tir/transforms/inject_rolling_buffer.cc | 6 +-
src/tir/transforms/inject_software_pipeline.cc | 67 +-
src/tir/transforms/lower_async_dma.cc | 7 +-
src/tir/transforms/make_packed_api.cc | 2 +-
src/tir/transforms/make_unpacked_api.cc | 2 +-
src/tir/transforms/narrow_datatype.cc | 154 +---
.../remove_weight_layout_rewrite_block.cc | 213 +++++-
src/tir/transforms/simplify.cc | 43 +-
src/tir/transforms/texture_flatten.cc | 3 +-
.../hexagon/hexagon_device_api_tests.cc | 39 +-
.../cpp-runtime/hexagon/hexagon_user_dma_tests.cc | 104 ++-
.../cpp-runtime/hexagon/hexagon_vtcm_pool_tests.cc | 177 +++++
tests/cpp-runtime/hexagon/ring_buffer_tests.cc | 30 +-
tests/cpp/data_type_rewriter_test.cc | 140 ++++
tests/cpp/name_transforms_test.cc | 1 +
.../backend/contrib/cmsisnn/buffer_size_test.cc | 86 ++-
tests/lint/check_file_type.py | 2 +
tests/micro/arduino/README.md | 4 +-
tests/micro/arduino/test_arduino_rpc_server.py | 2 +-
tests/micro/arduino/test_utils.py | 2 +-
tests/micro/common/test_tvmc.py | 14 +-
.../test_arduino_microtvm_api_server.py | 26 +-
tests/micro/project_api/test_project_api.py | 92 +++
tests/micro/zephyr/README.md | 6 +-
tests/micro/zephyr/test_utils.py | 2 +-
tests/micro/zephyr/test_zephyr.py | 6 +-
tests/micro/zephyr/test_zephyr_aot_exec.py | 2 +-
tests/python/contrib/test_cmsisnn/test_conv2d.py | 142 ++--
tests/python/contrib/test_cmsisnn/utils.py | 65 +-
.../contrib/test_ethosn/test_fullyconnected.py | 56 +-
tests/python/contrib/test_ethosn/test_networks.py | 4 +-
tests/python/contrib/test_ethosu/test_codegen.py | 2 +-
.../test_ethosu/test_replace_depthwise_conv2d.py | 5 +-
.../python/contrib/test_hexagon/infrastructure.py | 8 +-
.../test_hexagon/test_2d_physical_buffers.py | 16 +-
tests/python/contrib/test_hexagon/test_autotvm.py | 11 +-
.../test_hexagon/test_benchmark_elemwise_add.py | 8 +-
.../test_hexagon/test_benchmark_maxpool2d.py | 12 +-
.../contrib/test_hexagon/test_cache_read_write.py | 5 +-
.../test_hexagon/test_fixed_point_multiply.py | 8 +-
tests/python/contrib/test_hexagon/test_launcher.py | 166 ++++-
.../contrib/test_hexagon/test_memory_alloc.py | 11 +-
.../contrib/test_hexagon/test_meta_schedule.py | 304 +++++++-
tests/python/contrib/test_hexagon/test_models.py | 9 +-
.../contrib/test_hexagon/test_parallel_hvx.py | 11 +-
.../test_hexagon/test_parallel_hvx_load_vtcm.py | 16 +-
.../contrib/test_hexagon/test_parallel_scalar.py | 11 +-
tests/python/contrib/test_hexagon/test_sigmoid.py | 113 +++
.../test_hexagon/test_software_pipeline_async.py | 187 ++++-
.../contrib/test_hexagon/test_thread_pool.py | 10 +-
.../contrib/test_hexagon/test_vtcm_bandwidth.py | 12 +-
.../topi/test_add_subtract_multiply.py | 10 +-
.../contrib/test_hexagon/topi/test_argmax_slice.py | 6 +-
.../test_hexagon/topi/test_avg_pool2d_slice.py | 12 +-
.../contrib/test_hexagon/topi/test_batch_matmul.py | 11 +-
.../contrib/test_hexagon/topi/test_cast_slice.py | 14 +-
.../python/contrib/test_hexagon/topi/test_clip.py | 9 +-
.../test_hexagon/topi/test_conv2d_fp16_intrin.py | 9 +-
.../contrib/test_hexagon/topi/test_conv2d_nchw.py | 6 +-
.../contrib/test_hexagon/topi/test_conv2d_nhwc.py | 10 +-
.../contrib/test_hexagon/topi/test_conv2d_slice.py | 8 +-
.../test_hexagon/topi/test_conv2d_transpose.py | 11 +-
.../python/contrib/test_hexagon/topi/test_dense.py | 9 +-
.../test_hexagon/topi/test_depth_to_space.py | 7 +-
.../test_hexagon/topi/test_depthwise_conv2d.py | 7 +-
.../test_hexagon/topi/test_dequantize_slice.py | 11 +-
.../test_hexagon/topi/test_dwconv2d_slice.py | 8 +-
.../test_hexagon/topi/test_max_pool2d_slice.py | 9 +-
tests/python/contrib/test_hexagon/topi/test_pad.py | 7 +-
.../contrib/test_hexagon/topi/test_pooling.py | 12 +-
.../contrib/test_hexagon/topi/test_quantize.py | 11 +-
.../contrib/test_hexagon/topi/test_reduce.py | 8 +-
.../contrib/test_hexagon/topi/test_relu_slice.py | 9 +-
.../contrib/test_hexagon/topi/test_reshape.py | 10 +-
.../contrib/test_hexagon/topi/test_resize2d.py | 5 +-
.../contrib/test_hexagon/topi/test_softmax.py | 9 +-
.../test_hexagon/topi/test_softmax_slice.py | 3 +-
.../contrib/test_hexagon/topi/test_tanh_slice.py | 6 +-
...le_auto_tensorize.py => test_auto_tensorize.py} | 207 +++---
.../{test_tuning.py => test_legacy_tuning.py} | 0
tests/python/integration/test_tuning.py | 433 ++---------
tests/python/relay/aot/test_cpp_aot.py | 41 +-
tests/python/relay/aot/test_crt_aot_usmp.py | 2 +-
.../relay/collage/demo_collage_partitioner.py | 10 +-
.../relay/strategy/arm_cpu/test_conv2d_nhwc.py | 36 +-
.../strategy/arm_cpu/test_depthwise_conv2d.py | 27 +
tests/python/relay/test_dataflow_pattern.py | 30 +
tests/python/relay/test_name_transforms.py | 4 +-
tests/python/relay/test_op_level1.py | 2 +-
tests/python/relay/test_op_level10.py | 2 +-
tests/python/relay/test_op_level2.py | 65 +-
tests/python/relay/test_pass_merge_composite.py | 14 +-
tests/python/relay/test_pass_simplify_expr.py | 18 +
.../python/unittest/test_auto_scheduler_feature.py | 11 +
tests/python/unittest/test_index_map.py | 68 +-
.../unittest/test_meta_schedule_cost_model.py | 36 +-
.../test_meta_schedule_custom_rule_winograd_cpu.py | 4 -
...test_meta_schedule_custom_rule_winograd_cuda.py | 3 -
...schedule_feature_extractor_per_store_feature.py | 62 +-
.../unittest/test_meta_schedule_integration.py | 343 ---------
.../test_meta_schedule_measure_callback.py | 18 +-
...eta_schedule_mutator_mutate_compute_location.py | 17 +-
.../test_meta_schedule_mutator_mutate_parallel.py | 17 +-
..._meta_schedule_mutator_mutate_thread_binding.py | 17 +-
.../test_meta_schedule_mutator_mutate_tile_size.py | 15 +-
.../test_meta_schedule_mutator_mutate_unroll.py | 17 +-
.../test_meta_schedule_post_order_apply.py | 50 +-
...meta_schedule_postproc_disallow_dynamic_loop.py | 21 +-
..._schedule_postproc_rewrite_cooperative_fetch.py | 22 +-
.../test_meta_schedule_postproc_rewrite_layout.py | 83 ++-
...ta_schedule_postproc_rewrite_reduction_block.py | 21 +-
...est_meta_schedule_postproc_rewrite_tensorize.py | 26 +-
...meta_schedule_postproc_rewrite_unbound_block.py | 26 +-
.../test_meta_schedule_postproc_verify_gpu_code.py | 23 +-
.../test_meta_schedule_relay_integration.py | 572 +++++++++++++++
...test_meta_schedule_schedule_rule_add_rfactor.py | 23 +-
.../test_meta_schedule_schedule_rule_auto_bind.py | 33 +-
...test_meta_schedule_schedule_rule_auto_inline.py | 43 +-
...chedule_schedule_rule_cross_thread_reduction.py | 51 +-
.../test_meta_schedule_schedule_rule_mlt.py | 61 +-
.../test_meta_schedule_schedule_rule_mlt_intrin.py | 19 +-
.../test_meta_schedule_schedule_rule_mlt_tc.py | 430 +++++------
...dule_schedule_rule_parallel_vectorize_unroll.py | 21 +-
...hedule_schedule_rule_random_compute_location.py | 13 +-
.../unittest/test_meta_schedule_search_strategy.py | 45 +-
.../unittest/test_meta_schedule_space_cpu.py | 106 +--
.../unittest/test_meta_schedule_space_cuda.py | 116 +--
.../unittest/test_meta_schedule_space_generator.py | 7 +-
.../unittest/test_meta_schedule_task_scheduler.py | 155 ++--
.../unittest/test_meta_schedule_tune_relay.py | 553 --------------
.../python/unittest/test_meta_schedule_tune_te.py | 52 --
.../python/unittest/test_meta_schedule_tune_tir.py | 115 ++-
.../test_meta_schedule_vnni_integration.py | 250 +++++++
tests/python/unittest/test_micro_project_api.py | 35 +
tests/python/unittest/test_roofline.py | 119 ++--
tests/python/unittest/test_runtime_profiling.py | 2 +-
tests/python/unittest/test_target_target.py | 4 +-
tests/python/unittest/test_te_create_primfunc.py | 2 +
tests/python/unittest/test_tir_constructor.py | 11 +-
.../python/unittest/test_tir_schedule_analysis.py | 5 +-
.../unittest/test_tir_schedule_cache_read_write.py | 58 ++
tests/python/unittest/test_tir_schedule_reindex.py | 104 ++-
tests/python/unittest/test_tir_schedule_trace.py | 1 -
.../unittest/test_tir_schedule_transform_layout.py | 73 +-
.../test_tir_transform_inject_rolling_buffer.py | 25 +-
.../test_tir_transform_inject_software_pipeline.py | 67 ++
.../python/unittest/test_tir_transform_simplify.py | 271 ++++++-
tests/python/unittest/test_transform_layout.py | 13 +
.../unittest/test_tvm_testing_before_after.py | 49 +-
.../unittest/test_tvmscript_parser_evaluator.py | 63 ++
.../unittest/test_tvmscript_parser_source.py | 86 +++
tests/python/unittest/test_tvmscript_roundtrip.py | 46 ++
.../python/unittest/test_tvmscript_syntax_sugar.py | 26 +
.../release/PRERELEASE_NOTES.md} | 23 +-
tests/scripts/request_hook/request_hook.py | 40 +-
tests/scripts/task_config_build_arm.sh | 1 +
tests/scripts/task_config_build_cortexm.sh | 1 +
tests/scripts/task_config_build_cpu.sh | 1 +
tests/scripts/task_config_build_gpu.sh | 1 +
tests/scripts/task_config_build_gpu_other.sh | 1 +
tests/scripts/task_config_build_hexagon.sh | 1 +
tests/scripts/task_config_build_i386.sh | 1 +
tests/scripts/task_python_hexagon.sh | 2 +-
tests/scripts/task_python_microtvm.sh | 4 +-
version.py | 2 +-
web/package.json | 4 +-
472 files changed, 18162 insertions(+), 8007 deletions(-)
delete mode 100644 .github/ISSUE_TEMPLATE/ci-image.md
create mode 100644 apps/microtvm/arduino/template_project/Makefile.template
copy include/tvm/runtime/{crt/rpc_common/write_stream.h => name_transforms.h}
(58%)
delete mode 100644 python/tvm/meta_schedule/default_config.py
create mode 100644 python/tvm/meta_schedule/logging.py
delete mode 100644 python/tvm/meta_schedule/measure_callback/echo_statistics.py
delete mode 100644 python/tvm/meta_schedule/testing/schedule_rule.py
copy python/tvm/{contrib/cuda_graph =>
meta_schedule/testing/torchbench}/__init__.py (100%)
create mode 100644 python/tvm/meta_schedule/testing/torchbench/run.py
create mode 100644 python/tvm/meta_schedule/testing/torchbench/utils.py
create mode 100644 python/tvm/meta_schedule/tir_integration.py
copy python/tvm/{meta_schedule/measure_callback/__init__.py =>
runtime/name_transforms.py} (70%)
copy cmake/modules/contrib/CMSISNN.cmake =>
python/tvm/script/_parser/__init__.py (72%)
copy cmake/modules/contrib/CMSISNN.cmake => python/tvm/script/_parser/_core.py
(72%)
copy cmake/modules/contrib/CMSISNN.cmake =>
python/tvm/script/_parser/core/__init__.py (74%)
create mode 100644 python/tvm/script/_parser/core/diagnostics.py
create mode 100644 python/tvm/script/_parser/core/dispatch.py
create mode 100644 python/tvm/script/_parser/core/doc.py
rename python/tvm/script/{printer => _parser/core}/doc_core.py (100%)
copy python/tvm/{topi/hexagon/dense.py => script/_parser/core/entry.py} (54%)
create mode 100644 python/tvm/script/_parser/core/evaluator.py
create mode 100644 python/tvm/script/_parser/core/parser.py
create mode 100644 python/tvm/script/_parser/core/utils.py
create mode 100644
python/tvm/topi/arm_cpu/mprofile/dsp/micro_kernel/tensordot.py
create mode 100644 python/tvm/topi/arm_cpu/mprofile/dsp/tensordot_conv2ds.py
create mode 100644 python/tvm/topi/hexagon/conv2d_alter_op.py
create mode 100644 python/tvm/topi/hexagon/dense_alter_op.py
create mode 100644 src/arith/conjunctive_normal_form.cc
copy include/tvm/runtime/crt/rpc_common/write_stream.h =>
src/arith/conjunctive_normal_form.h (56%)
create mode 100644 src/arith/transitive_comparison_analyzer.cc
delete mode 100644 src/meta_schedule/measure_callback/echo_statistics.cc
copy include/tvm/runtime/crt/rpc_common/write_stream.h =>
src/meta_schedule/module_equality.cc (55%)
create mode 100644 src/meta_schedule/module_equality.h
create mode 100644 src/relay/backend/contrib/constant_transforms.cc
create mode 100644 src/relay/backend/contrib/constant_transforms.h
create mode 100644 src/runtime/hexagon/hexagon_vtcm_pool.cc
create mode 100644 src/runtime/hexagon/hexagon_vtcm_pool.h
copy include/tvm/runtime/crt/rpc_common/write_stream.h =>
src/runtime/name_transforms.cc (58%)
create mode 100644 src/tir/ir/data_type_rewriter.cc
create mode 100644 tests/cpp-runtime/hexagon/hexagon_vtcm_pool_tests.cc
create mode 100644 tests/cpp/data_type_rewriter_test.cc
rename {apps/microtvm/arduino/template_project/tests =>
tests/micro/project_api}/test_arduino_microtvm_api_server.py (89%)
create mode 100644 tests/micro/project_api/test_project_api.py
mode change 100755 => 100644
tests/python/contrib/test_hexagon/test_2d_physical_buffers.py
create mode 100644 tests/python/contrib/test_hexagon/test_sigmoid.py
rename tests/python/integration/{test_meta_schedule_auto_tensorize.py =>
test_auto_tensorize.py} (70%)
copy tests/python/integration/{test_tuning.py => test_legacy_tuning.py} (100%)
delete mode 100644 tests/python/unittest/test_meta_schedule_integration.py
create mode 100644
tests/python/unittest/test_meta_schedule_relay_integration.py
delete mode 100644 tests/python/unittest/test_meta_schedule_tune_relay.py
delete mode 100644 tests/python/unittest/test_meta_schedule_tune_te.py
create mode 100644 tests/python/unittest/test_meta_schedule_vnni_integration.py
create mode 100644 tests/python/unittest/test_tvmscript_parser_evaluator.py
create mode 100644 tests/python/unittest/test_tvmscript_parser_source.py
copy tests/{micro/arduino/README.md => scripts/release/PRERELEASE_NOTES.md}
(60%)
