You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by gi...@apache.org on 2022/07/15 00:00:37 UTC

[tvm] branch nightly-docker-update updated (0cf90d296f -> 27a40280e3)

This is an automated email from the ASF dual-hosted git repository.

github-bot pushed a change to branch nightly-docker-update
in repository https://gitbox.apache.org/repos/asf/tvm.git


 discard 0cf90d296f [ci][docker] Nightly Docker image update
     add a571bfbbca [TOPI] Allow conv definition to have custom kernel layout (#11936)
     add 3f9f41a103 [Relay] Add RecoverVirtualDeviceMap helper (#12085)
     add f6f90569bc Add tensorflow Einsum op converter (#12064)
     add de3c0f4b30 [WIN] export void Configure(...) symbol for Windows (#12091)
     add b9fa576ab3 [BugFix] Use shape dtype on ArgReduce to determine return type (#12083)
     add 09b607e354 [ci] Override Request in pytests (#11974)
     add 811373ddbb [ci] Add a manual retry for conda setup (#12058)
     add 484f602327 [ci] Re-run failed tests on failure (#12055)
     add 2fee86e43b [Releay] Fix on_device call for explicit virtual_device (#12088)
     add 9a16503443 adding Alexey (#12090)
     add e084791852 [MetaSchedule] Add MultiLevelTilingTensorCore rule for auto-tensorization on CUDA (#12059)
     add 7661ba8910 [Collage] CollagePartition pass (#12086)
     add 27a40280e3 [ci][docker] Nightly Docker image update

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   (0cf90d296f)
            \
             N -- N -- N   refs/heads/nightly-docker-update (27a40280e3)

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/actions/setup/action.yml                   |  12 +
 CONTRIBUTORS.md                                    |   1 +
 Jenkinsfile                                        |  16 +-
 ci/jenkins/Jenkinsfile.j2                          |  16 +-
 conftest.py                                        |  15 +-
 docker/install/ubuntu_install_python_package.sh    |   1 +
 include/tvm/meta_schedule/schedule_rule.h          |  24 +
 include/tvm/runtime/threading_backend.h            |   4 +-
 include/tvm/tir/stmt.h                             |   4 +
 python/tvm/meta_schedule/schedule_rule/__init__.py |   7 +-
 .../schedule_rule/multi_level_tiling.py            |  54 +-
 python/tvm/meta_schedule/testing/schedule_rule.py  |  34 ++
 python/tvm/relay/__init__.py                       |   1 +
 .../{_ffi/_ctypes => relay/collage}/__init__.py    |   9 +-
 python/tvm/{arith => relay/collage}/_ffi_api.py    |   4 +-
 python/tvm/relay/collage/collage.py                | 146 +++++
 python/tvm/relay/frontend/tensorflow_ops.py        |  10 +
 python/tvm/relay/op/annotation/annotation.py       |   2 +
 python/tvm/relay/op/strategy/arm_cpu.py            |  10 +-
 python/tvm/relay/op/strategy/cuda.py               |  11 +-
 python/tvm/relay/op/strategy/generic.py            |  15 +-
 python/tvm/relay/op/strategy/hls.py                |   2 +-
 python/tvm/relay/op/strategy/intel_graphics.py     |   8 +-
 python/tvm/relay/op/strategy/rocm.py               |   2 +-
 python/tvm/relay/op/strategy/x86.py                |  10 +-
 python/tvm/relay/transform/transform.py            |  23 +
 python/tvm/tir/tensor_intrin/cuda.py               |  73 ++-
 python/tvm/topi/nn/conv1d.py                       |  35 +-
 python/tvm/topi/nn/conv2d.py                       | 117 ++--
 python/tvm/topi/nn/conv3d.py                       |   3 +-
 .../postproc/rewrite_reduction_block.cc            |  17 +
 src/meta_schedule/postproc/rewrite_tensorize.cc    |   6 +-
 src/meta_schedule/schedule_rule/auto_inline.cc     |   3 +-
 .../schedule_rule/multi_level_tiling.cc            |  28 +-
 .../schedule_rule/multi_level_tiling.h             |  10 +-
 .../multi_level_tiling_tensor_core.cc              | 393 +++++++++++++
 src/relay/collage/candidate_partition.cc           |   3 +-
 src/relay/collage/candidate_partition_index.cc     | 150 +++++
 src/relay/collage/candidate_partition_index.h      | 102 ++++
 src/relay/collage/collage_partitioner.cc           | 352 ++++++++++++
 .../collage/{cost.cc => collage_partitioner.h}     |  33 +-
 src/relay/collage/cost_estimator.cc                |   8 +-
 src/relay/collage/cost_estimator.h                 |   4 +-
 src/relay/collage/gather_partition_specs.cc        | 214 +++++++
 src/relay/collage/gather_partition_specs.h         |  71 +++
 src/relay/collage/priority_queue.h                 |  72 +++
 src/relay/collage/utils.cc                         |   2 +-
 src/relay/op/tensor/reduce.cc                      |   2 +-
 src/relay/transforms/device_aware_visitors.cc      |  35 ++
 src/relay/transforms/device_aware_visitors.h       |   8 +
 src/runtime/thread_pool.cc                         |   4 +-
 src/runtime/vm/vm.cc                               |   8 +-
 src/tir/schedule/analysis.h                        |  18 +-
 src/tir/schedule/analysis/analysis.cc              |  33 +-
 src/tir/schedule/primitive/block_annotate.cc       |   5 +-
 src/tir/schedule/primitive/cache_read_write.cc     |   7 +-
 .../schedule/primitive/layout_transformation.cc    |   7 +-
 src/tir/schedule/transform.cc                      |   6 +-
 tests/python/frontend/tensorflow/test_forward.py   |  33 ++
 tests/python/integration/test_winograd_nnpack.py   |   2 +-
 tests/python/relay/test_pass_collage_partition.py  | 617 +++++++++++++++++++++
 tests/python/relay/test_pass_plan_devices.py       |  78 +++
 tests/python/relay/test_type_infer.py              |  38 ++
 tests/python/topi/python/test_topi_conv2d_nhwc.py  |  31 +-
 ...est_meta_schedule_postproc_rewrite_tensorize.py |   2 +-
 ...ta_schedule_schedule_rule_multi_level_tiling.py | 426 +++++++++++++-
 vta/python/vta/top/op.py                           |   2 +-
 67 files changed, 3308 insertions(+), 191 deletions(-)
 copy python/tvm/{_ffi/_ctypes => relay/collage}/__init__.py (83%)
 copy python/tvm/{arith => relay/collage}/_ffi_api.py (89%)
 create mode 100644 python/tvm/relay/collage/collage.py
 create mode 100644 src/meta_schedule/schedule_rule/multi_level_tiling_tensor_core.cc
 create mode 100644 src/relay/collage/candidate_partition_index.cc
 create mode 100644 src/relay/collage/candidate_partition_index.h
 create mode 100644 src/relay/collage/collage_partitioner.cc
 copy src/relay/collage/{cost.cc => collage_partitioner.h} (54%)
 create mode 100644 src/relay/collage/gather_partition_specs.cc
 create mode 100644 src/relay/collage/gather_partition_specs.h
 create mode 100644 src/relay/collage/priority_queue.h
 create mode 100644 tests/python/relay/test_pass_collage_partition.py