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 2022/09/29 00:03:31 UTC
[tvm] branch nightly-docker-update updated (60752ba428 -> 9ed93be1f0)
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 60752ba428 [ci][docker] Nightly Docker image update
add f64e933246 [LLVM] Emit fp16/fp32 builtins directly into target module (#12877)
add b61f633e10 [TVM PyTorch Integration] optimized_torch & as_torch how-to guide (#12318)
add 7a4c10c44a [TIR][Transform] Remove num_unpacked_args from MakePackedAPI (#12892)
add 7dbc68d108 [ONNX] Fix test_roi_align failure (#12906)
add 77d8eef514 [Runtime][Bugfix] Added type-checking for Array::insert (#12691)
add 9a673faa74 [ci] Initialize git during deploys (#12909)
add 332b1469b7 [Hexagon] depth_to_space slice op (#12669)
add 5a807e27c0 [Hexagon] [runtime] Add thread manager to resource management (#12905)
add 82e6fc41f8 [microTVM] add the option to open a saved micro project for debugging (#12495)
add a07a46ed19 [TIR] add unit-tests for upcoming primfunc-slicing (#12794)
add bec9f16d42 [TIR][Transform] Clear buffer_map during MakeUnpackedAPI (#12891)
add c89a8baeeb [usmp] Also remap VarNode to USMP-allocated buffer (#12880)
add 178f82dc48 [TOPI] Implement Einsum with reduction axes (#12913)
add d1c9febeca [ETHOSN] Remove support for 22.05 version of the driver stack (#12770)
add 17e4644019 [TIR][MetaSchedule] Add regression test for layout_rewrite extent=1 (#12916)
add e3a6cb6a1b [microTVM] Generalize depthwise_conv2d schedule (#12856)
add 9d1fe6d8d1 [Target] Add Ampere GPUs CUDA tags (#12930)
add 8af43d3c11 [Hexagon] [runtime] Add user DMA to device API resource management (#12918)
add 9ed93be1f0 [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 (60752ba428)
\
N -- N -- N refs/heads/nightly-docker-update (9ed93be1f0)
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 | 34 +-
ci/jenkins/Deploy.groovy.j2 | 10 +-
ci/jenkins/Jenkinsfile.j2 | 20 +-
gallery/how_to/work_with_pytorch/using_as_torch.py | 159 ++++
.../work_with_pytorch/using_optimized_torch.py | 149 ++++
include/tvm/runtime/container/array.h | 5 +
include/tvm/tir/transform.h | 9 +-
include/tvm/topi/einsum.h | 889 +--------------------
python/tvm/contrib/torch/as_torch.py | 9 +-
python/tvm/contrib/torch/optimize_torch.py | 4 +-
python/tvm/micro/build.py | 26 +-
python/tvm/micro/session.py | 56 +-
python/tvm/micro/testing/evaluation.py | 39 +-
.../tvm/relay/backend/contrib/ethosu/tir/utils.py | 30 +
.../backend/contrib/ethosu/tir_to_cs_translator.py | 37 +-
python/tvm/relay/op/contrib/ethosn.py | 2 +-
python/tvm/relay/op/strategy/arm_cpu.py | 22 +-
python/tvm/tir/transform/transform.py | 39 +-
python/tvm/topi/arm_cpu/conv2d_alter_op.py | 38 +-
.../topi/arm_cpu/mprofile/dsp/depthwise_conv2d.py | 126 +--
.../arm_cpu/mprofile/dsp/micro_kernel/common.py | 15 +
.../dsp/micro_kernel/multi_channel_convolve.py | 210 +++++
.../dsp/micro_kernel/quad_channel_convolve.py | 180 -----
python/tvm/topi/hexagon/slice_ops/__init__.py | 1 +
.../tvm/topi/hexagon/slice_ops/depth_to_space.py | 43 +
src/driver/driver_api.cc | 2 +-
src/meta_schedule/postproc/rewrite_layout.cc | 2 +
src/relay/op/tensor/math.cc | 2 +-
src/runtime/builtin_fp16.cc | 3 -
src/runtime/hexagon/hexagon_device_api.cc | 4 +-
src/runtime/hexagon/hexagon_device_api.h | 44 +-
src/runtime/hexagon/hexagon_user_dma.cc | 12 +-
src/runtime/hexagon/hexagon_user_dma.h | 21 +-
src/target/llvm/codegen_llvm.cc | 227 ++++++
src/target/llvm/codegen_llvm.h | 8 +
src/target/tag.cc | 9 +
src/te/schedule/schedule_lang.cc | 2 +-
src/tir/transforms/make_packed_api.cc | 109 +--
src/tir/transforms/make_unpacked_api.cc | 7 +-
src/tir/usmp/analysis/extract_buffer_info.cc | 20 +-
.../convert_pool_allocations_to_offsets.cc | 10 +
src/topi/einsum.cc | 353 ++++++++
src/topi/transform.cc | 4 -
.../hexagon/hexagon_device_api_tests.cc | 20 +
.../cpp-runtime/hexagon/hexagon_user_dma_tests.cc | 48 +-
tests/python/contrib/test_ethosn/test_networks.py | 28 +-
tests/python/contrib/test_ethosn/test_split.py | 13 -
.../python/contrib/test_ethosn/test_topologies.py | 17 +-
.../test_hexagon/topi/test_depth_to_space.py | 136 ++++
tests/python/frontend/onnx/test_forward.py | 1 +
.../strategy/arm_cpu/test_depthwise_conv2d.py | 43 +-
tests/python/topi/python/test_topi_einsum.py | 36 +-
.../test_meta_schedule_postproc_rewrite_layout.py | 156 ++--
tests/python/unittest/test_slice_tir.py | 216 +++++
tests/python/unittest/test_target_codegen_llvm.py | 7 +-
tests/python/unittest/test_target_codegen_x86.py | 74 +-
.../unittest/test_tir_transform_make_packed_api.py | 5 +-
...ransform_convert_pool_allocations_to_offsets.py | 93 +++
58 files changed, 2343 insertions(+), 1541 deletions(-)
create mode 100644 gallery/how_to/work_with_pytorch/using_as_torch.py
create mode 100644 gallery/how_to/work_with_pytorch/using_optimized_torch.py
create mode 100644 python/tvm/topi/arm_cpu/mprofile/dsp/micro_kernel/multi_channel_convolve.py
delete mode 100644 python/tvm/topi/arm_cpu/mprofile/dsp/micro_kernel/quad_channel_convolve.py
create mode 100644 python/tvm/topi/hexagon/slice_ops/depth_to_space.py
create mode 100644 src/topi/einsum.cc
create mode 100644 tests/python/contrib/test_hexagon/topi/test_depth_to_space.py
create mode 100644 tests/python/unittest/test_slice_tir.py