[GitHub] [tvm] juda commented on a diff in pull request #11911: TVM Vertical Integration with PyTorch

juda commented on code in PR #11911: URL: https://github.com/apache/tvm/pull/11911#discussion_r918542699 ## src/contrib/torch/pt_call_tvm/RuntimeModuleWrapper.cc: ## @@ -0,0 +1,262 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor

[tvm] branch last-successful updated (7baf4be2d4 -> fc419df32f)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch last-successful in repository https://gitbox.apache.org/repos/asf/tvm.git from 7baf4be2d4 [relay] Changed 'name' field to 'registry_name' for Executor and Runtime (#10466) add

[GitHub] [tvm] masahi commented on a diff in pull request #12059: [MetaSchedule] Add MultiLevelTilingTensorCore rule for auto-tensorization on CUDA

masahi commented on code in PR #12059: URL: https://github.com/apache/tvm/pull/12059#discussion_r918501599 ## python/tvm/meta_schedule/testing/schedule_rule.py: ## @@ -110,6 +112,38 @@ def multi_level_tiling(target: Target) -> ScheduleRule: raise

[GitHub] [tvm] sunggg commented on issue #11462: CI monitoring rotation schedule

sunggg commented on issue #11462: URL: https://github.com/apache/tvm/issues/11462#issuecomment-1181264493 07/11/2022: No CI failure -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific

[GitHub] [tvm] masahi commented on a diff in pull request #11911: TVM Vertical Integration with PyTorch

masahi commented on code in PR #11911: URL: https://github.com/apache/tvm/pull/11911#discussion_r918495614 ## src/contrib/torch/pt_call_tvm/RuntimeModuleWrapper.cc: ## @@ -0,0 +1,262 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor

[GitHub] [tvm] juda commented on a diff in pull request #11911: TVM Vertical Integration with PyTorch

juda commented on code in PR #11911: URL: https://github.com/apache/tvm/pull/11911#discussion_r918484968 ## python/tvm/contrib/torch/optimize_torch.py: ## @@ -0,0 +1,143 @@ +# pylint: disable=inconsistent-return-statements +#!/usr/bin/env python + +# Licensed to the Apache

[GitHub] [tvm] juda commented on a diff in pull request #11911: TVM Vertical Integration with PyTorch

juda commented on code in PR #11911: URL: https://github.com/apache/tvm/pull/11911#discussion_r918484571 ## python/tvm/contrib/torch/optimize_torch.py: ## @@ -0,0 +1,143 @@ +# pylint: disable=inconsistent-return-statements +#!/usr/bin/env python + +# Licensed to the Apache

[GitHub] [tvm] juda commented on a diff in pull request #11911: TVM Vertical Integration with PyTorch

juda commented on code in PR #11911: URL: https://github.com/apache/tvm/pull/11911#discussion_r918483831 ## src/contrib/torch/pt_call_tvm/RuntimeModuleWrapper.cc: ## @@ -0,0 +1,262 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor

[GitHub] [tvm] juda commented on a diff in pull request #11911: TVM Vertical Integration with PyTorch

juda commented on code in PR #11911: URL: https://github.com/apache/tvm/pull/11911#discussion_r918475519 ## python/tvm/contrib/torch/optimize_torch.py: ## @@ -0,0 +1,143 @@ +# pylint: disable=inconsistent-return-statements +#!/usr/bin/env python + +# Licensed to the Apache

[GitHub] [tvm] juda commented on a diff in pull request #11911: TVM Vertical Integration with PyTorch

juda commented on code in PR #11911: URL: https://github.com/apache/tvm/pull/11911#discussion_r918475388 ## python/tvm/contrib/torch/optimize_torch.py: ## @@ -0,0 +1,143 @@ +# pylint: disable=inconsistent-return-statements +#!/usr/bin/env python + +# Licensed to the Apache

[GitHub] [tvm] juda commented on a diff in pull request #11911: TVM Vertical Integration with PyTorch

juda commented on code in PR #11911: URL: https://github.com/apache/tvm/pull/11911#discussion_r918475091 ## python/tvm/contrib/torch/optimize_torch.py: ## @@ -0,0 +1,143 @@ +# pylint: disable=inconsistent-return-statements +#!/usr/bin/env python + +# Licensed to the Apache

[GitHub] [tvm] juda commented on a diff in pull request #11911: TVM Vertical Integration with PyTorch

juda commented on code in PR #11911: URL: https://github.com/apache/tvm/pull/11911#discussion_r918474760 ## python/tvm/contrib/torch/as_torch.py: ## @@ -0,0 +1,89 @@ +# pylint: disable=inconsistent-return-statements +#!/usr/bin/env python + +# Licensed to the Apache Software

[tvm] branch main updated (f3359d241f -> fc419df32f)

This is an automated email from the ASF dual-hosted git repository. masahi pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from f3359d241f [LLVM] Fix build errors in CodeGenCPU::AddDebugInformation (#12054) add fc419df32f [AOT][BUG] Only

[GitHub] [tvm] masahi merged pull request #12061: [AOT][BUG] Only include extra headers if the constants array is needed.

masahi merged PR #12061: URL: https://github.com/apache/tvm/pull/12061 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail:

[GitHub] [tvm] masahi closed issue #12049: [Bug] Build error with the current TVM main version

masahi closed issue #12049: [Bug] Build error with the current TVM main version URL: https://github.com/apache/tvm/issues/12049 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment.

[GitHub] [tvm] masahi commented on issue #12049: [Bug] Build error with the current TVM main version

masahi commented on issue #12049: URL: https://github.com/apache/tvm/issues/12049#issuecomment-1181216595 https://github.com/apache/tvm/pull/12054 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to

[GitHub] [tvm] github-actions[bot] commented on pull request #12055: [ci] Re-run failed tests on failure

github-actions[bot] commented on PR #12055: URL: https://github.com/apache/tvm/pull/12055#issuecomment-1181216548 Built docs for commit df4e21720a541a9ee9c93fd0945fee4bcade8197 can be found [here](https://pr-docs.tlcpack.ai/PR-12055/4/docs/index.html). -- This is an automated

[GitHub] [tvm] juda commented on a diff in pull request #11911: TVM Vertical Integration with PyTorch

juda commented on code in PR #11911: URL: https://github.com/apache/tvm/pull/11911#discussion_r918473912 ## apps/pt_tvmdsoop/tests/test_optimize_torch.py: ## @@ -0,0 +1,161 @@ +# pylint: disable=missing-class-docstring +#!/usr/bin/env python + +# Licensed to the Apache Software

[tvm] branch main updated (7baf4be2d4 -> f3359d241f)

This is an automated email from the ASF dual-hosted git repository. masahi pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from 7baf4be2d4 [relay] Changed 'name' field to 'registry_name' for Executor and Runtime (#10466) add f3359d241f [LLVM]

[GitHub] [tvm] masahi merged pull request #12054: [LLVM] Fix build errors in CodeGenCPU::AddDebugInformation

masahi merged PR #12054: URL: https://github.com/apache/tvm/pull/12054 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail:

[GitHub] [tvm] juda commented on a diff in pull request #11911: TVM Vertical Integration with PyTorch

juda commented on code in PR #11911: URL: https://github.com/apache/tvm/pull/11911#discussion_r918473322 ## apps/pt_tvmdsoop/tests/test_as_torch.py: ## @@ -0,0 +1,148 @@ +#!/usr/bin/env python + +# Licensed to the Apache Software Foundation (ASF) under one +# or more

[GitHub] [tvm] juda commented on a diff in pull request #11911: TVM Vertical Integration with PyTorch

juda commented on code in PR #11911: URL: https://github.com/apache/tvm/pull/11911#discussion_r918470369 ## src/contrib/torch/pt_call_tvm/RuntimeModuleWrapper.cc: ## @@ -0,0 +1,262 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor

[GitHub] [tvm] juda commented on a diff in pull request #11911: TVM Vertical Integration with PyTorch

juda commented on code in PR #11911: URL: https://github.com/apache/tvm/pull/11911#discussion_r918469708 ## src/contrib/torch/pt_call_tvm/RuntimeModuleWrapper.cc: ## @@ -0,0 +1,262 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor

[GitHub] [tvm] juda commented on a diff in pull request #11911: TVM Vertical Integration with PyTorch

juda commented on code in PR #11911: URL: https://github.com/apache/tvm/pull/11911#discussion_r918469535 ## src/contrib/torch/pt_call_tvm/RuntimeModuleWrapper.cc: ## @@ -0,0 +1,262 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor

[GitHub] [tvm] juda commented on a diff in pull request #11911: TVM Vertical Integration with PyTorch

juda commented on code in PR #11911: URL: https://github.com/apache/tvm/pull/11911#discussion_r918469338 ## src/contrib/torch/pt_call_tvm/RuntimeModuleWrapper.cc: ## @@ -0,0 +1,262 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor

[GitHub] [tvm] alanmacd commented on a diff in pull request #12063: [Pylint][microTVM] Applying Pylint rules to unittest folder

alanmacd commented on code in PR #12063: URL: https://github.com/apache/tvm/pull/12063#discussion_r918452702 ## tests/python/unittest/test_runtime_graph.py: ## @@ -14,21 +14,26 @@ # KIND, either express or implied. See the License for the # specific language governing

[GitHub] [tvm-rfcs] xqdan commented on pull request #79: [RFC] TVMScript Metaprogramming

xqdan commented on PR #79: URL: https://github.com/apache/tvm-rfcs/pull/79#issuecomment-1181184802 @yelite It's a great RFC，and this is what we need right now. the requirements we need: 1) For compute fusion. With TE compute, it's easy to concate TE computes with producer-comsuer

[GitHub] [tvm] alanmacd opened a new pull request, #12063: [Pylint][microTVM] Applying Pylint rules to unittest folder

alanmacd opened a new pull request, #12063: URL: https://github.com/apache/tvm/pull/12063 Apply Pylint to the following files: - tests/python/unittest/test_crt.py - tests/python/unittest/test_micro_model_library_format.py - tests/python/unittest/test_micro_project_api.py -

[GitHub] [tvm-rfcs] csullivan commented on pull request #77: [RFC] Buffer Layout Padding

csullivan commented on PR #77: URL: https://github.com/apache/tvm-rfcs/pull/77#issuecomment-1181142880 Thanks everyone for the very fruitful discussions! We indeed have a good path forward and are aligned on the principles that for the end to end optimization we will maintain function

[tvm] branch nightly-docker-update updated (1c0ee61453 -> 125446a862)

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 1c0ee61453 [ci][docker] Nightly Docker image update add 927620e20f [BYOC-DNNL] support more

[GitHub] [tvm] alanmacd commented on issue #11414: [Tracking Issue] pylint the tests!

alanmacd commented on issue #11414: URL: https://github.com/apache/tvm/issues/11414#issuecomment-1181110122 @Mousius Thanks, that works! It was slightly more complicated, but I think I figured it out.  `@pytest.mark.parametrize("enable_usmp,expect_exception", [(True, True),

[tvm] branch last-successful updated (d27f853ed5 -> 7baf4be2d4)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch last-successful in repository https://gitbox.apache.org/repos/asf/tvm.git from d27f853ed5 [ci][docker] fix the path of custom toolchain in ci_qemu for csinn2 (#11905) add

[tvm] branch last-successful updated (ae72e7e653 -> d27f853ed5)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch last-successful in repository https://gitbox.apache.org/repos/asf/tvm.git from ae72e7e653 Fix node.func to node.funcs on parser.py (#12053) add d27f853ed5 [ci][docker] fix the path

[GitHub] [tvm] junrushao1994 commented on a diff in pull request #11991: [release] Bump version numbers to v0.9.0

junrushao1994 commented on code in PR #11991: URL: https://github.com/apache/tvm/pull/11991#discussion_r918408783 ## python/setup.py: ## @@ -223,7 +223,7 @@ def long_description_contents(): setup( name="tvm", -version=__version__, +version="0.9.0", Review

[tvm] branch last-successful updated (b4c1cc02eb -> ae72e7e653)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch last-successful in repository https://gitbox.apache.org/repos/asf/tvm.git from b4c1cc02eb [CMSIS-NN][Perf] Converted Relay Conv2D into CMSIS-NN Depthwise (#12006) add 5be8e0a3de

[GitHub] [tvm] areusch commented on a diff in pull request #11991: [release] Bump version numbers to v0.9.0

areusch commented on code in PR #11991: URL: https://github.com/apache/tvm/pull/11991#discussion_r918401578 ## python/setup.py: ## @@ -223,7 +223,7 @@ def long_description_contents(): setup( name="tvm", -version=__version__, +version="0.9.0", Review Comment:

[GitHub] [tvm] areusch commented on pull request #12040: [docs][tvmc] Fix ResNet50 model URL in tvmc tutorials

areusch commented on PR #12040: URL: https://github.com/apache/tvm/pull/12040#issuecomment-1180948326 @leandron could you comment more about what's changing here? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the

[GitHub] [tvm-rfcs] areusch commented on pull request #84: [RFC] Name mangling in IRModules

areusch commented on PR #84: URL: https://github.com/apache/tvm-rfcs/pull/84#issuecomment-1180909624 i've put this RFC on the meeting agenda for this week's Community Meeting -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub

[GitHub] [tvm-rfcs] areusch commented on pull request #84: [RFC] Name mangling in IRModules

areusch commented on PR #84: URL: https://github.com/apache/tvm-rfcs/pull/84#issuecomment-1180909320 @tqchen i think one goal we should have with this RFC is to codify things like > When a function does not have global_symbol attribute, it means the LinkageType is private the

[GitHub] [tvm] areusch commented on a diff in pull request #12028: [WIP][Pylint] Making frontend tests pylint compliant

areusch commented on code in PR #12028: URL: https://github.com/apache/tvm/pull/12028#discussion_r918374814 ## tests/python/frontend/caffe/test_forward.py: ## @@ -37,9 +32,13 @@ import tvm import tvm.testing from tvm import relay -from tvm.contrib import utils,

[GitHub] [tvm] areusch commented on issue #12020: [Bug] Enable USE_DNNL cause Sphix crash when processing from_oneflow.py and from_paddle.py for documents

areusch commented on issue #12020: URL: https://github.com/apache/tvm/issues/12020#issuecomment-1180900035 @huajsj just curious why you needed USE_DNNL ON in your tutorial? is that closer to the use case for pipeline executor, or is it possible to demonstrate pipeline executor with just

[GitHub] [tvm] areusch commented on a diff in pull request #11953: [CI][docker] Add comment

areusch commented on code in PR #11953: URL: https://github.com/apache/tvm/pull/11953#discussion_r918371458 ## docker/bash.sh: ## @@ -138,6 +138,8 @@ CONTAINER_NAME= # "${REPO_DIR}". The consistent directory for Jenkins is currently # necessary to allow cmake build commands

[tvm] branch main updated (d27f853ed5 -> 7baf4be2d4)

This is an automated email from the ASF dual-hosted git repository. areusch pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from d27f853ed5 [ci][docker] fix the path of custom toolchain in ci_qemu for csinn2 (#11905) add 7baf4be2d4 [relay]

[GitHub] [tvm] areusch merged pull request #10466: [relay] Changed 'name' field to 'registry_name' for Executor and Runtime

areusch merged PR #10466: URL: https://github.com/apache/tvm/pull/10466 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail:

[GitHub] [tvm-rfcs] areusch commented on a diff in pull request #81: [USMP] Update RFC with constants pools

areusch commented on code in PR #81: URL: https://github.com/apache/tvm-rfcs/pull/81#discussion_r918369300 ## rfcs/0009_Unified_Static_Memory_Planning.md: ## @@ -461,19 +461,43 @@ Inputs : * IRModule containing * AoT TIR PrimFunc (the control function describing the call

[GitHub] [tvm-rfcs] areusch commented on a diff in pull request #81: [USMP] Update RFC with constants pools

areusch commented on code in PR #81: URL: https://github.com/apache/tvm-rfcs/pull/81#discussion_r918368975 ## rfcs/0009_Unified_Static_Memory_Planning.md: ## @@ -461,19 +461,43 @@ Inputs : * IRModule containing * AoT TIR PrimFunc (the control function describing the call

[GitHub] [tvm] areusch commented on pull request #11905: [ci][docker] fix the path of custom toolchain in ci_qemu for csinn2

areusch commented on PR #11905: URL: https://github.com/apache/tvm/pull/11905#issuecomment-1180892488 @alter-xp my apologies, we were out for a week and i didn't see this beforehand. are we good to build/push new ci_qemu images now? as an aside--there's been discussion on #11809

[tvm] branch main updated (ae72e7e653 -> d27f853ed5)

This is an automated email from the ASF dual-hosted git repository. areusch pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from ae72e7e653 Fix node.func to node.funcs on parser.py (#12053) add d27f853ed5 [ci][docker] fix the path of custom

[GitHub] [tvm] areusch merged pull request #11905: [ci][docker] fix the path of custom toolchain in ci_qemu for csinn2

areusch merged PR #11905: URL: https://github.com/apache/tvm/pull/11905 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail:

[GitHub] [tvm] gromero closed issue #11900: [Bug]

gromero closed issue #11900: [Bug] URL: https://github.com/apache/tvm/issues/11900 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail:

[GitHub] [tvm] gromero commented on issue #11900: [Bug]

gromero commented on issue #11900: URL: https://github.com/apache/tvm/issues/11900#issuecomment-1180879749 Since #12053 is merged I'm closing it. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to

[GitHub] [tvm-rfcs] areusch commented on a diff in pull request #71: Add Target Pre-processing RFC

areusch commented on code in PR #71: URL: https://github.com/apache/tvm-rfcs/pull/71#discussion_r918357125 ## rfcs/0071-target-json-parser.md: ## @@ -0,0 +1,172 @@ +- Feature Name: target-json-preprocessor +- Start Date: 2022-04-04 +- RFC PR:

[tvm] branch last-successful updated (04db26e8d9 -> b4c1cc02eb)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch last-successful in repository https://gitbox.apache.org/repos/asf/tvm.git from 04db26e8d9 [microtvm][RVM] Refactor Arduino/Zephyr into one RVM (#12023) add b4c1cc02eb

[GitHub] [tvm] mbs-octoml commented on pull request #11993: [Collage] PartitionRule (though without CombinePartitionRule)

mbs-octoml commented on PR #11993: URL: https://github.com/apache/tvm/pull/11993#issuecomment-1180871880 This is now based on main and ready for review, thanks. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL

[tvm] branch main updated (5be8e0a3de -> ae72e7e653)

This is an automated email from the ASF dual-hosted git repository. kparzysz pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from 5be8e0a3de [Collage] SubGraphs (#11981) add ae72e7e653 Fix node.func to node.funcs on parser.py (#12053) No new

[GitHub] [tvm] kparzysz-quic merged pull request #12053: Fix node.func to node.funcs on parser.py

kparzysz-quic merged PR #12053: URL: https://github.com/apache/tvm/pull/12053 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail:

[tvm] branch no_vta_for_0.9 updated (e6c1cb4506 -> c59f10657a)

This is an automated email from the ASF dual-hosted git repository. driazati pushed a change to branch no_vta_for_0.9 in repository https://gitbox.apache.org/repos/asf/tvm.git omit e6c1cb4506 [release][v0.9] Remove VTA submodule add c59f10657a [release][v0.9] Remove VTA submodule This

[tvm] branch main updated (b4c1cc02eb -> 5be8e0a3de)

This is an automated email from the ASF dual-hosted git repository. masahi pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from b4c1cc02eb [CMSIS-NN][Perf] Converted Relay Conv2D into CMSIS-NN Depthwise (#12006) add 5be8e0a3de [Collage]

[GitHub] [tvm] masahi merged pull request #11981: [Collage] SubGraphs

masahi merged PR #11981: URL: https://github.com/apache/tvm/pull/11981 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail:

[tvm] branch last-successful updated (c4dc41a0dd -> 04db26e8d9)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch last-successful in repository https://gitbox.apache.org/repos/asf/tvm.git from c4dc41a0dd [Frontend][TFLite] PreLU alpha can be an expr (#11879) add 04db26e8d9 [microtvm][RVM]

[GitHub] [tvm] ganler commented on issue #11794: [Bug] Int64 BroadCast-ArgMax triggers assertion error at graph runtime

ganler commented on issue #11794: URL: https://github.com/apache/tvm/issues/11794#issuecomment-1180852039 @everton1984 Thanks for the interest! There are many interger mismatch bugs in TVM happening in compilation stage: https://github.com/apache/tvm/pull/10519

[GitHub] [tvm] areusch commented on issue #11804: [ci] sphinx gallery issue

areusch commented on issue #11804: URL: https://github.com/apache/tvm/issues/11804#issuecomment-1180847643 ``` [2022-06-18T02:10:18.067Z] /workspace/docs/_staging/contribute/pull_request.rst:120: WARNING: Error in "code" directive: [2022-06-18T02:10:18.067Z] maximum 1 argument(s)

[GitHub] [tvm] areusch commented on issue #11462: CI monitoring rotation schedule

areusch commented on issue #11462: URL: https://github.com/apache/tvm/issues/11462#issuecomment-1180842150 -

[GitHub] [tvm] areusch commented on issue #11764: [ci] test wrapper scripts failure

areusch commented on issue #11764: URL: https://github.com/apache/tvm/issues/11764#issuecomment-1180839630 duplicate of #11749 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific

[GitHub] [tvm] areusch closed issue #11764: [ci] test wrapper scripts failure

areusch closed issue #11764: [ci] test wrapper scripts failure URL: https://github.com/apache/tvm/issues/11764 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe,

[GitHub] [tvm] areusch opened a new issue, #12062: [Flaky Test] `tests/python/frontend/onnx/test_forward.py::test_qlinearadd[cuda]`

areusch opened a new issue, #12062: URL: https://github.com/apache/tvm/issues/12062 These tests were found to be flaky (intermittently failing on `main` or failed in a PR with unrelated changes). See [the

[GitHub] [tvm] everton1984 commented on issue #11794: [Bug] Int64 BroadCast-ArgMax triggers assertion error at graph runtime

everton1984 commented on issue #11794: URL: https://github.com/apache/tvm/issues/11794#issuecomment-1180785153 @ganler Thanks for answering, I'm investigating a possible fix but just to be clear on what is expected here then is that `broadcast_to` should just ignore the data type used to

[GitHub] [tvm] gbonik commented on a diff in pull request #11977: [TVMScript] Add ObjectPath class

gbonik commented on code in PR #11977: URL: https://github.com/apache/tvm/pull/11977#discussion_r918272273 ## include/tvm/node/object_path.h: ## @@ -0,0 +1,281 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See

[GitHub] [tvm] mbs-octoml commented on issue #12033: [Bug] [microTVM] BYOC hardware test fails when used with Arduino

mbs-octoml commented on issue #12033: URL: https://github.com/apache/tvm/issues/12033#issuecomment-1180765249 My bad, I think the fix is https://github.com/apache/tvm/pull/12061. A unit test would be nice. -- This is an automated message from the Apache Git Service. To respond to the

[GitHub] [tvm] mbs-octoml opened a new pull request, #12061: [AOT][BUG] Only include extra headers if the constants array is needed.

mbs-octoml opened a new pull request, #12061: URL: https://github.com/apache/tvm/pull/12061 I broke AOT in #11981 since I misread the original code. Spotted by @guberti, thanks! -- This is an automated message from the Apache Git Service. To respond to the message, please log on to

[GitHub] [tvm] areusch opened a new issue, #12060: [Bug] AOTExecutor (in CRT and C++) should error with good message when not compiled with correct interface-api and unpacked-api

areusch opened a new issue, #12060: URL: https://github.com/apache/tvm/issues/12060 See https://discuss.tvm.apache.org/t/aot-problem-in-microtvm-utvm-host-driven-application/13100 ### Expected behavior Should error with a more informative assert. ### Actual behavior

[GitHub] [tvm] gbonik commented on a diff in pull request #11977: [TVMScript] Add ObjectPath class

gbonik commented on code in PR #11977: URL: https://github.com/apache/tvm/pull/11977#discussion_r918260734 ## src/node/object_path.cc: ## @@ -0,0 +1,322 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the

[tvm] branch last-successful updated (9ee25eb9f4 -> c4dc41a0dd)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch last-successful in repository https://gitbox.apache.org/repos/asf/tvm.git from 9ee25eb9f4 [MetaSchedule] Added a cost model (#11961) add c4dc41a0dd [Frontend][TFLite] PreLU alpha

[tvm] branch main updated: [CMSIS-NN][Perf] Converted Relay Conv2D into CMSIS-NN Depthwise (#12006)

This is an automated email from the ASF dual-hosted git repository. mousius pushed a commit to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git The following commit(s) were added to refs/heads/main by this push: new b4c1cc02eb [CMSIS-NN][Perf] Converted Relay Conv2D

[GitHub] [tvm] Mousius merged pull request #12006: [CMSIS-NN][Perf] Converted Relay Conv2D into CMSIS-NN Depthwise

Mousius merged PR #12006: URL: https://github.com/apache/tvm/pull/12006 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail:

[GitHub] [tvm] gbonik commented on a diff in pull request #11977: [TVMScript] Add ObjectPath class

gbonik commented on code in PR #11977: URL: https://github.com/apache/tvm/pull/11977#discussion_r918257461 ## src/node/object_path.cc: ## @@ -0,0 +1,322 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the

[GitHub] [tvm] Mousius commented on a diff in pull request #12006: [CMSIS-NN][Perf] Converted Relay Conv2D into CMSIS-NN Depthwise

Mousius commented on code in PR #12006: URL: https://github.com/apache/tvm/pull/12006#discussion_r918257288 ## tests/python/contrib/test_cmsisnn/test_conv2d.py: ## @@ -540,6 +547,135 @@ def test_depthwise_int8( ) +@tvm.testing.requires_cmsisnn

[GitHub] [tvm] Mousius commented on a diff in pull request #12006: [CMSIS-NN][Perf] Converted Relay Conv2D into CMSIS-NN Depthwise

Mousius commented on code in PR #12006: URL: https://github.com/apache/tvm/pull/12006#discussion_r918257011 ## tests/python/contrib/test_cmsisnn/test_conv2d.py: ## @@ -540,6 +542,111 @@ def test_depthwise_int8( ) +@tvm.testing.requires_cmsisnn

[GitHub] [tvm] gbonik commented on a diff in pull request #11977: [TVMScript] Add ObjectPath class

gbonik commented on code in PR #11977: URL: https://github.com/apache/tvm/pull/11977#discussion_r918253260 ## include/tvm/node/object_path.h: ## @@ -0,0 +1,281 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See

[GitHub] [tvm] gbonik commented on a diff in pull request #11977: [TVMScript] Add ObjectPath class

gbonik commented on code in PR #11977: URL: https://github.com/apache/tvm/pull/11977#discussion_r918245728 ## include/tvm/node/object_path.h: ## @@ -0,0 +1,281 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See

[GitHub] [tvm] gbonik commented on a diff in pull request #11977: [TVMScript] Add ObjectPath class

gbonik commented on code in PR #11977: URL: https://github.com/apache/tvm/pull/11977#discussion_r918241949 ## src/node/object_path.cc: ## @@ -0,0 +1,322 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the

[tvm] branch main updated: [microtvm][RVM] Refactor Arduino/Zephyr into one RVM (#12023)

This is an automated email from the ASF dual-hosted git repository. mehrdadh pushed a commit to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git The following commit(s) were added to refs/heads/main by this push: new 04db26e8d9 [microtvm][RVM] Refactor Arduino/Zephyr

[GitHub] [tvm] mehrdadh merged pull request #12023: [microtvm][RVM] Refactor Arduino/Zephyr into one RVM

mehrdadh merged PR #12023: URL: https://github.com/apache/tvm/pull/12023 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail:

[GitHub] [tvm] gbonik commented on a diff in pull request #11977: [TVMScript] Add ObjectPath class

gbonik commented on code in PR #11977: URL: https://github.com/apache/tvm/pull/11977#discussion_r918239877 ## src/node/object_path.cc: ## @@ -0,0 +1,322 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the

[GitHub] [tvm] gbonik commented on a diff in pull request #11977: [TVMScript] Add ObjectPath class

gbonik commented on code in PR #11977: URL: https://github.com/apache/tvm/pull/11977#discussion_r918236015 ## src/node/object_path.cc: ## @@ -0,0 +1,322 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the

[GitHub] [tvm] github-actions[bot] commented on pull request #12058: [ci] Add a manual retry for conda setup

github-actions[bot] commented on PR #12058: URL: https://github.com/apache/tvm/pull/12058#issuecomment-1180731292 Built docs for commit 68f94d29d962206a589e517c3d77bac740c7b6df can be found [here](https://pr-docs.tlcpack.ai/PR-12058/1/docs/index.html). -- This is an automated

[GitHub] [tvm] vinx13 opened a new pull request, #12059: [MetaSchedule] Add MultiLevelTilingTensorCore rule for auto-tensorization on CUDA

vinx13 opened a new pull request, #12059: URL: https://github.com/apache/tvm/pull/12059 Co-authored-by: Bohan Hou \<32121147+spectrometer...@users.noreply.github.com\> Co-authored-by: Hongyi Jin \<3231950...@qq.com\> cc @junrushao1994 @masahi @spectrometerHBH @jinhongyii --

[GitHub] [tvm] driazati opened a new pull request, #12058: [ci] Add a manual retry for conda setup

driazati opened a new pull request, #12058: URL: https://github.com/apache/tvm/pull/12058 This makes it so the conda setup will re-run entirely in case of failures like in https://github.com/apache/tvm/runs/7287493088. [This

[tvm] branch gha_conda_flakes created (now 68f94d29d9)

This is an automated email from the ASF dual-hosted git repository. driazati pushed a change to branch gha_conda_flakes in repository https://gitbox.apache.org/repos/asf/tvm.git at 68f94d29d9 [ci] Add a manual retry for conda setup No new revisions were added by this update.

[GitHub] [tvm] driazati closed pull request #12057: [ci] Add a manual retry for conda setup

driazati closed pull request #12057: [ci] Add a manual retry for conda setup URL: https://github.com/apache/tvm/pull/12057 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To

[GitHub] [tvm] driazati closed pull request #12056: [ci] Add a manual retry for conda setup

driazati closed pull request #12056: [ci] Add a manual retry for conda setup URL: https://github.com/apache/tvm/pull/12056 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To

[GitHub] [tvm] driazati opened a new pull request, #12056: [ci] Add a manual retry for conda setup

driazati opened a new pull request, #12056: URL: https://github.com/apache/tvm/pull/12056 (git log -1 --format='%b') -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To

[GitHub] [tvm] ganler commented on issue #11794: [Bug] Int64 BroadCast-ArgMax triggers assertion error at graph runtime

ganler commented on issue #11794: URL: https://github.com/apache/tvm/issues/11794#issuecomment-1180713035 > @ganler Sorry if I understood improperly but the situation here seems to be that the broadcast operation is mismatching, if you want it to output a `int64` tensor shouldn't you use

[GitHub] [tvm] mbs-octoml commented on pull request #11981: [Collage] SubGraphs

mbs-octoml commented on PR #11981: URL: https://github.com/apache/tvm/pull/11981#issuecomment-1180710973 Thanks @SebastianBoblest, every little bit helps. Thanks and PTAL @mbaret. -- This is an automated message from the Apache Git Service. To respond to the message, please log on

[GitHub] [tvm] mbs-octoml commented on a diff in pull request #11981: [Collage] SubGraphs

mbs-octoml commented on code in PR #11981: URL: https://github.com/apache/tvm/pull/11981#discussion_r918212927 ## src/relay/collage/sub_graph.h: ## @@ -0,0 +1,451 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements.

[GitHub] [tvm] mbs-octoml commented on a diff in pull request #11981: [Collage] SubGraphs

mbs-octoml commented on code in PR #11981: URL: https://github.com/apache/tvm/pull/11981#discussion_r918212033 ## src/relay/collage/sub_graph.h: ## @@ -0,0 +1,451 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements.

[GitHub] [tvm] everton1984 commented on issue #11794: [Bug] Int64 BroadCast-ArgMax triggers assertion error at graph runtime

everton1984 commented on issue #11794: URL: https://github.com/apache/tvm/issues/11794#issuecomment-1180702130 @ganler Sorry if I understood improperly but the situation here seems to be that the broadcast operation is mismatching, if you want it to output a `int64` tensor shouldn't you

[tvm] branch no_vta_for_0.9 updated (9e8973fe4a -> e6c1cb4506)

This is an automated email from the ASF dual-hosted git repository. driazati pushed a change to branch no_vta_for_0.9 in repository https://gitbox.apache.org/repos/asf/tvm.git omit 9e8973fe4a [release][v0.9] Remove VTA submodule add e6c1cb4506 [release][v0.9] Remove VTA submodule This

[GitHub] [tvm] mbs-octoml commented on a diff in pull request #11981: [Collage] SubGraphs

mbs-octoml commented on code in PR #11981: URL: https://github.com/apache/tvm/pull/11981#discussion_r918201673 ## src/relay/collage/sub_graph.h: ## @@ -0,0 +1,451 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements.

[GitHub] [tvm] mbs-octoml commented on a diff in pull request #11981: [Collage] SubGraphs

mbs-octoml commented on code in PR #11981: URL: https://github.com/apache/tvm/pull/11981#discussion_r918201354 ## src/relay/collage/README.md: ## @@ -0,0 +1,26 @@ + + + + + + + + + + + + + + + + + +The `CollagePartition` pass for finding optimal partitionings of Relay models. +

  1   2   >