[GitHub] [tvm-rfcs] masahi commented on a change in pull request #18: [RFC] Adding initial SVE implementation

masahi commented on a change in pull request #18: URL: https://github.com/apache/tvm-rfcs/pull/18#discussion_r697183013 ## File path: rfcs/0018-initial-sve-addition.md ## @@ -0,0 +1,221 @@ +- Feature Name: Adding Initial SVE Support to TVM +- Start Date: 2021-07-30 +- RFC PR:

[GitHub] [tvm-rfcs] masahi commented on a change in pull request #18: [RFC] Adding initial SVE implementation

masahi commented on a change in pull request #18: URL: https://github.com/apache/tvm-rfcs/pull/18#discussion_r697200267 ## File path: rfcs/0018-initial-sve-addition.md ## @@ -0,0 +1,221 @@ +- Feature Name: Adding Initial SVE Support to TVM +- Start Date: 2021-07-30 +- RFC PR:

[GitHub] [tvm] vvchernov commented on a change in pull request #8709: [Frontend] clean duplicates of infer_type and infer_shape in frontends

vvchernov commented on a change in pull request #8709: URL: https://github.com/apache/tvm/pull/8709#discussion_r697210306 ## File path: python/tvm/relay/frontend/pytorch.py ## @@ -148,29 +146,13 @@ def infer_type(self, node, mod=None): if node in self.types:

[GitHub] [tvm-rfcs] masahi edited a comment on pull request #18: [RFC] Adding initial SVE implementation

masahi edited a comment on pull request #18: URL: https://github.com/apache/tvm-rfcs/pull/18#issuecomment-906985030 Thanks @MeeraN7 @giuseros, I like the approach of making the vectorized loop explicit with `VL` parameter at the TIR level, in contrast to how the fixed-width vectorization

[GitHub] [tvm] xqdan commented on a change in pull request #8863: [TensorIR][M2a] CacheRead/Write

xqdan commented on a change in pull request #8863: URL: https://github.com/apache/tvm/pull/8863#discussion_r697224923 ## File path: tests/python/unittest/test_tir_schedule_cache_read_write.py ## @@ -0,0 +1,625 @@ +# Licensed to the Apache Software Foundation (ASF) under one +#

[GitHub] [tvm-rfcs] masahi commented on a change in pull request #18: [RFC] Adding initial SVE implementation

masahi commented on a change in pull request #18: URL: https://github.com/apache/tvm-rfcs/pull/18#discussion_r697187201 ## File path: rfcs/0018-initial-sve-addition.md ## @@ -0,0 +1,221 @@ +- Feature Name: Adding Initial SVE Support to TVM +- Start Date: 2021-07-30 +- RFC PR:

[GitHub] [tvm-rfcs] masahi commented on a change in pull request #18: [RFC] Adding initial SVE implementation

masahi commented on a change in pull request #18: URL: https://github.com/apache/tvm-rfcs/pull/18#discussion_r697183013 ## File path: rfcs/0018-initial-sve-addition.md ## @@ -0,0 +1,221 @@ +- Feature Name: Adding Initial SVE Support to TVM +- Start Date: 2021-07-30 +- RFC PR:

[GitHub] [tvm-rfcs] masahi commented on a change in pull request #18: [RFC] Adding initial SVE implementation

masahi commented on a change in pull request #18: URL: https://github.com/apache/tvm-rfcs/pull/18#discussion_r697187201 ## File path: rfcs/0018-initial-sve-addition.md ## @@ -0,0 +1,221 @@ +- Feature Name: Adding Initial SVE Support to TVM +- Start Date: 2021-07-30 +- RFC PR:

[GitHub] [tvm-rfcs] masahi commented on a change in pull request #18: [RFC] Adding initial SVE implementation

masahi commented on a change in pull request #18: URL: https://github.com/apache/tvm-rfcs/pull/18#discussion_r697205476 ## File path: rfcs/0018-initial-sve-addition.md ## @@ -0,0 +1,221 @@ +- Feature Name: Adding Initial SVE Support to TVM +- Start Date: 2021-07-30 +- RFC PR:

[GitHub] [tvm-rfcs] masahi commented on pull request #18: [RFC] Adding initial SVE implementation

masahi commented on pull request #18: URL: https://github.com/apache/tvm-rfcs/pull/18#issuecomment-906985030 Thanks @MeeraN7 @giuseros, I like the approach of making the vectorized loop explicit with `VL` parameter at the TIR level, in contrast to how the fixed-width vectorization is done

[tvm] branch main updated: Add a PaddlePaddle Frontend (#8645)

This is an automated email from the ASF dual-hosted git repository. leandron 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 e774fed Add a PaddlePaddle Frontend (#8645) e774fed

[GitHub] [tvm] leandron merged pull request #8645: Add a PaddlePaddle Frontend

leandron merged pull request #8645: URL: https://github.com/apache/tvm/pull/8645 -- 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 commented on a change in pull request #8709: [Frontend] clean duplicates of infer_type and infer_shape in frontends

masahi commented on a change in pull request #8709: URL: https://github.com/apache/tvm/pull/8709#discussion_r697250903 ## File path: python/tvm/relay/frontend/pytorch.py ## @@ -148,29 +146,13 @@ def infer_type(self, node, mod=None): if node in self.types:

[GitHub] [tvm] junrushao1994 commented on issue #7527: [RFC][Tracking Issue] TensorIR Scheduling

junrushao1994 commented on issue #7527: URL: https://github.com/apache/tvm/issues/7527#issuecomment-907032840 Hey @manupa-arm thanks for your interest! > Will the integration be using #7987 ? Most of the operators are defined with the TE DSL in TOPI, so in these cases we will

[GitHub] [tvm-rfcs] masahi commented on a change in pull request #18: [RFC] Adding initial SVE implementation

masahi commented on a change in pull request #18: URL: https://github.com/apache/tvm-rfcs/pull/18#discussion_r697195878 ## File path: rfcs/0018-initial-sve-addition.md ## @@ -0,0 +1,221 @@ +- Feature Name: Adding Initial SVE Support to TVM +- Start Date: 2021-07-30 +- RFC PR:

[GitHub] [tvm-rfcs] masahi commented on a change in pull request #18: [RFC] Adding initial SVE implementation

masahi commented on a change in pull request #18: URL: https://github.com/apache/tvm-rfcs/pull/18#discussion_r697187201 ## File path: rfcs/0018-initial-sve-addition.md ## @@ -0,0 +1,221 @@ +- Feature Name: Adding Initial SVE Support to TVM +- Start Date: 2021-07-30 +- RFC PR:

[GitHub] [tvm-rfcs] masahi commented on a change in pull request #18: [RFC] Adding initial SVE implementation

masahi commented on a change in pull request #18: URL: https://github.com/apache/tvm-rfcs/pull/18#discussion_r697183013 ## File path: rfcs/0018-initial-sve-addition.md ## @@ -0,0 +1,221 @@ +- Feature Name: Adding Initial SVE Support to TVM +- Start Date: 2021-07-30 +- RFC PR:

[GitHub] [tvm-rfcs] masahi commented on a change in pull request #18: [RFC] Adding initial SVE implementation

masahi commented on a change in pull request #18: URL: https://github.com/apache/tvm-rfcs/pull/18#discussion_r697187201 ## File path: rfcs/0018-initial-sve-addition.md ## @@ -0,0 +1,221 @@ +- Feature Name: Adding Initial SVE Support to TVM +- Start Date: 2021-07-30 +- RFC PR:

[GitHub] [tvm-rfcs] masahi commented on a change in pull request #18: [RFC] Adding initial SVE implementation

masahi commented on a change in pull request #18: URL: https://github.com/apache/tvm-rfcs/pull/18#discussion_r697183013 ## File path: rfcs/0018-initial-sve-addition.md ## @@ -0,0 +1,221 @@ +- Feature Name: Adding Initial SVE Support to TVM +- Start Date: 2021-07-30 +- RFC PR:

[GitHub] [tvm] echuraev opened a new pull request #8864: [RPC] Fix ios_rpc build

echuraev opened a new pull request #8864: URL: https://github.com/apache/tvm/pull/8864 Thanks for contributing to TVM! Please refer to guideline https://tvm.apache.org/docs/contribute/ for useful information and tips. After the pull request is submitted, please request code reviews from

[GitHub] [tvm-rfcs] masahi commented on a change in pull request #18: [RFC] Adding initial SVE implementation

masahi commented on a change in pull request #18: URL: https://github.com/apache/tvm-rfcs/pull/18#discussion_r697183013 ## File path: rfcs/0018-initial-sve-addition.md ## @@ -0,0 +1,221 @@ +- Feature Name: Adding Initial SVE Support to TVM +- Start Date: 2021-07-30 +- RFC PR:

[GitHub] [tvm] Hzfengsy commented on issue #7527: [RFC][Tracking Issue] TensorIR Scheduling

Hzfengsy commented on issue #7527: URL: https://github.com/apache/tvm/issues/7527#issuecomment-906972044 Thanks, @manupa-arm. Of course, we will! The ultimate goal of TensorIR is to replace the current TE schedule. Before integrating it to `relay`, we need to finish all of

[GitHub] [tvm] leandron commented on pull request #8645: Add a PaddlePaddle Frontend

leandron commented on pull request #8645: URL: https://github.com/apache/tvm/pull/8645#issuecomment-906994739 This is merged now. Thanks @jiangjiajun and reviewers @comaniac @junrushao1994 @areusch @jcf94! -- This is an automated message from the Apache Git Service. To respond to the

[GitHub] [tvm] manupa-arm commented on issue #7527: [RFC][Tracking Issue] TensorIR Scheduling

manupa-arm commented on issue #7527: URL: https://github.com/apache/tvm/issues/7527#issuecomment-906994980 Ack. Thanks. Out of curiosity for the planned relay integration, * Will the integration be using #7987 ? * If you guys have decided, please let us know what other the

[GitHub] [tvm] Hzfengsy opened a new pull request #8863: [TensorIR][M2a] CacheRead/Write

Hzfengsy opened a new pull request #8863: URL: https://github.com/apache/tvm/pull/8863 This PR is part of the TensorIR upstreaming effort (#7527), which adds a schedule primitive: cache_read/write. CC: @junrushao1994 @vinx13 @MasterJH5574 @comaniac @jcf94 Co-authored-by: Bohan

[GitHub] [tvm] manupa-arm commented on issue #7527: [RFC][Tracking Issue] TensorIR Scheduling

manupa-arm commented on issue #7527: URL: https://github.com/apache/tvm/issues/7527#issuecomment-906964640 Hi @junrushao1994 @Hzfengsy , Thanks for the effort that goes in here. Some of this scheduling primitives are really useful that are not there in TE. Will there be a

[GitHub] [tvm] vvchernov commented on a change in pull request #8709: [Frontend] clean duplicates of infer_type and infer_shape in frontends

vvchernov commented on a change in pull request #8709: URL: https://github.com/apache/tvm/pull/8709#discussion_r697263806 ## File path: python/tvm/relay/frontend/pytorch.py ## @@ -148,29 +146,13 @@ def infer_type(self, node, mod=None): if node in self.types:

[GitHub] [tvm-rfcs] masahi commented on a change in pull request #25: [RFC]PyTorchTVM

masahi commented on a change in pull request #25: URL: https://github.com/apache/tvm-rfcs/pull/25#discussion_r697321721 ## File path: rfcs/0025-add-pytorch-tvm.md ## @@ -0,0 +1,265 @@ +- Feature Name: PyTorchTVM +- Start Date: 2021-08-24 +- RFC PR:

[GitHub] [tvm] manupa-arm commented on issue #7527: [RFC][Tracking Issue] TensorIR Scheduling

manupa-arm commented on issue #7527: URL: https://github.com/apache/tvm/issues/7527#issuecomment-907200410 Ack, Many thanks for the info ! -- 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

[GitHub] [tvm] mehrdadh commented on pull request #8846: [microtvm][Zephyr] Increase timeout to fix flaky tests

mehrdadh commented on pull request #8846: URL: https://github.com/apache/tvm/pull/8846#issuecomment-907054610 @masahi no worries. @junrushao1994 I did test it with repeat and it still breaks. I need to do more investigation on this. -- This is an automated message from the Apache Git

[GitHub] [tvm] manupa-arm edited a comment on issue #7527: [RFC][Tracking Issue] TensorIR Scheduling

manupa-arm edited a comment on issue #7527: URL: https://github.com/apache/tvm/issues/7527#issuecomment-907065910 Hey @junrushao1994 , Thanks for the clarifications. Since the relay integration is supposed to be happening prior to meta schedule is being concluded, will there

[GitHub] [tvm] jcf94 commented on pull request #8497: [Runtime] add set_output_zero_copy

jcf94 commented on pull request #8497: URL: https://github.com/apache/tvm/pull/8497#issuecomment-907064130 Thanks! Now merged. @sunjiweiswift @huajsj @comaniac -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL

[GitHub] [tvm] masahi commented on a change in pull request #8497: [Runtime] add set_output_zero_copy

masahi commented on a change in pull request #8497: URL: https://github.com/apache/tvm/pull/8497#discussion_r697324406 ## File path: tests/cpp/build_module_test.cc ## @@ -199,3 +199,129 @@ TEST(BuildModule, Heterogeneous) { ICHECK_LT(std::fabs(p_out[i] - (i + (i + 1.0) -

[GitHub] [tvm] mehrdadh commented on pull request #8748: [microTVM][RVM] Add Arduino RVM

mehrdadh commented on pull request #8748: URL: https://github.com/apache/tvm/pull/8748#issuecomment-907059409 I built the image and tested it against Arduino Due. It fails on two tests, currently waiting for fixing those tests before merging. -- This is an automated message from the

[tvm] branch main updated: [Runtime] add set_output_zero_copy (#8497)

This is an automated email from the ASF dual-hosted git repository. jcf94 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 3306857 [Runtime] add set_output_zero_copy (#8497)

[GitHub] [tvm] jcf94 merged pull request #8497: [Runtime] add set_output_zero_copy

jcf94 merged pull request #8497: URL: https://github.com/apache/tvm/pull/8497 -- 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] Hzfengsy commented on issue #7527: [RFC][Tracking Issue] TensorIR Scheduling

Hzfengsy commented on issue #7527: URL: https://github.com/apache/tvm/issues/7527#issuecomment-907176062 Hey @manupa-arm, Don't worry. We will make TensorIR be an optional but not the default backend of relay as our first step. There must be many works (including meta schedule and

[GitHub] [tvm] jcf94 edited a comment on pull request #8790: [RELAY][FRONTEND] Initial OneFlow frontend support.

jcf94 edited a comment on pull request #8790: URL: https://github.com/apache/tvm/pull/8790#issuecomment-907055939 Thanks! @hhhfccz We've just merged the frontend support for PaddlePaddle in #8645, you can check about what that PR did to speed up the review of this PR. Several basic

[GitHub] [tvm] jcf94 commented on pull request #8790: [RELAY][FRONTEND] Initial OneFlow frontend support.

jcf94 commented on pull request #8790: URL: https://github.com/apache/tvm/pull/8790#issuecomment-907055939 Thanks! @hhhfccz We've just merged the frontend support for PaddlePaddle in #8645, you can check about what that PR did. Several basic things: 1. Clean the ci linting; 2. Add

[GitHub] [tvm] manupa-arm commented on issue #7527: [RFC][Tracking Issue] TensorIR Scheduling

manupa-arm commented on issue #7527: URL: https://github.com/apache/tvm/issues/7527#issuecomment-907065910 Hey @junrushao1994 , Thanks for the clarifications. Since the relay integration is supposed to be happening prior to meta schedule is being concluded, what would be the

[GitHub] [tvm] t-vi commented on a change in pull request #8709: [Frontend] clean duplicates of infer_type and infer_shape in frontends

t-vi commented on a change in pull request #8709: URL: https://github.com/apache/tvm/pull/8709#discussion_r697425658 ## File path: python/tvm/relay/frontend/pytorch.py ## @@ -148,29 +146,13 @@ def infer_type(self, node, mod=None): if node in self.types:

[GitHub] [tvm] junrushao1994 commented on pull request #8790: [RELAY][FRONTEND] Initial OneFlow frontend support.

junrushao1994 commented on pull request #8790: URL: https://github.com/apache/tvm/pull/8790#issuecomment-907058135 OneFlow installation in the CI’s docker image is required in a separate PR. After the docker image PR is merged, you guys may use https://github.com/apache/tvm/issues/8824 as

[GitHub] [tvm] Hzfengsy commented on a change in pull request #8863: [TensorIR][M2a] CacheRead/Write

Hzfengsy commented on a change in pull request #8863: URL: https://github.com/apache/tvm/pull/8863#discussion_r697403291 ## File path: tests/python/unittest/test_tir_schedule_cache_read_write.py ## @@ -0,0 +1,625 @@ +# Licensed to the Apache Software Foundation (ASF) under one

[GitHub] [tvm-rfcs] t-vi commented on pull request #25: [RFC]PyTorchTVM

t-vi commented on pull request #25: URL: https://github.com/apache/tvm-rfcs/pull/25#issuecomment-907192105 I wonder whether this would make the torch fallback op (https://github.com/apache/tvm/pull/7401) more or less useful (it would depend on what you (plan to) do with unsupported ops).

[GitHub] [tvm] vvchernov commented on a change in pull request #8709: [Frontend] clean duplicates of infer_type and infer_shape in frontends

vvchernov commented on a change in pull request #8709: URL: https://github.com/apache/tvm/pull/8709#discussion_r697456678 ## File path: python/tvm/relay/frontend/pytorch.py ## @@ -148,29 +146,13 @@ def infer_type(self, node, mod=None): if node in self.types:

[GitHub] [tvm-vta] aasorokiin commented on pull request #32: VTA Chisel Wide memory interface.

aasorokiin commented on pull request #32: URL: https://github.com/apache/tvm-vta/pull/32#issuecomment-907319156 I tried that, but checks keep failing. I cannot reproduce the problem. For me it looks like CI caching error. @vegaluisjose -- This is an automated message from the

[GitHub] [tvm-vta] vegaluisjose edited a comment on pull request #32: VTA Chisel Wide memory interface.

vegaluisjose edited a comment on pull request #32: URL: https://github.com/apache/tvm-vta/pull/32#issuecomment-907340569 I think I found what is causing the issue. Can you try to update the docker image numbers [here](https://github.com/apache/tvm-vta/blob/main/Jenkinsfile#L24-L26) with

[GitHub] [tvm-vta] vegaluisjose commented on pull request #32: VTA Chisel Wide memory interface.

vegaluisjose commented on pull request #32: URL: https://github.com/apache/tvm-vta/pull/32#issuecomment-907340569 I think I found what is causing the issue. Can you try to update the docker image numbers [here](https://github.com/apache/tvm-vta/blob/main/Jenkinsfile#L24-L26) with the

[GitHub] [tvm] jcf94 edited a comment on pull request #8790: [RELAY][FRONTEND] Initial OneFlow frontend support.

jcf94 edited a comment on pull request #8790: URL: https://github.com/apache/tvm/pull/8790#issuecomment-907055939 Thanks! @hhhfccz We've just merged the frontend support for PaddlePaddle in #8645, you can check about what that PR did to speed up the review of this PR. Several basic

[GitHub] [tvm] masahi merged pull request #8859: [Hexagon] Change declaration order of unique_ptr objects to fix crash

masahi merged pull request #8859: URL: https://github.com/apache/tvm/pull/8859 -- 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 main updated: [Hexagon] Change declaration order of unique_ptr objects to fix crash (#8859)

This is an automated email from the ASF dual-hosted git repository. masahi 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 cf19c88 [Hexagon] Change declaration order of

[GitHub] [tvm-vta] aasorokiin commented on pull request #32: VTA Chisel Wide memory interface.

aasorokiin commented on pull request #32: URL: https://github.com/apache/tvm-vta/pull/32#issuecomment-907298045 Changes were addressed. TSIM ci also should pass now. @vegaluisjose -- This is an automated message from the Apache Git Service. To respond to the message, please log on to

[GitHub] [tvm] jtuyls commented on pull request #8814: [CI][VitisAI] Update CI Vitis AI PyXIR version to v0.3.1

jtuyls commented on pull request #8814: URL: https://github.com/apache/tvm/pull/8814#issuecomment-907293887 @leandron I can add the dependencies to the `gen_requirements.py`. I am not familiar with how it's set up but I guess it would look like this? ``` ( "vitis-ai",

[GitHub] [tvm] luyaor opened a new pull request #8866: Set default value of p in LpPool as 2

luyaor opened a new pull request #8866: URL: https://github.com/apache/tvm/pull/8866 @jwfromm Sorry for very late update. Followed https://github.com/apache/tvm/pull/7242. I have already formatted the code based on https://github.com/apache/tvm/pull/7242#issuecomment-792789679.

[GitHub] [tvm] mikepapadim commented on pull request #8861: [Docker] Re-enabled automatic --tty flag when running bash.

mikepapadim commented on pull request #8861: URL: https://github.com/apache/tvm/pull/8861#issuecomment-907311405 @Lunderberg that's my bad. I split them for the pre-commit hooks to run through docker, but it needs only `tty` session to open. -- This is an automated message from the

[GitHub] [tvm-vta] vegaluisjose commented on pull request #32: VTA Chisel Wide memory interface.

vegaluisjose commented on pull request #32: URL: https://github.com/apache/tvm-vta/pull/32#issuecomment-907308252 could you push en empty commit to see if we can re-trigger CI? Thanks! @aasorokiin -- This is an automated message from the Apache Git Service. To respond to the

[GitHub] [tvm] AndrewZhaoLuo edited a comment on pull request #8862: [Autoscheduler] Change workload keys

AndrewZhaoLuo edited a comment on pull request #8862: URL: https://github.com/apache/tvm/pull/8862#issuecomment-907360364 I think I'll do this solution you describe. I will also change the default implementation to use the md5 hash but also produce a new file which associates the hashes

[GitHub] [tvm] comaniac commented on pull request #8862: [Autoscheduler] Change workload keys

comaniac commented on pull request #8862: URL: https://github.com/apache/tvm/pull/8862#issuecomment-907361893 Sounds like a plan to me. -- 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

[GitHub] [tvm] shingjan commented on a change in pull request #8851: [AutoTVM] Use popenpool in local_executor

shingjan commented on a change in pull request #8851: URL: https://github.com/apache/tvm/pull/8851#discussion_r697621817 ## File path: python/tvm/autotvm/measure/executor.py ## @@ -87,6 +87,25 @@ def get(self, timeout=None): raise NotImplementedError() +class

[GitHub] [tvm] comaniac commented on a change in pull request #8702: [Runtime] Pipeline Executor Initial patch.

comaniac commented on a change in pull request #8702: URL: https://github.com/apache/tvm/pull/8702#discussion_r697595674 ## File path: CMakeLists.txt ## @@ -388,6 +388,21 @@ if(GTEST_INCLUDE_DIR AND GTEST_LIB) include(GoogleTest) endif() +if(USE_PIPELINE_EXECUTOR) +

[GitHub] [tvm] yzhliu opened a new pull request #8868: [Community] @mdw-octoml -> Reviewer

yzhliu opened a new pull request #8868: URL: https://github.com/apache/tvm/pull/8868 Please join us to welcome @mdw-octoml as a new reviewer to TVM. Matt has contributed to microTVM, TVMC, etc. and has provided valuable feedback and suggestions to pull requests. - [Commits

[GitHub] [tvm] huajsj commented on pull request #8702: [Runtime] Pipeline Executor Initial patch.

huajsj commented on pull request #8702: URL: https://github.com/apache/tvm/pull/8702#issuecomment-907341646 @comaniac, the RFC design already applied into this PR, if you have time please review again. -- This is an automated message from the Apache Git Service. To respond to the

[GitHub] [tvm] Lunderberg edited a comment on pull request #8861: [Docker] Re-enabled automatic --tty flag when running bash.

Lunderberg edited a comment on pull request #8861: URL: https://github.com/apache/tvm/pull/8861#issuecomment-907346689 No problem at all, and sorry if that came across as finger-pointing. I like to link to previous commits/PRs where possible, so that it's easier to do code archaeology in

[GitHub] [tvm] Lunderberg commented on pull request #8861: [Docker] Re-enabled automatic --tty flag when running bash.

Lunderberg commented on pull request #8861: URL: https://github.com/apache/tvm/pull/8861#issuecomment-907346689 No problem at all, and sorry if that came across as finger-pointing. I like to link to previous commits/PRs where possible, so that it's easier to track when things were

[GitHub] [tvm] AndrewZhaoLuo commented on pull request #8862: [Autoscheduler] Change workload keys

AndrewZhaoLuo commented on pull request #8862: URL: https://github.com/apache/tvm/pull/8862#issuecomment-907360364 I think I'll do this solution you describe. I will also change the default implementation to use the md5 hash but also produce a new file which associates the hashes with the

[GitHub] [tvm] junrushao1994 commented on pull request #8862: [Autoscheduler] Change workload keys

junrushao1994 commented on pull request #8862: URL: https://github.com/apache/tvm/pull/8862#issuecomment-907361064 @comaniac @jcf94 let me know what you guys think -- 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-vta] aasorokiin commented on pull request #35: Jenkinsfile ci pipeline fix.

aasorokiin commented on pull request #35: URL: https://github.com/apache/tvm-vta/pull/35#issuecomment-907364591 This request has the same issue. @vegaluisjose -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL

[GitHub] [tvm] jroesch commented on pull request #5519: [RFC] Make tflite frontend more data driven / improve errors.

jroesch commented on pull request #5519: URL: https://github.com/apache/tvm/pull/5519#issuecomment-907387614 @u99127 I am doing triage on old PRs, going to close this, please feel free to follow up if you would like to still merge these changes. Thanks for your contributions!. -- This

[GitHub] [tvm] jroesch closed pull request #5519: [RFC] Make tflite frontend more data driven / improve errors.

jroesch closed pull request #5519: URL: https://github.com/apache/tvm/pull/5519 -- 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] jroesch closed pull request #5528: POC refactor tflite frontend

jroesch closed pull request #5528: URL: https://github.com/apache/tvm/pull/5528 -- 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] jroesch commented on pull request #6200: [WIP] [AutoQ] Framework coverage, Avg min max calibration, Bias add handling

jroesch commented on pull request #6200: URL: https://github.com/apache/tvm/pull/6200#issuecomment-907388104 I am doing triage on old PRs, going to close this, please feel free to follow up if you would like to still merge these changes. Thanks for your contributions!. -- This is an

[GitHub] [tvm] jroesch closed pull request #6200: [WIP] [AutoQ] Framework coverage, Avg min max calibration, Bias add handling

jroesch closed pull request #6200: URL: https://github.com/apache/tvm/pull/6200 -- 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] jroesch commented on pull request #6219: [Runtime][WIP] Add prototype Relay AoT compiler directly into TVM

jroesch commented on pull request #6219: URL: https://github.com/apache/tvm/pull/6219#issuecomment-907388307 I am doing triage on old PRs, going to close this, please feel free to follow up if you would like to still merge these changes. Thanks for your contributions!. -- This is an

[GitHub] [tvm] jroesch closed pull request #6871: [Rust] Revive wasm32 test

jroesch closed pull request #6871: URL: https://github.com/apache/tvm/pull/6871 -- 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] jroesch commented on pull request #6871: [Rust] Revive wasm32 test

jroesch commented on pull request #6871: URL: https://github.com/apache/tvm/pull/6871#issuecomment-907391599 I am doing triage on old PRs, going to close this, please feel free to follow up if you would like to still merge these changes. Thanks for your contributions! @kazum I

[GitHub] [tvm] jroesch commented on pull request #6435: [Training][Bugfix][VM] Fix reshape optimization being too agressive and is unsound on constant tensor.

jroesch commented on pull request #6435: URL: https://github.com/apache/tvm/pull/6435#issuecomment-907389662 I am doing triage on old PRs, going to close this, please feel free to follow up if you would like to still merge these changes. Thanks for your contributions!. @altanh did a

[GitHub] [tvm] jroesch closed pull request #6219: [Runtime][WIP] Add prototype Relay AoT compiler directly into TVM

jroesch closed pull request #6219: URL: https://github.com/apache/tvm/pull/6219 -- 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] jroesch closed pull request #6779: [WIP] Object Schema

jroesch closed pull request #6779: URL: https://github.com/apache/tvm/pull/6779 -- 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] jroesch commented on pull request #5528: POC refactor tflite frontend

jroesch commented on pull request #5528: URL: https://github.com/apache/tvm/pull/5528#issuecomment-907387792 I am doing triage on old PRs, going to close this, please feel free to follow up if you would like to still merge these changes. Thanks for your contributions!. -- This is an

[GitHub] [tvm] jroesch closed pull request #6435: [Training][Bugfix][VM] Fix reshape optimization being too agressive and is unsound on constant tensor.

jroesch closed pull request #6435: URL: https://github.com/apache/tvm/pull/6435 -- 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] jwfromm opened a new pull request #8867: [Frontend][Onnx] Simplify onnx input since name accesses are no longer supported.

jwfromm opened a new pull request #8867: URL: https://github.com/apache/tvm/pull/8867 A while back @mbrookhart and I implemented the `onnx_input` class to support accessing inputs by index or name. However, it turned out that onnx node names are inconsistent and shouldnt be used. Although

[GitHub] [tvm] jroesch closed pull request #6688: [TUTORIAL][FIX] Fix VTA autotuning from tutorial fails with one P…

jroesch closed pull request #6688: URL: https://github.com/apache/tvm/pull/6688 -- 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] guberti commented on pull request #8748: [microTVM][RVM] Add Arduino RVM

guberti commented on pull request #8748: URL: https://github.com/apache/tvm/pull/8748#issuecomment-907419125 > I built the image and tested it against Arduino Due. It fails on two tests, currently waiting for fixing those tests before merging. Those failures are due to a regression

[GitHub] [tvm-vta] aasorokiin opened a new pull request #35: Jenkinsfile ci pipeline fix.

aasorokiin opened a new pull request #35: URL: https://github.com/apache/tvm-vta/pull/35 Update CI docker -- 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] vinx13 commented on a change in pull request #8851: [AutoTVM] Use popenpool in local_executor

vinx13 commented on a change in pull request #8851: URL: https://github.com/apache/tvm/pull/8851#discussion_r697625308 ## File path: tests/python/unittest/test_autotvm_measure.py ## @@ -76,7 +76,7 @@ def submit(self, func, *args, **kwargs): self.ran_dummy_executor

[GitHub] [tvm] jroesch commented on pull request #6779: [WIP] Object Schema

jroesch commented on pull request #6779: URL: https://github.com/apache/tvm/pull/6779#issuecomment-907389275 I am doing triage on old PRs, going to close this, please feel free to follow up if you would like to still merge these changes. Thanks for your contributions!. re this

[GitHub] [tvm] jroesch commented on pull request #7006: [frontend][keras] Add support for TimeDistributed

jroesch commented on pull request #7006: URL: https://github.com/apache/tvm/pull/7006#issuecomment-907393964 @jwfromm @mbrookhart can you guys try to land this? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL

[GitHub] [tvm] AndrewZhaoLuo commented on pull request #8862: [Autoscheduler] Configurable workload keys

AndrewZhaoLuo commented on pull request #8862: URL: https://github.com/apache/tvm/pull/8862#issuecomment-907403818 PTAL @junrushao1994 @comaniac -- 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] guberti opened a new pull request #8869: Remove Arduino aot code

guberti opened a new pull request #8869: URL: https://github.com/apache/tvm/pull/8869 PR #8758 changed the way AOT is handled in microTVM, which caused my Arduino PR to stop working. This should have been caught by the unit tests since it caused Arduino project compilation to stop

[GitHub] [tvm] shingjan commented on a change in pull request #8851: [AutoTVM] Use popenpool in local_executor

shingjan commented on a change in pull request #8851: URL: https://github.com/apache/tvm/pull/8851#discussion_r697678994 ## File path: tests/python/unittest/test_autotvm_measure.py ## @@ -76,7 +76,7 @@ def submit(self, func, *args, **kwargs):

[GitHub] [tvm-vta] vegaluisjose commented on a change in pull request #32: VTA Chisel Wide memory interface.

vegaluisjose commented on a change in pull request #32: URL: https://github.com/apache/tvm-vta/pull/32#discussion_r697591552 ## File path: Jenkinsfile ## @@ -21,9 +21,13 @@ // Jenkins pipeline // See documents at https://jenkins.io/doc/book/pipeline/jenkinsfile/ -ci_lint =

[GitHub] [tvm-vta] vegaluisjose commented on pull request #32: VTA Chisel Wide memory interface.

vegaluisjose commented on pull request #32: URL: https://github.com/apache/tvm-vta/pull/32#issuecomment-907357263 Oh I see, what is going on now. Can you do this docker update on a separate PR? we will merge that before this one, so it can take the changes. -- This is an automated

[GitHub] [tvm] jroesch commented on pull request #6688: [TUTORIAL][FIX] Fix VTA autotuning from tutorial fails with one P…

jroesch commented on pull request #6688: URL: https://github.com/apache/tvm/pull/6688#issuecomment-907389907 I am doing triage on old PRs, going to close this, please feel free to follow up if you would like to still merge these changes. Thanks for your contributions!. -- This is an

[GitHub] [tvm] jwfromm commented on pull request #8867: [Frontend][Onnx] Simplify onnx input since name accesses are no longer supported.

jwfromm commented on pull request #8867: URL: https://github.com/apache/tvm/pull/8867#issuecomment-907390358 @cconvey can you take a look at this? -- 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

[GitHub] [tvm] jroesch commented on pull request #7495: [Relay VM] VM Profiler Supports Profiling Instruction Stats

jroesch commented on pull request #7495: URL: https://github.com/apache/tvm/pull/7495#issuecomment-907395484 I am doing triage on old PRs, going to close this, please feel free to follow up if you would like to still merge these changes. Thanks for your contributions! It looks like

[GitHub] [tvm] jroesch closed pull request #7495: [Relay VM] VM Profiler Supports Profiling Instruction Stats

jroesch closed pull request #7495: URL: https://github.com/apache/tvm/pull/7495 -- 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] tmoreau89 opened a new pull request #8870: [Community] @manupa-arm -> Committer

tmoreau89 opened a new pull request #8870: URL: https://github.com/apache/tvm/pull/8870 Please join us to welcome @manupa-arm as a new committer to TVM. Manupa has contributed to the AOT and Ethos-U efforts at ARM and proposed and prototyped designs for improved embedded-friendly memory

[GitHub] [tvm] electriclilies commented on pull request #8835: Change target string to Target object in the TE compiler and interpreter

electriclilies commented on pull request #8835: URL: https://github.com/apache/tvm/pull/8835#issuecomment-907513588 @mbs-octoml @junrushao1994 @Mousius I implemented what Junru suggested-- just using `std::unordered_map` instead of` tvm::Map` in a few places. It looks like this will go

[tvm-vta] branch ci-docker-staging updated: update cmake variable

This is an automated email from the ASF dual-hosted git repository. vega pushed a commit to branch ci-docker-staging in repository https://gitbox.apache.org/repos/asf/tvm-vta.git The following commit(s) were added to refs/heads/ci-docker-staging by this push: new ea6829f update cmake

[tvm-vta] branch ci-docker-staging updated: update llvm

This is an automated email from the ASF dual-hosted git repository. vega pushed a commit to branch ci-docker-staging in repository https://gitbox.apache.org/repos/asf/tvm-vta.git The following commit(s) were added to refs/heads/ci-docker-staging by this push: new 9f8a46e update llvm

  1   2   >