[GitHub] [tvm] junrushao1994 merged pull request #12344: [TVMScript] Text underlining in DocPrinter based on Doc's source_paths

junrushao1994 merged PR #12344: URL: https://github.com/apache/tvm/pull/12344 -- 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: commits-unsubscr...@tvm.ap

[tvm] branch main updated: [TVMScript] Text underlining in DocPrinter based on Doc's source_paths (#12344)

This is an automated email from the ASF dual-hosted git repository. junrushao 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 f5f5a75ae9 [TVMScript] Text underlining in DocPrinte

[GitHub] [tvm] junrushao1994 commented on pull request #12344: [TVMScript] Text underlining in DocPrinter based on Doc's source_paths

junrushao1994 commented on PR #12344: URL: https://github.com/apache/tvm/pull/12344#issuecomment-1211572753 Thanks for the PR! It's merged :-) -- 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] cxx122 opened a new issue, #12374: [Bug] Core dump caused by truncdiv and min_value of int64

cxx122 opened a new issue, #12374: URL: https://github.com/apache/tvm/issues/12374 ``` tensor_0 = te.placeholder([5], dtype="int8", name="TENSOR_0") tensor_1 = te.compute([5], lambda lcg:te.truncdiv(te.min_value("int64"), tensor_0[lcg]), name ="TENSOR_1") ``` ### Expected beh

[tvm] branch ir-builder updated (c0df9cf626 -> f1065cd751)

This is an automated email from the ASF dual-hosted git repository. junrushao pushed a change to branch ir-builder in repository https://gitbox.apache.org/repos/asf/tvm.git discard c0df9cf626 Squashed new f1065cd751 Squashed This update added new revisions after undoing existing revisions

[GitHub] [tvm] cxx122 opened a new issue, #12373: [Bug] Misuse of reduce_axis

cxx122 opened a new issue, #12373: URL: https://github.com/apache/tvm/issues/12373 ``` REDUCE_0 = te.reduce_axis((5, 0), name="REDUCE_0") REDUCE_1 = te.reduce_axis((7, 0), name="REDUCE_1") TENSOR_0 = te.placeholder([10,10,10,10], dtype="int8", name="TENSOR_0") TENSOR_1 = te.comput

[GitHub] [tvm] github-actions[bot] commented on pull request #12369: Add ci_riscv image, update all to 20220810-060142-fae79bbc3

github-actions[bot] commented on PR #12369: URL: https://github.com/apache/tvm/pull/12369#issuecomment-1211554418 Built docs for commit 41fd385fe680c4e9bc26e89032a19b47dcd917d5 can be found [here](https://pr-docs.tlcpack.ai/PR-12369/2/docs/index.html). -- This is an automated mess

[GitHub] [tvm] cxx122 opened a new issue, #12372: [Bug] Array out-of-bounds access of TE language

cxx122 opened a new issue, #12372: URL: https://github.com/apache/tvm/issues/12372 ``` tensor_1 = te.placeholder((2,2,), dtype="int16", name="A") tensor_2 = te.compute((2,3,), lambda dj,z,: tensor_1[z][dj], name ="IR") ``` ### Expected behavior An error message or a def

[tvm] branch last-successful updated (06ac5cde84 -> 1a3d36a28d)

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 06ac5cde84 [testing] Remove wrapper from @slow (#11566) add 1a3d36a28d [ci] Test pytest-forked boxing (

[tvm] branch ir-builder updated (b9688898e3 -> c0df9cf626)

This is an automated email from the ASF dual-hosted git repository. junrushao pushed a change to branch ir-builder in repository https://gitbox.apache.org/repos/asf/tvm.git discard b9688898e3 Squashed new c0df9cf626 Squashed This update added new revisions after undoing existing revisions

[GitHub] [tvm] yangulei commented on pull request #12371: [BYOC] [DNNL] enable in-place post-op sum in dnnl json runtime

yangulei commented on PR #12371: URL: https://github.com/apache/tvm/pull/12371#issuecomment-1211481211 @apeskov This is more like a work-around before the true solution based on memory planning is ready. -- This is an automated message from the Apache Git Service. To respond to the messag

[GitHub] [tvm] yangulei opened a new pull request, #12371: [BYOC] [DNNL] enable in-place post-op sum in dnnl json runtime

yangulei opened a new pull request, #12371: URL: https://github.com/apache/tvm/pull/12371 This PR enable in-place post-op sum in dnnl json runtime for performance concern. This feature only enabled if the `rhs` of `sum` is not `input` nor `output` of current dnnl partition. Thus the workloa

[tvm] branch ir-builder updated (d42b89972a -> b9688898e3)

This is an automated email from the ASF dual-hosted git repository. junrushao pushed a change to branch ir-builder in repository https://gitbox.apache.org/repos/asf/tvm.git discard d42b89972a Squashed new b9688898e3 Squashed This update added new revisions after undoing existing revisions

[GitHub] [tvm] yangulei opened a new pull request, #12370: [TIR] [bfloat16] add bfloat16 promotion for CallNode

yangulei opened a new pull request, #12370: URL: https://github.com/apache/tvm/pull/12370 So far, the `softmax` in the following test ``` python import tvm from tvm import relay data = relay.var("data", shape=(1, 1000), dtype="bfloat16") expr = relay.nn.softmax(data) mod = t

[tvm] branch main updated (06ac5cde84 -> 1a3d36a28d)

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 06ac5cde84 [testing] Remove wrapper from @slow (#11566) add 1a3d36a28d [ci] Test pytest-forked boxing (#12312) No ne

[GitHub] [tvm] areusch merged pull request #12312: [ci] Box all tests in pytest-forked

areusch merged PR #12312: URL: https://github.com/apache/tvm/pull/12312 -- 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: commits-unsubscr...@tvm.apache.o

[GitHub] [tvm] areusch commented on pull request #12369: Add ci_riscv image, update all to 20220810-060142-fae79bbc3

areusch commented on PR #12369: URL: https://github.com/apache/tvm/pull/12369#issuecomment-1211441610 progresses #11688 -- 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] areusch opened a new pull request, #12369: Add ci_riscv image, update all to 20220810-060142-fae79bbc3

areusch opened a new pull request, #12369: URL: https://github.com/apache/tvm/pull/12369 This PR updates all images to the most recent nightly build: https://ci.tlcpack.ai/blue/organizations/jenkins/docker-images-ci%2Fdaily-docker-image-rebuild/detail/daily-docker-image-rebuild/419/pipeline/

[GitHub] [tvm] areusch commented on pull request #11579: [Runtime] Add runtime.contrib.set_custom_cpu_allocator to use custom memalign() and free() functions

areusch commented on PR #11579: URL: https://github.com/apache/tvm/pull/11579#issuecomment-1211437719 @jsheng-jian could you address the question above? -- 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

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

areusch commented on PR #12028: URL: https://github.com/apache/tvm/pull/12028#issuecomment-1211437054 thanks @blackkker and sorry for yet another delay. i tried to resolve the merge conflicts myself here, hopefully it goes green and can merge. -- This is an automated message from the Apac

[tvm] branch last-successful updated (cd6d19f481 -> 06ac5cde84)

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 cd6d19f481 [ci][tvmbot] Fix authorization filtering (#12310) add 06ac5cde84 [testing] Remove wrapper fr

[GitHub] [tvm] github-actions[bot] commented on pull request #12367: [ci] Add linter for PR title and body

github-actions[bot] commented on PR #12367: URL: https://github.com/apache/tvm/pull/12367#issuecomment-1211427586 Built docs for commit 68cfdc1733f66c22ae88fee094f4b269dbc51eed can be found [here](https://pr-docs.tlcpack.ai/PR-12367/8/docs/index.html). -- This is an automated mess

[tvm] branch ir-builder updated (b1bbac798c -> d42b89972a)

This is an automated email from the ASF dual-hosted git repository. junrushao pushed a change to branch ir-builder in repository https://gitbox.apache.org/repos/asf/tvm.git discard b1bbac798c Squashed new d42b89972a Squashed This update added new revisions after undoing existing revisions

[tvm] branch ir-builder updated (18ab55fb71 -> b1bbac798c)

This is an automated email from the ASF dual-hosted git repository. junrushao pushed a change to branch ir-builder in repository https://gitbox.apache.org/repos/asf/tvm.git discard 18ab55fb71 Squashed new b1bbac798c Squashed This update added new revisions after undoing existing revisions

[GitHub] [tvm] farshidsp opened a new pull request, #12368: Farshidsp/hexagon/add parallelism

farshidsp opened a new pull request, #12368: URL: https://github.com/apache/tvm/pull/12368 Adding flag to `pytest` to skip creating a new RPC server on the android device. @mehrdadh -- This is an automated message from the Apache Git Service. To respond to the message, please log

[tvm] branch main updated (cd6d19f481 -> 06ac5cde84)

This is an automated email from the ASF dual-hosted git repository. driazati pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from cd6d19f481 [ci][tvmbot] Fix authorization filtering (#12310) add 06ac5cde84 [testing] Remove wrapper from @slow (#11

[GitHub] [tvm] driazati merged pull request #11566: [testing] Remove wrapper from @slow

driazati merged PR #11566: URL: https://github.com/apache/tvm/pull/11566 -- 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: commits-unsubscr...@tvm.apache.

[GitHub] [tvm] slyubomirsky commented on issue #12362: [Bug][Relay] TVM generates input name that cannot be imported with the parser

slyubomirsky commented on issue #12362: URL: https://github.com/apache/tvm/issues/12362#issuecomment-1211286866 Can you write a simple pass that will change ("sanitize") the variable names? I don't know if a colon is treated in any special way by the parser, but you could try replacing them

[GitHub] [tvm] tkonolige commented on a diff in pull request #12352: [TIR] Add pass to check for out of bounds memory access

tkonolige commented on code in PR #12352: URL: https://github.com/apache/tvm/pull/12352#discussion_r942888251 ## src/tir/analysis/oob_checker.cc: ## @@ -0,0 +1,122 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements.

[GitHub] [tvm] tkonolige commented on a diff in pull request #12352: [TIR] Add pass to check for out of bounds memory access

tkonolige commented on code in PR #12352: URL: https://github.com/apache/tvm/pull/12352#discussion_r942888251 ## src/tir/analysis/oob_checker.cc: ## @@ -0,0 +1,122 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements.

[GitHub] [tvm] yelite opened a new pull request, #12366: [TVMScript] Printer Frame

yelite opened a new pull request, #12366: URL: https://github.com/apache/tvm/pull/12366 This PR: - Implement Frame for the TVMScript Unified Printer Compared to the prototype version, this: - Removes the dependency of VarTable (SymbolTable) from Frame - Adds a callback

[GitHub] [tvm] github-actions[bot] commented on pull request #12361: [ci][tvmbot] Trigger GitHub Actions after merging

github-actions[bot] commented on PR #12361: URL: https://github.com/apache/tvm/pull/12361#issuecomment-1211240890 Built docs for commit 23f2113d7acd6e1205141bf5f6ca694310446c95 can be found [here](https://pr-docs.tlcpack.ai/PR-12361/3/docs/index.html). -- This is an automated mess

[tvm] 01/01: fix assert

This is an automated email from the ASF dual-hosted git repository. andrewzhaoluo pushed a commit to branch aluo/fix-int8-tensorcore in repository https://gitbox.apache.org/repos/asf/tvm.git commit 7eab14aace2c76cda9c4c1ee99e4c40f87c5d81c Author: Andrew Zhao Luo AuthorDate: Wed Aug 10 13:31:44 2

[tvm] branch aluo/fix-int8-tensorcore created (now 7eab14aace)

This is an automated email from the ASF dual-hosted git repository. andrewzhaoluo pushed a change to branch aluo/fix-int8-tensorcore in repository https://gitbox.apache.org/repos/asf/tvm.git at 7eab14aace fix assert This branch includes the following new commits: new 7eab14aace fix

[GitHub] [tvm] tkonolige opened a new pull request, #12364: [LowerVTCMAlloc] Move LowerVtcmAlloc to after StorageRewrite

tkonolige opened a new pull request, #12364: URL: https://github.com/apache/tvm/pull/12364 Vtcm allocations were being moved inside loops even if they were originally allocated outside of the loops. Normally PlanAndUpdateBufferAllocationLocation moves allocations as close to use as possible

[GitHub] [tvm] Lunderberg commented on a diff in pull request #12352: [TIR] Add pass to check for out of bounds memory access

Lunderberg commented on code in PR #12352: URL: https://github.com/apache/tvm/pull/12352#discussion_r942866132 ## src/tir/analysis/oob_checker.cc: ## @@ -0,0 +1,122 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements.

[tvm] branch ir-builder updated (66c2087bd0 -> 18ab55fb71)

This is an automated email from the ASF dual-hosted git repository. junrushao pushed a change to branch ir-builder in repository https://gitbox.apache.org/repos/asf/tvm.git discard 66c2087bd0 Squashed new 18ab55fb71 Squashed This update added new revisions after undoing existing revisions

[tvm] branch ir-builder updated (e36f448b62 -> 66c2087bd0)

This is an automated email from the ASF dual-hosted git repository. junrushao pushed a change to branch ir-builder in repository https://gitbox.apache.org/repos/asf/tvm.git discard e36f448b62 Squashed add fc411dc6fa Pass that removes reshapes post LowerTE (#12215) add 9b86009553 [Open

[GitHub] [tvm] gbonik commented on a diff in pull request #12344: [TVMScript] Text underlining in DocPrinter based on Doc's source_paths

gbonik commented on code in PR #12344: URL: https://github.com/apache/tvm/pull/12344#discussion_r942843203 ## src/script/printer/base_doc_printer.cc: ## @@ -23,19 +23,256 @@ namespace tvm { namespace script { namespace printer { -DocPrinter::DocPrinter(int indent_spaces) : i

[GitHub] [tvm] gbonik commented on a diff in pull request #12344: [TVMScript] Text underlining in DocPrinter based on Doc's source_paths

gbonik commented on code in PR #12344: URL: https://github.com/apache/tvm/pull/12344#discussion_r942842429 ## src/script/printer/base_doc_printer.cc: ## @@ -23,19 +23,256 @@ namespace tvm { namespace script { namespace printer { -DocPrinter::DocPrinter(int indent_spaces) : i

[GitHub] [tvm] yelite commented on a diff in pull request #12344: [TVMScript] Text underlining in DocPrinter based on Doc's source_paths

yelite commented on code in PR #12344: URL: https://github.com/apache/tvm/pull/12344#discussion_r942838191 ## src/script/printer/base_doc_printer.cc: ## @@ -23,19 +23,256 @@ namespace tvm { namespace script { namespace printer { -DocPrinter::DocPrinter(int indent_spaces) : i

[GitHub] [tvm] gbonik commented on a diff in pull request #12344: [TVMScript] Text underlining in DocPrinter based on Doc's source_paths

gbonik commented on code in PR #12344: URL: https://github.com/apache/tvm/pull/12344#discussion_r942807225 ## src/script/printer/base_doc_printer.cc: ## @@ -23,19 +23,256 @@ namespace tvm { namespace script { namespace printer { -DocPrinter::DocPrinter(int indent_spaces) : i

[GitHub] [tvm] tkonolige closed pull request #10474: [TOPHUB] Remove tophub context from VM and Graph Executor

tkonolige closed pull request #10474: [TOPHUB] Remove tophub context from VM and Graph Executor URL: https://github.com/apache/tvm/pull/10474 -- 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 spec

[GitHub] [tvm] tkonolige closed pull request #12084: [FIX,TIR] Handle allocations of variable size

tkonolige closed pull request #12084: [FIX,TIR] Handle allocations of variable size URL: https://github.com/apache/tvm/pull/12084 -- 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] cconvey commented on a diff in pull request #12340: [TOPI][Hexagon] Implement quantized avgpool

cconvey commented on code in PR #12340: URL: https://github.com/apache/tvm/pull/12340#discussion_r942794370 ## python/tvm/topi/hexagon/qnn/avg_pool2d.py: ## @@ -0,0 +1,180 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements.

[GitHub] [tvm] vinx13 commented on a diff in pull request #12134: [PyTorch] Fix pad_common for float pad_value

vinx13 commented on code in PR #12134: URL: https://github.com/apache/tvm/pull/12134#discussion_r942785399 ## python/tvm/relay/frontend/pytorch.py: ## @@ -1925,16 +1925,25 @@ def pad_common(self, mode, pad_value, inputs, input_types): for pad in paddings:

[GitHub] [tvm] cconvey commented on a diff in pull request #12204: [Hexagon] Add Hand written HVX conv2d

cconvey commented on code in PR #12204: URL: https://github.com/apache/tvm/pull/12204#discussion_r940401262 ## src/runtime/hexagon/ops/conv2d_hvx.cc: ## @@ -0,0 +1,468 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreement

[tvm] branch last-successful updated (22ba659438 -> cd6d19f481)

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 22ba659438 [BYOC-DNNL] add partition test on sum pattern (#12357) add cd6d19f481 [ci][tvmbot] Fix autho

[GitHub] [tvm] junrushao1994 commented on issue #10317: Auto-cc subscription tracker

junrushao1994 commented on issue #10317: URL: https://github.com/apache/tvm/issues/10317#issuecomment-1211090372 metaschedule @junrushao1994 arith @junrushao1994 tir @junrushao1994 -- This is an automated message from the Apache Git Service. To respond to the message, please log on

[GitHub] [tvm] driazati commented on pull request #12361: [ci][tvmbot] Trigger GitHub Actions after merging

driazati commented on PR #12361: URL: https://github.com/apache/tvm/pull/12361#issuecomment-1211083237 @tvm-bot rerun -- 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 uns

[GitHub] [tvm] shingjan commented on a diff in pull request #12134: [PyTorch] Fix pad_common for float pad_value

shingjan commented on code in PR #12134: URL: https://github.com/apache/tvm/pull/12134#discussion_r942757423 ## python/tvm/relay/frontend/pytorch.py: ## @@ -1925,16 +1925,25 @@ def pad_common(self, mode, pad_value, inputs, input_types): for pad in paddings:

[GitHub] [tvm] csullivan commented on pull request #12204: [Hexagon] Add Hand written HVX conv2d

csullivan commented on PR #12204: URL: https://github.com/apache/tvm/pull/12204#issuecomment-1211072314 Also @quic-sanirudh, if @kparzysz-quic is a co-author I suggest amending the commit message with `Co-authored-by: Krzysztof Parzyszek ` which will attribute mutual authorship in Git and

[GitHub] [tvm] csullivan commented on a diff in pull request #12204: [Hexagon] Add Hand written HVX conv2d

csullivan commented on code in PR #12204: URL: https://github.com/apache/tvm/pull/12204#discussion_r942612681 ## src/runtime/hexagon/ops/conv2d_hvx.cc: ## @@ -0,0 +1,468 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreeme

[GitHub] [tvm] driazati commented on pull request #12356: [TIR] Minor fix to tensor intrin description

driazati commented on PR #12356: URL: https://github.com/apache/tvm/pull/12356#issuecomment-1211068960 > I wonder why github automatically tagged gigiblender for a bit he was signed up to track tir on #10317 -- This is an automated message from the Apache Git Service. To respond to

[GitHub] [tvm] mehrdadh opened a new issue, #12362: [Bug][Relay] TVM generates input name that cannot be imported with the parser

mehrdadh opened a new issue, #12362: URL: https://github.com/apache/tvm/issues/12362 I'm compiling an ONNX model which I realized the input name has this format `input:0`. When I import the model and run it, it works fine, however if I try to import the same Relay text with the parser it wi

[GitHub] [tvm] jverma-quic commented on a diff in pull request #12340: [TOPI][Hexagon] Implement quantized avgpool

jverma-quic commented on code in PR #12340: URL: https://github.com/apache/tvm/pull/12340#discussion_r942649464 ## python/tvm/topi/hexagon/qnn/avg_pool2d.py: ## @@ -0,0 +1,180 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreeme

[GitHub] [tvm] driazati opened a new pull request, #12361: [ci][tvmbot] Trigger GitHub Actions after merging

driazati opened a new pull request, #12361: URL: https://github.com/apache/tvm/pull/12361 This fixes the issue where merging from GitHub Actions (i.e. with the default `GITHUB_TOKEN`) doesn't trigger post-merge GitHub Actions on the commit it creates in `main`. Instead these jobs are tr

[GitHub] [tvm] driazati commented on pull request #12178: Build and test TVM under minimal configuration

driazati commented on PR #12178: URL: https://github.com/apache/tvm/pull/12178#issuecomment-1210931462 @tvm-bot merge -- 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 uns

[tvm] branch main updated: [ci][tvmbot] Fix authorization filtering (#12310)

This is an automated email from the ASF dual-hosted git repository. areusch 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 cd6d19f481 [ci][tvmbot] Fix authorization filtering (#

[GitHub] [tvm] areusch merged pull request #12310: [ci][tvmbot] Fix authorization filtering

areusch merged PR #12310: URL: https://github.com/apache/tvm/pull/12310 -- 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: commits-unsubscr...@tvm.apache.o

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

tqchen commented on PR #18: URL: https://github.com/apache/tvm-rfcs/pull/18#issuecomment-1210920580 @areusch this can be something that is internal in TVM for now given this is only used as a compiler abstraction not runtime -- This is an automated message from the Apache Git Service. To

[GitHub] [tvm] ashutosh-arm commented on a diff in pull request #12353: [CMSIS-NN] Pad fusion with QNN Conv2D

ashutosh-arm commented on code in PR #12353: URL: https://github.com/apache/tvm/pull/12353#discussion_r942622944 ## python/tvm/relay/op/contrib/cmsisnn.py: ## @@ -122,6 +130,25 @@ def check_qnn_conv2d(pattern): conv2d_input = conv2d.args[0] conv2d_weight = conv

[GitHub] [tvm] ashutosh-arm commented on a diff in pull request #12353: [CMSIS-NN] Pad fusion with QNN Conv2D

ashutosh-arm commented on code in PR #12353: URL: https://github.com/apache/tvm/pull/12353#discussion_r942621766 ## tests/python/contrib/test_cmsisnn/test_fuse_pads.py: ## @@ -0,0 +1,285 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor lice

[GitHub] [tvm] ashutosh-arm commented on a diff in pull request #12353: [CMSIS-NN] Pad fusion with QNN Conv2D

ashutosh-arm commented on code in PR #12353: URL: https://github.com/apache/tvm/pull/12353#discussion_r942617075 ## tests/python/contrib/test_cmsisnn/test_fuse_pads.py: ## @@ -0,0 +1,285 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor lice

[GitHub] [tvm] ashutosh-arm commented on a diff in pull request #12353: [CMSIS-NN] Pad fusion with QNN Conv2D

ashutosh-arm commented on code in PR #12353: URL: https://github.com/apache/tvm/pull/12353#discussion_r942615935 ## tests/python/contrib/test_cmsisnn/test_conv2d.py: ## @@ -403,6 +401,109 @@ def test_conv2d_asymmetric_padding_int8( ) +@tvm.testing.requires_cmsisnn +@pyt

[GitHub] [tvm] SebastianBoblest commented on a diff in pull request #12353: [CMSIS-NN] Pad fusion with QNN Conv2D

SebastianBoblest commented on code in PR #12353: URL: https://github.com/apache/tvm/pull/12353#discussion_r942613452 ## python/tvm/relay/op/contrib/cmsisnn.py: ## @@ -122,6 +130,25 @@ def check_qnn_conv2d(pattern): conv2d_input = conv2d.args[0] conv2d_weight =

[GitHub] [tvm] ashutosh-arm commented on a diff in pull request #12353: [CMSIS-NN] Pad fusion with QNN Conv2D

ashutosh-arm commented on code in PR #12353: URL: https://github.com/apache/tvm/pull/12353#discussion_r942611235 ## python/tvm/relay/op/contrib/cmsisnn.py: ## @@ -122,6 +130,25 @@ def check_qnn_conv2d(pattern): conv2d_input = conv2d.args[0] conv2d_weight = conv

[GitHub] [tvm] driazati commented on pull request #12178: Build and test TVM under minimal configuration

driazati commented on PR #12178: URL: https://github.com/apache/tvm/pull/12178#issuecomment-1210879967 @tvm-bot merge -- 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 uns

[tvm] branch last-successful updated (4280d673f1 -> 22ba659438)

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 4280d673f1 [skip ci] [CI] Re-generate Jenkinsfile (#12360) add 22ba659438 [BYOC-DNNL] add partition tes

[GitHub] [tvm] SebastianBoblest commented on a diff in pull request #12353: [CMSIS-NN] Pad fusion with QNN Conv2D

SebastianBoblest commented on code in PR #12353: URL: https://github.com/apache/tvm/pull/12353#discussion_r942434869 ## python/tvm/relay/op/contrib/cmsisnn.py: ## @@ -122,6 +130,25 @@ def check_qnn_conv2d(pattern): conv2d_input = conv2d.args[0] conv2d_weight =

[tvm] branch last-successful updated (fae79bbc3e -> 4280d673f1)

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 fae79bbc3e [TVMScript] Printer Registry (#12237) add 22102063dc [ci] De-duplicate retry functions (#123

[tvm] branch main updated: [BYOC-DNNL] add partition test on sum pattern (#12357)

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 22ba659438 [BYOC-DNNL] add partition test on sum patter

[GitHub] [tvm] masahi merged pull request #12357: [BYOC-DNNL] add partition test on sum pattern

masahi merged PR #12357: URL: https://github.com/apache/tvm/pull/12357 -- 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: commits-unsubscr...@tvm.apache.or

[tvm] branch main updated: [skip ci] [CI] Re-generate Jenkinsfile (#12360)

This is an automated email from the ASF dual-hosted git repository. lukhut 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 4280d673f1 [skip ci] [CI] Re-generate Jenkinsfile (#123

[GitHub] [tvm] lhutton1 merged pull request #12360: [skip ci] [CI] Re-generate Jenkinsfile

lhutton1 merged PR #12360: URL: https://github.com/apache/tvm/pull/12360 -- 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: commits-unsubscr...@tvm.apache.

[GitHub] [tvm] Mousius opened a new pull request, #12360: [skip ci] [CI] Re-generate Jenkinsfile

Mousius opened a new pull request, #12360: URL: https://github.com/apache/tvm/pull/12360 Timing of merges resulted in the Jenkinsfile being out of sync -- 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 g

[GitHub] [tvm] Mousius commented on pull request #12346: [DOCS] fix `tvm.build` API doc layout.

Mousius commented on PR #12346: URL: https://github.com/apache/tvm/pull/12346#issuecomment-1210429438 Thanks for the fix @kuizhiqing, this is now merged 😸 -- 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

[tvm] branch main updated (5d72bc1a20 -> 7f100158a5)

This is an automated email from the ASF dual-hosted git repository. mousius pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from 5d72bc1a20 [ci][tvmbot] Ignore irrelevant Actions jobs (#12351) add 7f100158a5 [DOCS] Fix tvm.build API doc layout N

[GitHub] [tvm] Mousius merged pull request #12346: [DOCS] fix `tvm.build` API doc layout.

Mousius merged PR #12346: URL: https://github.com/apache/tvm/pull/12346 -- 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: commits-unsubscr...@tvm.apache.o

[GitHub] [tvm] Mousius merged pull request #12351: [ci][tvmbot] Ignore irrelevant Actions jobs

Mousius merged PR #12351: URL: https://github.com/apache/tvm/pull/12351 -- 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: commits-unsubscr...@tvm.apache.o

[tvm] branch main updated (22102063dc -> 5d72bc1a20)

This is an automated email from the ASF dual-hosted git repository. mousius pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from 22102063dc [ci] De-duplicate retry functions (#12325) add 5d72bc1a20 [ci][tvmbot] Ignore irrelevant Actions jobs (#12

[tvm] branch main updated: [ci] De-duplicate retry functions (#12325)

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 22102063dc [ci] De-duplicate retry functions (#12325)

[GitHub] [tvm] Mousius merged pull request #12325: [ci] De-duplicate retry functions

Mousius merged PR #12325: URL: https://github.com/apache/tvm/pull/12325 -- 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: commits-unsubscr...@tvm.apache.o

[GitHub] [tvm] AreopagX commented on issue #12359: [Frontend][ONNX][Tracking Issue] Add support for onnx::SequenceAt operator

AreopagX commented on issue #12359: URL: https://github.com/apache/tvm/issues/12359#issuecomment-1210371992 Related: #9116 -- 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] AreopagX opened a new issue, #12359: [Frontend][ONNX][Tracking Issue] Add support for onnx::SequenceAt operator

AreopagX opened a new issue, #12359: URL: https://github.com/apache/tvm/issues/12359 Hi all, I was trying to convert an ONNX model to TVM and the conversion fails. The error states: > tvm.error.OpNotImplemented: The following operators are not supported for frontend ONNX: Sequence

[GitHub] [tvm] ibsidorenko opened a new pull request, #12358: [QNN][Hexagon] Disable QNN canonicalization pass

ibsidorenko opened a new pull request, #12358: URL: https://github.com/apache/tvm/pull/12358 Main goals for this PR are the following: 1. E2E compilation should work without QNN canonicalization on the Hexagon target. 2. Enable int8 -> int8 computation for dense/conv2d operation.

[tvm] branch last-successful updated (1a98ea2e4f -> fae79bbc3e)

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 1a98ea2e4f [TIR] Minor fix to tensor intrin description (#12356) add fae79bbc3e [TVMScript] Printer Reg

[tvm] branch last-successful updated (a6c3b1f6c3 -> 1a98ea2e4f)

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 a6c3b1f6c3 [TIR] Add int8 CUDA tensor core intrinsics (#12354) add 1a98ea2e4f [TIR] Minor fix to tensor