[GitHub] [tvm] wrongtest-intellif commented on a diff in pull request #13033: [TIR][Primitive] Support rolling_buffer schedule primitive in TensorIR

wrongtest-intellif commented on code in PR #13033: URL: https://github.com/apache/tvm/pull/13033#discussion_r994085307 ## src/tir/schedule/primitive/rolling_buffer.cc: ## @@ -0,0 +1,443 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more

[GitHub] [tvm] wrongtest-intellif commented on pull request #13033: [TIR][Primitive] Support rolling_buffer schedule primitive in TensorIR

wrongtest-intellif commented on PR #13033: URL: https://github.com/apache/tvm/pull/13033#issuecomment-1277032075 cc @mbaret @junrushao -- 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

[tvm] branch last-successful updated (90c666f860 -> 61c9742ea7)

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 90c666f860 [Relay][Pass] ConcretizeCastLikeRewrite for SimplifyExpr (#12923) add 61c9742ea7 [Hexagon]

[GitHub] [tvm] mkatanbaf commented on pull request #13051: [microTVM] Improve code reuse in Corstone300 conv2d tests

mkatanbaf commented on PR #13051: URL: https://github.com/apache/tvm/pull/13051#issuecomment-1276970012 > * Would it make sense to create an even-more-abstract class that handled graph creation and running for all the tests in this directory, not just the `conv2d` variants? When I

[GitHub] [tvm] mkatanbaf commented on a diff in pull request #13051: [microTVM] Improve code reuse in Corstone300 conv2d tests

mkatanbaf commented on code in PR #13051: URL: https://github.com/apache/tvm/pull/13051#discussion_r994052471 ## tests/python/relay/strategy/arm_cpu/test_depthwise_conv2d.py: ## @@ -14,203 +14,125 @@ # KIND, either express or implied. See the License for the # specific

[GitHub] [tvm] Hzfengsy commented on issue #13026: [VOTE] Release Apache TVM v0.10.0.rc0

Hzfengsy commented on issue #13026: URL: https://github.com/apache/tvm/issues/13026#issuecomment-1276961044 +1 -- 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

[tvm] branch last-successful updated (b18f6425f9 -> 90c666f860)

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 b18f6425f9 [Hexagon] [runtime] Manage RPC and runtime buffers separately (#13028) add 90c666f860

[tvm] branch main updated (90c666f860 -> 61c9742ea7)

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 90c666f860 [Relay][Pass] ConcretizeCastLikeRewrite for SimplifyExpr (#12923) add 61c9742ea7 [Hexagon] Enable multi

[GitHub] [tvm] masahi merged pull request #13037: [Hexagon] Enable multi input Async DMA; same queue / stage

masahi merged PR #13037: URL: https://github.com/apache/tvm/pull/13037 -- 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] zhaoyang-star commented on pull request #11208: [AOT] Calculate used memory at the callsite of primitive functions

zhaoyang-star commented on PR #11208: URL: https://github.com/apache/tvm/pull/11208#issuecomment-1276903736 Hi @lhutton1 , thanks for your contributition. After running FuseOps pass, I want to get the memory usage per op or per primitive func by `AnnotateUsedMemory` pass for furture

[GitHub] [tvm] mehrdadh commented on a diff in pull request #13046: [AOT] Sanitize input/output name in runtime

mehrdadh commented on code in PR #13046: URL: https://github.com/apache/tvm/pull/13046#discussion_r994014937 ## include/tvm/runtime/name_mangling.h: ## @@ -0,0 +1,43 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] tvm-bot commented on pull request #13006: [MetaSchedule][UX] Support Interactive Performance Table Printing in Notebook

tvm-bot commented on PR #13006: URL: https://github.com/apache/tvm/pull/13006#issuecomment-1276886579 Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews

[GitHub] [tvm] tvm-bot commented on pull request #13037: [Hexagon] Enable multi input Async DMA; same queue / stage

tvm-bot commented on PR #13037: URL: https://github.com/apache/tvm/pull/13037#issuecomment-1276865071 Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews

[tvm] branch nightly-docker-update updated (66d7f10877 -> 95eddbd70f)

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 66d7f10877 [ci][docker] Nightly Docker image update add a752b74956 [Conda] Specify python

[GitHub] [tvm] tqchen commented on issue #13026: [VOTE] Release Apache TVM v0.10.0.rc0

tqchen commented on issue #13026: URL: https://github.com/apache/tvm/issues/13026#issuecomment-1276854573 +1. i checked - checksum - gpg - code compiles -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use

[tvm] branch last-successful updated (90c666f860 -> b18f6425f9)

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 omit 90c666f860 [Relay][Pass] ConcretizeCastLikeRewrite for SimplifyExpr (#12923) This update removed existing

[tvm] 01/01: Update use_pass_infra.py

This is an automated email from the ASF dual-hosted git repository. driazati pushed a commit to branch driazati-patch-1 in repository https://gitbox.apache.org/repos/asf/tvm.git commit 4cb7d38f309ce742151769d3b0f33fb39ff8b3fe Author: driazati <9407960+driaz...@users.noreply.github.com>

[tvm] branch driazati-patch-1 created (now 4cb7d38f30)

This is an automated email from the ASF dual-hosted git repository. driazati pushed a change to branch driazati-patch-1 in repository https://gitbox.apache.org/repos/asf/tvm.git at 4cb7d38f30 Update use_pass_infra.py This branch includes the following new commits: new 4cb7d38f30

[GitHub] [tvm] janetsc commented on a diff in pull request #13028: [Hexagon] [runtime] Manage RPC and runtime buffers separately

janetsc commented on code in PR #13028: URL: https://github.com/apache/tvm/pull/13028#discussion_r993977233 ## src/runtime/hexagon/hexagon_buffer_manager.h: ## @@ -85,6 +86,18 @@ class HexagonBufferManager { return hexagon_buffer_map_.empty(); } + //! \brief Returns

[tvm] branch test_branch created (now 90c666f860)

This is an automated email from the ASF dual-hosted git repository. driazati pushed a change to branch test_branch in repository https://gitbox.apache.org/repos/asf/tvm.git at 90c666f860 [Relay][Pass] ConcretizeCastLikeRewrite for SimplifyExpr (#12923) No new revisions were added by

[GitHub] [tvm] janetsc commented on a diff in pull request #13028: [Hexagon] [runtime] Manage RPC and runtime buffers separately

janetsc commented on code in PR #13028: URL: https://github.com/apache/tvm/pull/13028#discussion_r993978956 ## src/runtime/hexagon/hexagon_device_api.cc: ## @@ -115,13 +120,34 @@ void* HexagonDeviceAPI::AllocDataSpace(Device dev, size_t nbytes, size_t alignme if (alignment

[GitHub] [tvm] janetsc commented on a diff in pull request #13028: [Hexagon] [runtime] Manage RPC and runtime buffers separately

janetsc commented on code in PR #13028: URL: https://github.com/apache/tvm/pull/13028#discussion_r993978956 ## src/runtime/hexagon/hexagon_device_api.cc: ## @@ -115,13 +120,34 @@ void* HexagonDeviceAPI::AllocDataSpace(Device dev, size_t nbytes, size_t alignme if (alignment

[GitHub] [tvm] janetsc commented on a diff in pull request #13028: [Hexagon] [runtime] Manage RPC and runtime buffers separately

janetsc commented on code in PR #13028: URL: https://github.com/apache/tvm/pull/13028#discussion_r993977233 ## src/runtime/hexagon/hexagon_buffer_manager.h: ## @@ -85,6 +86,18 @@ class HexagonBufferManager { return hexagon_buffer_map_.empty(); } + //! \brief Returns

[GitHub] [tvm] vinx13 commented on issue #13026: [VOTE] Release Apache TVM v0.10.0.rc0

vinx13 commented on issue #13026: URL: https://github.com/apache/tvm/issues/13026#issuecomment-1276839145 +1 -- 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

[tvm] branch last-successful updated (b18f6425f9 -> 90c666f860)

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 b18f6425f9 [Hexagon] [runtime] Manage RPC and runtime buffers separately (#13028) add 90c666f860

[GitHub] [tvm] adstraw commented on a diff in pull request #13028: [Hexagon] [runtime] Manage RPC and runtime buffers separately

adstraw commented on code in PR #13028: URL: https://github.com/apache/tvm/pull/13028#discussion_r993952131 ## src/runtime/hexagon/hexagon_device_api.cc: ## @@ -115,13 +120,34 @@ void* HexagonDeviceAPI::AllocDataSpace(Device dev, size_t nbytes, size_t alignme if (alignment

[GitHub] [tvm] AndrewZhaoLuo commented on a diff in pull request #13056: [Relay] Fix simplifying consecutive casts when intermediate type has fewer bits

AndrewZhaoLuo commented on code in PR #13056: URL: https://github.com/apache/tvm/pull/13056#discussion_r993939468 ## src/relay/transforms/simplify_expr.cc: ## @@ -146,7 +146,7 @@ class SimplifyConsecutiveCast : public DFPatternRewrite { // BFloat cast cannot be omitted

[GitHub] [tvm] AndrewZhaoLuo commented on a diff in pull request #13056: [Relay] Fix simplifying consecutive casts when intermediate type has fewer bits

AndrewZhaoLuo commented on code in PR #13056: URL: https://github.com/apache/tvm/pull/13056#discussion_r993939468 ## src/relay/transforms/simplify_expr.cc: ## @@ -146,7 +146,7 @@ class SimplifyConsecutiveCast : public DFPatternRewrite { // BFloat cast cannot be omitted

[GitHub] [tvm] AndrewZhaoLuo commented on a diff in pull request #13056: [Relay] Fix simplifying consecutive casts when intermediate type has fewer bits

AndrewZhaoLuo commented on code in PR #13056: URL: https://github.com/apache/tvm/pull/13056#discussion_r993939468 ## src/relay/transforms/simplify_expr.cc: ## @@ -146,7 +146,7 @@ class SimplifyConsecutiveCast : public DFPatternRewrite { // BFloat cast cannot be omitted

[GitHub] [tvm] AndrewZhaoLuo commented on a diff in pull request #13056: [Relay] Fix simplifying consecutive casts when intermediate type has fewer bits

AndrewZhaoLuo commented on code in PR #13056: URL: https://github.com/apache/tvm/pull/13056#discussion_r993938788 ## src/relay/transforms/simplify_expr.cc: ## @@ -146,7 +146,7 @@ class SimplifyConsecutiveCast : public DFPatternRewrite { // BFloat cast cannot be omitted

[GitHub] [tvm] AndrewZhaoLuo commented on a diff in pull request #13056: [Relay] Fix simplifying consecutive casts when intermediate type has fewer bits

AndrewZhaoLuo commented on code in PR #13056: URL: https://github.com/apache/tvm/pull/13056#discussion_r993938350 ## src/relay/transforms/simplify_expr.cc: ## @@ -146,7 +146,7 @@ class SimplifyConsecutiveCast : public DFPatternRewrite { // BFloat cast cannot be omitted

[GitHub] [tvm] adstraw commented on pull request #13037: [Hexagon] Enable multi input Async DMA; same queue / stage

adstraw commented on PR #13037: URL: https://github.com/apache/tvm/pull/13037#issuecomment-1276785052 @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

[GitHub] [tvm] vinx13 commented on a diff in pull request #13056: [Relay] Fix simplifying consecutive casts when intermediate type has fewer bits

vinx13 commented on code in PR #13056: URL: https://github.com/apache/tvm/pull/13056#discussion_r993930056 ## src/relay/transforms/simplify_expr.cc: ## @@ -146,7 +146,7 @@ class SimplifyConsecutiveCast : public DFPatternRewrite { // BFloat cast cannot be omitted

[GitHub] [tvm] AndrewZhaoLuo commented on a diff in pull request #13056: [Relay] Fix simplifying consecutive casts when intermediate type has fewer bits

AndrewZhaoLuo commented on code in PR #13056: URL: https://github.com/apache/tvm/pull/13056#discussion_r993924333 ## src/relay/transforms/simplify_expr.cc: ## @@ -146,7 +146,7 @@ class SimplifyConsecutiveCast : public DFPatternRewrite { // BFloat cast cannot be omitted

[GitHub] [tvm] Mousius commented on a diff in pull request #13046: [AOT] Sanitize input/output name in runtime

Mousius commented on code in PR #13046: URL: https://github.com/apache/tvm/pull/13046#discussion_r993909453 ## include/tvm/runtime/name_mangling.h: ## @@ -0,0 +1,43 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements.

[tvm] branch last-successful updated (256075566b -> b18f6425f9)

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 256075566b [MetaSchedule] Allow skipping exact NDArray rewrite in RemoveWeightLayoutRewriteBlock (#13052)

[GitHub] [tvm] tkonolige opened a new issue, #13059: [Bug] Cannot print primfunc as script

tkonolige opened a new issue, #13059: URL: https://github.com/apache/tvm/issues/13059 ``` import tvm tvm.ir.load_json(open("cant_print.json").read()).script() ``` `cant_print.json`: ``` { "root": 1, "nodes": [ { "type_key": "" },

[GitHub] [tvm] driazati commented on pull request #13058: [ci] Check for trailing newlines and spaces

driazati commented on PR #13058: URL: https://github.com/apache/tvm/pull/13058#issuecomment-1276712630 @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

[GitHub] [tvm] tvm-bot commented on pull request #13058: Lint fixups to pass CI

tvm-bot commented on PR #13058: URL: https://github.com/apache/tvm/pull/13058#issuecomment-1276710895 Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews

[GitHub] [tvm] vinx13 merged pull request #12923: [Relay][Pass] ConcretizeCastLikeRewrite for SimplifyExpr

vinx13 merged PR #12923: URL: https://github.com/apache/tvm/pull/12923 -- 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 (b18f6425f9 -> 90c666f860)

This is an automated email from the ASF dual-hosted git repository. wuwei pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from b18f6425f9 [Hexagon] [runtime] Manage RPC and runtime buffers separately (#13028) add 90c666f860 [Relay][Pass]

[GitHub] [tvm] tvm-bot commented on pull request #12972: [Arith] Optional rewriting and simplification into AND of ORs

tvm-bot commented on PR #12972: URL: https://github.com/apache/tvm/pull/12972#issuecomment-1276685447 Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews

[tvm] branch last-successful updated (b8cfc4ca82 -> 256075566b)

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 b8cfc4ca82 [CI] Update Compute Library to v22.08 (#13035) add 256075566b [MetaSchedule] Allow skipping

[GitHub] [tvm] masahi commented on issue #13057: compile error

masahi commented on issue #13057: URL: https://github.com/apache/tvm/issues/13057#issuecomment-1276615413 Can you upload your `model-small.onnx`? -- 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

[tvm] branch main updated: [Hexagon] [runtime] Manage RPC and runtime buffers separately (#13028)

This is an automated email from the ASF dual-hosted git repository. driazati 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 b18f6425f9 [Hexagon] [runtime] Manage RPC and

[GitHub] [tvm] driazati merged pull request #13028: [Hexagon] [runtime] Manage RPC and runtime buffers separately

driazati merged PR #13028: URL: https://github.com/apache/tvm/pull/13028 -- 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] venkatesha2411 opened a new issue, #13057: compile error

venkatesha2411 opened a new issue, #13057: URL: https://github.com/apache/tvm/issues/13057 Facing problem with compiling the model. from tvm.driver import tvmc #basic config target = "llvm -mcpu=znver2" device = "cpu" do_tuning = True #load the model model =

[GitHub] [tvm] alanmacd commented on a diff in pull request #12851: [microTVM][Windows] Enable building TVM on Windows with USE_MICRO

alanmacd commented on code in PR #12851: URL: https://github.com/apache/tvm/pull/12851#discussion_r993780406 ## cmake/modules/StandaloneCrt.cmake: ## @@ -149,4 +173,6 @@ if(USE_MICRO) list(APPEND TVM_RUNTIME_LINKER_LIBS ${TVM_CRT_LINKER_LIB}) endif() -endif(USE_MICRO)

[GitHub] [tvm] mehrdadh commented on a diff in pull request #12851: [microTVM][Windows] Enable building TVM on Windows with USE_MICRO

mehrdadh commented on code in PR #12851: URL: https://github.com/apache/tvm/pull/12851#discussion_r993769531 ## cmake/modules/StandaloneCrt.cmake: ## @@ -149,4 +173,6 @@ if(USE_MICRO) list(APPEND TVM_RUNTIME_LINKER_LIBS ${TVM_CRT_LINKER_LIB}) endif() -endif(USE_MICRO)

[GitHub] [tvm] vinx13 opened a new pull request, #13056: [Relay] Fix simplifying consecutive casts when intermediate type has fewer bits

vinx13 opened a new pull request, #13056: URL: https://github.com/apache/tvm/pull/13056 Fixed #13048 cc @AndrewZhaoLuo @ganler -- 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] tvm-bot commented on pull request #13056: [Relay] Fix simplifying consecutive casts when intermediate type has fewer bits

tvm-bot commented on PR #13056: URL: https://github.com/apache/tvm/pull/13056#issuecomment-1276537515 Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews

[tvm] branch main updated (b8cfc4ca82 -> 256075566b)

This is an automated email from the ASF dual-hosted git repository. wuwei pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from b8cfc4ca82 [CI] Update Compute Library to v22.08 (#13035) add 256075566b [MetaSchedule] Allow skipping exact NDArray

[GitHub] [tvm] vinx13 closed pull request #13052: [MetaSchedule] Allow skipping exact NDArray rewrite in RemoveWeightLayoutRewriteBlock

vinx13 closed pull request #13052: [MetaSchedule] Allow skipping exact NDArray rewrite in RemoveWeightLayoutRewriteBlock URL: https://github.com/apache/tvm/pull/13052 -- 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] Lunderberg commented on a diff in pull request #12972: [Arith] Optional rewriting and simplification into AND of ORs

Lunderberg commented on code in PR #12972: URL: https://github.com/apache/tvm/pull/12972#discussion_r993706285 ## tests/python/unittest/test_tir_transform_simplify.py: ## @@ -686,5 +688,52 @@ def before(A: T.Buffer[1, "bool"], i: T.int32, j: T.int32, k: T.int32): expected

[GitHub] [tvm] Lunderberg commented on a diff in pull request #12972: [Arith] Optional rewriting and simplification into AND of ORs

Lunderberg commented on code in PR #12972: URL: https://github.com/apache/tvm/pull/12972#discussion_r993704527 ## tests/python/unittest/test_tir_transform_simplify.py: ## @@ -686,5 +688,52 @@ def before(A: T.Buffer[1, "bool"], i: T.int32, j: T.int32, k: T.int32): expected

[GitHub] [tvm] Lunderberg commented on a diff in pull request #12972: [Arith] Optional rewriting and simplification into AND of ORs

Lunderberg commented on code in PR #12972: URL: https://github.com/apache/tvm/pull/12972#discussion_r993696964 ## src/arith/conjunctive_normal_form.cc: ## @@ -0,0 +1,415 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] Lunderberg commented on a diff in pull request #12972: [Arith] Optional rewriting and simplification into AND of ORs

Lunderberg commented on code in PR #12972: URL: https://github.com/apache/tvm/pull/12972#discussion_r993682474 ## src/arith/conjunctive_normal_form.cc: ## @@ -0,0 +1,415 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] tvm-bot commented on pull request #13054: [TIR][FIX] Fix crash when using 'if' without 'else' in TVMScript

tvm-bot commented on PR #13054: URL: https://github.com/apache/tvm/pull/13054#issuecomment-1276430121 Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews

[GitHub] [tvm] janetsc commented on a diff in pull request #13028: [Hexagon] [runtime] Manage RPC and runtime buffers separately

janetsc commented on code in PR #13028: URL: https://github.com/apache/tvm/pull/13028#discussion_r993652802 ## src/runtime/hexagon/hexagon_device_api.cc: ## @@ -137,7 +166,7 @@ void* HexagonDeviceAPI::AllocWorkspace(Device dev, size_t size, DLDataType type_ void

[GitHub] [tvm] mehrdadh commented on a diff in pull request #13046: [AOT] Sanitize input/output name in runtime

mehrdadh commented on code in PR #13046: URL: https://github.com/apache/tvm/pull/13046#discussion_r993645188 ## tests/python/relay/aot/test_cpp_aot.py: ## @@ -203,5 +203,39 @@ def test_pass_wrong_device_arg(): # TODO write asserts for # and type of device.

[GitHub] [tvm] mehrdadh commented on a diff in pull request #13046: [AOT] Sanitize input/output name in runtime

mehrdadh commented on code in PR #13046: URL: https://github.com/apache/tvm/pull/13046#discussion_r993635598 ## include/tvm/runtime/name_mangling.h: ## @@ -0,0 +1,43 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] tvm-bot commented on pull request #13055: [microTVM] Improve kernel alignment on tensordot schedules

tvm-bot commented on PR #13055: URL: https://github.com/apache/tvm/pull/13055#issuecomment-1276386357 Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews

[GitHub] [tvm] guberti opened a new pull request, #13055: [microTVM] Improve kernel alignment on tensordot schedules

guberti opened a new pull request, #13055: URL: https://github.com/apache/tvm/pull/13055 This is a draft PR - please see the comment in 547c447437e0dccff23533e8eff76d2092fe3365 for more details. -- This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] [tvm] Lunderberg commented on a diff in pull request #12972: [Arith] Optional rewriting and simplification into AND of ORs

Lunderberg commented on code in PR #12972: URL: https://github.com/apache/tvm/pull/12972#discussion_r993599891 ## src/arith/conjunctive_normal_form.cc: ## @@ -0,0 +1,415 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] Lunderberg commented on a diff in pull request #12972: [Arith] Optional rewriting and simplification into AND of ORs

Lunderberg commented on code in PR #12972: URL: https://github.com/apache/tvm/pull/12972#discussion_r993613214 ## src/arith/conjunctive_normal_form.cc: ## @@ -0,0 +1,415 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] Lunderberg commented on a diff in pull request #12972: [Arith] Optional rewriting and simplification into AND of ORs

Lunderberg commented on code in PR #12972: URL: https://github.com/apache/tvm/pull/12972#discussion_r993599891 ## src/arith/conjunctive_normal_form.cc: ## @@ -0,0 +1,415 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] Lunderberg commented on a diff in pull request #12972: [Arith] Optional rewriting and simplification into AND of ORs

Lunderberg commented on code in PR #12972: URL: https://github.com/apache/tvm/pull/12972#discussion_r993589048 ## include/tvm/arith/analyzer.h: ## @@ -297,6 +297,14 @@ class RewriteSimplifier { * if_then_else(i if_then_else(i (a && c) || (b && c) Review Comment:

[GitHub] [tvm] WuSiYu opened a new pull request, #13054: [TIR][FIX] Fix crash when using 'if' without 'else' in TVMScript

WuSiYu opened a new pull request, #13054: URL: https://github.com/apache/tvm/pull/13054 This commit fix the tvm's crash during auto-tune when having a 'if' statement which doesn't have a 'else' part in TVMScript. It's cause in tvm's tir processing, in `VisitStmt_()` for IfThenElseNode

[GitHub] [tvm] cVladu opened a new issue, #13053: [Bug] relay.frontend.from_keras fails with AttributeError: 'LSTM_network' object has no attribute '_output_coordinates'

cVladu opened a new issue, #13053: URL: https://github.com/apache/tvm/issues/13053 I tried converting a LSTM model from keras to relay IR ### Expected behavior The network is converted ### Actual behavior `relay.frontend.from_keras` fails with `AttributeError:

[GitHub] [tvm] Lunderberg commented on a diff in pull request #13028: [Hexagon] [runtime] Manage RPC and runtime buffers separately

Lunderberg commented on code in PR #13028: URL: https://github.com/apache/tvm/pull/13028#discussion_r993437460 ## src/runtime/hexagon/hexagon_buffer_manager.h: ## @@ -85,6 +86,18 @@ class HexagonBufferManager { return hexagon_buffer_map_.empty(); } + //! \brief

[GitHub] [tvm-rfcs] tqchen commented on pull request #94: [RFC] CodeGenAArch64 backend with Scalable Vector Extension (SVE)

tqchen commented on PR #94: URL: https://github.com/apache/tvm-rfcs/pull/94#issuecomment-1276158707 Thanks @ekalda i don't have further comments at this pt -- 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

[GitHub] [tvm] ekalda commented on issue #12841: [Flaky Test] `tests/python/contrib/test_ethosu/test_replace_depthwise_conv2d.py::test_depthwise_conv2d_single[trial3]`

ekalda commented on issue #12841: URL: https://github.com/apache/tvm/issues/12841#issuecomment-1276148209 It looks like this is the caused by the same issue of segfaulting when calling Vela's weight encoder for which there are few issues raised and a fix in progress -

[GitHub] [tvm] tvm-bot commented on pull request #13052: [MetaSchedule] Allow skipping exact NDArray rewrite in RemoveWeightLayoutRewriteBlock

tvm-bot commented on PR #13052: URL: https://github.com/apache/tvm/pull/13052#issuecomment-1276006968 Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews

[GitHub] [tvm] masahi opened a new pull request, #13052: [MetaSchedule] Allow skipping exact NDArray rewrite in RemoveWeightLayoutRewriteBlock

masahi opened a new pull request, #13052: URL: https://github.com/apache/tvm/pull/13052 I found that, using MS evo search + `RewriteLayout` + `link-params=True`, where the latter combination is supported by https://github.com/apache/tvm/pull/12991 and

[tvm] branch last-successful updated (48829756e1 -> b8cfc4ca82)

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 48829756e1 [TIR] Fix the stride calculation in InjectRollingBuffer (#13016) add b8cfc4ca82 [CI]

[GitHub] [tvm] masahi commented on pull request #13037: [Hexagon] Enable multi input Async DMA; same queue / stage

masahi commented on PR #13037: URL: https://github.com/apache/tvm/pull/13037#issuecomment-1275978653 Interesting, for CUDA the former behavior is more natural - treating global to shared mem async copy for A and B matrices in GEMM as one "chunk". All CUDA async copy examples and

[GitHub] [tvm] guberti commented on pull request #13051: [microTVM] Improve code reuse in Corstone300 conv2d tests

guberti commented on PR #13051: URL: https://github.com/apache/tvm/pull/13051#issuecomment-1275918393 When reorganizing the tests, I noticed there was duplicate code in tests besides the `conv2d` ones. Each file has its own calls to `relay.op.nn.[operator]`, which are constructed the same

[GitHub] [tvm-rfcs] ekalda commented on pull request #94: [RFC] CodeGenAArch64 backend with Scalable Vector Extension (SVE)

ekalda commented on PR #94: URL: https://github.com/apache/tvm-rfcs/pull/94#issuecomment-1275917969 Thanks for your input and suggestions @tqchen, much appreciated! I added a paragraph about pattern matching TIR, see if it makes sense. Yes, this RFC propses A1 change. A2 style TIR

[tvm] branch last-successful updated (a752b74956 -> 48829756e1)

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 a752b74956 [Conda] Specify python version in build-environment (#13047) add 48829756e1 [TIR] Fix the

[GitHub] [tvm] tvm-bot commented on pull request #13051: [microTVM] Improve code reuse in Corstone300 conv2d tests

tvm-bot commented on PR #13051: URL: https://github.com/apache/tvm/pull/13051#issuecomment-1275858605 Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews

[GitHub] [tvm] guberti opened a new pull request, #13051: [microTVM] Improve code reuse in Corstone300 conv2d tests

guberti opened a new pull request, #13051: URL: https://github.com/apache/tvm/pull/13051 This pull request contains two changes: 1. Adds support for specifying an `out_layout` to the `Tensordot`-based convolution schedules for Arm Cortex-M. This is useful, as when compiling models

[GitHub] [tvm] Mousius commented on a diff in pull request #13046: [AOT] Sanitize input/output name in runtime

Mousius commented on code in PR #13046: URL: https://github.com/apache/tvm/pull/13046#discussion_r993185912 ## include/tvm/runtime/name_mangling.h: ## @@ -0,0 +1,43 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements.

[tvm] branch last-successful updated (cc235f88a4 -> a752b74956)

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 cc235f88a4 [TIR] Update ir_comparator message to be more clear about what is being compared (#13038)

[GitHub] [tvm] lhutton1 commented on a diff in pull request #13022: [ETHOSN] Throw error message when inference fails

lhutton1 commented on code in PR #13022: URL: https://github.com/apache/tvm/pull/13022#discussion_r993161820 ## tests/cpp/runtime/contrib/ethosn/inference_test.cc: ## @@ -0,0 +1,60 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor

[GitHub] [tvm] lhutton1 commented on a diff in pull request #13022: [ETHOSN] Throw error message when inference fails

lhutton1 commented on code in PR #13022: URL: https://github.com/apache/tvm/pull/13022#discussion_r993133454 ## src/runtime/contrib/ethosn/ethosn_device.cc: ## @@ -123,21 +133,26 @@ bool Inference(tvm::runtime::TVMArgs args, dl::Network* npu, } // Execute the

[GitHub] [tvm] lhutton1 commented on pull request #13035: [CI] Update Compute Library to v22.08

lhutton1 commented on PR #13035: URL: https://github.com/apache/tvm/pull/13035#issuecomment-1275749195 Thanks @leandron @Mousius! -- 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

[tvm] branch main updated (48829756e1 -> b8cfc4ca82)

This is an automated email from the ASF dual-hosted git repository. lukhut pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from 48829756e1 [TIR] Fix the stride calculation in InjectRollingBuffer (#13016) add b8cfc4ca82 [CI] Update Compute

[GitHub] [tvm] lhutton1 merged pull request #13035: [CI] Update Compute Library to v22.08

lhutton1 merged PR #13035: URL: https://github.com/apache/tvm/pull/13035 -- 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 (a752b74956 -> 48829756e1)

This is an automated email from the ASF dual-hosted git repository. junrushao pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from a752b74956 [Conda] Specify python version in build-environment (#13047) add 48829756e1 [TIR] Fix the stride

[GitHub] [tvm] junrushao merged pull request #13016: [TIR] Fix the stride calculation in InjectRollingBuffer

junrushao merged PR #13016: URL: https://github.com/apache/tvm/pull/13016 -- 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] driazati merged pull request #13047: [Conda] Specify python version in build-environment

driazati merged PR #13047: URL: https://github.com/apache/tvm/pull/13047 -- 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 (cc235f88a4 -> a752b74956)

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 cc235f88a4 [TIR] Update ir_comparator message to be more clear about what is being compared (#13038) add