[GitHub] [tvm] mehrdadh commented on pull request #8519: [microTVM][Cortex-R5] Add zephyr cortex-r5 board to Zephyr

mehrdadh commented on pull request #8519: URL: https://github.com/apache/tvm/pull/8519#issuecomment-883902130 cc @leandron -- 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] mehrdadh opened a new pull request #8519: [microTVM][Cortex-R5] Add zephyr cortex-r5 board to Zephyr

mehrdadh opened a new pull request #8519: URL: https://github.com/apache/tvm/pull/8519 This PR: - adds support for [cortex-r5](https://docs.zephyrproject.org/latest/boards/arm/qemu_cortex_r5/doc/index.html) board using qemu emulation @areusch `zynq_mp_r5` target name is chosen f

[GitHub] [tvm] mehrdadh opened a new pull request #8518: [Docker][QEMU] Remove QEMU Install

mehrdadh opened a new pull request #8518: URL: https://github.com/apache/tvm/pull/8518 Zephyr SDK has binary files for different qemu versions which are developed by Zephyr and we can use them for our qemu tests. We just need to add the binary path to PATH to enable them. Zephyr SDK als

[GitHub] [tvm] mehrdadh commented on a change in pull request #8487: [AOT][Stack Allocator] Fix Initial Memory Misalignment

mehrdadh commented on a change in pull request #8487: URL: https://github.com/apache/tvm/pull/8487#discussion_r673659054 ## File path: tests/crt/aot_memory_test.cc ## @@ -109,27 +145,50 @@ TEST(AOTMemory, OverAllocate) { } /* - * Test for out-of-order memory deallocation +

[GitHub] [tvm] mehrdadh commented on a change in pull request #8487: [AOT][Stack Allocator] Fix Initial Memory Misalignment

mehrdadh commented on a change in pull request #8487: URL: https://github.com/apache/tvm/pull/8487#discussion_r673658807 ## File path: src/runtime/crt/memory/stack_allocator.c ## @@ -79,8 +79,14 @@ tvm_crt_error_t StackMemoryManager_Free(tvm_workspace_t* tvm_runtime_workspace,

[GitHub] [tvm] GoodNight-bye commented on issue #7986: ImplementationError: The following operators are not implemented: ['aten::embedding_bag']

GoodNight-bye commented on issue #7986: URL: https://github.com/apache/tvm/issues/7986#issuecomment-883829191 embedding_bag is supported in relay/frontend/onnx.py, you can refer to this file. -- This is an automated message from the Apache Git Service. To respond to the message, please

[GitHub] [tvm] chiwwang edited a comment on pull request #8448: [Draft PR] Relay IR visualizer.

chiwwang edited a comment on pull request #8448: URL: https://github.com/apache/tvm/pull/8448#issuecomment-883824888 @elvin-n Thanks for suggestions! 1. The position and edge are provided by graphviz and after some googling.. yes it will mix order :( https://stackoverflow.com/questio

[GitHub] [tvm] chiwwang edited a comment on pull request #8448: [Draft PR] Relay IR visualizer.

chiwwang edited a comment on pull request #8448: URL: https://github.com/apache/tvm/pull/8448#issuecomment-883824888 @elvin-n Thanks for suggestions! 1. The position and edge are provided by graphviz and after some googling.. yes it will mix input order :( https://stackoverflow.com/q

[GitHub] [tvm] chiwwang commented on pull request #8448: [Draft PR] Relay IR visualizer.

chiwwang commented on pull request #8448: URL: https://github.com/apache/tvm/pull/8448#issuecomment-883824888 @elvin-n Thanks for suggestions! 1. The position and edge are provided by graphviz and after some googling.. yes it will mix input order :( https://stackoverflow.com/question

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

sunjiweiswift commented on a change in pull request #8497: URL: https://github.com/apache/tvm/pull/8497#discussion_r673603393 ## File path: src/runtime/graph_executor/graph_executor.cc ## @@ -389,6 +434,13 @@ void GraphExecutor::SetupOpExecs() { input_dltensors_[eid].

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

sunjiweiswift commented on a change in pull request #8497: URL: https://github.com/apache/tvm/pull/8497#discussion_r673599408 ## File path: src/runtime/graph_executor/graph_executor.cc ## @@ -139,6 +156,30 @@ void GraphExecutor::SetInputZeroCopy(int index, DLTensor* data_ref)

[GitHub] [tvm] junrushao1994 commented on a change in pull request #8515: [TIR] Bugfix for zero number arguments tir functions.

junrushao1994 commented on a change in pull request #8515: URL: https://github.com/apache/tvm/pull/8515#discussion_r673598569 ## File path: python/tvm/tir/transform/transform.py ## @@ -331,14 +331,15 @@ def LowerCustomDatatypes(): return _ffi_api.LowerCustomDatatypes() #

[GitHub] [tvm] masahi commented on pull request #8512: Organize CodeOwners File

masahi commented on pull request #8512: URL: https://github.com/apache/tvm/pull/8512#issuecomment-883795160 @tqchen Need to push again -- 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 specifi

[tvm] branch main updated (1a1be09 -> dbdfc44)

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 1a1be09 [Refactor] Remove scope attribute from Buffer class (#8463) add dbdfc44 Enable ONNX tests that needed onnxr

[GitHub] [tvm] masahi merged pull request #8502: Enable ONNX tests that needed onnxruntime 1.7.0

masahi merged pull request #8502: URL: https://github.com/apache/tvm/pull/8502 -- 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...

[GitHub] [tvm] areusch commented on pull request #8382: [Pre-commit] Add pre-commit configuration to perform minimal checks locally

areusch commented on pull request #8382: URL: https://github.com/apache/tvm/pull/8382#issuecomment-883738910 @mikepapadim how do you want to proceed here? -- 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] areusch commented on a change in pull request #8380: [microTVM] Project API infrastructure

areusch commented on a change in pull request #8380: URL: https://github.com/apache/tvm/pull/8380#discussion_r673520171 ## File path: python/tvm/micro/compiler.py ## @@ -1,361 +0,0 @@ -# Licensed to the Apache Software Foundation (ASF) under one Review comment: so i was

[GitHub] [tvm] areusch commented on a change in pull request #8380: [microTVM] Project API infrastructure

areusch commented on a change in pull request #8380: URL: https://github.com/apache/tvm/pull/8380#discussion_r673519098 ## File path: python/tvm/micro/project.py ## @@ -0,0 +1,97 @@ +"""Defines glue wrappers around the Project API which mate to TVM interfaces.""" + +from ..con

[GitHub] [tvm] AndrewZhaoLuo commented on pull request #8517: [TIR] cast disparate floating point types for binary ops

AndrewZhaoLuo commented on pull request #8517: URL: https://github.com/apache/tvm/pull/8517#issuecomment-883731563 By the way, please don't merge this anytime soon. I want to get some input from some others like @jroesch or @junrushao1994 -- This is an automated message from the Apache

[GitHub] [tvm] AndrewZhaoLuo edited a comment on pull request #8517: [TIR] cast disparate floating point types for binary ops

AndrewZhaoLuo edited a comment on pull request #8517: URL: https://github.com/apache/tvm/pull/8517#issuecomment-883729525 Thanks for the review @comaniac PTAL -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL ab

[GitHub] [tvm] AndrewZhaoLuo commented on pull request #8517: [TIR] cast disparate floating point types for binary ops

AndrewZhaoLuo commented on pull request #8517: URL: https://github.com/apache/tvm/pull/8517#issuecomment-883729525 Thanks for the review@comaniac PTAL -- 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] AndrewZhaoLuo commented on a change in pull request #8517: [TIR] cast disparate floating point types for binary ops

AndrewZhaoLuo commented on a change in pull request #8517: URL: https://github.com/apache/tvm/pull/8517#discussion_r673516044 ## File path: src/tir/op/op.cc ## @@ -112,12 +112,23 @@ void BinaryOpMatchTypes(PrimExpr& lhs, PrimExpr& rhs, Span span) { // NOLINT(*) ICHECK(lt

[GitHub] [tvm] comaniac commented on a change in pull request #8517: [TIR] cast disparate floating point types for binary ops

comaniac commented on a change in pull request #8517: URL: https://github.com/apache/tvm/pull/8517#discussion_r673511444 ## File path: tests/python/unittest/test_tir_base.py ## @@ -30,15 +32,25 @@ def build_tir_func(func): def test_scalar_add(): -a = tir.Var("a", "floa

[GitHub] [tvm] tkonolige closed pull request #6794: [TOPI] Add embedding op and gradient

tkonolige closed pull request #6794: URL: https://github.com/apache/tvm/pull/6794 -- 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

[GitHub] [tvm] tkonolige closed pull request #7994: [FIX,DOCS] Prevent segfault when building docs when VTA is not available

tkonolige closed pull request #7994: URL: https://github.com/apache/tvm/pull/7994 -- 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

[GitHub] [tvm] AndrewZhaoLuo commented on issue #8340: [AMP] Audit mixed precision operation schedules for correctness

AndrewZhaoLuo commented on issue #8340: URL: https://github.com/apache/tvm/issues/8340#issuecomment-883717259 https://github.com/apache/tvm/pull/8517 might fix this if it is not too controversial. -- This is an automated message from the Apache Git Service. To respond to the message, ple

[GitHub] [tvm] AndrewZhaoLuo opened a new pull request #8517: [TIR] upcast disparate floating point types for binary ops

AndrewZhaoLuo opened a new pull request #8517: URL: https://github.com/apache/tvm/pull/8517 Right now if we in TIR add a floating point type with an integer type, the integer type will be implicitly cast to the floating point type. E.g. `a: float32 + b: int32 ---> a: float32 + cast(f

[GitHub] [tvm] masahi closed issue #8140: [TEST][FLAKY] tests/python/frontend/tensorflow/test_forward.py::test_forward_combined_nms

masahi closed issue #8140: URL: https://github.com/apache/tvm/issues/8140 -- 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.

[GitHub] [tvm] masahi commented on issue #8140: [TEST][FLAKY] tests/python/frontend/tensorflow/test_forward.py::test_forward_combined_nms

masahi commented on issue #8140: URL: https://github.com/apache/tvm/issues/8140#issuecomment-883706777 This can be closed in the sense that the flaky test is now disabled. But the underlying problem with combined NMS converter for `q != 1` case (each class has its own box) should be addres

[GitHub] [tvm-rfcs] areusch commented on a change in pull request #11: [RFC] Arm Ethos-U Integration

areusch commented on a change in pull request #11: URL: https://github.com/apache/tvm-rfcs/pull/11#discussion_r673406162 ## File path: rfcs/0011_Arm_Ethos-U_Integration.md ## @@ -0,0 +1,233 @@ +Feature Name: Arm® Ethos™-U Integration +Start Date: 2020 May +RFC PR: h

[GitHub] [tvm] masahi opened a new pull request #8516: [Refactor] Remove AttrStmt with storage_scope key

masahi opened a new pull request #8516: URL: https://github.com/apache/tvm/pull/8516 The final follow-up to https://github.com/apache/tvm/pull/8366 After #8366, storage scopes associated with AttrStmt and PointerType are identical. To consolidate storage scope information into one pl

[GitHub] [tvm] tqchen closed pull request #7629: [Runtime] OpenCL Texture Memory Support

tqchen closed pull request #7629: URL: https://github.com/apache/tvm/pull/7629 -- 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...

[GitHub] [tvm] tqchen commented on pull request #7686: [Texture support][Part 1] TIR lowering and OpenCL support

tqchen commented on pull request #7686: URL: https://github.com/apache/tvm/pull/7686#issuecomment-883670484 cc @Lunderberg @masahi @ZihengJiang would be great if we can followup on this. -- This is an automated message from the Apache Git Service. To respond to the message, please log on

[GitHub] [tvm] tqchen commented on pull request #8463: [Refactor] Remove scope attribute from Buffer class

tqchen commented on pull request #8463: URL: https://github.com/apache/tvm/pull/8463#issuecomment-883668687 Thanks @masahi @jroesch @kparzysz-quic -- 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 t

[tvm] branch main updated (1141709 -> 1a1be09)

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git. from 1141709 Fix 8093, Enhance Buffer Index Simplify (#8204) add 1a1be09 [Refactor] Remove scope attribute from Buffer c

[GitHub] [tvm] tqchen merged pull request #8463: [Refactor] Remove scope attribute from Buffer class

tqchen merged pull request #8463: URL: https://github.com/apache/tvm/pull/8463 -- 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...

[GitHub] [tvm] tqchen commented on issue #8432: [Bug.Relay.InferType] Type Inference Report Mismatch after One Operator Is Removed

tqchen commented on issue #8432: URL: https://github.com/apache/tvm/issues/8432#issuecomment-883668312 @slyubomirsky @jroesch @altanh would be great if you can help followup on this -- This is an automated message from the Apache Git Service. To respond to the message, please log on to G

[GitHub] [tvm] tqchen commented on issue #7724: Minor bug in storage_access.cc

tqchen commented on issue #7724: URL: https://github.com/apache/tvm/issues/7724#issuecomment-883666742 cc @masahi it would be great if you can help follow up -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL abo

[GitHub] [tvm] tqchen commented on issue #8140: [TEST][FLAKY] tests/python/frontend/tensorflow/test_forward.py::test_forward_combined_nms

tqchen commented on issue #8140: URL: https://github.com/apache/tvm/issues/8140#issuecomment-883664861 @masahi please followup to see if we can close this issue -- 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] tqchen commented on issue #8182: Support conv2d transpose operator with groups > 1

tqchen commented on issue #8182: URL: https://github.com/apache/tvm/issues/8182#issuecomment-883664486 Thanks @Lyken17 , do you mind to send a PR to resolve the case? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use th

[GitHub] [tvm] ZihengJiang commented on pull request #8515: [TIR] Bugfix for zero number arguments tir functions.

ZihengJiang commented on pull request #8515: URL: https://github.com/apache/tvm/pull/8515#issuecomment-883664479 Hi @areusch ! The motivation of the PR is here: https://discuss.tvm.apache.org/t/tvm-primfunc-returning-a-constant-cannot-compile/10577/2 It is mainly because we cannot distin

[GitHub] [tvm] ZihengJiang commented on a change in pull request #8515: [TIR] Bugfix for zero number arguments tir functions.

ZihengJiang commented on a change in pull request #8515: URL: https://github.com/apache/tvm/pull/8515#discussion_r673453914 ## File path: src/tir/transforms/make_packed_api.cc ## @@ -120,6 +120,11 @@ PrimFunc MakePackedAPI(PrimFunc&& func, int num_unpacked_args) { int num_a

[tvm] branch main updated (ca28dff -> 1141709)

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git. from ca28dff Rename runtime-config to executor-config and add documentation for Model Library Format (#8270) add 1141709

[GitHub] [tvm] tqchen closed issue #8093: Bug in StorageFlatten Cause Mussy Index

tqchen closed issue #8093: URL: https://github.com/apache/tvm/issues/8093 -- 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.

[GitHub] [tvm] tqchen merged pull request #8204: Fix #8093, Enhance Buffer Index Simplify

tqchen merged pull request #8204: URL: https://github.com/apache/tvm/pull/8204 -- 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...

[GitHub] [tvm] tqchen commented on pull request #8204: Fix #8093, Enhance Buffer Index Simplify

tqchen commented on pull request #8204: URL: https://github.com/apache/tvm/pull/8204#issuecomment-883661993 Thanks @Johnson9009 ! -- 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 com

[GitHub] [tvm] tqchen commented on pull request #8225: Implement bitserial_conv1d operator

tqchen commented on pull request #8225: URL: https://github.com/apache/tvm/pull/8225#issuecomment-883660401 cc @jwfromm , would be great if you can help to manage this PR -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and us

[GitHub] [tvm] tqchen commented on issue #8510: Bug: Golang Frontend expects wrong output size

tqchen commented on issue #8510: URL: https://github.com/apache/tvm/issues/8510#issuecomment-883656640 Thanks @Shpionus can you confirm if it is because the mobile net number of classes get updated, it is a bit strange to see 1001 classes. cc @srkreddy1238 -- This is an automated messag

[tvm] branch ci-docker-staging updated (9fc03b9 -> 843d1c4)

This is an automated email from the ASF dual-hosted git repository. mbrookhart pushed a change to branch ci-docker-staging in repository https://gitbox.apache.org/repos/asf/tvm.git. from 9fc03b9 Change jenkinsfile add c8f54f9 [Bugfix, CuDNN] fix segfault when cudnnDestroy called with

[GitHub] [tvm] areusch commented on a change in pull request #8515: [TIR] Bugfix for zero number arguments tir functions.

areusch commented on a change in pull request #8515: URL: https://github.com/apache/tvm/pull/8515#discussion_r673430696 ## File path: src/tir/transforms/make_packed_api.cc ## @@ -120,6 +120,11 @@ PrimFunc MakePackedAPI(PrimFunc&& func, int num_unpacked_args) { int num_args

[tvm] branch ci-docker-staging updated (5fbe3be -> 9fc03b9)

This is an automated email from the ASF dual-hosted git repository. mbrookhart pushed a change to branch ci-docker-staging in repository https://gitbox.apache.org/repos/asf/tvm.git. omit 5fbe3be support explicit padding for NCHW TF padding test omit ccf68b7 manage TF memory use in TF1

[GitHub] [tvm] leandron commented on pull request #8270: Rename runtime-config to executor-config and add documentation for Model Library Format

leandron commented on pull request #8270: URL: https://github.com/apache/tvm/pull/8270#issuecomment-883631515 Thanks @areusch @mehrdadh @Mousius @manupa-arm, this is merged now! -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub

[tvm] branch main updated (d5818b4 -> ca28dff)

This is an automated email from the ASF dual-hosted git repository. leandron pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git. from d5818b4 [Frontend, pytorch] Vc/pytorch lstm (#8447) add ca28dff Rename runtime-config to executor-config and add

[GitHub] [tvm] leandron merged pull request #8270: Rename runtime-config to executor-config and add documentation for Model Library Format

leandron merged pull request #8270: URL: https://github.com/apache/tvm/pull/8270 -- 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.

[GitHub] [tvm] tqchen commented on a change in pull request #8515: [TIR] Bugfix for zero number arguments tir functions.

tqchen commented on a change in pull request #8515: URL: https://github.com/apache/tvm/pull/8515#discussion_r673399611 ## File path: src/tir/transforms/make_packed_api.cc ## @@ -120,6 +120,11 @@ PrimFunc MakePackedAPI(PrimFunc&& func, int num_unpacked_args) { int num_args =

[GitHub] [tvm] tqchen commented on a change in pull request #8515: [TIR] Bugfix for zero number arguments tir functions.

tqchen commented on a change in pull request #8515: URL: https://github.com/apache/tvm/pull/8515#discussion_r673399253 ## File path: tests/python/unittest/test_tir_base.py ## @@ -36,11 +36,22 @@ def test_scalar_add(): c = tir.ret(c) c = tir.Evaluate(c) func = tir

[GitHub] [tvm] tqchen commented on pull request #8499: [TVMSCRIPT]Fix script printters StructuralEqual check failed

tqchen commented on pull request #8499: URL: https://github.com/apache/tvm/pull/8499#issuecomment-883620580 Thanks @Beya2019 , can you please add a regression test to cover the case? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to G

[GitHub] [tvm] rohanmukh commented on a change in pull request #8506: [TensorRT, BYOC] Handling a corner case in TRT RemoveDropout pass

rohanmukh commented on a change in pull request #8506: URL: https://github.com/apache/tvm/pull/8506#discussion_r673394213 ## File path: python/tvm/relay/op/contrib/tensorrt.py ## @@ -1033,6 +1033,8 @@ def visit_tuple_getitem(self, op): visit = super().visit_tuple_getit

[GitHub] [tvm] ZihengJiang opened a new pull request #8515: [TIR] Bugfix for zero number arguments tir functions.

ZihengJiang opened a new pull request #8515: URL: https://github.com/apache/tvm/pull/8515 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 fr

[GitHub] [tvm] tkonolige commented on a change in pull request #8514: Docker env for Arm® Ethos™-U55 Port

tkonolige commented on a change in pull request #8514: URL: https://github.com/apache/tvm/pull/8514#discussion_r673389174 ## File path: docker/Dockerfile.ci_cpu ## @@ -109,3 +109,13 @@ RUN bash /install/ubuntu_install_androidsdk.sh ENV ANDROID_HOME=/opt/android-sdk-linux/ ENV

[GitHub] [tvm] electriclilies commented on a change in pull request #8373: [BUILD] Add caching to CMake

electriclilies commented on a change in pull request #8373: URL: https://github.com/apache/tvm/pull/8373#discussion_r673375800 ## File path: CMakeLists.txt ## @@ -635,3 +635,32 @@ if(APPLE AND TVM_IS_DEBUG_BUILD) VERBATIM ) endif() + +#Caches the build

[GitHub] [tvm] grant-arm opened a new pull request #8514: Docker env for Arm® Ethos™-U55 Port

grant-arm opened a new pull request #8514: URL: https://github.com/apache/tvm/pull/8514 This is the first PR for the Arm Ethos-U Integration RFC (https://github.com/apache/tvm-rfcs/pull/11) * Added Arm® Corstone™-300 Reference System for testing * Added Arm® Ethos™-U driver stack

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

huajsj commented on a change in pull request #8497: URL: https://github.com/apache/tvm/pull/8497#discussion_r673356905 ## File path: src/runtime/graph_executor/graph_executor.cc ## @@ -389,6 +434,13 @@ void GraphExecutor::SetupOpExecs() { input_dltensors_[eid].push_ba

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

huajsj commented on a change in pull request #8497: URL: https://github.com/apache/tvm/pull/8497#discussion_r673354423 ## File path: src/runtime/graph_executor/graph_executor.cc ## @@ -139,6 +156,30 @@ void GraphExecutor::SetInputZeroCopy(int index, DLTensor* data_ref) {

[GitHub] [tvm] shingjan commented on a change in pull request #8492: [TIR] Use PopenPool instead of multiprocessing.pool

shingjan commented on a change in pull request #8492: URL: https://github.com/apache/tvm/pull/8492#discussion_r673350211 ## File path: python/tvm/autotvm/tuner/xgboost_cost_model.py ## @@ -161,7 +162,8 @@ def _reset_pool(self, space, target, task): _extract_space = spa

[GitHub] [tvm] comaniac commented on a change in pull request #8454: [Frontend, Tensorflow2] Added support for TensorList ops

comaniac commented on a change in pull request #8454: URL: https://github.com/apache/tvm/pull/8454#discussion_r673343332 ## File path: python/tvm/relay/frontend/tensorflow2.py ## @@ -66,6 +74,12 @@ def set_span(sym, node_name): return sym +def is_tensor_list_constuctor

[GitHub] [tvm] shingjan commented on a change in pull request #8492: [TIR] Use PopenPool instead of multiprocessing.pool

shingjan commented on a change in pull request #8492: URL: https://github.com/apache/tvm/pull/8492#discussion_r673350114 ## File path: python/tvm/autotvm/tuner/xgboost_cost_model.py ## @@ -161,7 +162,8 @@ def _reset_pool(self, space, target, task): _extract_space = spa

[GitHub] [tvm] vinx13 commented on a change in pull request #8492: [TIR] Use PopenPool instead of multiprocessing.pool

vinx13 commented on a change in pull request #8492: URL: https://github.com/apache/tvm/pull/8492#discussion_r673338148 ## File path: python/tvm/auto_scheduler/utils.py ## @@ -306,6 +307,7 @@ def _func_wrapper(que, func, args, kwargs, add_thread_wrapper): def call_func_with_ti

[GitHub] [tvm] manupa-arm commented on pull request #8472: [TIR] Introduce tir.allocate_const to TIR

manupa-arm commented on pull request #8472: URL: https://github.com/apache/tvm/pull/8472#issuecomment-883568689 Hi @tqchen, Thanks for explanation of the concerns. I agree that they are valid concerns. > Q0: What is the interaction model with the high level TIR schedule(e.g. do

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

huajsj commented on a change in pull request #8497: URL: https://github.com/apache/tvm/pull/8497#discussion_r673336938 ## File path: src/runtime/graph_executor/graph_executor.cc ## @@ -139,6 +156,30 @@ void GraphExecutor::SetInputZeroCopy(int index, DLTensor* data_ref) {

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

huajsj commented on a change in pull request #8497: URL: https://github.com/apache/tvm/pull/8497#discussion_r673334665 ## File path: src/runtime/graph_executor/graph_executor.cc ## @@ -463,6 +515,15 @@ PackedFunc GraphExecutor::GetFunction(const std::string& name, thi

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

huajsj commented on a change in pull request #8497: URL: https://github.com/apache/tvm/pull/8497#discussion_r672767636 ## File path: src/runtime/graph_executor/graph_executor.cc ## @@ -463,6 +515,15 @@ PackedFunc GraphExecutor::GetFunction(const std::string& name, thi

[GitHub] [tvm] tkonolige commented on a change in pull request #8492: [TIR] Use PopenPool instead of multiprocessing.pool

tkonolige commented on a change in pull request #8492: URL: https://github.com/apache/tvm/pull/8492#discussion_r673331852 ## File path: python/tvm/autotvm/tuner/xgboost_cost_model.py ## @@ -161,7 +162,8 @@ def _reset_pool(self, space, target, task): _extract_space = sp

[GitHub] [tvm] tkonolige commented on a change in pull request #8479: [TOPI][CUDA] Improve the performance of scatter_nd

tkonolige commented on a change in pull request #8479: URL: https://github.com/apache/tvm/pull/8479#discussion_r673323689 ## File path: python/tvm/topi/cuda/scatter.py ## @@ -787,42 +786,42 @@ def gen_ir(data_ptr, indices_ptr, updates_ptr, out_ptr): for i in data_ptr.s

[GitHub] [tvm] areusch commented on a change in pull request #8487: [AOT][Stack Allocator] Fix Initial Memory Misalignment

areusch commented on a change in pull request #8487: URL: https://github.com/apache/tvm/pull/8487#discussion_r673276401 ## File path: tests/crt/aot_memory_test.cc ## @@ -24,83 +25,118 @@ // Check with LIFO checks enabled for stack allocator #define TVM_CRT_STACK_ALLOCATOR_E

[GitHub] [tvm] tkonolige commented on a change in pull request #8503: [PROFILING] Add json output to profiling reports

tkonolige commented on a change in pull request #8503: URL: https://github.com/apache/tvm/pull/8503#discussion_r673284379 ## File path: src/runtime/profiling.cc ## @@ -228,6 +228,71 @@ String ReportNode::AsCSV() const { return s.str(); } +namespace { +void print_metric(st

[GitHub] [tvm] junrushao1994 commented on a change in pull request #8467: [TensorIR][M2a] Fuse, Split

junrushao1994 commented on a change in pull request #8467: URL: https://github.com/apache/tvm/pull/8467#discussion_r673243019 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -0,0 +1,388 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one

[GitHub] [tvm] mbrookhart opened a new pull request #8513: add onnx to cpu docker image

mbrookhart opened a new pull request #8513: URL: https://github.com/apache/tvm/pull/8513 @areusch @electriclilies @tkonolige Uploading a CI image with this change to [v3 here](https://hub.docker.com/repository/docker/mbrookhart/ci-cpu), will make a staging job for #8390 -- This

[GitHub] [tvm] mbaret commented on a change in pull request #8368: [TFLite][Testing] Add infra to write TFLite model buffers directly

mbaret commented on a change in pull request #8368: URL: https://github.com/apache/tvm/pull/8368#discussion_r673217549 ## File path: python/tvm/relay/testing/tflite.py ## @@ -0,0 +1,468 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor lice

[GitHub] [tvm] jinhongyii edited a comment on pull request #8467: [TensorIR][M2a] Fuse, Split

jinhongyii edited a comment on pull request #8467: URL: https://github.com/apache/tvm/pull/8467#issuecomment-883430097 Sorry for bothering you. I don't know what happened and I don't know how to cancel it. -- This is an automated message from the Apache Git Service. To respond to the mes

[GitHub] [tvm] jinhongyii commented on a change in pull request #8467: [TensorIR][M2a] Fuse, Split

jinhongyii commented on a change in pull request #8467: URL: https://github.com/apache/tvm/pull/8467#discussion_r673178406 ## File path: src/arith/rewrite_simplify.cc ## @@ -881,7 +883,7 @@ PrimExpr RewriteSimplifier::Impl::VisitExpr_(const FloorModNode* op) { TVM_TRY_R

[GitHub] [tvm] jinhongyii edited a comment on pull request #8467: [TensorIR][M2a] Fuse, Split

jinhongyii edited a comment on pull request #8467: URL: https://github.com/apache/tvm/pull/8467#issuecomment-883430097 Sorry for bothering you. This is a mistaken action and I don't know how to cancel it. -- This is an automated message from the Apache Git Service. To respond to the mess

[GitHub] [tvm] elvin-n commented on pull request #8448: [Draft PR] Relay IR visualizer.

elvin-n commented on pull request #8448: URL: https://github.com/apache/tvm/pull/8448#issuecomment-883431883 @chiwwang that's great feature, thank you for adding it! some comments - Is my impression correct that visual representation can mix the order of the inputs into the node?

[GitHub] [tvm] jinhongyii commented on pull request #8467: [TensorIR][M2a] Fuse, Split

jinhongyii commented on pull request #8467: URL: https://github.com/apache/tvm/pull/8467#issuecomment-883430596 @junrushao1994 I've addressed your new comments -- 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] jinhongyii edited a comment on pull request #8467: [TensorIR][M2a] Fuse, Split

jinhongyii edited a comment on pull request #8467: URL: https://github.com/apache/tvm/pull/8467#issuecomment-883430097 Sorry for requesting you reviews. This is a mistaken action and I don't know how to cancel it. -- This is an automated message from the Apache Git Service. To respond to

[GitHub] [tvm] jinhongyii commented on pull request #8467: [TensorIR][M2a] Fuse, Split

jinhongyii commented on pull request #8467: URL: https://github.com/apache/tvm/pull/8467#issuecomment-883430097 Sorry for requesting you reviews. This is a mistaken action -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and u

[GitHub] [tvm] jinhongyii commented on a change in pull request #8467: [TensorIR][M2a] Fuse, Split

jinhongyii commented on a change in pull request #8467: URL: https://github.com/apache/tvm/pull/8467#discussion_r673157414 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -0,0 +1,388 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + *

[GitHub] [tvm] tqchen commented on pull request #8466: [CUDA] Initial support for dynamic shared memory

tqchen commented on pull request #8466: URL: https://github.com/apache/tvm/pull/8466#issuecomment-883405197 @masahi this is done by the code owner mechanism automatically by github. As long as there is one codeowner approval i think we are fine here. -- This is an automated message from

[GitHub] [tvm] tqchen opened a new pull request #8512: Make codeowner always matches prefix

tqchen opened a new pull request #8512: URL: https://github.com/apache/tvm/pull/8512 Seems wildcard matching in the prefix does not work too well with github, updates the file to always mark the prefix -- This is an automated message from the Apache Git Service. To respond to the message

[GitHub] [tvm] masahi commented on a change in pull request #8466: [CUDA] Initial support for dynamic shared memory

masahi commented on a change in pull request #8466: URL: https://github.com/apache/tvm/pull/8466#discussion_r673123145 ## File path: src/runtime/thread_storage_scope.h ## @@ -194,13 +198,17 @@ struct ThreadWorkLoad { inline size_t grid_dim(size_t i) const { return work_size[

[GitHub] [tvm] masahi commented on pull request #8466: [CUDA] Initial support for dynamic shared memory

masahi commented on pull request #8466: URL: https://github.com/apache/tvm/pull/8466#issuecomment-883387968 Sorry apparently I made a bunch of review requests to code owners, but that was not intended. -- This is an automated message from the Apache Git Service. To respond to the messag

[GitHub] [tvm] Shpionus opened a new pull request #8511: Fix #8510

Shpionus opened a new pull request #8511: URL: https://github.com/apache/tvm/pull/8511 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] Shpionus opened a new issue #8510: Bug: Golang Frontend expects wrong output size

Shpionus opened a new issue #8510: URL: https://github.com/apache/tvm/issues/8510 I tested ` Golang Frontend for TVM Runtime` tutorial (tvm/golang/README.md) Running of ```./complex``` raise and error: ``` $ ./complex TVM Version : v0.8.dev0 DLPACK Version: v40 G

[GitHub] [tvm] tqchen commented on a change in pull request #8466: [CUDA] Initial support for dynamic shared memory

tqchen commented on a change in pull request #8466: URL: https://github.com/apache/tvm/pull/8466#discussion_r673111931 ## File path: src/runtime/thread_storage_scope.h ## @@ -194,13 +198,17 @@ struct ThreadWorkLoad { inline size_t grid_dim(size_t i) const { return work_size[

[GitHub] [tvm] tqchen commented on a change in pull request #8466: [CUDA] Initial support for dynamic shared memory

tqchen commented on a change in pull request #8466: URL: https://github.com/apache/tvm/pull/8466#discussion_r673108940 ## File path: src/runtime/file_utils.cc ## @@ -64,12 +66,14 @@ void FunctionInfo::Save(dmlc::Stream* writer) const { writer->Write(name); writer->Write(a

[GitHub] [tvm] tqchen commented on pull request #8463: [Refactor] Remove scope attribute from Buffer class

tqchen commented on pull request #8463: URL: https://github.com/apache/tvm/pull/8463#issuecomment-883380855 Trying to move this convo forward and conclude: - I think we can remove the scope from buffer if we felt strongly about it - Ideally it would be great to keep the dtype field

[GitHub] [tvm] masahi commented on pull request #8466: [CUDA] Initial support for dynamic shared memory

masahi commented on pull request #8466: URL: https://github.com/apache/tvm/pull/8466#issuecomment-883376561 @tqchen Backward compat issue has been addressed. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL abo

[GitHub] [tvm] masahi commented on a change in pull request #8466: [CUDA] Initial support for dynamic shared memory

masahi commented on a change in pull request #8466: URL: https://github.com/apache/tvm/pull/8466#discussion_r673103103 ## File path: src/runtime/file_utils.cc ## @@ -64,12 +66,14 @@ void FunctionInfo::Save(dmlc::Stream* writer) const { writer->Write(name); writer->Write(a

[GitHub] [tvm] masahi commented on issue #6474: [Torch] Support aten::lstm

masahi commented on issue #6474: URL: https://github.com/apache/tvm/issues/6474#issuecomment-883358492 https://github.com/apache/tvm/pull/8447 -- 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] masahi closed issue #6474: [Torch] Support aten::lstm

masahi closed issue #6474: URL: https://github.com/apache/tvm/issues/6474 -- 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.

  1   2   3   >