[GitHub] [tvm] elvin-nnov commented on a change in pull request #7903: [Graph Executor Debugger] Fix parameter dump

elvin-nnov commented on a change in pull request #7903: URL: https://github.com/apache/tvm/pull/7903#discussion_r620878239 ## File path: src/runtime/graph_executor/debug/graph_executor_debug.cc ## @@ -207,6 +200,42 @@ class GraphExecutorDebug : public GraphExecutor { retur

[tvm] branch main updated (82fecbf -> f681359)

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 82fecbf [CodeGenC] Fix bugs when calling extern functions (#7911) add f681359 [FIX] skip_conv_layers will affect qua

[GitHub] [tvm] vinx13 merged pull request #7795: [FIX] skip_conv_layers will affect quantization of nn.dense

vinx13 merged pull request #7795: URL: https://github.com/apache/tvm/pull/7795 -- 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. For queries about this service, please cont

[GitHub] [tvm] wzelang removed a comment on issue #7930: "CRT not found. Ensure you have built the standalone_crt target and try again"

wzelang removed a comment on issue #7930: URL: https://github.com/apache/tvm/issues/7930#issuecomment-827254116 when I try "../tests/scripts/task_cpp_unittest.sh" for testing TVM. It turned out "CRT not found. Ensure you have built the standalone_crt target and try again". I followed TV

[GitHub] [tvm] wzelang commented on issue #7930: "CRT not found. Ensure you have built the standalone_crt target and try again"

wzelang commented on issue #7930: URL: https://github.com/apache/tvm/issues/7930#issuecomment-827254116 when I try "../tests/scripts/task_cpp_unittest.sh" for testing TVM. It turned out "CRT not found. Ensure you have built the standalone_crt target and try again". I followed TVM INSTAL

[GitHub] [tvm] wzelang closed issue #7930: crt

wzelang closed issue #7930: URL: https://github.com/apache/tvm/issues/7930 -- 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. For queries about this service, please contact

[GitHub] [tvm] wzelang opened a new issue #7930: crt

wzelang opened a new issue #7930: URL: https://github.com/apache/tvm/issues/7930 Thanks for participating in the TVM community! We use https://discuss.tvm.ai for any general usage questions and discussions. The issue tracker is used for actionable items such as feature proposals discussion

[GitHub] [tvm] xqdan commented on a change in pull request #7923: [TensorIR][PASS][M1c] CompactBufferAllocation

xqdan commented on a change in pull request #7923: URL: https://github.com/apache/tvm/pull/7923#discussion_r620763914 ## File path: python/tvm/tir/transform/transform.py ## @@ -560,3 +560,53 @@ def PlanAndUpdateBufferAllocationLocation(): The result pass """

[GitHub] [tvm] Hzfengsy commented on a change in pull request #7923: [TensorIR][PASS][M1c] CompactBufferAllocation

Hzfengsy commented on a change in pull request #7923: URL: https://github.com/apache/tvm/pull/7923#discussion_r620755304 ## File path: include/tvm/tir/stmt.h ## @@ -991,13 +992,22 @@ class BufferRegion : public ObjectRef { TVM_DLL explicit BufferRegion(Buffer buffer, Array r

[GitHub] [tvm] jwfromm commented on pull request #7929: [Topi] Fix arm_cpu bitserial schedule with elemwise ops.

jwfromm commented on pull request #7929: URL: https://github.com/apache/tvm/pull/7929#issuecomment-827224609 Yes, just added a test that confirms the issue is resolved. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use

[GitHub] [tvm] jwfromm commented on issue #7909: [arm_cpu] bitserial conv2d TOPI schedule does not support post elementwise ops

jwfromm commented on issue #7909: URL: https://github.com/apache/tvm/issues/7909#issuecomment-827218332 fixed in #7929 -- 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. For

[GitHub] [tvm] jwfromm commented on pull request #7929: [Topi] Fix arm_cpu bitserial schedule with elemwise ops.

jwfromm commented on pull request #7929: URL: https://github.com/apache/tvm/pull/7929#issuecomment-827218254 @bernhardklein can you give this PR a shot to confirm everything now works as expected? -- This is an automated message from the Apache Git Service. To respond to the message, ple

[GitHub] [tvm] jwfromm opened a new pull request #7929: [Topi] Fix arm_cpu bitserial schedule with elemwise ops.

jwfromm opened a new pull request #7929: URL: https://github.com/apache/tvm/pull/7929 As noted in #7909, the arm bitserial_conv2d schedule has issues with fused elemwise operators. This turned out to be due to a simple mistake that this PR corrects. -- This is an automated message from

[GitHub] [tvm] comaniac commented on a change in pull request #7923: [TensorIR][PASS][M1c] CompactBufferAllocation

comaniac commented on a change in pull request #7923: URL: https://github.com/apache/tvm/pull/7923#discussion_r620643414 ## File path: include/tvm/tir/transform.h ## @@ -360,6 +360,52 @@ TVM_DLL Pass LowerInitBlock(); */ TVM_DLL Pass PlanAndUpdateBufferAllocationLocation();

[GitHub] [tvm] AndrewZhaoLuo opened a new pull request #7928: [WIP] Support dilations in pooling operators

AndrewZhaoLuo opened a new pull request #7928: URL: https://github.com/apache/tvm/pull/7928 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

[GitHub] [tvm] mehrdadh commented on a change in pull request #7903: [Graph Executor Debugger] Fix parameter dump

mehrdadh commented on a change in pull request #7903: URL: https://github.com/apache/tvm/pull/7903#discussion_r620697672 ## File path: src/runtime/graph_executor/debug/graph_executor_debug.cc ## @@ -210,65 +210,42 @@ class GraphExecutorDebug : public GraphExecutor { return

[GitHub] [tvm] tkonolige commented on a change in pull request #7927: [TOPI][RELAY][ONNX] Scatter ND

tkonolige commented on a change in pull request #7927: URL: https://github.com/apache/tvm/pull/7927#discussion_r620690363 ## File path: python/tvm/topi/cuda/scatter.py ## @@ -723,7 +723,7 @@ def update_func(dst_ptr, dst_index, update): return out -def scatter_nd(data,

[GitHub] [tvm] mbrookhart opened a new pull request #7927: [TOPI][RELAY][ONNX] Scatter ND

mbrookhart opened a new pull request #7927: URL: https://github.com/apache/tvm/pull/7927 This PR refactors the relay scatter_nd op in two ways: 1) scatter_nd now takes in a data tensor, and updates are scattered based on indices. The previous API assumed zero-initialization for the data.

[GitHub] [tvm] areusch commented on a change in pull request #7903: [Graph Executor Debugger] Fix parameter dump

areusch commented on a change in pull request #7903: URL: https://github.com/apache/tvm/pull/7903#discussion_r620669445 ## File path: src/runtime/graph_executor/debug/graph_executor_debug.cc ## @@ -210,65 +210,42 @@ class GraphExecutorDebug : public GraphExecutor { return

[GitHub] [tvm] tqchen commented on pull request #7918: [RUNTIME][BUGFIX] Fix DSO module problem when its parent get destructed.

tqchen commented on pull request #7918: URL: https://github.com/apache/tvm/pull/7918#issuecomment-827141931 First of all, I do not think such a hierachy is useful. In terms of the naming overlap, while it might be possible to do so in the compiler for artifacts checks to the extent when po

[GitHub] [tvm] areusch commented on pull request #7918: [RUNTIME][BUGFIX] Fix DSO module problem when its parent get destructed.

areusch commented on pull request #7918: URL: https://github.com/apache/tvm/pull/7918#issuecomment-827106711 @tqchen the practical ramifications of this are that if somehow CUDA module had code to `GetFunction("foo", query_imports=true)` and `dso0` defined that function, the module might b

[GitHub] [tvm] tqchen edited a comment on pull request #7918: [RUNTIME][BUGFIX] Fix DSO module problem when its parent get destructed.

tqchen edited a comment on pull request #7918: URL: https://github.com/apache/tvm/pull/7918#issuecomment-827091366 This is only hypothetical, but we can create a module that looks like this in the import chain:`dso0 <- cuda <-dso1`. Of course right now I cannot think of a practical setting

[GitHub] [tvm] tqchen edited a comment on pull request #7918: [RUNTIME][BUGFIX] Fix DSO module problem when its parent get destructed.

tqchen edited a comment on pull request #7918: URL: https://github.com/apache/tvm/pull/7918#issuecomment-827091366 This is only hypothetical, but we can create a module that looks like this in the import chain:`dso0 <- cuda <-dso1`. Of course right now I cannot think of a practical setting

[GitHub] [tvm] tqchen commented on pull request #7918: [RUNTIME][BUGFIX] Fix DSO module problem when its parent get destructed.

tqchen commented on pull request #7918: URL: https://github.com/apache/tvm/pull/7918#issuecomment-827091366 This is only hypothetical, but we can create a module that looks like this in the import chain:`dso0 <- cuda <-dso1` -- This is an automated message from the Apache Git Service. To

[GitHub] [tvm] giuseros commented on a change in pull request #7785: [AOT] Introducing AOT in TVM

giuseros commented on a change in pull request #7785: URL: https://github.com/apache/tvm/pull/7785#discussion_r620586188 ## File path: python/tvm/relay/build_module.py ## @@ -213,7 +218,7 @@ def _build_module_no_factory(mod, target=None, target_host=None, params=None, mo

[GitHub] [tvm] giuseros commented on a change in pull request #7785: [AOT] Introducing AOT in TVM

giuseros commented on a change in pull request #7785: URL: https://github.com/apache/tvm/pull/7785#discussion_r620584765 ## File path: src/target/source/source_module.cc ## @@ -191,17 +192,36 @@ class CSourceCrtMetadataModuleNode : public runtime::ModuleNode { << "}

[GitHub] [tvm] giuseros commented on a change in pull request #7785: [AOT] Introducing AOT in TVM

giuseros commented on a change in pull request #7785: URL: https://github.com/apache/tvm/pull/7785#discussion_r620580137 ## File path: src/target/source/codegen_source_base.h ## @@ -155,7 +156,8 @@ runtime::Module CSourceModuleCreate(const String& code, const String& fmt, */

[GitHub] [tvm] huajsj commented on pull request #7892: [Runtime]Pipeline Compute Graph With New Subgraph Executor

huajsj commented on pull request #7892: URL: https://github.com/apache/tvm/pull/7892#issuecomment-827073131 Thanks @tqchen for the follow up, I proposed a RFC(https://discuss.tvm.apache.org/t/rfc-compute-graph-pipeline-with-new-subgraph-executor/9839) to explain the motivation and solution

[GitHub] [tvm] areusch commented on pull request #7918: [RUNTIME][BUGFIX] Fix DSO module problem when its parent get destructed.

areusch commented on pull request #7918: URL: https://github.com/apache/tvm/pull/7918#issuecomment-827067378 @tqchen if the import tree has no cycles to begin with, could you give an example where merging could create such a cycle? I think one of the children of a DSO-exportable module wou

[GitHub] [tvm] mehrdadh commented on a change in pull request #7903: [Graph Executor Debugger] Fix parameter dump

mehrdadh commented on a change in pull request #7903: URL: https://github.com/apache/tvm/pull/7903#discussion_r620555111 ## File path: src/runtime/graph_executor/debug/graph_executor_debug.cc ## @@ -282,17 +324,17 @@ class GraphExecutorDebug : public GraphExecutor { PackedFunc

[GitHub] [tvm] areusch commented on a change in pull request #7785: [AOT] Introducing AOT in TVM

areusch commented on a change in pull request #7785: URL: https://github.com/apache/tvm/pull/7785#discussion_r620537842 ## File path: src/relay/backend/build_module.cc ## @@ -131,14 +164,14 @@ class RelayBuildModule : public runtime::ModuleNode { PackedFunc GetFunction(const

[GitHub] [tvm] tqchen edited a comment on pull request #7918: [RUNTIME][BUGFIX] Fix DSO module problem when its parent get destructed.

tqchen edited a comment on pull request #7918: URL: https://github.com/apache/tvm/pull/7918#issuecomment-827051612 Thanks @areusch, there are indeed no cycles in the import relation before we export, and a cycles will only be created when we try to merge all dso modules into one. The load

[GitHub] [tvm] tqchen edited a comment on pull request #7918: [RUNTIME][BUGFIX] Fix DSO module problem when its parent get destructed.

tqchen edited a comment on pull request #7918: URL: https://github.com/apache/tvm/pull/7918#issuecomment-827051612 Thanks @areusch, there is indeed no cycles in the import tree before we export, and the cycle will only be created when we try to merge dso modules. The load process always a

[GitHub] [tvm] tqchen edited a comment on pull request #7918: [RUNTIME][BUGFIX] Fix DSO module problem when its parent get destructed.

tqchen edited a comment on pull request #7918: URL: https://github.com/apache/tvm/pull/7918#issuecomment-827051612 Thanks @areusch, there is indeed no cycles in the import tree before we export, and the cycle will only be created when we try to merge dso modules. The load process always a

[GitHub] [tvm] tqchen commented on pull request #7918: [RUNTIME][BUGFIX] Fix DSO module problem when its parent get destructed.

tqchen commented on pull request #7918: URL: https://github.com/apache/tvm/pull/7918#issuecomment-827051612 Thanks @areusch, there is indeed no cycles in the import tree before we export, and the cycle will only be created when we try to merge dso modules -- This is an automated message

[GitHub] [tvm] tqchen commented on a change in pull request #7918: [RUNTIME][BUGFIX] Fix DSO module problem when its parent get destructed.

tqchen commented on a change in pull request #7918: URL: https://github.com/apache/tvm/pull/7918#discussion_r620538434 ## File path: src/target/codegen.cc ## @@ -110,32 +115,100 @@ class ModuleSerializer { // invariance: root module is always at location 0. // The modul

[GitHub] [tvm] tqchen commented on a change in pull request #7918: [RUNTIME][BUGFIX] Fix DSO module problem when its parent get destructed.

tqchen commented on a change in pull request #7918: URL: https://github.com/apache/tvm/pull/7918#discussion_r620534314 ## File path: src/runtime/library_module.cc ## @@ -125,10 +125,11 @@ Module LoadModuleFromBinary(const std::string& type_key, dmlc::Stream* stream) { * \bri

[GitHub] [tvm] tqchen commented on a change in pull request #7918: [RUNTIME][BUGFIX] Fix DSO module problem when its parent get destructed.

tqchen commented on a change in pull request #7918: URL: https://github.com/apache/tvm/pull/7918#discussion_r620534032 ## File path: src/runtime/library_module.cc ## @@ -143,14 +144,20 @@ runtime::Module ProcessModuleBlob(const char* mblob, ObjectPtr lib) { std::vector modu

[GitHub] [tvm] areusch commented on a change in pull request #7918: [RUNTIME][BUGFIX] Fix DSO module problem when its parent get destructed.

areusch commented on a change in pull request #7918: URL: https://github.com/apache/tvm/pull/7918#discussion_r620461701 ## File path: src/runtime/library_module.cc ## @@ -143,14 +144,20 @@ runtime::Module ProcessModuleBlob(const char* mblob, ObjectPtr lib) { std::vector mod

[GitHub] [tvm] mehrdadh commented on pull request #7915: [microTVM] Refactor zephyr installation + Update Zephyr RVM doc

mehrdadh commented on pull request #7915: URL: https://github.com/apache/tvm/pull/7915#issuecomment-827006956 @areusch PTAL. Thanks! -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specif

[GitHub] [tvm] Mousius commented on a change in pull request #7926: Use new SBT Debian Repo before bintray is shutdown

Mousius commented on a change in pull request #7926: URL: https://github.com/apache/tvm/pull/7926#discussion_r620461475 ## File path: docker/install/ubuntu_install_sbt.sh ## @@ -25,7 +25,7 @@ set -o pipefail apt-get update && apt-get install -y apt-transport-https # Install

[GitHub] [tvm] areusch commented on a change in pull request #7917: allow libbacktrace to be used when cross compiling the runtime

areusch commented on a change in pull request #7917: URL: https://github.com/apache/tvm/pull/7917#discussion_r620458713 ## File path: cmake/libs/Libbacktrace.cmake ## @@ -21,7 +21,16 @@ ExternalProject_Add(project_libbacktrace SOURCE_DIR ${CMAKE_CURRENT_LIST_DIR}/../../3rdpa

[GitHub] [tvm] eed3si9n commented on a change in pull request #7926: Use new SBT Debian Repo before bintray is shutdown

eed3si9n commented on a change in pull request #7926: URL: https://github.com/apache/tvm/pull/7926#discussion_r620452728 ## File path: docker/install/ubuntu_install_sbt.sh ## @@ -25,7 +25,7 @@ set -o pipefail apt-get update && apt-get install -y apt-transport-https # Instal

[GitHub] [tvm] eed3si9n commented on a change in pull request #7926: Use new SBT Debian Repo before bintray is shutdown

eed3si9n commented on a change in pull request #7926: URL: https://github.com/apache/tvm/pull/7926#discussion_r620451752 ## File path: docker/install/ubuntu_install_sbt.sh ## @@ -25,7 +25,7 @@ set -o pipefail apt-get update && apt-get install -y apt-transport-https # Instal

[GitHub] [tvm] eed3si9n commented on a change in pull request #7926: Use new SBT Debian Repo before bintray is shutdown

eed3si9n commented on a change in pull request #7926: URL: https://github.com/apache/tvm/pull/7926#discussion_r620451752 ## File path: docker/install/ubuntu_install_sbt.sh ## @@ -25,7 +25,7 @@ set -o pipefail apt-get update && apt-get install -y apt-transport-https # Instal

[GitHub] [tvm] Mousius opened a new pull request #7926: Use new SBT Debian Repo before bintray is shutdown

Mousius opened a new pull request #7926: URL: https://github.com/apache/tvm/pull/7926 We started seeing issues with building the TVM Docker images, and they stemmed from the SBT (Scala Build Tool) installation which was using bintray instead of a later SBT Debian URL. Thankfully bintray we

[GitHub] [tvm] tkonolige commented on pull request #7917: allow libbacktrace to be used when cross compiling the runtime

tkonolige commented on pull request #7917: URL: https://github.com/apache/tvm/pull/7917#issuecomment-826937491 btw, I tried these same changes when debugging #7705 but they did not fix it. -- This is an automated message from the Apache Git Service. To respond to the message, please log o

[GitHub] [tvm] giuseros commented on a change in pull request #7785: [AOT] Introducing AOT in TVM

giuseros commented on a change in pull request #7785: URL: https://github.com/apache/tvm/pull/7785#discussion_r620355300 ## File path: src/runtime/crt/memory/stack_allocator.c ## @@ -16,17 +16,22 @@ * specific language governing permissions and limitations * under the Licen

[GitHub] [tvm] tqchen closed pull request #7884: [ignore] Dummy commit to test ci

tqchen closed pull request #7884: URL: https://github.com/apache/tvm/pull/7884 -- 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. For queries about this service, please cont

[GitHub] [tvm] tqchen closed pull request #6764: Add decorator for extending a native generic function

tqchen closed pull request #6764: URL: https://github.com/apache/tvm/pull/6764 -- 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. For queries about this service, please cont

[GitHub] [tvm] tqchen commented on pull request #6764: Add decorator for extending a native generic function

tqchen commented on pull request #6764: URL: https://github.com/apache/tvm/pull/6764#issuecomment-826874089 close due to inactive status, feel free to reopen a new pr -- 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] mbaret opened a new pull request #7925: Add a 'rolling_buffer' scheduling primitive

mbaret opened a new pull request #7925: URL: https://github.com/apache/tvm/pull/7925 Support declaring a tensor as a rolling buffer so that it will be realized as a circular buffer without the need to recompute elements. For further detail you can take a look at this RFC: [TBA] -- T

[GitHub] [tvm] tqchen commented on pull request #7924: [DOCS] Remove stale Auto TensorCore CodeGen tutorial

tqchen commented on pull request #7924: URL: https://github.com/apache/tvm/pull/7924#issuecomment-826812856 cc @jcf94 @junrushao1994 @Hzfengsy -- 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 th

[GitHub] [tvm] tqchen opened a new pull request #7924: [DOCS] Remove stale Auto TensorCore CodeGen tutorial

tqchen opened a new pull request #7924: URL: https://github.com/apache/tvm/pull/7924 The auto tensorcore tutorial is stale and can cause confusions, removing this for now. We will recover a new tutorial once updated support for tensor core in available in tensorir -- This is an a

[GitHub] [tvm] tqchen commented on pull request #7373: Add Support for aten::index

tqchen commented on pull request #7373: URL: https://github.com/apache/tvm/pull/7373#issuecomment-826807958 gentle ping @codeislife99 @kevinthesun @comaniac -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL abo

[GitHub] [tvm] tqchen commented on pull request #7892: [Runtime]Pipeline Compute Graph With New Subgraph Executor

tqchen commented on pull request #7892: URL: https://github.com/apache/tvm/pull/7892#issuecomment-826806851 cc @comaniac @areusch @tmoreau89 -- 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] tqchen commented on issue #7010: [TEST][FLAKY] test_op_grad_level2.py::test_conv2d_grad.py

tqchen commented on issue #7010: URL: https://github.com/apache/tvm/issues/7010#issuecomment-826806166 closing for now as original flaky issue is fixed -- 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] tqchen closed issue #7010: [TEST][FLAKY] test_op_grad_level2.py::test_conv2d_grad.py

tqchen closed issue #7010: URL: https://github.com/apache/tvm/issues/7010 -- 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. For queries about this service, please contact I

[GitHub] [tvm] tqchen commented on issue #6389: Performance step about memory reduced for sgx

tqchen commented on issue #6389: URL: https://github.com/apache/tvm/issues/6389#issuecomment-826805692 Thanks for good discussions, let us move followups to https://discuss.tvm.apache.org/ -- This is an automated message from the Apache Git Service. To respond to the message, please log

[GitHub] [tvm] tqchen closed issue #6389: Performance step about memory reduced for sgx

tqchen closed issue #6389: URL: https://github.com/apache/tvm/issues/6389 -- 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. For queries about this service, please contact I

[GitHub] [tvm] tqchen closed issue #5827: [TEST][FLAKY][RUST] rust threading::tests::test_parallel_launch

tqchen closed issue #5827: URL: https://github.com/apache/tvm/issues/5827 -- 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. For queries about this service, please contact I

[GitHub] [tvm] tqchen commented on issue #7744: [CI] Fix Rust Docs Build

tqchen commented on issue #7744: URL: https://github.com/apache/tvm/issues/7744#issuecomment-826803653 cc @jroesch -- 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. For que

[GitHub] [tvm] tqchen closed issue #7921: TVM support for A100

tqchen closed issue #7921: URL: https://github.com/apache/tvm/issues/7921 -- 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. For queries about this service, please contact I

[GitHub] [tvm] tqchen commented on issue #7921: TVM support for A100

tqchen commented on issue #7921: URL: https://github.com/apache/tvm/issues/7921#issuecomment-826803346 Thanks for asking the question, the community uses for trouble shooting and discussions, please bring a new discussion topic on https://discuss.tvm.apache.org/, where more people will be

[GitHub] [tvm] tqchen commented on issue #7856: Undefined symbols when running task_cpp_unittest.sh on MacOS

tqchen commented on issue #7856: URL: https://github.com/apache/tvm/issues/7856#issuecomment-826802701 Thanks @CleverFat for reporting the problem, do you mind open a new PR to add the include of logging.cc? Thank you! cc @tkonolige @junrushao1994 -- This is an automated message

[GitHub] [tvm] tqchen closed issue #7878: PyTorch quantized group conv2d raise an error when converting to tvm

tqchen closed issue #7878: URL: https://github.com/apache/tvm/issues/7878 -- 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. For queries about this service, please contact I

[GitHub] [tvm] tqchen commented on issue #7878: PyTorch quantized group conv2d raise an error when converting to tvm

tqchen commented on issue #7878: URL: https://github.com/apache/tvm/issues/7878#issuecomment-826801482 Thanks for asking the question, the community uses for trouble shooting and discussions, please bring a new discussion topic on https://discuss.tvm.apache.org/, where more people will be

[GitHub] [tvm] tqchen closed issue #7798: How can I run TVM over RPC on Xilinx ultra96v2?

tqchen closed issue #7798: URL: https://github.com/apache/tvm/issues/7798 -- 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. For queries about this service, please contact I

[GitHub] [tvm] tqchen commented on issue #7798: How can I run TVM over RPC on Xilinx ultra96v2?

tqchen commented on issue #7798: URL: https://github.com/apache/tvm/issues/7798#issuecomment-826801077 Thanks for asking the question, the community uses for trouble shooting , please bring a new discussion topic on https://discuss.tvm.apache.org/, where more people will be able to watch a

[GitHub] [tvm] tqchen commented on issue #7909: [arm_cpu] bitserial conv2d TOPI schedule does not support post elementwise ops

tqchen commented on issue #7909: URL: https://github.com/apache/tvm/issues/7909#issuecomment-826800570 cc @jwfromm -- 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. For que

[GitHub] [tvm] tqchen closed issue #7857: pynq z1 fails to make the VTA, caused by the libbacktrace

tqchen closed issue #7857: URL: https://github.com/apache/tvm/issues/7857 -- 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. For queries about this service, please contact I

[GitHub] [tvm] tqchen commented on issue #7857: pynq z1 fails to make the VTA, caused by the libbacktrace

tqchen commented on issue #7857: URL: https://github.com/apache/tvm/issues/7857#issuecomment-826800109 Thanks for asking the question, the community uses for trouble shooting , please bring a new discussion topic on https://discuss.tvm.apache.org/, where more people will be able to watch a

[GitHub] [tvm] tqchen commented on issue #7811: BYODT- Error: my custom data type and float32 do not match in BroadcastRel

tqchen commented on issue #7811: URL: https://github.com/apache/tvm/issues/7811#issuecomment-826799639 Thanks for asking the question, the community uses for related support , please bring a new discussion topic on https://discuss.tvm.apache.org/, where more people will be able to watch an

[GitHub] [tvm] tqchen closed issue #7811: BYODT- Error: my custom data type and float32 do not match in BroadcastRel

tqchen closed issue #7811: URL: https://github.com/apache/tvm/issues/7811 -- 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. For queries about this service, please contact I

[GitHub] [tvm] tqchen commented on issue #7865: Check failed: (context != nullptr) is false: No OpenCL device

tqchen commented on issue #7865: URL: https://github.com/apache/tvm/issues/7865#issuecomment-826798959 Thanks for reporting the problem, please use https://discuss.tvm.apache.org/ for related topics. in your particular case, it could due to the fact that the opencl driver is not properly l

[GitHub] [tvm] tqchen closed issue #7865: Check failed: (context != nullptr) is false: No OpenCL device

tqchen closed issue #7865: URL: https://github.com/apache/tvm/issues/7865 -- 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. For queries about this service, please contact I

[GitHub] [tvm] tqchen closed issue #7849: [Bug][WASM] Cannot find function tvm.contrib.sort.argsort_nms in the imported modules or global registry

tqchen closed issue #7849: URL: https://github.com/apache/tvm/issues/7849 -- 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. For queries about this service, please contact I

[GitHub] [tvm] tqchen commented on issue #7849: [Bug][WASM] Cannot find function tvm.contrib.sort.argsort_nms in the imported modules or global registry

tqchen commented on issue #7849: URL: https://github.com/apache/tvm/issues/7849#issuecomment-826797678 Thanks for easking the question, please use https://discuss.tvm.apache.org/ for related questions. In your particular case, it could due to the original runtime bundle does not in

[GitHub] [tvm] tqchen closed issue #7855: AttributeError: module 'tvm.contrib.graph_runtime' has no attribute 'GraphModule'

tqchen closed issue #7855: URL: https://github.com/apache/tvm/issues/7855 -- 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. For queries about this service, please contact I

[GitHub] [tvm] tqchen commented on issue #7855: AttributeError: module 'tvm.contrib.graph_runtime' has no attribute 'GraphModule'

tqchen commented on issue #7855: URL: https://github.com/apache/tvm/issues/7855#issuecomment-826796391 Thanks for easking the question, please use https://discuss.tvm.apache.org/ for related questions -- This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] [tvm] tqchen closed issue #7596: [BUG][CodeGenC]“tvm_call_packed” crashes when build with target “c”

tqchen closed issue #7596: URL: https://github.com/apache/tvm/issues/7596 -- 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. For queries about this service, please contact I

[GitHub] [tvm] tqchen commented on issue #7596: [BUG][CodeGenC]“tvm_call_packed” crashes when build with target “c”

tqchen commented on issue #7596: URL: https://github.com/apache/tvm/issues/7596#issuecomment-826795852 Closed by https://github.com/apache/tvm/pull/7911. Thanks @zhuwenxi @leeexyz ! -- This is an automated message from the Apache Git Service. To respond to the message, please log on to G

[tvm] branch main updated (54fdcc5 -> 82fecbf)

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 54fdcc5 Enable StackVM in AutoTVM (#7897) add 82fecbf [CodeGenC] Fix bugs when calling extern functions (#7911) No

[GitHub] [tvm] tqchen commented on pull request #7911: [CodeGenC] Fix bugs when calling extern functions

tqchen commented on pull request #7911: URL: https://github.com/apache/tvm/pull/7911#issuecomment-826795578 Thanks @leeexyz ! -- 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] tqchen merged pull request #7911: [CodeGenC] Fix bugs when calling extern functions

tqchen merged pull request #7911: URL: https://github.com/apache/tvm/pull/7911 -- 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. For queries about this service, please cont

[GitHub] [tvm] tqchen commented on a change in pull request #7834: [OpenCL] Refactor cl_program generation

tqchen commented on a change in pull request #7834: URL: https://github.com/apache/tvm/pull/7834#discussion_r620245071 ## File path: src/runtime/opencl/opencl_common.h ## @@ -315,6 +315,13 @@ class OpenCLModuleNode : public ModuleNode { cl_kernel InstallKernel(cl::OpenCLWork

[tvm] branch main updated (f92d7fc -> 54fdcc5)

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 f92d7fc [BugFix]: Convert tuple to int (#7880) add 54fdcc5 Enable StackVM in AutoTVM (#7897) No new revisions were

[GitHub] [tvm] tqchen commented on pull request #7897: Enable StackVM in AutoTVM

tqchen commented on pull request #7897: URL: https://github.com/apache/tvm/pull/7897#issuecomment-826791607 THanks @PENGUINLIONG ! -- 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 co

[GitHub] [tvm] tqchen merged pull request #7897: Enable StackVM in AutoTVM

tqchen merged pull request #7897: URL: https://github.com/apache/tvm/pull/7897 -- 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. For queries about this service, please cont

[GitHub] [tvm] leandron commented on a change in pull request #7917: allow libbacktrace to be used when cross compiling the runtime

leandron commented on a change in pull request #7917: URL: https://github.com/apache/tvm/pull/7917#discussion_r620145762 ## File path: cmake/libs/Libbacktrace.cmake ## @@ -21,7 +21,16 @@ ExternalProject_Add(project_libbacktrace SOURCE_DIR ${CMAKE_CURRENT_LIST_DIR}/../../3rdp

[GitHub] [tvm] leandron commented on a change in pull request #7917: allow libbacktrace to be used when cross compiling the runtime

leandron commented on a change in pull request #7917: URL: https://github.com/apache/tvm/pull/7917#discussion_r620145762 ## File path: cmake/libs/Libbacktrace.cmake ## @@ -21,7 +21,16 @@ ExternalProject_Add(project_libbacktrace SOURCE_DIR ${CMAKE_CURRENT_LIST_DIR}/../../3rdp

[GitHub] [tvm] leandron commented on pull request #7917: allow libbacktrace to be used when cross compiling the runtime

leandron commented on pull request #7917: URL: https://github.com/apache/tvm/pull/7917#issuecomment-826689569 > On another note, i think an end to end example on inference on a remote machine would be great. i could probably contribute that end to end example if it does not exist, does

[GitHub] [tvm] mherkazandjian commented on a change in pull request #7917: allow libbacktrace to be used when cross compiling the runtime

mherkazandjian commented on a change in pull request #7917: URL: https://github.com/apache/tvm/pull/7917#discussion_r620138520 ## File path: cmake/libs/Libbacktrace.cmake ## @@ -21,7 +21,16 @@ ExternalProject_Add(project_libbacktrace SOURCE_DIR ${CMAKE_CURRENT_LIST_DIR}/../.

[GitHub] [tvm] leandron commented on a change in pull request #7917: allow libbacktrace to be used when cross compiling the runtime

leandron commented on a change in pull request #7917: URL: https://github.com/apache/tvm/pull/7917#discussion_r620130294 ## File path: cmake/libs/Libbacktrace.cmake ## @@ -21,7 +21,16 @@ ExternalProject_Add(project_libbacktrace SOURCE_DIR ${CMAKE_CURRENT_LIST_DIR}/../../3rdp

[GitHub] [tvm] vinceab commented on pull request #7922: [TVMC] add the support of the cross compiler options

vinceab commented on pull request #7922: URL: https://github.com/apache/tvm/pull/7922#issuecomment-826668571 > The change looks OK to me. Apart from my comments below, is there a chance to add a test case for this PR? I will do the update of the patch. Regarding the test of this P

[GitHub] [tvm] Hzfengsy opened a new pull request #7923: [TensorIR][PASS][M1c] CompactBufferAllocation

Hzfengsy opened a new pull request #7923: URL: https://github.com/apache/tvm/pull/7923 This PR is part of the TensorIR upstreaming effort (#7527) as one of the M1c stages, introducing the pass `ConvertBlocksToOpaque` and `CompactBufferAllocation` Pass `ConvertBlocksToOpaque` will su

[GitHub] [tvm] leandron commented on a change in pull request #7922: [TVMC] add the support of the cross compiler options

leandron commented on a change in pull request #7922: URL: https://github.com/apache/tvm/pull/7922#discussion_r620106210 ## File path: python/tvm/driver/tvmc/compiler.py ## @@ -271,7 +277,7 @@ def save_module(module_path, graph, lib, params, cross=None): The parameter

[GitHub] [tvm] manupa-arm commented on a change in pull request #7785: [AOT] Introducing AOT in TVM

manupa-arm commented on a change in pull request #7785: URL: https://github.com/apache/tvm/pull/7785#discussion_r620080936 ## File path: src/runtime/crt/memory/stack_allocator.c ## @@ -16,17 +16,22 @@ * specific language governing permissions and limitations * under the Lic

[GitHub] [tvm] manupa-arm commented on a change in pull request #7785: [AOT] Introducing AOT in TVM

manupa-arm commented on a change in pull request #7785: URL: https://github.com/apache/tvm/pull/7785#discussion_r620080936 ## File path: src/runtime/crt/memory/stack_allocator.c ## @@ -16,17 +16,22 @@ * specific language governing permissions and limitations * under the Lic

[GitHub] [tvm] manupa-arm commented on a change in pull request #7785: [AOT] Introducing AOT in TVM

manupa-arm commented on a change in pull request #7785: URL: https://github.com/apache/tvm/pull/7785#discussion_r620080936 ## File path: src/runtime/crt/memory/stack_allocator.c ## @@ -16,17 +16,22 @@ * specific language governing permissions and limitations * under the Lic

  1   2   >