[GitHub] [tvm] junrushao1994 commented on issue #9476: [Bug] Auto scheduler tuning on CPU fails on `batch_matmul(x, x)`

junrushao1994 commented on issue #9476: URL: https://github.com/apache/tvm/issues/9476#issuecomment-1102039958 This issue is going to be fixed with Relax. CC: @jinhongyii. Not sure if we have any bandwidth on AutoScheduler / Relay -- This is an automated message from the Apache Git

[GitHub] [tvm] junrushao1994 commented on pull request #11054: [TE][TIR] Enable CreatePrimFunc to properly handle 'layout_free_placeholder'

junrushao1994 commented on PR #11054: URL: https://github.com/apache/tvm/pull/11054#issuecomment-1102033520 CC: @Hzfengsy @jinhongyii -- 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] junrushao1994 opened a new pull request, #11054: [TE][TIR] Enable CreatePrimFunc to properly handle 'layout_free_placeholder'

junrushao1994 opened a new pull request, #11054: URL: https://github.com/apache/tvm/pull/11054 `layout_free_placeholder` is used to guide proper layout transformation on TE/TIR-level. However, previously it is not properly supported on upstream AutoTIR. This PR introduces legalization of

[GitHub] [tvm] masahi commented on a diff in pull request #11050: [TIR] Utility function to decide loop mapping for auto tensorization

masahi commented on code in PR #11050: URL: https://github.com/apache/tvm/pull/11050#discussion_r852574197 ## python/tvm/tir/schedule/analysis.py: ## @@ -56,3 +60,28 @@ def suggest_index_map( loops, predicate, ) + +

[tvm] branch last-successful updated (1542286826 -> 0cd4fa688c)

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 1542286826 [ci] Add branch protections to .asf.yaml (#10964) add c2eb51311c [COMMUNITY] New Committer

[GitHub] [tvm] MasterJH5574 commented on a diff in pull request #11050: [TIR] Utility function to decide loop mapping for auto tensorization

MasterJH5574 commented on code in PR #11050: URL: https://github.com/apache/tvm/pull/11050#discussion_r852547688 ## python/tvm/tir/schedule/analysis.py: ## @@ -56,3 +60,28 @@ def suggest_index_map( loops, predicate, ) + +

[GitHub] [tvm] huajsj commented on a diff in pull request #11018: [Hexagon][Runtime] Add QuRT thread pool backend

huajsj commented on code in PR #11018: URL: https://github.com/apache/tvm/pull/11018#discussion_r852556302 ## src/runtime/threading_backend.cc: ## @@ -34,13 +34,61 @@ #endif #if defined(__hexagon__) #include +#include +#include +#define HEXAGON_STACK_SIZE 65536 +#define

[GitHub] [tvm] BBuf commented on pull request #11052: Upgrad oneflow version

BBuf commented on PR #11052: URL: https://github.com/apache/tvm/pull/11052#issuecomment-1101941202 > Please keep pushing jobs until the build succeeds. Unfortunately there is a very frequent error `unknown parent image ID` #10832 ok. -- This is an automated message from the Apache

[GitHub] [tvm] junrushao1994 commented on a diff in pull request #11050: [TIR] Utility function to decide loop mapping for auto tensorization

junrushao1994 commented on code in PR #11050: URL: https://github.com/apache/tvm/pull/11050#discussion_r852550848 ## src/tir/schedule/analysis/analysis.cc: ## @@ -2028,5 +2034,107 @@ bool NeedsRFactorOrCrossThreadReduction(const tir::ScheduleState& self, // } }

[GitHub] [tvm] masahi commented on pull request #11052: Upgrad oneflow version

masahi commented on PR #11052: URL: https://github.com/apache/tvm/pull/11052#issuecomment-1101938624 Please keep pushing jobs until the build succeeds. Unfortunately there is a very frequent error `unknown parent image ID` https://github.com/apache/tvm/issues/10832 -- This is an

[GitHub] [tvm] Maximilianxu opened a new pull request, #11053: Let remote RPCModule get function recursively

Maximilianxu opened a new pull request, #11053: URL: https://github.com/apache/tvm/pull/11053 Same reason as we changed the get_function to recursive searching #10881 . This is useful when we treat the local module as a data segment and wrap it in another empty LLVM module to reuse

[GitHub] [tvm] masahi commented on a diff in pull request #10504: Complete pytorch grid_sample

masahi commented on code in PR #10504: URL: https://github.com/apache/tvm/pull/10504#discussion_r852532644 ## tests/python/relay/test_op_level5.py: ## @@ -1423,10 +1442,23 @@ def verify_grid_sample(data_shape, grid_shape, padding_mode="zeros"): )

[GitHub] [tvm] BBuf commented on pull request #11036: Add oneflow fronted tutorials

BBuf commented on PR #11036: URL: https://github.com/apache/tvm/pull/11036#issuecomment-1101915290 > Changing the docker script won't immediately change CI environments. We need to go through proper steps to update CI. > > Please send the docker script change in a separate PR.

[GitHub] [tvm] BBuf opened a new pull request, #11052: Upgrad oneflow version

BBuf opened a new pull request, #11052: URL: https://github.com/apache/tvm/pull/11052 Because the oneflow v0.7.0 interface change will cause the oneflow front-end test to fail, so comment out the tests in this pr to upgrade the onheflow version. The bug caused by the interface change has

[tvm] branch main updated (0cd4fa688c -> 4178617fcf)

This is an automated email from the ASF dual-hosted git repository. areusch pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from 0cd4fa688c Register PackedFuncObj with type registry. (#11039) add 4178617fcf [AOT] Support LLVM backend with C++

[GitHub] [tvm] areusch merged pull request #10753: [AOT] Support LLVM backend with C++ runtime

areusch merged PR #10753: URL: https://github.com/apache/tvm/pull/10753 -- 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] BBuf commented on pull request #11036: Add oneflow fronted tutorials

BBuf commented on PR #11036: URL: https://github.com/apache/tvm/pull/11036#issuecomment-1101910658 > Changing the docker script won't immediately change CI environments. We need to go through proper steps to update CI. > > Please send the docker script change in a separate PR.

[GitHub] [tvm] masahi commented on pull request #11036: Add oneflow fronted tutorials

masahi commented on PR #11036: URL: https://github.com/apache/tvm/pull/11036#issuecomment-1101910215 Changing the docker script won't immediately change CI environments. We need to go through proper steps to update CI. Please send the docker script change in a separate PR. -- This

[GitHub] [tvm] BBuf commented on pull request #11036: Add oneflow fronted tutorials

BBuf commented on PR #11036: URL: https://github.com/apache/tvm/pull/11036#issuecomment-1101905292 Hello, I want to upgrade the version of oneflow to 0.7.0 and solve the bug caused by the interface change. The local tests has passed. But the tests of running oneflow front-end in ci would

[tvm] branch main updated (f6171914a7 -> 0cd4fa688c)

This is an automated email from the ASF dual-hosted git repository. areusch pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from f6171914a7 [Hexagon] Adjust RPC read buffer size from python (#11022) add 0cd4fa688c Register PackedFuncObj with

[GitHub] [tvm] areusch merged pull request #11039: Register PackedFuncObj with type registry

areusch merged PR #11039: URL: https://github.com/apache/tvm/pull/11039 -- 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] areusch commented on pull request #11040: Downgrade some LOG(FATAL) in structural_equal

areusch commented on PR #11040: URL: https://github.com/apache/tvm/pull/11040#issuecomment-1101882555 actually on second thought, i think the problem debugging tests came from a double-check fail solved in #11039 . closing for now. -- This is an automated message from the Apache Git

[GitHub] [tvm] areusch closed pull request #11040: Downgrade some LOG(FATAL) in structural_equal

areusch closed pull request #11040: Downgrade some LOG(FATAL) in structural_equal URL: https://github.com/apache/tvm/pull/11040 -- 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.

[tvm] branch main updated (c2eb51311c -> f6171914a7)

This is an automated email from the ASF dual-hosted git repository. areusch pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from c2eb51311c [COMMUNITY] New Committer -- lhutton1 (#11049) add f6171914a7 [Hexagon] Adjust RPC read buffer size from

[GitHub] [tvm] areusch merged pull request #11022: [Hexagon] Adjust RPC read buffer size from python

areusch merged PR #11022: URL: https://github.com/apache/tvm/pull/11022 -- 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 (1542286826 -> c2eb51311c)

This is an automated email from the ASF dual-hosted git repository. jroesch pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from 1542286826 [ci] Add branch protections to .asf.yaml (#10964) add c2eb51311c [COMMUNITY] New Committer -- lhutton1

[GitHub] [tvm] jroesch merged pull request #11049: [COMMUNITY] New Committer -- lhutton1

jroesch merged PR #11049: URL: https://github.com/apache/tvm/pull/11049 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail:

[GitHub] [tvm] masahi commented on a diff in pull request #11050: [TIR] Utility function to decide loop mapping for auto tensorization

masahi commented on code in PR #11050: URL: https://github.com/apache/tvm/pull/11050#discussion_r852501884 ## src/tir/schedule/analysis/analysis.cc: ## @@ -2028,5 +2034,107 @@ bool NeedsRFactorOrCrossThreadReduction(const tir::ScheduleState& self, // } }

[GitHub] [tvm] sfvaroglu commented on pull request #11047: [ONNX] Reshape op

sfvaroglu commented on PR #11047: URL: https://github.com/apache/tvm/pull/11047#issuecomment-1101850446 @AndrewZhaoLuo mind taking a look? -- 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] sfvaroglu commented on a diff in pull request #10952: [QNN] Support input scale and zp of 1-element vector in qnn.conv2d_transpose

sfvaroglu commented on code in PR #10952: URL: https://github.com/apache/tvm/pull/10952#discussion_r852479246 ## python/tvm/relay/qnn/op/legalizations.py: ## @@ -92,12 +92,30 @@ def qnn_conv2d_transpose_legalize(attrs, inputs, types): # Collect the input exprs. data,

[GitHub] [tvm] AndrewZhaoLuo commented on a diff in pull request #10952: [QNN] Support input scale and zp of 1-element vector in qnn.conv2d_transpose

AndrewZhaoLuo commented on code in PR #10952: URL: https://github.com/apache/tvm/pull/10952#discussion_r852476788 ## python/tvm/relay/qnn/op/legalizations.py: ## @@ -92,12 +92,30 @@ def qnn_conv2d_transpose_legalize(attrs, inputs, types): # Collect the input exprs.

[GitHub] [tvm] AndrewZhaoLuo commented on a diff in pull request #10952: [QNN] Support input scale and zp of 1-element vector in qnn.conv2d_transpose

AndrewZhaoLuo commented on code in PR #10952: URL: https://github.com/apache/tvm/pull/10952#discussion_r852476644 ## python/tvm/relay/qnn/op/legalizations.py: ## @@ -91,12 +91,30 @@ def qnn_conv2d_transpose_legalize(attrs, inputs, types): # Collect the input exprs.

[GitHub] [tvm] tkonolige commented on a diff in pull request #10958: modify CMakeLists.txt and *.cmake in folder cmake to generate debuggable workspace and projects

tkonolige commented on code in PR #10958: URL: https://github.com/apache/tvm/pull/10958#discussion_r852469823 ## cmake/libs/Libbacktrace.cmake: ## @@ -68,3 +68,6 @@ set_property(TARGET libbacktrace PROPERTY IMPORTED_LOCATION

[GitHub] [tvm] sfvaroglu commented on a diff in pull request #10952: [QNN] Support input scale and zp of 1-element vector in qnn.conv2d_transpose

sfvaroglu commented on code in PR #10952: URL: https://github.com/apache/tvm/pull/10952#discussion_r852475648 ## python/tvm/relay/qnn/op/legalizations.py: ## @@ -91,12 +91,30 @@ def qnn_conv2d_transpose_legalize(attrs, inputs, types): # Collect the input exprs. data,

[GitHub] [tvm] tkonolige commented on a diff in pull request #11000: [Graph Debugger] Expose way to benchmark individual nodes.

tkonolige commented on code in PR #11000: URL: https://github.com/apache/tvm/pull/11000#discussion_r852464986 ## python/tvm/contrib/debugger/debug_executor.py: ## @@ -281,6 +282,42 @@ def run_individual(self, number, repeat=1, min_repeat_ms=0): ret =

[GitHub] [tvm] AndrewZhaoLuo commented on a diff in pull request #11000: [Graph Debugger] Expose way to benchmark individual nodes.

AndrewZhaoLuo commented on code in PR #11000: URL: https://github.com/apache/tvm/pull/11000#discussion_r852460978 ## python/tvm/contrib/debugger/debug_executor.py: ## @@ -281,6 +282,42 @@ def run_individual(self, number, repeat=1, min_repeat_ms=0): ret =

[GitHub] [tvm] masahi commented on a diff in pull request #10753: [AOT] Support LLVM backend with C++ runtime

masahi commented on code in PR #10753: URL: https://github.com/apache/tvm/pull/10753#discussion_r852461956 ## src/target/llvm/codegen_llvm.h: ## @@ -481,6 +518,8 @@ void CodeGenLLVM::AddFunctionsOrdered(IterType begin, IterType end, ConvType pfu return name_a < name_b;

[GitHub] [tvm] masahi commented on a diff in pull request #10753: [AOT] Support LLVM backend with C++ runtime

masahi commented on code in PR #10753: URL: https://github.com/apache/tvm/pull/10753#discussion_r852459814 ## src/target/llvm/codegen_llvm.h: ## @@ -481,6 +518,8 @@ void CodeGenLLVM::AddFunctionsOrdered(IterType begin, IterType end, ConvType pfu return name_a < name_b;

[tvm] branch ci-docker-staging updated (a4797a6a94 -> e293746128)

This is an automated email from the ASF dual-hosted git repository. mehrdadh pushed a change to branch ci-docker-staging in repository https://gitbox.apache.org/repos/asf/tvm.git from a4797a6a94 fix merge conflict add e293746128 rever removing the script No new revisions were added by

[GitHub] [tvm] CircleSpin commented on a diff in pull request #11047: [ONNX] Reshape op

CircleSpin commented on code in PR #11047: URL: https://github.com/apache/tvm/pull/11047#discussion_r852454098 ## include/tvm/topi/transform.h: ## @@ -321,6 +321,11 @@ inline Tensor reshape(const Tensor& x, Array newshape, std::string nam auto x_shape = x->shape; Array

[GitHub] [tvm] CircleSpin commented on a diff in pull request #11047: [ONNX] Reshape op

CircleSpin commented on code in PR #11047: URL: https://github.com/apache/tvm/pull/11047#discussion_r852453933 ## src/relay/op/tensor/transform.cc: ## @@ -634,11 +636,17 @@ Array InferNewShape(const Array& data_shape, const Attrs& oshape.push_back(newshape[i]);

[GitHub] [tvm] CircleSpin commented on a diff in pull request #11047: [ONNX] Reshape op

CircleSpin commented on code in PR #11047: URL: https://github.com/apache/tvm/pull/11047#discussion_r852453104 ## src/relay/op/make_op.h: ## @@ -67,7 +67,7 @@ Expr MakeReduce(Expr data, Array axis, bool keepdims, bool exclude, Str Expr MakeRepeat(Expr data, int repeats, int

[tvm] branch aluo/run-individual-node updated: formatting returned double str, move to typed packed func

This is an automated email from the ASF dual-hosted git repository. andrewzhaoluo pushed a commit to branch aluo/run-individual-node in repository https://gitbox.apache.org/repos/asf/tvm.git The following commit(s) were added to refs/heads/aluo/run-individual-node by this push: new

[GitHub] [tvm] kparzysz-quic commented on pull request #11022: [Hexagon] Adjust RPC read buffer size from python

kparzysz-quic commented on PR #11022: URL: https://github.com/apache/tvm/pull/11022#issuecomment-1101806589 > @kparzysz-quic I don't see a reason. Would std::stringbuf be slower? Probably, but it wouldn't affect model performance, since the only time it would run would be during the

[GitHub] [tvm] driazati opened a new pull request, #11051: [ci] Add local test re-run info

driazati opened a new pull request, #11051: URL: https://github.com/apache/tvm/pull/11051 This adds a note about `ci.py` with relevant tests when failures are detected. This should help advertise `ci.py` and make it more clear how to reproduce failures locally. This also adds a long link

[GitHub] [tvm] driazati closed pull request #10476: [ci] Add test local re-run command after failures

driazati closed pull request #10476: [ci] Add test local re-run command after failures URL: https://github.com/apache/tvm/pull/10476 -- 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 ci-docker-staging updated (902e40a163 -> a4797a6a94)

This is an automated email from the ASF dual-hosted git repository. mehrdadh pushed a change to branch ci-docker-staging in repository https://gitbox.apache.org/repos/asf/tvm.git from 902e40a163 Refactor hexagon test scripts add 29774bddd8 [RELAY][FRONTEND] Initial OneFlow frontend

[tvm] 01/02: initial

This is an automated email from the ASF dual-hosted git repository. andrewzhaoluo pushed a commit to branch aluo/run-individual-node in repository https://gitbox.apache.org/repos/asf/tvm.git commit f8718669194308921dc2d0a4b9137a2f739a45d0 Author: Andrew Zhao Luo AuthorDate: Tue Apr 12 15:42:43

[tvm] 02/02: secondary commit

This is an automated email from the ASF dual-hosted git repository. andrewzhaoluo pushed a commit to branch aluo/run-individual-node in repository https://gitbox.apache.org/repos/asf/tvm.git commit 85751bce55737cb4645898e40e247dc830bab60a Author: Andrew Zhao Luo AuthorDate: Tue Apr 12 16:29:44

[tvm] branch aluo/run-individual-node updated (ab98b150be -> 85751bce55)

This is an automated email from the ASF dual-hosted git repository. andrewzhaoluo pushed a change to branch aluo/run-individual-node in repository https://gitbox.apache.org/repos/asf/tvm.git omit ab98b150be secondary commit omit a1f26fc5b6 initial add a2d973d350 change Hexagon

[GitHub] [tvm] AndrewZhaoLuo commented on a diff in pull request #10952: [QNN] Support input scale and zp of 1-element vector in qnn.conv2d_transpose

AndrewZhaoLuo commented on code in PR #10952: URL: https://github.com/apache/tvm/pull/10952#discussion_r852437203 ## python/tvm/relay/qnn/op/legalizations.py: ## @@ -91,12 +91,30 @@ def qnn_conv2d_transpose_legalize(attrs, inputs, types): # Collect the input exprs.

[GitHub] [tvm] jwfromm commented on a diff in pull request #11047: [ONNX] Reshape op

jwfromm commented on code in PR #11047: URL: https://github.com/apache/tvm/pull/11047#discussion_r852434782 ## tests/python/frontend/onnx/test_forward.py: ## @@ -5077,7 +5077,7 @@ def verify_eyelike(indata): "test_reduce_sum_keepdims_random",

[GitHub] [tvm-rfcs] tkonolige commented on a diff in pull request #59: Remove tophub from being the default configuration when compiling

tkonolige commented on code in PR #59: URL: https://github.com/apache/tvm-rfcs/pull/59#discussion_r852435732 ## rfcs/0059-remove-tophub-from-default.md: ## @@ -0,0 +1,57 @@ +- Feature Name: remove_tophub_from_default +- Start Date: 2021-03-04 +- RFC PR:

[GitHub] [tvm] AndrewZhaoLuo commented on a diff in pull request #10952: [QNN] Support input scale and zp of 1-element vector in qnn.conv2d_transpose

AndrewZhaoLuo commented on code in PR #10952: URL: https://github.com/apache/tvm/pull/10952#discussion_r852434189 ## python/tvm/relay/qnn/op/legalizations.py: ## @@ -91,12 +91,30 @@ def qnn_conv2d_transpose_legalize(attrs, inputs, types): # Collect the input exprs.

[GitHub] [tvm] tkonolige commented on a diff in pull request #11000: [Graph Debugger] Expose way to benchmark individual nodes.

tkonolige commented on code in PR #11000: URL: https://github.com/apache/tvm/pull/11000#discussion_r852420219 ## python/tvm/contrib/debugger/debug_executor.py: ## @@ -281,6 +282,42 @@ def run_individual(self, number, repeat=1, min_repeat_ms=0): ret =

[GitHub] [tvm] mehrdadh commented on pull request #11022: [Hexagon] Adjust RPC read buffer size from python

mehrdadh commented on PR #11022: URL: https://github.com/apache/tvm/pull/11022#issuecomment-1101786592 @kparzysz-quic I don't see a reason. Would std::stringbuf be slower? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and

[GitHub] [tvm] tqchen opened a new pull request, #11049: [COMMUNITY] New Committer -- lhutton1

tqchen opened a new pull request, #11049: URL: https://github.com/apache/tvm/pull/11049 Dear community: Please join us to welcome @lhutton1 as a new committer. He has contributed lots of improvements to the microNPU, CMSIS-NN efforts as well as general support for AOT. -

[GitHub] [tvm] driazati opened a new pull request, #11048: [ci] Remove TensorCore node name

driazati opened a new pull request, #11048: URL: https://github.com/apache/tvm/pull/11048 This is the same as `GPU` so this consolidates on that to simplify reporting Thanks for contributing to TVM! Please refer to guideline https://tvm.apache.org/docs/contribute/ for useful

[GitHub] [tvm] CircleSpin opened a new pull request, #11047: [ONNX] Reshape op

CircleSpin opened a new pull request, #11047: URL: https://github.com/apache/tvm/pull/11047 It's been a journey. @jwfromm @MargaretQian @anwang2009 @sfvaroglu -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and

[GitHub] [tvm] areusch commented on a diff in pull request #10753: [AOT] Support LLVM backend with C++ runtime

areusch commented on code in PR #10753: URL: https://github.com/apache/tvm/pull/10753#discussion_r852422643 ## src/target/llvm/codegen_llvm.h: ## @@ -481,6 +518,8 @@ void CodeGenLLVM::AddFunctionsOrdered(IterType begin, IterType end, ConvType pfu return name_a < name_b;

[GitHub] [tvm] kparzysz-quic commented on pull request #11022: [Hexagon] Adjust RPC read buffer size from python

kparzysz-quic commented on PR #11022: URL: https://github.com/apache/tvm/pull/11022#issuecomment-1101770782 Is there a reason not to use `std::stringbuf` instead of the preallocated array? (the code for simulator does that) -- This is an automated message from the Apache Git Service. To

[GitHub] [tvm] kparzysz-quic commented on a diff in pull request #11022: [Hexagon] Adjust RPC read buffer size from python

kparzysz-quic commented on code in PR #11022: URL: https://github.com/apache/tvm/pull/11022#discussion_r852415525 ## src/runtime/hexagon/rpc/simulator/session.cc: ## @@ -1311,6 +1311,8 @@ detail::Optional SimulatorRPCChannel::to_nullptr(const detail::M

[GitHub] [tvm] kparzysz-quic opened a new pull request, #11046: [Hexagon] Pass stack size to simulator

kparzysz-quic opened a new pull request, #11046: URL: https://github.com/apache/tvm/pull/11046 Increase the default stack size to 256kB, since this is the minimum main thread stack size in QuRT on simulator. -- This is an automated message from the Apache Git Service. To respond to

[GitHub] [tvm] kparzysz-quic commented on a diff in pull request #10967: [rpc] Implemented rpc logging

kparzysz-quic commented on code in PR #10967: URL: https://github.com/apache/tvm/pull/10967#discussion_r852412804 ## src/runtime/minrpc/minrpc_server.h: ## @@ -524,94 +523,249 @@ class MinRPCServer { int call_ecode = TVMSetStream(dev.device_type, dev.device_id, handle);

[GitHub] [tvm] gromero commented on pull request #10504: Complete pytorch grid_sample

gromero commented on PR #10504: URL: https://github.com/apache/tvm/pull/10504#issuecomment-1101764931 @masahi I did a first pass on this and it looks alright, but I'm not totally confident I can review it thoroughly. Now that tests passed, could we get reviews from you? -- This is an

[GitHub] [tvm] driazati opened a new pull request, #11045: [docs] Add lightweight docs image

driazati opened a new pull request, #11045: URL: https://github.com/apache/tvm/pull/11045 This image includes everything necessary to build the docs without any tutorials and is just about 1.5 GB which is significantly less than the CPU/GPU images. TODO: * PR to

[GitHub] [tvm] sfvaroglu commented on a diff in pull request #10952: [QNN] Support input scale and zp of 1-element vector in qnn.conv2d_transpose

sfvaroglu commented on code in PR #10952: URL: https://github.com/apache/tvm/pull/10952#discussion_r852397609 ## python/tvm/relay/qnn/op/legalizations.py: ## @@ -91,12 +91,30 @@ def qnn_conv2d_transpose_legalize(attrs, inputs, types): # Collect the input exprs. data,

[GitHub] [tvm] alanmacd opened a new pull request, #11044: [microTVM] Add support for host-driven AoT Executor

alanmacd opened a new pull request, #11044: URL: https://github.com/apache/tvm/pull/11044 It is currently not possible to use the AoT executor with microTVM over the RPC channel. Add support for host-driven AoT executor using the generated AoT metadata to allocate working buffer memory for

[tvm] branch last-successful updated (28aac0e48f -> 1542286826)

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 28aac0e48f [Hexagon] Deprecate USE_HEXAGON_DEVICE, introduce USE_HEXAGON (#11025) add 1e524d0670 [ci]

[GitHub] [tvm] guberti opened a new pull request, #11043: Better version handling for Arduino

guberti opened a new pull request, #11043: URL: https://github.com/apache/tvm/pull/11043 This pull request contains three small changes: - Fix bug allowing `microTVM` to be used with Arduino version `v0.20` and above (see changes to `_parse_connected_boards`) and adds relevant unit

[GitHub] [tvm] mkatanbaf commented on a diff in pull request #10967: [rpc] Implemented rpc logging

mkatanbaf commented on code in PR #10967: URL: https://github.com/apache/tvm/pull/10967#discussion_r852333108 ## src/runtime/minrpc/minrpc_server_logging.h: ## @@ -0,0 +1,150 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] mkatanbaf commented on a diff in pull request #10967: [rpc] Implemented rpc logging

mkatanbaf commented on code in PR #10967: URL: https://github.com/apache/tvm/pull/10967#discussion_r852332951 ## src/runtime/minrpc/minrpc_logger.h: ## @@ -0,0 +1,219 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] mkatanbaf commented on a diff in pull request #10967: [rpc] Implemented rpc logging

mkatanbaf commented on code in PR #10967: URL: https://github.com/apache/tvm/pull/10967#discussion_r852332759 ## src/runtime/minrpc/minrpc_logger.h: ## @@ -0,0 +1,219 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] mkatanbaf commented on a diff in pull request #10967: [rpc] Implemented rpc logging

mkatanbaf commented on code in PR #10967: URL: https://github.com/apache/tvm/pull/10967#discussion_r852332550 ## src/runtime/minrpc/minrpc_logger.h: ## @@ -0,0 +1,219 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] mkatanbaf commented on a diff in pull request #10967: [rpc] Implemented rpc logging

mkatanbaf commented on code in PR #10967: URL: https://github.com/apache/tvm/pull/10967#discussion_r852332312 ## src/runtime/minrpc/minrpc_logger.cc: ## @@ -0,0 +1,307 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] mkatanbaf commented on a diff in pull request #10967: [rpc] Implemented rpc logging

mkatanbaf commented on code in PR #10967: URL: https://github.com/apache/tvm/pull/10967#discussion_r852332133 ## src/runtime/minrpc/minrpc_logger.cc: ## @@ -0,0 +1,307 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] mkatanbaf commented on a diff in pull request #10967: [rpc] Implemented rpc logging

mkatanbaf commented on code in PR #10967: URL: https://github.com/apache/tvm/pull/10967#discussion_r852331956 ## src/runtime/minrpc/minrpc_logger.cc: ## @@ -0,0 +1,307 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] mkatanbaf commented on a diff in pull request #10967: [rpc] Implemented rpc logging

mkatanbaf commented on code in PR #10967: URL: https://github.com/apache/tvm/pull/10967#discussion_r852330859 ## src/runtime/minrpc/minrpc_logger.cc: ## @@ -0,0 +1,307 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] mkatanbaf commented on a diff in pull request #10967: [rpc] Implemented rpc logging

mkatanbaf commented on code in PR #10967: URL: https://github.com/apache/tvm/pull/10967#discussion_r852330594 ## src/runtime/minrpc/minrpc_intrfc.h: ## @@ -0,0 +1,63 @@ +/* Review Comment: done -- This is an automated message from the Apache Git Service. To respond to

[tvm] branch last-successful updated (240fac437c -> 28aac0e48f)

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 240fac437c Fix typo in tutorial doc (#10974) add 28aac0e48f [Hexagon] Deprecate USE_HEXAGON_DEVICE,

[tvm] branch main updated: [ci] Add branch protections to .asf.yaml (#10964)

This is an automated email from the ASF dual-hosted git repository. areusch pushed a commit to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git The following commit(s) were added to refs/heads/main by this push: new 1542286826 [ci] Add branch protections to .asf.yaml

[GitHub] [tvm] areusch merged pull request #10964: [ci] Add branch protections to .asf.yaml

areusch merged PR #10964: URL: https://github.com/apache/tvm/pull/10964 -- 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: [ci] Migrate all test steps to macros (#10968)

This is an automated email from the ASF dual-hosted git repository. areusch pushed a commit to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git The following commit(s) were added to refs/heads/main by this push: new 1e524d0670 [ci] Migrate all test steps to macros

[GitHub] [tvm] areusch merged pull request #10968: [ci] Migrate all test steps to macros

areusch merged PR #10968: URL: https://github.com/apache/tvm/pull/10968 -- 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 opened a new pull request, #11042: [onnx] Relax tolerance for qlinearleakyrelu test

driazati opened a new pull request, #11042: URL: https://github.com/apache/tvm/pull/11042 This has failed on `main`: https://ci.tlcpack.ai/blue/organizations/jenkins/tvm/detail/main/3068/tests -- This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] [tvm] areusch opened a new pull request, #11041: Add vlogging for type-table registration

areusch opened a new pull request, #11041: URL: https://github.com/apache/tvm/pull/11041 This PR makes it easy to determine the order in which `GetOrAllocRuntimeTypeIndex` is called and to make sure it is called. Adding as VLOG(3) since we mostly don't want to do this unless something is

[GitHub] [tvm] areusch opened a new pull request, #11040: Downgrade some LOG(FATAL) in structural_equal

areusch opened a new pull request, #11040: URL: https://github.com/apache/tvm/pull/11040 It seems like these should be at most `CHECK` or `LOG(ERROR)`; having them as `LOG(FATAL)` made it difficult to debug a pytest crashing on this code. -- This is an automated message from the Apache

[GitHub] [tvm] areusch opened a new pull request, #11039: Register PackedFuncObj with type registry

areusch opened a new pull request, #11039: URL: https://github.com/apache/tvm/pull/11039 * Fixes CHECK-fails when trying to use reflection dispatch with PackedFuncObj. * Triggered by trying to add PrettyPrint() to StructuralEqual checks. @junrushao1994 @cyx-6 -- This is an

[GitHub] [tvm] driazati commented on pull request #11026: [Hexagon] Refactor test scripts

driazati commented on PR #11026: URL: https://github.com/apache/tvm/pull/11026#issuecomment-1101568070 You’ll need to close this PR and make a new one from the branch under apache/tvm instead of the one in your tvm fork (Jenkins doesn’t read the Jenkinsfile from forked repos for security

[tvm] branch ci-docker-staging updated (e4303d6f40 -> 902e40a163)

This is an automated email from the ASF dual-hosted git repository. mehrdadh pushed a change to branch ci-docker-staging in repository https://gitbox.apache.org/repos/asf/tvm.git discard e4303d6f40 Validating new gpu image add 61a9269d85 [FIX] resolve int64/32 for AttrStmtNode (#10983)

[tvm] branch ci--docker-staging created (now 902e40a163)

This is an automated email from the ASF dual-hosted git repository. mehrdadh pushed a change to branch ci--docker-staging in repository https://gitbox.apache.org/repos/asf/tvm.git at 902e40a163 Refactor hexagon test scripts No new revisions were added by this update.

[GitHub] [tvm] AndrewZhaoLuo closed pull request #10987: [WIP] Graph debugger run individual node

AndrewZhaoLuo closed pull request #10987: [WIP] Graph debugger run individual node URL: https://github.com/apache/tvm/pull/10987 -- 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: [Hexagon] Deprecate USE_HEXAGON_DEVICE, introduce USE_HEXAGON (#11025)

This is an automated email from the ASF dual-hosted git repository. mehrdadh 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 28aac0e48f [Hexagon] Deprecate USE_HEXAGON_DEVICE,

[GitHub] [tvm] mehrdadh merged pull request #11025: [Hexagon] Deprecate USE_HEXAGON_DEVICE, introduce USE_HEXAGON

mehrdadh merged PR #11025: URL: https://github.com/apache/tvm/pull/11025 -- 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] mehrdadh commented on a diff in pull request #11016: [Hexagon] Add test for registered schedules

mehrdadh commented on code in PR #11016: URL: https://github.com/apache/tvm/pull/11016#discussion_r852235645 ## tests/python/contrib/test_hexagon/test_batch_matmul.py: ## @@ -0,0 +1,138 @@ +# Licensed to the Apache Software Foundation (ASF) under one Review Comment: makes

[GitHub] [tvm] mbs-octoml commented on a diff in pull request #11003: [VirtualMachine] Zero copy in set_input when input is DLTensor

mbs-octoml commented on code in PR #11003: URL: https://github.com/apache/tvm/pull/11003#discussion_r852230388 ## src/runtime/vm/vm.cc: ## @@ -303,13 +303,12 @@ void VirtualMachine::SetInputTensorWithIndex(std::vector& tensors, if (inp_tensor.type_code() ==

[GitHub] [tvm] mehrdadh commented on a diff in pull request #11016: [Hexagon] Add test for registered schedules

mehrdadh commented on code in PR #11016: URL: https://github.com/apache/tvm/pull/11016#discussion_r852218603 ## tests/scripts/task_build_hexagon_api.sh: ## @@ -19,8 +19,18 @@ set -e set -u +use_cache=false +if [ $# -ge 1 ] && [[ "$1" == "--use-cache" ]]; then +

[GitHub] [tvm] mehrdadh commented on a diff in pull request #11016: [Hexagon] Add test for registered schedules

mehrdadh commented on code in PR #11016: URL: https://github.com/apache/tvm/pull/11016#discussion_r852218129 ## src/runtime/hexagon/rpc/hexagon/rpc_server.cc: ## @@ -41,7 +41,7 @@ extern "C" { #include "hexagon_rpc.h" // TODO(mehrdadh): make this configurable. -#define

[GitHub] [tvm] mehrdadh commented on a diff in pull request #11016: [Hexagon] Add test for registered schedules

mehrdadh commented on code in PR #11016: URL: https://github.com/apache/tvm/pull/11016#discussion_r852217674 ## python/tvm/script/tir/__init__.pyi: ## @@ -226,6 +226,7 @@ def alloc_buffer( """ special_stmt - Reads/Writes """ + Review Comment: This is done by the clang

[GitHub] [tvm] mehrdadh commented on pull request #11026: [Hexagon] Refactor test scripts

mehrdadh commented on PR #11026: URL: https://github.com/apache/tvm/pull/11026#issuecomment-1101518725 @driazati looks like the CI is buster because it is using old Jenkinsfile. How can I fix this? -- This is an automated message from the Apache Git Service. To respond to the message,

  1   2   >