[GitHub] [tvm] comaniac commented on a change in pull request #7297: [FIX,AUTOTVM] Add flop counts to cublas

comaniac commented on a change in pull request #7297: URL: https://github.com/apache/tvm/pull/7297#discussion_r558761663 ## File path: python/tvm/topi/cuda/dense.py ## @@ -44,6 +44,8 @@ def dense_cublas(cfg, data, weight, bias=None, out_dtype=None): matmul =

[GitHub] [tvm] tkonolige commented on a change in pull request #7297: [FIX,AUTOTVM] Add flop counts to cublas

tkonolige commented on a change in pull request #7297: URL: https://github.com/apache/tvm/pull/7297#discussion_r558726514 ## File path: python/tvm/topi/cuda/dense.py ## @@ -44,6 +44,8 @@ def dense_cublas(cfg, data, weight, bias=None, out_dtype=None): matmul =

[GitHub] [tvm] comaniac commented on a change in pull request #7297: [FIX,AUTOTVM] Add flop counts to cublas

comaniac commented on a change in pull request #7297: URL: https://github.com/apache/tvm/pull/7297#discussion_r558723487 ## File path: python/tvm/topi/cuda/dense.py ## @@ -44,6 +44,8 @@ def dense_cublas(cfg, data, weight, bias=None, out_dtype=None): matmul =

[GitHub] [tvm] FrozenGene commented on pull request #6998: [TFLite] Strided slice handling of shrink_axis_mask improved

FrozenGene commented on pull request #6998: URL: https://github.com/apache/tvm/pull/6998#issuecomment-761277427 @d-smirnov I will review it today, thanks! This is an automated message from the Apache Git Service. To respond

[GitHub] [tvm] tkonolige opened a new pull request #7298: [TUTORIAL] Add gpu instructions and results to deploy_sparse

tkonolige opened a new pull request #7298: URL: https://github.com/apache/tvm/pull/7298 I've just mentioned that GPU + CUDA was supported and added some sample results. @jwfromm @binarybana This is an automated

[GitHub] [tvm] tkonolige opened a new pull request #7297: [FIX,AUTOTVM] Add flop counts to cublas

tkonolige opened a new pull request #7297: URL: https://github.com/apache/tvm/pull/7297 Fixes a bug when trying to autotune with cublas enabled. I've also added the op name to the error messages. @comaniac @merrymercy

[GitHub] [tvm] mdw-octoml commented on pull request #7296: Add QEMU setup to uTVM tutorial.

mdw-octoml commented on pull request #7296: URL: https://github.com/apache/tvm/pull/7296#issuecomment-761274179 > Thank you @mdw-octoml for improving TVM tutorials! > > Is the change in permissions of the ubuntu_install_qemu.sh file intentional? Yes because it is a script

[GitHub] [tvm] mdw-octoml commented on pull request #7296: Add QEMU setup to uTVM tutorial.

mdw-octoml commented on pull request #7296: URL: https://github.com/apache/tvm/pull/7296#issuecomment-761271750 @areusch @tmoreau89 This is an automated message from the Apache Git Service. To respond to the message, please

[GitHub] [tvm] mdw-octoml opened a new pull request #7296: Add QEMU setup to uTVM tutorial.

mdw-octoml opened a new pull request #7296: URL: https://github.com/apache/tvm/pull/7296 As it says on the tin. This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and

[tvm] branch main updated (637c9da -> ccccac2)

This is an automated email from the ASF dual-hosted git repository. zhic pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git. from 637c9da A few typo fixes in the uTVM design doc. (#7291) add ac2 Change const to used dtype if it is passed in

[GitHub] [tvm] zhiics merged pull request #7285: Change const to used dtype if it is passed in

zhiics merged pull request #7285: URL: https://github.com/apache/tvm/pull/7285 This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the

[tvm] branch main updated (b52267e -> 637c9da)

This is an automated email from the ASF dual-hosted git repository. zhic pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git. from b52267e [uTVM] Initial BYOC support with c-source module (#6950) add 637c9da A few typo fixes in the uTVM design

[GitHub] [tvm] zhiics merged pull request #7291: A few typo fixes in the uTVM design doc.

zhiics merged pull request #7291: URL: https://github.com/apache/tvm/pull/7291 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] zhiics commented on pull request #6950: [uTVM] Initial BYOC support with c-source module

zhiics commented on pull request #6950: URL: https://github.com/apache/tvm/pull/6950#issuecomment-761258851 Thanks @manupa-arm @areusch This is an automated message from the Apache Git Service. To respond to the message,

[tvm] branch main updated (2992e9b -> b52267e)

This is an automated email from the ASF dual-hosted git repository. zhic pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git. from 2992e9b [VM] Per-input, data dependence specification for shape func (#7210) add b52267e [uTVM] Initial BYOC

[GitHub] [tvm] zhiics merged pull request #6950: [uTVM] Initial BYOC support with c-source module

zhiics merged pull request #6950: URL: https://github.com/apache/tvm/pull/6950 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] d-smirnov commented on pull request #7293: [TFLite] Added ability to infer shapes for arguments

d-smirnov commented on pull request #7293: URL: https://github.com/apache/tvm/pull/7293#issuecomment-761226359 cc @anijain2305 This is an automated message from the Apache Git Service. To respond to the message, please log

[GitHub] [tvm] mbrookhart commented on pull request #7295: Fix an issue with dynamic functions overwritting call arg types

mbrookhart commented on pull request #7295: URL: https://github.com/apache/tvm/pull/7295#issuecomment-761220335 Yeah, it's a little weird. I'd like to run through CI just in case I introduced a bug somewhere, but once that's complete I can take a look at unifying the different versions of

[GitHub] [tvm] altanh commented on pull request #7295: Fix an issue with dynamic functions overwritting call arg types

altanh commented on pull request #7295: URL: https://github.com/apache/tvm/pull/7295#issuecomment-761217256 not the fault of this PR, but I've noticed some confusing and inconsistent usage of `src` and `dst` in different `Unify` functions (along with non-matching implementation argument

[GitHub] [tvm] tkonolige closed pull request #7235: [FIX] Fix make format to work with arbitrary upstream names

tkonolige closed pull request #7235: URL: https://github.com/apache/tvm/pull/7235 This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to

[tvm] branch main updated (4f1f591 -> 2992e9b)

This is an automated email from the ASF dual-hosted git repository. haichen pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git. from 4f1f591 Bring back numbered lists to TVM docs. (#7290) add 2992e9b [VM] Per-input, data dependence specification

[GitHub] [tvm] icemelon9 merged pull request #7210: [VM] Per-input, data dependence specification for shape func

icemelon9 merged pull request #7210: URL: https://github.com/apache/tvm/pull/7210 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

[GitHub] [tvm] icemelon9 commented on pull request #7210: [VM] Per-input, data dependence specification for shape func

icemelon9 commented on pull request #7210: URL: https://github.com/apache/tvm/pull/7210#issuecomment-761204133 Thanks @masahi @zhiics @mbrookhart This is an automated message from the Apache Git Service. To respond to the

[GitHub] [tvm] mbrookhart opened a new pull request #7295: Fix an issue with dynamic functions overwritting call arg types

mbrookhart opened a new pull request #7295: URL: https://github.com/apache/tvm/pull/7295 cc @jroesch @altanh @tmoreau89 This is an automated message from the Apache Git Service. To respond to the message, please log on to

[tvm] branch main updated (c3f50ff -> 4f1f591)

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 c3f50ff [µTVM] Add ST STM32F746 disco board to tflite tutorial script (#7254) add 4f1f591 Bring back numbered

[GitHub] [tvm] tqchen merged pull request #7290: Bring back numbered lists to TVM docs.

tqchen merged pull request #7290: URL: https://github.com/apache/tvm/pull/7290 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 pull request #7235: [FIX] Fix make format to work with arbitrary upstream names

tqchen commented on pull request #7235: URL: https://github.com/apache/tvm/pull/7235#issuecomment-761175149 Unfortunately these scripts are being used in the CI and CI needs `origin/main`. Let us just encourage developers to use ```bash ./tests/lint/git-clang-format.sh -i

[GitHub] [tvm] masahi commented on pull request #7210: [VM] Per-input, data dependence specification for shape func

masahi commented on pull request #7210: URL: https://github.com/apache/tvm/pull/7210#issuecomment-761173735 @icemelon9 it should be ready This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] [tvm] leandron opened a new pull request #7294: [TEST] Fix test_topi_batch_matmul_tensorcore.py:test_batch_matmul requirement

leandron opened a new pull request #7294: URL: https://github.com/apache/tvm/pull/7294 `test_topi_batch_matmul_tensorcore.py:test_batch_matmul` currently sets a requirement to "uses_gpu", which causes it to fail on cpu-only machine. This patch changes it to be "requires_tensorcore",

[GitHub] [tvm] d-smirnov opened a new pull request #7293: [TFLite] Added ability to infer shapes for arguments

d-smirnov opened a new pull request #7293: URL: https://github.com/apache/tvm/pull/7293 Added an ability to infer argument shapes if shapes are not present in TFLite files. The set of networks on which the patch was tested is internal to Arm. Any help with creating unit tests would be

[GitHub] [tvm] tkonolige opened a new issue #7292: [BUG] Passing wrong Tensor to tvm.build segfaults

tkonolige opened a new issue #7292: URL: https://github.com/apache/tvm/issues/7292 On d7a9a7c5bc5952e43db2ac0113fd40cc77453fbe, the following script segfaults: ```python import tvm import numpy as np def gen_ir(out_ptr): irb = tvm.tir.ir_builder.create() out =

[GitHub] [tvm] areusch commented on pull request #7266: [µTVM] Add TVMPlatformGenerateRandom, a non-cryptographic random number generator.

areusch commented on pull request #7266: URL: https://github.com/apache/tvm/pull/7266#issuecomment-761121252 @liangfu thanks for the review, please take another look when you have a minute This is an automated message from

[GitHub] [tvm] areusch commented on a change in pull request #7266: [µTVM] Add TVMPlatformGenerateRandom, a non-cryptographic random number generator.

areusch commented on a change in pull request #7266: URL: https://github.com/apache/tvm/pull/7266#discussion_r558513482 ## File path: tests/micro/qemu/zephyr-runtime/src/main.c ## @@ -161,6 +162,25 @@ tvm_crt_error_t TVMPlatformTimerStop(double* elapsed_time_seconds) {

[GitHub] [tvm] areusch commented on a change in pull request #7289: Generate requirements.txt from Python spec

areusch commented on a change in pull request #7289: URL: https://github.com/apache/tvm/pull/7289#discussion_r558498794 ## File path: python/gen_requirements.py ## @@ -0,0 +1,391 @@ +#!/usr/bin/env python3 +# Licensed to the Apache Software Foundation (ASF) under one +# or

[GitHub] [tvm] areusch commented on a change in pull request #7289: Generate requirements.txt from Python spec

areusch commented on a change in pull request #7289: URL: https://github.com/apache/tvm/pull/7289#discussion_r558498417 ## File path: python/gen_requirements.py ## @@ -0,0 +1,391 @@ +#!/usr/bin/env python3 +# Licensed to the Apache Software Foundation (ASF) under one +# or

[GitHub] [tvm] tkonolige commented on issue #7277: [TEST] Skip TensorCore TOPI test when it is not available

tkonolige commented on issue #7277: URL: https://github.com/apache/tvm/issues/7277#issuecomment-761112339 Yes, it should be `requires_tensorcore`. This is an automated message from the Apache Git Service. To respond to the

[GitHub] [tvm] tqchen commented on issue #7277: [TEST] Skip TensorCore TOPI test when it is not available

tqchen commented on issue #7277: URL: https://github.com/apache/tvm/issues/7277#issuecomment-761108796 I think `requires_tensorcore` is the right macro cc @tkonolige This is an automated message from the Apache Git Service.

[GitHub] [tvm] mdw-octoml commented on a change in pull request #7289: Generate requirements.txt from Python spec

mdw-octoml commented on a change in pull request #7289: URL: https://github.com/apache/tvm/pull/7289#discussion_r558479917 ## File path: python/gen_requirements.py ## @@ -0,0 +1,391 @@ +#!/usr/bin/env python3 +# Licensed to the Apache Software Foundation (ASF) under one +# or

[GitHub] [tvm] altanh commented on a change in pull request #7287: [PRNG] Add check to PRNG to make sure that unsigned integer arithmetic is wrapping

altanh commented on a change in pull request #7287: URL: https://github.com/apache/tvm/pull/7287#discussion_r558481865 ## File path: python/tvm/topi/random/kernel.py ## @@ -20,6 +20,8 @@ from ... import tir from ...tir import ir_builder +import numpy as np Review comment:

[GitHub] [tvm] tkonolige commented on a change in pull request #7287: [PRNG] Add check to PRNG to make sure that unsigned integer arithmetic is wrapping

tkonolige commented on a change in pull request #7287: URL: https://github.com/apache/tvm/pull/7287#discussion_r558479119 ## File path: python/tvm/topi/random/kernel.py ## @@ -20,6 +20,8 @@ from ... import tir from ...tir import ir_builder +import numpy as np Review

[GitHub] [tvm] areusch commented on pull request #7254: [µTVM] Add ST STM32F746 disco board to tflite tutorial script

areusch commented on pull request #7254: URL: https://github.com/apache/tvm/pull/7254#issuecomment-761094488 this will get updated on docs.tvm.ai when we regenerate the site, will probably happen later on today or tomorrow.

[GitHub] [tvm] altanh commented on a change in pull request #7287: [PRNG] Add check to PRNG to make sure that unsigned integer arithmetic is wrapping

altanh commented on a change in pull request #7287: URL: https://github.com/apache/tvm/pull/7287#discussion_r558477345 ## File path: python/tvm/topi/random/kernel.py ## @@ -20,6 +20,8 @@ from ... import tir from ...tir import ir_builder +import numpy as np Review comment:

[GitHub] [tvm] mdw-octoml commented on a change in pull request #7289: Generate requirements.txt from Python spec

mdw-octoml commented on a change in pull request #7289: URL: https://github.com/apache/tvm/pull/7289#discussion_r558478062 ## File path: python/gen_requirements.py ## @@ -0,0 +1,391 @@ +#!/usr/bin/env python3 +# Licensed to the Apache Software Foundation (ASF) under one +# or

[GitHub] [tvm] areusch commented on a change in pull request #7289: Generate requirements.txt from Python spec

areusch commented on a change in pull request #7289: URL: https://github.com/apache/tvm/pull/7289#discussion_r558477221 ## File path: python/gen_requirements.py ## @@ -0,0 +1,391 @@ +#!/usr/bin/env python3 +# Licensed to the Apache Software Foundation (ASF) under one +# or

[GitHub] [tvm] areusch commented on pull request #7289: Generate requirements.txt from Python spec

areusch commented on pull request #7289: URL: https://github.com/apache/tvm/pull/7289#issuecomment-761090330 thanks for the reviews so far everyone! one quick note: the actual list of dependencies (and their version constraints) are still WIP. so definitely feel free to comment on

[tvm] branch main updated: [µTVM] Add ST STM32F746 disco board to tflite tutorial script (#7254)

This is an automated email from the ASF dual-hosted git repository. moreau 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 c3f50ff [µTVM] Add ST STM32F746 disco board to tflite

[GitHub] [tvm] tmoreau89 merged pull request #7254: [µTVM] Add ST STM32F746 disco board to tflite tutorial script

tmoreau89 merged pull request #7254: URL: https://github.com/apache/tvm/pull/7254 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

[GitHub] [tvm] comaniac commented on pull request #7284: [ConvertLayout] Resize operator

comaniac commented on pull request #7284: URL: https://github.com/apache/tvm/pull/7284#issuecomment-761083068 @anijain2305 please fix the CI error that seems failed at an OD model. This is an automated message from the

[GitHub] [tvm] gromero commented on pull request #7254: [µTVM] Add ST STM32F746 disco board to tflite tutorial script

gromero commented on pull request #7254: URL: https://github.com/apache/tvm/pull/7254#issuecomment-761081715 @areusch you're right, it was just a matter of re-triggering the build. Tests seem ok now :) This is an automated

[GitHub] [tvm] comaniac commented on a change in pull request #7289: Generate requirements.txt from Python spec

comaniac commented on a change in pull request #7289: URL: https://github.com/apache/tvm/pull/7289#discussion_r558462379 ## File path: python/gen_requirements.py ## @@ -0,0 +1,391 @@ +#!/usr/bin/env python3 +# Licensed to the Apache Software Foundation (ASF) under one +# or

[GitHub] [tvm] mdw-octoml commented on pull request #7291: A few typo fixes in the uTVM design doc.

mdw-octoml commented on pull request #7291: URL: https://github.com/apache/tvm/pull/7291#issuecomment-761075052 @areusch Quick PR for you. This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] [tvm] mdw-octoml opened a new pull request #7291: A few typo fixes in the uTVM design doc.

mdw-octoml opened a new pull request #7291: URL: https://github.com/apache/tvm/pull/7291 A few small typo fixes in the uTVM design doc. This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] [tvm] areusch opened a new pull request #7290: Bring back numbered lists to TVM docs.

areusch opened a new pull request #7290: URL: https://github.com/apache/tvm/pull/7290 * Upstream fix in https://github.com/tlc-pack/tlcpack-sphinx-addon/commit/995178d81e6e38eabbc28da2b285b68583c88769 @tqchen @mdw-octoml

[GitHub] [tvm] tkonolige commented on a change in pull request #7289: Generate requirements.txt from Python spec

tkonolige commented on a change in pull request #7289: URL: https://github.com/apache/tvm/pull/7289#discussion_r558447476 ## File path: python/gen_requirements.py ## @@ -0,0 +1,391 @@ +#!/usr/bin/env python3 +# Licensed to the Apache Software Foundation (ASF) under one +# or

[GitHub] [tvm] d-smirnov commented on pull request #6998: [TFLite] Strided slice handling of shrink_axis_mask improved

d-smirnov commented on pull request #6998: URL: https://github.com/apache/tvm/pull/6998#issuecomment-761053826 How can we make a progress here? This is an automated message from the Apache Git Service. To respond to the

[GitHub] [tvm] leandron commented on a change in pull request #7289: Generate requirements.txt from Python spec

leandron commented on a change in pull request #7289: URL: https://github.com/apache/tvm/pull/7289#discussion_r558247110 ## File path: python/gen_requirements.py ## @@ -0,0 +1,391 @@ +#!/usr/bin/env python3 +# Licensed to the Apache Software Foundation (ASF) under one +# or

[GitHub] [tvm] leandron edited a comment on issue #7277: [TEST] Skip TensorCore TOPI test when it is not available

leandron edited a comment on issue #7277: URL: https://github.com/apache/tvm/issues/7277#issuecomment-761043520 (Posting here as I think it is related) When running TVM tests on a CPU only machine, I'm getting an error when running

[GitHub] [tvm] leandron edited a comment on issue #7277: [TEST] Skip TensorCore TOPI test when it is not available

leandron edited a comment on issue #7277: URL: https://github.com/apache/tvm/issues/7277#issuecomment-761043520 (Posting here as I think it is related) When running TVM tests on a CPU only machine, I'm getting an error when running

[GitHub] [tvm] leandron commented on issue #7277: [TEST] Skip TensorCore TOPI test when it is not available

leandron commented on issue #7277: URL: https://github.com/apache/tvm/issues/7277#issuecomment-761043520 (Posting here as I think it is related) When running TVM tests on a CPU only machine, I'm getting an error when running `/test_topi_batch_matmul_tensorcore.py:test_batch_matmul`:

[GitHub] [tvm] tqchen commented on a change in pull request #7084: [TIR] Support Return in TIR

tqchen commented on a change in pull request #7084: URL: https://github.com/apache/tvm/pull/7084#discussion_r558329105 ## File path: src/target/llvm/codegen_llvm.cc ## @@ -927,6 +927,12 @@ llvm::Value* CodeGenLLVM::CreateIntrinsic(const CallNode* op) {

[GitHub] [tvm] tqchen commented on issue #7246: [BUG][Tensorize] race condition when using "tvm.tir.call_packed()" in a parallel schedule.

tqchen commented on issue #7246: URL: https://github.com/apache/tvm/issues/7246#issuecomment-760961517 It will be generated in singele threaded as well. In that case the space will be allocated in the begining of the function, and i believe LLVM optimization should be able to collapse

[tvm] branch main updated: [BYOC][bugfix] Handle empty tuples in annotation pass (#7288)

This is an automated email from the ASF dual-hosted git repository. tqchen 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 4c5c086 [BYOC][bugfix] Handle empty tuples in

[GitHub] [tvm] tqchen merged pull request #7288: [BYOC][bugfix] Handle empty tuples in annotation pass

tqchen merged pull request #7288: URL: https://github.com/apache/tvm/pull/7288 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] liangfu commented on a change in pull request #7266: [µTVM] Add TVMPlatformGenerateRandom, a non-cryptographic random number generator.

liangfu commented on a change in pull request #7266: URL: https://github.com/apache/tvm/pull/7266#discussion_r558228081 ## File path: tests/micro/qemu/zephyr-runtime/src/main.c ## @@ -161,6 +162,25 @@ tvm_crt_error_t TVMPlatformTimerStop(double* elapsed_time_seconds) {

[GitHub] [tvm] manupa-arm commented on pull request #6950: [uTVM] Initial BYOC support with c-source module

manupa-arm commented on pull request #6950: URL: https://github.com/apache/tvm/pull/6950#issuecomment-760798261 Hi @areusch @zhiics , I've addressed all the minor comments. The current limitation of making this generator a pure C code generator is that is carries a Array\ for

[GitHub] [tvm] masahi closed issue #7281: Frontend PT cannot accept List[Tensor] as input

masahi closed issue #7281: URL: https://github.com/apache/tvm/issues/7281 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