[GitHub] [tvm] wyc-ruiker opened a new pull request #8402: [CUDA] add int8/int4 tensorcore for dense/batch_matmul

wyc-ruiker opened a new pull request #8402: URL: https://github.com/apache/tvm/pull/8402 Let dense_tensorcore and batch_matmul_tensorcore support int8/int4. Before this pr, the [vision transform (vit)](https://arxiv.org/abs/2010.11929) latency (https://github.com/apache/tvm/pull/7814) i

[GitHub] [tvm] huajsj commented on a change in pull request #7892: [Runtime]Pipeline Executor For Compute graph pipeline

huajsj commented on a change in pull request #7892: URL: https://github.com/apache/tvm/pull/7892#discussion_r663615114 ## File path: src/runtime/pipeline/pipeline_executor.h ## @@ -0,0 +1,183 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more cont

[GitHub] [tvm] wrongtest commented on a change in pull request #8338: [TIR] fix storage rewrite index remap

wrongtest commented on a change in pull request #8338: URL: https://github.com/apache/tvm/pull/8338#discussion_r663608412 ## File path: src/tir/transforms/storage_rewrite.cc ## @@ -504,8 +503,13 @@ class StoragePlanRewriter : public StmtExprMutator { return MergeNest(nest,

[GitHub] [tvm] jcf94 edited a comment on issue #8392: [BUG][FRONTEND] Pytorch Importer mm

jcf94 edited a comment on issue #8392: URL: https://github.com/apache/tvm/issues/8392#issuecomment-873726740 Sorry for my late reply, what's the latest update to this bug? The `nn.matmul` should currently only be enabled in TensorFlow frontend with extra parameters. Ok, seems it h

[GitHub] [tvm] jcf94 edited a comment on issue #8392: [BUG][FRONTEND] Pytorch Importer mm

jcf94 edited a comment on issue #8392: URL: https://github.com/apache/tvm/issues/8392#issuecomment-873726740 -- 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,

[GitHub] [tvm] jcf94 commented on issue #8392: [BUG][FRONTEND] Pytorch Importer mm

jcf94 commented on issue #8392: URL: https://github.com/apache/tvm/issues/8392#issuecomment-873726740 Sorry for my later replay, what's the latest update to this problem? The `nn.matmul` should currently only be enabled in TensorFlow frontend with extra parameters. -- This is an autom

[GitHub] [tvm] chiwwang commented on pull request #8220: [DOCS] Add docs for Pass Instrument

chiwwang commented on pull request #8220: URL: https://github.com/apache/tvm/pull/8220#issuecomment-873715155 Thanks @zackcquic @tkonolige !! @areusch you might want to check if this PR is ready for merging. Thanks :) -- This is an automated message from the Apache Git Service. To resp

[GitHub] [tvm] elvin-n opened a new pull request #8401: Add qnn batch_matmul operator

elvin-n opened a new pull request #8401: URL: https://github.com/apache/tvm/pull/8401 + added support of the different out type for x86 batch_matmul -- 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-rfcs] areusch commented on a change in pull request #8: Add Project API RFC

areusch commented on a change in pull request #8: URL: https://github.com/apache/tvm-rfcs/pull/8#discussion_r663520135 ## File path: rfcs/0008-microtvm-project-api.md ## @@ -0,0 +1,509 @@ +- Feature Name: microtvm_project_api +- Start Date: 2020-06-09 +- RFC PR: [apache/tvm-rfc

[GitHub] [tvm] u99127 commented on pull request #8394: Add Compute Library tests to Jenkins for AArch64 CI

u99127 commented on pull request #8394: URL: https://github.com/apache/tvm/pull/8394#issuecomment-873599307 > > @u99127 @leandron @tqchen ARM compute lib tests are failing in CI: > > https://ci.tlcpack.ai/blue/organizations/jenkins/tvm/detail/main/1255/pipeline > > https://ci.tlcpack

[GitHub] [tvm] u99127 opened a new pull request #8400: Revert "Actually add Compute Library tests to the Jenkins File (#8394)"

u99127 opened a new pull request #8400: URL: https://github.com/apache/tvm/pull/8400 This reverts commit a00d21156ea749d65c9a59b3a79dac501723086b till we can understand why this is failing in CI but not on developer machines. @tqchen @leandron @masahi Sorry about the breaka

[GitHub] [tvm] tqchen commented on pull request #8384: [Arith] Inverse affine map

tqchen commented on pull request #8384: URL: https://github.com/apache/tvm/pull/8384#issuecomment-873581831 @vinx13 let us followup with another PR to address the last comment @spectrometerHBH bought up -- This is an automated message from the Apache Git Service. To respond to the messag

[GitHub] [tvm] u99127 opened a new pull request #8399: Cleanup more uses of np.bool and np.int.

u99127 opened a new pull request #8399: URL: https://github.com/apache/tvm/pull/8399 In a similar vein to previous pull requests replacing deprecated use of np.bool and np.int from numpy with bool and int. -- This is an automated message from the Apache Git Service. To respond to the

[GitHub] [tvm] u99127 edited a comment on pull request #8394: Add Compute Library tests to Jenkins for AArch64 CI

u99127 edited a comment on pull request #8394: URL: https://github.com/apache/tvm/pull/8394#issuecomment-873556799 > @u99127 @leandron @tqchen ARM compute lib tests are failing in CI: > https://ci.tlcpack.ai/blue/organizations/jenkins/tvm/detail/main/1255/pipeline > https://ci.tlcpac

[GitHub] [tvm] u99127 edited a comment on pull request #8394: Add Compute Library tests to Jenkins for AArch64 CI

u99127 edited a comment on pull request #8394: URL: https://github.com/apache/tvm/pull/8394#issuecomment-873556799 > @u99127 @leandron @tqchen ARM compute lib tests are failing in CI: > https://ci.tlcpack.ai/blue/organizations/jenkins/tvm/detail/main/1255/pipeline > https://ci.tlcpac

[GitHub] [tvm] u99127 commented on pull request #8394: Add Compute Library tests to Jenkins for AArch64 CI

u99127 commented on pull request #8394: URL: https://github.com/apache/tvm/pull/8394#issuecomment-873556799 > @u99127 @leandron @tqchen ARM compute lib tests are failing in CI: > https://ci.tlcpack.ai/blue/organizations/jenkins/tvm/detail/main/1255/pipeline > https://ci.tlcpack.ai/bl