# TVM Monthly - May 2023

**URL:** <https://discuss.tvm.apache.org/t/tvm-monthly-may-2023/15054>\
**Category:** Announcements\
**Created:** [June 3, 2023, 5:19am UTC](https://discuss.tvm.apache.org/t/tvm-monthly-may-2023/15054 "2023-06-03T05:19:48Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![ysh329](https://discuss.tvm.apache.org/user_avatar/discuss.tvm.apache.org/ysh329/32/632_2.png) [@ysh329](https://discuss.tvm.apache.org/u/ysh329)\
**Post date:** [June 3, 2023, 5:19am UTC](https://discuss.tvm.apache.org/t/tvm-monthly-may-2023/15054/1 "2023-06-03T05:19:48Z")

</div>

# TVM Monthly - May 2023

**Note: This montly report contains main branch only.**

As discussed by the TVM PMC, our goal is to provide a monthly summary of the project so users and developers can get a better understanding of the goings on of the TVM community.

Feedback and suggestions are welcomed so that we can further improve these updates.

## Community

During May of 2023 we welcomed many new contributors to the project. Importantly we welcomed [Anirudh Sundar Subramaniam](https://github.com/quic-sanirudh) as new committers. Thanks to everyone for the hard work and contributions!

We continue to improve TensorIR, MetaSchedule, Frontend and other runtimes.

## Pull Requests

The below is high-level summary of the PRs closed in the last month grouped by area.

### Community

- [#14853](https://github.com/apache/incubator-tvm/pull/14853) - Anirudh Sundar Subramaniam → Committer
- [#14772](https://github.com/apache/incubator-tvm/pull/14772) - Add new key for release signing

### Arith

- [#14532](https://github.com/apache/incubator-tvm/pull/14532) - Implement statistics counters for RewriteSimplifier

### CI

- [#14908](https://github.com/apache/incubator-tvm/pull/14908) - Update the expected CI jobs list in the update\_branch script
- [#14847](https://github.com/apache/incubator-tvm/pull/14847) - Update CPU image to install PyTorch
- [#14808](https://github.com/apache/incubator-tvm/pull/14808) - [Testing] Use TVMScript’s “name” argument for error messages
- [#14780](https://github.com/apache/incubator-tvm/pull/14780) - fix doc deploy issue
- [#14651](https://github.com/apache/incubator-tvm/pull/14651) - Modify test cases to accommodate the CI upgrades
- [#14666](https://github.com/apache/incubator-tvm/pull/14666) - sccache support while using ci.py under multi user environments
- [#14635](https://github.com/apache/incubator-tvm/pull/14635) - Upgrade CI

### Docker

- [#14912](https://github.com/apache/incubator-tvm/pull/14912) - Add polly package
- [#14842](https://github.com/apache/incubator-tvm/pull/14842) - Install PyTorch on cpu image

### Frontend

- [#14963](https://github.com/apache/incubator-tvm/pull/14963) - [Bugfix][PyTorch] Support use\_input\_stats in instance\_norm
- [#14826](https://github.com/apache/incubator-tvm/pull/14826) - [Paddle] [PaddlePaddle Hackathon 4] add convert support for p\_norm/roi\_align/softmax\_with\_cross\_entropy
- [#14819](https://github.com/apache/incubator-tvm/pull/14819) - [TFLite]Generate name when tensor name is missing
- [#14830](https://github.com/apache/incubator-tvm/pull/14830) - Use f-strings for string formatting, NFC
- [#14747](https://github.com/apache/incubator-tvm/pull/14747) - [PyTorch] Add aten::new\_zeros

### Hexagon

- [#14948](https://github.com/apache/incubator-tvm/pull/14948) - Update instructions to compile hexagon runtime
- [#14965](https://github.com/apache/incubator-tvm/pull/14965) - Add support for v73, make v68 default

### LLVM

- [#14958](https://github.com/apache/incubator-tvm/pull/14958) - [Codegen]Allow void return type from PackedFunc
- [#14946](https://github.com/apache/incubator-tvm/pull/14946) - Expose Host CPU Feature Detection
- [#14901](https://github.com/apache/incubator-tvm/pull/14901) - Codegen subroutine call when CallNode::op is GlobalVar
- [#14570](https://github.com/apache/incubator-tvm/pull/14570) - Use Var annotation in LetStmt for pointer type
- [#14843](https://github.com/apache/incubator-tvm/pull/14843) - [RUNTIME] Enable multi systemlib with device code

### MetaSchedule

- [#14784](https://github.com/apache/incubator-tvm/pull/14784) - Enhance `tune_tir` to tune IRModule of TIR Collections
- [#14783](https://github.com/apache/incubator-tvm/pull/14783) - Add an API to dump a pruned database
- [#14785](https://github.com/apache/incubator-tvm/pull/14785) - Clear screen only when specified
- [#14781](https://github.com/apache/incubator-tvm/pull/14781) - [Bugfix]RPC port needs to be an integer

### OpenCL & CLML

- [#14995](https://github.com/apache/incubator-tvm/pull/14995) - [OpenCL] Improve diagnostic message
- [#14833](https://github.com/apache/incubator-tvm/pull/14833) - [Codegen][OpenCL] fix amibiguous selection operator call
- [#14792](https://github.com/apache/incubator-tvm/pull/14792) - [OpenCL] Refactor OpenCL runtime to support SPIRV binary ingestion

### Relay

- [#14925](https://github.com/apache/incubator-tvm/pull/14925) - add a dimension check to reject invalid input
- [#14858](https://github.com/apache/incubator-tvm/pull/14858) - [simplify\_expr]: Add pass to remove trivial transpose ops
- [#14838](https://github.com/apache/incubator-tvm/pull/14838) - Use f-strings for string formatting, NFC
- [#14831](https://github.com/apache/incubator-tvm/pull/14831) - [Relay/Op] Use f-strings for string formatting, NFC
- [#14580](https://github.com/apache/incubator-tvm/pull/14580) - Simplify the square of a binomial

### Runtime

- [#14967](https://github.com/apache/incubator-tvm/pull/14967) - [Runtime,RPC] Use f-strings for string formatting, NFC
- [#14887](https://github.com/apache/incubator-tvm/pull/14887) - Make systemlib unique per prefix
- [#14775](https://github.com/apache/incubator-tvm/pull/14775) - Added **str** for tvm.\_ffi.runtime\_ctypes.TVMArray

### TIR

- [#14993](https://github.com/apache/incubator-tvm/pull/14993) - Fix incorrect construction of block frames
- [#14982](https://github.com/apache/incubator-tvm/pull/14982) - [Bugfix][VTA] Update host-side target, even without device func
- [#14952](https://github.com/apache/incubator-tvm/pull/14952) - Avoid re-defining `var = arg_var` in ArgBinder
- [#14918](https://github.com/apache/incubator-tvm/pull/14918) - SplitHostDevice, handle subroutines
- [#14943](https://github.com/apache/incubator-tvm/pull/14943) - Restrict tir.transform.InstallDebugSpans to host functions
- [#14942](https://github.com/apache/incubator-tvm/pull/14942) - Preserve existing kTarget function attribute in BindTarget
- [#14945](https://github.com/apache/incubator-tvm/pull/14945) - Restrict tir.transform.CombineContextCall to host functions
- [#14914](https://github.com/apache/incubator-tvm/pull/14914) - Handle subroutine calls in MakeUnpackedAPI
- [#14913](https://github.com/apache/incubator-tvm/pull/14913) - Handle subroutine calls in MakePackedAPI
- [#14892](https://github.com/apache/incubator-tvm/pull/14892) - Expand unit tests for ConvertSSA
- [#14866](https://github.com/apache/incubator-tvm/pull/14866) - Avoid too complex predicate in compaction
- [#14766](https://github.com/apache/incubator-tvm/pull/14766) - [Schedule] Improve blockize to support blockizing multiple blocks
- [#14776](https://github.com/apache/incubator-tvm/pull/14776) - Improved parameter name in DLTensor unpacking error messages
- [#14562](https://github.com/apache/incubator-tvm/pull/14562) - [Driver] Move ShouldAnnotateEntryFunc logic into transform
- [#14741](https://github.com/apache/incubator-tvm/pull/14741) - Keep block annotations from tensorization

### TOPI

- [#14856](https://github.com/apache/incubator-tvm/pull/14856) - Fix in interpretation of empty axis parameter in reduction fun…
- [#14483](https://github.com/apache/incubator-tvm/pull/14483) - [Target] Add SVE specific convolution
- [#14839](https://github.com/apache/incubator-tvm/pull/14839) - Use f-strings for string formatting, NFC
- [#14822](https://github.com/apache/incubator-tvm/pull/14822) - Use f-strings for string formatting, NFC
- [#14519](https://github.com/apache/incubator-tvm/pull/14519) - Vectorize depthwise conv2d output operator

### TVMScript

- [#14919](https://github.com/apache/incubator-tvm/pull/14919) - [TIR] Parse subroutine calls with no arguments
- [#14941](https://github.com/apache/incubator-tvm/pull/14941) - Prevent bool to int conversion in T.Assert condition
- [#14915](https://github.com/apache/incubator-tvm/pull/14915) - Allow T.target(“device”, host=“host”) to specify host
- [#14900](https://github.com/apache/incubator-tvm/pull/14900) - Round-trip DeclBuffer with undefined data pointer
- [#14889](https://github.com/apache/incubator-tvm/pull/14889) - [TIR]Added format/parsing of subroutine calls
- [#14874](https://github.com/apache/incubator-tvm/pull/14874) - Use default fallback for un-registered type
- [#14849](https://github.com/apache/incubator-tvm/pull/14849) - [Bugfix]Capture fails if var appears only in annotation
- [#14840](https://github.com/apache/incubator-tvm/pull/14840) - Print Executor, Runtime, and FunctionInfo as metadata
- [#14812](https://github.com/apache/incubator-tvm/pull/14812) - Handle AllocatedPoolInfo, ConstantPoolInfo, ConstantInfo
- [#14786](https://github.com/apache/incubator-tvm/pull/14786) - Add ` __name__ ` attr for parsed PrimFunc and IRModule
- [#14531](https://github.com/apache/incubator-tvm/pull/14531) - Preserve LetStmt of constants

### cuda & cutlass & tensorrt

- [#14751](https://github.com/apache/incubator-tvm/pull/14751) - [CUDA] Fixed the call of the min function in the schedule for cuda
- [#14798](https://github.com/apache/incubator-tvm/pull/14798) - [CUTLASS] Add NDEBUG option to CUTLASS compile to speed up attention kernel

### metal

- [#14962](https://github.com/apache/incubator-tvm/pull/14962) - Fix int8 vectorized cast
- [#14846](https://github.com/apache/incubator-tvm/pull/14846) - Fix vectorized select

### micoNPU

- [#14861](https://github.com/apache/incubator-tvm/pull/14861) - [microNPU][ETHOSU] Add offloading to the NPU the nn.avg\_pool2d operator with a stride \> 3
- [#14765](https://github.com/apache/incubator-tvm/pull/14765) - [microNPU][ETHOSU] Channel pad offloaded to NPU
- [#14774](https://github.com/apache/incubator-tvm/pull/14774) - [microNPU][ETHOSU] Fix Softmax quantization parameters

### microTVM

- [#14872](https://github.com/apache/incubator-tvm/pull/14872) - Use self.close\_transport() on error

### Misc

- [#14975](https://github.com/apache/incubator-tvm/pull/14975) - [CMAKE] Add a dummy target to defer libtvm dep
- [#14574](https://github.com/apache/incubator-tvm/pull/14574) - [IR][SIBuilder]
- [#14930](https://github.com/apache/incubator-tvm/pull/14930) - Fix pytorch axis
- [#14939](https://github.com/apache/incubator-tvm/pull/14939) - [Target] Add target to all TVM callbacks
- [#14937](https://github.com/apache/incubator-tvm/pull/14937) - [BUILD] Enable log before throw message in windows
- [#14934](https://github.com/apache/incubator-tvm/pull/14934) - [TestCases] fix unreachable test cases due to outside the for-loop
- [#14916](https://github.com/apache/incubator-tvm/pull/14916) - [TypoFix] fix some typo problem in keras frontend
- [#14893](https://github.com/apache/incubator-tvm/pull/14893) - [Contrib] Use f-strings for string formatting, NFC
- [#14817](https://github.com/apache/incubator-tvm/pull/14817) - [Vulkan] Add cooperative matrix support
- [#14884](https://github.com/apache/incubator-tvm/pull/14884) - [AutoTVM] Use f-strings for string formatting, NFC
- [#14876](https://github.com/apache/incubator-tvm/pull/14876) - [CONTRIB] Enable create\_staticlib to take in tar files
- [#14875](https://github.com/apache/incubator-tvm/pull/14875) - [Quantization]: Update simulated\_quantize to infer correct layout
- [#14867](https://github.com/apache/incubator-tvm/pull/14867) - Fix f-string typo
- [#14536](https://github.com/apache/incubator-tvm/pull/14536) - [QNN] Implement ‘qnn.softmax’
- [#14851](https://github.com/apache/incubator-tvm/pull/14851) - Add v0.12.0 docs
- [#14813](https://github.com/apache/incubator-tvm/pull/14813) - [BUILD] Removed the duplicated MACROs in config.cmake
- [#14743](https://github.com/apache/incubator-tvm/pull/14743) - [SUPPORT] Fix RingBuffer ReadWithCallback
- [#14767](https://github.com/apache/incubator-tvm/pull/14767) - [OpenCLML] Transposed convolution support and other fixes
- [#14799](https://github.com/apache/incubator-tvm/pull/14799) - [LINT] Fix clang-format script for newest clang-format
- [#14797](https://github.com/apache/incubator-tvm/pull/14797) - [NDArray] Allow arbitrary stride when the corresponding shape is 1
- [#14790](https://github.com/apache/incubator-tvm/pull/14790) - More clear ref of thirdparty license
- [#14779](https://github.com/apache/incubator-tvm/pull/14779) - fix: use arm on demand instead of spot
- [#14762](https://github.com/apache/incubator-tvm/pull/14762) - [Target][Minor] Add A6000 Target Tag

### BugFix

- [#14991](https://github.com/apache/incubator-tvm/pull/14991) - Make ThreadAllReduce pass compatible with int64
- [#14950](https://github.com/apache/incubator-tvm/pull/14950) - Avoid symbol conflicts in MakePackedAPI/MakeUnpackedAPI
- [#14903](https://github.com/apache/incubator-tvm/pull/14903) - [Test Cases]Add some version check to make test cases run in all PyTorch versions
- [#14885](https://github.com/apache/incubator-tvm/pull/14885) - [Relay] fix broadcast in PyTorch frontend
- [#14890](https://github.com/apache/incubator-tvm/pull/14890) - [Fix] Fix typo in error message
- [#14879](https://github.com/apache/incubator-tvm/pull/14879) - fix the undeclared identifier ‘f’
- [#14857](https://github.com/apache/incubator-tvm/pull/14857) - Fix batch\_norm
- [#14787](https://github.com/apache/incubator-tvm/pull/14787) - [FIX] fix typo in comment
- [#14845](https://github.com/apache/incubator-tvm/pull/14845) - [Relay] Fix softplus in paddlepaddle frontend
- [#14837](https://github.com/apache/incubator-tvm/pull/14837) - [Relay] Fix AdaptiveAvgPool2d about wrong dtype prasing
- [#14821](https://github.com/apache/incubator-tvm/pull/14821) - [Relay] Fix softplus about the wrong calculation formula in Relay PyTorch frontend
- [#14820](https://github.com/apache/incubator-tvm/pull/14820) - [Relay] Fix threshold calculation logic in PyTorch frontend
- [#14824](https://github.com/apache/incubator-tvm/pull/14824) - [Relay]fix a bug about ReLu in the threshold attribute which causes a different results with keras
- [#14796](https://github.com/apache/incubator-tvm/pull/14796) - [relay] fix wrong calculate logic about celu
- [#14773](https://github.com/apache/incubator-tvm/pull/14773) - [Relay] fix `scatter_nd` type relation
- [#14782](https://github.com/apache/incubator-tvm/pull/14782) - [Codegen][CUDA] Wrong casting in ASM
- [#14742](https://github.com/apache/incubator-tvm/pull/14742) - [relay] Fix alpha attribute with None in ELU

## Contributors Whose Pull Requests were Updated

Note: The format is name (number of activities).

Disclaimer: number of activities do not directly correspond to the community’s view about the significance of contributions. Data is from [2023-05-01&to=2023-05-31](https://github.com/apache/tvm/graphs/contributors?from=2023-05-01&to=2023-05-31&type=c).

Lunderberg(25), jikechao(12), kparzysz-quic(11), tqchen(11), masahi(5), junrushao(4), Hzfengsy(3), vinx13(2), yzh119(2), srkreddy1238(2), echuraev(2), wrongtest-intellif(1), ashutosh-arm(1), quic-sanirudh(1), spectrometerHBH(1), gigiblender(1).
