LLVM Weekly - #511, October 16th 2023

Welcome to the five hundred and eleventh issue of LLVM Weekly, a weekly newsletter (published every Monday) covering developments in LLVM, Clang, and related projects. LLVM Weekly is brought to you by Alex Bradbury. Subscribe to future issues at http://llvmweekly.org and pass it on to anyone else you think may be interested. Please send any tips or feedback to asb@asbradbury.org, @llvmweekly or @asbradbury on Twitter, or @llvmweekly@fosstodon.org or @asb@fosstodon.org.

News and articles from around the web and events

I wrote up a summary of some of the main RISC-V related developments for LLVM 17.

Vitor M. de Sousa Pereira blogged about trying out C++20’s modules with Clang and CMake.

Fangrui Song wrote about global variable instrumentation in AddressSanitizer.

According to the LLVM calendar in the coming week there will be:

• Office hours with the following hosts: Phoebe Wang, Johannes Doerfert, Aaron Ballman.
• Online sync-ups on the following topics: Flang, SYCL, LLVM security group public meeting, classic Flang, C/C++ language working group, loop optimisations, OpenMP in Flang, MLIR, PowerPC, SPIR-V.
• For more details see the LLVM calendar, getting involved documentation on online sync ups and office hours.

On the forums

• Intel have posted a collection of SYCL related RFCs, all handily linked to from this overview RFC on adding full support for the SYCL programming model.

• Chris Bieneman posted an RFC proposal for project governance. This was co-authored with Mehdi Amini, Aaron Ballman, Eric Christopher, and Reid Kleckner, and presented at the LLVM Dev Meeting last week.

• Various notes from discussions at the LLVM Dev Meeting were shared: the LLVM libc roundtable, Community.o workshop discussions, and ML-guided compiler optimization in LLVM workshop.

• MLIR News #57 is out, and as always provides a great rundown of recent MLIR developments.

• Some of the slides were shared from the Compiler Optimizations for Warehouse Scale Computing Workshop at the LLVM Dev Meeting.

• Jonas Devlieghere proposed renaming lldb-vscode to lldb-dap. So far, all respondents are in favour.

• Mingming Liu shared a proposal for improving profile-guided call graph sort.

LLVM commits

• Code to print embedded bitcode sections in MachO files was removed, as it’s no longer possible to submit bitcode apps to the Apple App Store and the relevant tools have been discontinued. eb60143.

• LLVM support was added for HIP parallel aogorithm offload on AMDGPU targets. This involves adding HipstdParAcceleratorCodeSelectionPass and HipStdParAllocationInterpositionPass. 0ce6255.

• Various users of llvm:support::endianness were updated to use llvm::endianness. a9d5056.

• X86’s datalayout now explicitly aligns i128 to 16 bytes. a21abc7.

• The newly added TableGen !repr represents the content of a variable or of a record as string. cacfac4.

• llvm-remarkutil gained a new count subcommand. 31c2cf1.

Clang commits

• A C++ array delete checker was added for the Clang Static Analyzer. 0e246bb.

• Clang’s getting involved page was updated. 2ef1587.

Other project commits

• TSan gained support for 64-bit RISC-V. 46cb8d9.

• An FIR type-based alias analysis (TBAA) pass was added to Flang. df5c278.

• A new stop-at-user-entry-point option was added for LLDB process launch, allowing execution to be paused at the entry point (e.g. main for C-based languages). ac0dda8.

• The first commit adding a ‘Mesh’ dialect to MLIR landed. 08545e8.

• A new mlir-query tool was introduced, with autocomplete support. 02d9f4d.