[llvm-dev] LLVM Weekly - #266, February 4th 2019

Alex Bradbury via llvm-dev llvm-dev at lists.llvm.org
Mon Feb 4 12:11:28 PST 2019

LLVM Weekly - #266, February 4th 2019

If you prefer, you can read a HTML version of this email at

Welcome to the two hundred and sixty-sixth 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](https://www.linkedin.com/in/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 at asbradbury.org>, or
@llvmweekly or @asbradbury on Twitter.

## News and articles from around the web

LLVM 8.0.0 RC1 [is

# On the mailing lists

* Tim Northover has written an RFC on [supporting `arm64_32` for
This includes AArch64 ILP32 support and support for the custom ABI used to
allow armv7k IR to be compiled to be compatible with `arm64_32`.

* Tom Stellard started an RFC thread on [linear history vs merge
after migrating to git. This generated copious discussion.

* Simon Moll
[shared](http://lists.llvm.org/pipermail/llvm-dev/2019-January/129791.html) an
RFC on adding native vector predication support to LLVM. It is accompanied by
a prototype implementation.

* Stephen Kelly [has revived the discussion on the use of auto in LLVM

* Vedant Kumar is [curious about the status of the function merging

* David Greene
that a much improved 'zipping' tool has been released, to help people migrate
their downstream repositories to the new combined git repo.

* Davide Italiano started an RFC thread on [the future of pexpect in

## LLVM commits

* The LLVM WebAssembly backend was updated to use the new exception handling
proposal. [r352598](http://reviews.llvm.org/rL352598).

* RISC-V RV64F and RV64D codegen (single and double-precision floating point)
support was committed. [r352807](http://reviews.llvm.org/rL352807).

* New variadic avx512 compress/expand intrinsics were added.

* The SelectionDAG will now work to avoid re-ordering `DBG_VALUEs`.

* A new 'apple-latest' CPU alias was added.

* The in-tree gn build system gained a get.py script to download an use a
prebuilt gn. [r352420](http://reviews.llvm.org/rL352420).

* The .align directive now correctly inserts NOPs and the `R_RISCV_ALIGN`
relocation when appropriate for RISC-V.

* A 'dynamic' parameter was added to the objectsize intrinsic.

* A new FunctionCallee wrapper type was added.

## Clang commits

* A new `-fapply-gobal-visibility-to-externs` option was added.

* clangd implemented textDocument/declaration from the Language Server
Protocol 3.14. [r352864](http://reviews.llvm.org/rL352864).

* clangd gained new interfaces for writing "code tweaks".

## Other project commits

* The CoMD Classical Molecular Dynamics Mini-app was added to the LLVM test
suite. This is another DOE proxy app.

* libFuzzer gained a new experimental performance optimisation,
`lazy_counters` which is off by default.

* The LLD COFF linker supports range extension thunks for ARM64.

More information about the llvm-dev mailing list