[llvm-dev] LLVM Weekly - #149, Nov 7th 2016

Alex Bradbury via llvm-dev llvm-dev at lists.llvm.org
Mon Nov 7 06:49:57 PST 2016

LLVM Weekly - #149, Nov 7th 2016

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

Welcome to the one hundred and forty-ninth 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](http://asbradbury.org). 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

The LLVM Developers' Meeting was a huge success last week. Thank you to
everyone involved for making it such a great event. Talks were recorded,
though I don't know the expected timeline for them being posted online.

The PVS-Studio folks have written up a blog post on [bugs in LLVM's
implementation found using PVS-Studio](http://www.viva64.com/en/b/0446/).

## On the mailing lists

* Initial results from the survey on the move to GitHub have been shared, both
and from

* Graham Hunter from ARM has shared a monster patch and design document for
[supporting ARM's Scalable Vector Extension in
This will be split up in to smaller patches for code review.

* I wrote up an RFC based on the lightning talk I did back in September at the
LLVM Cauldron. [Improving the experience of first-time LLVM
The proposal is essentially that a small group of volunteers (or '[first
work together to improve the experience of first-time contributors by putting
aside time to review the patches from newcomers in particular. From the
discussion in the BoF on 'raising the next generation of LLVM developers', it
sounds like Chandler Carruth has been thinking along similar lines.

* Sanjoy Das has
dropping support for old-style scalar TBAA. If this would pose a problem for
you, now is the time to speak up.

* John Regehr is interested in [reports from anyone who implemented
optimisations derived by

* The thread posted in preparation for the Birds of a Feather session on debug
info evolved to discuss [the challenges of debug info for VLIW

* Zachary Turner's proposed general purpose type-safe formatting library now
has a [prototype
He [revised it
after feedback.

* The thread on changing the LLVM license has seen a flurry of activity this
week, mostly discussing the [proposed exception to Apache's attribution

## LLVM commits

* Documentation for Global ISel has been written and committed.

* The first patches of my RISC-V backend have begun to land.

* The OperandMatchResultTy enum is now part of MCTargetAsmParser.h rather than
emitted through tablegen. Out-of-tree backends will want to make a matching
adjustment. [r285705](http://reviews.llvm.org/rL285705).

* Instruction selection lowering for AVR has been added.

## Clang commits

* The new `-cl-ext` option can be used to override supported OpenCL

* Clang-side changes for the X86 'regcall' calling convention have been
committed. [r285849](http://reviews.llvm.org/rL285849).

* The tests for the libclang python bindings (cindex.py) now pass with Python 3.

* The new readability-redundant-declaration checker will find redundant
variable and function declarations.

## Other project commits

* LLD gained support for the MIPS N32 ABI.

* The LLVM test suite gained a new tidff.py utility, which makes it easy to
compare assembly files of two test-suite builds.

More information about the llvm-dev mailing list