[LLVMdev] LLVM Weekly - #46, Nov 17th 2014

[Alex Bradbury]

LLVM Weekly - #46, Nov 17th 2014
================================

If you prefer, you can read a HTML version of this email at
<http://llvmweekly.org/issue/46>.

Welcome to the forty-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](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

Chrome on Linux [now uses Clang for production
builds](http://article.gmane.org/gmane.comp.compilers.clang.devel/39700).
Clang has of course been used on OS X Chrome for quite some time. The switch
saw reduction in binary size of ~8%, but this was vs GCC 4.6 rather than
something more up-to-date.

The LLVM in HPC workshop at SC14 is taking place on Monday and the full agenda
with abstracts [is available online](http://llvm-hpc-workshop.github.io/)


## On the mailing lists

* Duncan P.N. Exon Smith has posted an [RFC on splitting out metadata from the
Value
hierarchy](http://article.gmane.org/gmane.comp.compilers.llvm.devel/78738).
There seems to be general support for the idea. If you have concerns, now is
the time to speak up.

* Tom Stellard has posted a proposed [LLVM/Clang 3.5.1 release
schedule](http://article.gmane.org/gmane.comp.compilers.llvm.devel/78752). RC1
is currently planned for November 26th.

* Zachary Turner raised the issue [referencing Apple rdar bugs in commit
messages](http://article.gmane.org/gmane.comp.debugging.lldb.devel/5710). The
concern is that sometimes the commit messages are hard to work out without the
context of the bug, which many of us do not have access to.


## LLVM commits

* Work on call lowering for MIPS FastISel has started.
[r221948](http://reviews.llvm.org/rL221948).

* Work has started on an assembler for the R600 backend.
[r221994](http://reviews.llvm.org/rL221994).

* A pass implementing forward control-flow integrity as been added.
[r221708](http://reviews.llvm.org/rL221708).

* A whole slew of patches that made MDNode a Value have been reverted due to a
change in plan. The aim is now to separate metadata from the Value hierarchy.
[r221711](http://reviews.llvm.org/rL221711).

* There are two ways to inform the optimizer the result of a load is never
null. Either with metadata or via assume. The latter is now canonicalized into
the former. [r221737](http://reviews.llvm.org/rL221737).

* `vec_vsx_ld` and `vec_vsx_st` intrinsics have been added for PowerPC.
[r221767](http://reviews.llvm.org/rL221767).

* PowerPC gained support for small-model PIC.
[r221791](http://reviews.llvm.org/rL221791).

* The llvm.arm.space intrinsic was added to make it easier to write tests for
ARM ConstantIslands. [r221903](http://reviews.llvm.org/rL221903).


## Clang commits

* The constant trickle of OpenMP patches continues. Codegen for threadprivate
variables has been added. [r221663](http://reviews.llvm.org/rL221663).

* Support for `__has_cpp_attribute` is now present.
[r221991](http://reviews.llvm.org/rL221991).


## Other project commits

* Breakpoint stop/resume has been implemented on Windows for LLDB.
[r221642](http://reviews.llvm.org/rL221642).

* The [libcxx status page](http://libcxx.llvm.org/cxx1z_status.html) has been
updated with the current state of C++1z support.
[r221601](http://reviews.llvm.org/rL221601)).