[LLVMdev] LLVM Weekly - #72, May 18th 2015

[Alex Bradbury]

LLVM Weekly - #72, May 18th 2015
================================

If you prefer, you can read a HTML version of this email at
<http://llvmweekly.org/issue/72>.

Welcome to the seventy-second 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.

Some of you may be interested that over at the lowRISC project, we've
announced the [full set of summer student
projects](http://www.lowrisc.org/blog/2015/05/summer-of-code-students-for-lowrisc/)
we're supporting.


## News and articles from around the web

The Rust programming language, which of course uses LLVM as its compiler
backend, has just [released version
1.0](http://blog.rust-lang.org/2015/05/15/Rust-1.0.html).

The next Cambridge LLVM Social will [take place on Wednesday 20th
May](http://article.gmane.org/gmane.comp.compilers.clang.devel/42786) at the
Cambridge Beer Festival.


## On the mailing lists

* Teresa Johnson has posted an [RFC on her ThinLTO implementation
plan](http://article.gmane.org/gmane.comp.compilers.llvm.devel/85628). It's
seen a lot of feedback, too much for me to hope to summarise, though much of
it was around emitting bitcode wrapped in ELF.

* Yaxun Liu is [looking to upstream the Khronos Group LLM to SPIR-V
converter](http://article.gmane.org/gmane.comp.compilers.llvm.devel/85609).
There was some discussion as to whether SPIR-V makes most sense as a backend
or as serialisation format alongside the LLVM bitcode output.

* Reid Kleckner has written up an RFC on a [new exception handling
representation for
MSVC](http://article.gmane.org/gmane.comp.compilers.llvm.devel/85783), given
that the Itanium EH representation has been found insufficient.

* Chris Matthews proposes [changing LNT's regression detection
algorithm](http://article.gmane.org/gmane.comp.compilers.llvm.devel/85775),
with the aim being to reduce the number of false positives to the extent that
people will be motivated to investigate reported regressions.

* Matt Arsenault is proposing [upstreaming an LLVM backend for
HSAIL](http://article.gmane.org/gmane.comp.compilers.llvm.devel/85616). HSAIL
presents a virtual target machine (in a way similar to NVPTX), and is defined
by the HSA Foundation.

* Owen Anderson proposes a new [convergent
attribute](http://article.gmane.org/gmane.comp.compilers.llvm.devel/85633)
aiming to make LLVM more suitable for SPMD/SIMT programming models. So far,
all feedback has been very positive.

## LLVM commits

* The ARM backend has been updated to use AEABI aligned function variants.
[r237127](http://reviews.llvm.org/rL237127).

* The heuristic for estimating the effect of complete loop unrolling has been
reimplemented. [r237156](http://reviews.llvm.org/rL237156).

* Statepoints are now 'patchable'.
[r237214](http://reviews.llvm.org/rL237214).

* Support for function entry count metadata has been added.
[r237260](http://reviews.llvm.org/rL237260).

* A new loop distribution pass has been born. It is off by default.
[r237358](http://reviews.llvm.org/rL237358).

* New SelectionDAG nodes have been added for signed/unsigned min/max.
[r237423](http://reviews.llvm.org/rL237423).

* A simple speculative execution pass, targeted mainly at GPUs has been added.
[r237459](http://reviews.llvm.org/rL237459).


## Clang commits

* The little endian SPARC target has been added to clang.
[r237001](http://reviews.llvm.org/rL237001).

* clang-format's formatter has undergone some refactoring, which also led to a
few bug fixes.[r237104](http://reviews.llvm.org/rL237104).

* Documentation on adding new attributes has seen a significant update.
[r237268](http://reviews.llvm.org/rL237268).


## Other project commits

* libcxx learnt `std::experimental::sample`
[r237264](http://reviews.llvm.org/rL237264).

* lldb has enabled multithreaded debugging on Windows.
[r237392](http://reviews.llvm.org/rL237392).

* lldb can now set and clear hardware watchpoints and breakpoints on AArch64.
[r237419](http://reviews.llvm.org/rL237419).

* lldb gained an assembly profiler for mips32.
[r237420](http://reviews.llvm.org/rL237420).