[llvm-dev] LLVM GPU News Issue #5, February 5 2021
Jakub (Kuba) Kuderski via llvm-dev
llvm-dev at lists.llvm.org
Fri Feb 5 13:52:44 PST 2021
The fifth issue of LLVM GPU News, a bi-weekly newsletter on all the GPU things
under the LLVM umbrella, is now available at:
I'm also pasting the content below, in case you prefer to read in your
# LLVM GPU News Issue #5, February 5 2021
Authors: Jakub Kuderski, Lei Zhang, Johannes Doerfert
Welcome to the fifth issue of LLVM GPU News, a bi-weekly newsletter on all
the GPU things under the LLVM umbrella.
This issue covers the period from January 22 to February 4 2021.
We welcome your feedback and suggestions. Let us know if we missed anything
interesting, or want us to bring attention to your (sub)project, revisions
under review, or proposals. Please see the bottom of the page for details
on how to submit suggestions and contribute.
## Industry News and Conference Talks
* Alyssa Rosenzweig published a [second blog post in the series dedicated
to investigating the Apple M1 GPU](
https://rosenzweig.io/blog/asahi-gpu-part-2.html). The project reached a
new milestone: being able to draw a triangle with some hand-written machine
code. The code is [available on GitHub](https://github.com/AsahiLinux/gpu).
## LLVM and Clang
* Sameer Sahasrabuddhe is looking into [enabling divergence analysis in
the new pass manager and its interaction with loop unswitching passes](
* Lowering of memory intrinsics (`memcpy`, `memmove`, `memset`) [will be
moved from `Combiner` to `Legalizer` in `GlobalISel`](
is motivated by the needs of the AMDGPU backend.
* A new intrinsic [`llvm.set.rounding` for setting floating point rounding
mode](https://reviews.llvm.org/D74729) has finally landed. AMDGPU targets
support changing the rounding mode at runtime and can make use of it.
* [Fixes around CUDA/HIP static variables in host, device, and global
* (In-review) [Initial plumbing for CUDA sm_86 GPU targets.](
* Weiwei [expressed interests](
in contributing to SPIR-V dialect for graphics usage and started sending
out patches for it.
* [Integration tests](https://reviews.llvm.org/D94421) using CUDA runner
are added to demonstrate async lowering.
* Serialization and deserialization for [image types](
https://reviews.llvm.org/D95580) are now supported in the SPIR-V dialect.
* A few [new patterns](https://reviews.llvm.org/D96042) were added to
lower vector ops to their SPIR-V counterparts.
## OpenMP (Target Offloading)
* Initial [AMDGPU offloading toolchain for OpenMP offloading](
https://reviews.llvm.org/D94961) has landed.
## External Compilers
* The standalone SPIR-V to GCN compiler `amdllpc` can now [compile
individual shaders](https://github.com/GPUOpen-Drivers/llpc/pull/1121) into
[relocatable ELF files](
These ELFs can be then [bundled as Vulkan Pipeline Cache files](
https://github.com/GPUOpen-Drivers/xgl/pull/94/files), all offline, i.e.,
on a system without a GPU. The produced caches cannot be loaded by the
AMDVLK driver yet.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the llvm-dev