[llvm-dev] LLVM GPU News #18, August 20 2021
Jakub (Kuba) Kuderski via llvm-dev
llvm-dev at lists.llvm.org
Sun Aug 22 16:11:01 PDT 2021
The 18th issue of LLVM GPU News, a bi-weekly newsletter on all the GPU
things under the LLVM umbrella, is out:
I also pasted the content below, in case you prefer to read in your email
# LLVM GPU News #18, Aug 20 2021
Authors: Jakub Kuderski, Lei Zhang
Welcome to LLVM GPU News, a bi-weekly newsletter on all the GPU things
under the LLVM umbrella.
This issue covers the period from August 6 to August 19 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 Conferences
## LLVM and Clang
* Henry Linjamäki posted an RFC on ['Emitting HIP device code as SPIR-V'](
initial focus is to use SPIR-V binaries for OpenCL-like environments:
HIPCL, and HIPLZ (Intel Level Zero API). Others point out that this
proposal has a large overlap with what changes would be necessary for
OpenCL and OpenMP offloading and that it would be preferred to keep non-HIP
parts generic. Initially, the project may use the SPIRV-LLVM translator as
an external tool until the SPIR-V backend (SPIR-V BE) is available.
[Anastasia Stulova suggests waiting for the SPIR-V BE](
https://lists.llvm.org/pipermail/llvm-dev/2021-August/152228.html), to help
with its adoption and avoid having to redesign parts of the codebase.
* Frank Winter is working on a JIT compiler targeting AMDGPU GCN and asked
for [help with using LLD to link files passed through memory instead of
[Fangrui Song responded](
https://lists.llvm.org/pipermail/llvm-dev/2021-August/152139.html) that LLD
is generally not ready to be used as a library and gave a few suggestions.
* Dustin Favorite suspects that [`__nvvm_redux_sync_*` CUDA intrinsics are
not working as expected](
result in 'Illegal instruction' errors. There are no replies at the time of
* New Clang language defines for C++ for OpenCL version 2021. The language
standard is `lang_openclcpp2021`, enabled with the `-cl-std=clc++2021` flag.
* Rematerialization of virtual register uses is not allowed. The change
affected AMDGPU, ARM/Thumb, MIPS, RISC-V, and x86. [D106408](
- Machine LICM (Loop Invariant Code Motion) will not rematterialzie
instructions with virtual register uses. Such an instruction will not be
hoisted out of the loop in a belief that the register allocator will sink
it back if needed. This impacts AMDGPU. [D107677](
* `alias.scope` metadata is added to lowered AMDGPU LDS struct to improve
alias analysis. [D108315](https://reviews.llvm.org/D108315)
* `GlobalISel` gained a combine for `PTR_ADD` with register banks.
* There are questions about [marking](
`gpu::LaunchFuncOp` async and how to [lower and run](
`linalg.matmul` using CUDA.
* `spv.(InBounds)PtrAccessChain` ops are [defined](
## OpenMP (Target Offloading)
* A bunch of refactoring patches by Jon Chesterfield:
- Lane masks are now represented with `uint64_t` instead of
`__kmpc_impl_lanemask_t` in `deviceRTLs`. [D108317](
- Replace `OMPGridValues` array with struct. [D108339](
- Refactor `GridValues`: remove redundant fields and replace pointer
with virtual function. [D108380](https://reviews.llvm.org/D108380)
## External Compilers
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the llvm-dev