-
| | |
|--------------------|----|
| Bugzilla Link | [PR19393](https://bugs.llvm.org/show_bug.cgi?id=19393) |
| Status | NEW |
| Importance | P normal |
|…
-
| | |
|--------------------|----|
| Bugzilla Link | [PR24108](https://bugs.llvm.org/show_bug.cgi?id=24108) |
| Status | NEW |
| Importance | P normal |
|…
-
Until now, editing the page directory / tables is not thread safe. We should use Atomics when accessing page tables.
-
Hi,
TLDR : I have a small proxmox server at home and planned on using ROCm for some deep learning tasks I have. I have followed all the different information I could find but still cannot use ROCm…
-
I'm in the process of benchmarking the latencies of different MPI RMA operations such as `put`, `get`, `fetch-op`, `accumulate`, etc. I'm using the Open MPI 4.0.x branch with the UCX 1.6.x branch on a…
-
Recently someone in the channel has tried to do Atomic[SharedPtr], but it's not currently possible. Of course you don't need the encapsulation, could also use an Atomic[bool] for this. But an idea wou…
-
# Summary
On armel, compiling cmake projects using onetbb fails because of missing `-latomic`. We worked around this in Debian using CMake snippets like this:
```cmake
# link with -latomic if n…
josch updated
4 months ago
-
This was introduced on v1.8:
https://buildkite.com/julialang/ordinarydiffeq-dot-jl/builds/1225#0182b0ca-b9b0-4a29-bc10-90bbbab0036d/861-1887
```
| Got exception outside of a @test
| MethodEr…
-
Current `ch4/ucx` netmod does not use hardware-based atomics in RMA. All accumulate-like operations are sent via active message. We should enable hw-based atomics when possible similar to `ch4/ofi`(i.…
-
clspv currently replaces OpenCL atomic builtins with SPIR-V builtins. Instead we should use llvm atomic instructions throughout the flow and translate them in SPIRVProducerPass. Currently some of the …