-
### Steps to reproduce
I'm still not sure about exact conditions, despite I experience this for many months. It happens with GIFs and sometimes with static images too. It's enough to switch to a chat…
-
[3509557.600364] oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0,global_oom,task_memcg=/user.slice/user-1000.slice/session-4362.scope,task=classifyExact,pid=889353,uid=1000…
-
Doing this generally may require first add Nv to the type parameter space (https://github.com/CliMA/ClimaCore.jl/issues/11).
-
I think the [Wiki page "Hierarchical Parallelism"](https://kokkos.org/kokkos-core-wiki/ProgrammingGuide/HierarchicalParallelism.html) is potentially very valuable to users, as it tackles some slightly…
-
Does KernelAbstractions.jl support auto-setting workgroupsize when the kernel has local memory size that depends on groupsize? For example, `CUDA.launch_configuration` takes a `shmem` callback that ma…
-
## Problem
As per #194, we are planning to separate the main Fortran replacement interface proposal into multiple smaller parts. This issue tracks the first part of the proposal - to add and support …
-
The shmem library uses `shm_open()` to create shared memory files in `/dev/shm`. These files are mapped into shadow and the shim.
We should consider using memfd files instead of `/dev/shm` filesyst…
-
Calling `DLB_TALP_Attach()` from outside calls `shmem_cpuinfo_ext__init()` and `shmem_procinfo_ext__init()`. They both call `open_shmem()` -> `shmem_init()` -> `shm_open()`+ `ftruncate()`. This create…
-
Under Ubuntu-18.04.5, with gcc-8.3, and the ucx master branch,
```
$./autogen.sh
$./contrib/configure-release --prefix=$HOME/soft/ucx --with-cuda CC=gcc CXX=g++
$make
...
Making install in test/…
-
## Problem
OpenSHMEM specifies only the string `SHMEM_VENDOR_STRING` for implementations to convey version information. This does not give programmers an easy way to determine what version of an Op…