-
# Summary
When `std::sort(std::execution::par_unseq,...)` is called repeatedly it leaks memory. Ultimately it will exhaust the machine's memory and the kernel will kill the process.
Single thread…
-
| | |
| --- | --- |
| Bugzilla Link | [27837](https://llvm.org/bz27837) |
| Version | unspecified |
| OS | Windows NT |
| Reporter | LLVM Bugzilla Contributor |
| CC | @hfinkel,@mclow |
## Extended…
-
| | |
|--------------------|----|
| Bugzilla Link | [PR27837](https://bugs.llvm.org/show_bug.cgi?id=27837) |
| Status | NEW |
| Importance | P normal |
|…
-
## Current issue
- [ ] Infeasible to merge multiple views
- [ ] Cannot support multiple views (e.g. [A, B] -> View -> Shape cannot be inferenced)
- [ ] Parallelize Attention QKV Projection
…
-
## Relations
## Description
Currently we always divide our plane into 4 chunks. I we need to be able to split into variable amount of chunks.
rexim updated
8 years ago
-
### Generalizing tensor variable elimination to dynamic factor graphs
- [x] #398 A version of `modified_partial_sum_product` that works for time lag 1, and tests that compare it to `partial_sum_pro…
-
### 🚀 The feature, motivation and pitch
It would be great to have a general parallel prefix sum (associative scan) operation in PyTorch, something like [associative_scan](https://jax.readthedocs.io…
-
### Discussed in https://github.com/scikit-learn/scikit-learn/discussions/20651
In forest algorithms, the preferred parallelization backend is `threading`. However, it looks that it is not anymore …
-
**Describe the feature request**
Use a smarter selection algorithm when proxying DNS requests upstream serially. Examples include using the original destination IP or randomly iterating through the l…
-
While developing a modification for the CCDCiel internal guider we noticed the following:
RA & DEC dither pulse commands are executed sequential. First the RA then DEC dither pulse is processed.
…