-
LLVM does not do any vectorization in cases like below.
GCC does better job here.
```
auto foo(unsigned const * __restrict in, unsigned const *__restrict sel, unsigned *__restrict out)
{
fo…
-
This is a collection of further possible auto-tuning parameters:
- [ ] Traversals: chunk size of OpenMP schedule dynamic. This is currently set to 1, but with many threads, this can be problematic (e…
-
my laptop screen is dead and i cant be arsed to type on a phone or i would be sending a PR already
anyway apple does this extensively with clang, and I don't expect the performance to be terrible …
-
Hi !
Would it make sense to have the ability to build for Neon with auto-vectorization (which should be rather well supported by gcc these days), to get improved performances on things like Raspber…
-
If ```-ffast-math``` is included on the gcc compile line, the resulting code segfaults. This option seems desirable because it might improve mkFit's timing performance, by enabling a higher degree of …
-
Hello experts,
I am trying to compile blingfire wbd tokenizer (latest release) on Mac following https://github.com/Microsoft/BlingFire/wiki/How-to-change-linguistic-resources
I am encountering an …
-
## Problem statement
Creating of new nodes is quite a big part of Sverckok development. It is not optimized now and has some problems. This issue is intend to clear what problems should be solved d…
-
Hotspot has support for [auto-vectorization](https://cr.openjdk.java.net/~vlivanov/talks/2017_Vectorization_in_HotSpot_JVM.pdf). That is, the jvm will identify certain looping patterns on arrays and g…
-
| | |
|--------------------|----|
| Bugzilla Link | [PR27625](https://bugs.llvm.org/show_bug.cgi?id=27625) |
| Status | NEW |
| Importance | P normal |
|…
-
### Describe the enhancement requested
Users should generally prefer using the checked versions of arithmetic functions such as "add", because unchecked arithmetic is inherently unsafe and can silent…