-
We've iteratively exposed support for various levels of hardware intrinsics since .NET Core 3.1. In .NET 7, we exposed the new "cross platform" hardware intrinsics which aim to help simplify the writi…
-
https://crates.io/crates/crc32fast
Implementation of crc32 with hardware acceleration where applicable. Over 9000 downloads per day. Exposed to untrusted input through a zillion of compression/deco…
-
Could we have support for PowerPC as well? Currently only x86 and arm are there.
-
Vector, Matrix and Quaternion types form `System.Numeric` should implement newly added generic math interfaces like `IMultiplyOperators`, `ISubtractionOperators` etc.
Since these types already use …
-
### Background and motivation
It's been a long time .NET first added `Half` type.
But there's no support for hardware acceleration of conversion, that might be a common use case.
In Ivy Bridg…
-
Hi,
From what I understand Blake2b and then Blake3 were designed with the modern CPUs in mind, and BLAKE3 is currently the fastest secure hash function in modern CPUs.
I wonder how would they comp…
-
Most modern CPUs offer specialized instructions for computing the SHA family of functions. For optimal power usage and maximum throughput in SSZ merkleization, we can benefit from such a hardware-acce…
-
It's that time of year again, and having done some work with Nvidia's new hardware sporting "TensorRT" cores, I thought it'd make a nice project to evaluate whether or not Accelerate programs would be…
-
| | |
|--------------------|----|
| Bugzilla Link | [PR18090](https://bugs.llvm.org/show_bug.cgi?id=18090) |
| Status | NEW |
| Importance | P release bl…
-
I tried this code:
```rust
pub fn foo(x: usize) -> usize {
core::arch::riscv64::clrs8(x)
}
```
I expected to see this happen: it should compile
Instead, this happened: LLVM codegen rejects th…