LLVM's libc Gets Much Faster memcpy For RISC-V

[phoronix]

Phoronix: LLVM's libc Gets Much Faster memcpy For RISC-V

Being merged into LLVM's libc library recently are an optimized memory copy function "memcpy" implementation for the RISC-V architecture...

https://www.phoronix.com/news/LLVM-libc-Faster-memcpy

[scottishduck]

When this libc is production ready it will be a very interesting time.

[FireBurn]

I'm looking forward to libc supporting shared libraries

[zexelon]

Now if only we could get much faster RISC-V parts!

[tuxd3v]

Now if only we could get much faster RISC-V parts!

The Chinese opensource Xiangshan 2nd release@14nm is already faster than CortexA76.

It is expected to be taped out at SMIC for engineering samples, so that they can fine tune it in June 2023.
I saw some things saying they will target ARM neoverse-N2...maybe for next gen.

But we need to understand that a SoC is much more than just the cpu itself, its a lot of things that are around him, that really add value to the product.