Linux 7.4 Slated To Land The AVX-512 Optimized xor_gen For A Big Performance Win

Over the past few months there have been patches floating around for an AVX-512 optimized xor_gen() for the Linux kernel for that functuon used by the Linux software RAID code. It's been shhowing very nice performance gains with this function being used for generating and validating parity blocks such as for RAID5 and RAID6. The upcoming Linux 7.4 cycle is slated to merge this nice enhancement for benefiting AMD Zen 4 and newer or current Intel server CPUs with AVX-512.

Eric Biggers of Google has been leading this effort for delivering nice xor_gen() performance gains with AVX-512, just as he's done for a lot of other Linux crypto code with stellar performance improvements over recent years, especially around AVX-512 and other modern CPU ISA capabilities.

After the latest revision of the patches, they are now queued for being submitted with the Linux 7.4 merge window happening in the second half of October.

Queued last week via the TIP's x86/fpu Git branch is the patch enabling the AVX-512 xor_gen() to very nice gains as shown on an AMD Ryzen 9 9950X:

With it now in a TIP branch, look for it with the very exciting Linux 7.4 kernel barring any last minute surprises.