Re: cleanup the RAID6 P/Q library v3

"H. Peter Anvin" <[email protected]>
Newsgroups gmane.linux.ports.riscv,gmane.linux.kernel,gmane.linux.ports.arm.kernel,gmane.linux.ports.ppc64.devel,gmane.linux.kernel.cryptoapi,gmane.comp.file-systems.btrfs,gmane.linux.kernel.cross-arch,gmane.linux.raid
Message-ID <[email protected]>
On June 29, 2026 5:21:52 AM PDT, Christoph Hellwig <[email protected]> wrote:
>On Fri, Jun 26, 2026 at 05:52:18PM -0700, H. Peter Anvin wrote:
>> On 2026-05-19 01:24, Christoph Hellwig wrote:
>> > 
>> > raid6: rework registration of optimized algorithms
>> > 
>> >  - avx2 instead of avx512 is probably the right thing for no
>> >    benchmarking, but if it was intentional (it wasn't), that should
>> >    be document.  So I'll just switch back to the previous version to
>> >    keep the state of the art
>> 
>> It is unlikely to be the right thing *going forward*, though.
>> 
>> The very unfortunate performance inversion is likely model-specific. It is one
>> of those things where you largely would have to have a list of quirks :(
>
>Note that the default is still to benchmark the implementations.
>This is just about the case where the user explicitly disabled that
>benchmark-based selection.
>

The real issue though is that the benchmarking isn't really helpful, either. 

It has to do with power across the whole chip, so it perversely enough depends on what *else* is going on...
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.