The code is working on addressing a Linux issue of small reads of one file from many threads contend on the folio reference counting. So the goal of the branch / patch series is to read from the page cache without folio references.

Improvements seem to be 2x - 2.5x in benchmarks

you are viewing a single comment's thread
view the rest of the comments
[–] 3 points 1 month ago

I'm completely onboard with that. I, too, am fatigued by it all, honestly. Of course we can't be completely free of it, because everything has implications, but I appreciate your efforts of keeping the focus on what we as a community are passionate about and interested in. The passion is also that which can provoke some rather emotional responses sometimes.

  • source
  • parent