On Tue, Feb 27, 2024 at 08:34:39AM -0800, Linus Torvalds wrote: > On Mon, 26 Feb 2024 at 23:22, Kent Overstreet <kent.overstreet@xxxxxxxxx> wrote: > > > > Only rough testing, but this is looking like around a 25% performance > > increase doing 4k random reads on a 1G file with fio, 8 jobs, on my > > Ryzen 5950x - 16.7M -> 21.4M iops, very roughly. fio's a pig and we're > > only spending half our cpu time in the kernel, so the buffered read path > > is actually getting 40% or 50% faster. > > > > So I'd say that's substantial. > > No, you're doing something wrong. The new fastread logic only > triggers for reads <= 128 bytes, so you must have done some other > major change (like built a kernel without the mitigations, and > compared it to one with mitigations - that would easily be 25% > depending on hardware). Like I replied to willy - the "4k" was a typo from force of habit, I was doing 64 byte random reads.