On Sat, Sep 20, 2025 at 4:42 PM David Laight <david.laight.li...@gmail.com> wrote: > > On Wed, 17 Sep 2025 18:05:45 +0200 > Mateusz Guzik <mjgu...@gmail.com> wrote: > > I can agree the current state concerning ep_events_available() is > > avoidably error prone and something(tm) should be done. fwiw the > > refcount thing is almost free on amd64, I have no idea how this pans > > out on arm64. > > Atomic operations are anything but free.... > They are likely to be a similar cost to an uncontested spinlock entry. >
In this context it was supposed to be s/refcount/seqcount/ and on amd64 that's loading the same var twice + a branch for the read thing. Not *free* but not in the same galaxy comped to acquiring a spinlock (even assuming it is uncontested).
