> On 12. Sep 2017, at 20:52, Joerg Sonnenberger <jo...@bec.de> wrote: > > Hello Juergen, > the new VSTATE_ASSERT_UNLOCKED code you added to genfs_lock is very > heavy as it significantly increases lock contention for something > already highly contended. I'd like to push the lock down to critical > cases and try to avoid it for common cases. > > Thoughts? > > Joerg > <state-assert-perf.diff>
Looks good to me, cannot test anything as I am on vacation. Please commit and request pullup to -8. -- J. Hannken-Illjes - hann...@eis.cs.tu-bs.de - TU Braunschweig (Germany)