On 5/29/2023 10:58 AM, Yin Fengwei wrote: > Hi Andrew, > > On 5/4/23 19:48, kirill@xxxxxxxxxxxxx wrote: >> On Sat, Apr 29, 2023 at 04:27:58PM +0800, Yin Fengwei wrote: >>> free_transhuge_page() acquires split queue lock then check >>> whether the THP was added to deferred list or not. It brings >>> high deferred queue lock contention. >>> >>> It's safe to check whether the THP is in deferred list or not >>> without holding the deferred queue lock in free_transhuge_page() >>> because when code hit free_transhuge_page(), there is no one >>> tries to add the folio to _deferred_list. >>> >>> Running page_fault1 of will-it-scale + order 2 folio for anonymous >>> mapping with 96 processes on an Ice Lake 48C/96T test box, we could >>> see the 61% split_queue_lock contention: >>> - 63.02% 0.01% page_fault1_pro [kernel.kallsyms] [k] free_transhuge_page >>> - 63.01% free_transhuge_page >>> + 62.91% _raw_spin_lock_irqsave >>> >>> With this patch applied, the split_queue_lock contention is less >>> than 1%. >>> >>> Signed-off-by: Yin Fengwei <fengwei.yin@xxxxxxxxx> >> >> Acked-by: Kirill A. Shutemov <kirill.shutemov@xxxxxxxxxxxxxxx> > I didn't get the green light for patch2 (which was trying to reduce lru lock contention) > from Matthew. It may need more time to figure out how to reduce lru lock contention. Andrew, thanks a lot for picking this patch. Regards Yin, Fengwei > > I am wondering whether this patch1 (without patch2) can be picked as > - It has nothing to do with patch2 > - It could reduce the deferred queue lock contention. > - It got acked-by from Kirill. > > Let me know if you want me to resend this patch1. Thanks. > > > Regards > Yin, Fengwei > >>
