Messages in this thread | | | Date | Wed, 12 Oct 2022 19:22:21 +0800 | Subject | Re: [External] Re: [RFC] mm: add new syscall pidfd_set_mempolicy() | From | Zhongkun He <> |
| |
> > Yes, this will require some refactoring and one potential way is to make > mpol ref counting unconditional. The conditional ref. counting has > already caused issues in the past and the code is rather hard to follow > anyway. I am not really sure this optimization is worth it. > > Another option would be to block the pidfd side of things on completion > which would wake it up from the task_work context but I would rather > explore the ref counting approach first and only if this is proven to be > too expensive to go with hacks like this.
Hi Michal
The counting approach means executing mpol_get/put() when start/finish using mempolicy,right? With the addition of lock add/dec on the hot path, the performance may be degraded.
I'll try it to see its performance impact in detail.
Thanks.
| |