On Tue, Feb 27, 2024 at 5:45 PM Michal Hocko <mhocko@xxxxxxxx> wrote: > > On Tue 27-02-24 17:39:01, Yafang Shao wrote: > > On Tue, Feb 27, 2024 at 5:05 PM Michal Hocko <mhocko@xxxxxxxx> wrote: > > > > > > On Tue 27-02-24 13:48:31, Yafang Shao wrote: > > > > On Mon, Feb 26, 2024 at 10:05 PM Michal Hocko <mhocko@xxxxxxxx> wrote: > > > [...] > > > > > > To manage disk > > > > > > storage efficiently, we employ an agent that identifies container images > > > > > > eligible for destruction once all instances of that image exit. > > > > > > > > > > > > However, during destruction, dealing with directories containing numerous > > > > > > negative dentries can significantly impact performance. > > > > > > > > > > Performance of what. I have to say I am kind of lost here. We are > > > > > talking about memory or a disk storage? > > > > > > > > Removing an empty directory with numerous dentries can significantly > > > > prolong the process of freeing associated dentries, leading to high > > > > system CPU usage that adversely affects overall system performance. > > > > > > Is there anything that prevents you from reclaiming the memcg you are > > > about to remove? We do have interfaces for that. > > > > Reclaiming numerous dentries through force_empty can also lead to > > potential issues, which is why we attempt to shrink the slab gradually > > to mitigate them. However, it's important to note that the underlying > > causes of the issues in force_empty and rmdir are not identical, as > > they involve different locks. > > Please be more specific about those issues. Both of these issues stem from lock contention: - rmdir When executing rmdir, the lock of the inode of the empty directory is held. If this directory contains numerous negative dentries, this lock is held for an extended duration. Consequently, if other processes attempt to acquire this lock, they are blocked. A simple reproducer involves: 1. Generating numerous negative dentries in an empty directory. 2. Running `rmdir ~/test` and `ls ~/` concurrently. This setup demonstrates that ls takes a significant amount of time to complete due to lock contention. - force_empty Force_empty holds the lock of super_block->dentry_list. However, I haven't yet had the opportunity to produce a specific example to illustrate this issue. > > > > > > > To mitigate this > > > > > > issue, we aim to proactively reclaim these dentries using a user agent. > > > > > > Extending the memory.reclaim functionality to specifically target slabs > > > > > > aligns with our requirements. > > > > > > > > > > Matthew has already pointed out that this has been proposed several > > > > > times already and rejected. > > > > > > > > With that being said, we haven't come up with any superior solutions > > > > compared to the proposals mentioned. > > > > > > > > > Dedicated slab shrinking interface is > > > > > especially tricky because you would need a way to tell which shrinkers > > > > > to invoke and that would be very kernel version specific. > > > > > > > > The persistence of this issue over several years without any > > > > discernible progress suggests that we might be heading in the wrong > > > > direction. Perhaps we could consider providing a kernel interface to > > > > users, allowing them to tailor the reclamation process based on their > > > > workload requirements. > > > > > > There are clear problems identified with type specific reclaim and those > > > might easily strike back with future changes. Once we put an interface > > > in place we won't be able remove it and that could lead to problems with > > > future changes in the memory reclaim. > > > > That shouldn't deter us from actively seeking a resolution to an issue > > that has persisted for tens of years. > > Right, I do not believe we would deter anybody from doing that. This is > just not an easy problem to tackle. So either you find solid arguments > that previous conclusions do not hold anymore or you need to look into > options which haven't been discussed so far. I do realize that chasing > previous discussions in email archives is not fun but maybe a good (re)start > would be documenting those problems somewhere under Documentation/. > > > As observed, numerous memcg interfaces have been deprecated in recent years. > > yes, for very good reasons -- Regards Yafang
