> > Yes, "should be". I could understand that the presence of huge > > nunmbers of -ve dentries could result in undesirable reclaim of > > pagecache, etc. Triggering oom-killings is very bad, and presumably > > has the same cause. > > > > Before we go and add a large amount of code to do the shrinker's job > > for it, we should get a full understanding of what's going wrong. Is > > it because the dentry_lru had a mixture of +ve and -ve dentries? > > Should we have a separate LRU for -ve dentries? Are we appropriately > > aging the various dentries? etc. > > > > It could be that tuning/fixing the current code will fix whatever > > problems inspired this patchset. > > What I think is contributing to the problems and could lead to reclaim > oddities is the internal fragmentation of dentry slab cache. Dentries are > relatively small, you get 21 per page on my system, so if trivial to > reclaim negative dentries get mixed with a small amount of unreclaimable > positive dentries, you can get a lot of pages in dentry slab cache that are > unreclaimable. Could we allocate -ve entries from separate slab? -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
