![]() | |||||||||||||
Messages in this thread |
David, >>The proposed fix prevents this issue by using per-sb dentry LRU list. It >>provides very quickly search for the unused dentries for given super block thus >>forcing shrinking always making good progress. > > > We've been down this path before: > > http://marc.theaimsgroup.com/?l=linux-kernel&m=114861109717260&w=2 > > A lot of comments on per-sb unused dentry lists were made in > the threads associated with the above. other solutions were > found to the problem that the above patch addressed, but I don't > think any of them have made it to mainline yet. You might want > to have a bit of a read of these threads first... The major difference between our patch and the one discussed in the link it that we keep both global and per-sb dentry LRU lists. Thus, when needed normal LRU is used and prune logic is unchanged, while umount/remount use per-sb list and do its job faster. Thanks, Kirill - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ | ||||||||||||
| Last update: 2006-10-30 07:25 [from the cache] ©2003-2008 | |||||||||||||