Messages in this thread |  | | From | (david parsons) | | Subject | Re: Deletion of big files... | | Date | 21 May 1999 11:51:37 -0700 |
| |
In article <linux.kernel.Pine.LNX.4.10.9905180149420.8976-100000@tahallah.demon.co.uk>, Alex Buell <alex.buell@tahallah.demon.co.uk> wrote: >On Mon, 17 May 1999, Alexander Viro wrote: > >> Hell knows. I can see one thing where it would be a clear win - if we >> are deep down on free blocks we will be better off taking them from >> the queue instead of searching through bitmaps. This would require a >> bit different eater on the blocks queue (different behaviour when fs >> is low on free space; instead of marking blocks free put them into the >> small cache where they can be found fast; finding free block would >> involve checking the cache if we are low on free blocks). > >Why not just queue them onto a kernel thread (i.e krmd) and let it delete >the files when the system is idle?
That would be good for implementing a proper undelete; when an inode is freed, instead of trashing it just give it to the delete daemon and have it cling to it (or write it into a trashcan), then delete it at whatever the configured deletion criteria are.
____ david parsons \bi/ The "ohmygod" command is sorely missed. \/
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
|  |