If filesystem has to guarantee write ordering, it has to serialize write request. So, only when 'A' is written to the block (completion received from block driver) filesystem can dispatch 'B' not before that. Remember that there could be failures too. So if 'A' is failed, request for 'B' is pending, would a filesystem fail that too? I guess not. Usually page cache absorbs such overwrites first than issues a single merged IO to block device, but what about direct IO? If applications need consistency guarantee against such failures, it should serialize writes. So to summarize, write ordering effort from each layers Block driver: none Filesystem: best effort Application: full
Rajat ------------------------------ From: neha naik Sent: 03-08-2013 02:40 To: Greg Freemyer Cc: Rajat Sharma; kernelnewbies Subject: Re: Conceptual questions about device driver Thanks for the responses. I have one more question for Greg. I come from filesystem background and not device driver so i may be a bit confused about the write order fidelity. I know that filesystems guarantee that. Looking from filesystem perspective, no write will be allowed on the same block until the first write finishes. So, if 'B' is written after 'A' you can always guarantee that you will see 'B' at the end of the two writes. Now imagine not having a filesystem, and doing a write directly on the device. Do device drivers honour it. Should they? I imagine device driver as a kind of queue. So any writes are always queued up one after the other so that it gives write order fidelity whether it wants to or not. Am i missing something here. Regards, Neha On Fri, Aug 2, 2013 at 1:56 PM, Greg Freemyer <[email protected]>wrote: > On Fri, Aug 2, 2013 at 1:32 AM, Rajat Sharma <[email protected]> wrote: > > On Fri, Aug 2, 2013 at 2:25 AM, neha naik <[email protected]> wrote: > >> Hi, > >> I have some conceptual questions about device driver : > >> > >> 1. Write order fidelity should be maintained when submitting requests > from > >> device driver to disk below. > >> However, acknowledging these requests it is okay if we don't > necessarily > >> maintain that order, right? > >> > > > > Yes it should not matter as long as application can rely on data being > > written is in order of submission. > > But it can't ..... unless the write cache is turned off and it is > known the the cache is truly off. > > There is no guarantee of write order in the block stack. Not between > the filesystem and the driver. Not between the driver and the drive. > > There are at least 2 elevators shuffling the order of writes to > optimize performance. > > Rajat, did you get confused? Or were you trying to say something else? > > Greg >
_______________________________________________ Kernelnewbies mailing list [email protected] http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
