Re: [PATCH] Give kjournald a IOPRIO_CLASS_RT io priority

On Thu, Oct 02, 2008 at 05:32:04PM +0200, Bodo Eggert wrote:
Jens Axboe <jens.axboe@xxxxxxxxxx> wrote:
On Thu, Oct 02 2008, Dave Chinner wrote:
On Thu, Oct 02, 2008 at 09:55:11AM +0200, Jens Axboe wrote:

Good point. I think we should mark the IO as sync, and maintain the same
priority level. Any IO that ends up being waited on is sync by
definition, we just need to expand the coverage a bit.

That's what XFS has always done - mark the journal I/O as sync.
Still, once you load up the elevator, the sync I/O can still get
delayed for hundreds of milliseconds before dispatch, which was
why I started looking at boosting the priority of the log I/O.
It proved to be much more effective at getting the log I/O
dispatched than the existing "mark it sync" technique....

Sure, just marking it as sync is not a magic bullet. It'll be in the
first priority for that class, but it'll share bandwidth with other
processes. So if you have lots of IO going on, it can take hundreds of
miliseconds before being dispatched.

Sounds like you need a priority class besides sync and async.

There's BIO_META now as well, which I was testing at the same time
as RT priority. Marking all the metadata I/O as BIO_META did help,
but once again I never got to determining if that was a result of
the different tagging or the priority increase.

However, given that only CFQ understand BIO_META, I suspect that
changing the way XFS uses BIO_SYNC to be a combination of BIO_META
and BIO_SYNC would cause significant regressions on other
schedulers.....

Cheers,

Dave.
--
Dave Chinner
david@xxxxxxxxxxxxx
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Relevant Pages

  • Re: [PATCH] Give kjournald a IOPRIO_CLASS_RT io priority
    ... So it would really need a new priority class between RT and normal priority. ... I think we should mark the IO as sync, ... delayed for hundreds of milliseconds before dispatch, ... It proved to be much more effective at getting the log I/O ...
    (Linux-Kernel)
  • Re: [PATCH] Give kjournald a IOPRIO_CLASS_RT io priority
    ... So it would really need a new priority class between RT and normal priority. ... I think we should mark the IO as sync, ... That's what XFS has always done - mark the journal I/O as sync. ...
    (Linux-Kernel)
  • Re: Background thread with lower priority
    ... records and saves them to the local database. ... background sync takes about 20 seconds. ... Running it at a lower priority would allow the UI to get quantum before ...
    (microsoft.public.dotnet.framework.compactframework)
  • Re: Tasks losing information between 2 PC syncs
    ... I'm using ActiveSync 4.2 and date/time/zone is consistent. ... The priority change is reflected on the attached device immediately, ... ActiveSync", as either 4.2 or 4.5beta might fit that description. ... I only sync Contacts, Tasks, Notes. ...
    (microsoft.public.pocketpc.activesync)
  • Re: get_fs_excl/put_fs_excl/has_fs_excl
    ... I/O request is processed? ... I tend to think of it more as an I/O priority issue, and specifically, ... filesystem in a ionice'd process, I could imagine a badly written ... FREEZE ioctl, where the filesystem has been frozen --- the very ...
    (Linux-Kernel)
Loading