lkml.org 
[lkml]   [2013]   [Jan]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
SubjectRe: [Xen-devel] [PATCH 3/3] xen/blkback: Check for insane amounts of request on the ring.
On Mon, Jan 28, 2013 at 04:03:49PM +0000, David Vrabel wrote:
> On 28/01/13 15:44, Konrad Rzeszutek Wilk wrote:
> > On Mon, Jan 28, 2013 at 11:38:50AM +0000, David Vrabel wrote:
> >> On 25/01/13 18:43, Konrad Rzeszutek Wilk wrote:
> >>>
> >>> Check that the ring does not have an insane amount of requests
> >>> (more than there could fit on the ring).
> >> [...]
> >>> --- a/drivers/block/xen-blkback/blkback.c
> >>> +++ b/drivers/block/xen-blkback/blkback.c
> >> [...]
> >>> @@ -415,8 +415,12 @@ int xen_blkif_schedule(void *arg)
> >>> blkif->waiting_reqs = 0;
> >>> smp_mb(); /* clear flag *before* checking for work */
> >>>
> >>> - if (do_block_io_op(blkif))
> >>> + rc = do_block_io_op(blkif);
> >>> + if (rc > 0)
> >>> blkif->waiting_reqs = 1;
> >>> + if (rc == -EACCES)
> >>> + wait_event_interruptible(blkif->shutdown_wq,
> >>> + kthread_should_stop());
> >>
> >> So if this happens then the backend silently stops processing any
> >> further requests? I think it should print an error and disconnect from
> >> the front end.
> >
> > I was not sure what it should print?
> >
> > "Frontend provided bogus ring requests. Halting ring processing"?
>
> That would be fine but you may want to print the bad values, in case
> they're useful for tracking down bugs in frontends.

OK.
>
> > It does eventually disconnect from the front-end as the XenBus enters
> > the Disconnect phase. It seemed the strike the balance - so that if
> > the frontend negotiates again the ring can start again and penalize
> > the frontend for giving bogus data.
>
> Ok.
>
> David


\
 
 \ /
  Last update: 2013-01-28 18:42    [W:2.005 / U:1.436 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site