Re: Drivers: scsi: FLUSH timeout

[James Bottomley <jbottomley@parallels.com>]

On Wed, 2013-10-02 at 18:29 +0000, KY Srinivasan wrote:
> 
> > -----Original Message-----
> > From: geert.uytterhoeven@gmail.com [mailto:geert.uytterhoeven@gmail.com]
> > On Behalf Of Geert Uytterhoeven
> > Sent: Wednesday, September 25, 2013 1:40 AM
> > To: KY Srinivasan
> > Cc: Mike Christie; Jack Wang; Greg KH; linux-kernel@vger.kernel.org;
> > devel@linuxdriverproject.org; ohering@suse.com; jbottomley@parallels.com;
> > hch@infradead.org; linux-scsi@vger.kernel.org
> > Subject: Re: Drivers: scsi: FLUSH timeout
> > 
> > On Tue, Sep 24, 2013 at 11:53 PM, KY Srinivasan <kys@microsoft.com> wrote:
> > > I am not sure how that magic number was arrived at (the 60HZ number). We
> > want this to be little higher -
> > 
> > "60 * HZ" means "60 seconds".
> > 
> > > would there be any issues raising this to say  180 seconds. This is the value we
> > currently have for I/O
> > > timeout.
> > 
> > So you want to replace it by "180 * HZ", which is ... another magic number?
> 
> Ideally, I want this to be adjustable like the way we can change the I/O timeout.
> Since that has been attempted earlier and rejected (not clear what the reasons were),
> I was suggesting that we pick a larger number. James, let me know how I should proceed here.

It's only one problem device with its own driver, right?  how about a
per target value you set in target_configure?  That way there's no
damage to the rest of the system even in a mixed device environment.  I
don't see a particular need to expose this via sysfs unless someone else
has a use case.

James