On Thu, Apr 01, 2004 at 10:36:04AM +0200, Oliver Neukum wrote:
> Am Donnerstag, 1. April 2004 02:46 schrieb Matthew Dharm:
> > On Thu, Apr 01, 2004 at 12:44:41AM +0200, Oliver Neukum wrote:
> > > Am Mittwoch, 31. M�rz 2004 22:42 schrieb Matthew Dharm:
> > > > On Wed, Mar 31, 2004 at 11:17:04AM +0100, Richard Curnow wrote:
> > > > > On Mon, 29 Mar, 2004 at 12:22pm, Matthew Dharm wrote:
> > > > > > I've seen some buggy devices act this way.
> > > > > >
> > > > > > They can't handle the control requests interspersed between
> > > > > > usb-storage transaction requests.  It's a spec violation.
> > > > >
> > > > > So any platform/version combinations where it appears to work are
> > > > > just likely to be luck, then?
> > > >
> > > > That would be my theory.
> > >
> > > What would you propose to do? Grab dev->serialize in both code
> > > paths?
> >
> > To be honest, I don't have a proposal.
> >
> > We've talked about this before on the mailing lists... we've never come up
> > with a good solution.
> 
> It seems to me that the problem screams for a semaphore. The question is
> just which semaphore.

That, and coming up with reasonable semantics to impose on every driver.

And this question:  Do devices with multiple interfaces (and different
drivers attached to those different interfaces) need to interlock on that
semaphore?

Matt

-- 
Matthew Dharm                              Home: [EMAIL PROTECTED] 
Maintainer, Linux USB Mass Storage Driver

SP: I sell software for Microsoft.  Can you set me free?
DP: Natural Selection says I shouldn't.
                                        -- MS Salesman and Dust Puppy
User Friendly, 4/2/1998

Attachment: pgp00000.pgp
Description: PGP signature

Reply via email to