If the driver submits a group of requests to of
sizes 16 bytes, 10 bytes, 14 bytes ... I'd expect
the driver to know the device would handle that!
And to be responsible for recovering in any case
where it couldn't.
The USB transfer model is "chunky", and it'd
be error-prone (IMO) to try to present any
kind of non-chunky model to drivers.
----- Original Message -----
From: "Johannes Erdfelt" <[EMAIL PROTECTED]>
To: "David Brownell" <[EMAIL PROTECTED]>; "USB Developers"
<[EMAIL PROTECTED]>
Sent: Tuesday, September 18, 2001 2:47 PM
Subject: Re: [linux-usb-devel] URB with scatter-gather?
> Yes, but I wasn't worried about underrun in the device to host case. It
> can be a problem in the host to device case tho.
>
> If we want to transfer 40 bytes, I wouldn't trust every device to handle
> a 16 byte packet, followed by a 10, finished by a 14 if the endpoint
> size is 16 bytes.
>
> JE
>
> On Tue, Sep 18, 2001, Matthew Dharm <[EMAIL PROTECTED]> wrote:
> > Overrun is an error in all cases... but underrun (at least on device to
> > host) is not. In fact, it's pretty normal.
> >
> > Matt
> >
> > On Tue, Sep 18, 2001 at 05:21:08PM -0400, Johannes Erdfelt wrote:
> > > On Tue, Sep 18, 2001, David Brownell <[EMAIL PROTECTED]> wrote:
> > > > > > > Not really, since for lots of devices, short packets mean something
> > > > > > > special if we send one.
> > > > >
> > > > > > I think you were assuming something different about s/g semantics
> > > > > > than I was, then. If the driver chooses to put a short packet into
> > > > > > an s/g list, I expected it did so to get that special effect. Perhaps
> > > > > > you were thinking it was expecting the HCD to coalesce some
> > > > > > buffers? How about just returning EINVAL? :)
> > > > >
> > > > > I think he means short packet in, not short packet out.
> > > >
> > > > Except he said "send" ... ;)
> > > >
> > > > For "in", why shouldn't a short packet terminate the whole series
> > > > of transfers? Underrun, overrun ... in both cases, the device and
> > > > driver are significantly out-of-sync.
> > >
> > > I think you're talking about something else. Obviously an underrun or
> > > overrun should terminate all of the transfers.
> > >
> > > What I was talking about was not putting ourselves into the situation
> > > where we would have overrun's.
> > >
> > > An overrun is a bug, in every case. It might be the host or the device,
> > > but it should never happen.
> > >
> > > JE
> >
> > --
> > Matthew Dharm Home: [EMAIL PROTECTED]
> > Maintainer, Linux USB Mass Storage Driver
> >
> > C: Like the Furby?
> > DP: He gives me the creeps. Think the SPCA will take him?
> > -- Cobb and Dust Puppy
> > User Friendly, 1/2/1999
>
>
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
