On Tue, 6 Sep 2005, Franck wrote:
> Alan,
>
> 2005/9/5, Alan Stern <[EMAIL PROTECTED]>:
> > > > On the third hand, who would ever make an embedded hub with only one,
> > > > fixed downstream port? Why not leave out the hub and attach the
> > > > upstream
> > > > port directly to the downstream
Alan,
2005/9/5, Alan Stern <[EMAIL PROTECTED]>:
> > > On the third hand, who would ever make an embedded hub with only one,
> > > fixed downstream port? Why not leave out the hub and attach the upstream
> > > port directly to the downstream device?
> > >
> >
> > surely because the downstream devi
On Tue, 6 Sep 2005, Franck wrote:
> > > Can you point out in the code where is the setup of 4 times per second
> > > interval ? I can't find it.
> >
> > It's in drivers/usb/core/hub.c. The URB is set up near the end of
> > hub_configure(), and it is submitted in hub_activate(). Search for
> > o
2005/9/6, Alan Stern <[EMAIL PROTECTED]>:
> On Mon, 5 Sep 2005, Franck wrote:
>
> > Alan,
> >
> > 2005/9/5, Alan Stern <[EMAIL PROTECTED]>:
> > > On Mon, 5 Sep 2005, Franck wrote:
> > >
> > > > > On the other hand, it's not really all that much useless work. 4
> > > > > times
> > > > > per secon
On Mon, 5 Sep 2005, Franck wrote:
> Alan,
>
> 2005/9/5, Alan Stern <[EMAIL PROTECTED]>:
> > On Mon, 5 Sep 2005, Franck wrote:
> >
> > > > On the other hand, it's not really all that much useless work. 4 times
> > > > per second or so there's a single, very short transaction (IN followed
> > >
On Mon, 5 Sep 2005, Franck wrote:
> 2005/9/1, Alan Stern <[EMAIL PROTECTED]>:
> > On Thu, 1 Sep 2005, Franck wrote:
> >
> > Sure. All the information in the hub descriptor is available, including
> > the number of ports and whether they are fixed or removable.
> >
>
> BTW, I looked at sl811
Alan,
2005/9/5, Alan Stern <[EMAIL PROTECTED]>:
> On Mon, 5 Sep 2005, Franck wrote:
>
> > > On the other hand, it's not really all that much useless work. 4 times
> > > per second or so there's a single, very short transaction (IN followed by
> > > NAK). Unless you've got some good reason for a
On Mon, 5 Sep 2005, Franck wrote:
> > On the other hand, it's not really all that much useless work. 4 times
> > per second or so there's a single, very short transaction (IN followed by
> > NAK). Unless you've got some good reason for avoiding it, why bother?
> >
>
> Well, if work can be avoi
2005/9/1, Alan Stern <[EMAIL PROTECTED]>:
> On Thu, 1 Sep 2005, Franck wrote:
>
> Sure. All the information in the hub descriptor is available, including
> the number of ports and whether they are fixed or removable.
>
BTW, I looked at sl811 host driver and it seems to me that the root
hub desc
On Thu, 1 Sep 2005, Alan Stern wrote:
> On the third hand, who would ever make an embedded hub with only one,
> fixed downstream port? Why not leave out the hub and attach the upstream
> port directly to the downstream device?
Power Quotient International Co., Ltd has. I have here a
flash d
On Thu, 1 Sep 2005, Franck wrote:
> Hi,
>
> Is it really needed to poll an embedded hub that has only one
> downstream port on which a device is plugged forever ?
It depends. If the device can be suspended and is a wakeup source then
the polling is needed, at least whenever the device is suspe
Hi,
Is it really needed to poll an embedded hub that has only one
downstream port on which a device is plugged forever ? If no, Is there
a way to detect it and to avoid useless work ?
Thanks.
--
Franck
---
SF.Net email i
12 matches
Mail list logo
