riginal Message-
From: thomas chen <[EMAIL PROTECTED]>
To: David Brownell <[EMAIL PROTECTED]>
Cc: Greg KH <[EMAIL PROTECTED]>; [EMAIL PROTECTED]
<[EMAIL PROTECTED]>
Date: Thursday, January 16, 2003 7:49 PM
Subject: Re: [linux-usb-devel] NACK from device
>
>thanks
thanks for the pointers it turned out that it is being queued
but for the wrong reason...
in the submit_urb.. (this is the SL811)... it checks the
RH.DEVNUM (which is 1)... against the PIPEDEVICE which is 2...
so it does not send it out and queue it up... and i am now trying
to find out wher
thomas chen wrote:
ohci can handle two urb_submit by DELAYing one for the next frame?
No, it handles them by _queueing_ the second one so that it'll
start after the first one finishes. In the case of OHCI, the
typical case is that the hardware automatically advances that
queue ... no software a
ohci can handle two urb_submit by DELAYing one for the next frame?
i guess the question is that can two packets be sent in the same frame?
if not.. then the driver is responsible for delaying one??? i guess that
is the problem... the driver i have does not seem to do that properly
and slave start
On Thu, Jan 16, 2003 at 07:33:14PM -0500, thomas chen wrote:
> can TWO submit_urb be performed within the same FRAME?
Well UHCI and OHCI can do it, I don't know if your driver can :)
greg k-h
---
This SF.NET email is sponsored by: Thawte.com
U
can TWO submit_urb be performed within the same FRAME?
i basically tracked it down to usb.c making TWO submit_urb request
within the same 1MS frame can this be done? or should the lower
level driver take responsibility to queue them up and delay second
one by one frame ???
basically when thi
On Thu, Jan 16, 2003 at 03:27:15PM -0500, thomas chen wrote:
>
> and there is a prolific 2303 serial converter... it is hooked up to a
> small GPS (look like a mouse) receiver running at 4800 bps...
So the control messages sent to the device are not completing? The bulk
urbs that are sent to the
here is the set up
i have a ARM7TDMI based board with SL811HS chip
running uclinux 2.4.19
and there is a prolific 2303 serial converter... it is hooked up to a
small GPS (look like a mouse) receiver running at 4800 bps...
Greg KH wrote:
>
> On Thu, Jan 16, 2003 at 03:06:01PM -0500, thom
On Thu, Jan 16, 2003 at 03:06:01PM -0500, thomas chen wrote:
>
> well... i am getting usb.c complaining about "bulk_msg: timeout"
>
> and i traced it down to the chip level where SL811 is getting NACK
> from the slave (device)... and because of that... the urb->complete is
> not getting
> called
well... i am getting usb.c complaining about "bulk_msg: timeout"
and i traced it down to the chip level where SL811 is getting NACK
from the slave (device)... and because of that... the urb->complete is
not getting
called... and usb task is then spitting out that timeout message
is this a non-i
On Thu, Jan 16, 2003 at 01:58:39PM -0500, thomas chen wrote:
> i am working on a SL811 host controller in a ARM7 based embeded system
> running uclinux
>
> i am continusouly getting NACK from slave device after a while...
Nothing wrong with that, it means the device doesn't have any data to
send
i am working on a SL811 host controller in a ARM7 based embeded system
running uclinux
i am continusouly getting NACK from slave device after a while...
can someone point me to a good resource about the low level usb
handshaking??? why is this happening? the same device seems to work
ok with intel
12 matches
Mail list logo
