Re: 1394 Listen

From: Bill McKenzie (bill.mckenzie_at_nospam.conexant.com)
Date: 08/04/04 

Date: Wed, 4 Aug 2004 08:47:19 -0500

I have written isoch 1394 drivers on 2000, and they worked just fine. It
has been my experience that the 1394 stack is SEVERLY flawed on all
platforms, but you can work around it in many cases. 1394 has been working
for Windows since 98 so I think writing off 2000 is a poor idea.

I would be interested to see the OP's code where the transfer is setup,
perhaps there is something there that we can point out an error in. 

-- 
Bill McKenzie
Software Engineer - Prism 802.11 Wireless Solutions
Conexant Systems, Inc.
"Raj" <r_konjeti@mailcity.com> wrote in message
news:8509fde8.0408040520.77638a67@posting.google.com...
> 1) Are you using Windows 2000 or Windows XP. Microsoft stack for
> Windows 2000 is not perfect. Dont waste time with 2000 driver. Move to
> Windows XP stack instead, if you want to use microsoft stack.
>
> 2) How is your isoch listen header. Do you want to receive the header
> or not. If you dont want header, follw instructions like ORing correct
> flag. You have provide extra 8 bytes if you need header. If you dont
> need header,
>
>   AllocRsrc.fulFlags |= RESOURCE_STRIP_ADDITIONAL_QUADLETS;
>   AllocRsrc.nQuadletsToStrip = 1;
>
> 3. Packet based or stream based. Packet based is used for variable
> length packets.
>
> 4. Keep all data lengths in quadlets(multiples of 4).
>
> 5. Another issue is driver issue. You have to free your buffer and
> complete Irp in detach complete callback. Since your Isoch Tx is
> working, I hope you know exactly what you are doing in driver.
>
> Thanks.
>
> "Eddie Hoffman" <kpeter6448@hotmail.com> wrote in message
news:<uTD4VEaeEHA.1692@tk2msftngp13.phx.gbl>...
> > Greetings,
> >     Can anybody explain how to get 1394 Listening working?  I followed
the
> > exact procedure outlined on:
> >
> >
http://msdn.microsoft.com/library/en-us/buses/hh/buses/1394-isoch_5b296f25-f7e4-47f9-b724-8a03419363ed.xml.asp?frame=true
> >
> > After attaching a buffer using  REQUEST_ISOCH_ATTACH_BUFFERS  and then
using
> > REQUEST_ISOCH_LISTEN  to start listening, my ISOCH_DESCRIPTOR callback
that
> > I assigned when attaching the buffer never gets called.  I have
"talking"
> > implemented on a seperate channel and I see that working properly, but I
> > have not been able to listen successfully.  On the bus analyzer I can
see
> > the data that I'm supposed to be listening to going across, but I never
> > receive any notification that my buffer was filled in.  All functions
during
> > channel setup report STATUS_SUCCESS.  Anybody know of anything I may be
over
> > looking?
> >
> > Thanks,
> > Eddie

Relevant Pages

  • Re: [PATCH 0/3] New firewire stack
    ... My main point about ohci1394 (the old stacks PCI driver) is, ... The big problems in the ohci1394 drivers is the irq_handler, bus reset ... reset, so there is no need to complicate the core stack with this extra state, ... interfaces have slightly disjoint feature sets and can't really be phased out. ...
    (Linux-Kernel)
  • Re: [PATCH 0/3] New firewire stack
    ... My main point about ohci1394 (the old stacks PCI driver) is, ... The big problems in the ohci1394 drivers is the irq_handler, bus reset ... reset, so there is no need to complicate the core stack with this extra state, ... interfaces have slightly disjoint feature sets and can't really be phased out. ...
    (Linux-Kernel)
  • [crash] BUG: unable to handle kernel NULL pointer dereference at (null), last sysfs file: /sys/devic
    ... Stack: ... Call Trace: ... # SCSI device support ... # You can enable one or both FireWire driver stacks. ...
    (Linux-Kernel)
  • Re: Driver getting Page Fault 0Eh Fault=0000, only when 2 PCI cards present in system
    ... There certainly are stack limits that a poorly designed function can exceed, ... > Is there a size issue on the size of a subroutine in a driver? ... With only one of our PCI cards ... >> area in Memory Space and also uses 1 PCI interrupt. ...
    (microsoft.public.development.device.drivers)
  • Re: how to test RAM?
    ... Driver Verifier. ... REPEAT THE LAST TABLE O N L Y) or match the fault address to the driver. ... Memory corruption, other hardware ... A kernel stack overflow. ...
    (microsoft.public.windowsxp.general)
Loading