[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Fw: Re: USB host driver update



Bjorn,

    You are right, prepare_rx_descriptor() exists if kernel 2.4.18 is 
patched with patch-2.4.19-rc2. Next problem (probably a silly one) is 
to add Axis' patches included in  devboard_lx-R2_1_0.tgz  
(linux-2_4_14.diff) to the new kernel. "install" script fails 
miserably, and if "make kernel" is executed it starts asking for 
Processor family (etrax is not included). Besides, all those Etrax 
patches are not included in the standard kernel distribution since 
they don't appear with "make xconfig".

   Any ideas? Thanks.


> On Tue, 16 Jul 2002, Oscar Gueta wrote:
> > Anyone from axis who can provide the missing routine 
> > ("prepare_rx_descriptor()") for the update sent by Orjan?
> 
> Download the official latest 2.4 kernel (2.4.19-rc1 ?) and you'll 
find it 
> there. In arch/cris/mm/init.c if I recall correctly...
> 
> /Bjorn
> 
> > ------------- Forwarded message follows -------------
> > 
> > Orjan,
> > 
> >    Seems a routine is missing in usb-host.c you sent. Kernel is 
not 
> > compilable with those new files. The error is as follows:
> >         
> > /root/etrax/devboard/axis/devboard_lx/os/linux/arch/cris/lib/lib.a 
> > /root/etrax/devboard/axis/devboard_lx/os/linux/lib/lib.a 
> > /root/etrax/devboard/axis/devboard_lx/os/linux/arch/cris/lib/lib.a 
> > /usr/local/cris/lib/gcc-lib/cris/2.96/linux/libgcc.a \
> >         --end-group \
> >         -o vmlinux
> > arch/cris/drivers/drivers.o: In function `init_rx_buffers':
> > arch/cris/drivers/drivers.o(.text+0x4972): undefined reference to 
> > `prepare_rx_descriptor'
> > arch/cris/drivers/drivers.o: In function `etrax_usb_rx_interrupt':
> > arch/cris/drivers/drivers.o(.text+0x58e0): undefined reference to 
> > `prepare_rx_descriptor'
> > collect2: ld returned 1 exit status
> > make[1]: *** [vmlinux] Error 1
> > 
> > > This is a multi-part message in MIME format.
> > > 
> > > Hi all,
> > > 
> > > Here's an update of the USB host driver.  It's still 
> > work-in-progress, easily
> > > noted by the number of FIXMEs still in there.  Nevertheless, it 
> > might be worth
> > > the effort of trying it.  (Apologies to anyone whose name I 
grabbed 
> > off
> > > dev-etrax in a USB context not interested in this.)
> > > 
> > > Bulk traffic should be working better now.  There is basic 
support 
> > for
> > > isochronous in traffic (for cameras, microphones etc), but no 
> > isochronous out
> > > traffic yet.
> > > 
> > > I recommend you to backup the old usb-host.[hc] files, just in 
case.
> > > 
> > > If you turn on any of the debug-printing macros, I recommend you 
to 
> > disable the
> > > debug port, and read the debug messages from /proc/kmsg instead.
> > > 
> > > I will be gone until the 23nd of July, and won't be able to 
answer 
> > any e-mails
> > > until then.  Still, I'd be interested in hearing about problems 
> > and/or success
> > > stories.
> > > 
> > > Best regards,
> > > Orjan
> > > 
> > > 
> > > (FYI: If any of you plan to use the bluetooth.c driver for 
running 
> > the OpenBT
> > > stack, please apply the following patch which fixes a critical 
> > buffer bug.)
> > > 
> > > Index: bluetooth.c
> > > 
===================================================================
> > > RCS file: /n/cvsroot/os/linux/drivers/usb/bluetooth.c,v
> > > retrieving revision 1.11
> > > diff -u -r1.11 bluetooth.c
> > > --- bluetooth.c 7 Dec 2001 16:54:35 -0000       1.11
> > > +++ bluetooth.c 5 Jul 2002 19:21:09 -0000
> > > @@ -1136,6 +1136,7 @@
> > >  
> > >         endpoint = bulk_out_endpoint[0];
> > >         bluetooth->bulk_out_endpointAddress = 
> > endpoint->bEndpointAddress;
> > > +       bluetooth->bulk_out_buffer_size = 
endpoint->wMaxPacketSize * 
> > 2;
> > >  
> > >         /* create our write urb pool */ 
> > >         for (i = 0; i < NUM_BULK_URBS; ++i) {
> > > @@ -1152,8 +1153,6 @@
> > >                 bluetooth->write_urb_pool[i] = urb;
> > >         }
> > >  
> > > -       bluetooth->bulk_out_buffer_size = 
endpoint->wMaxPacketSize * 
> > 2;
> > > -
> > >         endpoint = interrupt_in_endpoint[0];
> > >         bluetooth->interrupt_in_urb = usb_alloc_urb(0);
> > >         if (!bluetooth->interrupt_in_urb) {
> > > 
> > > 
> > > -- 
> > > Orjan Friberg
> > > Axis Communications AB
> > > 
> > 
> >   
> >  Regards,
> > 
> >   Oscar H. Gueta 
> >   (33) 33453904
> >   Guadalajara, Mex
> > 
> 
> 

 
 Regards,

  Oscar H. Gueta 
  (33) 33453904
  Guadalajara, Mex