hverkuil at xs4all
Aug 27, 2008, 11:56 PM
Post #2 of 2
(1608 views)
Permalink
|
|
Re: cx18: please confirm bogus PCI setup lines....
[In reply to]
|
|
On Thursday 28 August 2008 04:45:56 Andy Walls wrote:
> Hans,
>
> These look like bogus holdovers from ivtv; can you please confirm?
>
> 1. In cx18-driver.h:cx18_setup_pci():
>
> cmd |= PCI_COMMAND_IO | PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER;
> ^^^^^^^^^^^^^^
> I'm almost sure the CX23418 doesn't have any I/O space ports, and
> trying to turn them on doesn't appear to have any affect according to
> lspci. Bogus, right?
This code comes from the Conexant sources. I agree that it doesn't seem
to make any sense.
> 2. Also in that same function:
>
> /* This config space value relates to DMA latencies. The
> default value 0x8080 is too low however and will lead
> to DMA errors. 0xffff is the max value which solves
> these problems. */
> pci_write_config_dword(dev, 0x40, 0xffff);
>
> At boot up, the CX23418 (on the Raptor card) defaults to
>
> 40: 30 10 10 01 (or something like that, I can't remember)
>
> After the module loads it looks similar to this:
>
> 40: 7b 1f 00 01 (I can't remember exactly)
>
> It's obvious the value 0x0000ffff didn't take. So that line of code
> is bogus too, right?
Oops, that's from ivtv. Vitally important there but not applicable to
cx18. Definitely should be removed.
Regards,
Hans
_______________________________________________
ivtv-devel mailing list
ivtv-devel [at] ivtvdriver
http://ivtvdriver.org/mailman/listinfo/ivtv-devel
|
