Re: [linux-dvb] CI/CAM fixes for knc1 dvb-s cards

[Christoph Pfister <christophpfister@gmail.com>]

Am Samstag 22 März 2008 schrieb e9hack:
> Christoph Pfister schrieb:
> > Hi,
> >
> > Can somebody please pick up those patches (descriptions inlined)?
> >
> > Thanks,
> >
> > Christoph
>
> diff -r 1886a5ea2f84 -r f252381440c1
> linux/drivers/media/dvb/ttpci/budget-av.c ---
> a/linux/drivers/media/dvb/ttpci/budget-av.c	Fri Mar 21 08:04:55 2008 -0300
> +++ b/linux/drivers/media/dvb/ttpci/budget-av.c	Fri Mar 21 19:29:15 2008
> +0100 @@ -178,7 +178,7 @@ static int ciintf_read_cam_control(struc
>   	udelay(1);
>
>   	result = ttpci_budget_debiread(&budget_av->budget, DEBICICAM, address &
> 3, 1, 0, 0); -	if ((result == -ETIMEDOUT) || ((result == 0xff) && ((address
> & 3) < 2))) { +	if ((result == -ETIMEDOUT) || ((result == 0xff) &&
> ((address & 3) == 1))) { ciintf_slot_shutdown(ca, slot);
>   		printk(KERN_INFO "budget-av: cam ejected 3\n");
>   		return -ETIMEDOUT;
>
>
> IMHO you should remove the test for 0xff . Without your patch, it wasn't
> possible to read 0xff from address 0 and 1. Now it isn't possible to read
> 0xff from address 1.

Address 1 is the status register; bits 2-5 are reserved according to en50221 
and should be zero, so this case is less problematic with regards to 0xff 
checking.

On second thoughts it's probably better to remove the check altogether, 
because a) budget-av isn't here to check standards conformance - the higher 
layers know better how to deal with the content and b) who should care if the 
other status bits work correctly ;)

> I've described this problem some time ago:
> http://linuxtv.org/pipermail/linux-dvb/2007-July/019436.html

Argh, would have saved some work for me ...

> -Hartmut

Christoph

_______________________________________________
linux-dvb mailing list
linux-dvb@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb