synclink_cs 2.6.11 diff

Dominik Brodowski linux at dominikbrodowski.net
Sat Mar 19 16:56:34 EST 2005


On Tue, Mar 15, 2005 at 12:29:34PM -0300, Marcelo Tosatti wrote:
> Hi folks,
> 
> While wandering through the 2.6.11 diff I've noticed:
> 
> diff -Nru a/drivers/char/pcmcia/synclink_cs.c b/drivers/char/pcmcia/synclink_cs.c
> --- a/drivers/char/pcmcia/synclink_cs.c 2005-03-01 23:38:55 -08:00
> +++ b/drivers/char/pcmcia/synclink_cs.c 2005-03-01 23:38:55 -08:00
> @@ -592,11 +582,6 @@
>      /* Interrupt setup */
>      link->irq.Attributes = IRQ_TYPE_EXCLUSIVE;
>      link->irq.IRQInfo1   = IRQ_INFO2_VALID | IRQ_LEVEL_ID;
> -    if (irq_list[0] == -1)
> -           link->irq.IRQInfo2 = irq_mask;
> -    else
> -           for (i = 0; i < 4; i++)
> -                   link->irq.IRQInfo2 |= 1 << irq_list[i];
>      link->irq.Handler = NULL;
>                                                                                         
>      link->conf.Attributes = 0;
> @@ -608,7 +593,6 @@
> 
> 
> Isnt the patch supposed to remove "IRQ_INFO2_VALID" from the link->irq.IRQInfo1
> assignment ? 

Yes, it was supposed to do so... Thanks for noticing it. I'll send out a fix
taking care of it soon. Fortunately, nothing breaks because of it...

> On a related note, isnt drivers/pcmcia/cistpl.c parse_irq() function handling
> of IRQ_INFO2_VALID now obsolete? 

Unfortunately, no. It is still used by userspace "parse_cis", and I still
want to maintain backwards compatibility as far as possible until new-style 
PCMCIA support is fully deployable.

Many thanks,
	Dominik



More information about the linux-pcmcia mailing list