Cesar Eduardo Barros wrote:
>
> This patch detects the bogus '255' values (meaning no IRQ/DMA) in the VIA
> parport code.
>
> Not tested or compiled.
Patch looks good to me. Self-contained inside the Via code I wrote,
which is important.
Longer term, I would like to fix up the values and ensure an IRQ and DMA
setting, if such was the intention (as with ECP/EPP).
Jeff
> diff -Naur linux-2.4.0-test8.orig/drivers/parport/parport_pc.c
>linux-2.4.0-test8/drivers/parport/parport_pc.c
> --- linux-2.4.0-test8.orig/drivers/parport/parport_pc.c Sat Sep 9 17:09:42 2000
> +++ linux-2.4.0-test8/drivers/parport/parport_pc.c Sat Sep 9 17:20:37 2000
> @@ -2239,11 +2239,13 @@
> irq = ((irq >> 4) & 0x0F);
>
> /* filter bogus IRQs */
> + /* 255 means NONE, and is bogus as well */
> switch (irq) {
> case 0:
> case 2:
> case 8:
> case 13:
> + case 255:
> irq = PARPORT_IRQ_NONE;
> break;
>
> @@ -2252,7 +2254,9 @@
> }
>
> /* if ECP not enabled, DMA is not enabled, assumed bogus 'dma' value */
> - if (!have_eppecp)
> + /* 255 means NONE. Looks like some BIOS don't set the DMA correctly
> + * even on ECP mode */
> + if (!have_eppecp || dma == 255)
> dma = PARPORT_DMA_NONE;
>
> /* finally, do the probe with values obtained */
>
> --
> Cesar Eduardo Barros
> [EMAIL PROTECTED]
> [EMAIL PROTECTED]
--
Jeff Garzik | Isn't it strange that the same people
Building 1024 | that laugh at gypsy fortune tellers
MandrakeSoft, Inc. | take economists seriously?
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
Please read the FAQ at http://www.tux.org/lkml/
