[patch] Re: linux-2.4.12 / linux-2.4.13 parallel port problem

[Tim Waugh <twaugh@redhat.com>]

On Fri, Oct 26, 2001 at 12:51:48AM -0700, junio@siamese.dhis.twinsun.com wrote:

> >From the original poster's description, 2.4.10 claimed to have
> detected both address and irq for parport0, while 2.4.12,
> according to the your response, could not tell that IRQ=7.  Do
> you mean that the logic which made 2.4.10 to claime to have
> detected IRQ=7 was faulty and the logic in 2.4.12 is being
> careful not to misdetect?

Oh, I see.  No, this is a regression.  Please try this patch:

--- linux/drivers/parport/ChangeLog.irq	Fri Oct 26 10:11:02 2001
+++ linux/drivers/parport/ChangeLog	Fri Oct 26 10:37:36 2001
@@ -0,0 +1,8 @@
+2001-10-26  Tim Waugh  <twaugh@redhat.com>
+
+	* parport_pc.c (parport_irq_probe): When ECR programmable IRQ
+	support fails, generate interrupts using the FIFO even if we don't
+	want to use the FIFO for real data transfers.
+	(parport_pc_probe_port): Display the ECR address if we have an
+	ECR, not just if we will use the FIFO.
+
--- linux/drivers/parport/parport_pc.c.irq	Fri Oct 26 10:11:06 2001
+++ linux/drivers/parport/parport_pc.c	Fri Oct 26 10:35:49 2001
@@ -2119,10 +2119,9 @@
 
 	if (priv->ecr) {
 		pb->irq = programmable_irq_support(pb);
-	}
 
-	if (pb->modes & PARPORT_MODE_ECP) {
-		pb->irq = irq_probe_ECP(pb);
+		if (pb->irq == PARPORT_IRQ_NONE)
+			pb->irq = irq_probe_ECP(pb);
 	}
 
 	if ((pb->irq == PARPORT_IRQ_NONE) && priv->ecr &&
@@ -2255,7 +2254,7 @@
 	p->private_data = priv;
 
 	printk(KERN_INFO "%s: PC-style at 0x%lx", p->name, p->base);
-	if (p->base_hi && (p->modes & PARPORT_MODE_ECP))
+	if (p->base_hi && priv->ecr)
 		printk(" (0x%lx)", p->base_hi);
 	p->irq = irq;
 	p->dma = dma;

Tim.
*/