Hello!
I'm using Linux 2.4.19rc5aa1 and having problems to compile Digi's cpci
driver for multiport serialboard. This driver is not part of the kernel,
but available from Digi's website.
The problem seems to be related to changes in include/linux/sched.h.
Digi's driver wants to use member called "counter" which doesn't exist
anymore in the -aa kernel.
The code goes like this (cpci.c line 3847):
current->state = TASK_INTERRUPTIBLE;
current->counter = 0; /* make us low-priority */
current is task_struct.
Sources available from http://nrg.joroinen.fi/digi/
Any ideas how to fix this? I'm not familiar with kernel driver
development..
- Pasi K?rkk?inen
^
. .
Linux
/ - \
Choice.of.the
.Next.Generation.
On Thu, 2002-09-12 at 17:30, Pasi K?rkk?inen wrote
> The code goes like this (cpci.c line 3847):
>
> current->state = TASK_INTERRUPTIBLE;
> current->counter = 0; /* make us low-priority */
>
> current is task_struct.
Its assuming old old scheduler bits. I suspect just removing the
current->counter junk will make it happy
On 12 Sep 2002, Alan Cox wrote:
> On Thu, 2002-09-12 at 17:30, Pasi K?rkk?inen wrote
> > The code goes like this (cpci.c line 3847):
> >
> > current->state = TASK_INTERRUPTIBLE;
> > current->counter = 0; /* make us low-priority */
> >
> > current is task_struct.
>
> Its assuming old old scheduler bits. I suspect just removing the
> current->counter junk will make it happy
>
I removed that line, and then it compiled OK and it seems to work too.
Digi International ClassicBoard PCI and Neo Driver v1.1.16 Loaded
Copyright (C) 1999-2001 Digi International, Inc.
PCI Digi Neo 8 Port (Card #0;IRQ17) Found
ttyCP0 at 0xd084a000 (irq = 17) is a XR17C158
ttyCP1 at 0xd084a200 (irq = 17) is a XR17C158
ttyCP2 at 0xd084a400 (irq = 17) is a XR17C158
ttyCP3 at 0xd084a600 (irq = 17) is a XR17C158
ttyCP4 at 0xd084a800 (irq = 17) is a XR17C158
ttyCP5 at 0xd084aa00 (irq = 17) is a XR17C158
ttyCP6 at 0xd084ac00 (irq = 17) is a XR17C158
ttyCP7 at 0xd084ae00 (irq = 17) is a XR17C158
Thank you!
- Pasi K?rkk?inen
ps. If somebody from Digi is reading this, why don't you send these
drivers for inclusion to the kernel?
^
. .
Linux
/ - \
Choice.of.the
.Next.Generation.