I'm writing a driver for DVB-C PCI receiving cards these days and
I found a tricky problem of interrupts. Since kernel-2.6.19 the
interrupt handlers get rid of "struct pt_regs*" argument and I think
it will be easy for me to modify the driver to fit for the recent
kernel.
I simply delete the "struct pt_regs *" argument of my interrupt
handler and compile the driver again. But then I met a rather tricky
problem. I cann't get the PCI intterrupt now! I found that the chip on
the card actually had generate an interrupt and it should not be the
hardware problem.
I'm get seriously confused. My driver works well under kernel-2.6.18
but not generates a single interrupt signal when works above
kernel-2.6.19.
Does anybody meet similar problems?
Sincerely
Yours, Lost Graden.
1,21. 2008
Lost Garden wrote:
> I'm writing a driver for DVB-C PCI receiving cards these days and
> I found a tricky problem of interrupts. Since kernel-2.6.19 the
> interrupt handlers get rid of "struct pt_regs*" argument and I think
> it will be easy for me to modify the driver to fit for the recent
> kernel.
> I simply delete the "struct pt_regs *" argument of my interrupt
> handler and compile the driver again. But then I met a rather tricky
> problem. I cann't get the PCI intterrupt now! I found that the chip on
> the card actually had generate an interrupt and it should not be the
> hardware problem.
> I'm get seriously confused. My driver works well under kernel-2.6.18
> but not generates a single interrupt signal when works above
> kernel-2.6.19.
> Does anybody meet similar problems?
> Sincerely
> Yours, Lost Graden.
> 1,21. 2008
Need more details, like dmesg output. The source code of your driver
would also be useful. One common problem is not calling
pci_enable_device before requesting the interrupt..
--
Robert Hancock Saskatoon, SK, Canada
To email, remove "nospam" from [email protected]
Home Page: http://www.roberthancock.com/