As far as I know, there's no way to prevent this no-interrupt problem
at the moment. This is seen more frequently but not always when the
frontend is unstable, like when it's raining :-)
Just to make sure if it's of the same symptom, can you compile and run
the following code?
---
#include <linux/module.h>
#include <linux/types.h>
#include <linux/pci.h>
#include <asm/uaccess.h>
#define PCI_VENDOR_ID_B2C2 0x13d0
#define PCI_DEVICE_ID_B2C2 0x2103
#define write_reg_dw(reg,value) writel(value, iomem + reg)
#define read_reg_dw(reg) readl(iomem + reg)
static int b2c2_reg = 0;
static int b2c2_dump(void)
{
struct pci_dev *dev = NULL;
u32 iomem;
dev = pci_find_device(PCI_VENDOR_ID_B2C2, PCI_DEVICE_ID_B2C2, 0);
if (dev == NULL) {
printk("Device not found!\n");
return -EINVAL;
}
iomem = (u32) ioremap(dev->resource[0].start, 0x800);
if (!iomem)
return 0;
printk("Rx[0x%03x] = 0x%08x\n", b2c2_reg, read_reg_dw(b2c2_reg));
iounmap((void *)iomem);
return 0;
}
static int flexcop_main(void)
{
b2c2_dump();
return -EINVAL; /* so we don't need to rmmod */
}
module_init(flexcop_main);
MODULE_PARM(b2c2_reg, "i");
---
On a 2.4 kernel, compile it and run it with
insmod this_module.o b2c2_reg=0x20c.
I use the script like the following
---
if [ $# -eq 1 ]; then
insmod b2c2reg.o b2c2_reg=$1 2>/dev/null
dmesg | tail -n1
else
echo "you need to pass the register offset"
fi
---
If it's of the same symptom, it will probably return 0x0000100. And I
really want b2c2 guys to start looking at this problem or upgrade the
silicon.
Q-ha.
On 5/27/05, Anssi Hannula <anssi.hannula@xxxxxxxxx> wrote:
> Anssi Hannula wrote:
> > It happened that it is always the card which vdr thinks is the second
> > card, i.e. it the IRQ stop is initiated by something VDR does to the
> > second card, probably EPG scan.
> >>> This issue does not exist when using the skystar2 driver, I've used
> >>> that for few days now without problems.
>
> I just managed to stop the IRQ:s of the first card as well with some
> channel swapping (hasn't happened before).
>
> I'll look at the differences of the b2c2 and skystar2 drivers and do
> some testing when I have some time.
>
> --
> Anssi Hannula
>
> _______________________________________________
>
> linux-dvb@xxxxxxxxxxx
> http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
>
