Mailing List archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[mpeg2] Re: kfir and linux 2.4.X



Ralph Metzler writes:
 > Thanks, that might be the reason why the driver fails on some boards.
 > So, I should insert a pci_set_master() in the initial Kfir config.
 > I guess I was lucky this was always set by the BIOS on the boards I used.
 > 
 > Do you also get the SERR after busmastering is enabled?

Now I've moved the bus master enable to the tail of init_plx9054(),
the SERR disable seems bogus.

 > Or is this error unrelated? Then we would still need to clear the SERR
 > flag.

The reasons I disabled SERR was I was trying to figure out why I was getting
parity errors.. 

kfir: irq 4524 plx f040900 PCIsr 2900117
kfir: irq 4525 plx f048980 PCIsr 2900117
kfir: irq 4526 plx f040980 PCIsr 2900117
kfir: irq 4527 plx f240980 PCIsr 2900117
kfir: irq 4528 plx f040900 PCIsr 2900117
kfir: irq 4529 plx f240980 PCIsr 2900117
kfir: irq 4530 plx f040900 PCIsr 2900117

So, picking this lot apart:

f040900
bit 8: pci interrupt enable
bit 11: local int enable
bit 18: local dma channel 0 enable
bit 24-27: various target abort bits - set to 1 after reset..

f048980
bit 7: pci initiator write/pci target read local data parity check error status
    why is this being set? Hence fool around trying to stop this (eg.
    by disabling SERR :-))
bit 15: local int active

f040980:
     ^ parity error again

f240980
     ^ parity error
bit 21: DMA channel 0 interupt

f040900 etc..

I'm now wondering whether the parity errors are benign. Any idea?

I've been able to play the streams that I've captured, so this
error maybe bogus. 

Andrew





Home | Main Index | Thread Index