I guess the reservation of dma channel 4 for "cascade" is
PC or chipset specific and we don't have such a thing in the
CRIS (ETRAX100LX) chip and channel 4 clashes with external dma0.
Perhaps a better fix is to #ifdef on something else or remove
the cascade stuff entirely from this file, but I leave that
to those who know better.
Have no other arch been bitten by this?
Please apply to both 2.4 and 2.5.
/Johan
diff -u -p -r1.3 dma.c
--- linux/kernel/dma.c 23 Feb 2001 13:50:32 -0000 1.3
+++ linux/kernel/dma.c 4 Mar 2003 18:46:51 -0000
@@ -59,7 +59,11 @@ static struct dma_chan dma_chan_busy[MAX
{ 0, 0 },
{ 0, 0 },
{ 0, 0 },
+#ifndef __CRIS__
{ 1, "cascade" },
+#else
+ { 0, 0 },
+#endif
{ 0, 0 },
{ 0, 0 },
{ 0, 0 }
On Tue, 2003-03-04 at 18:49, Johan Adolfsson wrote:
> I guess the reservation of dma channel 4 for "cascade" is
> PC or chipset specific and we don't have such a thing in the
> CRIS (ETRAX100LX) chip and channel 4 clashes with external dma0.
> Perhaps a better fix is to #ifdef on something else or remove
> the cascade stuff entirely from this file, but I leave that
> to those who know better.
> Have no other arch been bitten by this?
I don't know of any PC cards that can support ISA DMA channel 4 so I
guess simply because of that it hasn't happened. Do you actually
know of any DMA 4 capable ISA devices or is it used for onboard
ISA devices ?
The ifdef is ugly. There should be a nicer way to do this.
> > I guess the reservation of dma channel 4 for "cascade" is
> > PC or chipset specific and we don't have such a thing in the
> > CRIS (ETRAX100LX) chip and channel 4 clashes with external dma0.
> > Perhaps a better fix is to #ifdef on something else or remove
> > the cascade stuff entirely from this file, but I leave that
> > to those who know better.
> > Have no other arch been bitten by this?
>
> I don't know of any PC cards that can support ISA DMA channel 4
As far as I know, there is no pin defined for request or
acknowledgement of DMA channel 4 on the ISA bus - or am I missing
something?
John.