James,
I pushed the latest au1000 code. I had to add a new vec0 handler for the au1000
in head.S, much like the other vec0 ones. I also checked in the vr41xx patch.
Jun indicated that he and/or Ralf reworked some of the 41xx so the 41xx will
probably have to be updated again, but at least the kernel compiles.
To compile the code right at this moment, you need to make this local change in
kernel/Makefile:
[ppopov@zeus linux-dev]$ diff ../stock/kernel/Makefile kernel/Makefile
14c14
< obj-y = sched.o dma.o fork.o exec_domain.o panic.o printk.o \
---
> obj-y = sched.o fork.o exec_domain.o panic.o printk.o \
17a18,21
>
> ifndef CONFIG_MIPS_AU1000
> obj-y += dma.o
> endif
and, this local change in drivers/net/Space.c:
diff ../stock/drivers/net/Space.c drivers/net/Space.c
103a104,105
> extern int gt96100_probe(struct net_device *);
> extern int au1000_probe(struct net_device *dev);
382a385,390
> #endif
> #ifdef CONFIG_MIPS_GT96100ETH
> {gt96100_probe, 0},
> #endif
> #ifdef CONFIG_MIPS_AU1000_ENET
> {au1000_probe, 0},
The Space.c patch really needs to go to Alan. I'll see if he'll take it. The
kernel/Makefile patch probably needs to be reworked so that we don't have to
touch kernel/Makefile.
I tested the kernel and it seems to work fine, so I think I got everything.
However, I haven't tested the fb driver, pcmcia, usb, etc, etc with this kernel.
Also, note that if anyone needs usb on this board or any other au1000 board,
we'll need to send a local usb patch for non-pci devices. Steve already had most
of his work accepted in the usb tree, but the non-pci patch wasn't accepted
because the usb code is going through some reorg right now. So, without this usb
patch which is only in the MontaVista tree at the moment, usb is not usable on
this board.
There are Au1000 customers showing up all over the place, so having the latest
code in sourceforge is a good thing. Now I have to find the time to create a
patch against the oss tree...
Pete
|