Re: questions about undi patch (fwd)

[Thomas Reifferscheid]

On Thu, 22 Jul 2004, Michael Brown wrote:

[...]

> The patch at http://134.93.168.49/~reiffert/grub-0.95.undi.patch.gz
> ends up calling PXENV_UNDI_ISR with undi_isr.FuncFlag set to 0, which is
> invalid as an entry parameter.  If it works, it's purely by coincidence.
> I suspect that you end up simply discarding packets; this will end up
> being corrected by the higher layer protocols.



2 => 0
1 => 0
2 => 3
3 => 3
3 =>

That is what I can see of undi_isr.FuncFlag
before => after
undi_call

Why does it die only and only after
3 => 3

What should be done in this case?

The patch just does like this:
3 => 3 (*beep* FuncFlag = 0)


> the problem is likely to be that grub is using an altered version of the
> Etherboot UNDI driver code.  If they don't work, they should print out

I'll check this next week, maybe Fengshuo can tell us more
about the altering meanwhile.