On Fri, 9 Feb 1996, Robert Black wrote:

[animated discussion about bounce-buffer problem deleted]
> > The hardware is broken, not the OS.

I agree. But I also agree that we need to deal with the hardware
users have already, not just what is the best or where the market is going.

> This suggests that the fix should be in the i386 port machine-dependant code.
> Presumably it is felt that the bounce-buffers would have to be implementented
> in machine-independant code otherwise I can't see any reason not to do them. I
> would have thought it would be easy enough for the i386 port to reserve a block
> of physical memory at boot time which is then used for a DMA bounce buffer. The
> only things which would need changing this way are specific to the port or
> drivers. In other words it isn't a job for the core, but for the i386 port
> maintainer. Am I missing something?

As I see it, yes and no. The problem is definitely (as I understand it, 
please correct me if I'm wrong) isa-bus specific. The reason it's not 
machine-dependant is that ports other than just i386 have isa busses. I 
believe the alpha port does too. The isa code certainly is considered 
machine-independant; it's in /sys/dev/isa, not /sys/arch/i386/dev/isa.

With the release of the CHRP standard, the PowerPC clone standard Apple, 
Motorola, and IBM have agreed on, we Mac users have learned we soon will 
get the joys of isa hardware. A pci-to-isa bridge is part of the standard 
(including other fun things like SVGA, parallel printer ports, and 
SoundBlaster audio on the isa side).

So we (future) CHRP users will be interested in isa support too, and 
these problems, and thus it's a bit bigger than machine-dep. Though I do 
think it's probably just the isa code, not the whole OS, which is ill.

Take care,

Bill