DragonFly kernel List (threaded) for 2004-11
Re: RFC bus_dma_tag_parent
On Fri, Nov 12, 2004 at 11:04:27AM -0800, Matthew Dillon wrote:
> :This is useful if the bus DMA engine supports scatter/gather itself.
> :This can be done IIRC on the Alpha, maybe on other architectures too.
> :I don't mind if we have a UNKNOWN value, e.g. 0. This is not a problem
> :here, but for many drivers we know this in advantage and it might be
> :used to allocate the segment maps in the parent tag.
> Yes, that makes sense. I think the best approach to the allocation issue
> is to solve the real problem, and that is to allow a dma tag to be
> flagged as being just a topological specification, and panicing if
> someone actually tries to initiate DMA using the topological
> specification. I would rather not delay the allocation of the
> scatter/gather array to hack around the lack of knowledge.
Even better, we just have to fail any attempt to load or allocate DMA
mappings, perhaps with a KASSERT for the debugging case :)
> Matthew Dillon