Messages in this thread | | | Date | Tue, 7 Mar 2000 14:15:27 +0100 | From | almesber@lrc ... | Subject | Re: DMA mapping |
| |
David S. Miller wrote: > The bus_to_virt() interface is just for lazy driver authors :-)
As long as it stays around, I'm happy ;-)
I depend on it in the ZeitNet driver. That card takes a lot of pre-allocated buffers, fills them, and returns them in more or less arbitrary order. What I get from the card is a pointer to the buffer (in card address space).
Now I could of course introduce hashes and caches to implement a fast lookup that makes this work with the new scheme, but the problem is that the card is pretty obsolete (original company (ZeitNet) got bought, new company (Cabletron) killed the product, and it only works in Pentium 1 or older, so the assumptions made by bus_to_virt are intact anyway), so anything beyond trivial maintenance is likely to cause more damage then improvement.
Maybe a case for #define OBSOLETE_HARDWARE #include <linux/ugly-old-stuff.h> ;-)
- Werner
-- _________________________________________________________________________ / Werner Almesberger, ICA, EPFL, CH werner.almesberger@ica.epfl.ch / /_IN_N_032__Tel_+41_21_693_6621__Fax_+41_21_693_6610_____________________/
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |