Messages in this thread | | | Date | Fri, 07 Jul 2006 13:39:51 -0400 | From | Jeff Garzik <> | Subject | Re: 2.6.17-mm6 |
| |
Alan Cox wrote: > Ar Gwe, 2006-07-07 am 13:14 -0400, ysgrifennodd Jeff Garzik: >> Most older controllers always fall into pata_, most newer into sata_, >> and an odd few ata_ >> >> Its a bug if you don't help maintain these assumptions :) > > It would be very hard to do so. Almost anything can and (alas) did have > sata bridges nailed to it early on. Almost every later highpoint, > promise and ati chip has been found with SATA bridges attached. > > I've tried to follow the convention on the basis of "not usually found > nailed to a SATA bridge chip". If we want to be strict then most of > pata_ is ata_ and the prefix really isn't useful.
If there is a SATA receptable -- for plugging in SATA cables -- hardwired onto the controller, it should not be called pata_
Its about what the _user_ knows, and sees.
The user doesn't know about soldered bridge chips. The user knows if he is plugging a PATA or SATA cable into his controller.
> And what about IDE/SATA convertor boards ?
Covered, in the "for the purposes of" sentence.
Jeff
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |