Messages in this thread | | | From | Russell King <> | Subject | Re: IDE driver porting problems | Date | Sun, 6 Aug 2000 09:00:27 +0100 (BST) |
| |
Andre Hedrick writes: > On Mon, 24 Jul 2000, Russell King wrote: > So Russell, you want a no assumption call?
Ideally yes, since writing to the drive/head register and then reading it back and checking that it is actually what we wrote will fail if the disk is not spun up.
However, there is a problem - bad interfaces which don't have a pull-down resistor on IDE dataline 7 (as specified in the ATA specs). In this case, we may end up having to wait around 30 seconds or more before deciding that there isn't actually a drive present. (Note that some drives seem to take 45 seconds - you'll be glad to know that gb has one). Note that the timeout should be timed from the point when Linux starts, not the point at which you start probing for the drive.
Maybe it ought to be a configuration option, so that architectures which can guarantee that the drive is spun up (and therefore not busy, eg, x86) miss out this check? (or maybe an config option which is derived from the architecture).
Whatever it is, if it appears in 2.4, then it should start of benign, and the various architecture maintainers should be given the power to turn it on (the users would complain at the 45 second timeout). Otherwise, it should probably be a 2.5 thing. _____ |_____| ------------------------------------------------- ---+---+- | | Russell King rmk@arm.linux.org.uk --- --- | | | | http://www.arm.linux.org.uk/personal/aboutme.html / / | | +-+-+ --- -+- / | THE developer of ARM Linux |+| /|\ / | | | --- | +-+-+ ------------------------------------------------- /\\\ |
- 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/
| |