Søren Schmidt wrote:
> OK, easy enough, this is what I want to do:
>
> Boot from an ata disk on major# 30, device name "ad", plain and simple.
I'd be inclined to handle this outside the boot code, by treating the
passed in major# as describing the device rather than specifying
the driver.
The point about the boot code is it is deliberately intended to be
usable when completely out of sync with any actual kernel it is
booting. (I expect to be able to use 2.0 bootblocks with 4.0, and
also that loader will be able to boot a 2.0 kernel.)
I assume at some stage that some stage the new driver will take over
completely, and the older driver will disappear. Before that, as
people grow accustomed to thinking "ad" rather than "wd", it will
probably make sense for the boot code to accept (say)
0:ad(0,a)boot/loader
rather than
0:wd(0,a)boot/loader
However, I'd *still* expect it to pass a major# of 0 rather than
30. Why? Because a 2.0 kernel knows only 0. And if a 5.0 kernel
knows only 30, it is -- at least -- in a position to know what
0 meant, and simply substitute one for the other (under the
influence of a kernel configuration option, if necessary).
As an example of the kind of reasoning that goes into the above,
consider the case of booting from CD-ROM. If the boot image used
is that of a floppy, the major# used is 2; if the boot image uses
is that of a hard disk (quite probably a old 20M MFM 615x4x17 hard
disk), the major# used is 0. So in both cases, the major# doesn't
relate to what was booted from, and may not even describe the
underlying technology correctly.
> As the bootcode is now this wont work. If its as simple as me adding
> the pair 30 & "ad" somewhere, I'm satisfied, if not, I'm dissapointed :)
--
Robert Nordier
To Unsubscribe: send mail to majord...@freebsd.org
with "unsubscribe freebsd-current" in the body of the message
