Device layering work patch #3
Matthew Dillon
dillon at apollo.backplane.com
Mon May 17 09:44:32 PDT 2004
:On Sun, May 16, 2004 at 04:52:03PM -0700, Matthew Dillon wrote:
:> Sorry about that, patch #2 doesn't apply cleanly due to other recent
:> changes (e.g. to bktr, and buildworld had a fault due to vinum's attempt
:> to overload userland and kerneland definitions). I have generated a
:> new patch, #3, which does apply cleanly.
:
:Are you sure about the patch to bktr_os.c? Doesn't the cdevsw_remove
:destroy all devices, even if there are multiple bktr(4) devices in the
:tree? I haven't check the rest yet.
:
:Joerg
I'm definitely not sure, which is why it's still a patch set and not
yet committed.
I think you have hit upon an interesting problem. We probably need to
specify a unit argument or some such to cdevsw_remove(), or perhaps
even go as far as to specify a minor device number mask and match
value since 'unit' number formatting is not really formalized (other
then by the DISK subsystem). For that matter, we might have to specify
some sort of ranging mask for cdevsw_add(). I'll have to think about
it.
-Matt
Matthew Dillon
<dillon at xxxxxxxxxxxxx>
More information about the Kernel
mailing list