Subject: RE: Calling a function in pciide() from user space?
To: 'Jon Lindgren' <jlindgren@slk.com>
From: Bruce Martin <brucem@cat.co.za>
List: tech-kern
Date: 04/11/2001 15:36:22
Thanks, I didn't even know atactl() existed, so it was interesting to take a
look at it. However, it leaves me with the same problem: it needs a device
name (like /dev/wd0d), and so, although something like "atactl /dev/wd1d
detach" makes sense, something like "atactl /dev/wd1d reattach" does not
make sense, as wd1 does not exist after the detach. I need some way to call
in on the pciide level, or at least the wdc level, to say "detach drive x on
channel y", or "check your channels for new drives"...

Thanks
 Bruce

-----Original Message-----
From: Jon Lindgren [mailto:jlindgren@slk.com]
Sent: 11 April 2001 02:13
To: Bruce Martin
Cc: tech-kern@netbsd.org
Subject: Re: Calling a function in pciide() from user space?


On Wed, 11 Apr 2001, Bruce Martin wrote:

> Hi All
>
> We have an application that allows us to Hot Swap PCI IDE drives, at the
> hardware level. I am currently working on the NetBSD kernel and user
> software to enable this (similar to the PCMCIA detach and re-attach). I
have
> written two functions that do this - they call modified versions of
> wdcdetach() and wdcattach(). However, the question is, where to call these
> from? The logical place to me is pciide() - the controller calls the
> channel, which calls the disk to be detached/reattached.
>
> However, I need to trigger the detach and reattach manually from user
> space - does anybody have any suggestions of a good way to do this?
ioctls()
> etc?

A shot in the dark, but perhaps atactl(8) might be a starting point?

-
Jon
 --------------------------------------------------------------------
 - The opinions expressed are not necesarily those of my employer.
   "I wonder how many people actually read my .sig?"