Re: Reattach/redetect allways connected umass device - is it possible ?

From: Bernd Walter <ticso_at_cicely12.cicely.de>
Date: Mon, 28 Mar 2005 15:40:21 +0200
On Mon, Mar 28, 2005 at 03:33:38PM +0200, Poul-Henning Kamp wrote:
> In message <20050328131318.GC14532_at_cicely12.cicely.de>, Bernd Walter writes:
> >On Mon, Mar 28, 2005 at 02:13:16PM +0200, Poul-Henning Kamp wrote:
> >> In message <20050328114633.GZ14532_at_cicely12.cicely.de>, Bernd Walter writes:
> >> 
> >> >> camcontrol detach da0; camcontrol rescan all 
> >> >> helps, but, it should be much better if it will be issued automatically.
> >> >
> >> >Yes - GEOM seems to ignore media change signals from drives.
> >> >I've added PHK to the recipient list - maybe he has an idea about this
> >> >problem.
> >> 
> >> No, GEOM doesn't ignore any such thing, because as far as I know
> >> GEOM doesn't get any such thing to ignore in the first place.
> >
> >At least with (most) SCSI devices you get a media changed exception
> >on the first access after that.
> 
> Yes, but sane people would argue that is one turn too late in the
> game, whouldn' they ?

Ack.

> We should get an async event as soon as the media disappears.

Well - the without a request the drive has no transaction handle to
contact the host.
And we don't always have a physcial SCSI bus - e.g. USB devices
can't contact the host without the host issuing a transfer.
If you want media change events earlier we have to explizitly ask
the devices.

-- 
B.Walter                   BWCT                http://www.bwct.de
bernd_at_bwct.de                                  info_at_bwct.de
Received on Mon Mar 28 2005 - 11:40:45 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:30 UTC