[PATCH] 0/4: Add the ability to splice a list into another list

Heinz Mauelshagen mauelshagen at redhat.com
Tue Jun 6 23:51:13 UTC 2006


On Tue, Jun 06, 2006 at 10:22:52AM -0700, Darrick J. Wong wrote:
> Heinz Mauelshagen wrote:
> 
> > that's what the t_group type is for.
> 
> I'm confused about what you meant by that--in the isw handler, the
> raid_devs created in _create_rd are suppsoed to have
> 
> > Is there a reason, why you couldn't use that as already down
> > for Intel Software RAID (iswc.[ch]) and designed for other flavours
> > such as DDF ?
> 
> Actually, I had no idea that isw was creating extra raid_devs in
> group_rd; from my perspective, I had simply thought that it made sense
> for read_raid_dev to create all the raid_devs that were required and to
> push all of those raid_devs through setup_rd, instead of creating one
> raid_dev per disk in read_raid_dev and adding the rest later.  However,
> it seems a little weird to me that under that scheme one doesn't see all
> the raid_devs until after isw_group is called.

Well, a raid_dev always is a whole single device for all possible mappings.

OTOH there can be multiple raid_sets mapped to a particular raid_dev
(of course to a group of those), which should cover what you want and
is already done in isw.
E.g. 2 mirrors one after the other mapped to 2 raid_devs.

> 
> But, I suppose it wouldn't be too difficult to reroll the patches to do
> it that way too.

That would be great, thanks.

Heinz

> 
> --D
> 

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

Heinz Mauelshagen                                 Red Hat GmbH
Consulting Development Engineer                   Am Sonnenhang 11
Storage Development                               56242 Marienrachdorf
                                                  Germany
Mauelshagen at RedHat.com                            PHONE +49  171 7803392
                                                  FAX   +49 2626 924446
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-




More information about the Ataraid-list mailing list