From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vivek Goyal Subject: Re: [PATCH] Add option to bring up a device in initramfs Date: Tue, 14 Aug 2012 15:36:07 -0400 Message-ID: <20120814193607.GB5760@redhat.com> References: <20120809030848.GA10460@dhcp-16-143.nay.redhat.com> <20120813151029.GE24876@redhat.com> <5029EFCF.4060005@redhat.com> Mime-Version: 1.0 Return-path: Content-Disposition: inline In-Reply-To: <5029EFCF.4060005-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> Sender: initramfs-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Dave Young Cc: chaowang-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org, harald-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org, initramfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org On Tue, Aug 14, 2012 at 02:27:27PM +0800, Dave Young wrote: > On 08/13/2012 11:10 PM, Vivek Goyal wrote: > > > On Thu, Aug 09, 2012 at 11:08:48AM +0800, Dave Young wrote: > >> > >> Kdump support dump to raw device which could be on top of complex storage > >> such as multipath and iscsi which are standalone dracut modules. > >> > >> Add a --device option to dracut which will add the device to host_devs so > >> dracut can add the dependent modules automaticlly and enable them in initramfs. > >> > > > > You have not specified any format for --device argument? What do you expect > > there? Device name, UUID, udev persistent link, or what? > > > It's a device name, should metion it in doc as well. > > > > > Secondly, do we need some extra code to make sure somebody waits for > > this device to come up? Who makes sure that initqueue does not exit > > till this device node comes up. > > > I see there's a wait-mount-dev.sh in 95fstab-sys module which is for > waiting for the mount needed dev. > > Currently dracut maintains host_devs and host_fs_types, host_fs_types > are mainly used inst the fs kernel modules. Other than the filesystem > modules we should use host_devs instead, ie. introduce > for_each_host_devs for modules such as multipath and iscsi using. I see that iscsi is using for_each_host_dev_fs, which has both device and filesystem info. Same is the case with multipath. May be add raw device in host_dev_fs with filesystem type as "raw" and special case "raw" whoever relies on this info. There might be other better ways to handle this though. Thanks Vivek