linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Keith Owens <kaos@ocs.com.au>
To: linux-kernel@vger.kernel.org
Cc: rgooch@atnf.csiro.au, viro@math.psu.edu
Subject: 2.4.19 breaks devfs mapping for root=
Date: Mon, 21 Oct 2002 21:02:29 +1000	[thread overview]
Message-ID: <565.1035198149@ocs3.intra.ocs.com.au> (raw)

Resend #2 - no response to previous mails.

A change from 2.4.18 to 2.4.19 has broken the way that devfs maps
root=.  2.4.18 init/main.c::mount_root() has

        devfs_make_root (root_device_name);
        handle = devfs_find_handle (NULL, ROOT_DEVICE_NAME,
                                    MAJOR (ROOT_DEV), MINOR (ROOT_DEV),
                                    DEVFS_SPECIAL_BLK, 1);

where ROOT_DEVICE_NAME maps to the value of root= for non-initrd.  This
allowed devfs to remap an entry such as sda3 to whatever driver was
implementing sda3, even if that driver used a different major number.
The correct major was returned in handle.

2.4.19 init/do_mounts.c::mount_root() has

        devfs_make_root(root_device_name);
        create_dev("/dev/root", ROOT_DEV, root_device_name);

create_dev() has

	handle = devfs_find_handle(NULL, dev ? NULL : devfs_name,
                    MAJOR(dev), MINOR(dev), DEVFS_SPECIAL_BLK, 1);

The difference in 2.4.19 is that if dev is already set from
root_dev_names[] then devfs does NOT get the value of root=, forcing
the use of major from root_dev_names[].  If a driver reimplements one
of the standard device names and uses a different major or minor number
then it no longer works in 2.4.19 because devfs is given incomplete
information.

Quick and dirty workaround

--- 2.4.19/init/do_mounts.c
+++ 2.4.19/init/do_mounts.c
@@ -368,7 +368,7 @@
 	if (!do_devfs)
 		return sys_mknod(name, S_IFBLK|0600, kdev_t_to_nr(dev));
 
-	handle = devfs_find_handle(NULL, dev ? NULL : devfs_name,
+	handle = devfs_find_handle(NULL, devfs_name,
 				MAJOR(dev), MINOR(dev), DEVFS_SPECIAL_BLK, 1);
 	if (!handle)
 		return -1;

But that probably breaks initrd.  What should that code be doing to
cope with both initrd and still allow devfs to remap root=?


             reply	other threads:[~2002-10-21 10:56 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-10-21 11:02 Keith Owens [this message]
  -- strict thread matches above, loose matches on Subject: below --
2002-10-16 11:03 2.4.19 breaks devfs mapping for root= Keith Owens
2002-10-14  6:26 Keith Owens

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=565.1035198149@ocs3.intra.ocs.com.au \
    --to=kaos@ocs.com.au \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rgooch@atnf.csiro.au \
    --cc=viro@math.psu.edu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).