All of lore.kernel.org
 help / color / mirror / Atom feed
From: Vivek Goyal <vgoyal@redhat.com>
To: Christoph Hellwig <hch@lst.de>
Cc: viro@zeniv.linux.org.uk, linux-fsdevel@vger.kernel.org,
	linux-kernel@vger.kernel.org, virtio-fs@redhat.com
Subject: Re: [PATCH 2/2] init: allow mounting arbitrary non-blockdevice filesystems as root
Date: Thu, 17 Jun 2021 12:26:10 -0400	[thread overview]
Message-ID: <20210617162610.GC1142820@redhat.com> (raw)
In-Reply-To: <20210617153649.1886693-3-hch@lst.de>

On Thu, Jun 17, 2021 at 05:36:49PM +0200, Christoph Hellwig wrote:

[..]
> +static int __init try_mount_nodev(char *fstype)
> +{
> +	struct file_system_type *fs = get_fs_type(fstype);
> +	int err = -EINVAL;
> +
> +	if (!fs)
> +		return -EINVAL;
> +	if (!(fs->fs_flags & (FS_REQUIRES_DEV | FS_BINARY_MOUNTDATA)))

Not sure what FS_BINARY_MOUNTDATA is why fs should not have that set. nfs
seems to set it too. So that means they can't use try_mount_nodev().

> +		err = do_mount_root(root_device_name, fstype, root_mountflags,
> +					root_mount_data);
> +	put_filesystem(fs);
> +
> +	if (err != -EACCES && err != -EINVAL)

In case of success err == 0, but we still panic(). We will need to
check for success as well.
	if (err && err != -EACCES && err != -EINVAL)

> +		panic("VFS: Unable to mount root \"%s\" (%s), err=%d\n",
> +			      root_device_name, fstype, err);
> +	return err;
> +}
> +
> +static int __init mount_nodev_root(void)
> +{
> +	char *fs_names, *p;
> +	int err = -EINVAL;
> +
> +	fs_names = (void *)__get_free_page(GFP_KERNEL);
> +	if (!fs_names)
> +		return -EINVAL;
> +	split_fs_names(fs_names, root_fs_names);

root_fs_names can be NULL and it crashes with NULL pointer dereference.

Vivek

> +
> +	for (p = fs_names; *p; p += strlen(p) + 1) {
> +		err = try_mount_nodev(p);
> +		if (!err)
> +			break;
> +	}
> +
> +	free_page((unsigned long)fs_names);
> +	return err;
> +}
> +
>  void __init mount_root(void)
>  {
>  #ifdef CONFIG_ROOT_NFS
> @@ -550,6 +588,8 @@ void __init mount_root(void)
>  		return;
>  	}
>  #endif
> +	if (ROOT_DEV == 0 && mount_nodev_root() == 0)
> +		return;
>  #ifdef CONFIG_BLOCK
>  	{
>  		int err = create_dev("/dev/root", ROOT_DEV);
> -- 
> 2.30.2
> 


WARNING: multiple messages have this Message-ID (diff)
From: Vivek Goyal <vgoyal@redhat.com>
To: Christoph Hellwig <hch@lst.de>
Cc: linux-fsdevel@vger.kernel.org, virtio-fs@redhat.com,
	viro@zeniv.linux.org.uk, linux-kernel@vger.kernel.org
Subject: Re: [Virtio-fs] [PATCH 2/2] init: allow mounting arbitrary non-blockdevice filesystems as root
Date: Thu, 17 Jun 2021 12:26:10 -0400	[thread overview]
Message-ID: <20210617162610.GC1142820@redhat.com> (raw)
In-Reply-To: <20210617153649.1886693-3-hch@lst.de>

On Thu, Jun 17, 2021 at 05:36:49PM +0200, Christoph Hellwig wrote:

[..]
> +static int __init try_mount_nodev(char *fstype)
> +{
> +	struct file_system_type *fs = get_fs_type(fstype);
> +	int err = -EINVAL;
> +
> +	if (!fs)
> +		return -EINVAL;
> +	if (!(fs->fs_flags & (FS_REQUIRES_DEV | FS_BINARY_MOUNTDATA)))

Not sure what FS_BINARY_MOUNTDATA is why fs should not have that set. nfs
seems to set it too. So that means they can't use try_mount_nodev().

> +		err = do_mount_root(root_device_name, fstype, root_mountflags,
> +					root_mount_data);
> +	put_filesystem(fs);
> +
> +	if (err != -EACCES && err != -EINVAL)

In case of success err == 0, but we still panic(). We will need to
check for success as well.
	if (err && err != -EACCES && err != -EINVAL)

> +		panic("VFS: Unable to mount root \"%s\" (%s), err=%d\n",
> +			      root_device_name, fstype, err);
> +	return err;
> +}
> +
> +static int __init mount_nodev_root(void)
> +{
> +	char *fs_names, *p;
> +	int err = -EINVAL;
> +
> +	fs_names = (void *)__get_free_page(GFP_KERNEL);
> +	if (!fs_names)
> +		return -EINVAL;
> +	split_fs_names(fs_names, root_fs_names);

root_fs_names can be NULL and it crashes with NULL pointer dereference.

Vivek

> +
> +	for (p = fs_names; *p; p += strlen(p) + 1) {
> +		err = try_mount_nodev(p);
> +		if (!err)
> +			break;
> +	}
> +
> +	free_page((unsigned long)fs_names);
> +	return err;
> +}
> +
>  void __init mount_root(void)
>  {
>  #ifdef CONFIG_ROOT_NFS
> @@ -550,6 +588,8 @@ void __init mount_root(void)
>  		return;
>  	}
>  #endif
> +	if (ROOT_DEV == 0 && mount_nodev_root() == 0)
> +		return;
>  #ifdef CONFIG_BLOCK
>  	{
>  		int err = create_dev("/dev/root", ROOT_DEV);
> -- 
> 2.30.2
> 


  reply	other threads:[~2021-06-17 16:26 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-17 15:36 support booting of arbitrary non-blockdevice file systems Christoph Hellwig
2021-06-17 15:36 ` [Virtio-fs] " Christoph Hellwig
2021-06-17 15:36 ` [PATCH 1/2] init: split get_fs_names Christoph Hellwig
2021-06-17 15:36   ` [Virtio-fs] " Christoph Hellwig
2021-06-17 15:36 ` [PATCH 2/2] init: allow mounting arbitrary non-blockdevice filesystems as root Christoph Hellwig
2021-06-17 15:36   ` [Virtio-fs] " Christoph Hellwig
2021-06-17 16:26   ` Vivek Goyal [this message]
2021-06-17 16:26     ` Vivek Goyal
2021-06-18 13:20     ` Christoph Hellwig
2021-06-18 13:20       ` [Virtio-fs] " Christoph Hellwig
2021-06-18 14:10       ` Vivek Goyal
2021-06-18 14:10         ` [Virtio-fs] " Vivek Goyal
2021-06-18 18:03 ` [Virtio-fs] support booting of arbitrary non-blockdevice file systems Stefan Hajnoczi
2021-06-18 18:03   ` Stefan Hajnoczi
2021-06-21  6:26 support booting of arbitrary non-blockdevice file systems v2 Christoph Hellwig
2021-06-21  6:26 ` [PATCH 2/2] init: allow mounting arbitrary non-blockdevice filesystems as root Christoph Hellwig

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=20210617162610.GC1142820@redhat.com \
    --to=vgoyal@redhat.com \
    --cc=hch@lst.de \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=viro@zeniv.linux.org.uk \
    --cc=virtio-fs@redhat.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.