linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Alexander Lobakin <alobakin@pm.me>
To: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Kees Cook <keescook@chromium.org>,
	Sami Tolvanen <samitolvanen@google.com>,
	Miguel Ojeda <ojeda@kernel.org>, Johan Hovold <johan@kernel.org>,
	Bjorn Helgaas <bhelgaas@google.com>,
	Masahiro Yamada <masahiroy@kernel.org>,
	Joe Perches <joe@perches.com>, Alexander Lobakin <alobakin@pm.me>,
	John Wood <john.wood@gmx.com>, Jens Axboe <axboe@kernel.dk>,
	Jan Kara <jack@suse.cz>, Hannes Reinecke <hare@suse.de>,
	Tejun Heo <tj@kernel.org>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH] do_mounts: always prefer tmpfs for rootfs when available
Date: Fri, 02 Jul 2021 23:44:35 +0000	[thread overview]
Message-ID: <20210702233727.21301-1-alobakin@pm.me> (raw)

Inspired by the situation from [0].

The roots of choosing tmpfs/ramfs backend for rootfs go far back
in history, and it's unclear at all why it was decided to select
full-blown tmpfs when "root=" is not specified and feature-poor
ramfs otherwise.
There are several cases when "root=" is not needed at all to work,
and it doesn't break anything or make any [negative] sense. On the
other hand, such separation is rather counter-intuitive and makes
debugging more difficult.
Simply always use tmpfs when it's available -- just like devtmpfs
does [for over a decade].

[0] https://lore.kernel.org/kernel-hardening/20210701234807.50453-1-alobakin@pm.me/

Signed-off-by: Alexander Lobakin <alobakin@pm.me>
---
 fs/namespace.c       |  2 --
 include/linux/init.h |  1 -
 init/do_mounts.c     | 26 +++++++-------------------
 3 files changed, 7 insertions(+), 22 deletions(-)

diff --git a/fs/namespace.c b/fs/namespace.c
index ab4174a3c802..310ab44fdbe7 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -17,7 +17,6 @@
 #include <linux/security.h>
 #include <linux/cred.h>
 #include <linux/idr.h>
-#include <linux/init.h>		/* init_rootfs */
 #include <linux/fs_struct.h>	/* get_fs_root et.al. */
 #include <linux/fsnotify.h>	/* fsnotify_vfsmount_delete */
 #include <linux/file.h>
@@ -4248,7 +4247,6 @@ void __init mnt_init(void)
 	if (!fs_kobj)
 		printk(KERN_WARNING "%s: kobj create error\n", __func__);
 	shmem_init();
-	init_rootfs();
 	init_mount_tree();
 }

diff --git a/include/linux/init.h b/include/linux/init.h
index d82b4b2e1d25..10839922a1d3 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -148,7 +148,6 @@ extern unsigned int reset_devices;
 /* used by init/main.c */
 void setup_arch(char **);
 void prepare_namespace(void);
-void __init init_rootfs(void);
 extern struct file_system_type rootfs_fs_type;

 #if defined(CONFIG_STRICT_KERNEL_RWX) || defined(CONFIG_STRICT_MODULE_RWX)
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 74aede860de7..c00b05015a66 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -611,24 +611,12 @@ void __init prepare_namespace(void)
 	init_chroot(".");
 }

-static bool is_tmpfs;
-static int rootfs_init_fs_context(struct fs_context *fc)
-{
-	if (IS_ENABLED(CONFIG_TMPFS) && is_tmpfs)
-		return shmem_init_fs_context(fc);
-
-	return ramfs_init_fs_context(fc);
-}
-
 struct file_system_type rootfs_fs_type = {
-	.name		= "rootfs",
-	.init_fs_context = rootfs_init_fs_context,
-	.kill_sb	= kill_litter_super,
+	.name			= "rootfs",
+#ifdef CONFIG_TMPFS
+	.init_fs_context	= shmem_init_fs_context,
+#else
+	.init_fs_context	= ramfs_init_fs_context,
+#endif
+	.kill_sb		= kill_litter_super,
 };
-
-void __init init_rootfs(void)
-{
-	if (IS_ENABLED(CONFIG_TMPFS) && !saved_root_name[0] &&
-		(!root_fs_names || strstr(root_fs_names, "tmpfs")))
-		is_tmpfs = true;
-}
--
2.32.0



             reply	other threads:[~2021-07-02 23:44 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-02 23:44 Alexander Lobakin [this message]
2021-08-04 21:14 ` [PATCH] do_mounts: always prefer tmpfs for rootfs when available Alexander Lobakin

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=20210702233727.21301-1-alobakin@pm.me \
    --to=alobakin@pm.me \
    --cc=akpm@linux-foundation.org \
    --cc=axboe@kernel.dk \
    --cc=bhelgaas@google.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=hare@suse.de \
    --cc=jack@suse.cz \
    --cc=joe@perches.com \
    --cc=johan@kernel.org \
    --cc=john.wood@gmx.com \
    --cc=keescook@chromium.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=masahiroy@kernel.org \
    --cc=ojeda@kernel.org \
    --cc=samitolvanen@google.com \
    --cc=tj@kernel.org \
    --cc=viro@zeniv.linux.org.uk \
    /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).