linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Christian Brauner <brauner@kernel.org>
To: James Bottomley <James.Bottomley@hansenpartnership.com>
Cc: Ard Biesheuvel <ardb@kernel.org>,
	Al Viro <viro@zeniv.linux.org.uk>,
	 linux-fsdevel@vger.kernel.org, torvalds@linux-foundation.org,
	jack@suse.cz, raven@themaw.net,  miklos@szeredi.hu,
	neil@brown.name, a.hindborg@kernel.org, linux-mm@kvack.org,
	 linux-efi@vger.kernel.org, ocfs2-devel@lists.linux.dev,
	kees@kernel.org,  rostedt@goodmis.org,
	gregkh@linuxfoundation.org, linux-usb@vger.kernel.org,
	 paul@paul-moore.com, casey@schaufler-ca.com,
	linuxppc-dev@lists.ozlabs.org,  john.johansen@canonical.com,
	selinux@vger.kernel.org, borntraeger@linux.ibm.com,
	 bpf@vger.kernel.org
Subject: Re: [PATCH v2 22/50] convert efivarfs
Date: Wed, 5 Nov 2025 14:43:34 +0100	[thread overview]
Message-ID: <20251105-vorbild-zutreffen-fe00d1dd98db@brauner> (raw)
In-Reply-To: <20251105-sohlen-fenster-e7c5af1204c4@brauner>

[-- Attachment #1: Type: text/plain, Size: 765 bytes --]

> > > And suspend/resume works just fine with freeze/thaw. See commit
> > > eacfbf74196f ("power: freeze filesystems during suspend/resume")
> > > which implements exactly that.
> > > 
> > > The reason this didn't work for you is very likely:
> > > 
> > > cat /sys/power/freeze_filesystems
> > > 0
> > > 
> > > which you must set to 1.
> > 
> > Actually, no, that's not correct.  The efivarfs freeze/thaw logic must
> > run unconditionally regardless of this setting to fix the systemd bug,
> > so all the variable resyncing is done in the thaw call, which isn't
> > conditioned on the above (or at least it shouldn't be).
> 
> It is conditioned on the above currently but we can certainly fix it
> easily to not be.

Something like the appended patch would do it.

[-- Attachment #2: 0001-power-always-freeze-efivarfs.patch --]
[-- Type: text/x-diff, Size: 4586 bytes --]

From 1f9dc293cebb10b18d9ec8e01b60c014664c98ab Mon Sep 17 00:00:00 2001
From: Christian Brauner <brauner@kernel.org>
Date: Wed, 5 Nov 2025 14:39:45 +0100
Subject: [PATCH] power: always freeze efivarfs

The efivarfs filesystems must always be frozen and thawed to resync
variable state. Make it so.

Signed-off-by: Christian Brauner <brauner@kernel.org>
---
 fs/efivarfs/super.c      |  1 +
 fs/super.c               | 14 +++++++++++---
 include/linux/fs.h       |  3 ++-
 kernel/power/hibernate.c |  9 +++------
 kernel/power/suspend.c   |  3 +--
 5 files changed, 18 insertions(+), 12 deletions(-)

diff --git a/fs/efivarfs/super.c b/fs/efivarfs/super.c
index 1f4d8ce56667..6f9cd18ce6f0 100644
--- a/fs/efivarfs/super.c
+++ b/fs/efivarfs/super.c
@@ -512,6 +512,7 @@ static int efivarfs_init_fs_context(struct fs_context *fc)
 	sfi->mount_opts.gid = GLOBAL_ROOT_GID;
 
 	fc->s_fs_info = sfi;
+	fc->s_iflags |= SB_I_FREEZE_POWER;
 	fc->ops = &efivarfs_context_ops;
 
 	return 0;
diff --git a/fs/super.c b/fs/super.c
index 5bab94fb7e03..017f0be22c61 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -1183,11 +1183,19 @@ static inline bool get_active_super(struct super_block *sb)
 
 static const char *filesystems_freeze_ptr = "filesystems_freeze";
 
-static void filesystems_freeze_callback(struct super_block *sb, void *unused)
+static void filesystems_freeze_callback(struct super_block *sb, void *bool_freeze_all)
 {
+	bool freeze_all = *(bool *)bool_freeze_all;
+
 	if (!sb->s_op->freeze_fs && !sb->s_op->freeze_super)
 		return;
 
+	if (!freeze_all) {
+		if (!(sb->s_iflags & SB_I_FREEZE_POWER))
+			return;
+		pr_info("VFS: Freezing %s filesystems\n", sb->s_type->name);
+	}
+
 	if (!get_active_super(sb))
 		return;
 
@@ -1201,9 +1209,9 @@ static void filesystems_freeze_callback(struct super_block *sb, void *unused)
 	deactivate_super(sb);
 }
 
-void filesystems_freeze(void)
+void filesystems_freeze(bool freeze_all)
 {
-	__iterate_supers(filesystems_freeze_callback, NULL,
+	__iterate_supers(filesystems_freeze_callback, &freeze_all,
 			 SUPER_ITER_UNLOCKED | SUPER_ITER_REVERSE);
 }
 
diff --git a/include/linux/fs.h b/include/linux/fs.h
index c895146c1444..bde4967fdb68 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1419,6 +1419,7 @@ extern int send_sigurg(struct file *file);
 #define SB_I_NOUMASK	0x00001000	/* VFS does not apply umask */
 #define SB_I_NOIDMAP	0x00002000	/* No idmapped mounts on this superblock */
 #define SB_I_ALLOW_HSM	0x00004000	/* Allow HSM events on this superblock */
+#define SB_I_FREEZE_POWER	0x00008000	/* Always freeze on suspend/hibernate */
 
 /* Possible states of 'frozen' field */
 enum {
@@ -3606,7 +3607,7 @@ extern void drop_super_exclusive(struct super_block *sb);
 extern void iterate_supers(void (*f)(struct super_block *, void *), void *arg);
 extern void iterate_supers_type(struct file_system_type *,
 			        void (*)(struct super_block *, void *), void *);
-void filesystems_freeze(void);
+void filesystems_freeze(bool freeze_all);
 void filesystems_thaw(void);
 
 extern int dcache_dir_open(struct inode *, struct file *);
diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index 14e85ff23551..1f250ce036a0 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -825,8 +825,7 @@ int hibernate(void)
 		goto Restore;
 
 	ksys_sync_helper();
-	if (filesystem_freeze_enabled)
-		filesystems_freeze();
+	filesystems_freeze(filesystem_freeze_enabled);
 
 	error = freeze_processes();
 	if (error)
@@ -932,8 +931,7 @@ int hibernate_quiet_exec(int (*func)(void *data), void *data)
 	if (error)
 		goto restore;
 
-	if (filesystem_freeze_enabled)
-		filesystems_freeze();
+	filesystems_freeze(filesystem_freeze_enabled);
 
 	error = freeze_processes();
 	if (error)
@@ -1083,8 +1081,7 @@ static int software_resume(void)
 	if (error)
 		goto Restore;
 
-	if (filesystem_freeze_enabled)
-		filesystems_freeze();
+	filesystems_freeze(filesystem_freeze_enabled);
 
 	pm_pr_dbg("Preparing processes for hibernation restore.\n");
 	error = freeze_processes();
diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c
index 4bb4686c1c08..c933a63a9718 100644
--- a/kernel/power/suspend.c
+++ b/kernel/power/suspend.c
@@ -375,8 +375,7 @@ static int suspend_prepare(suspend_state_t state)
 	if (error)
 		goto Restore;
 
-	if (filesystem_freeze_enabled)
-		filesystems_freeze();
+	filesystems_freeze(filesystem_freeze_enabled);
 	trace_suspend_resume(TPS("freeze_processes"), 0, true);
 	error = suspend_freeze_processes();
 	trace_suspend_resume(TPS("freeze_processes"), 0, false);
-- 
2.47.3


  parent reply	other threads:[~2025-11-05 13:43 UTC|newest]

Thread overview: 101+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-10-28  0:45 [PATCH v2 00/50] tree-in-dcache stuff Al Viro
2025-10-28  0:45 ` [PATCH v2 01/50] fuse_ctl_add_conn(): fix nlink breakage in case of early failure Al Viro
2025-10-28  0:45 ` [PATCH v2 02/50] tracefs: fix a leak in eventfs_create_events_dir() Al Viro
2025-10-28  1:15   ` Steven Rostedt
2025-10-28  0:45 ` [PATCH v2 03/50] new helper: simple_remove_by_name() Al Viro
2025-10-28  0:45 ` [PATCH v2 04/50] new helper: simple_done_creating() Al Viro
2025-10-28  0:45 ` [PATCH v2 05/50] introduce a flag for explicitly marking persistently pinned dentries Al Viro
2025-10-28  0:45 ` [PATCH v2 06/50] primitives for maintaining persisitency Al Viro
2025-10-28 12:38   ` James Bottomley
2025-10-29  5:10     ` Al Viro
2025-10-29 15:25       ` James Bottomley
2025-10-28  0:45 ` [PATCH v2 07/50] convert simple_{link,unlink,rmdir,rename,fill_super}() to new primitives Al Viro
2025-10-29 14:02   ` [External] : " Mark Tinguely
2025-10-29 17:55     ` Al Viro
2025-10-28  0:45 ` [PATCH v2 08/50] convert ramfs and tmpfs Al Viro
2025-10-28  0:45 ` [PATCH v2 09/50] procfs: make /self and /thread_self dentries persistent Al Viro
2025-10-28  0:45 ` [PATCH v2 10/50] configfs, securityfs: kill_litter_super() not needed Al Viro
2025-10-28 23:58   ` Paul Moore
2025-10-29  6:18   ` Andreas Hindborg
2025-10-28  0:45 ` [PATCH v2 11/50] convert xenfs Al Viro
2025-10-28  0:45 ` [PATCH v2 12/50] convert smackfs Al Viro
2025-10-28  0:45 ` [PATCH v2 13/50] convert hugetlbfs Al Viro
2025-10-28  0:45 ` [PATCH v2 14/50] convert mqueue Al Viro
2025-10-28  0:45 ` [PATCH v2 15/50] convert bpf Al Viro
2025-10-28  0:45 ` [PATCH v2 16/50] convert dlmfs Al Viro
2025-10-28  0:45 ` [PATCH v2 17/50] convert fuse_ctl Al Viro
2025-10-28  0:45 ` [PATCH v2 18/50] convert pstore Al Viro
2025-11-04  1:43   ` Kees Cook
2025-10-28  0:45 ` [PATCH v2 19/50] convert tracefs Al Viro
2025-10-28 15:37   ` Steven Rostedt
2025-10-28  0:45 ` [PATCH v2 20/50] convert debugfs Al Viro
2025-10-28  0:45 ` [PATCH v2 21/50] debugfs: remove duplicate checks in callers of start_creating() Al Viro
2025-10-28  0:45 ` [PATCH v2 22/50] convert efivarfs Al Viro
2025-10-28 12:53   ` James Bottomley
2025-10-28 17:45     ` Al Viro
2025-10-28 21:08       ` Al Viro
2025-10-28 21:34         ` Ard Biesheuvel
2025-10-29 18:08           ` Al Viro
2025-10-29 18:26             ` Ard Biesheuvel
2025-10-29 18:57           ` James Bottomley
2025-10-29 19:37             ` Al Viro
2025-10-29 19:48               ` James Bottomley
2025-10-30 13:35               ` Ard Biesheuvel
2025-11-05 11:47                 ` Christian Brauner
2025-11-05 13:09                   ` James Bottomley
2025-11-05 13:16                     ` Christian Brauner
2025-11-05 13:33                       ` James Bottomley
2025-11-05 13:46                         ` Christian Brauner
2025-11-05 14:01                           ` James Bottomley
2025-11-05 15:23                             ` Christian Brauner
2025-11-05 13:43                       ` Christian Brauner [this message]
2025-11-09 20:40                         ` Al Viro
2025-11-11 10:56                           ` Christian Brauner
2025-11-19 21:15                         ` [REGRESSION] " Chris Bainbridge
2025-11-25  9:00                           ` Christian Brauner
2025-11-05 14:34                   ` Ard Biesheuvel
2025-10-28  0:45 ` [PATCH v2 23/50] convert spufs Al Viro
2025-10-28  1:15   ` bot+bpf-ci
2025-10-28  1:33     ` Al Viro
2025-10-28  0:45 ` [PATCH v2 24/50] convert ibmasmfs Al Viro
2025-10-28  0:45 ` [PATCH v2 25/50] ibmasmfs: get rid of ibmasmfs_dir_ops Al Viro
2025-10-28  0:45 ` [PATCH v2 26/50] convert devpts Al Viro
2025-10-28  0:45 ` [PATCH v2 27/50] binderfs: use simple_start_creating() Al Viro
2025-10-28  0:45 ` [PATCH v2 28/50] binderfs_binder_ctl_create(): kill a bogus check Al Viro
2025-10-28  0:45 ` [PATCH v2 29/50] convert binderfs Al Viro
2025-10-28  0:45 ` [PATCH v2 30/50] autofs_{rmdir,unlink}: dentry->d_fsdata->dentry == dentry there Al Viro
2025-10-28  0:45 ` [PATCH v2 31/50] convert autofs Al Viro
2025-10-28  1:55   ` Al Viro
2025-10-28  5:32     ` Linus Torvalds
2025-10-28  0:45 ` [PATCH v2 32/50] convert binfmt_misc Al Viro
2025-10-28  0:45 ` [PATCH v2 33/50] selinuxfs: don't stash the dentry of /policy_capabilities Al Viro
2025-10-29  0:08   ` Paul Moore
2025-10-29 15:19   ` Stephen Smalley
2025-10-28  0:45 ` [PATCH v2 34/50] selinuxfs: new helper for attaching files to tree Al Viro
2025-10-28 23:51   ` Paul Moore
2025-10-29 15:22   ` Stephen Smalley
2025-10-28  0:45 ` [PATCH v2 35/50] convert selinuxfs Al Viro
2025-10-29  0:02   ` Paul Moore
2025-10-29  3:24     ` Al Viro
2025-10-29 14:49       ` Paul Moore
2025-10-29 15:06   ` Stephen Smalley
2025-10-28  0:45 ` [PATCH v2 36/50] functionfs: switch to simple_remove_by_name() Al Viro
2025-10-28  8:47   ` Greg KH
2025-10-28  0:45 ` [PATCH v2 37/50] convert functionfs Al Viro
2025-10-28  0:45 ` [PATCH v2 38/50] gadgetfs: switch to simple_remove_by_name() Al Viro
2025-10-28  0:45 ` [PATCH v2 39/50] convert gadgetfs Al Viro
2025-10-28  0:45 ` [PATCH v2 40/50] hypfs: don't pin dentries twice Al Viro
2025-10-28  0:46 ` [PATCH v2 41/50] hypfs: switch hypfs_create_str() to returning int Al Viro
2025-10-28  0:46 ` [PATCH v2 42/50] hypfs: swich hypfs_create_u64() " Al Viro
2025-10-28  0:46 ` [PATCH v2 43/50] convert hypfs Al Viro
2025-10-28  0:46 ` [PATCH v2 44/50] convert rpc_pipefs Al Viro
2025-10-28  0:46 ` [PATCH v2 45/50] convert nfsctl Al Viro
2025-10-28  0:46 ` [PATCH v2 46/50] convert rust_binderfs Al Viro
2025-10-28  0:46 ` [PATCH v2 47/50] get rid of kill_litter_super() Al Viro
2025-10-28  0:46 ` [PATCH v2 48/50] convert securityfs Al Viro
2025-10-29  0:10   ` Paul Moore
2025-10-28  0:46 ` [PATCH v2 49/50] kill securityfs_recursive_remove() Al Viro
2025-10-29  0:04   ` Paul Moore
2025-10-28  0:46 ` [PATCH v2 50/50] d_make_discardable(): warn if given a non-persistent dentry Al Viro
2025-10-28  0:59 ` [PATCH v2 00/50] tree-in-dcache stuff Al Viro
2025-10-28  5:33 ` Linus Torvalds

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=20251105-vorbild-zutreffen-fe00d1dd98db@brauner \
    --to=brauner@kernel.org \
    --cc=James.Bottomley@hansenpartnership.com \
    --cc=a.hindborg@kernel.org \
    --cc=ardb@kernel.org \
    --cc=borntraeger@linux.ibm.com \
    --cc=bpf@vger.kernel.org \
    --cc=casey@schaufler-ca.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=jack@suse.cz \
    --cc=john.johansen@canonical.com \
    --cc=kees@kernel.org \
    --cc=linux-efi@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-usb@vger.kernel.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=miklos@szeredi.hu \
    --cc=neil@brown.name \
    --cc=ocfs2-devel@lists.linux.dev \
    --cc=paul@paul-moore.com \
    --cc=raven@themaw.net \
    --cc=rostedt@goodmis.org \
    --cc=selinux@vger.kernel.org \
    --cc=torvalds@linux-foundation.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