[syzbot] [fs?] KASAN: slab-use-after-free Read in clone_mnt

[syzbot] [fs?] KASAN: slab-use-after-free Read in clone_mnt

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit:    c22e26bd0906 Merge tag 'landlock-7.0-rc1' of git://git.ker..
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=12c6a6e6580000
kernel config:  https://syzkaller.appspot.com/x/.config?x=6428d17febdfb14e
dashboard link: https://syzkaller.appspot.com/bug?extid=a89f9434fb5a001ccd58
compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8

Unfortunately, I don't have any reproducer for this issue yet.

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/b33c549157ca/disk-c22e26bd.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/34c7ded19553/vmlinux-c22e26bd.xz
kernel image: https://storage.googleapis.com/syzbot-assets/66faec2158ed/bzImage-c22e26bd.xz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+a89f9434fb5a001ccd58@syzkaller.appspotmail.com

==================================================================
BUG: KASAN: slab-use-after-free in __list_add_valid_or_report+0x4e/0x130 lib/list_debug.c:29
Read of size 8 at addr ffff8880341bcb40 by task syz.6.5169/22925

CPU: 1 UID: 0 PID: 22925 Comm: syz.6.5169 Tainted: G             L      syzkaller #0 PREEMPT_{RT,(full)} 
Tainted: [L]=SOFTLOCKUP
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/24/2026
Call Trace:
 <TASK>
 dump_stack_lvl+0xe8/0x150 lib/dump_stack.c:120
 print_address_description mm/kasan/report.c:378 [inline]
 print_report+0xba/0x230 mm/kasan/report.c:482
 kasan_report+0x117/0x150 mm/kasan/report.c:595
 __list_add_valid_or_report+0x4e/0x130 lib/list_debug.c:29
 __list_add_valid include/linux/list.h:96 [inline]
 __list_add include/linux/list.h:158 [inline]
 list_add include/linux/list.h:177 [inline]
 clone_mnt+0x447/0x9a0 fs/namespace.c:1275
 copy_tree+0xde/0x930 fs/namespace.c:2159
 copy_mnt_ns+0x24d/0x990 fs/namespace.c:4246
 create_new_namespaces+0xcf/0x6a0 kernel/nsproxy.c:98
 unshare_nsproxy_namespaces+0x11a/0x160 kernel/nsproxy.c:226
 ksys_unshare+0x4f4/0x900 kernel/fork.c:3174
 __do_sys_unshare kernel/fork.c:3245 [inline]
 __se_sys_unshare kernel/fork.c:3243 [inline]
 __x64_sys_unshare+0x38/0x50 kernel/fork.c:3243
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x14d/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7ff48d79bf79
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 e8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ff48b9f6028 EFLAGS: 00000246 ORIG_RAX: 0000000000000110
RAX: ffffffffffffffda RBX: 00007ff48da15fa0 RCX: 00007ff48d79bf79
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 000000002a020400
RBP: 00007ff48d8327e0 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 00007ff48da16038 R14: 00007ff48da15fa0 R15: 00007ffc92e1ac28
 </TASK>

Allocated by task 22855:
 kasan_save_stack mm/kasan/common.c:57 [inline]
 kasan_save_track+0x3e/0x80 mm/kasan/common.c:78
 unpoison_slab_object mm/kasan/common.c:340 [inline]
 __kasan_slab_alloc+0x6c/0x80 mm/kasan/common.c:366
 kasan_slab_alloc include/linux/kasan.h:253 [inline]
 slab_post_alloc_hook mm/slub.c:4459 [inline]
 slab_alloc_node mm/slub.c:4788 [inline]
 kmem_cache_alloc_noprof+0x33b/0x680 mm/slub.c:4795
 alloc_vfsmnt+0x23/0x420 fs/namespace.c:287
 clone_mnt+0x4b/0x9a0 fs/namespace.c:1246
 create_new_namespace fs/namespace.c:3098 [inline]
 open_new_namespace fs/namespace.c:3168 [inline]
 vfs_open_tree+0x507/0x1040 fs/namespace.c:3217
 __do_sys_open_tree fs/namespace.c:3227 [inline]
 __se_sys_open_tree fs/namespace.c:3225 [inline]
 __x64_sys_open_tree+0x96/0x110 fs/namespace.c:3225
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x14d/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Freed by task 20:
 kasan_save_stack mm/kasan/common.c:57 [inline]
 kasan_save_track+0x3e/0x80 mm/kasan/common.c:78
 kasan_save_free_info+0x46/0x50 mm/kasan/generic.c:584
 poison_slab_object mm/kasan/common.c:253 [inline]
 __kasan_slab_free+0x5c/0x80 mm/kasan/common.c:285
 kasan_slab_free include/linux/kasan.h:235 [inline]
 slab_free_hook mm/slub.c:2670 [inline]
 slab_free mm/slub.c:6082 [inline]
 kmem_cache_free+0x185/0x690 mm/slub.c:6212
 rcu_do_batch kernel/rcu/tree.c:2617 [inline]
 rcu_core kernel/rcu/tree.c:2869 [inline]
 rcu_cpu_kthread+0x99e/0x1470 kernel/rcu/tree.c:2957
 smpboot_thread_fn+0x541/0xa50 kernel/smpboot.c:160
 kthread+0x388/0x470 kernel/kthread.c:467
 ret_from_fork+0x51e/0xb90 arch/x86/kernel/process.c:158
 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245

Last potentially related work creation:
 kasan_save_stack+0x3e/0x60 mm/kasan/common.c:57
 kasan_record_aux_stack+0xbd/0xd0 mm/kasan/generic.c:556
 __call_rcu_common kernel/rcu/tree.c:3131 [inline]
 call_rcu+0xee/0x890 kernel/rcu/tree.c:3251
 task_work_run+0x1d9/0x270 kernel/task_work.c:233
 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline]
 __exit_to_user_mode_loop kernel/entry/common.c:67 [inline]
 exit_to_user_mode_loop+0xed/0x480 kernel/entry/common.c:98
 __exit_to_user_mode_prepare include/linux/irq-entry-common.h:226 [inline]
 syscall_exit_to_user_mode_prepare include/linux/irq-entry-common.h:256 [inline]
 syscall_exit_to_user_mode include/linux/entry-common.h:325 [inline]
 do_syscall_64+0x32d/0xf80 arch/x86/entry/syscall_64.c:100
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Second to last potentially related work creation:
 kasan_save_stack+0x3e/0x60 mm/kasan/common.c:57
 kasan_record_aux_stack+0xbd/0xd0 mm/kasan/generic.c:556
 task_work_add+0xb6/0x440 kernel/task_work.c:70
 mntput_no_expire_slowpath+0x70c/0xbd0 fs/namespace.c:1379
 create_new_namespace fs/namespace.c:3162 [inline]
 open_new_namespace fs/namespace.c:3168 [inline]
 vfs_open_tree+0xe17/0x1040 fs/namespace.c:3217
 __do_sys_open_tree fs/namespace.c:3227 [inline]
 __se_sys_open_tree fs/namespace.c:3225 [inline]
 __x64_sys_open_tree+0x96/0x110 fs/namespace.c:3225
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x14d/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

The buggy address belongs to the object at ffff8880341bca80
 which belongs to the cache mnt_cache of size 352
The buggy address is located 192 bytes inside of
 freed 352-byte region [ffff8880341bca80, ffff8880341bcbe0)

The buggy address belongs to the physical page:
page: refcount:0 mapcount:0 mapping:0000000000000000 index:0xffff8880341bda40 pfn:0x341bc
head: order:1 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0
memcg:ffff8880341bc171
flags: 0x80000000000240(workingset|head|node=0|zone=1)
page_type: f5(slab)
raw: 0080000000000240 ffff888140412780 ffffea0001036010 ffffea000100bc10
raw: ffff8880341bda40 000001c00012000f 00000000f5000000 ffff8880341bc171
head: 0080000000000240 ffff888140412780 ffffea0001036010 ffffea000100bc10
head: ffff8880341bda40 000001c00012000f 00000000f5000000 ffff8880341bc171
head: 0080000000000001 ffffea0000d06f01 00000000ffffffff 00000000ffffffff
head: 0000000000000000 0000000000000000 00000000ffffffff 0000000000000002
page dumped because: kasan: bad access detected
page_owner tracks the page as allocated
page last allocated via order 1, migratetype Unmovable, gfp_mask 0xd20c0(__GFP_IO|__GFP_FS|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 18378, tgid 18375 (syz.6.3692), ts 1535437491766, free_ts 1534179924298
 set_page_owner include/linux/page_owner.h:32 [inline]
 post_alloc_hook+0x228/0x280 mm/page_alloc.c:1884
 prep_new_page mm/page_alloc.c:1892 [inline]
 get_page_from_freelist+0x28bb/0x2950 mm/page_alloc.c:3950
 __alloc_frozen_pages_noprof+0x18d/0x380 mm/page_alloc.c:5245
 alloc_slab_page mm/slub.c:3238 [inline]
 allocate_slab+0x77/0x660 mm/slub.c:3411
 new_slab mm/slub.c:3469 [inline]
 refill_objects+0x334/0x3c0 mm/slub.c:7091
 refill_sheaf mm/slub.c:2787 [inline]
 __pcs_replace_empty_main+0x328/0x5f0 mm/slub.c:4536
 alloc_from_pcs mm/slub.c:4639 [inline]
 slab_alloc_node mm/slub.c:4773 [inline]
 kmem_cache_alloc_noprof+0x433/0x680 mm/slub.c:4795
 alloc_vfsmnt+0x23/0x420 fs/namespace.c:287
 clone_mnt+0x4b/0x9a0 fs/namespace.c:1246
 copy_tree+0x3d4/0x930 fs/namespace.c:2194
 get_detached_copy fs/namespace.c:3044 [inline]
 open_detached_copy+0x23e/0x5a0 fs/namespace.c:3060
 vfs_open_tree+0x490/0x1040 fs/namespace.c:3220
 __do_sys_open_tree fs/namespace.c:3227 [inline]
 __se_sys_open_tree fs/namespace.c:3225 [inline]
 __x64_sys_open_tree+0x96/0x110 fs/namespace.c:3225
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x14d/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
page last free pid 5166 tgid 5166 stack trace:
 reset_page_owner include/linux/page_owner.h:25 [inline]
 free_pages_prepare mm/page_alloc.c:1433 [inline]
 __free_frozen_pages+0xfd0/0x1160 mm/page_alloc.c:2973
 __slab_free+0x24f/0x2a0 mm/slub.c:5490
 qlink_free mm/kasan/quarantine.c:163 [inline]
 qlist_free_all+0x97/0x100 mm/kasan/quarantine.c:179
 kasan_quarantine_reduce+0x148/0x160 mm/kasan/quarantine.c:286
 __kasan_slab_alloc+0x22/0x80 mm/kasan/common.c:350
 kasan_slab_alloc include/linux/kasan.h:253 [inline]
 slab_post_alloc_hook mm/slub.c:4459 [inline]
 slab_alloc_node mm/slub.c:4788 [inline]
 kmem_cache_alloc_noprof+0x33b/0x680 mm/slub.c:4795
 alloc_filename fs/namei.c:142 [inline]
 do_getname+0x2e/0x250 fs/namei.c:182
 getname include/linux/fs.h:2512 [inline]
 class_filename_constructor include/linux/fs.h:2539 [inline]
 do_sys_openat2+0xca/0x200 fs/open.c:1365
 do_sys_open fs/open.c:1372 [inline]
 __do_sys_openat fs/open.c:1388 [inline]
 __se_sys_openat fs/open.c:1383 [inline]
 __x64_sys_openat+0x138/0x170 fs/open.c:1383
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x14d/0xf80 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Memory state around the buggy address:
 ffff8880341bca00: 00 00 00 00 fc fc fc fc fc fc fc fc fc fc fc fc
 ffff8880341bca80: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff8880341bcb00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                                           ^
 ffff8880341bcb80: fb fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc
 ffff8880341bcc00: fc fc fc fc fc fc fc fc 00 00 00 00 00 00 00 00
==================================================================


---
This report is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzkaller@googlegroups.com.

syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.

If the report is already addressed, let syzbot know by replying with:
#syz fix: exact-commit-title

If you want to overwrite report's subsystems, reply with:
#syz set subsystems: new-subsystem
(See the list of subsystem names on the web dashboard)

If the report is a duplicate of another one, reply with:
#syz dup: exact-subject-of-another-report

If you want to undo deduplication, reply with:
#syz undup

Re: [syzbot] [fs?] KASAN: slab-use-after-free Read in clone_mnt

[Christian Brauner]

[-- Attachment #1: Type: text/plain, Size: 1668 bytes --]

On Sat, Feb 14, 2026 at 02:01:26AM -0800, syzbot wrote:
> Hello,
> 
> syzbot found the following issue on:
> 
> HEAD commit:    c22e26bd0906 Merge tag 'landlock-7.0-rc1' of git://git.ker..
> git tree:       upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=12c6a6e6580000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=6428d17febdfb14e
> dashboard link: https://syzkaller.appspot.com/bug?extid=a89f9434fb5a001ccd58
> compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
> 
> Unfortunately, I don't have any reproducer for this issue yet.
> 
> Downloadable assets:
> disk image: https://storage.googleapis.com/syzbot-assets/b33c549157ca/disk-c22e26bd.raw.xz
> vmlinux: https://storage.googleapis.com/syzbot-assets/34c7ded19553/vmlinux-c22e26bd.xz
> kernel image: https://storage.googleapis.com/syzbot-assets/66faec2158ed/bzImage-c22e26bd.xz
> 
> IMPORTANT: if you fix the issue, please add the following tag to the commit:
> Reported-by: syzbot+a89f9434fb5a001ccd58@syzkaller.appspotmail.com

Right, this should be fixed by the appended patch. If someone had the
bright idea to make the real rootfs a shared or dependent mount and it
is later copied the copy will become a peer of the old real rootfs mount
or a dependent mount of it. If locking the new rootfs mount for mounting
fails or the subsequent do_loopback() fails we rely on the copy of the
real root mount to be cleaned up by path_put(). The problem is that this
doesn't deal with mount propagation and will leave the mounts linked in
the propagation lists. Fix this by actually unmounting the copied tree.



[-- Attachment #2: 0001-mount-unmount-copied-tree.patch --]
[-- Type: text/x-diff, Size: 4131 bytes --]

From 3b3dc6a956af1afc6b95c68550b99083d47ce5fa Mon Sep 17 00:00:00 2001
From: Christian Brauner <brauner@kernel.org>
Date: Sat, 14 Feb 2026 16:22:13 +0100
Subject: [PATCH] mount: unmount copied tree

Fix an oversight when creating a new mount namespace. If someone had the
bright idea to make the real rootfs a shared or dependent mount and it
is later copied the copy will become a peer of the old real rootfs mount
or a dependent mount of it. The namespace semaphore is dropped and we
use mount lock exact to lock the new real root mount. If that fails or
the subsequent do_loopback() fails we rely on the copy of the real root
mount to be cleaned up by path_put(). The problem is that this doesn't
deal with mount propagation and will leave the mounts linked in the
propagation lists. Fix this by actually unmounting the copied tree.

Reported-by: syzbot+a89f9434fb5a001ccd58@syzkaller.appspotmail.com
Link: https:/lore.kernel.org/699047f6.050a0220.2757fb.0024.GAE@google.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
---
 fs/namespace.c | 37 ++++++++++++++++++++++---------------
 1 file changed, 22 insertions(+), 15 deletions(-)

diff --git a/fs/namespace.c b/fs/namespace.c
index a67cbe42746d..89cac95e6c9b 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -3073,16 +3073,13 @@ static struct file *open_detached_copy(struct path *path, unsigned int flags)
 	return file;
 }
 
-DEFINE_FREE(put_empty_mnt_ns, struct mnt_namespace *,
-	    if (!IS_ERR_OR_NULL(_T)) free_mnt_ns(_T))
-
 static struct mnt_namespace *create_new_namespace(struct path *path, unsigned int flags)
 {
-	struct mnt_namespace *new_ns __free(put_empty_mnt_ns) = NULL;
-	struct path to_path __free(path_put) = {};
 	struct mnt_namespace *ns = current->nsproxy->mnt_ns;
 	struct user_namespace *user_ns = current_user_ns();
+	struct mnt_namespace *new_ns;
 	struct mount *new_ns_root;
+	struct path to_path;
 	struct mount *mnt;
 	unsigned int copy_flags = 0;
 	bool locked = false;
@@ -3096,8 +3093,10 @@ static struct mnt_namespace *create_new_namespace(struct path *path, unsigned in
 
 	scoped_guard(namespace_excl) {
 		new_ns_root = clone_mnt(ns->root, ns->root->mnt.mnt_root, copy_flags);
-		if (IS_ERR(new_ns_root))
+		if (IS_ERR(new_ns_root)) {
+			emptied_ns = new_ns;
 			return ERR_CAST(new_ns_root);
+		}
 
 		/*
 		 * If the real rootfs had a locked mount on top of it somewhere
@@ -3122,12 +3121,17 @@ static struct mnt_namespace *create_new_namespace(struct path *path, unsigned in
 
 	/* Borrow the reference from clone_mnt(). */
 	to_path.mnt = &new_ns_root->mnt;
-	to_path.dentry = dget(new_ns_root->mnt.mnt_root);
+	to_path.dentry = new_ns_root->mnt.mnt_root;
 
 	/* Now lock for actual mounting. */
 	LOCK_MOUNT_EXACT(mp, &to_path);
-	if (unlikely(IS_ERR(mp.parent)))
+	if (unlikely(IS_ERR(mp.parent))) {
+		guard(namespace_excl)();
+		emptied_ns = new_ns;
+		guard(mount_writer)();
+		umount_tree(new_ns_root, 0);
 		return ERR_CAST(mp.parent);
+	}
 
 	/*
 	 * We don't emulate unshare()ing a mount namespace. We stick to the
@@ -3135,10 +3139,13 @@ static struct mnt_namespace *create_new_namespace(struct path *path, unsigned in
 	 * saner and simpler semantics.
 	 */
 	mnt = __do_loopback(path, flags, copy_flags);
-	if (IS_ERR(mnt))
-		return ERR_CAST(mnt);
-
 	scoped_guard(mount_writer) {
+		if (IS_ERR(mnt)) {
+			emptied_ns = new_ns;
+			umount_tree(new_ns_root, 0);
+			return ERR_CAST(mnt);
+		}
+
 		if (locked)
 			mnt->mnt.mnt_flags |= MNT_LOCKED;
 		/*
@@ -3151,14 +3158,14 @@ static struct mnt_namespace *create_new_namespace(struct path *path, unsigned in
 	}
 
 	/* Add all mounts to the new namespace. */
-	for (struct mount *p = new_ns_root; p; p = next_mnt(p, new_ns_root)) {
-		mnt_add_to_ns(new_ns, p);
+	for (mnt = new_ns_root; mnt; mnt = next_mnt(mnt, new_ns_root)) {
+		mnt_add_to_ns(new_ns, mnt);
 		new_ns->nr_mounts++;
 	}
 
-	new_ns->root = real_mount(no_free_ptr(to_path.mnt));
+	new_ns->root = real_mount(to_path.mnt);
 	ns_tree_add_raw(new_ns);
-	return no_free_ptr(new_ns);
+	return new_ns;
 }
 
 static struct file *open_new_namespace(struct path *path, unsigned int flags)
-- 
2.47.3

Re: [syzbot] [fs?] KASAN: slab-use-after-free Read in clone_mnt

[Christian Brauner]

[-- Attachment #1: Type: text/plain, Size: 1923 bytes --]

On Sat, Feb 14, 2026 at 04:59:11PM +0100, Christian Brauner wrote:
> On Sat, Feb 14, 2026 at 02:01:26AM -0800, syzbot wrote:
> > Hello,
> > 
> > syzbot found the following issue on:
> > 
> > HEAD commit:    c22e26bd0906 Merge tag 'landlock-7.0-rc1' of git://git.ker..
> > git tree:       upstream
> > console output: https://syzkaller.appspot.com/x/log.txt?x=12c6a6e6580000
> > kernel config:  https://syzkaller.appspot.com/x/.config?x=6428d17febdfb14e
> > dashboard link: https://syzkaller.appspot.com/bug?extid=a89f9434fb5a001ccd58
> > compiler:       Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
> > 
> > Unfortunately, I don't have any reproducer for this issue yet.
> > 
> > Downloadable assets:
> > disk image: https://storage.googleapis.com/syzbot-assets/b33c549157ca/disk-c22e26bd.raw.xz
> > vmlinux: https://storage.googleapis.com/syzbot-assets/34c7ded19553/vmlinux-c22e26bd.xz
> > kernel image: https://storage.googleapis.com/syzbot-assets/66faec2158ed/bzImage-c22e26bd.xz
> > 
> > IMPORTANT: if you fix the issue, please add the following tag to the commit:
> > Reported-by: syzbot+a89f9434fb5a001ccd58@syzkaller.appspotmail.com
> 
> Right, this should be fixed by the appended patch. If someone had the
> bright idea to make the real rootfs a shared or dependent mount and it
> is later copied the copy will become a peer of the old real rootfs mount
> or a dependent mount of it. If locking the new rootfs mount for mounting
> fails or the subsequent do_loopback() fails we rely on the copy of the
> real root mount to be cleaned up by path_put(). The problem is that this
> doesn't deal with mount propagation and will leave the mounts linked in
> the propagation lists. Fix this by actually unmounting the copied tree.

Actually lets just refactor this to be a bit cleaner as in the appended
patch. I plan on getting this upstream during this week.

[-- Attachment #2: 0001-mount-hold-namespace_sem-across-copy-in-create_new_n.patch --]
[-- Type: text/x-diff, Size: 8666 bytes --]

From 186af1baac0e03b0915287d31d8fcf237d9e5410 Mon Sep 17 00:00:00 2001
From: Christian Brauner <brauner@kernel.org>
Date: Sat, 14 Feb 2026 16:22:13 +0100
Subject: [PATCH] mount: hold namespace_sem across copy in
 create_new_namespace()

Fix an oversight when creating a new mount namespace. If someone had the
bright idea to make the real rootfs a shared or dependent mount and it
is later copied the copy will become a peer of the old real rootfs mount
or a dependent mount of it. The namespace semaphore is dropped and we
use mount lock exact to lock the new real root mount. If that fails or
the subsequent do_loopback() fails we rely on the copy of the real root
mount to be cleaned up by path_put(). The problem is that this doesn't
deal with mount propagation and will leave the mounts linked in the
propagation lists.

When creating a new mount namespace create_new_namespace() first
acquires namespace_sem to clone the nullfs root, drops it, then
reacquires it via LOCK_MOUNT_EXACT which takes inode_lock first to
respect the inode_lock -> namespace_sem lock ordering. This
drop-and-reacquire pattern is fragile and was the source of the
propagation cleanup bug fixed in the preceding commit.

Extend lock_mount_exact() with a copy_mount mode that clones the mount
under the locks atomically. When copy_mount is true, path_overmounted()
is skipped since we're copying the mount, not mounting on top of it -
the nullfs root always has rootfs mounted on top so the check would
always fail. If clone_mnt() fails after get_mountpoint() has pinned the
mountpoint, __unlock_mount() is used to properly unpin the mountpoint
and release both locks.

This allows create_new_namespace() to use LOCK_MOUNT_EXACT_COPY which
takes inode_lock and namespace_sem once and holds them throughout the
clone and subsequent mount operations, eliminating the
drop-and-reacquire pattern entirely.

Reported-by: syzbot+a89f9434fb5a001ccd58@syzkaller.appspotmail.com
Fixes: 9b8a0ba68246 ("mount: add OPEN_TREE_NAMESPACE") # mainline only
Link: https://lore.kernel.org/699047f6.050a0220.2757fb.0024.GAE@google.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
---
 fs/namespace.c | 112 +++++++++++++++++++++++++------------------------
 1 file changed, 58 insertions(+), 54 deletions(-)

diff --git a/fs/namespace.c b/fs/namespace.c
index 90700df65f0d..188be116f6e6 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -2791,7 +2791,8 @@ static inline void unlock_mount(struct pinned_mountpoint *m)
 }
 
 static void lock_mount_exact(const struct path *path,
-			     struct pinned_mountpoint *mp);
+			     struct pinned_mountpoint *mp, bool copy_mount,
+			     unsigned int copy_flags);
 
 #define LOCK_MOUNT_MAYBE_BENEATH(mp, path, beneath) \
 	struct pinned_mountpoint mp __cleanup(unlock_mount) = {}; \
@@ -2799,7 +2800,10 @@ static void lock_mount_exact(const struct path *path,
 #define LOCK_MOUNT(mp, path) LOCK_MOUNT_MAYBE_BENEATH(mp, (path), false)
 #define LOCK_MOUNT_EXACT(mp, path) \
 	struct pinned_mountpoint mp __cleanup(unlock_mount) = {}; \
-	lock_mount_exact((path), &mp)
+	lock_mount_exact((path), &mp, false, 0)
+#define LOCK_MOUNT_EXACT_COPY(mp, path, copy_flags) \
+	struct pinned_mountpoint mp __cleanup(unlock_mount) = {}; \
+	lock_mount_exact((path), &mp, true, (copy_flags))
 
 static int graft_tree(struct mount *mnt, const struct pinned_mountpoint *mp)
 {
@@ -3073,16 +3077,13 @@ static struct file *open_detached_copy(struct path *path, unsigned int flags)
 	return file;
 }
 
-DEFINE_FREE(put_empty_mnt_ns, struct mnt_namespace *,
-	    if (!IS_ERR_OR_NULL(_T)) free_mnt_ns(_T))
-
 static struct mnt_namespace *create_new_namespace(struct path *path, unsigned int flags)
 {
-	struct mnt_namespace *new_ns __free(put_empty_mnt_ns) = NULL;
-	struct path to_path __free(path_put) = {};
 	struct mnt_namespace *ns = current->nsproxy->mnt_ns;
 	struct user_namespace *user_ns = current_user_ns();
-	struct mount *new_ns_root;
+	struct mnt_namespace *new_ns;
+	struct mount *new_ns_root, *old_ns_root;
+	struct path to_path;
 	struct mount *mnt;
 	unsigned int copy_flags = 0;
 	bool locked = false;
@@ -3094,71 +3095,64 @@ static struct mnt_namespace *create_new_namespace(struct path *path, unsigned in
 	if (IS_ERR(new_ns))
 		return ERR_CAST(new_ns);
 
-	scoped_guard(namespace_excl) {
-		new_ns_root = clone_mnt(ns->root, ns->root->mnt.mnt_root, copy_flags);
-		if (IS_ERR(new_ns_root))
-			return ERR_CAST(new_ns_root);
+	old_ns_root = ns->root;
+	to_path.mnt = &old_ns_root->mnt;
+	to_path.dentry = old_ns_root->mnt.mnt_root;
 
-		/*
-		 * If the real rootfs had a locked mount on top of it somewhere
-		 * in the stack, lock the new mount tree as well so it can't be
-		 * exposed.
-		 */
-		mnt = ns->root;
-		while (mnt->overmount) {
-			mnt = mnt->overmount;
-			if (mnt->mnt.mnt_flags & MNT_LOCKED)
-				locked = true;
-		}
+	VFS_WARN_ON_ONCE(old_ns_root->mnt_id_unique != 1);
+	VFS_WARN_ON_ONCE(old_ns_root->mnt.mnt_sb->s_type != &nullfs_fs_type);
+
+	LOCK_MOUNT_EXACT_COPY(mp, &to_path, copy_flags);
+	if (IS_ERR(mp.parent)) {
+		free_mnt_ns(new_ns);
+		return ERR_CAST(mp.parent);
 	}
+	new_ns_root = mp.parent;
 
 	/*
-	 * We dropped the namespace semaphore so we can actually lock
-	 * the copy for mounting. The copied mount isn't attached to any
-	 * mount namespace and it is thus excluded from any propagation.
-	 * So realistically we're isolated and the mount can't be
-	 * overmounted.
+	 * If the real rootfs had a locked mount on top of it somewhere
+	 * in the stack, lock the new mount tree as well so it can't be
+	 * exposed.
 	 */
-
-	/* Borrow the reference from clone_mnt(). */
-	to_path.mnt = &new_ns_root->mnt;
-	to_path.dentry = dget(new_ns_root->mnt.mnt_root);
-
-	/* Now lock for actual mounting. */
-	LOCK_MOUNT_EXACT(mp, &to_path);
-	if (unlikely(IS_ERR(mp.parent)))
-		return ERR_CAST(mp.parent);
+	mnt = old_ns_root;
+	while (mnt->overmount) {
+		mnt = mnt->overmount;
+		if (mnt->mnt.mnt_flags & MNT_LOCKED)
+			locked = true;
+	}
 
 	/*
-	 * We don't emulate unshare()ing a mount namespace. We stick to the
-	 * restrictions of creating detached bind-mounts. It has a lot
-	 * saner and simpler semantics.
+	 * We don't emulate unshare()ing a mount namespace. We stick
+	 * to the restrictions of creating detached bind-mounts. It
+	 * has a lot saner and simpler semantics.
 	 */
 	mnt = __do_loopback(path, flags, copy_flags);
-	if (IS_ERR(mnt))
-		return ERR_CAST(mnt);
-
 	scoped_guard(mount_writer) {
+		if (IS_ERR(mnt)) {
+			emptied_ns = new_ns;
+			umount_tree(new_ns_root, 0);
+			return ERR_CAST(mnt);
+		}
+
 		if (locked)
 			mnt->mnt.mnt_flags |= MNT_LOCKED;
 		/*
-		 * Now mount the detached tree on top of the copy of the
-		 * real rootfs we created.
+		 * now mount the detached tree on top of the copy
+		 * of the real rootfs we created.
 		 */
 		attach_mnt(mnt, new_ns_root, mp.mp);
 		if (user_ns != ns->user_ns)
 			lock_mnt_tree(new_ns_root);
 	}
 
-	/* Add all mounts to the new namespace. */
-	for (struct mount *p = new_ns_root; p; p = next_mnt(p, new_ns_root)) {
-		mnt_add_to_ns(new_ns, p);
+	for (mnt = new_ns_root; mnt; mnt = next_mnt(mnt, new_ns_root)) {
+		mnt_add_to_ns(new_ns, mnt);
 		new_ns->nr_mounts++;
 	}
 
-	new_ns->root = real_mount(no_free_ptr(to_path.mnt));
+	new_ns->root = new_ns_root;
 	ns_tree_add_raw(new_ns);
-	return no_free_ptr(new_ns);
+	return new_ns;
 }
 
 static struct file *open_new_namespace(struct path *path, unsigned int flags)
@@ -3840,16 +3834,20 @@ static int do_new_mount(const struct path *path, const char *fstype,
 }
 
 static void lock_mount_exact(const struct path *path,
-			     struct pinned_mountpoint *mp)
+			     struct pinned_mountpoint *mp, bool copy_mount,
+			     unsigned int copy_flags)
 {
 	struct dentry *dentry = path->dentry;
 	int err;
 
+	/* Assert that inode_lock() locked the correct inode. */
+	VFS_WARN_ON_ONCE(copy_mount && !path_mounted(path));
+
 	inode_lock(dentry->d_inode);
 	namespace_lock();
 	if (unlikely(cant_mount(dentry)))
 		err = -ENOENT;
-	else if (path_overmounted(path))
+	else if (!copy_mount && path_overmounted(path))
 		err = -EBUSY;
 	else
 		err = get_mountpoint(dentry, mp);
@@ -3857,9 +3855,15 @@ static void lock_mount_exact(const struct path *path,
 		namespace_unlock();
 		inode_unlock(dentry->d_inode);
 		mp->parent = ERR_PTR(err);
-	} else {
-		mp->parent = real_mount(path->mnt);
+		return;
 	}
+
+	if (copy_mount)
+		mp->parent = clone_mnt(real_mount(path->mnt), dentry, copy_flags);
+	else
+		mp->parent = real_mount(path->mnt);
+	if (unlikely(IS_ERR(mp->parent)))
+		__unlock_mount(mp);
 }
 
 int finish_automount(struct vfsmount *__m, const struct path *path)
-- 
2.47.3