[syzbot] possible deadlock in mi_read

[syzbot] possible deadlock in mi_read

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit:    bbed346d5a96 Merge branch 'for-next/core' into for-kernelci
git tree:       git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux.git for-kernelci
console output: https://syzkaller.appspot.com/x/log.txt?x=111f3904880000
kernel config:  https://syzkaller.appspot.com/x/.config?x=aae2d21e7dd80684
dashboard link: https://syzkaller.appspot.com/bug?extid=bc7ca0ae4591cb2550f9
compiler:       Debian clang version 13.0.1-++20220126092033+75e33f71c2da-1~exp1~20220126212112.63, GNU ld (GNU Binutils for Debian) 2.35.2
userspace arch: arm64
syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=13a49000880000
C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=14ace23f080000

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/11078f50b80b/disk-bbed346d.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/398e5f1e6c84/vmlinux-bbed346d.xz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+bc7ca0ae4591cb2550f9@syzkaller.appspotmail.com

loop0: detected capacity change from 0 to 4096
ntfs3: loop0: Different NTFS' sector size (1024) and media sector size (512)
============================================
WARNING: possible recursive locking detected
6.0.0-rc7-syzkaller-18095-gbbed346d5a96 #0 Not tainted
--------------------------------------------
syz-executor243/3030 is trying to acquire lock:
ffff0000caa09ca0 (&ni->ni_lock/4){+.+.}-{3:3}, at: ni_lock fs/ntfs3/ntfs_fs.h:1108 [inline]
ffff0000caa09ca0 (&ni->ni_lock/4){+.+.}-{3:3}, at: mi_read+0x140/0x274 fs/ntfs3/record.c:148

but task is already holding lock:
ffff0000caa0d3e0 (&ni->ni_lock/4){+.+.}-{3:3}, at: ntfs_lookup+0x84/0xe8 fs/ntfs/namei.c:111

other info that might help us debug this:
 Possible unsafe locking scenario:

       CPU0
       ----
  lock(&ni->ni_lock/4);
  lock(&ni->ni_lock/4);

 *** DEADLOCK ***

 May be due to missing lock nesting notation

2 locks held by syz-executor243/3030:
 #0: ffff0000caa0d680 (&type->i_mutex_dir_key#6){++++}-{3:3}, at: inode_lock_shared include/linux/fs.h:766 [inline]
 #0: ffff0000caa0d680 (&type->i_mutex_dir_key#6){++++}-{3:3}, at: lookup_slow+0x34/0x68 fs/namei.c:1701
 #1: ffff0000caa0d3e0 (&ni->ni_lock/4){+.+.}-{3:3}, at: ntfs_lookup+0x84/0xe8 fs/ntfs/namei.c:111

stack backtrace:
CPU: 1 PID: 3030 Comm: syz-executor243 Not tainted 6.0.0-rc7-syzkaller-18095-gbbed346d5a96 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 08/26/2022
Call trace:
 dump_backtrace+0x1c4/0x1f0 arch/arm64/kernel/stacktrace.c:156
 show_stack+0x2c/0x54 arch/arm64/kernel/stacktrace.c:163
 __dump_stack lib/dump_stack.c:88 [inline]
 dump_stack_lvl+0x104/0x16c lib/dump_stack.c:106
 dump_stack+0x1c/0x58 lib/dump_stack.c:113
 __lock_acquire+0x808/0x30a4
 lock_acquire+0x100/0x1f8 kernel/locking/lockdep.c:5666
 __mutex_lock_common+0xd4/0xca8 kernel/locking/mutex.c:603
 __mutex_lock kernel/locking/mutex.c:747 [inline]
 mutex_lock_nested+0x38/0x44 kernel/locking/mutex.c:799
 ni_lock fs/ntfs3/ntfs_fs.h:1108 [inline]
 mi_read+0x140/0x274 fs/ntfs3/record.c:148
 ntfs_read_mft fs/ntfs3/inode.c:69 [inline]
 ntfs_iget5+0x15c/0x138c fs/ntfs3/inode.c:501
 dir_search_u+0x18c/0x1e0 fs/ntfs3/dir.c:264
 ntfs_lookup+0x94/0xe8 fs/ntfs/namei.c:111
 __lookup_slow+0x14c/0x204 fs/namei.c:1685
 lookup_slow+0x44/0x68 fs/namei.c:1702
 walk_component+0x178/0x1b0 fs/namei.c:1993
 lookup_last fs/namei.c:2450 [inline]
 path_lookupat+0xc4/0x208 fs/namei.c:2474
 filename_lookup+0xf8/0x264 fs/namei.c:2503
 user_path_at_empty+0x5c/0x114 fs/namei.c:2876
 user_path_at include/linux/namei.h:57 [inline]
 __do_sys_open_tree fs/namespace.c:2537 [inline]
 __se_sys_open_tree fs/namespace.c:2504 [inline]
 __arm64_sys_open_tree+0x130/0x4a4 fs/namespace.c:2504
 __invoke_syscall arch/arm64/kernel/syscall.c:38 [inline]
 invoke_syscall arch/arm64/kernel/syscall.c:52 [inline]
 el0_svc_common+0x138/0x220 arch/arm64/kernel/syscall.c:142
 do_el0_svc+0x48/0x164 arch/arm64/kernel/syscall.c:206
 el0_svc+0x58/0x150 arch/arm64/kernel/entry-common.c:636
 el0t_64_sync_handler+0x84/0xf0 arch/arm64/kernel/entry-common.c:654
 el0t_64_sync+0x18c/0x190 arch/arm64/kernel/entry.S:581


---
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.
syzbot can test patches for this issue, for details see:
https://goo.gl/tpsmEJ#testing-patches

Re: [syzbot] possible deadlock in mi_read

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.

***

Subject: possible deadlock in mi_read
Author: almaz.alexandrovich@paragon-software.com

#syz test: https://github.com/Paragon-Software-Group/linux-ntfs3.git master

Re: [syzbot] possible fix (linux-ntfs3)

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.

***

Subject: possible fix (linux-ntfs3)
Author: almaz.alexandrovich@paragon-software.com

#syz test: https://github.com/Paragon-Software-Group/linux-ntfs3.git master

diff --git a/fs/ntfs3/super.c b/fs/ntfs3/super.c
index 128d49512f5d..c08bbacc0030 100644
--- a/fs/ntfs3/super.c
+++ b/fs/ntfs3/super.c
@@ -125,8 +125,9 @@ void ntfs_inode_printk(struct inode *inode, const 
char *fmt, ...)

          if (de) {
              spin_lock(&de->d_lock);
-            snprintf(name, sizeof(s_name_buf), " \"%s\"",
-                 de->d_name.name);
+            if (snprintf(name, sizeof(s_name_buf), " \"%s\"",
+                     de->d_name.name) >= sizeof(s_name_buf))
+                name[sizeof(s_name_buf) - 1] = 0;
              spin_unlock(&de->d_lock);
          } else {
              name[0] = 0;

Re: [syzbot] possible fix (linux-ntfs3)

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.

***

Subject: possible fix (linux-ntfs3)
Author: almaz.alexandrovich@paragon-software.com

#syz test: https://github.com/Paragon-Software-Group/linux-ntfs3.git master

diff --git a/fs/ntfs3/inode.c b/fs/ntfs3/inode.c
index 1963a8928360..ab2abd323859 100644
--- a/fs/ntfs3/inode.c
+++ b/fs/ntfs3/inode.c
@@ -1679,7 +1679,10 @@ int ntfs_create_inode(struct mnt_idmap *idmap, struct inode *dir,
 	attr = ni_find_attr(ni, NULL, NULL, ATTR_EA, NULL, 0, NULL, NULL);
 	if (attr && attr->non_res) {
 		/* Delete ATTR_EA, if non-resident. */
-		attr_set_size(ni, ATTR_EA, NULL, 0, NULL, 0, NULL, false, NULL);
+		struct runs_tree run;
+		run_init(&run);
+		attr_set_size(ni, ATTR_EA, NULL, 0, &run, 0, NULL, false, NULL);
+		run_close(&run);
 	}
 
 	if (rp_inserted)

Re: [syzbot] possible fix (linux-ntfs3)

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.

***

Subject: possible fix (linux-ntfs3)
Author: almaz.alexandrovich@paragon-software.com

#syz test: https://github.com/Paragon-Software-Group/linux-ntfs3.git master

diff --git a/fs/ntfs3/namei.c b/fs/ntfs3/namei.c
index fc720ad9c57a..aa7e9d70e32e 100644
--- a/fs/ntfs3/namei.c
+++ b/fs/ntfs3/namei.c
@@ -81,7 +81,7 @@ static struct dentry *ntfs_lookup(struct inode *dir, struct dentry *dentry,
 		if (err < 0)
 			inode = ERR_PTR(err);
 		else {
-			ni_lock(ni);
+			ni_lock_dir(ni);
 			inode = dir_search_u(dir, uni, NULL);
 			ni_unlock(ni);
 		}

Re: [syzbot] possible fix (linux-ntfs3)

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.

***

Subject: possible fix (linux-ntfs3)
Author: almaz.alexandrovich@paragon-software.com

#syz test: https://github.com/Paragon-Software-Group/linux-ntfs3.git master

diff --git a/fs/ntfs3/frecord.c b/fs/ntfs3/frecord.c
index 7d4e54161291..41c7ffad2790 100644
--- a/fs/ntfs3/frecord.c
+++ b/fs/ntfs3/frecord.c
@@ -102,7 +102,9 @@ void ni_clear(struct ntfs_inode *ni)
 {
 	struct rb_node *node;
 
-	if (!ni->vfs_inode.i_nlink && ni->mi.mrec && is_rec_inuse(ni->mi.mrec))
+	if (!ni->vfs_inode.i_nlink && ni->mi.mrec &&
+	    is_rec_inuse(ni->mi.mrec) &&
+	    !(ni->mi.sbi->flags & NTFS_FLAGS_LOG_REPLAYING))
 		ni_delete_all(ni);
 
 	al_destroy(ni);

Re: [syzbot] possible fix (linux-ntfs3)

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.

***

Subject: possible fix (linux-ntfs3)
Author: almaz.alexandrovich@paragon-software.com

#syz test: https://github.com/Paragon-Software-Group/linux-ntfs3.git master

diff --git a/fs/ntfs3/super.c b/fs/ntfs3/super.c
index 128d49512f5d..4a2062e481a4 100644
--- a/fs/ntfs3/super.c
+++ b/fs/ntfs3/super.c
@@ -124,10 +124,15 @@ void ntfs_inode_printk(struct inode *inode, const char *fmt, ...)
 		struct dentry *de = d_find_alias(inode);
 
 		if (de) {
+			int len;
 			spin_lock(&de->d_lock);
-			snprintf(name, sizeof(s_name_buf), " \"%s\"",
-				 de->d_name.name);
+			len = snprintf(name, sizeof(s_name_buf), " \"%s\"",
+				       de->d_name.name);
 			spin_unlock(&de->d_lock);
+			if (len <= 0)
+				name[0] = 0;
+			else if (len >= sizeof(s_name_buf))
+				name[sizeof(s_name_buf) - 1] = 0;
 		} else {
 			name[0] = 0;
 		}

Re: [syzbot] possible fix (linux-ntfs3)

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.

***

Subject: possible fix (linux-ntfs3)
Author: almaz.alexandrovich@paragon-software.com

Not sure about all indexes on volume must have the same index_block_size.

#syz test: https://github.com/Paragon-Software-Group/linux-ntfs3.git master

diff --git a/fs/ntfs3/index.c b/fs/ntfs3/index.c
index 9089c58a005c..63fbb8ba6e1b 100644
--- a/fs/ntfs3/index.c
+++ b/fs/ntfs3/index.c
@@ -892,7 +892,9 @@ int indx_init(struct ntfs_index *indx, struct ntfs_sb_info *sbi,
 	indx->idx2vbn_bits = __ffs(root->index_block_clst);
 
 	t32 = le32_to_cpu(root->index_block_size);
-	indx->index_bits = blksize_bits(t32);
+	if (t32 != sbi->index_size)
+		goto out;
+	indx->index_bits = sbi->index_bits;
 
 	/* Check index record size. */
 	if (t32 < sbi->cluster_size) {
diff --git a/fs/ntfs3/ntfs_fs.h b/fs/ntfs3/ntfs_fs.h
index 3dd6215316e4..73d72fa8ab65 100644
--- a/fs/ntfs3/ntfs_fs.h
+++ b/fs/ntfs3/ntfs_fs.h
@@ -222,8 +222,9 @@ struct ntfs_sb_info {
 	u32 record_size;
 	u32 index_size;
 
-	u8 cluster_bits;
-	u8 record_bits;
+	u8 cluster_bits; // log2(cluster_size)
+	u8 record_bits; // log2(record_size)
+	u8 index_bits; // log2(index_size)
 
 	u64 maxbytes; // Maximum size for normal files.
 	u64 maxbytes_sparse; // Maximum size for sparse file.
diff --git a/fs/ntfs3/super.c b/fs/ntfs3/super.c
index 128d49512f5d..f96641b80869 100644
--- a/fs/ntfs3/super.c
+++ b/fs/ntfs3/super.c
@@ -993,6 +993,7 @@ static int ntfs_init_from_boot(struct super_block *sb, u32 sector_size,
 			 sbi->index_size);
 		goto out;
 	}
+	sbi->index_bits = blksize_bits(sbi->index_size);
 
 	sbi->volume.size = sectors * boot_sector_size;
 

Re: [syzbot] check upstream

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com.

***

Subject: check upstream
Author: almaz.alexandrovich@paragon-software.com

#syz test