[PATCH AUTOSEL 6.0 07/18] chardev: Fix potential memory leak when cdev_add() failed

[PATCH AUTOSEL 6.0 07/18] chardev: Fix potential memory leak when cdev_add() failed

[Sasha Levin]

From: Shang XiaoJing <shangxiaojing@huawei.com>

[ Upstream commit 4634c973096a64662a24d9914c47cebc2a8b72f4 ]

Some init function of cdev(like comedi) will call kobject_set_name()
before cdev_add(), but won't free the cdev.kobj.name or put the ref cnt
of cdev.kobj when cdev_add() failed. As the result, cdev.kobj.name will
be leaked.

Free the name of kobject in cdev_add() fail path to prevent memleak. With
this fix, the callers don't need to care about freeing the name of
kobject if cdev_add() fails.

unreferenced object 0xffff8881000fa8c0 (size 8):
  comm "modprobe", pid 239, jiffies 4294905173 (age 51.308s)
  hex dump (first 8 bytes):
    63 6f 6d 65 64 69 00 ff                          comedi..
  backtrace:
    [<000000005f9878f7>] __kmalloc_node_track_caller+0x4c/0x1c0
    [<000000000fd70302>] kstrdup+0x3f/0x70
    [<000000009428bc33>] kstrdup_const+0x46/0x60
    [<00000000ed50d9de>] kvasprintf_const+0xdb/0xf0
    [<00000000b2766964>] kobject_set_name_vargs+0x3c/0xe0
    [<00000000f2424ef7>] kobject_set_name+0x62/0x90
    [<000000005d5a125b>] 0xffffffffa0013098
    [<00000000f331e663>] do_one_initcall+0x7a/0x380
    [<00000000aa7bac96>] do_init_module+0x5c/0x230
    [<000000005fd72335>] load_module+0x227d/0x2420
    [<00000000ad550cf1>] __do_sys_finit_module+0xd5/0x140
    [<00000000069a60c5>] do_syscall_64+0x3f/0x90
    [<00000000c5e0d521>] entry_SYSCALL_64_after_hwframe+0x63/0xcd

Suggested-by: Greg KH <gregkh@linuxfoundation.org>
Signed-off-by: Shang XiaoJing <shangxiaojing@huawei.com>
Link: https://lore.kernel.org/r/20221102072659.23671-1-shangxiaojing@huawei.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/char_dev.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/fs/char_dev.c b/fs/char_dev.c
index ba0ded7842a7..340e4543b24a 100644
--- a/fs/char_dev.c
+++ b/fs/char_dev.c
@@ -483,17 +483,24 @@ int cdev_add(struct cdev *p, dev_t dev, unsigned count)
 	p->dev = dev;
 	p->count = count;
 
-	if (WARN_ON(dev == WHITEOUT_DEV))
-		return -EBUSY;
+	if (WARN_ON(dev == WHITEOUT_DEV)) {
+		error = -EBUSY;
+		goto err;
+	}
 
 	error = kobj_map(cdev_map, dev, count, NULL,
 			 exact_match, exact_lock, p);
 	if (error)
-		return error;
+		goto err;
 
 	kobject_get(p->kobj.parent);
 
 	return 0;
+
+err:
+	kfree_const(p->kobj.name);
+	p->kobj.name = NULL;
+	return error;
 }
 
 /**
-- 
2.35.1

[PATCH AUTOSEL 6.0 16/18] writeback: Add asserts for adding freed inode to lists

[Sasha Levin]

From: Jan Kara <jack@suse.cz>

[ Upstream commit a9438b44bc7015b18931e312bbd249a25bb59a65 ]

In the past we had several use-after-free issues with inodes getting
added to writeback lists after evict() removed them. These are painful
to debug so add some asserts to catch the problem earlier. The only
non-obvious change in the commit is that we need to tweak
redirty_tail_locked() to avoid triggering assertion in
inode_io_list_move_locked().

Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20221212113633.29181-1-jack@suse.cz
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/fs-writeback.c | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 9958d4020771..7ee235680006 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -121,6 +121,7 @@ static bool inode_io_list_move_locked(struct inode *inode,
 {
 	assert_spin_locked(&wb->list_lock);
 	assert_spin_locked(&inode->i_lock);
+	WARN_ON_ONCE(inode->i_state & I_FREEING);
 
 	list_move(&inode->i_io_list, head);
 
@@ -280,6 +281,7 @@ static void inode_cgwb_move_to_attached(struct inode *inode,
 {
 	assert_spin_locked(&wb->list_lock);
 	assert_spin_locked(&inode->i_lock);
+	WARN_ON_ONCE(inode->i_state & I_FREEING);
 
 	inode->i_state &= ~I_SYNC_QUEUED;
 	if (wb != &wb->bdi->wb)
@@ -1129,6 +1131,7 @@ static void inode_cgwb_move_to_attached(struct inode *inode,
 {
 	assert_spin_locked(&wb->list_lock);
 	assert_spin_locked(&inode->i_lock);
+	WARN_ON_ONCE(inode->i_state & I_FREEING);
 
 	inode->i_state &= ~I_SYNC_QUEUED;
 	list_del_init(&inode->i_io_list);
@@ -1294,6 +1297,17 @@ static void redirty_tail_locked(struct inode *inode, struct bdi_writeback *wb)
 {
 	assert_spin_locked(&inode->i_lock);
 
+	inode->i_state &= ~I_SYNC_QUEUED;
+	/*
+	 * When the inode is being freed just don't bother with dirty list
+	 * tracking. Flush worker will ignore this inode anyway and it will
+	 * trigger assertions in inode_io_list_move_locked().
+	 */
+	if (inode->i_state & I_FREEING) {
+		list_del_init(&inode->i_io_list);
+		wb_io_lists_depopulated(wb);
+		return;
+	}
 	if (!list_empty(&wb->b_dirty)) {
 		struct inode *tail;
 
@@ -1302,7 +1316,6 @@ static void redirty_tail_locked(struct inode *inode, struct bdi_writeback *wb)
 			inode->dirtied_when = jiffies;
 	}
 	inode_io_list_move_locked(inode, wb, &wb->b_dirty);
-	inode->i_state &= ~I_SYNC_QUEUED;
 }
 
 static void redirty_tail(struct inode *inode, struct bdi_writeback *wb)
-- 
2.35.1

[PATCH AUTOSEL 6.0 17/18] exfat: fix overflow in sector and cluster conversion

[Sasha Levin]

From: Yuezhang Mo <Yuezhang.Mo@sony.com>

[ Upstream commit 40306b4d1ba25970dafd53432e8daa5d591ebd99 ]

According to the exFAT specification, there are at most 2^32-11
clusters in a volume. so using 'int' is not enough for cluster
index, the return value type of exfat_sector_to_cluster() should
be 'unsigned int'.

Signed-off-by: Yuezhang Mo <Yuezhang.Mo@sony.com>
Reviewed-by: Sungjong Seo <sj1557.seo@samsung.com>
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/exfat/exfat_fs.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/exfat/exfat_fs.h b/fs/exfat/exfat_fs.h
index a8f8eee4937c..7f6b1d5bceab 100644
--- a/fs/exfat/exfat_fs.h
+++ b/fs/exfat/exfat_fs.h
@@ -375,7 +375,7 @@ static inline sector_t exfat_cluster_to_sector(struct exfat_sb_info *sbi,
 		sbi->data_start_sector;
 }
 
-static inline int exfat_sector_to_cluster(struct exfat_sb_info *sbi,
+static inline unsigned int exfat_sector_to_cluster(struct exfat_sb_info *sbi,
 		sector_t sec)
 {
 	return ((sec - sbi->data_start_sector) >> sbi->sect_per_clus_bits) +
-- 
2.35.1