f2fs: fix to avoid invalid wait context issue

[ Upstream commit 90d5c9b ]

=============================
[ BUG: Invalid wait context ]
6.13.0-rc1 torvalds#84 Tainted: G           O
-----------------------------
cat/56160 is trying to lock:
ffff888105c86648 (&cprc->stat_lock){+.+.}-{3:3}, at: update_general_status+0x32a/0x8c0 [f2fs]
other info that might help us debug this:
context-{5:5}
2 locks held by cat/56160:
 #0: ffff88810a002a98 (&p->lock){+.+.}-{4:4}, at: seq_read_iter+0x56/0x4c0
 #1: ffffffffa0462638 (f2fs_stat_lock){....}-{2:2}, at: stat_show+0x29/0x1020 [f2fs]
stack backtrace:
CPU: 0 UID: 0 PID: 56160 Comm: cat Tainted: G           O       6.13.0-rc1 torvalds#84
Tainted: [O]=OOT_MODULE
Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006
Call Trace:
 <TASK>
 dump_stack_lvl+0x88/0xd0
 dump_stack+0x14/0x20
 __lock_acquire+0x8d4/0xbb0
 lock_acquire+0xd6/0x300
 _raw_spin_lock+0x38/0x50
 update_general_status+0x32a/0x8c0 [f2fs]
 stat_show+0x50/0x1020 [f2fs]
 seq_read_iter+0x116/0x4c0
 seq_read+0xfa/0x130
 full_proxy_read+0x66/0x90
 vfs_read+0xc4/0x350
 ksys_read+0x74/0xf0
 __x64_sys_read+0x1d/0x20
 x64_sys_call+0x17d9/0x1b80
 do_syscall_64+0x68/0x130
 entry_SYSCALL_64_after_hwframe+0x67/0x6f
RIP: 0033:0x7f2ca53147e2

- seq_read
 - stat_show
  - raw_spin_lock_irqsave(&f2fs_stat_lock, flags)
  : f2fs_stat_lock is raw_spinlock_t type variable
  - update_general_status
   - spin_lock(&sbi->cprc_info.stat_lock);
   : stat_lock is spinlock_t type variable

The root cause is the lock order is incorrect [1], we should not acquire
spinlock_t lock after raw_spinlock_t lock, as if CONFIG_PREEMPT_LOCK is
on, spinlock_t is implemented based on rtmutex, which can sleep after
holding the lock.

To fix this issue, let's use change f2fs_stat_lock lock type from
raw_spinlock_t to spinlock_t, it's safe due to:
- we don't need to use raw version of spinlock as the path is not
performance sensitive.
- we don't need to use irqsave version of spinlock as it won't be
used in irq context.

Quoted from [1]:

"Extend lockdep to validate lock wait-type context.

The current wait-types are:

	LD_WAIT_FREE,		/* wait free, rcu etc.. */
	LD_WAIT_SPIN,		/* spin loops, raw_spinlock_t etc.. */
	LD_WAIT_CONFIG,		/* CONFIG_PREEMPT_LOCK, spinlock_t etc.. */
	LD_WAIT_SLEEP,		/* sleeping locks, mutex_t etc.. */

Where lockdep validates that the current lock (the one being acquired)
fits in the current wait-context (as generated by the held stack).

This ensures that there is no attempt to acquire mutexes while holding
spinlocks, to acquire spinlocks while holding raw_spinlocks and so on. In
other words, its a more fancy might_sleep()."

[1] https://lore.kernel.org/all/[email protected]

Fixes: 98237fc ("f2fs: use spin_lock to avoid hang")
Signed-off-by: Chao Yu <[email protected]>
Signed-off-by: Jaegeuk Kim <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>

Author: chaseyu

fs/f2fs/debug.c

@@ -21,7 +21,7 @@
 #include "gc.h"
 
 static LIST_HEAD(f2fs_stat_list);
-static DEFINE_RAW_SPINLOCK(f2fs_stat_lock);
+static DEFINE_SPINLOCK(f2fs_stat_lock);
 #ifdef CONFIG_DEBUG_FS
 static struct dentry *f2fs_debugfs_root;
 #endif
@@ -439,9 +439,8 @@ static int stat_show(struct seq_file *s, void *v)
 {
 	struct f2fs_stat_info *si;
 	int i = 0, j = 0;
-	unsigned long flags;
 
-	raw_spin_lock_irqsave(&f2fs_stat_lock, flags);
+	spin_lock(&f2fs_stat_lock);
 	list_for_each_entry(si, &f2fs_stat_list, stat_list) {
 		struct f2fs_sb_info *sbi = si->sbi;
 
@@ -753,7 +752,7 @@ static int stat_show(struct seq_file *s, void *v)
 		seq_printf(s, "  - paged : %llu KB\n",
 				si->page_mem >> 10);
 	}
-	raw_spin_unlock_irqrestore(&f2fs_stat_lock, flags);
+	spin_unlock(&f2fs_stat_lock);
 	return 0;
 }
 
@@ -765,7 +764,6 @@ int f2fs_build_stats(struct f2fs_sb_info *sbi)
 	struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi);
 	struct f2fs_stat_info *si;
 	struct f2fs_dev_stats *dev_stats;
-	unsigned long flags;
 	int i;
 
 	si = f2fs_kzalloc(sbi, sizeof(struct f2fs_stat_info), GFP_KERNEL);
@@ -817,21 +815,20 @@ int f2fs_build_stats(struct f2fs_sb_info *sbi)
 
 	atomic_set(&sbi->max_aw_cnt, 0);
 
-	raw_spin_lock_irqsave(&f2fs_stat_lock, flags);
+	spin_lock(&f2fs_stat_lock);
 	list_add_tail(&si->stat_list, &f2fs_stat_list);
-	raw_spin_unlock_irqrestore(&f2fs_stat_lock, flags);
+	spin_unlock(&f2fs_stat_lock);
 
 	return 0;
 }
 
 void f2fs_destroy_stats(struct f2fs_sb_info *sbi)
 {
 	struct f2fs_stat_info *si = F2FS_STAT(sbi);
-	unsigned long flags;
 
-	raw_spin_lock_irqsave(&f2fs_stat_lock, flags);
+	spin_lock(&f2fs_stat_lock);
 	list_del(&si->stat_list);
-	raw_spin_unlock_irqrestore(&f2fs_stat_lock, flags);
+	spin_unlock(&f2fs_stat_lock);
 
 	kfree(si->dev_stats);
 	kfree(si);