From e096d0c7e2e4e5893792db865dd065ac73cf1f00 Mon Sep 17 00:00:00 2001 From: Josh Boyer Date: Thu, 25 Aug 2011 07:48:12 -0400 Subject: lockdep: Add helper function for dir vs file i_mutex annotation Purely in-memory filesystems do not use the inode hash as the dcache tells us if an entry already exists. As a result, they do not call unlock_new_inode, and thus directory inodes do not get put into a different lockdep class for i_sem. We need the different lockdep classes, because the locking order for i_mutex is different for directory inodes and regular inodes. Directory inodes can do "readdir()", which takes i_mutex *before* possibly taking mm->mmap_sem (due to a page fault while copying the directory entry to user space). In contrast, regular inodes can be mmap'ed, which takes mm->mmap_sem before accessing i_mutex. The two cases can never happen for the same inode, so no real deadlock can occur, but without the different lockdep classes, lockdep cannot understand that. As a result, if CONFIG_DEBUG_LOCK_ALLOC is set, this can lead to false positives from lockdep like below: find/645 is trying to acquire lock: (&mm->mmap_sem){++++++}, at: [] might_fault+0x5c/0xac but task is already holding lock: (&sb->s_type->i_mutex_key#15){+.+.+.}, at: [] vfs_readdir+0x5b/0xb4 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&sb->s_type->i_mutex_key#15){+.+.+.}: [] lock_acquire+0xbf/0x103 [] __mutex_lock_common+0x4c/0x361 [] mutex_lock_nested+0x40/0x45 [] hugetlbfs_file_mmap+0x82/0x110 [] mmap_region+0x258/0x432 [] do_mmap_pgoff+0x2ac/0x306 [] sys_mmap_pgoff+0x118/0x16a [] sys_mmap+0x22/0x24 [] system_call_fastpath+0x16/0x1b -> #0 (&mm->mmap_sem){++++++}: [] __lock_acquire+0xa1a/0xcf7 [] lock_acquire+0xbf/0x103 [] might_fault+0x89/0xac [] filldir+0x6f/0xc7 [] dcache_readdir+0x67/0x205 [] vfs_readdir+0x7b/0xb4 [] sys_getdents+0x7e/0xd1 [] system_call_fastpath+0x16/0x1b This patch moves the directory vs file lockdep annotation into a helper function that can be called by in-memory filesystems and has hugetlbfs call it. Signed-off-by: Josh Boyer Acked-by: Peter Zijlstra Signed-off-by: Linus Torvalds --- fs/inode.c | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) (limited to 'fs/inode.c') diff --git a/fs/inode.c b/fs/inode.c index 73920d555c8..ec7924696a1 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -848,16 +848,9 @@ struct inode *new_inode(struct super_block *sb) } EXPORT_SYMBOL(new_inode); -/** - * unlock_new_inode - clear the I_NEW state and wake up any waiters - * @inode: new inode to unlock - * - * Called when the inode is fully initialised to clear the new state of the - * inode and wake up anyone waiting for the inode to finish initialisation. - */ -void unlock_new_inode(struct inode *inode) -{ #ifdef CONFIG_DEBUG_LOCK_ALLOC +void lockdep_annotate_inode_mutex_key(struct inode *inode) +{ if (S_ISDIR(inode->i_mode)) { struct file_system_type *type = inode->i_sb->s_type; @@ -873,7 +866,20 @@ void unlock_new_inode(struct inode *inode) &type->i_mutex_dir_key); } } +} +EXPORT_SYMBOL(lockdep_annotate_inode_mutex_key); #endif + +/** + * unlock_new_inode - clear the I_NEW state and wake up any waiters + * @inode: new inode to unlock + * + * Called when the inode is fully initialised to clear the new state of the + * inode and wake up anyone waiting for the inode to finish initialisation. + */ +void unlock_new_inode(struct inode *inode) +{ + lockdep_annotate_inode_mutex_key(inode); spin_lock(&inode->i_lock); WARN_ON(!(inode->i_state & I_NEW)); inode->i_state &= ~I_NEW; -- cgit v1.2.3