On Sat, Jul 12, 2025 at 7:16 PM Greg KH gregkh@linuxfoundation.org wrote:
On Mon, Jun 30, 2025 at 02:05:42PM +0530, Pranav Tyagi wrote:
From: Mohammed Anees pvmohammedanees2003@gmail.com
[ Upstream commit 7bf1823e010e8db2fb649c790bd1b449a75f52d8 ]
syzbot has found a possible deadlock in ocfs2_get_system_file_inode [1].
The scenario is depicted here,
CPU0 CPU1lock(&ocfs2_file_ip_alloc_sem_key); lock(&osb->system_file_mutex); lock(&ocfs2_file_ip_alloc_sem_key); lock(&osb->system_file_mutex);
The function calls which could lead to this are:
CPU0 ocfs2_mknod - lock(&ocfs2_file_ip_alloc_sem_key); . . . ocfs2_get_system_file_inode - lock(&osb->system_file_mutex);
CPU1 - ocfs2_fill_super - lock(&osb->system_file_mutex); . . . ocfs2_read_virt_blocks - lock(&ocfs2_file_ip_alloc_sem_key);
This issue can be resolved by making the down_read -> down_read_try in the ocfs2_read_virt_blocks.
[1] https://syzkaller.appspot.com/bug?extid=e0055ea09f1f5e6fabdd
[ Backport to 5.15: context cleanly applied with no semantic changes. Build-tested. ]
We can't take a 5.15.y version, without it being in 6.1.y first, sorry :(
Hi,
Thanks for pointing that out. I wasn’t aware of the 6.1.y prerequisite for 5.15.y backports. Learned something new today. I’ll make sure to get it into 6.1.y first.
Regards Pranav Tyagi