fs: WARNING in locks_unlink_lock_ctx (not holding proper lock)

20 views
Skip to first unread message

Dmitry Vyukov

unread,
Oct 7, 2016, 4:03:36 PM10/7/16
to Jeff Layton, Bruce Fields, Al Viro, linux-...@vger.kernel.org, LKML, syzkaller
Hello,

I am hitting lots of the following warnings while running syzkaller
fuzzer. Seems that path does not hold proper lock.

WARNING: CPU: 1 PID: 12090 at fs/locks.c:610 locks_unlink_lock_ctx+0x2c7/0x370
CPU: 1 PID: 12090 Comm: syz-executor Not tainted 4.8.0+ #28
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011
ffff880038ba7728 ffffffff82d2b849 ffffffff00000016 fffffbfff10971e8
ffffffff86e8c000 ffff880038ba7800 ffffffff86f42400 dffffc0000000000
0000000000000009 ffff880038ba77f0 ffffffff816a229a 0000000041b58ab3
Call Trace:
[< inline >] __dump_stack lib/dump_stack.c:15
[<ffffffff82d2b849>] dump_stack+0x12e/0x185 lib/dump_stack.c:51
[<ffffffff816a229a>] panic+0x1e9/0x3f4 kernel/panic.c:153
[<ffffffff81354fb9>] __warn+0x1c9/0x1e0 kernel/panic.c:509
[<ffffffff813551a1>] warn_slowpath_null+0x31/0x40 kernel/panic.c:552
[< inline >] locks_delete_global_locks fs/locks.c:610
[<ffffffff8193b247>] locks_unlink_lock_ctx+0x2c7/0x370 fs/locks.c:739
[<ffffffff8193b30f>] locks_delete_lock_ctx+0x1f/0x80 fs/locks.c:751
[<ffffffff8193d329>] lease_modify+0x229/0x2e0 fs/locks.c:1370
[< inline >] locks_remove_lease fs/locks.c:2528
[<ffffffff81947408>] locks_remove_file+0x2d8/0x380 fs/locks.c:2551
[<ffffffff8182eea6>] __fput+0x1a6/0x780 fs/file_table.c:200
[<ffffffff8182f50a>] ____fput+0x1a/0x20 fs/file_table.c:244
[<ffffffff813bae68>] task_work_run+0xf8/0x170 kernel/task_work.c:116
[< inline >] exit_task_work include/linux/task_work.h:21
[<ffffffff81364de4>] do_exit+0x864/0x2ad0 kernel/exit.c:828
[<ffffffff813671cd>] do_group_exit+0x10d/0x330 kernel/exit.c:931
[<ffffffff8138a57f>] get_signal+0x62f/0x15e0 kernel/signal.c:2307
[<ffffffff811cf344>] do_signal+0x84/0x18f0 arch/x86/kernel/signal.c:807
[<ffffffff8100629b>] exit_to_usermode_loop+0x13b/0x200
arch/x86/entry/common.c:156
[< inline >] prepare_exit_to_usermode arch/x86/entry/common.c:190
[< inline >] syscall_return_slowpath arch/x86/entry/common.c:259
[<ffffffff81008a4f>] do_syscall_64+0x49f/0x620 arch/x86/entry/common.c:285

On commit a6930aaee06755d1bdcfd943fbf614e4d92bb0c7 (Oct 5).

Jeff Layton

unread,
Oct 7, 2016, 7:26:41 PM10/7/16
to Dmitry Vyukov, Bruce Fields, Al Viro, linux-...@vger.kernel.org, LKML, syzkaller, Peter Zijlstra
(cc'ing Peter...)

Well spotted. Yeah, I think you're right. The assertion is this:

    percpu_rwsem_assert_held(&file_rwsem);

I'm guessing this is probably fallout from the lglock to rwsem
conversion (commit aba376607383).

From a quick glance, I think we probably just need to down_read the
file_rwsem in locks_remove_lease, prior to taking the flc_lock, and
release it just afterward. I do want to go over the code a little more
closely though to make sure other codepaths aren't missing that lock
though.

Thanks,
--
Jeff Layton <jla...@poochiereds.net>

Peter Zijlstra

unread,
Oct 8, 2016, 4:12:32 AM10/8/16
to Jeff Layton, Dmitry Vyukov, Bruce Fields, Al Viro, linux-...@vger.kernel.org, LKML, syzkaller
On Fri, Oct 07, 2016 at 07:26:36PM -0400, Jeff Layton wrote:

> Well spotted. Yeah, I think you're right. The assertion is this:
>
>     percpu_rwsem_assert_held(&file_rwsem);
>
> I'm guessing this is probably fallout from the lglock to rwsem
> conversion (commit aba376607383).
>
> From a quick glance, I think we probably just need to down_read the
> file_rwsem in locks_remove_lease, prior to taking the flc_lock, and
> release it just afterward.

Correct on all that.

> I do want to go over the code a little more
> closely though to make sure other codepaths aren't missing that lock
> though.

Urg, sorry for missing these, I went through it again and found the
below to be missing.

---
Subject: fs/locks: Add missing file_sem locks

I overlooked a few code-paths that can lead to
locks_delete_global_locks().

Reported-by: Dmitry Vyukov <dvy...@google.com>
Signed-off-by: Peter Zijlstra (Intel) <pet...@infradead.org>
---

diff --git a/fs/locks.c b/fs/locks.c
index 133fb2543d21..c623490863a9 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -1604,6 +1604,7 @@ int fcntl_getlease(struct file *filp)

ctx = smp_load_acquire(&inode->i_flctx);
if (ctx && !list_empty_careful(&ctx->flc_lease)) {
+ percpu_down_read_preempt_disable(&file_rwsem);
spin_lock(&ctx->flc_lock);
time_out_leases(file_inode(filp), &dispose);
list_for_each_entry(fl, &ctx->flc_lease, fl_list) {
@@ -1613,6 +1614,8 @@ int fcntl_getlease(struct file *filp)
break;
}
spin_unlock(&ctx->flc_lock);
+ percpu_up_read_preempt_enable(&file_rwsem);
+
locks_dispose_list(&dispose);
}
return type;
@@ -2522,11 +2525,14 @@ locks_remove_lease(struct file *filp, struct file_lock_context *ctx)
if (list_empty(&ctx->flc_lease))
return;

+ percpu_down_read_preempt_disable(&file_rwsem);
spin_lock(&ctx->flc_lock);
list_for_each_entry_safe(fl, tmp, &ctx->flc_lease, fl_list)
if (filp == fl->fl_file)
lease_modify(fl, F_UNLCK, &dispose);
spin_unlock(&ctx->flc_lock);
+ percpu_up_read_preempt_enable(&file_rwsem);
+
locks_dispose_list(&dispose);
}

Jeff Layton

unread,
Oct 8, 2016, 6:47:27 AM10/8/16
to Peter Zijlstra, Dmitry Vyukov, Bruce Fields, Al Viro, linux-...@vger.kernel.org, LKML, syzkaller
Looks correct to me.

Peter, do you need me to pick this patch up or do you plan to get it to
Linus another way?

Thanks, and:

Reviewed-by: Jeff Layton <jla...@poochiereds.net>

tip-bot for Peter Zijlstra

unread,
Oct 18, 2016, 5:31:10 AM10/18/16
to linux-ti...@vger.kernel.org, tg...@linutronix.de, dvy...@google.com, pet...@infradead.org, torv...@linux-foundation.org, jla...@poochiereds.net, bfi...@fieldses.org, linux-...@vger.kernel.org, mi...@kernel.org, h...@zytor.com, syzk...@googlegroups.com, vi...@zeniv.linux.org.uk
Commit-ID: 1c7df9c183278052aedc3dcb9fecb8bf2b24a659
Gitweb: http://git.kernel.org/tip/1c7df9c183278052aedc3dcb9fecb8bf2b24a659
Author: Peter Zijlstra <pet...@infradead.org>
AuthorDate: Sat, 8 Oct 2016 10:12:28 +0200
Committer: Ingo Molnar <mi...@kernel.org>
CommitDate: Tue, 18 Oct 2016 11:19:05 +0200

locking, fs/locks: Add missing file_sem locks

I overlooked a few code-paths that can lead to
locks_delete_global_locks().

Reported-by: Dmitry Vyukov <dvy...@google.com>
Signed-off-by: Peter Zijlstra (Intel) <pet...@infradead.org>
Reviewed-by: Jeff Layton <jla...@poochiereds.net>
Cc: Al Viro <vi...@zeniv.linux.org.uk>
Cc: Bruce Fields <bfi...@fieldses.org>
Cc: Linus Torvalds <torv...@linux-foundation.org>
Cc: Peter Zijlstra <pet...@infradead.org>
Cc: Thomas Gleixner <tg...@linutronix.de>
Cc: linux-...@vger.kernel.org
Cc: syzkaller <syzk...@googlegroups.com>
Link: http://lkml.kernel.org/r/2016100808...@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mi...@kernel.org>
---
fs/locks.c | 6 ++++++
1 file changed, 6 insertions(+)

diff --git a/fs/locks.c b/fs/locks.c
index ce93b41..22c5b4a 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -1609,6 +1609,7 @@ int fcntl_getlease(struct file *filp)

ctx = smp_load_acquire(&inode->i_flctx);
if (ctx && !list_empty_careful(&ctx->flc_lease)) {
+ percpu_down_read_preempt_disable(&file_rwsem);
spin_lock(&ctx->flc_lock);
time_out_leases(inode, &dispose);
list_for_each_entry(fl, &ctx->flc_lease, fl_list) {
@@ -1618,6 +1619,8 @@ int fcntl_getlease(struct file *filp)
break;
}
spin_unlock(&ctx->flc_lock);
+ percpu_up_read_preempt_enable(&file_rwsem);
+
locks_dispose_list(&dispose);
}
return type;
@@ -2529,11 +2532,14 @@ locks_remove_lease(struct file *filp, struct file_lock_context *ctx)

tip-bot for Peter Zijlstra

unread,
Oct 18, 2016, 6:25:41 AM10/18/16
to linux-ti...@vger.kernel.org, torv...@linux-foundation.org, jla...@poochiereds.net, bfi...@fieldses.org, dvy...@google.com, tg...@linutronix.de, vi...@zeniv.linux.org.uk, linux-...@vger.kernel.org, pet...@infradead.org, h...@zytor.com, mi...@kernel.org, syzk...@googlegroups.com
Commit-ID: 5f43086bb9224987010460dcf3dee68fbd4f574d
Gitweb: http://git.kernel.org/tip/5f43086bb9224987010460dcf3dee68fbd4f574d
Author: Peter Zijlstra <pet...@infradead.org>
AuthorDate: Sat, 8 Oct 2016 10:12:28 +0200
Committer: Ingo Molnar <mi...@kernel.org>
CommitDate: Tue, 18 Oct 2016 12:21:28 +0200

locking, fs/locks: Add missing file_sem locks

I overlooked a few code-paths that can lead to
locks_delete_global_locks().

Reported-by: Dmitry Vyukov <dvy...@google.com>
Signed-off-by: Peter Zijlstra (Intel) <pet...@infradead.org>
Reviewed-by: Jeff Layton <jla...@poochiereds.net>
Cc: Al Viro <vi...@zeniv.linux.org.uk>
Cc: Bruce Fields <bfi...@fieldses.org>
Cc: Linus Torvalds <torv...@linux-foundation.org>
Cc: Peter Zijlstra <pet...@infradead.org>
Cc: Thomas Gleixner <tg...@linutronix.de>
Cc: linux-...@vger.kernel.org
Cc: syzkaller <syzk...@googlegroups.com>
Link: http://lkml.kernel.org/r/2016100808...@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mi...@kernel.org>
---
fs/locks.c | 6 ++++++
1 file changed, 6 insertions(+)

diff --git a/fs/locks.c b/fs/locks.c
index ce93b41..22c5b4a 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -1609,6 +1609,7 @@ int fcntl_getlease(struct file *filp)

ctx = smp_load_acquire(&inode->i_flctx);
if (ctx && !list_empty_careful(&ctx->flc_lease)) {
+ percpu_down_read_preempt_disable(&file_rwsem);
spin_lock(&ctx->flc_lock);
time_out_leases(inode, &dispose);
list_for_each_entry(fl, &ctx->flc_lease, fl_list) {
@@ -1618,6 +1619,8 @@ int fcntl_getlease(struct file *filp)
break;
}
spin_unlock(&ctx->flc_lock);
+ percpu_up_read_preempt_enable(&file_rwsem);
+
locks_dispose_list(&dispose);
}
return type;
@@ -2529,11 +2532,14 @@ locks_remove_lease(struct file *filp, struct file_lock_context *ctx)
Reply all
Reply to author
Forward
0 new messages