KASAN: use-after-free Read in __hrtimer_run_queues

KASAN: use-after-free Read in __hrtimer_run_queues

[syzbot]

Hello,

syzbot found the following crash on:

HEAD commit:    673b41e0 staging/octeon: fix up merge error
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=1488940be00000
kernel config:  https://syzkaller.appspot.com/x/.config?x=3e9c99d33a5ab36f
dashboard link: https://syzkaller.appspot.com/bug?extid=62c155c276e580cfb606
compiler:       gcc (GCC) 9.0.0 20181231 (experimental)
userspace arch: i386
syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=158ed16de00000
C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=12b41b9de00000

The bug was bisected to:

commit 40db173965c05a1d803451240ed41707d5bd978d
Author: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date:   Sat Mar 21 11:26:02 2020 +0000

    lockdep: Add hrtimer context tracing bits

bisection log:  https://syzkaller.appspot.com/x/bisect.txt?x=15b40697e00000
final crash:    https://syzkaller.appspot.com/x/report.txt?x=17b40697e00000
console output: https://syzkaller.appspot.com/x/log.txt?x=13b40697e00000

IMPORTANT: if you fix the bug, please add the following tag to the commit:
Reported-by: syzbot+62c155c276e580cfb606@syzkaller.appspotmail.com
Fixes: 40db173965c0 ("lockdep: Add hrtimer context tracing bits")

vxcan0: j1939_tp_rxtimer: 0x00000000008c9d66: rx timeout, send abort
vxcan0: j1939_tp_rxtimer: 0x00000000008c9d66: abort rx timeout. Force session deactivation
==================================================================
BUG: KASAN: use-after-free in __run_hrtimer kernel/time/hrtimer.c:1521 [inline]
BUG: KASAN: use-after-free in __hrtimer_run_queues+0xe18/0xf10 kernel/time/hrtimer.c:1583
Read of size 1 at addr ffff8880934d1d73 by task swapper/1/0

CPU: 1 PID: 0 Comm: swapper/1 Not tainted 5.6.0-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
 <IRQ>
 __dump_stack lib/dump_stack.c:77 [inline]
 dump_stack+0x188/0x20d lib/dump_stack.c:118
 print_address_description.constprop.0.cold+0xd3/0x315 mm/kasan/report.c:374
 __kasan_report.cold+0x1a/0x32 mm/kasan/report.c:506
 kasan_report+0xe/0x20 mm/kasan/common.c:641
 __run_hrtimer kernel/time/hrtimer.c:1521 [inline]
 __hrtimer_run_queues+0xe18/0xf10 kernel/time/hrtimer.c:1583
 hrtimer_run_softirq+0x16d/0x250 kernel/time/hrtimer.c:1600
 __do_softirq+0x26c/0x9f7 kernel/softirq.c:292
 invoke_softirq kernel/softirq.c:373 [inline]
 irq_exit+0x192/0x1d0 kernel/softirq.c:413
 exiting_irq arch/x86/include/asm/apic.h:546 [inline]
 smp_apic_timer_interrupt+0x19e/0x600 arch/x86/kernel/apic/apic.c:1140
 apic_timer_interrupt+0xf/0x20 arch/x86/entry/entry_64.S:829
 </IRQ>
RIP: 0010:native_safe_halt+0xe/0x10 arch/x86/include/asm/irqflags.h:61
Code: cc cc cc cc cc cc cc cc cc cc cc cc e9 07 00 00 00 0f 00 2d 94 59 4c 00 f4 c3 66 90 e9 07 00 00 00 0f 00 2d 84 59 4c 00 fb f4 <c3> cc 41 56 41 55 41 54 55 53 e8 63 8d a4 f9 e8 9e 9f d7 fb 0f 1f
RSP: 0018:ffffc90000d3fdb8 EFLAGS: 00000282 ORIG_RAX: ffffffffffffff13
RAX: 1ffffffff12e90e7 RBX: ffff8880a9638340 RCX: 0000000000000000
RDX: dffffc0000000000 RSI: 0000000000000006 RDI: ffff8880a9638bfc
RBP: dffffc0000000000 R08: ffff8880a9638340 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: ffffed10152c7068
R13: 0000000000000001 R14: ffffffff8a673000 R15: 0000000000000000
 arch_safe_halt arch/x86/include/asm/paravirt.h:144 [inline]
 default_idle+0x49/0x350 arch/x86/kernel/process.c:695
 cpuidle_idle_call kernel/sched/idle.c:154 [inline]
 do_idle+0x393/0x690 kernel/sched/idle.c:269
 cpu_startup_entry+0x14/0x20 kernel/sched/idle.c:361
 start_secondary+0x2f3/0x400 arch/x86/kernel/smpboot.c:268
 secondary_startup_64+0xa4/0xb0 arch/x86/kernel/head_64.S:242

Allocated by task 7030:
 save_stack+0x1b/0x80 mm/kasan/common.c:72
 set_track mm/kasan/common.c:80 [inline]
 __kasan_kmalloc mm/kasan/common.c:515 [inline]
 __kasan_kmalloc.constprop.0+0xbf/0xd0 mm/kasan/common.c:488
 kmem_cache_alloc_trace+0x153/0x7d0 mm/slab.c:3551
 kmalloc include/linux/slab.h:555 [inline]
 kzalloc include/linux/slab.h:669 [inline]
 j1939_session_new+0x7c/0x3f0 net/can/j1939/transport.c:1418
 j1939_tp_send+0x22f/0x800 net/can/j1939/transport.c:1877
 j1939_sk_send_loop net/can/j1939/socket.c:1037 [inline]
 j1939_sk_sendmsg+0xabf/0x1360 net/can/j1939/socket.c:1160
 sock_sendmsg_nosec net/socket.c:652 [inline]
 sock_sendmsg+0xcf/0x120 net/socket.c:672
 ____sys_sendmsg+0x6bf/0x7e0 net/socket.c:2362
 ___sys_sendmsg+0x100/0x170 net/socket.c:2416
 __sys_sendmsg+0xec/0x1b0 net/socket.c:2449
 do_syscall_32_irqs_on arch/x86/entry/common.c:337 [inline]
 do_fast_syscall_32+0x270/0xe8f arch/x86/entry/common.c:408
 entry_SYSENTER_compat+0x70/0x7f arch/x86/entry/entry_64_compat.S:139

Freed by task 0:
 save_stack+0x1b/0x80 mm/kasan/common.c:72
 set_track mm/kasan/common.c:80 [inline]
 kasan_set_free_info mm/kasan/common.c:337 [inline]
 __kasan_slab_free+0xf7/0x140 mm/kasan/common.c:476
 __cache_free mm/slab.c:3426 [inline]
 kfree+0x109/0x2b0 mm/slab.c:3757
 j1939_session_destroy net/can/j1939/transport.c:276 [inline]
 __j1939_session_release net/can/j1939/transport.c:284 [inline]
 kref_put include/linux/kref.h:65 [inline]
 j1939_session_put+0x25c/0x330 net/can/j1939/transport.c:289
 j1939_tp_rxtimer+0x2e9/0x2f4 net/can/j1939/transport.c:1194
 __run_hrtimer kernel/time/hrtimer.c:1519 [inline]
 __hrtimer_run_queues+0x3a2/0xf10 kernel/time/hrtimer.c:1583
 hrtimer_run_softirq+0x16d/0x250 kernel/time/hrtimer.c:1600
 __do_softirq+0x26c/0x9f7 kernel/softirq.c:292

The buggy address belongs to the object at ffff8880934d1c00
 which belongs to the cache kmalloc-512 of size 512
The buggy address is located 371 bytes inside of
 512-byte region [ffff8880934d1c00, ffff8880934d1e00)
The buggy address belongs to the page:
page:ffffea00024d3440 refcount:1 mapcount:0 mapping:ffff8880aa000a80 index:0x0
flags: 0xfffe0000000200(slab)
raw: 00fffe0000000200 ffffea00024fbd48 ffffea00024fb888 ffff8880aa000a80
raw: 0000000000000000 ffff8880934d1000 0000000100000004 0000000000000000
page dumped because: kasan: bad access detected

Memory state around the buggy address:
 ffff8880934d1c00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff8880934d1c80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff8880934d1d00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                                                             ^
 ffff8880934d1d80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff8880934d1e00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
==================================================================


---
This bug is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzkaller@googlegroups.com.

syzbot will keep track of this bug report. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
For information about bisection process see: https://goo.gl/tpsmEJ#bisection
syzbot can test patches for this bug, for details see:
https://goo.gl/tpsmEJ#testing-patches

[PATCH] lockdep: Don't access the hrtimer after the callback

[Sebastian Andrzej Siewior]

A hrtimer can be released in its callback as reported by syzbot.

Retrieve the context in which the hrtimer expires before its callback is
invoked and use it in lockdep_hrtimer_exit().

Reported-by: syzbot+62c155c276e580cfb606@syzkaller.appspotmail.com
Fixes: 40db173965c0 ("lockdep: Add hrtimer context tracing bits")
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
 include/linux/irqflags.h | 22 +++++++++++++---------
 kernel/time/hrtimer.c    |  5 +++--
 2 files changed, 16 insertions(+), 11 deletions(-)

diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h
index a16adbb58f66a..808d10de852f9 100644
--- a/include/linux/irqflags.h
+++ b/include/linux/irqflags.h
@@ -58,16 +58,20 @@ do {						\
 } while (0)
 
 # define lockdep_hrtimer_enter(__hrtimer)		\
-	  do {						\
-		  if (!__hrtimer->is_hard)		\
-			current->irq_config = 1;	\
-	  } while (0)
+({							\
+	bool __expires_hardirq = true;			\
+	if (!__hrtimer->is_hard) {			\
+		current->irq_config = 1;		\
+		__expires_hardirq = false;		\
+	}						\
+	__expires_hardirq;				\
+})
 
-# define lockdep_hrtimer_exit(__hrtimer)		\
-	  do {						\
-		  if (!__hrtimer->is_hard)		\
+# define lockdep_hrtimer_exit(__expires_hardirq)	\
+	do {						\
+		if (__expires_hardirq == false)	\
 			current->irq_config = 0;	\
-	  } while (0)
+	} while (0)
 
 # define lockdep_posixtimer_enter()				\
 	  do {							\
@@ -102,7 +106,7 @@ do {						\
 # define trace_hardirq_exit()		do { } while (0)
 # define lockdep_softirq_enter()	do { } while (0)
 # define lockdep_softirq_exit()		do { } while (0)
-# define lockdep_hrtimer_enter(__hrtimer)		do { } while (0)
+# define lockdep_hrtimer_enter(__hrtimer)		false
 # define lockdep_hrtimer_exit(__hrtimer)		do { } while (0)
 # define lockdep_posixtimer_enter()		do { } while (0)
 # define lockdep_posixtimer_exit()		do { } while (0)
diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c
index 8cce72501aea5..817a9c7be47e3 100644
--- a/kernel/time/hrtimer.c
+++ b/kernel/time/hrtimer.c
@@ -1481,6 +1481,7 @@ static void __run_hrtimer(struct hrtimer_cpu_base *cpu_base,
 {
 	enum hrtimer_restart (*fn)(struct hrtimer *);
 	int restart;
+	bool expires_in_hardirq;
 
 	lockdep_assert_held(&cpu_base->lock);
 
@@ -1514,11 +1515,11 @@ static void __run_hrtimer(struct hrtimer_cpu_base *cpu_base,
 	 */
 	raw_spin_unlock_irqrestore(&cpu_base->lock, flags);
 	trace_hrtimer_expire_entry(timer, now);
-	lockdep_hrtimer_enter(timer);
+	expires_in_hardirq = lockdep_hrtimer_enter(timer);
 
 	restart = fn(timer);
 
-	lockdep_hrtimer_exit(timer);
+	lockdep_hrtimer_exit(expires_in_hardirq);
 	trace_hrtimer_expire_exit(timer);
 	raw_spin_lock_irq(&cpu_base->lock);
 
-- 
2.26.0

[tip: timers/urgent] hrtimer: Don't dereference the hrtimer pointer after the callback

[tip-bot2 for Sebastian Andrzej Siewior]

The following commit has been merged into the timers/urgent branch of tip:

Commit-ID:     73d20564e0dcae003e0d79977f044d5e57496304
Gitweb:        https://git.kernel.org/tip/73d20564e0dcae003e0d79977f044d5e57496304
Author:        Sebastian Andrzej Siewior <bigeasy@linutronix.de>
AuthorDate:    Tue, 31 Mar 2020 22:18:49 +02:00
Committer:     Thomas Gleixner <tglx@linutronix.de>
CommitterDate: Wed, 01 Apr 2020 13:20:14 +02:00

hrtimer: Don't dereference the hrtimer pointer after the callback

A hrtimer can be released in its callback, but lockdep_hrtimer_exit()
dereferences the pointer after the callback returns, i.e. a potential use
after free.

Retrieve the context in which the hrtimer expires before the callback is
invoked and use it in lockdep_hrtimer_exit().

Fixes: 40db173965c0 ("lockdep: Add hrtimer context tracing bits")
Reported-by: syzbot+62c155c276e580cfb606@syzkaller.appspotmail.com
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20200331201849.fkp2siy3vcdqvqlz@linutronix.de
---
 include/linux/irqflags.h | 27 ++++++++++++++++-----------
 kernel/time/hrtimer.c    |  5 +++--
 2 files changed, 19 insertions(+), 13 deletions(-)

diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h
index ceca42d..61a9ced 100644
--- a/include/linux/irqflags.h
+++ b/include/linux/irqflags.h
@@ -58,16 +58,21 @@ do {						\
 } while (0)
 
 # define lockdep_hrtimer_enter(__hrtimer)		\
-	  do {						\
-		  if (!__hrtimer->is_hard)		\
-			current->irq_config = 1;	\
-	  } while (0)
-
-# define lockdep_hrtimer_exit(__hrtimer)		\
-	  do {						\
-		  if (!__hrtimer->is_hard)		\
+({							\
+	bool __expires_hardirq = true;			\
+							\
+	if (!__hrtimer->is_hard) {			\
+		current->irq_config = 1;		\
+		__expires_hardirq = false;		\
+	}						\
+	__expires_hardirq;				\
+})
+
+# define lockdep_hrtimer_exit(__expires_hardirq)	\
+	do {						\
+		if (!__expires_hardirq)			\
 			current->irq_config = 0;	\
-	  } while (0)
+	} while (0)
 
 # define lockdep_posixtimer_enter()				\
 	  do {							\
@@ -102,8 +107,8 @@ do {						\
 # define lockdep_hardirq_exit()		do { } while (0)
 # define lockdep_softirq_enter()	do { } while (0)
 # define lockdep_softirq_exit()		do { } while (0)
-# define lockdep_hrtimer_enter(__hrtimer)		do { } while (0)
-# define lockdep_hrtimer_exit(__hrtimer)		do { } while (0)
+# define lockdep_hrtimer_enter(__hrtimer)	false
+# define lockdep_hrtimer_exit(__context)	do { } while (0)
 # define lockdep_posixtimer_enter()		do { } while (0)
 # define lockdep_posixtimer_exit()		do { } while (0)
 # define lockdep_irq_work_enter(__work)		do { } while (0)
diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c
index d0a5ba3..d89da1c 100644
--- a/kernel/time/hrtimer.c
+++ b/kernel/time/hrtimer.c
@@ -1480,6 +1480,7 @@ static void __run_hrtimer(struct hrtimer_cpu_base *cpu_base,
 			  unsigned long flags) __must_hold(&cpu_base->lock)
 {
 	enum hrtimer_restart (*fn)(struct hrtimer *);
+	bool expires_in_hardirq;
 	int restart;
 
 	lockdep_assert_held(&cpu_base->lock);
@@ -1514,11 +1515,11 @@ static void __run_hrtimer(struct hrtimer_cpu_base *cpu_base,
 	 */
 	raw_spin_unlock_irqrestore(&cpu_base->lock, flags);
 	trace_hrtimer_expire_entry(timer, now);
-	lockdep_hrtimer_enter(timer);
+	expires_in_hardirq = lockdep_hrtimer_enter(timer);
 
 	restart = fn(timer);
 
-	lockdep_hrtimer_exit(timer);
+	lockdep_hrtimer_exit(expires_in_hardirq);
 	trace_hrtimer_expire_exit(timer);
 	raw_spin_lock_irq(&cpu_base->lock);