UBSAN: shift-out-of-bounds in load_balance

syzbot has found a reproducer for the following issue on:

HEAD commit:    31caf8b2 Merge branch 'linus' of git://git.kernel.org/pub/..
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=16ab2682d00000
kernel config:  https://syzkaller.appspot.com/x/.config?x=b81388f0b32761d4
dashboard link: https://syzkaller.appspot.com/bug?extid=d7581744d5fd27c9fbe1
syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1277457f500000

IMPORTANT: if you fix the issue, please add the following tag to the commit:

================================================================================

UBSAN: shift-out-of-bounds in kernel/sched/fair.c:7712:14
shift exponent 149 is too large for 64-bit type 'long unsigned int'
CPU: 0 PID: 12 Comm: ksoftirqd/0 Not tainted 5.11.0-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
 __dump_stack lib/dump_stack.c:79 [inline]
 dump_stack+0xfa/0x151 lib/dump_stack.c:120
 ubsan_epilogue+0xb/0x5a lib/ubsan.c:148
 __ubsan_handle_shift_out_of_bounds.cold+0xb1/0x181 lib/ubsan.c:395
 detach_tasks kernel/sched/fair.c:7712 [inline]
 load_balance.cold+0x1d/0x2e kernel/sched/fair.c:9641
 rebalance_domains+0x5cc/0xdb0 kernel/sched/fair.c:10029
 __do_softirq+0x29b/0x9f6 kernel/softirq.c:343
 run_ksoftirqd kernel/softirq.c:650 [inline]
 run_ksoftirqd+0x2d/0x60 kernel/softirq.c:642
 smpboot_thread_fn+0x655/0x9e0 kernel/smpboot.c:165
 kthread+0x3b1/0x4a0 kernel/kthread.c:292
 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:294
================================================================================
Kernel panic - not syncing: panic_on_warn set ...
CPU: 0 PID: 12 Comm: ksoftirqd/0 Not tainted 5.11.0-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
 __dump_stack lib/dump_stack.c:79 [inline]
 dump_stack+0xfa/0x151 lib/dump_stack.c:120
 panic+0x306/0x73d kernel/panic.c:231
 ubsan_epilogue+0x54/0x5a lib/ubsan.c:162
 __ubsan_handle_shift_out_of_bounds.cold+0xb1/0x181 lib/ubsan.c:395
 detach_tasks kernel/sched/fair.c:7712 [inline]
 load_balance.cold+0x1d/0x2e kernel/sched/fair.c:9641
 rebalance_domains+0x5cc/0xdb0 kernel/sched/fair.c:10029
 __do_softirq+0x29b/0x9f6 kernel/softirq.c:343
 run_ksoftirqd kernel/softirq.c:650 [inline]
 run_ksoftirqd+0x2d/0x60 kernel/softirq.c:642
 smpboot_thread_fn+0x655/0x9e0 kernel/smpboot.c:165
 kthread+0x3b1/0x4a0 kernel/kthread.c:292
 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:294

======================================================

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     95d3920902d198b78d489db7e11732661ef79357
Gitweb:        https://git.kernel.org/tip/95d3920902d198b78d489db7e11732661ef79357
Author:        Valentin Schneider <[email protected]>
AuthorDate:    Thu, 25 Feb 2021 17:56:56 
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Mon, 01 Mar 2021 18:17:26 +01:00

sched/fair: Fix shift-out-of-bounds in load_balance()

Syzbot reported a handful of occurrences where an sd->nr_balance_failed can grow to much higher values than one would expect.

A successful load_balance() resets it to 0; a failed one increments it. Once it gets to sd->cache_nice_tries + 3, this *should* trigger an active balance, which will either set it to sd->cache_nice_tries+1 or reset it to 0. However, in case the to-be-active-balanced task is not allowed to run on env->dst_cpu, then the increment is done without any further modification.

This could then be repeated ad nauseam, and would explain the absurdly high values reported by syzbot (86, 149). VincentG noted there is value in letting sd->cache_nice_tries grow, so the shift itself should be fixed. That means preventing:

"""
If the value of the right operand is negative or is greater than or equal to the width of the promoted left operand, the behavior is undefined. """

Thus we need to cap the shift exponent to
BITS_PER_TYPE(typeof(lefthand)) - 1.

I had a look around for other similar cases via coccinelle:

@expr@
position pos;
expression E1;
expression E2;
@@
(
E1 >> E2@pos
|
E1 >> E2@pos
)

@cst depends on expr@
position pos;
expression expr.E1;
constant cst;
@@
(
E1 >> cst@pos
|
E1 << cst@pos
)

@script:python depends on !cst@
pos << expr.pos;
exp << expr.E2;
@@
# Dirty hack to ignore constexpr
if exp.upper() != exp:
coccilib.report.print_report(pos[0], "Possible UB shift here")

The only other match in kernel/sched is rq_clock_thermal() which employs sched_thermal_decay_shift, and that exponent is already capped to 10, so that one is fine.

---
 kernel/sched/fair.c  | 3 +--
 kernel/sched/sched.h | 7 +++++++
 2 files changed, 8 insertions(+), 2 deletions(-)

index 7b2fac0..1af51a6 100644
@@ -7722,8 +7722,7 @@ static int detach_tasks(struct lb_env *env)
 			 * scheduler fails to find a good waiting task to
 			 * migrate.
 			 */
-
-			if ((load >> env->sd->nr_balance_failed) > env->imbalance)
+			if (shr_bound(load, env->sd->nr_balance_failed) > env->imbalance)
 				goto next;
 
 			env->imbalance -= load;

index 0ddc9a6..bb8bb06 100644
@@ -205,6 +205,13 @@ static inline void update_avg(u64 *avg, u64 sample)
 }
 
 /*
+ * Shifting a value by an exponent greater *or equal* to the size of said value
+ * is UB; cap at size-1.
+ */
+#define shr_bound(val, shift)							\
+	(val >> min_t(typeof(shift), shift, BITS_PER_TYPE(typeof(val)) - 1))
+
+/*
  * !! For sched_setattr_nocheck() (kernel) only !!
  *
  * This is actually gross. :(

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     9ab8f620eea3a797667add72eae5e235d2ca2fc8
Gitweb:        https://git.kernel.org/tip/9ab8f620eea3a797667add72eae5e235d2ca2fc8
Author:        Valentin Schneider <[email protected]>
AuthorDate:    Thu, 25 Feb 2021 17:56:56 
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Wed, 03 Mar 2021 10:33:00 +01:00

sched/fair: Fix shift-out-of-bounds in load_balance()

Syzbot reported a handful of occurrences where an sd->nr_balance_failed can grow to much higher values than one would expect.

A successful load_balance() resets it to 0; a failed one increments it. Once it gets to sd->cache_nice_tries + 3, this *should* trigger an active balance, which will either set it to sd->cache_nice_tries+1 or reset it to 0. However, in case the to-be-active-balanced task is not allowed to run on env->dst_cpu, then the increment is done without any further modification.

This could then be repeated ad nauseam, and would explain the absurdly high values reported by syzbot (86, 149). VincentG noted there is value in letting sd->cache_nice_tries grow, so the shift itself should be fixed. That means preventing:

"""
If the value of the right operand is negative or is greater than or equal to the width of the promoted left operand, the behavior is undefined. """

Thus we need to cap the shift exponent to
BITS_PER_TYPE(typeof(lefthand)) - 1.

I had a look around for other similar cases via coccinelle:

@expr@
position pos;
expression E1;
expression E2;
@@
(
E1 >> E2@pos
|
E1 >> E2@pos
)

@cst depends on expr@
position pos;
expression expr.E1;
constant cst;
@@
(
E1 >> cst@pos
|
E1 << cst@pos
)

@script:python depends on !cst@
pos << expr.pos;
exp << expr.E2;
@@
# Dirty hack to ignore constexpr
if exp.upper() != exp:
coccilib.report.print_report(pos[0], "Possible UB shift here")

The only other match in kernel/sched is rq_clock_thermal() which employs sched_thermal_decay_shift, and that exponent is already capped to 10, so that one is fine.

---
 kernel/sched/fair.c  | 3 +--
 kernel/sched/sched.h | 7 +++++++
 2 files changed, 8 insertions(+), 2 deletions(-)

index 7b2fac0..1af51a6 100644
@@ -7722,8 +7722,7 @@ static int detach_tasks(struct lb_env *env)
 			 * scheduler fails to find a good waiting task to
 			 * migrate.
 			 */
-
-			if ((load >> env->sd->nr_balance_failed) > env->imbalance)
+			if (shr_bound(load, env->sd->nr_balance_failed) > env->imbalance)
 				goto next;
 
 			env->imbalance -= load;

index 0ddc9a6..bb8bb06 100644
@@ -205,6 +205,13 @@ static inline void update_avg(u64 *avg, u64 sample)
 }
 
 /*
+ * Shifting a value by an exponent greater *or equal* to the size of said value
+ * is UB; cap at size-1.
+ */
+#define shr_bound(val, shift)							\
+	(val >> min_t(typeof(shift), shift, BITS_PER_TYPE(typeof(val)) - 1))
+
+/*
  * !! For sched_setattr_nocheck() (kernel) only !!
  *
  * This is actually gross. :(

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     39a2a6eb5c9b66ea7c8055026303b3aa681b49a5
Gitweb:        https://git.kernel.org/tip/39a2a6eb5c9b66ea7c8055026303b3aa681b49a5
Author:        Valentin Schneider <[email protected]>
AuthorDate:    Thu, 25 Feb 2021 17:56:56 
Committer:     Ingo Molnar <[email protected]>
CommitterDate: Sat, 06 Mar 2021 12:40:22 +01:00

sched/fair: Fix shift-out-of-bounds in load_balance()

Syzbot reported a handful of occurrences where an sd->nr_balance_failed can grow to much higher values than one would expect.

A successful load_balance() resets it to 0; a failed one increments it. Once it gets to sd->cache_nice_tries + 3, this *should* trigger an active balance, which will either set it to sd->cache_nice_tries+1 or reset it to 0. However, in case the to-be-active-balanced task is not allowed to run on env->dst_cpu, then the increment is done without any further modification.

This could then be repeated ad nauseam, and would explain the absurdly high values reported by syzbot (86, 149). VincentG noted there is value in letting sd->cache_nice_tries grow, so the shift itself should be fixed. That means preventing:

"""
If the value of the right operand is negative or is greater than or equal to the width of the promoted left operand, the behavior is undefined. """

Thus we need to cap the shift exponent to
BITS_PER_TYPE(typeof(lefthand)) - 1.

I had a look around for other similar cases via coccinelle:

@expr@
position pos;
expression E1;
expression E2;
@@
(
E1 >> E2@pos
|
E1 >> E2@pos
)

@cst depends on expr@
position pos;
expression expr.E1;
constant cst;
@@
(
E1 >> cst@pos
|
E1 << cst@pos
)

@script:python depends on !cst@
pos << expr.pos;
exp << expr.E2;
@@
# Dirty hack to ignore constexpr
if exp.upper() != exp:
coccilib.report.print_report(pos[0], "Possible UB shift here")

The only other match in kernel/sched is rq_clock_thermal() which employs sched_thermal_decay_shift, and that exponent is already capped to 10, so that one is fine.

---
 kernel/sched/fair.c  | 3 +--
 kernel/sched/sched.h | 7 +++++++
 2 files changed, 8 insertions(+), 2 deletions(-)

index 7b2fac0..1af51a6 100644
@@ -7722,8 +7722,7 @@ static int detach_tasks(struct lb_env *env)
 			 * scheduler fails to find a good waiting task to
 			 * migrate.
 			 */
-
-			if ((load >> env->sd->nr_balance_failed) > env->imbalance)
+			if (shr_bound(load, env->sd->nr_balance_failed) > env->imbalance)
 				goto next;
 
 			env->imbalance -= load;

index 0ddc9a6..bb8bb06 100644
@@ -205,6 +205,13 @@ static inline void update_avg(u64 *avg, u64 sample)
 }
 
 /*
+ * Shifting a value by an exponent greater *or equal* to the size of said value
+ * is UB; cap at size-1.
+ */
+#define shr_bound(val, shift)							\
+	(val >> min_t(typeof(shift), shift, BITS_PER_TYPE(typeof(val)) - 1))
+
+/*
  * !! For sched_setattr_nocheck() (kernel) only !!
  *
  * This is actually gross. :(