[PATCHv2 bpf 1/2] bpf: Disable preemption in bpf_perf_event_output

[PATCHv2 bpf 1/2] bpf: Disable preemption in bpf_perf_event_output

[Jiri Olsa]

The nesting protection in bpf_perf_event_output relies on disabled
preemption, which is guaranteed for kprobes and tracepoints.

However bpf_perf_event_output can be also called from uprobes context
through bpf_prog_run_array_sleepable function which disables migration,
but keeps preemption enabled.

This can cause task to be preempted by another one inside the nesting
protection and lead eventually to two tasks using same perf_sample_data
buffer and cause crashes like:

  kernel tried to execute NX-protected page - exploit attempt? (uid: 0)
  BUG: unable to handle page fault for address: ffffffff82be3eea
  ...
  Call Trace:
   ? __die+0x1f/0x70
   ? page_fault_oops+0x176/0x4d0
   ? exc_page_fault+0x132/0x230
   ? asm_exc_page_fault+0x22/0x30
   ? perf_output_sample+0x12b/0x910
   ? perf_event_output+0xd0/0x1d0
   ? bpf_perf_event_output+0x162/0x1d0
   ? bpf_prog_c6271286d9a4c938_krava1+0x76/0x87
   ? __uprobe_perf_func+0x12b/0x540
   ? uprobe_dispatcher+0x2c4/0x430
   ? uprobe_notify_resume+0x2da/0xce0
   ? atomic_notifier_call_chain+0x7b/0x110
   ? exit_to_user_mode_prepare+0x13e/0x290
   ? irqentry_exit_to_user_mode+0x5/0x30
   ? asm_exc_int3+0x35/0x40

Fixing this by disabling preemption in bpf_perf_event_output.

Cc: stable@vger.kernel.org
Fixes: 8c7dcb84e3b7 ("bpf: implement sleepable uprobes by chaining gps")
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
---
 kernel/trace/bpf_trace.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index 5f2dcabad202..14c9a1a548c9 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -661,8 +661,7 @@ static DEFINE_PER_CPU(int, bpf_trace_nest_level);
 BPF_CALL_5(bpf_perf_event_output, struct pt_regs *, regs, struct bpf_map *, map,
 	   u64, flags, void *, data, u64, size)
 {
-	struct bpf_trace_sample_data *sds = this_cpu_ptr(&bpf_trace_sds);
-	int nest_level = this_cpu_inc_return(bpf_trace_nest_level);
+	struct bpf_trace_sample_data *sds;
 	struct perf_raw_record raw = {
 		.frag = {
 			.size = size,
@@ -670,7 +669,12 @@ BPF_CALL_5(bpf_perf_event_output, struct pt_regs *, regs, struct bpf_map *, map,
 		},
 	};
 	struct perf_sample_data *sd;
-	int err;
+	int nest_level, err;
+
+	preempt_disable();
+
+	sds = this_cpu_ptr(&bpf_trace_sds);
+	nest_level = this_cpu_inc_return(bpf_trace_nest_level);
 
 	if (WARN_ON_ONCE(nest_level > ARRAY_SIZE(sds->sds))) {
 		err = -EBUSY;
@@ -691,6 +695,7 @@ BPF_CALL_5(bpf_perf_event_output, struct pt_regs *, regs, struct bpf_map *, map,
 
 out:
 	this_cpu_dec(bpf_trace_nest_level);
+	preempt_enable();
 	return err;
 }
 
-- 
2.41.0

[PATCHv2 bpf 2/2] bpf: Disable preemption in bpf_event_output

[Jiri Olsa]

We received report [1] of kernel crash, which is caused by
using nesting protection without disabled preemption.

The bpf_event_output can be called by programs executed by
bpf_prog_run_array_cg function that disabled migration but
keeps preemption enabled.

This can cause task to be preempted by another one inside the
nesting protection and lead eventually to two tasks using same
perf_sample_data buffer and cause crashes like:

  BUG: kernel NULL pointer dereference, address: 0000000000000001
  #PF: supervisor instruction fetch in kernel mode
  #PF: error_code(0x0010) - not-present page
  ...
  ? perf_output_sample+0x12a/0x9a0
  ? finish_task_switch.isra.0+0x81/0x280
  ? perf_event_output+0x66/0xa0
  ? bpf_event_output+0x13a/0x190
  ? bpf_event_output_data+0x22/0x40
  ? bpf_prog_dfc84bbde731b257_cil_sock4_connect+0x40a/0xacb
  ? xa_load+0x87/0xe0
  ? __cgroup_bpf_run_filter_sock_addr+0xc1/0x1a0
  ? release_sock+0x3e/0x90
  ? sk_setsockopt+0x1a1/0x12f0
  ? udp_pre_connect+0x36/0x50
  ? inet_dgram_connect+0x93/0xa0
  ? __sys_connect+0xb4/0xe0
  ? udp_setsockopt+0x27/0x40
  ? __pfx_udp_push_pending_frames+0x10/0x10
  ? __sys_setsockopt+0xdf/0x1a0
  ? __x64_sys_connect+0xf/0x20
  ? do_syscall_64+0x3a/0x90
  ? entry_SYSCALL_64_after_hwframe+0x72/0xdc

Fixing this by disabling preemption in bpf_event_output.

[1] https://github.com/cilium/cilium/issues/26756
Cc: stable@vger.kernel.org
Reported-by:  Oleg "livelace" Popov <o.popov@livelace.ru>
Fixes: 2a916f2f546c bpf: Use migrate_disable/enable in array macros and cgroup/lirc code.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
---
 kernel/trace/bpf_trace.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index 14c9a1a548c9..6826ebf750b0 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -720,7 +720,6 @@ static DEFINE_PER_CPU(struct bpf_trace_sample_data, bpf_misc_sds);
 u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size,
 		     void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy)
 {
-	int nest_level = this_cpu_inc_return(bpf_event_output_nest_level);
 	struct perf_raw_frag frag = {
 		.copy		= ctx_copy,
 		.size		= ctx_size,
@@ -737,8 +736,13 @@ u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size,
 	};
 	struct perf_sample_data *sd;
 	struct pt_regs *regs;
+	int nest_level;
 	u64 ret;
 
+	preempt_disable();
+
+	nest_level = this_cpu_inc_return(bpf_event_output_nest_level);
+
 	if (WARN_ON_ONCE(nest_level > ARRAY_SIZE(bpf_misc_sds.sds))) {
 		ret = -EBUSY;
 		goto out;
@@ -753,6 +757,7 @@ u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size,
 	ret = __bpf_perf_event_output(regs, map, flags, sd);
 out:
 	this_cpu_dec(bpf_event_output_nest_level);
+	preempt_enable();
 	return ret;
 }
 
-- 
2.41.0

Re: [PATCHv2 bpf 1/2] bpf: Disable preemption in bpf_perf_event_output

[Hou Tao]

On 7/20/2023 4:57 PM, Jiri Olsa wrote:
> The nesting protection in bpf_perf_event_output relies on disabled
> preemption, which is guaranteed for kprobes and tracepoints.
>
> However bpf_perf_event_output can be also called from uprobes context
> through bpf_prog_run_array_sleepable function which disables migration,
> but keeps preemption enabled.
>
> This can cause task to be preempted by another one inside the nesting
> protection and lead eventually to two tasks using same perf_sample_data
> buffer and cause crashes like:
>
>   kernel tried to execute NX-protected page - exploit attempt? (uid: 0)
>   BUG: unable to handle page fault for address: ffffffff82be3eea
>   ...
>   Call Trace:
>    ? __die+0x1f/0x70
>    ? page_fault_oops+0x176/0x4d0
>    ? exc_page_fault+0x132/0x230
>    ? asm_exc_page_fault+0x22/0x30
>    ? perf_output_sample+0x12b/0x910
>    ? perf_event_output+0xd0/0x1d0
>    ? bpf_perf_event_output+0x162/0x1d0
>    ? bpf_prog_c6271286d9a4c938_krava1+0x76/0x87
>    ? __uprobe_perf_func+0x12b/0x540
>    ? uprobe_dispatcher+0x2c4/0x430
>    ? uprobe_notify_resume+0x2da/0xce0
>    ? atomic_notifier_call_chain+0x7b/0x110
>    ? exit_to_user_mode_prepare+0x13e/0x290
>    ? irqentry_exit_to_user_mode+0x5/0x30
>    ? asm_exc_int3+0x35/0x40
>
> Fixing this by disabling preemption in bpf_perf_event_output.
>
> Cc: stable@vger.kernel.org
> Fixes: 8c7dcb84e3b7 ("bpf: implement sleepable uprobes by chaining gps")
> Signed-off-by: Jiri Olsa <jolsa@kernel.org>

Acked-by: Hou Tao <houtao1@huawei.com>

Re: [PATCHv2 bpf 2/2] bpf: Disable preemption in bpf_event_output

[Hou Tao]

On 7/20/2023 4:57 PM, Jiri Olsa wrote:
> We received report [1] of kernel crash, which is caused by
> using nesting protection without disabled preemption.
>
> The bpf_event_output can be called by programs executed by
> bpf_prog_run_array_cg function that disabled migration but
> keeps preemption enabled.
>
> This can cause task to be preempted by another one inside the
> nesting protection and lead eventually to two tasks using same
> perf_sample_data buffer and cause crashes like:
>
>   BUG: kernel NULL pointer dereference, address: 0000000000000001
>   #PF: supervisor instruction fetch in kernel mode
>   #PF: error_code(0x0010) - not-present page
>   ...
>   ? perf_output_sample+0x12a/0x9a0
>   ? finish_task_switch.isra.0+0x81/0x280
>   ? perf_event_output+0x66/0xa0
>   ? bpf_event_output+0x13a/0x190
>   ? bpf_event_output_data+0x22/0x40
>   ? bpf_prog_dfc84bbde731b257_cil_sock4_connect+0x40a/0xacb
>   ? xa_load+0x87/0xe0
>   ? __cgroup_bpf_run_filter_sock_addr+0xc1/0x1a0
>   ? release_sock+0x3e/0x90
>   ? sk_setsockopt+0x1a1/0x12f0
>   ? udp_pre_connect+0x36/0x50
>   ? inet_dgram_connect+0x93/0xa0
>   ? __sys_connect+0xb4/0xe0
>   ? udp_setsockopt+0x27/0x40
>   ? __pfx_udp_push_pending_frames+0x10/0x10
>   ? __sys_setsockopt+0xdf/0x1a0
>   ? __x64_sys_connect+0xf/0x20
>   ? do_syscall_64+0x3a/0x90
>   ? entry_SYSCALL_64_after_hwframe+0x72/0xdc
>
> Fixing this by disabling preemption in bpf_event_output.
>
> [1] https://github.com/cilium/cilium/issues/26756
> Cc: stable@vger.kernel.org
> Reported-by:  Oleg "livelace" Popov <o.popov@livelace.ru>
> Fixes: 2a916f2f546c bpf: Use migrate_disable/enable in array macros and cgroup/lirc code.
> Signed-off-by: Jiri Olsa <jolsa@kernel.org>

Acked-by: Hou Tao <houtao1@huawei.com>

With one nit above. The format of the Fixes tags should be 2a916f2f546c
("bpf: Use migrate_disable/enable in array macros and cgroup/lirc code.")

Re: [PATCHv2 bpf 2/2] bpf: Disable preemption in bpf_event_output

[Jiri Olsa]

On Fri, Jul 21, 2023 at 08:16:14PM +0800, Hou Tao wrote:
> 
> 
> On 7/20/2023 4:57 PM, Jiri Olsa wrote:
> > We received report [1] of kernel crash, which is caused by
> > using nesting protection without disabled preemption.
> >
> > The bpf_event_output can be called by programs executed by
> > bpf_prog_run_array_cg function that disabled migration but
> > keeps preemption enabled.
> >
> > This can cause task to be preempted by another one inside the
> > nesting protection and lead eventually to two tasks using same
> > perf_sample_data buffer and cause crashes like:
> >
> >   BUG: kernel NULL pointer dereference, address: 0000000000000001
> >   #PF: supervisor instruction fetch in kernel mode
> >   #PF: error_code(0x0010) - not-present page
> >   ...
> >   ? perf_output_sample+0x12a/0x9a0
> >   ? finish_task_switch.isra.0+0x81/0x280
> >   ? perf_event_output+0x66/0xa0
> >   ? bpf_event_output+0x13a/0x190
> >   ? bpf_event_output_data+0x22/0x40
> >   ? bpf_prog_dfc84bbde731b257_cil_sock4_connect+0x40a/0xacb
> >   ? xa_load+0x87/0xe0
> >   ? __cgroup_bpf_run_filter_sock_addr+0xc1/0x1a0
> >   ? release_sock+0x3e/0x90
> >   ? sk_setsockopt+0x1a1/0x12f0
> >   ? udp_pre_connect+0x36/0x50
> >   ? inet_dgram_connect+0x93/0xa0
> >   ? __sys_connect+0xb4/0xe0
> >   ? udp_setsockopt+0x27/0x40
> >   ? __pfx_udp_push_pending_frames+0x10/0x10
> >   ? __sys_setsockopt+0xdf/0x1a0
> >   ? __x64_sys_connect+0xf/0x20
> >   ? do_syscall_64+0x3a/0x90
> >   ? entry_SYSCALL_64_after_hwframe+0x72/0xdc
> >
> > Fixing this by disabling preemption in bpf_event_output.
> >
> > [1] https://github.com/cilium/cilium/issues/26756
> > Cc: stable@vger.kernel.org
> > Reported-by:  Oleg "livelace" Popov <o.popov@livelace.ru>
> > Fixes: 2a916f2f546c bpf: Use migrate_disable/enable in array macros and cgroup/lirc code.
> > Signed-off-by: Jiri Olsa <jolsa@kernel.org>
> 
> Acked-by: Hou Tao <houtao1@huawei.com>
> 
> With one nit above. The format of the Fixes tags should be 2a916f2f546c
> ("bpf: Use migrate_disable/enable in array macros and cgroup/lirc code.")
> 

right, sorry about that.. should I resend?

thanks,
jirka

Re: [PATCHv2 bpf 2/2] bpf: Disable preemption in bpf_event_output

[Hou Tao]

On 7/21/2023 8:45 PM, Jiri Olsa wrote:
> On Fri, Jul 21, 2023 at 08:16:14PM +0800, Hou Tao wrote:
>>
>> On 7/20/2023 4:57 PM, Jiri Olsa wrote:
>>> We received report [1] of kernel crash, which is caused by
>>> using nesting protection without disabled preemption.
>>>
>>> The bpf_event_output can be called by programs executed by
>>> bpf_prog_run_array_cg function that disabled migration but
>>> keeps preemption enabled.
>>>
>>> This can cause task to be preempted by another one inside the
>>> nesting protection and lead eventually to two tasks using same
>>> perf_sample_data buffer and cause crashes like:
>>>
>>>   BUG: kernel NULL pointer dereference, address: 0000000000000001
>>>   #PF: supervisor instruction fetch in kernel mode
>>>   #PF: error_code(0x0010) - not-present page
>>>   ...
>>>   ? perf_output_sample+0x12a/0x9a0
>>>   ? finish_task_switch.isra.0+0x81/0x280
>>>   ? perf_event_output+0x66/0xa0
>>>   ? bpf_event_output+0x13a/0x190
>>>   ? bpf_event_output_data+0x22/0x40
>>>   ? bpf_prog_dfc84bbde731b257_cil_sock4_connect+0x40a/0xacb
>>>   ? xa_load+0x87/0xe0
>>>   ? __cgroup_bpf_run_filter_sock_addr+0xc1/0x1a0
>>>   ? release_sock+0x3e/0x90
>>>   ? sk_setsockopt+0x1a1/0x12f0
>>>   ? udp_pre_connect+0x36/0x50
>>>   ? inet_dgram_connect+0x93/0xa0
>>>   ? __sys_connect+0xb4/0xe0
>>>   ? udp_setsockopt+0x27/0x40
>>>   ? __pfx_udp_push_pending_frames+0x10/0x10
>>>   ? __sys_setsockopt+0xdf/0x1a0
>>>   ? __x64_sys_connect+0xf/0x20
>>>   ? do_syscall_64+0x3a/0x90
>>>   ? entry_SYSCALL_64_after_hwframe+0x72/0xdc
>>>
>>> Fixing this by disabling preemption in bpf_event_output.
>>>
>>> [1] https://github.com/cilium/cilium/issues/26756
>>> Cc: stable@vger.kernel.org
>>> Reported-by:  Oleg "livelace" Popov <o.popov@livelace.ru>
>>> Fixes: 2a916f2f546c bpf: Use migrate_disable/enable in array macros and cgroup/lirc code.
>>> Signed-off-by: Jiri Olsa <jolsa@kernel.org>
>> Acked-by: Hou Tao <houtao1@huawei.com>
>>
>> With one nit above. The format of the Fixes tags should be 2a916f2f546c
>> ("bpf: Use migrate_disable/enable in array macros and cgroup/lirc code.")
>>
> right, sorry about that.. should I resend?

We can wait for the comments from Alexei. Maybe the maintainer can fix
it for you.
>
> thanks,
> jirka

Re: [PATCHv2 bpf 2/2] bpf: Disable preemption in bpf_event_output

[Alexei Starovoitov]

On Fri, Jul 21, 2023 at 5:45 AM Jiri Olsa <olsajiri@gmail.com> wrote:
>
> On Fri, Jul 21, 2023 at 08:16:14PM +0800, Hou Tao wrote:
> >
> >
> > On 7/20/2023 4:57 PM, Jiri Olsa wrote:
> > > We received report [1] of kernel crash, which is caused by
> > > using nesting protection without disabled preemption.
> > >
> > > The bpf_event_output can be called by programs executed by
> > > bpf_prog_run_array_cg function that disabled migration but
> > > keeps preemption enabled.
> > >
> > > This can cause task to be preempted by another one inside the
> > > nesting protection and lead eventually to two tasks using same
> > > perf_sample_data buffer and cause crashes like:
> > >
> > >   BUG: kernel NULL pointer dereference, address: 0000000000000001
> > >   #PF: supervisor instruction fetch in kernel mode
> > >   #PF: error_code(0x0010) - not-present page
> > >   ...
> > >   ? perf_output_sample+0x12a/0x9a0
> > >   ? finish_task_switch.isra.0+0x81/0x280
> > >   ? perf_event_output+0x66/0xa0
> > >   ? bpf_event_output+0x13a/0x190
> > >   ? bpf_event_output_data+0x22/0x40
> > >   ? bpf_prog_dfc84bbde731b257_cil_sock4_connect+0x40a/0xacb
> > >   ? xa_load+0x87/0xe0
> > >   ? __cgroup_bpf_run_filter_sock_addr+0xc1/0x1a0
> > >   ? release_sock+0x3e/0x90
> > >   ? sk_setsockopt+0x1a1/0x12f0
> > >   ? udp_pre_connect+0x36/0x50
> > >   ? inet_dgram_connect+0x93/0xa0
> > >   ? __sys_connect+0xb4/0xe0
> > >   ? udp_setsockopt+0x27/0x40
> > >   ? __pfx_udp_push_pending_frames+0x10/0x10
> > >   ? __sys_setsockopt+0xdf/0x1a0
> > >   ? __x64_sys_connect+0xf/0x20
> > >   ? do_syscall_64+0x3a/0x90
> > >   ? entry_SYSCALL_64_after_hwframe+0x72/0xdc
> > >
> > > Fixing this by disabling preemption in bpf_event_output.
> > >
> > > [1] https://github.com/cilium/cilium/issues/26756
> > > Cc: stable@vger.kernel.org
> > > Reported-by:  Oleg "livelace" Popov <o.popov@livelace.ru>
> > > Fixes: 2a916f2f546c bpf: Use migrate_disable/enable in array macros and cgroup/lirc code.
> > > Signed-off-by: Jiri Olsa <jolsa@kernel.org>
> >
> > Acked-by: Hou Tao <houtao1@huawei.com>
> >
> > With one nit above. The format of the Fixes tags should be 2a916f2f546c
> > ("bpf: Use migrate_disable/enable in array macros and cgroup/lirc code.")
> >
>
> right, sorry about that.. should I resend?

Please resend with fixes and acks.