* The same symbol is printed twice when use tracepoint to get stack
@ 2026-01-14 15:09 Tao Chen
2026-01-14 17:35 ` Alan Maguire
0 siblings, 1 reply; 4+ messages in thread
From: Tao Chen @ 2026-01-14 15:09 UTC (permalink / raw)
To: Alexei Starovoitov, Andrii Nakryiko, Eduard Zingerman, Song Liu,
Yonghong Song, Jiri Olsa, LKML, bpf
Hi guys,
When using tracepoints to retrieve stack information, I observed that
perf_trace_sched_migrate_task was printed twice. And the issue also
occurs with tools using libbpf.
sudo bpftrace -e '
tracepoint:sched:sched_migrate_task {
printf("Task %s migrated by:\n", args->comm);
print(kstack);
}'
Task kcompactd0 migrated by:
perf_trace_sched_migrate_task+9
perf_trace_sched_migrate_task+9
set_task_cpu+353
detach_task+77
detach_tasks+281
sched_balance_rq+452
sched_balance_newidle+504
pick_next_task_fair+84
__pick_next_task+66
pick_next_task+43
__schedule+332
schedule+41
schedule_hrtimeout_range+239
do_poll.constprop.0+668
do_sys_poll+499
__x64_sys_ppoll+220
x64_sys_call+5722
do_syscall_64+126
entry_SYSCALL_64_after_hwframe+118
Task jbd2/sda2-8 migrated by:
perf_trace_sched_migrate_task+9
perf_trace_sched_migrate_task+9
set_task_cpu+353
try_to_wake_up+365
default_wake_function+26
autoremove_wake_function+18
__wake_up_common+118
__wake_up+55
__jbd2_log_start_commit+195
env:
bpftrace v0.21.2
ubuntu24.04，6.14.0-36-generic
The issue is as follows:
https://github.com/bpftrace/bpftrace/issues/4949
It seems that there is no special handling in the kernel.
Does anyone has thoughts on this issue. Thanks.
BPF_CALL_4(bpf_get_stack_raw_tp, struct bpf_raw_tracepoint_args *, args,
void *, buf, u32, size, u64, flags)
{
struct pt_regs *regs = get_bpf_raw_tp_regs();
int ret;
if (IS_ERR(regs))
return PTR_ERR(regs);
perf_fetch_caller_regs(regs);
ret = bpf_get_stack((unsigned long) regs, (unsigned long) buf,
(unsigned long) size, flags, 0);
put_bpf_raw_tp_regs();
return ret;
}
--
Best Regards
Tao Chen
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: The same symbol is printed twice when use tracepoint to get stack
2026-01-14 15:09 The same symbol is printed twice when use tracepoint to get stack Tao Chen
@ 2026-01-14 17:35 ` Alan Maguire
2026-01-14 21:06 ` Jiri Olsa
0 siblings, 1 reply; 4+ messages in thread
From: Alan Maguire @ 2026-01-14 17:35 UTC (permalink / raw)
To: Tao Chen, Alexei Starovoitov, Andrii Nakryiko, Eduard Zingerman,
Song Liu, Yonghong Song, Jiri Olsa, LKML, bpf
On 14/01/2026 15:09, Tao Chen wrote:
> Hi guys,
>
> When using tracepoints to retrieve stack information, I observed that perf_trace_sched_migrate_task was printed twice. And the issue also occurs with tools using libbpf.
>
You may need the fix Jiri provided for x86_64 [1]. Eugene mentioned that
the issue persists for arm64 however [2].
Alan
[1] https://lore.kernel.org/bpf/20251104215405.168643-2-jolsa@kernel.org/
[2] https://lore.kernel.org/all/a38fed68-67bc-98ce-8e12-743342121ae3@oracle.com/
> sudo bpftrace -e '
> tracepoint:sched:sched_migrate_task {
> printf("Task %s migrated by:\n", args->comm);
> print(kstack);
> }'
>
> Task kcompactd0 migrated by:
>
> perf_trace_sched_migrate_task+9
> perf_trace_sched_migrate_task+9
> set_task_cpu+353
> detach_task+77
> detach_tasks+281
> sched_balance_rq+452
> sched_balance_newidle+504
> pick_next_task_fair+84
> __pick_next_task+66
> pick_next_task+43
> __schedule+332
> schedule+41
> schedule_hrtimeout_range+239
> do_poll.constprop.0+668
> do_sys_poll+499
> __x64_sys_ppoll+220
> x64_sys_call+5722
> do_syscall_64+126
> entry_SYSCALL_64_after_hwframe+118
>
> Task jbd2/sda2-8 migrated by:
>
> perf_trace_sched_migrate_task+9
> perf_trace_sched_migrate_task+9
> set_task_cpu+353
> try_to_wake_up+365
> default_wake_function+26
> autoremove_wake_function+18
> __wake_up_common+118
> __wake_up+55
> __jbd2_log_start_commit+195
>
> env:
> bpftrace v0.21.2
> ubuntu24.04，6.14.0-36-generic
>
> The issue is as follows:
> https://github.com/bpftrace/bpftrace/issues/4949
>
>
> It seems that there is no special handling in the kernel.
> Does anyone has thoughts on this issue. Thanks.
>
> BPF_CALL_4(bpf_get_stack_raw_tp, struct bpf_raw_tracepoint_args *, args,
> void *, buf, u32, size, u64, flags)
> {
> struct pt_regs *regs = get_bpf_raw_tp_regs();
> int ret;
>
> if (IS_ERR(regs))
> return PTR_ERR(regs);
>
> perf_fetch_caller_regs(regs);
> ret = bpf_get_stack((unsigned long) regs, (unsigned long) buf,
> (unsigned long) size, flags, 0);
> put_bpf_raw_tp_regs();
> return ret;
> }
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: The same symbol is printed twice when use tracepoint to get stack
2026-01-14 17:35 ` Alan Maguire
@ 2026-01-14 21:06 ` Jiri Olsa
2026-01-15 2:18 ` Tao Chen
0 siblings, 1 reply; 4+ messages in thread
From: Jiri Olsa @ 2026-01-14 21:06 UTC (permalink / raw)
To: Alan Maguire
Cc: Tao Chen, Alexei Starovoitov, Andrii Nakryiko, Eduard Zingerman,
Song Liu, Yonghong Song, LKML, bpf
On Wed, Jan 14, 2026 at 05:35:45PM +0000, Alan Maguire wrote:
> On 14/01/2026 15:09, Tao Chen wrote:
> > Hi guys,
> >
> > When using tracepoints to retrieve stack information, I observed that perf_trace_sched_migrate_task was printed twice. And the issue also occurs with tools using libbpf.
> >
>
> You may need the fix Jiri provided for x86_64 [1]. Eugene mentioned that
> the issue persists for arm64 however [2].
yep, there's also follow patchset up for kprobe multi [1]
jirka
[1] https://lore.kernel.org/bpf/20260112214940.1222115-1-jolsa@kernel.org/
>
> Alan
>
> [1] https://lore.kernel.org/bpf/20251104215405.168643-2-jolsa@kernel.org/
> [2] https://lore.kernel.org/all/a38fed68-67bc-98ce-8e12-743342121ae3@oracle.com/
>
> > sudo bpftrace -e '
> > tracepoint:sched:sched_migrate_task {
> > printf("Task %s migrated by:\n", args->comm);
> > print(kstack);
> > }'
> >
> > Task kcompactd0 migrated by:
> >
> > perf_trace_sched_migrate_task+9
> > perf_trace_sched_migrate_task+9
> > set_task_cpu+353
> > detach_task+77
> > detach_tasks+281
> > sched_balance_rq+452
> > sched_balance_newidle+504
> > pick_next_task_fair+84
> > __pick_next_task+66
> > pick_next_task+43
> > __schedule+332
> > schedule+41
> > schedule_hrtimeout_range+239
> > do_poll.constprop.0+668
> > do_sys_poll+499
> > __x64_sys_ppoll+220
> > x64_sys_call+5722
> > do_syscall_64+126
> > entry_SYSCALL_64_after_hwframe+118
> >
> > Task jbd2/sda2-8 migrated by:
> >
> > perf_trace_sched_migrate_task+9
> > perf_trace_sched_migrate_task+9
> > set_task_cpu+353
> > try_to_wake_up+365
> > default_wake_function+26
> > autoremove_wake_function+18
> > __wake_up_common+118
> > __wake_up+55
> > __jbd2_log_start_commit+195
> >
> > env:
> > bpftrace v0.21.2
> > ubuntu24.04，6.14.0-36-generic
> >
> > The issue is as follows:
> > https://github.com/bpftrace/bpftrace/issues/4949
> >
> >
> > It seems that there is no special handling in the kernel.
> > Does anyone has thoughts on this issue. Thanks.
> >
> > BPF_CALL_4(bpf_get_stack_raw_tp, struct bpf_raw_tracepoint_args *, args,
> > void *, buf, u32, size, u64, flags)
> > {
> > struct pt_regs *regs = get_bpf_raw_tp_regs();
> > int ret;
> >
> > if (IS_ERR(regs))
> > return PTR_ERR(regs);
> >
> > perf_fetch_caller_regs(regs);
> > ret = bpf_get_stack((unsigned long) regs, (unsigned long) buf,
> > (unsigned long) size, flags, 0);
> > put_bpf_raw_tp_regs();
> > return ret;
> > }
> >
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: The same symbol is printed twice when use tracepoint to get stack
2026-01-14 21:06 ` Jiri Olsa
@ 2026-01-15 2:18 ` Tao Chen
0 siblings, 0 replies; 4+ messages in thread
From: Tao Chen @ 2026-01-15 2:18 UTC (permalink / raw)
To: Jiri Olsa, Alan Maguire
Cc: Alexei Starovoitov, Andrii Nakryiko, Eduard Zingerman, Song Liu,
Yonghong Song, LKML, bpf
在 2026/1/15 05:06, Jiri Olsa 写道:
> On Wed, Jan 14, 2026 at 05:35:45PM +0000, Alan Maguire wrote:
>> On 14/01/2026 15:09, Tao Chen wrote:
>>> Hi guys,
>>>
>>> When using tracepoints to retrieve stack information, I observed that perf_trace_sched_migrate_task was printed twice. And the issue also occurs with tools using libbpf.
>>>
>>
>> You may need the fix Jiri provided for x86_64 [1]. Eugene mentioned that
>> the issue persists for arm64 however [2].
>
> yep, there's also follow patchset up for kprobe multi [1]
>
> jirka
>
>
> [1] https://lore.kernel.org/bpf/20260112214940.1222115-1-jolsa@kernel.org/
>
>>
>> Alan
>>
>> [1] https://lore.kernel.org/bpf/20251104215405.168643-2-jolsa@kernel.org/
>> [2] https://lore.kernel.org/all/a38fed68-67bc-98ce-8e12-743342121ae3@oracle.com/
>>
>>> sudo bpftrace -e '
>>> tracepoint:sched:sched_migrate_task {
>>> printf("Task %s migrated by:\n", args->comm);
>>> print(kstack);
>>> }'
>>>
>>> Task kcompactd0 migrated by:
>>>
>>> perf_trace_sched_migrate_task+9
>>> perf_trace_sched_migrate_task+9
>>> set_task_cpu+353
>>> detach_task+77
>>> detach_tasks+281
>>> sched_balance_rq+452
>>> sched_balance_newidle+504
>>> pick_next_task_fair+84
>>> __pick_next_task+66
>>> pick_next_task+43
>>> __schedule+332
>>> schedule+41
>>> schedule_hrtimeout_range+239
>>> do_poll.constprop.0+668
>>> do_sys_poll+499
>>> __x64_sys_ppoll+220
>>> x64_sys_call+5722
>>> do_syscall_64+126
>>> entry_SYSCALL_64_after_hwframe+118
>>>
>>> Task jbd2/sda2-8 migrated by:
>>>
>>> perf_trace_sched_migrate_task+9
>>> perf_trace_sched_migrate_task+9
>>> set_task_cpu+353
>>> try_to_wake_up+365
>>> default_wake_function+26
>>> autoremove_wake_function+18
>>> __wake_up_common+118
>>> __wake_up+55
>>> __jbd2_log_start_commit+195
>>>
>>> env:
>>> bpftrace v0.21.2
>>> ubuntu24.04，6.14.0-36-generic
>>>
>>> The issue is as follows:
>>> https://github.com/bpftrace/bpftrace/issues/4949
>>>
>>>
>>> It seems that there is no special handling in the kernel.
>>> Does anyone has thoughts on this issue. Thanks.
>>>
>>> BPF_CALL_4(bpf_get_stack_raw_tp, struct bpf_raw_tracepoint_args *, args,
>>> void *, buf, u32, size, u64, flags)
>>> {
>>> struct pt_regs *regs = get_bpf_raw_tp_regs();
>>> int ret;
>>>
>>> if (IS_ERR(regs))
>>> return PTR_ERR(regs);
>>>
>>> perf_fetch_caller_regs(regs);
>>> ret = bpf_get_stack((unsigned long) regs, (unsigned long) buf,
>>> (unsigned long) size, flags, 0);
>>> put_bpf_raw_tp_regs();
>>> return ret;
>>> }
>>>
>>
Alan, Jiri
Thanks for your reply, i see.
--
Best Regards
Tao Chen
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2026-01-15 2:19 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-01-14 15:09 The same symbol is printed twice when use tracepoint to get stack Tao Chen
2026-01-14 17:35 ` Alan Maguire
2026-01-14 21:06 ` Jiri Olsa
2026-01-15 2:18 ` Tao Chen
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox