lkml.org 
[lkml]   [2022]   [Jul]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH] sched: fix rq lock recursion issue
From
Date

On 7/1/22 8:54 AM, Steven Rostedt wrote:
> On Thu, 30 Jun 2022 18:37:04 -0400
> Steven Rostedt <rostedt@goodmis.org> wrote:
>
>>> Fix the issue by switching to preempt_enable/disable() for non-RT
>>> Kernels.
>>>
>>> -010 |spin_bug(lock = ???, msg = ???)
>>> -011 |debug_spin_lock_before(inline)
>>> -011 |do_raw_spin_lock(lock = 0xFFFFFF89323BB600)
>>> -012 |_raw_spin_lock(inline)
>>> -012 |raw_spin_rq_lock_nested(inline)
>>> -012 |raw_spin_rq_lock(inline)
>>> -012 |task_rq_lock(p = 0xFFFFFF88CFF1DA00, rf = 0xFFFFFFC03707BBE8)
>>> -013 |__set_cpus_allowed_ptr(inline)
>>> -013 |migrate_enable()
>>> -014 |trace_call_bpf(call = ?, ctx = 0xFFFFFFFDEF954600)
>>> -015 |perf_trace_run_bpf_submit(inline)
>>> -015 |perf_trace_sched_switch(__data = 0xFFFFFFE82CF0BCB8, preempt = FALSE, prev = ?, next = ?)
>>> -016 |__traceiter_sched_switch(inline)
>>> -016 |trace_sched_switch(inline)
>> trace_sched_switch() disables preemption.
>>
>> So how is this a fix?
> Let me rephrase my question.
>
> As trace_sched_switch() disables preemption, why is trace_call_bpf()
> calling migrate_disable()?
I'm not sure, why we have migrate_disable/enable(). I will need to cross
check further.

>
> Looks like you could modify the code to include a __bpf_prog_run_array()
> that skips the migrate_disable(). You even have a "cant_sleep()" call in
> trace_call_bpf().
Thanks for the inputs. I still need to cross check and find a better way
to repro
the issue to be able to comment further. Will cross check on your
suggestion.

>
> -- Steve

\
 
 \ /
  Last update: 2022-07-06 05:45    [W:1.635 / U:0.072 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site