mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Satya Durga Srinivasu Prabhala <quic_satyap@quicinc.com>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: <mingo@redhat.com>, <peterz@infradead.org>,
	<juri.lelli@redhat.com>, <vincent.guittot@linaro.org>,
	<dietmar.eggemann@arm.com>, <bsegall@google.com>,
	<mgorman@suse.de>, <bristot@redhat.com>, <vschneid@redhat.com>,
	<linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] sched: fix rq lock recursion issue
Date: Tue, 5 Jul 2022 20:47:33 -0700	[thread overview]
Message-ID: <037be4d3-0474-ba9f-fd0f-4bd9af3e835d@quicinc.com> (raw)
In-Reply-To: <Yr4lkBjU6Dy6Wb57@home.goodmis.org>


On 6/30/22 3:37 PM, Steven Rostedt wrote:
> On Fri, Jun 24, 2022 at 12:42:40AM -0700, Satya Durga Srinivasu Prabhala wrote:
>> Below recursion is observed in a rare scenario where __schedule()
>> takes rq lock, at around same time task's affinity is being changed,
>> bpf function for tracing sched_switch calls migrate_enabled(),
>> checks for affinity change (cpus_ptr != cpus_mask) lands into
>> __set_cpus_allowed_ptr which tries acquire rq lock and causing the
>> recursion bug.
>>
>> 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?
Thanks for your time and comments.
I was more looking at non-RT Kernel where switching to 
preempt_disable/enable() helps as it's
just increment/decrement of count. I agree, this isn't a right fix.
I'm still cross checking on easy way to repro the issue. Will cross 
check further and get back.

>
> -- Steve
>
>> -016 |__schedule(sched_mode = ?)
>> -017 |schedule()
>> -018 |arch_local_save_flags(inline)
>> -018 |arch_irqs_disabled(inline)
>> -018 |__raw_spin_lock_irq(inline)
>> -018 |_raw_spin_lock_irq(inline)
>> -018 |worker_thread(__worker = 0xFFFFFF88CE251300)
>> -019 |kthread(_create = 0xFFFFFF88730A5A80)
>> -020 |ret_from_fork(asm)
>>
>> Signed-off-by: Satya Durga Srinivasu Prabhala <quic_satyap@quicinc.com>
>> ---
>>   kernel/sched/core.c | 8 ++++++++
>>   1 file changed, 8 insertions(+)
>>
>> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
>> index bfa7452ca92e..e254e9227341 100644
>> --- a/kernel/sched/core.c
>> +++ b/kernel/sched/core.c
>> @@ -2223,6 +2223,7 @@ static void migrate_disable_switch(struct rq *rq, struct task_struct *p)
>>   
>>   void migrate_disable(void)
>>   {
>> +#ifdef CONFIG_PREEMPT_RT
>>   	struct task_struct *p = current;
>>   
>>   	if (p->migration_disabled) {
>> @@ -2234,11 +2235,15 @@ void migrate_disable(void)
>>   	this_rq()->nr_pinned++;
>>   	p->migration_disabled = 1;
>>   	preempt_enable();
>> +#else
>> +	preempt_disable();
>> +#endif
>>   }
>>   EXPORT_SYMBOL_GPL(migrate_disable);
>>   
>>   void migrate_enable(void)
>>   {
>> +#ifdef CONFIG_PREEMPT_RT
>>   	struct task_struct *p = current;
>>   
>>   	if (p->migration_disabled > 1) {
>> @@ -2265,6 +2270,9 @@ void migrate_enable(void)
>>   	p->migration_disabled = 0;
>>   	this_rq()->nr_pinned--;
>>   	preempt_enable();
>> +#else
>> +	preempt_enable();
>> +#endif
>>   }
>>   EXPORT_SYMBOL_GPL(migrate_enable);
>>   
>> -- 
>> 2.36.1

      parent reply	other threads:[~2022-07-06  3:50 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-06-24  7:42 Satya Durga Srinivasu Prabhala
2022-06-30 21:53 ` Qais Yousef
2022-07-01  8:33   ` Peter Zijlstra
2022-07-01 11:48     ` Qais Yousef
2022-07-04  8:32       ` Dietmar Eggemann
2022-07-06  3:42         ` Satya Durga Srinivasu Prabhala
2022-07-06  3:45       ` Satya Durga Srinivasu Prabhala
2022-07-06  3:46     ` Satya Durga Srinivasu Prabhala
2022-07-05  2:06   ` Satya Durga Srinivasu Prabhala
2022-06-30 22:37 ` Steven Rostedt
2022-07-01 15:54   ` Steven Rostedt
2022-07-06  3:44     ` Satya Durga Srinivasu Prabhala
     [not found]   ` <000101d89013$f2157320$d6405960$@quicinc.com>
2022-07-05  7:24     ` Peter Zijlstra
2022-07-06  3:40       ` Satya Durga Srinivasu Prabhala
2022-07-06  3:47   ` Satya Durga Srinivasu Prabhala [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=037be4d3-0474-ba9f-fd0f-4bd9af3e835d@quicinc.com \
    --to=quic_satyap@quicinc.com \
    --cc=bristot@redhat.com \
    --cc=bsegall@google.com \
    --cc=dietmar.eggemann@arm.com \
    --cc=juri.lelli@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mgorman@suse.de \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=vincent.guittot@linaro.org \
    --cc=vschneid@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®