From: Ankur Arora <ankur.a.arora@oracle.com>
To: Joel Fernandes <joel@joelfernandes.org>
Cc: paulmck@kernel.org, Ankur Arora <ankur.a.arora@oracle.com>,
linux-kernel@vger.kernel.org, tglx@linutronix.de,
peterz@infradead.org, torvalds@linux-foundation.org,
akpm@linux-foundation.org, luto@kernel.org, bp@alien8.de,
dave.hansen@linux.intel.com, hpa@zytor.com, mingo@redhat.com,
juri.lelli@redhat.com, vincent.guittot@linaro.org,
willy@infradead.org, mgorman@suse.de, jpoimboe@kernel.org,
mark.rutland@arm.com, jgross@suse.com, andrew.cooper3@citrix.com,
bristot@kernel.org, mathieu.desnoyers@efficios.com,
geert@linux-m68k.org, glaubitz@physik.fu-berlin.de,
anton.ivanov@cambridgegreys.com, mattst88@gmail.com,
krypton@ulrich-teichert.org, rostedt@goodmis.org,
David.Laight@aculab.com, richard@nod.at, mjguzik@gmail.com,
jon.grimm@amd.com, bharata@amd.com, raghavendra.kt@amd.com,
boris.ostrovsky@oracle.com, konrad.wilk@oracle.com,
rcu@vger.kernel.org
Subject: Re: [PATCH 15/30] rcu: handle quiescent states for PREEMPT_RCU=n, PREEMPT_COUNT=y
Date: Mon, 11 Mar 2024 13:51:59 -0700 [thread overview]
Message-ID: <877ci8h4pc.fsf@oracle.com> (raw)
In-Reply-To: <ffc5019a-b593-4dd9-b4e1-4f7755040f23@joelfernandes.org>
Joel Fernandes <joel@joelfernandes.org> writes:
> On 3/10/2024 11:56 PM, Paul E. McKenney wrote:
>> On Sun, Mar 10, 2024 at 08:48:28PM -0400, Joel Fernandes wrote:
>>> On 3/10/2024 2:56 PM, Paul E. McKenney wrote:
>>>> On Sun, Mar 10, 2024 at 06:03:30AM -0400, Joel Fernandes wrote:
>>>>> Hello Ankur and Paul,
>>>>>
>>>>> On Mon, Feb 12, 2024 at 09:55:39PM -0800, Ankur Arora wrote:
>>>>>> With PREEMPT_RCU=n, cond_resched() provides urgently needed quiescent
>>>>>> states for read-side critical sections via rcu_all_qs().
>>>>>> One reason why this was necessary: lacking preempt-count, the tick
>>>>>> handler has no way of knowing whether it is executing in a read-side
>>>>>> critical section or not.
>>>>>>
>>>>>> With PREEMPT_AUTO=y, there can be configurations with (PREEMPT_COUNT=y,
>>>>>> PREEMPT_RCU=n). This means that cond_resched() is a stub which does
>>>>>> not provide for quiescent states via rcu_all_qs().
>>>>>>
>>>>>> So, use the availability of preempt_count() to report quiescent states
>>>>>> in rcu_flavor_sched_clock_irq().
>>>>>>
>>>>>> Suggested-by: Paul E. McKenney <paulmck@kernel.org>
>>>>>> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com>
>>>>>> ---
>>>>>> kernel/rcu/tree_plugin.h | 11 +++++++----
>>>>>> 1 file changed, 7 insertions(+), 4 deletions(-)
>>>>>>
>>>>>> diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
>>>>>> index 26c79246873a..9b72e9d2b6fe 100644
>>>>>> --- a/kernel/rcu/tree_plugin.h
>>>>>> +++ b/kernel/rcu/tree_plugin.h
>>>>>> @@ -963,13 +963,16 @@ static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp)
>>>>>> */
>>>>>> static void rcu_flavor_sched_clock_irq(int user)
>>>>>> {
>>>>>> - if (user || rcu_is_cpu_rrupt_from_idle()) {
>>>>>> + if (user || rcu_is_cpu_rrupt_from_idle() ||
>>>>>> + (IS_ENABLED(CONFIG_PREEMPT_COUNT) &&
>>>>>> + !(preempt_count() & (PREEMPT_MASK | SOFTIRQ_MASK)))) {
>>>>>
>>>>> I was wondering if it makes sense to even support !PREEMPT_RCU under
>>>>> CONFIG_PREEMPT_AUTO.
>>>>>
>>>>> AFAIU, this CONFIG_PREEMPT_AUTO series preempts the kernel on
>>>>> the next tick boundary in the worst case, with all preempt modes including
>>>>> the preempt=none mode.
>>>>>
>>>>> Considering this, does it makes sense for RCU to be non-preemptible in
>>>>> CONFIG_PREEMPT_AUTO=y? Because if that were the case, and a read-side critical
>>>>> section extended beyond the tick, then it prevents the PREEMPT_AUTO preemption
>>>>> from happening, because rcu_read_lock() would preempt_disable().
>>>>
>>>> Yes, it does make sense for RCU to be non-preemptible in kernels
>>>> built with CONFIG_PREEMPT_AUTO=y and either CONFIG_PREEMPT_NONE=y or
>>>> CONFIG_PREEMPT_VOLUNTARY=y.
>>>> As noted in earlier discussions, there are
>>>
>>> Sorry if I missed a discussion, appreciate a link.
>>
>> It is part of the discussion of the first version of this patch series,
>> if I recall correctly.
>>
>>>> systems that are adequately but not abundantly endowed with memory.
>>>> Such systems need non-preemptible RCU to avoid preempted-reader OOMs.
>>>
>>> Then why don't such systems have a problem with CONFIG_PREEMPT_DYNAMIC=y and
>>> preempt=none mode? CONFIG_PREEMPT_DYNAMIC forces CONFIG_PREEMPT_RCU=y. There's
>>> no way to set CONFIG_PREEMPT_RCU=n with CONFIG_PREEMPT_DYNAMIC=y and
>>> preempt=none boot parameter. IMHO, if this feature is inconsistent with
>>> CONFIG_PREEMPT_DYNAMIC, that makes it super confusing. In fact, I feel
>>> CONFIG_PREEMPT_AUTO should instead just be another "preempt=auto" boot parameter
>>> mode added to CONFIG_PREEMPT_DYNAMIC feature, otherwise the proliferation of
>>> CONFIG_PREEMPT config options is getting a bit insane. And likely going to be
>>> burden to the users configuring the PREEMPT Kconfig option IMHO.
>>
>> Because such systems are built with CONFIG_PREEMPT_DYNAMIC=n.
>>
>> You could argue that we should just build with CONFIG_PREEMPT_AUTO=n,
>> but the long-term goal of eliminating cond_resched() will make that
>> ineffective.
>
> I see what you mean. We/I could also highlight some of the differences in RCU
> between DYNAMIC vs AUTO.
>
>>
>>>> Note well that non-preemptible RCU explicitly disables preemption across
>>>> all RCU readers.
>>>
>>> Yes, I mentioned this 'disabling preemption' aspect in my last email. My point
>>> being, unlike CONFIG_PREEMPT_NONE, CONFIG_PREEMPT_AUTO allows for kernel
>>> preemption in preempt=none. So the "Don't preempt the kernel" behavior has
>>> changed. That is, preempt=none under CONFIG_PREEMPT_AUTO is different from
>>> CONFIG_PREEMPT_NONE=y already. Here we *are* preempting. And RCU is getting on
>>> the way. It is like saying, you want an option for CONFIG_PREEMPT_RCU to be set
>>> to =n for CONFIG_PREEMPT=y kernels, sighting users who want a fully-preemptible
>>> kernel but are worried about reader preemptions.
>>
>> Such users can simply avoid building with either CONFIG_PREEMPT_NONE=y
>> or with CONFIG_PREEMPT_VOLUNTARY=y. They might also experiment with
>> CONFIG_RCU_BOOST=y, and also with short timeouts until boosting.
>> If that doesn't do what they need, we talk with them and either help
>> them configure their kernels, make RCU do what they need, or help work
>> out some other way for them to get their jobs done.
>
> Makes sense.
>
>>> That aside, as such, I do agree your point of view, that preemptible readers
>>> presents a problem to folks using preempt=none in this series and we could
>>> decide to keep CONFIG_PREEMPT_RCU optional for whoever wants it that way. I was
>>> just saying that I want CONFIG_PREEMPT_AUTO's preempt=none mode to be somewhat
>>> consistent with CONFIG_PREEMPT_DYNAMIC's preempt=none. Because I'm pretty sure a
>>> week from now, no one will likely be able to tell the difference ;-). So IMHO
>>> either CONFIG_PREEMPT_DYNAMIC should be changed to make CONFIG_PREEMPT_RCU
>>> optional, or this series should be altered to force CONFIG_PREEMPT_RCU=y.
>>>
>>> Let me know if I missed something.
>>
>> Why not key off of the value of CONFIG_PREEMPT_DYNAMIC? That way,
>> if both CONFIG_PREEMPT_AUTO=y and CONFIG_PREEMPT_DYNAMIC=y, RCU is
>> always preemptible. Then CONFIG_PREEMPT_DYNAMIC=y enables boot-time
>> (and maybe even run-time) switching between preemption flavors, while
>> CONFIG_PREEMPT_AUTO instead enables unconditional preemption of any
>> region of code that has not explicitly disabled preemption (or irq or
>> bh or whatever).
Currently CONFIG_PREEMPT_DYNAMIC does a few things:
1. dynamic selection of preemption model
2. dynamically toggling explicit preemption points
3. PREEMPT_RCU=y (though maybe this should be fixed to also
also allow PREEMPT_RCU=n)
Of these 3, PREEMPT_AUTO only really needs (1).
Maybe combining gives us the option of switching between the old and the
new models:
preempt=none | voluntary | full | auto-none | auto-voluntary
Where the last two provide the new auto semantics. But, the mixture
seems too rich.
This just complicates all the CONFIG_PREEMPT_* configurations more than
they were before when the end goal is to actually reduce and simplify
the number of options.
> That could be done. But currently, these patches disable DYNAMIC if AUTO is
> enabled in the config. I think the reason is the 2 features are incompatible.
> i.e. DYNAMIC wants to override the preemption mode at boot time, where as AUTO
> wants the scheduler to have a say in it using the need-resched LAZY bit.
Yeah exactly. That's why I originally made PREEMPT_AUTO and
PREEMPT_DYNAMIC exclusive of each other.
Thanks
--
ankur
next prev parent reply other threads:[~2024-03-11 20:52 UTC|newest]
Thread overview: 155+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-02-13 5:55 [PATCH 00/30] PREEMPT_AUTO: support lazy rescheduling Ankur Arora
2024-02-13 5:55 ` [PATCH 01/30] preempt: introduce CONFIG_PREEMPT_AUTO Ankur Arora
2024-02-13 5:55 ` [PATCH 02/30] thread_info: selector for TIF_NEED_RESCHED[_LAZY] Ankur Arora
2024-02-19 15:16 ` Thomas Gleixner
2024-02-20 22:50 ` Ankur Arora
2024-02-21 17:05 ` Thomas Gleixner
2024-02-21 18:26 ` Steven Rostedt
2024-02-21 20:03 ` Thomas Gleixner
2024-02-13 5:55 ` [PATCH 03/30] thread_info: tif_need_resched() now takes resched_t as param Ankur Arora
2024-02-14 3:17 ` kernel test robot
2024-02-14 14:08 ` Mark Rutland
2024-02-15 4:08 ` Ankur Arora
2024-02-19 12:30 ` Mark Rutland
2024-02-20 22:09 ` Ankur Arora
2024-02-19 15:21 ` Thomas Gleixner
2024-02-20 22:21 ` Ankur Arora
2024-02-21 17:07 ` Thomas Gleixner
2024-02-21 21:22 ` Ankur Arora
2024-02-13 5:55 ` [PATCH 04/30] sched: make test_*_tsk_thread_flag() return bool Ankur Arora
2024-02-14 14:12 ` Mark Rutland
2024-02-15 2:04 ` Ankur Arora
2024-02-13 5:55 ` [PATCH 05/30] sched: *_tsk_need_resched() now takes resched_t as param Ankur Arora
2024-02-19 15:26 ` Thomas Gleixner
2024-02-20 22:37 ` Ankur Arora
2024-02-21 17:10 ` Thomas Gleixner
2024-02-13 5:55 ` [PATCH 06/30] entry: handle lazy rescheduling at user-exit Ankur Arora
2024-02-19 15:29 ` Thomas Gleixner
2024-02-20 22:38 ` Ankur Arora
2024-02-13 5:55 ` [PATCH 07/30] entry/kvm: handle lazy rescheduling at guest-entry Ankur Arora
2024-02-13 5:55 ` [PATCH 08/30] entry: irqentry_exit only preempts for TIF_NEED_RESCHED Ankur Arora
2024-02-13 5:55 ` [PATCH 09/30] sched: __schedule_loop() doesn't need to check for need_resched_lazy() Ankur Arora
2024-02-13 5:55 ` [PATCH 10/30] sched: separate PREEMPT_DYNAMIC config logic Ankur Arora
2024-02-13 5:55 ` [PATCH 11/30] sched: runtime preemption config under PREEMPT_AUTO Ankur Arora
2024-02-13 5:55 ` [PATCH 12/30] rcu: limit PREEMPT_RCU to full preemption " Ankur Arora
2024-02-13 5:55 ` [PATCH 13/30] rcu: fix header guard for rcu_all_qs() Ankur Arora
2024-02-13 5:55 ` [PATCH 14/30] preempt,rcu: warn on PREEMPT_RCU=n, preempt=full Ankur Arora
2024-02-13 5:55 ` [PATCH 15/30] rcu: handle quiescent states for PREEMPT_RCU=n, PREEMPT_COUNT=y Ankur Arora
2024-03-10 10:03 ` Joel Fernandes
2024-03-10 18:56 ` Paul E. McKenney
2024-03-11 0:48 ` Joel Fernandes
2024-03-11 3:56 ` Paul E. McKenney
2024-03-11 15:01 ` Joel Fernandes
2024-03-11 20:51 ` Ankur Arora [this message]
2024-03-11 22:12 ` Thomas Gleixner
2024-03-11 5:18 ` Ankur Arora
2024-03-11 15:25 ` Joel Fernandes
2024-03-11 19:12 ` Thomas Gleixner
2024-03-11 19:53 ` Paul E. McKenney
2024-03-11 20:29 ` Thomas Gleixner
2024-03-12 0:01 ` Paul E. McKenney
2024-03-12 0:08 ` Joel Fernandes
2024-03-12 3:16 ` Ankur Arora
2024-03-12 3:24 ` Joel Fernandes
2024-03-12 5:23 ` Ankur Arora
2024-02-13 5:55 ` [PATCH 16/30] rcu: force context-switch " Ankur Arora
2024-02-13 5:55 ` [PATCH 17/30] x86/thread_info: define TIF_NEED_RESCHED_LAZY Ankur Arora
2024-02-14 13:25 ` Mark Rutland
2024-02-14 20:31 ` Ankur Arora
2024-02-19 12:32 ` Mark Rutland
2024-02-13 5:55 ` [PATCH 18/30] sched: prepare for lazy rescheduling in resched_curr() Ankur Arora
2024-02-13 5:55 ` [PATCH 19/30] sched: default preemption policy for PREEMPT_AUTO Ankur Arora
2024-02-13 5:55 ` [PATCH 20/30] sched: handle idle preemption " Ankur Arora
2024-02-13 5:55 ` [PATCH 21/30] sched: schedule eagerly in resched_cpu() Ankur Arora
2024-02-13 5:55 ` [PATCH 22/30] sched/fair: refactor update_curr(), entity_tick() Ankur Arora
2024-02-13 5:55 ` [PATCH 23/30] sched/fair: handle tick expiry under lazy preemption Ankur Arora
2024-02-21 21:38 ` Steven Rostedt
2024-02-28 13:47 ` Juri Lelli
2024-02-29 6:43 ` Ankur Arora
2024-02-29 9:33 ` Juri Lelli
2024-02-29 23:54 ` Ankur Arora
2024-03-01 0:28 ` Paul E. McKenney
2024-02-13 5:55 ` [PATCH 24/30] sched: support preempt=none under PREEMPT_AUTO Ankur Arora
2024-02-13 5:55 ` [PATCH 25/30] sched: support preempt=full " Ankur Arora
2024-02-13 5:55 ` [PATCH 26/30] sched: handle preempt=voluntary " Ankur Arora
2024-03-03 1:08 ` Joel Fernandes
2024-03-05 8:11 ` Ankur Arora
2024-03-06 20:42 ` Joel Fernandes
2024-03-07 19:01 ` Paul E. McKenney
2024-03-08 0:15 ` Joel Fernandes
2024-03-08 0:42 ` Paul E. McKenney
2024-03-08 4:22 ` Ankur Arora
2024-03-08 21:33 ` Paul E. McKenney
2024-03-11 4:50 ` Ankur Arora
2024-03-11 19:26 ` Paul E. McKenney
2024-03-11 20:09 ` Ankur Arora
2024-03-11 20:23 ` Linus Torvalds
2024-03-11 21:03 ` Ankur Arora
2024-03-12 0:03 ` Paul E. McKenney
2024-03-12 12:14 ` Thomas Gleixner
2024-03-12 19:40 ` Paul E. McKenney
2024-03-08 3:49 ` Ankur Arora
2024-03-08 5:29 ` Joel Fernandes
2024-03-08 6:54 ` Juri Lelli
2024-03-11 5:34 ` Ankur Arora
2024-02-13 5:55 ` [PATCH 27/30] sched: latency warn for TIF_NEED_RESCHED_LAZY Ankur Arora
2024-02-13 5:55 ` [PATCH 28/30] tracing: support lazy resched Ankur Arora
2024-02-13 5:55 ` [PATCH 29/30] Documentation: tracing: add TIF_NEED_RESCHED_LAZY Ankur Arora
2024-02-21 21:43 ` Steven Rostedt
2024-02-21 23:22 ` Ankur Arora
2024-02-21 23:53 ` Steven Rostedt
2024-03-01 23:33 ` Joel Fernandes
2024-03-02 3:09 ` Ankur Arora
2024-03-03 19:32 ` Joel Fernandes
2024-02-13 5:55 ` [PATCH 30/30] osnoise: handle quiescent states for PREEMPT_RCU=n, PREEMPTION=y Ankur Arora
2024-02-13 9:47 ` [PATCH 00/30] PREEMPT_AUTO: support lazy rescheduling Geert Uytterhoeven
2024-02-13 21:46 ` Ankur Arora
2024-02-14 23:57 ` Paul E. McKenney
2024-02-15 2:03 ` Ankur Arora
2024-02-15 3:45 ` Paul E. McKenney
2024-02-15 19:28 ` Paul E. McKenney
2024-02-15 20:04 ` Thomas Gleixner
2024-02-15 20:54 ` Paul E. McKenney
2024-02-15 20:53 ` Ankur Arora
2024-02-15 20:55 ` Paul E. McKenney
2024-02-15 21:24 ` Ankur Arora
2024-02-15 22:54 ` Paul E. McKenney
2024-02-15 22:56 ` Paul E. McKenney
2024-02-16 0:45 ` Ankur Arora
2024-02-16 2:59 ` Paul E. McKenney
2024-02-17 0:55 ` Paul E. McKenney
2024-02-17 3:59 ` Ankur Arora
2024-02-18 18:17 ` Paul E. McKenney
2024-02-19 16:48 ` Paul E. McKenney
2024-02-21 18:19 ` Steven Rostedt
2024-02-21 19:41 ` Paul E. McKenney
2024-02-21 20:11 ` Steven Rostedt
2024-02-21 20:22 ` Paul E. McKenney
2024-02-22 15:50 ` Mark Rutland
2024-02-22 19:11 ` Paul E. McKenney
2024-02-23 11:05 ` Mark Rutland
2024-02-23 15:31 ` Paul E. McKenney
2024-03-02 1:16 ` Paul E. McKenney
2024-03-19 11:45 ` Tasks RCU, ftrace, and trampolines (was: Re: [PATCH 00/30] PREEMPT_AUTO: support lazy rescheduling) Mark Rutland
2024-03-19 23:33 ` Paul E. McKenney
2024-02-21 6:48 ` [PATCH 00/30] PREEMPT_AUTO: support lazy rescheduling Ankur Arora
2024-02-21 17:44 ` Paul E. McKenney
2024-02-16 0:45 ` Ankur Arora
2024-02-21 12:23 ` Raghavendra K T
2024-02-21 17:15 ` Thomas Gleixner
2024-02-21 17:27 ` Raghavendra K T
2024-02-21 21:16 ` Ankur Arora
2024-02-22 4:05 ` Raghavendra K T
2024-02-22 21:23 ` Thomas Gleixner
2024-02-23 3:14 ` Ankur Arora
2024-02-23 6:28 ` Raghavendra K T
2024-02-24 3:15 ` Raghavendra K T
2024-02-27 17:45 ` Ankur Arora
2024-02-22 13:04 ` Raghavendra K T
2024-04-23 15:21 ` Shrikanth Hegde
2024-04-23 16:13 ` Linus Torvalds
2024-04-26 7:46 ` Shrikanth Hegde
2024-04-26 19:00 ` Ankur Arora
2024-05-07 11:16 ` Shrikanth Hegde
2024-05-08 5:18 ` Ankur Arora
2024-05-15 14:31 ` Shrikanth Hegde
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=877ci8h4pc.fsf@oracle.com \
--to=ankur.a.arora@oracle.com \
--cc=David.Laight@aculab.com \
--cc=akpm@linux-foundation.org \
--cc=andrew.cooper3@citrix.com \
--cc=anton.ivanov@cambridgegreys.com \
--cc=bharata@amd.com \
--cc=boris.ostrovsky@oracle.com \
--cc=bp@alien8.de \
--cc=bristot@kernel.org \
--cc=dave.hansen@linux.intel.com \
--cc=geert@linux-m68k.org \
--cc=glaubitz@physik.fu-berlin.de \
--cc=hpa@zytor.com \
--cc=jgross@suse.com \
--cc=joel@joelfernandes.org \
--cc=jon.grimm@amd.com \
--cc=jpoimboe@kernel.org \
--cc=juri.lelli@redhat.com \
--cc=konrad.wilk@oracle.com \
--cc=krypton@ulrich-teichert.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@kernel.org \
--cc=mark.rutland@arm.com \
--cc=mathieu.desnoyers@efficios.com \
--cc=mattst88@gmail.com \
--cc=mgorman@suse.de \
--cc=mingo@redhat.com \
--cc=mjguzik@gmail.com \
--cc=paulmck@kernel.org \
--cc=peterz@infradead.org \
--cc=raghavendra.kt@amd.com \
--cc=rcu@vger.kernel.org \
--cc=richard@nod.at \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=vincent.guittot@linaro.org \
--cc=willy@infradead.org \
/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;
as well as URLs for NNTP newsgroup(s).