From: Peter Zijlstra <peterz@infradead.org>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
linux-kernel@vger.kernel.org, mingo@kernel.org,
jiangshanlai@gmail.com, dipankar@in.ibm.com,
akpm@linux-foundation.org, mathieu.desnoyers@efficios.com,
josh@joshtriplett.org, tglx@linutronix.de, dhowells@redhat.com,
edumazet@google.com, fweisbec@gmail.com, oleg@redhat.com,
joel.opensrc@gmail.com
Subject: Re: [PATCH tip/core/rcu 01/22] sched: Make non-production PREEMPT cond_resched() help Tasks RCU
Date: Mon, 23 Apr 2018 16:10:38 +0200 [thread overview]
Message-ID: <20180423141038.GA4064@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <20180423094742.6582e19d@gandalf.local.home>
On Mon, Apr 23, 2018 at 09:47:42AM -0400, Steven Rostedt wrote:
> On Mon, 23 Apr 2018 05:40:00 -0700
> "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> wrote:
>
> > > I'm confused.. why is having this conditional on TRACEPOINT_BENCHMARK a
> > > sane idea?
> >
> > Because the TRACEPOINT_BENCHMARK tests are insane, so a similar
> > level of insanity is required to make things work. Plus having this
> > be unconditional would not be good for performance, as 0day has been
> > telling me frequently over the past couple of years.
>
> Just for some context. The tracepoint benchmark (which should never be
> enabled in any production machine), will start a thread when the
> benchmark trace event is enabled. This thread will never exit (until
> the trace event is disabled), and does a benchmark loop and constantly
> calls "cond_resched()" to allow other tasks to run. The point is, this
> thread will never have a quiescent state for task_rcu, unless we tell
> rcu that cond_resched() is a quiescent state. But this is only required
> because the tracepoint benchmark has this nasty thread, that is only
> used for debugging and benchmarking the tracepoint (during development).
>
> I also suggested having a direct call into RCU from the thread to tell
> RCU that it entered a quiescent state, but Paul didn't like that idea
> as it caused the tracepoint benchmark to call too deep into RCU
> internals.
>
> http://lkml.kernel.org/r/20180227153646.GD3777@linux.vnet.ibm.com
That thread using cond_resched_task_rcu_qs() seems like a _lot_ better
than having cond_resched() semantics change depending on random
!scheduler config parameters.
next prev parent reply other threads:[~2018-04-23 14:12 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-04-23 2:31 [PATCH tip/core/rcu 0/22] Miscellaneous fixes for v4.18 Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 01/22] sched: Make non-production PREEMPT cond_resched() help Tasks RCU Paul E. McKenney
2018-04-23 8:51 ` Peter Zijlstra
2018-04-23 12:40 ` Paul E. McKenney
2018-04-23 13:47 ` Steven Rostedt
2018-04-23 14:10 ` Peter Zijlstra [this message]
2018-04-23 14:35 ` Steven Rostedt
2018-04-23 15:47 ` Paul E. McKenney
2018-04-23 19:12 ` Paul E. McKenney
2018-04-23 14:03 ` Peter Zijlstra
2018-04-23 2:32 ` [PATCH tip/core/rcu 02/22] rcu: Inline rcu_preempt_do_callback() into its sole caller Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 03/22] rcu: Don't allocate rcu_nocb_mask if no one needs it Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 04/22] rcu: Call wake_nocb_leader_defer() with 'FORCE' when nocb_q_count is high Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 05/22] rcu: Remove deprecated RCU debugfs tracing code Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 06/22] rcu: Rename cond_resched_rcu_qs() to cond_resched_tasks_rcu_qs() Paul E. McKenney
2018-04-23 8:53 ` Peter Zijlstra
2018-04-23 12:32 ` Paul E. McKenney
2018-04-23 12:48 ` Peter Zijlstra
2018-04-23 13:12 ` Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 07/22] softirq: Eliminate unused cond_resched_softirq() macro Paul E. McKenney
2018-04-23 8:54 ` Peter Zijlstra
2018-04-23 13:25 ` Eric Dumazet
2018-04-23 19:10 ` Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 08/22] rcu: Move __rcu_read_lock() and __rcu_read_unlock() to tree_plugin.h Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 09/22] rcu: Update rcu_bind_gp_kthread() header comment Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 10/22] rcu: Declare rcu_eqs_special_set() in public header Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 11/22] srcu: Add cleanup_srcu_struct_quiesced() Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 12/22] nvme: Avoid flush dependency in delete controller flow Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 13/22] rcu: Exclude near-simultaneous RCU CPU stall warnings Paul E. McKenney
2018-05-03 18:22 ` Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 14/22] rcu: Add leaf-node macros Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 15/22] doc: Ensure whatisRCU.txt actually says what RCU is Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 16/22] EXP: rcu: Add debugging info to assertion Paul E. McKenney
2018-05-03 18:23 ` Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 17/22] EXP: rcu: Abstract addition of debugging information " Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 18/22] EXP: rcu: Add ->boost_tasks " Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 19/22] EXP: rcu: Add debugging info to other assertion Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 20/22] EXP rcu: Add ->qsmask to assertion Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 21/22] EXP rcu: Add checks for setting ->gp_flags Paul E. McKenney
2018-04-23 2:32 ` [PATCH tip/core/rcu 22/22] rcu: Use the proper lockdep annotation in dump_blkd_tasks() Paul E. McKenney
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=20180423141038.GA4064@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=akpm@linux-foundation.org \
--cc=dhowells@redhat.com \
--cc=dipankar@in.ibm.com \
--cc=edumazet@google.com \
--cc=fweisbec@gmail.com \
--cc=jiangshanlai@gmail.com \
--cc=joel.opensrc@gmail.com \
--cc=josh@joshtriplett.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=mingo@kernel.org \
--cc=oleg@redhat.com \
--cc=paulmck@linux.vnet.ibm.com \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
/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