From: Peter Zijlstra <peterz@infradead.org>
To: Waiman Long <longman@redhat.com>
Cc: Ingo Molnar <mingo@redhat.com>, Will Deacon <will.deacon@arm.com>,
Thomas Gleixner <tglx@linutronix.de>,
linux-kernel@vger.kernel.org, x86@kernel.org,
Davidlohr Bueso <dave@stgolabs.net>,
Linus Torvalds <torvalds@linux-foundation.org>,
Tim Chen <tim.c.chen@linux.intel.com>,
huang ying <huang.ying.caritas@gmail.com>
Subject: Re: [PATCH-tip v3 04/14] locking/rwsem: Implement lock handoff to prevent lock starvation
Date: Wed, 10 Apr 2019 21:38:44 +0200 [thread overview]
Message-ID: <20190410193844.GY4038@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <20190410184231.6748-5-longman@redhat.com>
Hurph, I was still looking at v2.. I suppose I'll go stare at this
verison, I don't think you said there were many changes, right?
This version seems to still suffer that HANDOFF issue I found on v2.
On Wed, Apr 10, 2019 at 02:42:21PM -0400, Waiman Long wrote:
> Because of writer lock stealing, it is possible that a constant
> stream of incoming writers will cause a waiting writer or reader to
> wait indefinitely leading to lock starvation.
>
> The mutex code has a lock handoff mechanism to prevent lock starvation.
> This patch implements a similar lock handoff mechanism to disable
> lock stealing and force lock handoff to the first waiter in the queue
> after at least a 4ms waiting period unless it is a RT writer task which
> doesn't need to wait. The waiting period is used to avoid discouraging
> lock stealing too much to affect performance.
>
> A rwsem microbenchmark was run for 5 seconds on a 2-socket 40-core
> 80-thread Skylake system with a v5.1 based kernel and 240 write_lock
> threads with 5us sleep critical section.
>
> Before the patch, the min/mean/max numbers of locking operations for
> the locking threads were 1/7,792/173,696. After the patch, the figures
> became 5,842/6,542/7,458. It can be seen that the rwsem became much
> more fair, though there was a drop of about 16% in the mean locking
> operations done which was a tradeoff of having better fairness.
>
> Making the waiter set the handoff bit right after the first wakeup can
What does 'right after the first wakeup' mean? If that the top-waiter
setting it if it fails to acquire the lock due to steals?
> impact performance especially with a mixed reader/writer workload. With
> the same microbenchmark with short critical section and equal number of
> reader and writer threads (40/40), the reader/writer locking operation
> counts with the current patch were:
>
> 40 readers, Iterations Min/Mean/Max = 1,793/1,794/1,796
> 40 writers, Iterations Min/Mean/Max = 1,793/34,956/86,081
>
> By making waiter set handoff bit immediately after wakeup:
>
> 40 readers, Iterations Min/Mean/Max = 43/44/46
> 40 writers, Iterations Min/Mean/Max = 43/1,263/3,191
next prev parent reply other threads:[~2019-04-10 19:38 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-04-10 18:42 [PATCH-tip v3 00/14] locking/rwsem: Rwsem rearchitecture part 2 Waiman Long
2019-04-10 18:42 ` [PATCH-tip v3 01/14] locking/rwsem: Prevent unneeded warning during locking selftest Waiman Long
2019-04-10 18:42 ` [PATCH-tip v3 02/14] locking/rwsem: Make owner available even if !CONFIG_RWSEM_SPIN_ON_OWNER Waiman Long
2019-04-11 8:12 ` Peter Zijlstra
2019-04-11 16:03 ` Waiman Long
2019-04-12 7:02 ` Ingo Molnar
2019-04-12 7:05 ` Peter Zijlstra
2019-04-12 7:09 ` Ingo Molnar
2019-04-12 14:04 ` Waiman Long
2019-04-12 14:07 ` Waiman Long
2019-04-12 14:22 ` Waiman Long
2019-04-12 16:41 ` Ingo Molnar
2019-04-12 18:05 ` Waiman Long
2019-04-13 2:24 ` Waiman Long
2019-04-15 13:43 ` Waiman Long
2019-04-16 7:53 ` Ingo Molnar
2019-04-12 18:21 ` Peter Zijlstra
2019-04-10 18:42 ` [PATCH-tip v3 03/14] locking/rwsem: Implement a new locking scheme Waiman Long
2019-04-10 18:42 ` [PATCH-tip v3 04/14] locking/rwsem: Implement lock handoff to prevent lock starvation Waiman Long
2019-04-10 19:38 ` Peter Zijlstra [this message]
2019-04-10 20:28 ` Waiman Long
2019-04-10 18:42 ` [PATCH-tip v3 05/14] locking/rwsem: Remove rwsem_wake() wakeup optimization Waiman Long
2019-04-11 7:25 ` Peter Zijlstra
2019-04-11 15:55 ` Waiman Long
2019-04-10 18:42 ` [PATCH-tip v3 06/14] locking/rwsem: Make rwsem_spin_on_owner() return owner state Waiman Long
2019-04-10 18:42 ` [PATCH-tip v3 07/14] locking/rwsem: Ensure an RT task will not spin on reader Waiman Long
2019-04-10 18:42 ` [PATCH-tip v3 08/14] locking/rwsem: Wake up almost all readers in wait queue Waiman Long
2019-04-10 18:42 ` [PATCH-tip v3 09/14] locking/rwsem: Enable readers spinning on writer Waiman Long
2019-04-10 18:42 ` [PATCH-tip v3 10/14] locking/rwsem: Enable time-based spinning on reader-owned rwsem Waiman Long
2019-04-10 18:42 ` [PATCH-tip v3 11/14] locking/rwsem: Add more rwsem owner access helpers Waiman Long
2019-04-10 18:42 ` [PATCH-tip v3 12/14] locking/rwsem: Guard against making count negative Waiman Long
2019-04-10 18:42 ` [PATCH-tip v3 13/14] locking/rwsem: Merge owner into count on x86-64 Waiman Long
2019-04-10 18:42 ` [PATCH-tip v3 14/14] locking/rwsem: Remove redundant computation of writer lock word Waiman Long
2019-04-11 8:37 ` [PATCH-tip v3 00/14] locking/rwsem: Rwsem rearchitecture part 2 Peter Zijlstra
2019-04-11 16:09 ` Waiman Long
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=20190410193844.GY4038@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=dave@stgolabs.net \
--cc=huang.ying.caritas@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=longman@redhat.com \
--cc=mingo@redhat.com \
--cc=tglx@linutronix.de \
--cc=tim.c.chen@linux.intel.com \
--cc=torvalds@linux-foundation.org \
--cc=will.deacon@arm.com \
--cc=x86@kernel.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
all inboxes | Powered by JetHome®