linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Leonardo Bras <leobras@redhat.com>
To: Boqun Feng <boqun.feng@gmail.com>
Cc: Leonardo Bras <leobras@redhat.com>,
	Vlastimil Babka <vbabka@suse.cz>,
	Johannes Weiner <hannes@cmpxchg.org>,
	Michal Hocko <mhocko@kernel.org>,
	Roman Gushchin <roman.gushchin@linux.dev>,
	Shakeel Butt <shakeel.butt@linux.dev>,
	Muchun Song <muchun.song@linux.dev>,
	Andrew Morton <akpm@linux-foundation.org>,
	Christoph Lameter <cl@linux.com>,
	Pekka Enberg <penberg@kernel.org>,
	David Rientjes <rientjes@google.com>,
	Joonsoo Kim <iamjoonsoo.kim@lge.com>,
	Hyeonggon Yoo <42.hyeyoo@gmail.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Marcelo Tosatti <mtosatti@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@redhat.com>, Will Deacon <will@kernel.org>,
	Waiman Long <longman@redhat.com>,
	linux-kernel@vger.kernel.org, cgroups@vger.kernel.org,
	linux-mm@kvack.org
Subject: Re: [RFC PATCH v1 0/4] Introduce QPW for per-cpu operations
Date: Mon, 24 Jun 2024 23:57:57 -0300	[thread overview]
Message-ID: <ZnoyNQLQdyAcMxjP@LeoBras> (raw)
In-Reply-To: <Znn5FgqoCAUAfQhu@boqun-archlinux>

On Mon, Jun 24, 2024 at 03:54:14PM -0700, Boqun Feng wrote:
> On Mon, Jun 24, 2024 at 09:31:51AM +0200, Vlastimil Babka wrote:
> > Hi,
> > 
> > you've included tglx, which is great, but there's also LOCKING PRIMITIVES
> > section in MAINTAINERS so I've added folks from there in my reply.
> 
> Thanks!
> 
> > Link to full series:
> > https://lore.kernel.org/all/20240622035815.569665-1-leobras@redhat.com/
> > 
> 
> And apologies to Leonardo... I think this is a follow-up of:
> 
> 	https://lpc.events/event/17/contributions/1484/
> 
> and I did remember we had a quick chat after that which I suggested it's
> better to change to a different name, sorry that I never found time to
> write a proper rely to your previous seriese [1] as promised.
> 
> [1]: https://lore.kernel.org/lkml/20230729083737.38699-2-leobras@redhat.com/

That's correct, I commented about this in the end of above presentation.
Don't worry, and thanks for suggesting the per-cpu naming, it was very 
helpful on designing this solution.

> 
> > On 6/22/24 5:58 AM, Leonardo Bras wrote:
> > > The problem:
> > > Some places in the kernel implement a parallel programming strategy
> > > consisting on local_locks() for most of the work, and some rare remote
> > > operations are scheduled on target cpu. This keeps cache bouncing low since
> > > cacheline tends to be mostly local, and avoids the cost of locks in non-RT
> > > kernels, even though the very few remote operations will be expensive due
> > > to scheduling overhead.
> > > 
> > > On the other hand, for RT workloads this can represent a problem: getting
> > > an important workload scheduled out to deal with remote requests is
> > > sure to introduce unexpected deadline misses.
> > > 
> > > The idea:
> > > Currently with PREEMPT_RT=y, local_locks() become per-cpu spinlocks.
> > > In this case, instead of scheduling work on a remote cpu, it should
> > > be safe to grab that remote cpu's per-cpu spinlock and run the required
> > > work locally. Tha major cost, which is un/locking in every local function,
> > > already happens in PREEMPT_RT.
> > 
> > I've also noticed this a while ago (likely in the context of rewriting SLUB
> > to use local_lock) and asked about it on IRC, and IIRC tglx wasn't fond of
> > the idea. But I forgot the details about why, so I'll let the the locking
> > experts reply...
> > 
> 
> I think it's a good idea, especially the new name is less confusing ;-)
> So I wonder Thomas' thoughts as well.

Thanks!

> 
> And I think a few (micro-)benchmark numbers will help.

Last year I got some numbers on how replacing local_locks with 
spinlocks would impact memcontrol.c cache operations:

https://lore.kernel.org/all/20230125073502.743446-1-leobras@redhat.com/

tl;dr: It increased clocks spent in the most common this_cpu operations, 
while reducing clocks spent in remote operations (drain_all_stock).

In RT case, since local locks are already spinlocks, this cost is 
already paid, so we can get results like these:

drain_all_stock
cpus	Upstream 	Patched		Diff (cycles)	Diff(%)
1	44331.10831	38978.03581	-5353.072507	-12.07520567
8	43992.96512	39026.76654	-4966.198572	-11.2886198
128	156274.6634	58053.87421	-98220.78915	-62.85138425

Upstream: Clocks to schedule work on remote CPU (performing not accounted)
Patched:  Clocks to grab remote cpu's spinlock and perform the needed work 
	  locally.

Do you have other suggestions to use as (micro-) benchmarking?

Thanks!
Leo


> 
> Regards,
> Boqun
> 
> > > Also, there is no need to worry about extra cache bouncing:
> > > The cacheline invalidation already happens due to schedule_work_on().
> > > 
> > > This will avoid schedule_work_on(), and thus avoid scheduling-out an 
> > > RT workload. 
> > > 
> > > For patches 2, 3 & 4, I noticed just grabing the lock and executing
> > > the function locally is much faster than just scheduling it on a
> > > remote cpu.
> > > 
> > > Proposed solution:
> > > A new interface called Queue PerCPU Work (QPW), which should replace
> > > Work Queue in the above mentioned use case. 
> > > 
> > > If PREEMPT_RT=n, this interfaces just wraps the current 
> > > local_locks + WorkQueue behavior, so no expected change in runtime.
> > > 
> > > If PREEMPT_RT=y, queue_percpu_work_on(cpu,...) will lock that cpu's
> > > per-cpu structure and perform work on it locally. This is possible
> > > because on functions that can be used for performing remote work on
> > > remote per-cpu structures, the local_lock (which is already
> > > a this_cpu spinlock()), will be replaced by a qpw_spinlock(), which
> > > is able to get the per_cpu spinlock() for the cpu passed as parameter.
> > > 
> > > Patch 1 implements QPW interface, and patches 2, 3 & 4 replaces the
> > > current local_lock + WorkQueue interface by the QPW interface in
> > > swap, memcontrol & slub interface.
> > > 
> > > Please let me know what you think on that, and please suggest
> > > improvements.
> > > 
> > > Thanks a lot!
> > > Leo
> > > 
> > > Leonardo Bras (4):
> > >   Introducing qpw_lock() and per-cpu queue & flush work
> > >   swap: apply new queue_percpu_work_on() interface
> > >   memcontrol: apply new queue_percpu_work_on() interface
> > >   slub: apply new queue_percpu_work_on() interface
> > > 
> > >  include/linux/qpw.h | 88 +++++++++++++++++++++++++++++++++++++++++++++
> > >  mm/memcontrol.c     | 20 ++++++-----
> > >  mm/slub.c           | 26 ++++++++------
> > >  mm/swap.c           | 26 +++++++-------
> > >  4 files changed, 127 insertions(+), 33 deletions(-)
> > >  create mode 100644 include/linux/qpw.h
> > > 
> > > 
> > > base-commit: 50736169ecc8387247fe6a00932852ce7b057083
> > 
> 



  reply	other threads:[~2024-06-25  2:58 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-22  3:58 [RFC PATCH v1 0/4] Introduce QPW for per-cpu operations Leonardo Bras
2024-06-22  3:58 ` [RFC PATCH v1 1/4] Introducing qpw_lock() and per-cpu queue & flush work Leonardo Bras
2024-09-04 21:39   ` Waiman Long
2024-09-05  0:08     ` Waiman Long
2024-09-11  7:18       ` Leonardo Bras
2024-09-11  7:17     ` Leonardo Bras
2024-09-11 13:39       ` Waiman Long
2024-06-22  3:58 ` [RFC PATCH v1 2/4] swap: apply new queue_percpu_work_on() interface Leonardo Bras
2024-06-22  3:58 ` [RFC PATCH v1 3/4] memcontrol: " Leonardo Bras
2024-06-22  3:58 ` [RFC PATCH v1 4/4] slub: " Leonardo Bras
2024-06-24  7:31 ` [RFC PATCH v1 0/4] Introduce QPW for per-cpu operations Vlastimil Babka
2024-06-24 22:54   ` Boqun Feng
2024-06-25  2:57     ` Leonardo Bras [this message]
2024-06-25 17:51       ` Boqun Feng
2024-06-26 16:40         ` Leonardo Bras
2024-06-28 18:47       ` Marcelo Tosatti
2024-06-25  2:36   ` Leonardo Bras
2024-07-15 18:38   ` Marcelo Tosatti
2024-07-23 17:14 ` Marcelo Tosatti
2024-09-05 22:19   ` Hillf Danton
2024-09-11  3:04     ` Marcelo Tosatti
2024-09-15  0:30       ` Hillf Danton
2024-09-11  6:42     ` Leonardo Bras

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=ZnoyNQLQdyAcMxjP@LeoBras \
    --to=leobras@redhat.com \
    --cc=42.hyeyoo@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=boqun.feng@gmail.com \
    --cc=cgroups@vger.kernel.org \
    --cc=cl@linux.com \
    --cc=hannes@cmpxchg.org \
    --cc=iamjoonsoo.kim@lge.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=longman@redhat.com \
    --cc=mhocko@kernel.org \
    --cc=mingo@redhat.com \
    --cc=mtosatti@redhat.com \
    --cc=muchun.song@linux.dev \
    --cc=penberg@kernel.org \
    --cc=peterz@infradead.org \
    --cc=rientjes@google.com \
    --cc=roman.gushchin@linux.dev \
    --cc=shakeel.butt@linux.dev \
    --cc=tglx@linutronix.de \
    --cc=vbabka@suse.cz \
    --cc=will@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;
as well as URLs for NNTP newsgroup(s).