From: Vlastimil Babka <vbabka@suse.cz>
To: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
Christoph Lameter <cl@linux.com>,
David Rientjes <rientjes@google.com>,
Pekka Enberg <penberg@kernel.org>,
Joonsoo Kim <iamjoonsoo.kim@lge.com>,
Thomas Gleixner <tglx@linutronix.de>,
Mel Gorman <mgorman@techsingularity.net>,
Jesper Dangaard Brouer <brouer@redhat.com>,
Peter Zijlstra <peterz@infradead.org>,
Jann Horn <jannh@google.com>
Subject: Re: [RFC v2 00/34] SLUB: reduce irq disabled scope and make it RT compatible
Date: Fri, 2 Jul 2021 22:25:05 +0200 [thread overview]
Message-ID: <35b26e48-a96a-41b0-826e-43e43660c9d6@suse.cz> (raw)
In-Reply-To: <20210702182944.lqa7o2a25to6czju@linutronix.de>
On 7/2/21 8:29 PM, Sebastian Andrzej Siewior wrote:
> I replaced my slub changes with slub-local-lock-v2r3.
> I haven't seen any complains from lockdep or so which is good. Then I
> did this with RT enabled (and no debug):
Thanks for testing!
> - A "time make -j32" run of allmodconfig on /dev/shm.
> Old:
> | real 20m6,217s
> | user 568m22,553s
> | sys 48m33,126s
>
> New:
> | real 20m9,049s
> | user 569m32,096s
> | sys 48m47,670s
>
> These 3 seconds here are probably in the noise range.
>
> - perf_5.10 stat -r 10 hackbench -g200 -s 4096 -l500
> Old:
> | 464.967,20 msec task-clock # 27,220 CPUs utilized ( +- 0,16% )
> | 7.683.944 context-switches # 0,017 M/sec ( +- 0,86% )
> | 931.380 cpu-migrations # 0,002 M/sec ( +- 4,94% )
> | 219.569 page-faults # 0,472 K/sec ( +- 0,39% )
> | 1.104.727.599.918 cycles # 2,376 GHz ( +- 0,18% )
> | 941.428.898.087 stalled-cycles-frontend # 85,22% frontend cycles idle ( +- 0,24% )
> | 729.016.546.572 stalled-cycles-backend # 65,99% backend cycles idle ( +- 0,32% )
> | 340.133.571.519 instructions # 0,31 insn per cycle
> | # 2,77 stalled cycles per insn ( +- 0,12% )
> | 73.746.821.314 branches # 158,607 M/sec ( +- 0,13% )
> | 377.838.006 branch-misses # 0,51% of all branches ( +- 1,01% )
> |
> | 17,0820 +- 0,0202 seconds time elapsed ( +- 0,12% )
>
> New:
> | 422.865,71 msec task-clock # 4,782 CPUs utilized ( +- 0,34% )
> | 14.594.238 context-switches # 0,035 M/sec ( +- 0,43% )
> | 3.737.926 cpu-migrations # 0,009 M/sec ( +- 0,46% )
> | 218.474 page-faults # 0,517 K/sec ( +- 0,74% )
> | 940.715.812.020 cycles # 2,225 GHz ( +- 0,34% )
> | 716.593.827.820 stalled-cycles-frontend # 76,18% frontend cycles idle ( +- 0,39% )
> | 550.730.862.839 stalled-cycles-backend # 58,54% backend cycles idle ( +- 0,43% )
> | 417.274.588.907 instructions # 0,44 insn per cycle
> | # 1,72 stalled cycles per insn ( +- 0,17% )
> | 92.814.150.290 branches # 219,488 M/sec ( +- 0,17% )
> | 822.102.170 branch-misses # 0,89% of all branches ( +- 0,41% )
> |
> | 88,427 +- 0,618 seconds time elapsed ( +- 0,70% )
>
> So this is outside of the noise range.
> I'm not sure where this is coming from. My guess would be higher lock
> contention within the memory allocator.
The series shouldn't significantly change the memory allocator
interaction, though.
Seems there's less cycles, but more time elapsed, thus more sleeping -
is it locks becoming mutexes on RT?
My first guess - the last, local_lock patch. What would happen if you
take that one out? Should be still RT-compatible. If it improves a lot,
maybe that conversion to local_lock is not worth it then.
My second guess - list_lock remains spinlock with my series, thus RT
mutex, but the current RT tree converts it to raw_spinlock. I'd hope
leaving that one as non-raw spinlock would still be much better for RT
goals, even if hackbench (which is AFAIK very slab intensive) throughput
regresses - hopefully not that much.
>> The remaining patches to upstream from the RT tree are small ones related to
>> KConfig. The patch that restricts PREEMPT_RT to SLUB (not SLAB or SLOB) makes
>> sense. The patch that disables CONFIG_SLUB_CPU_PARTIAL with PREEMPT_RT could
>> perhaps be re-evaluated as the series also addresses some latency issues with
>> percpu partial slabs.
>
> With that series the PARTIAL slab can be indeed enabled. I have (had) a
> half done series where I had PARTIAL enabled and noticed a slight
> increase in latency so made it "default y on !RT". It wasn't dramatic
> but appeared to be outside of noise.
>
> Sebastian
>
next prev parent reply other threads:[~2021-07-02 20:25 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-09 11:38 [RFC v2 00/34] SLUB: reduce irq disabled scope and make it RT compatible Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 01/34] mm, slub: don't call flush_all() from list_locations() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 02/34] mm, slub: allocate private object map for sysfs listings Vlastimil Babka
2021-06-09 13:29 ` Christoph Lameter
2021-06-09 11:38 ` [RFC v2 03/34] mm, slub: allocate private object map for validate_slab_cache() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 04/34] mm, slub: don't disable irq for debug_check_no_locks_freed() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 05/34] mm, slub: remove redundant unfreeze_partials() from put_cpu_partial() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 06/34] mm, slub: unify cmpxchg_double_slab() and __cmpxchg_double_slab() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 07/34] mm, slub: extract get_partial() from new_slab_objects() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 08/34] mm, slub: dissolve new_slab_objects() into ___slab_alloc() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 09/34] mm, slub: return slab page from get_partial() and set c->page afterwards Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 10/34] mm, slub: restructure new page checks in ___slab_alloc() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 11/34] mm, slub: simplify kmem_cache_cpu and tid setup Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 12/34] mm, slub: move disabling/enabling irqs to ___slab_alloc() Vlastimil Babka
2021-07-06 4:38 ` Mike Galbraith
2021-06-09 11:38 ` [RFC v2 13/34] mm, slub: do initial checks in ___slab_alloc() with irqs enabled Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 14/34] mm, slub: move disabling irqs closer to get_partial() in ___slab_alloc() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 15/34] mm, slub: restore irqs around calling new_slab() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 16/34] mm, slub: validate slab from partial list or page allocator before making it cpu slab Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 17/34] mm, slub: check new pages with restored irqs Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 18/34] mm, slub: stop disabling irqs around get_partial() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 19/34] mm, slub: move reset of c->page and freelist out of deactivate_slab() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 20/34] mm, slub: make locking in deactivate_slab() irq-safe Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 21/34] mm, slub: call deactivate_slab() without disabling irqs Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 22/34] mm, slub: move irq control into unfreeze_partials() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 23/34] mm, slub: discard slabs in unfreeze_partials() without irqs disabled Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 24/34] mm, slub: detach whole partial list at once in unfreeze_partials() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 25/34] mm, slub: detach percpu partial list in unfreeze_partials() using this_cpu_cmpxchg() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 26/34] mm, slub: only disable irq with spin_lock in __unfreeze_partials() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 27/34] mm, slub: don't disable irqs in slub_cpu_dead() Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 28/34] mm, slab: make flush_slab() possible to call with irqs enabled Vlastimil Babka
2021-06-09 11:38 ` [RFC v2 29/34] mm: slub: Move flush_cpu_slab() invocations __free_slab() invocations out of IRQ context Vlastimil Babka
2021-06-09 22:29 ` Cyrill Gorcunov
2021-06-10 8:32 ` Vlastimil Babka
2021-06-10 8:36 ` Cyrill Gorcunov
2021-06-09 11:38 ` [RFC v2 30/34] mm: slub: Make object_map_lock a raw_spinlock_t Vlastimil Babka
2021-06-09 11:39 ` [RFC v2 31/34] mm, slub: optionally save/restore irqs in slab_[un]lock()/ Vlastimil Babka
2021-07-02 12:17 ` Sebastian Andrzej Siewior
2021-06-09 11:39 ` [RFC v2 32/34] mm, slub: make slab_lock() disable irqs with PREEMPT_RT Vlastimil Babka
2021-06-09 11:39 ` [RFC v2 33/34] mm, slub: use migrate_disable() on PREEMPT_RT Vlastimil Babka
2021-06-14 11:07 ` Vlastimil Babka
2021-06-14 11:16 ` Sebastian Andrzej Siewior
2021-06-14 11:33 ` Vlastimil Babka
2021-06-14 12:54 ` Vlastimil Babka
2021-06-14 14:01 ` Sebastian Andrzej Siewior
2021-06-09 11:39 ` [RFC v2 34/34] mm, slub: convert kmem_cpu_slab protection to local_lock Vlastimil Babka
2021-06-14 9:49 ` [RFC v2 00/34] SLUB: reduce irq disabled scope and make it RT compatible Mel Gorman
2021-06-14 11:31 ` Mel Gorman
2021-06-14 11:10 ` Vlastimil Babka
2021-07-02 18:29 ` Sebastian Andrzej Siewior
2021-07-02 20:25 ` Vlastimil Babka [this message]
2021-07-29 13:49 ` Sebastian Andrzej Siewior
2021-07-29 14:17 ` Vlastimil Babka
2021-07-29 14:37 ` Sebastian Andrzej Siewior
2021-07-03 7:24 ` Mike Galbraith
2021-07-03 15:47 ` Mike Galbraith
2021-07-04 5:37 ` Mike Galbraith
2021-07-18 7:41 ` Vlastimil Babka
2021-07-18 8:29 ` Mike Galbraith
2021-07-18 12:09 ` Mike Galbraith
2021-07-05 16:00 ` Mike Galbraith
2021-07-06 17:56 ` Mike Galbraith
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=35b26e48-a96a-41b0-826e-43e43660c9d6@suse.cz \
--to=vbabka@suse.cz \
--cc=bigeasy@linutronix.de \
--cc=brouer@redhat.com \
--cc=cl@linux.com \
--cc=iamjoonsoo.kim@lge.com \
--cc=jannh@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@techsingularity.net \
--cc=penberg@kernel.org \
--cc=peterz@infradead.org \
--cc=rientjes@google.com \
--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;
as well as URLs for NNTP newsgroup(s).