From: Uladzislau Rezki <urezki@gmail.com>
To: Dave Chinner <david@fromorbit.com>
Cc: "Uladzislau Rezki (Sony)" <urezki@gmail.com>,
linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>,
LKML <linux-kernel@vger.kernel.org>, Baoquan He <bhe@redhat.com>,
Lorenzo Stoakes <lstoakes@gmail.com>,
Christoph Hellwig <hch@infradead.org>,
Matthew Wilcox <willy@infradead.org>,
"Liam R . Howlett" <Liam.Howlett@oracle.com>,
"Paul E . McKenney" <paulmck@kernel.org>,
Joel Fernandes <joel@joelfernandes.org>,
Oleksiy Avramchenko <oleksiy.avramchenko@sony.com>
Subject: Re: [PATCH v3 10/11] mm: vmalloc: Set nr_nodes based on CPUs in a system
Date: Mon, 15 Jan 2024 20:09:29 +0100 [thread overview]
Message-ID: <ZaWC6TRs4P1vq9TQ@pc636> (raw)
In-Reply-To: <ZZ+z8vBl645FvxPq@dread.disaster.area>
> On Tue, Jan 02, 2024 at 07:46:32PM +0100, Uladzislau Rezki (Sony) wrote:
> > A number of nodes which are used in the alloc/free paths is
> > set based on num_possible_cpus() in a system. Please note a
> > high limit threshold though is fixed and corresponds to 128
> > nodes.
>
> Large CPU count machines are NUMA machines. ALl of the allocation
> and reclaim is NUMA node based i.e. a pgdat per NUMA node.
>
> Shrinkers are also able to be run in a NUMA aware mode so that
> per-node structures can be reclaimed similar to how per-node LRU
> lists are scanned for reclaim.
>
> Hence I'm left to wonder if it would be better to have a vmalloc
> area per pgdat (or sub-node cluster) rather than just base the
> number on CPU count and then have an arbitrary maximum number when
> we get to 128 CPU cores. We can have 128 CPU cores in a
> single socket these days, so not being able to scale the vmalloc
> areas beyond a single socket seems like a bit of a limitation.
>
>
> Hence I'm left to wonder if it would be better to have a vmalloc
> area per pgdat (or sub-node cluster) rather than just base the
>
> Scaling out the vmalloc areas in a NUMA aware fashion allows the
> shrinker to be run in numa aware mode, which gets rid of the need
> for the global shrinker to loop over every single vmap area in every
> shrinker invocation. Only the vm areas on the node that has a memory
> shortage need to be scanned and reclaimed, it doesn't need reclaim
> everything globally when a single node runs out of memory.
>
> Yes, this may not give quite as good microbenchmark scalability
> results, but being able to locate each vm area in node local memory
> and have operation on them largely isolated to node-local tasks and
> vmalloc area reclaim will work much better on large multi-socket
> NUMA machines.
>
Currently i fix the max nodes number to 128. This is because i do not
have an access to such big NUMA systems whereas i do have an access to
around ~128 ones. That is why i have decided to stop on that number as
of now.
We can easily set nr_nodes to num_possible_cpus() and let it scale for
anyone. But before doing this, i would like to give it a try as a first
step because i have not tested it well on really big NUMA systems.
Thanks for you NUMA-aware input.
--
Uladzislau Rezki
next prev parent reply other threads:[~2024-01-15 19:09 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-02 18:46 [PATCH v3 00/11] Mitigate a vmap lock contention v3 Uladzislau Rezki (Sony)
2024-01-02 18:46 ` [PATCH v3 01/11] mm: vmalloc: Add va_alloc() helper Uladzislau Rezki (Sony)
2024-01-02 18:46 ` [PATCH v3 02/11] mm: vmalloc: Rename adjust_va_to_fit_type() function Uladzislau Rezki (Sony)
2024-01-02 18:46 ` [PATCH v3 03/11] mm: vmalloc: Move vmap_init_free_space() down in vmalloc.c Uladzislau Rezki (Sony)
2024-01-02 18:46 ` [PATCH v3 04/11] mm: vmalloc: Remove global vmap_area_root rb-tree Uladzislau Rezki (Sony)
2024-01-05 8:10 ` Wen Gu
2024-01-05 10:50 ` Uladzislau Rezki
2024-01-06 9:17 ` Wen Gu
2024-01-06 16:36 ` Uladzislau Rezki
2024-01-07 6:59 ` Hillf Danton
2024-01-08 7:45 ` Wen Gu
2024-01-08 18:37 ` Uladzislau Rezki
2024-01-16 23:25 ` Lorenzo Stoakes
2024-01-18 13:15 ` Uladzislau Rezki
2024-01-20 12:55 ` Lorenzo Stoakes
2024-01-22 17:44 ` Uladzislau Rezki
2024-01-02 18:46 ` [PATCH v3 05/11] mm/vmalloc: remove vmap_area_list Uladzislau Rezki (Sony)
2024-01-16 23:36 ` Lorenzo Stoakes
2024-01-02 18:46 ` [PATCH v3 06/11] mm: vmalloc: Remove global purge_vmap_area_root rb-tree Uladzislau Rezki (Sony)
2024-01-02 18:46 ` [PATCH v3 07/11] mm: vmalloc: Offload free_vmap_area_lock lock Uladzislau Rezki (Sony)
2024-01-03 11:08 ` Hillf Danton
2024-01-03 15:47 ` Uladzislau Rezki
2024-01-11 9:02 ` Dave Chinner
2024-01-11 15:54 ` Uladzislau Rezki
2024-01-11 20:37 ` Dave Chinner
2024-01-12 12:18 ` Uladzislau Rezki
2024-01-16 22:12 ` Dave Chinner
2024-01-18 18:15 ` Uladzislau Rezki
2024-02-08 0:25 ` Baoquan He
2024-02-08 13:57 ` Uladzislau Rezki
2024-02-28 9:48 ` Baoquan He
2024-02-28 10:39 ` Uladzislau Rezki
2024-02-28 12:26 ` Baoquan He
2024-03-22 18:21 ` Guenter Roeck
2024-03-22 19:03 ` Uladzislau Rezki
2024-03-22 20:53 ` Guenter Roeck
2024-01-02 18:46 ` [PATCH v3 08/11] mm: vmalloc: Support multiple nodes in vread_iter Uladzislau Rezki (Sony)
2024-01-02 18:46 ` [PATCH v3 09/11] mm: vmalloc: Support multiple nodes in vmallocinfo Uladzislau Rezki (Sony)
2024-01-02 18:46 ` [PATCH v3 10/11] mm: vmalloc: Set nr_nodes based on CPUs in a system Uladzislau Rezki (Sony)
2024-01-11 9:25 ` Dave Chinner
2024-01-15 19:09 ` Uladzislau Rezki [this message]
2024-01-16 22:06 ` Dave Chinner
2024-01-18 18:23 ` Uladzislau Rezki
2024-01-18 21:28 ` Dave Chinner
2024-01-19 10:32 ` Uladzislau Rezki
2024-01-02 18:46 ` [PATCH v3 11/11] mm: vmalloc: Add a shrinker to drain vmap pools Uladzislau Rezki (Sony)
2024-02-22 8:35 ` [PATCH v3 00/11] Mitigate a vmap lock contention v3 Uladzislau Rezki
2024-02-22 23:15 ` Pedro Falcato
2024-02-23 9:34 ` Uladzislau Rezki
2024-02-23 10:26 ` Baoquan He
2024-02-23 11:06 ` Uladzislau Rezki
2024-02-23 15:57 ` Baoquan He
2024-02-23 18:55 ` Uladzislau Rezki
2024-02-28 9:27 ` Baoquan He
2024-02-29 10:38 ` Uladzislau Rezki
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=ZaWC6TRs4P1vq9TQ@pc636 \
--to=urezki@gmail.com \
--cc=Liam.Howlett@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=bhe@redhat.com \
--cc=david@fromorbit.com \
--cc=hch@infradead.org \
--cc=joel@joelfernandes.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lstoakes@gmail.com \
--cc=oleksiy.avramchenko@sony.com \
--cc=paulmck@kernel.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).