From: Jesper Dangaard Brouer <brouer@redhat.com>
To: Michal Hocko <mhocko@kernel.org>
Cc: netdev@vger.kernel.org, lirongqing@baidu.com,
linyunsheng@huawei.com,
Ilias Apalodimas <ilias.apalodimas@linaro.org>,
Saeed Mahameed <saeedm@mellanox.com>,
peterz@infradead.org, linux-kernel@vger.kernel.org,
brouer@redhat.com
Subject: Re: [net-next v4 PATCH] page_pool: handle page recycle for NUMA_NO_NODE condition
Date: Thu, 19 Dec 2019 14:35:35 +0100 [thread overview]
Message-ID: <20191219143535.6c7bc880@carbon> (raw)
In-Reply-To: <20191219120925.GD26945@dhcp22.suse.cz>
On Thu, 19 Dec 2019 13:09:25 +0100
Michal Hocko <mhocko@kernel.org> wrote:
> On Wed 18-12-19 09:01:35, Jesper Dangaard Brouer wrote:
> [...]
> > For the NUMA_NO_NODE case, when a NIC IRQ is moved to another NUMA
> > node, then ptr_ring will be emptied in 65 (PP_ALLOC_CACHE_REFILL+1)
> > chunks per allocation and allocation fall-through to the real
> > page-allocator with the new nid derived from numa_mem_id(). We accept
> > that transitioning the alloc cache doesn't happen immediately.
Oh, I just realized that the drivers usually refill several RX
packet-pages at once, this means that this is called N times, meaning
during a NUMA change this will result in N * 65 pages returned.
> Could you explain what is the expected semantic of NUMA_NO_NODE in this
> case? Does it imply always the preferred locality? See my other email[1] to
> this matter.
I do think we want NUMA_NO_NODE to mean preferred locality. My code
allow the page to come from a remote NUMA node, but once it is returned
via the ptr_ring, we return pages not belonging to the local NUMA node
(determined by the CPU processing RX packets from the drivers RX-ring).
> [1] http://lkml.kernel.org/r/20191219115338.GC26945@dhcp22.suse.cz
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
LinkedIn: http://www.linkedin.com/in/brouer
next prev parent reply other threads:[~2019-12-19 13:35 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-12-17 23:17 [net-next v3 PATCH] page_pool: handle page recycle for NUMA_NO_NODE condition Jesper Dangaard Brouer
2019-12-18 7:44 ` Jesper Dangaard Brouer
2019-12-18 8:01 ` [net-next v4 " Jesper Dangaard Brouer
2019-12-18 14:27 ` 答复: " Li,Rongqing
2019-12-19 12:00 ` Jesper Dangaard Brouer
2019-12-19 12:47 ` 答复: " Li,Rongqing
2019-12-19 1:52 ` Yunsheng Lin
2019-12-19 12:15 ` Jesper Dangaard Brouer
2019-12-19 12:09 ` Michal Hocko
2019-12-19 13:35 ` Jesper Dangaard Brouer [this message]
2019-12-19 14:52 ` Michal Hocko
2019-12-19 15:28 ` Ilias Apalodimas
2019-12-19 14:20 ` [net-next v5 " Jesper Dangaard Brouer
2019-12-20 10:23 ` Ilias Apalodimas
2019-12-20 10:41 ` Jesper Dangaard Brouer
2019-12-20 10:49 ` Ilias Apalodimas
2019-12-20 15:22 ` Jesper Dangaard Brouer
2019-12-20 16:06 ` Ilias Apalodimas
2019-12-23 7:57 ` Ilias Apalodimas
2019-12-23 16:52 ` Jesper Dangaard Brouer
2019-12-23 22:10 ` Saeed Mahameed
2019-12-24 9:34 ` Ilias Apalodimas
2019-12-24 7:41 ` Ilias Apalodimas
2019-12-20 21:27 ` Saeed Mahameed
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=20191219143535.6c7bc880@carbon \
--to=brouer@redhat.com \
--cc=ilias.apalodimas@linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linyunsheng@huawei.com \
--cc=lirongqing@baidu.com \
--cc=mhocko@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=peterz@infradead.org \
--cc=saeedm@mellanox.com \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.