From: Jakub Kicinski <kuba@kernel.org>
To: Thomas Karlsson <thomas.karlsson@paneda.se>
Cc: "netdev@vger.kernel.org" <netdev@vger.kernel.org>,
"davem@davemloft.net" <davem@davemloft.net>, <jiri@resnulli.us>,
<kaber@trash.net>, <edumazet@google.com>, <vyasevic@redhat.com>,
<alexander.duyck@gmail.com>
Subject: Re: [PATCH net-next v4] macvlan: Support for high multicast packet rate
Date: Thu, 3 Dec 2020 08:20:38 -0800 [thread overview]
Message-ID: <20201203082038.3dab8511@kicinski-fedora-pc1c0hjn.DHCP.thefacebook.com> (raw)
In-Reply-To: <dd4673b2-7eab-edda-6815-85c67ce87f63@paneda.se>
On Wed, 2 Dec 2020 19:49:58 +0100 Thomas Karlsson wrote:
> Background:
> Broadcast and multicast packages are enqueued for later processing.
> This queue was previously hardcoded to 1000.
>
> This proved insufficient for handling very high packet rates.
> This resulted in packet drops for multicast.
> While at the same time unicast worked fine.
>
> The change:
> This patch make the queue length adjustable to accommodate
> for environments with very high multicast packet rate.
> But still keeps the default value of 1000 unless specified.
>
> The queue length is specified as a request per macvlan
> using the IFLA_MACVLAN_BC_QUEUE_LEN parameter.
>
> The actual used queue length will then be the maximum of
> any macvlan connected to the same port. The actual used
> queue length for the port can be retrieved (read only)
> by the IFLA_MACVLAN_BC_QUEUE_LEN_USED parameter for verification.
>
> This will be followed up by a patch to iproute2
> in order to adjust the parameter from userspace.
>
> Signed-off-by: Thomas Karlsson <thomas.karlsson@paneda.se>
> @@ -1658,6 +1680,8 @@ static const struct nla_policy macvlan_policy[IFLA_MACVLAN_MAX + 1] = {
> [IFLA_MACVLAN_MACADDR] = { .type = NLA_BINARY, .len = MAX_ADDR_LEN },
> [IFLA_MACVLAN_MACADDR_DATA] = { .type = NLA_NESTED },
> [IFLA_MACVLAN_MACADDR_COUNT] = { .type = NLA_U32 },
> + [IFLA_MACVLAN_BC_QUEUE_LEN] = { .type = NLA_U32 },
I wonder whether we should require that the queue_len is > 0? Is there
a valid use case for the queue to be completely disabled? If you agree
please follow up with a simple patch which adds a NLA_POLICY_MIN() here.
Applied to net-next, thank you!
next prev parent reply other threads:[~2020-12-03 16:21 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <485531aec7e243659ee4e3bb7fa2186d@paneda.se>
2020-11-23 14:22 ` Hardcoded multicast queue length in macvlan.c driver causes poor multicast receive performance Thomas Karlsson
2020-11-23 22:30 ` Jakub Kicinski
2020-11-25 12:51 ` Thomas Karlsson
2020-11-25 16:57 ` [PATCH] macvlan: Support for high multicast packet rate Thomas Karlsson
2020-11-25 21:55 ` [PATCH net-next v2] " Thomas Karlsson
2020-11-25 16:58 ` Hardcoded multicast queue length in macvlan.c driver causes poor multicast receive performance Jakub Kicinski
2020-11-25 17:12 ` Thomas Karlsson
2020-11-25 18:07 ` Jakub Kicinski
2020-11-25 22:15 ` Thomas Karlsson
2020-11-25 23:01 ` Jakub Kicinski
2020-11-26 20:00 ` Thomas Karlsson
2020-11-27 17:27 ` Jakub Kicinski
2020-11-27 23:13 ` Thomas Karlsson
2020-11-30 14:00 ` [PATCH net-next v3] macvlan: Support for high multicast packet rate Thomas Karlsson
2020-12-01 19:11 ` Jakub Kicinski
2020-12-02 11:28 ` Thomas Karlsson
2020-12-02 16:44 ` Jakub Kicinski
2020-11-30 14:23 ` [PATCH iproute2-next v1] iplink macvlan: Added bcqueuelen parameter Thomas Karlsson
2020-12-10 16:07 ` Thomas Karlsson
2020-12-11 0:36 ` Jakub Kicinski
2020-12-02 18:49 ` [PATCH net-next v4] macvlan: Support for high multicast packet rate Thomas Karlsson
2020-12-03 16:20 ` Jakub Kicinski [this message]
2020-12-03 16:53 ` Thomas Karlsson
2020-12-14 10:42 ` [PATCH iproute2-next v2] iplink:macvlan: Added bcqueuelen parameter Thomas Karlsson
2020-12-14 17:03 ` David Ahern
2020-12-14 17:50 ` Thomas Karlsson
2020-12-16 4:07 ` David Ahern
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=20201203082038.3dab8511@kicinski-fedora-pc1c0hjn.DHCP.thefacebook.com \
--to=kuba@kernel.org \
--cc=alexander.duyck@gmail.com \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=jiri@resnulli.us \
--cc=kaber@trash.net \
--cc=netdev@vger.kernel.org \
--cc=thomas.karlsson@paneda.se \
--cc=vyasevic@redhat.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 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).