From: Greg KH <gregkh@linuxfoundation.org>
To: "Pali Rohár" <pali@kernel.org>
Cc: stable@vger.kernel.org,
"Alexander Lobakin" <bloodyreaper@yandex.ru>,
"David S . Miller" <davem@davemloft.net>,
"Marek Behún" <kabel@kernel.org>
Subject: Re: [PATCH] net: dsa: add GRO support via gro_cells
Date: Tue, 9 Mar 2021 11:38:19 +0100 [thread overview]
Message-ID: <YEdQG34U32LQW+R+@kroah.com> (raw)
In-Reply-To: <20210309102455.qxwu7sokv2qhcz46@pali>
On Tue, Mar 09, 2021 at 11:24:55AM +0100, Pali Rohár wrote:
> On Monday 08 March 2021 20:55:20 Greg KH wrote:
> > On Mon, Mar 08, 2021 at 06:57:57PM +0100, Pali Rohár wrote:
> > > From: Alexander Lobakin <bloodyreaper@yandex.ru>
> > >
> > > commit e131a5634830047923c694b4ce0c3b31745ff01b upstream.
> > >
> > > gro_cells lib is used by different encapsulating netdevices, such as
> > > geneve, macsec, vxlan etc. to speed up decapsulated traffic processing.
> > > CPU tag is a sort of "encapsulation", and we can use the same mechs to
> > > greatly improve overall DSA performance.
> > > skbs are passed to the GRO layer after removing CPU tags, so we don't
> > > need any new packet offload types as it was firstly proposed by me in
> > > the first GRO-over-DSA variant [1].
> > >
> > > The size of struct gro_cells is sizeof(void *), so hot struct
> > > dsa_slave_priv becomes only 4/8 bytes bigger, and all critical fields
> > > remain in one 32-byte cacheline.
> > > The other positive side effect is that drivers for network devices
> > > that can be shipped as CPU ports of DSA-driven switches can now use
> > > napi_gro_frags() to pass skbs to kernel. Packets built that way are
> > > completely non-linear and are likely being dropped without GRO.
> > >
> > > This was tested on to-be-mainlined-soon Ethernet driver that uses
> > > napi_gro_frags(), and the overall performance was on par with the
> > > variant from [1], sometimes even better due to minimal overhead.
> > > net.core.gro_normal_batch tuning may help to push it to the limit
> > > on particular setups and platforms.
> > >
> > > iperf3 IPoE VLAN NAT TCP forwarding (port1.218 -> port0) setup
> > > on 1.2 GHz MIPS board:
> > >
> > > 5.7-rc2 baseline:
> > >
> > > [ID] Interval Transfer Bitrate Retr
> > > [ 5] 0.00-120.01 sec 9.00 GBytes 644 Mbits/sec 413 sender
> > > [ 5] 0.00-120.00 sec 8.99 GBytes 644 Mbits/sec receiver
> > >
> > > Iface RX packets TX packets
> > > eth0 7097731 7097702
> > > port0 426050 6671829
> > > port1 6671681 425862
> > > port1.218 6671677 425851
> > >
> > > With this patch:
> > >
> > > [ID] Interval Transfer Bitrate Retr
> > > [ 5] 0.00-120.01 sec 12.2 GBytes 870 Mbits/sec 122 sender
> > > [ 5] 0.00-120.00 sec 12.2 GBytes 870 Mbits/sec receiver
> > >
> > > Iface RX packets TX packets
> > > eth0 9474792 9474777
> > > port0 455200 353288
> > > port1 9019592 455035
> > > port1.218 353144 455024
> > >
> > > v2:
> > > - Add some performance examples in the commit message;
> > > - No functional changes.
> > >
> > > [1] https://lore.kernel.org/netdev/20191230143028.27313-1-alobakin@dlink.ru/
> > >
> > > Signed-off-by: Alexander Lobakin <bloodyreaper@yandex.ru>
> > > Signed-off-by: David S. Miller <davem@davemloft.net>
> > >
> > > ---
> > > This patch radically increase network performance on DSA setup.
> > >
> > > Please include this patch into stable releases.
> > >
> > > I have done following tests:
> > >
> > > NAT is a tested Espressobin board (ARM64 Marvell Armada 3720 SoC with
> > > Marvell 88E6141 DSA switch) which was configured for IPv4 masquerade.
> > > WAN and LAN are another two static boxes on which was running iperf3.
> > >
> > > 4.19.179 without e131a5634830047923c694b4ce0c3b31745ff01b
> > >
> > > WAN --> NAT --> LAN
> > > [ ID] Interval Transfer Bitrate Retr
> > > [ 5] 0.00-10.01 sec 440 MBytes 369 Mbits/sec 12 sender
> > > [ 5] 0.00-10.00 sec 437 MBytes 367 Mbits/sec receiver
> > >
> > > WAN <-- NAT <-- LAN
> > > [ ID] Interval Transfer Bitrate Retr
> > > [ 5] 0.00-10.00 sec 390 MBytes 327 Mbits/sec 90 sender
> > > [ 5] 0.00-10.01 sec 388 MBytes 326 Mbits/sec receiver
> > >
> > > 4.19.179 with e131a5634830047923c694b4ce0c3b31745ff01b
> > >
> > > WAN --> NAT --> LAN
> > > [ ID] Interval Transfer Bitrate Retr
> > > [ 5] 0.00-10.01 sec 616 MBytes 516 Mbits/sec 18 sender
> > > [ 5] 0.00-10.00 sec 613 MBytes 515 Mbits/sec receiver
> > >
> > > WAN <-- NAT <-- LAN
> > > [ ID] Interval Transfer Bitrate Retr
> > > [ 5] 0.00-10.00 sec 573 MBytes 480 Mbits/sec 32 sender
> > > [ 5] 0.00-10.01 sec 570 MBytes 478 Mbits/sec receiver
> > >
> > > 5.4.103 without e131a5634830047923c694b4ce0c3b31745ff01b
> > >
> > > WAN --> NAT --> LAN
> > > [ ID] Interval Transfer Bitrate Retr
> > > [ 5] 0.00-10.01 sec 454 MBytes 380 Mbits/sec 62 sender
> > > [ 5] 0.00-10.00 sec 451 MBytes 378 Mbits/sec receiver
> > >
> > > WAN <-- NAT <-- LAN
> > > [ ID] Interval Transfer Bitrate Retr
> > > [ 5] 0.00-10.00 sec 425 MBytes 356 Mbits/sec 155 sender
> > > [ 5] 0.00-10.01 sec 422 MBytes 354 Mbits/sec receiver
> > >
> > > 5.4.103 with e131a5634830047923c694b4ce0c3b31745ff01b
> > >
> > > WAN --> NAT --> LAN
> > > [ ID] Interval Transfer Bitrate Retr
> > > [ 5] 0.00-10.01 sec 604 MBytes 506 Mbits/sec 8 sender
> > > [ 5] 0.00-10.00 sec 601 MBytes 504 Mbits/sec receiver
> > >
> > > WAN <-- NAT <-- LAN
> > > [ ID] Interval Transfer Bitrate Retr
> > > [ 5] 0.00-10.00 sec 578 MBytes 485 Mbits/sec 79 sender
> > > [ 5] 0.00-10.01 sec 575 MBytes 482 Mbits/sec receiver
> > > ---
> > > net/dsa/Kconfig | 1 +
> > > net/dsa/dsa.c | 2 +-
> > > net/dsa/dsa_priv.h | 3 +++
> > > net/dsa/slave.c | 10 +++++++++-
> > > 4 files changed, 14 insertions(+), 2 deletions(-)
> >
> > So this patch should be applied to the 4.19 and 5.4 stable queues?
>
> Yes! Patch was introduced in 5.8 and applies cleanly for 4.19 and 5.4
> stable releases without any modifications. Trying to apply it for 4.14
> results in patch conflicts. So I have done tests only for 4.19 and 5.4.
Great, now queued up, thanks.
greg k-h
prev parent reply other threads:[~2021-03-09 10:39 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-08 17:57 [PATCH] net: dsa: add GRO support via gro_cells Pali Rohár
2021-03-08 19:55 ` Greg KH
2021-03-09 10:24 ` Pali Rohár
2021-03-09 10:38 ` Greg KH [this message]
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=YEdQG34U32LQW+R+@kroah.com \
--to=gregkh@linuxfoundation.org \
--cc=bloodyreaper@yandex.ru \
--cc=davem@davemloft.net \
--cc=kabel@kernel.org \
--cc=pali@kernel.org \
--cc=stable@vger.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).