From: Russell King - ARM Linux admin <linux@armlinux.org.uk>
To: Sascha Hauer <s.hauer@pengutronix.de>
Cc: linux-arm-kernel@lists.infradead.org, netdev@vger.kernel.org,
linux-kernel@vger.kernel.org,
Thomas Petazzoni <thomas.petazzoni@bootlin.com>,
kernel@pengutronix.de
Subject: Re: [PATCH v2] net: mvneta: Fix Serdes configuration for 2.5Gbps modes
Date: Fri, 12 Jun 2020 14:02:50 +0100 [thread overview]
Message-ID: <20200612130250.GK1551@shell.armlinux.org.uk> (raw)
In-Reply-To: <20200612115250.GS11869@pengutronix.de>
On Fri, Jun 12, 2020 at 01:52:50PM +0200, Sascha Hauer wrote:
> On Fri, Jun 12, 2020 at 12:30:31PM +0100, Russell King - ARM Linux admin wrote:
> > On Fri, Jun 12, 2020 at 12:22:13PM +0100, Russell King - ARM Linux admin wrote:
> > > On Fri, Jun 12, 2020 at 11:42:08AM +0100, Russell King - ARM Linux admin wrote:
> > > > With the obvious mistakes fixed (extraneous 'i' and lack of default
> > > > case), it seems to still work on Armada 388 Clearfog Pro with 2.5G
> > > > modules.
> > >
> > > ... and the other bug fixed - mvneta_comphy_init() needs to be passed
> > > the interface mode.
> >
> > Unrelated to the patch, has anyone noticed that mvneta's performance
> > seems to have reduced? I've only just noticed it (which makes 2.5Gbps
> > rather pointless). This is iperf between two clearfogs with a 2.5G
> > fibre link:
> >
> > root@clearfog21:~# iperf -V -c fe80::250:43ff:fe02:303%eno2
> > ------------------------------------------------------------
> > Client connecting to fe80::250:43ff:fe02:303%eno2, TCP port 5001
> > TCP window size: 43.8 KByte (default)
> > ------------------------------------------------------------
> > [ 3] local fe80::250:43ff:fe21:203 port 48928 connected with fe80::250:43ff:fe02:303 port 5001
> > [ ID] Interval Transfer Bandwidth
> > [ 3] 0.0-10.0 sec 553 MBytes 464 Mbits/sec
> >
> > I checked with Jon Nettleton, and he confirms my recollection that
> > mvneta on Armada 388 used to be able to fill a 2.5Gbps link.
> >
> > If Armada 388 can't manage, then I suspect Armada XP will have worse
> > performance being an earlier revision SoC.
>
> I only have one board with a Armada XP here which has a loopback cable
> between two ports. It gives me:
>
> [ 3] local 172.16.1.4 port 47002 connected with 172.16.1.0 port 5001
> [ ID] Interval Transfer Bandwidth
> [ 3] 0.0-10.0 sec 1.27 GBytes 1.09 Gbits/sec
>
> Still not 2.5Gbps, but at least twice the data rate you get, plus my
> board has to handle both ends of the link.
It turns out I still had various locking debugs and DMA API debug
enabled:
[ 3] local fe80::250:43ff:fe21:203 port 39326 connected with fe80::250:43ff:fe02:303 port 5001
[ ID] Interval Transfer Bandwidth
[ 3] 0.0-10.0 sec 2.70 GBytes 2.32 Gbits/sec
... which is more like it for this platform!
--
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
next prev parent reply other threads:[~2020-06-12 13:03 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-06-12 8:38 [PATCH v2] net: mvneta: Fix Serdes configuration for 2.5Gbps modes Sascha Hauer
2020-06-12 8:47 ` Russell King - ARM Linux admin
2020-06-12 10:01 ` Russell King - ARM Linux admin
2020-06-12 10:18 ` Russell King - ARM Linux admin
2020-06-12 10:42 ` Russell King - ARM Linux admin
2020-06-12 11:22 ` Russell King - ARM Linux admin
2020-06-12 11:30 ` Russell King - ARM Linux admin
2020-06-12 11:52 ` Sascha Hauer
2020-06-12 13:02 ` Russell King - ARM Linux admin [this message]
2020-06-12 12:06 ` Sascha Hauer
2020-06-12 12:12 ` Russell King - ARM Linux admin
2020-06-12 10:15 ` Sascha Hauer
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=20200612130250.GK1551@shell.armlinux.org.uk \
--to=linux@armlinux.org.uk \
--cc=kernel@pengutronix.de \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=s.hauer@pengutronix.de \
--cc=thomas.petazzoni@bootlin.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).