[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAOMZO5By8=j7gZRAdf2-h-5ak-Qo34rqwDL4+HA_97cMXTAnEQ@mail.gmail.com>
Date: Wed, 20 Mar 2019 14:04:42 -0300
From: Fabio Estevam <festevam@...il.com>
To: Steve Twiss <stwiss.opensource@...semi.com>,
Andrew Lunn <andrew@...n.ch>
Cc: Rob Herring <robh+dt@...nel.org>,
Mark Rutland <mark.rutland@....com>,
Shawn Guo <shawnguo@...nel.org>,
Sascha Hauer <s.hauer@...gutronix.de>,
Pengutronix Kernel Team <kernel@...gutronix.de>,
NXP Linux Team <linux-imx@....com>,
DEVICETREE <devicetree@...r.kernel.org>,
LINUX-ARM-KERNEL <linux-arm-kernel@...ts.infradead.org>,
LKML <linux-kernel@...r.kernel.org>,
Support Opensource <Support.Opensource@...semi.com>
Subject: Re: [PATCH] ARM: dts: imx6qdl-sabresd: change phy-mode to use rgmii-id
Hi Steve,
[Adding Andrew]
On Wed, Mar 20, 2019 at 1:03 PM Steve Twiss
<stwiss.opensource@...semi.com> wrote:
>
> Hi Fabio,
>
> On 20 March 2019 12:17, Fabio Estevam wrote:
> > Subject: Re: [PATCH] ARM: dts: imx6qdl-sabresd: change phy-mode to use rgmii-id
> >
> > Hi Steve,
> >
> > On Wed, Mar 20, 2019 at 9:06 AM Steve Twiss
> > <stwiss.opensource@...semi.com> wrote:
> > >
> > > The PHY used on the Freescale i.MX6Q/DL SABRE boards is qualified as
> > > 'rgmii' instead of 'rgmii-id'. Meaning the RX and TX delays that were
> >
> > This patch declares it as 'rgmii-id', which contradicts the commit log.
>
> The PHY used on the Freescale i.MX6Q/DL SABRE boards is qualified as
> 'rgmii' instead of 'rgmii-id'.
>
> Can that be read two ways?
> I meant the phy-mode is currently qualified as 'rgmii' instead of what it
> should be: 'rgmii-id'.
Thanks for the clarification.
>
> [...]
>
> > > This patch fixes the network problems seen on the Freescale i.MX6Q/DL
> >
> > Please provide a Fixes tag. It would be good to know if this fix needs
> > to be applied to older kernels.
>
> I didn't put a Fixes tag in because I find this patch fixes v5.1-rc1 (which I see as broken).
> But the patch is not needed in v5.0 because I see that as working.
> So didn't see the need for a Cc: stable.
It seems we have other boards that need to be fixed and we can not
have an old dtb with functional Ethernet with a new kernel.
Does anyone know if this issue is AR8031 specific?
Thanks
Powered by blists - more mailing lists