kern/177032: [arge] arge1 fails to attach on UBNT Routerstation

[Luiz Otavio O Souza]

The following reply was made to PR kern/177032; it has been noted by GNATS.

From: Luiz Otavio O Souza <lists.br at gmail.com>
To: Petko Bordjukov <bordjukov at gmail.com>
Cc: freebsd-mips at freebsd.org, bug-followup at FreeBSD.org
Subject: Re: kern/177032: [arge] arge1 fails to attach on UBNT Routerstation
Date: Mon, 1 Apr 2013 17:32:59 -0300

 --047d7bfd0bd6db028f04d9528645
 Content-Type: text/plain; charset=ISO-8859-1
 
 On 1 April 2013 14:00, Petko Bordjukov <bordjukov at gmail.com> wrote:
 
 > The following reply was made to PR kern/177032; it has been noted by GNATS.
 >
 > From: Petko Bordjukov <bordjukov at gmail.com>
 > To: bug-followup at FreeBSD.org
 > Cc:
 > Subject: Re: kern/177032: [arge] arge1 fails to attach on UBNT
 > Routerstation
 > Date: Mon, 01 Apr 2013 19:50:09 +0300
 >
 >  Hello Luiz,
 >
 >  Thank you for your response.
 >
 >  Using a 0x0 PHY mask does indeed eliminate the error and the interface
 >  is present and operational. However, what are the implications of not
 >  using a directly mapped PHY?
 >
 >  Also, to my knowledge the phymask should be 15 << 16 and the miimode
 >  should be RMII, however I was not able to get the interface running with
 >  these values.
 >
 >  --
 >  Best regards,
 >  Petko
 >
 >
 
 There are no implications for this change on ROUTERSTATION as the switch
 (ADM9669) cannot be managed from its MDIO interface and all you have is a
 static two port switch on arge1.
 
 Even with the correct phymask, you cannot attach and use two simultaneous
 PHY on a single ethernet interface so i think that the phymask=0 is correct
 for this board.
 
 And no, the phymask on the ROUTERSTATION.hints is correct (0x3 << 16 == PHY
 16 and 17) as there are only two ports connected on the switch (all the
 others PHYs on the switch are disconnected).
 
 I'm not sure about the correct miimode for ROUTERSTATION, but i think
 you're right, it should be RMII. I don't have physical access to this board
 anymore to test it myself (although a friend of mine has one of these).
 
 Regards,
 Luiz
 
 --047d7bfd0bd6db028f04d9528645
 Content-Type: text/html; charset=ISO-8859-1
 Content-Transfer-Encoding: quoted-printable
 
 <div dir=3D"ltr"><div class=3D"gmail_extra">On 1 April 2013 14:00, Petko Bo=
 rdjukov <span dir=3D"ltr"><<a href=3D"mailto:bordjukov at gmail.com" target=
 =3D"_blank">bordjukov at gmail.com</a>></span> wrote:<br><div class=3D"gmai=
 l_quote">
 <blockquote class=3D"gmail_quote" style=3D"margin-top:0px;margin-right:0px;=
 margin-bottom:0px;margin-left:0.8ex;border-left-width:1px;border-left-color=
 :rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div class=3D"i=
 m">
 The following reply was made to PR kern/177032; it has been noted by GNATS.=
 <br>
 <br>
 </div>From: Petko Bordjukov <<a href=3D"mailto:bordjukov at gmail.com">bord=
 jukov at gmail.com</a>><br>
 To: bug-followup at FreeBSD.org<br>
 <div class=3D"im">Cc:<br>
 Subject: Re: kern/177032: [arge] arge1 fails to attach on UBNT Routerstatio=
 n<br>
 </div>Date: Mon, 01 Apr 2013 19:50:09 +0300<br>
 <br>
 =A0Hello Luiz,<br>
 <br>
 =A0Thank you for your response.<br>
 <br>
 =A0Using a 0x0 PHY mask does indeed eliminate the error and the interface<b=
 r>
 =A0is present and operational. However, what are the implications of not<br=
 >
 =A0using a directly mapped PHY?<br>
 <br>
 =A0Also, to my knowledge the phymask should be 15 << 16 and the miimo=
 de<br>
 =A0should be RMII, however I was not able to get the interface running with=
 <br>
 =A0these values.<br>
 <br>
 =A0--<br>
 =A0Best regards,<br>
 =A0Petko<br>
 <div class=3D""><div class=3D"h5"><br></div></div></blockquote><div><br></d=
 iv><div><br></div><div>There are no implications for this change on ROUTERS=
 TATION as the switch (ADM9669) cannot be managed from its MDIO interface an=
 d all you have is a static two port switch on arge1.</div>
 <div><br></div><div>Even with the correct phymask, you cannot attach and us=
 e two simultaneous PHY on a single ethernet interface so i think that the p=
 hymask=3D0 is correct for this board.</div><div><br></div><div>And no, the =
 phymask on the ROUTERSTATION.hints is correct (0x3 << 16 =3D=3D PHY 1=
 6 and 17) as there are only two ports connected on the switch (all the othe=
 rs PHYs on the switch are disconnected).</div>
 <div><br></div><div>I'm not sure about the correct miimode for ROUTERST=
 ATION, but i think you're right, it should be RMII. I don't have ph=
 ysical access to this board anymore to test it myself (although a friend of=
  mine has one of these).</div>
 <div><br></div><div>Regards,</div><div>Luiz</div><div><br></div></div></div=
 ></div>
 
 --047d7bfd0bd6db028f04d9528645--