From owner-freebsd-arm@freebsd.org Tue Jan 16 17:13:13 2018 Return-Path: Delivered-To: freebsd-arm@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id C3CDBE7BA79 for ; Tue, 16 Jan 2018 17:13:13 +0000 (UTC) (envelope-from freebsd-arm@dino.sk) Received: from mailhost.netlabit.sk (mailhost.netlabit.sk [84.245.65.72]) (using TLSv1 with cipher DHE-RSA-CAMELLIA256-SHA (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 5A0C683678; Tue, 16 Jan 2018 17:13:12 +0000 (UTC) (envelope-from freebsd-arm@dino.sk) Received: from zeta.dino.sk (fw3.dino.sk [84.245.95.254]) (AUTH: LOGIN milan) by mailhost.netlabit.sk with ESMTPA; Tue, 16 Jan 2018 18:08:00 +0100 id 00DD603A.5A5E3170.000122BA Date: Tue, 16 Jan 2018 18:08:00 +0100 From: Milan Obuch To: Kyle Evans Cc: unto.foru13@gmail.com, freebsd-arm@freebsd.org Subject: Re: Orange pi one ,receive panic: bad pte2 Message-ID: <20180116180800.07a8f8ef@zeta.dino.sk> In-Reply-To: References: <8DD425C1-FEF4-4BC0-9A80-916B50620896@dsl-only.net> X-Mailer: Claws Mail 3.16.0 (GTK+ 2.24.31; i386-portbld-freebsd10.4) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-BeenThere: freebsd-arm@freebsd.org X-Mailman-Version: 2.1.25 Precedence: list List-Id: "Porting FreeBSD to ARM processors." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 16 Jan 2018 17:13:13 -0000 On Tue, 16 Jan 2018 08:21:47 -0600 Kyle Evans wrote: > On Tue, Jan 16, 2018 at 7:19 AM, =E9=98=BF=E9=87=91 wrote: > > OKOK below is booting message. =20 >=20 > Thanks! >=20 > > [ .. snip .. ] > > syscon_generic0: mem 0x1c00000-0x1c00033 on simplebus0 > > [.. snip ..] > > awg0: mem > > 0x1c30000-0x1c30103,0x1c00030-0x1c00033 irq 23 on simplebus0 > > awg0: cannot get tx clock > > [.. snip .. ] =20 >=20 > That explains it- I hadn't encountered this in practice, so it broke. > =3D( Your DTS here has a /soc/syscon *and* uses our former hacky method > for passing syscon, so syscon_generic takes the syscon registers and > awg can't. You have I think three choices for recourse here, any of > which should fix your problem: >=20 > 1.) Remove the /soc/syscon from the DTB you're using > 2.) Update your DTB to actually use /soc/syscon, removing the syscon > reg from /soc/ethernet and adding a 'syscon =3D <&syscon>' attribute to > that node. > 3.) Update to r327825 or later and pull new DTS from Linux. If pulling > new from Linux is a pain, I can generate a new DTB for you based on > what we'll be importing from Linux in a couple weeks. >=20 > r327825 added support for the new stable emac bindings, so after DTS > update this won't break again. > In my case, using Orange Pi Zero, I observed similar behavior. Using older DTB ethernet works, and with tips mentioned above I was quickly able to modify my DTS, so ethernet works again. It just uses different MAC. Question is, where does it actually come from, and, maybe, where should it come from. This is from dmesg: awg0: mem 0x1c30000-0x1c30103 irq 38 on simple= bus0 simplebus0: no default resources for rid =3D 1, type =3D 3 awg0: PHY type: mii, conf mode: reg awg0: EMAC clock: 0x00168000 awg0: AHB frequency 300000000 Hz, MDC div: 0x3 Any idea what resource is not found here? IS it something to worry about? Regards, Milan