From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from vms173003pub.verizon.net (vms173003pub.verizon.net [206.46.173.3]) by ozlabs.org (Postfix) with ESMTP id A7A82B6F9D for ; Mon, 13 Feb 2012 12:40:23 +1100 (EST) Received: from [192.168.1.3] ([unknown] [108.25.174.63]) by vms173003.mailsrvcs.net (Sun Java(tm) System Messaging Server 7u2-7.02 32bit (built Apr 16 2009)) with ESMTPA id <0LZB00BUW5YWAPK1@vms173003.mailsrvcs.net> for linuxppc-dev@lists.ozlabs.org; Sun, 12 Feb 2012 19:40:09 -0600 (CST) Message-id: <4F3869F8.6080802@mindchasers.com> Date: Sun, 12 Feb 2012 20:40:08 -0500 From: Bob Cochran MIME-version: 1.0 To: b29983@freescale.com, linuxppc-dev@lists.ozlabs.org, Tang Yuantian , Jin Qing Subject: Re: [PATCH 1/2 V3] powerpc/85xx: Add P1024rdb dts support References: <1328860851-392-1-git-send-email-b29983@freescale.com> In-reply-to: <1328860851-392-1-git-send-email-b29983@freescale.com> Content-type: text/plain; charset=ISO-8859-1; format=flowed List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On 02/10/2012 03:00 AM, b29983@freescale.com wrote: > From: Tang Yuantian > > Signed-off-by: Jin Qing > Signed-off-by: Li Yang > Signed-off-by: Tang Yuantian > --- > v3: I'm curious how fsl_pq_mdio_probe returns successfully when probing the phys on the first pass (mdio@24000). I don't have a P1024 to test with, but I believe it has the same ETSEC configuration as the P1010 that I work with. Inside the fsl_pq_mdio_probe routine (fsl_pq_mdio.c), a successful search of a tbi child node is required with a tbi->type of "tby-phy". If it doesn't find this node, the probe routine will return an error of EBUSY. The logic from the routine is provided below: for_each_child_of_node(np, tbi) { if (!strncmp(tbi->type, "tbi-phy", 8)) break; } if (tbi) { const u32 *prop = of_get_property(tbi, "reg", NULL); if (prop) tbiaddr = *prop; } if (tbiaddr == -1) { err = -EBUSY; goto err_free_irqs; } else { out_be32(tbipa, tbiaddr); } If I look at the p1010rdb.dtsi file, I see that a tbi-phy child node was added inside mdio@24000; however, I don't see this inside your p1024rdb.dtsi. On top of this, the current p1010rdb mdio@24000 node in p1024rdb.dtsi doesn't yield a successful probe (as I reported on this mail list on 2/7/12). So, it seems to me that a tbi child node needs to be added to your mdio@24000 node similar to what was done with the p1010rdb, but it also requires some additional tweaking or perhaps a fix inside the fsl_pq_mdio_probe routine or of library (It's not clear to me how to best produce a successful probe given the circumstances). > + > + mdio@24000 { > + phy0: ethernet-phy@0 { > + interrupts =<3 1 0 0>; > + reg =<0x0>; > + }; > + phy1: ethernet-phy@1 { > + interrupts =<2 1 0 0>; > + reg =<0x1>; > + }; > + phy2: ethernet-phy@2 { > + interrupts =<1 1 0 0>; > + reg =<0x2>; > + }; > + }; > + > + mdio@25000 { > + tbi0: tbi-phy@11 { > + reg =<0x11>; > + device_type = "tbi-phy"; > + }; > + }; > + > + mdio@26000 { > + tbi1: tbi-phy@11 { > + reg =<0x11>; > + device_type = "tbi-phy"; > + }; > + }; > + > + ethernet@b0000 { > + phy-handle =<&phy2>; > + phy-connection-type = "rgmii-id"; > + }; > +