From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from na01-by2-obe.outbound.protection.outlook.com (mail-by2on0113.outbound.protection.outlook.com [207.46.100.113]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by lists.ozlabs.org (Postfix) with ESMTPS id 998E21A0332 for ; Fri, 27 Mar 2015 09:16:03 +1100 (AEDT) Message-ID: <1427408150.22867.135.camel@freescale.com> Subject: Re: [PATCH 2/3] powerpc/85xx: Create dts of each core in CAMP mode for P1021RDB-PC From: Scott Wood To: Date: Thu, 26 Mar 2015 17:15:50 -0500 In-Reply-To: <1427361317-4733-1-git-send-email-ying.zhang@freescale.com> References: <1427361317-4733-1-git-send-email-ying.zhang@freescale.com> Content-Type: text/plain; charset="UTF-8" MIME-Version: 1.0 Cc: linuxppc-dev@lists.ozlabs.org, Ying Zhang List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Thu, 2015-03-26 at 17:15 +0800, ying.zhang@freescale.com wrote: > From: Ying Zhang > > Create the dts files for each core and splits the devices between > the two cores for P1021RDB-PC. > > Core0 has l2, serial0, i2c, spi, gpio, tdm,dma, usb, eth0, eth1, > sdhc, crypto, global-util, message, pci0, pci1, msi, crypto. > Core1 has l2, serial1, eth2. Are we going to have this split for every board (I hope not)? We already have a couple examples of CAMP; why do we need p1021 specifically? > Signed-off-by: Ying Zhang > Change-Id: I92614abdca3db3ab6083c4443ad563fd687050ec > Reviewed-on: http://git.am.freescale.net:8181/1179 > Reviewed-by: Fleming Andrew-AFLEMING > Tested-by: Fleming Andrew-AFLEMING Get rid of the gerrit stuff. > + mpic: pic@40000 { > + protected-sources = < > + 42 /* serial1 */ > + 31 32 33 /* enet2-queue-group0 */ > + 25 26 27 /* enet2-queue-group1 */ > + >; > + pic-no-reset; > + }; If you have pic-no-reset you don't strictly need protected-sources. Having it can protect against some relatively unlikely software errors, at the expense of allowing more likely device tree errors. Or at least, it makes it harder to modify the device tree to assign devices differently. -Scott