From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from krieglstein.org ([176.28.13.145] helo=lvps176-28-13-145.dedicated.hosteurope.de) by bombadil.infradead.org with esmtp (Exim 4.80.1 #2 (Red Hat Linux)) id 1aTsR6-0002Zi-AX for barebox@lists.infradead.org; Thu, 11 Feb 2016 14:41:49 +0000 From: Tim Sander Date: Thu, 11 Feb 2016 15:40:41 +0100 Message-ID: <3833870.VaYZ1EDOWB@dabox> In-Reply-To: References: <2936535.PfmXjTYsqy@dabox> MIME-Version: 1.0 List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: "barebox" Errors-To: barebox-bounces+u.kleine-koenig=pengutronix.de@lists.infradead.org Subject: Re: [PATCH v4] Terasic DE0-Nano-SoC: add support To: Andrey Smirnov Cc: "barebox@lists.infradead.org" , Steffen Trumtrar , Trent Piepho Hi Am Montag, 1. Februar 2016, 11:08:10 schrieb Andrey Smirnov: ... > > diff --git a/arch/arm/boards/terasic-de0-nano-soc/board.c > > b/arch/arm/boards/terasic-de0-nano-soc/board.c new file mode 100644 > > index 0000000..22f8291 > > --- /dev/null > > +++ b/arch/arm/boards/terasic-de0-nano-soc/board.c > > @@ -0,0 +1,37 @@ > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > +#include > > + > > +static int phy_fixup(struct phy_device *dev) > > +{ > > + /* min rx data delay */ > > + phy_write(dev, 0x0b, 0x8105); > > + phy_write(dev, 0x0c, 0x0000); > > + > > + /* max rx/tx clock delay, min rx/tx control delay */ > > + phy_write(dev, 0x0b, 0x8104); > > + phy_write(dev, 0x0c, 0xa0d0); > > + phy_write(dev, 0x0b, 0x104); > > + > > + return 0; > > +} > > + > > +static int socfpga_console_init(void) > > I know that this names comes from the source code for SoCKit, but I'd > argue that the name of this function should be changed to something > more descriptive. This function doesn't really initialize console in > any way and just happen to be executed on "console_initcall" level of > execution. > > > +{ > > + if (!of_machine_is_compatible("altr,socfpga-cyclone5")) The string above will be replaced with "terasic,de0-nano-soc" but see below... > > + return 0; > > Since Altera is not really a vendor of this board and, I'd suggest the > compatibility string be renamed to "terrasic,de0-nano-soc" Besides that the company is named Terasic i concur with you but there is one problem: I have now just patched the kernel with the folowing patch to add the device tree compatiblility information needed: diff --git a/arch/arm/boot/dts/socfpga_cyclone5_de0_sockit.dts b/arch/arm/boot/dts/socfpga_cyclone5_de0_sockit.dts index 555e9caf21e1..3a427423168e 100644 --- a/arch/arm/boot/dts/socfpga_cyclone5_de0_sockit.dts +++ b/arch/arm/boot/dts/socfpga_cyclone5_de0_sockit.dts @@ -18,7 +18,7 @@ / { model = "Terasic DE-0(Atlas)"; - compatible = "altr,socfpga-cyclone5", "altr,socfpga"; + compatible = "terasic,de0-nano-soc"," altr,socfpga-cyclone5", "altr,socfpga"; chosen { bootargs = "earlyprintk"; -- 1.9.1 But now i have the barebox bootmessage which states the following on boot: -- blspec: blspec_scan_directory: mmc loader/entries blspec: blspec_scan_directory: /mnt/mmc0.2 loader/entries blspec: ignoring entry with incompatible devicetree "terasic,de0-nano-soc" -- Looking into loader/entries on the SD-card shows the follwing contents of socfpga_cyclone5_de0_sockit.conf: -- title PTXdist - HBM-Altera Soc Test socfpga_cyclone5_de0_sockit version 4.4.1 options root=/dev/mmcblk0p3 rootwait rootfstype=ext4 rw linux /boot/zImage devicetree /boot/socfpga_cyclone5_de0_sockit.dtb -- dtc -I dtb -O dts socfpga_cyclone5_de0_sockit.dtb |head outputs -- /dts-v1/; /memreserve/ 0x0000000000000000 0x0000000000001000; / { #address-cells = <0x1>; #size-cells = <0x1>; model = "Terasic DE-0(Atlas)"; compatible = "terasic,de0-nano-soc", " altr,socfpga-cyclone5", "altr,socfpga"; -- All files are taken directly from the SD-card so i am pretty sure i made no errors there. But currently i fail to see why this error occurs? I seems to me that all the device tree magic is in place but unfortunatly barebox does not seem to know... Best Regards Tim _______________________________________________ barebox mailing list barebox@lists.infradead.org http://lists.infradead.org/mailman/listinfo/barebox