From: Wolfgang Denk Date: Thu, 18 Feb 2010 13:32:32 +0100 Subject: [U-Boot] [PATCH] AT91: Added support for taskit Stamp9G20 and PortuxG20 In-Reply-To: <4B7C14EC.7050209@taskit.de> References: <4B7C14EC.7050209@taskit.de> Message-ID: <20100218123232.DEC05A87D17@gemini.denx.de> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de Dear Achim Ehrlich, In message <4B7C14EC.7050209@taskit.de> you wrote: > Signed-off-by: Achim Ehrlich > --- ... > --- a/Makefile > +++ b/Makefile > @@ -2906,6 +2906,15 @@ TNY_A9260_config : unconfig > @echo "#define CONFIG_$(@:_config=) 1" >$(obj)include/config.h > @$(MKCONFIG) -a tny_a9260 arm arm926ejs tny_a9260 calao at91 > > +portuxG20_config \ > +stamp9G20_config : unconfig > + @mkdir -p $(obj)include > + @if [ "$(findstring portux,$@)" ] ; then \ > + echo "#define CONFIG_PORTUXG20 1" >$(obj)include/config.h; \ > + $(XECHO) "... PortuxG20";\ > + fi; NAK. We don't accept such scripting in the Makefile any more. Please use the available mkconfig features instead. > diff --git a/board/taskit/stamp9G20/partition.c b/board/taskit/stamp9G20/partition.c > new file mode 100644 > index 0000000..2629c67 > --- /dev/null > +++ b/board/taskit/stamp9G20/partition.c So we are adding yet another copy of this file. We already have 12 of these. Mostly identical. Isn't there a more intelligent way to handle this? ... > +int dram_init(void) > +{ > + gd->bd->bi_dram[0].start = PHYS_SDRAM; > + gd->bd->bi_dram[0].size = PHYS_SDRAM_SIZE; > + return 0; > +} Please consider using get_ram_size() for memory auto-sizing and testing. > +#ifdef CONFIG_RESET_PHY_R > +void reset_phy(void) > +{ > +#ifdef CONFIG_MACB CONFIG_MACB seems to be an undocumented variable. What is it supposed to do? > diff --git a/include/configs/stamp9G20.h b/include/configs/stamp9G20.h > new file mode 100644 > index 0000000..5008554 > --- /dev/null > +++ b/include/configs/stamp9G20.h ... > +/* ARM asynchronous clock */ > +#define AT91_MAIN_CLOCK 18432000 /* 18.432 MHz crystal */ This should be changed into a CONFIG_SYS_ (globally). > +#define AT91_CPU_NAME "AT91SAM9G20" Ditto. > +#define CONFIG_ATMEL_USART 1 > +#undef CONFIG_USART0 > +#undef CONFIG_USART1 > +#undef CONFIG_USART2 > +#define CONFIG_USART3 1 /* USART 3 is DBGU */ Do not undefine what is not defined anyway. > +/* LED */ > + > +#undef CONFIG_AT91_LED Ditto. > +/* > + * Command line configuration. > + */ > +#include > +#undef CONFIG_CMD_BDI > +#undef CONFIG_CMD_FPGA > +#undef CONFIG_CMD_IMI > +#undef CONFIG_CMD_IMLS > +#undef CONFIG_CMD_SOURCE Is there any specific reason for disabling the "bdi", "imi", "imls" and "source" commands? They are pretty useful, and you don't seem to be especially short on resources... ... > +/* USB */ > +#define CONFIG_USB_ATMEL > +#define CONFIG_USB_OHCI_NEW 1 > +#define CONFIG_DOS_PARTITION 1 > +#define CONFIG_SYS_USB_OHCI_CPU_INIT 1 > +#define CONFIG_SYS_USB_OHCI_REGS_BASE 0x00500000 /* AT91SAM9260_UHP_BASE */ Line too long. Please fix globally. ... > + "basicargs=console=ttyS0,115200 mem=64M\0" \ > + "" > + > + > + Excessive white space. Drop 3 of the empty lines. Best regards, Wolfgang Denk -- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de "Do we define evil as the absence of goodness? It seems only logical that shit happens--we discover this by the process of elimination." -- Larry Wall