July 2006 - U-Boot

[U-Boot-Users] [PATCH] logic error waiting for init in virtex2/spartan2/spartan3
by Andrew Dyer 18 Jul '06

18 Jul '06

In the xilinx download code there is a logic error in the section where it is waiting for the init signal to go inactive. There is a while loop which loop if init is active /or/ busy is active, instead it checks if init is active /and/ busy active. In most designs busy is never used (only needed for clocks > 50 MHz) and often not connected. If the unused busy function returns 0 (not busy) the while loop will exit early and configuration will start too soon. This behaviour is masked if the debugging code is turned on by the printf run time. CHANGELOG fix init wait loop exit conditions for xilinx downloads to properly handle case where busy is low Patch by Andrew Dyer <amdyer(a)gmail.com>, 18 July 2006 Signed-off-by: Andrew Dyer <amdyer(a)gmail.com> -- Hardware, n.: The parts of a computer system that can be kicked.

1 0

[U-Boot-Users] Clarification on /linuxrc Script.
by Raja Chidambaram 18 Jul '06

18 Jul '06

Hi all, I have a 512mb compact flash formatted with FAT32 filesytem.Requirement is to boot linux from comapct flash through System ace controller.I had just follwed the instrutions given in http://www.denx.de/wiki/view/DULG/RootFileSystemInAReadOnlyFile I was able to boot the linux kernel after follwing th instructions, but kernel hangs with kernel panic such as VFS: Mounted root (ext2 filesystem). Freeing unused kernel memory: 104k init Failed to execute /linuxrc. Attempting defaults... Kernel panic - not syncing: No init found. Try passing init= option to kernel. i had just set bootargs as >setenv bootargs 'console=ttyS0, 115200 root=/dev/ram rw init=/linuxrc rw' & >fatload ace 0 ${ram_addr} linux.img;bootm ${ram_addr} Is that anything more i need other than the instructions given.please help me in this. with regards raja __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 1

[U-Boot-Users] clarifications u-boot for arm
by Ram 18 Jul '06

18 Jul '06

Hi, I am using a omap5912osk board which uses arm926-ejs CPU. I am trying to understand the flow of u-boot - what really happens when the board is powered on till i get the u-boot prompt which accepts commands from the user. I started with the file u-boot-1.1.4/cpu/arm926ejs/start.S Is this the place where u-boot actually starts?. It tries switch the cpu mode to SVC32 mode and relocates u-boot to RAM, disables MMU and does so many things which i dont really understand. Could anyone explain it to me what really is happening out there and why?. Then i figured out that a function start_armboot() is called. This function is present in lib_arm/board.c is this function called by start.S? Where exactly it is called from?. There is a reference to start_armboot in start.S not sure if it is the one?. ldr pc, _start_armboot _start_armboot: .word start_armboot Does this actually invoke start_armboot?. I have got so far. could you please tell me am i right in my understarting?. I understand that start_armboot initialises the the cpu, board, dram.... Once this is done - we can say that the board has been initialised. Now start_armboot provides a prompt to accept and download the kernel. Am i right in assuming this?. Any other information and pointers u provide will be helpful I am trying to understand u-boot sources and any help will be highly appreciated. I just wanted to clarify my understanding. Thanks and Regards, sriram

5 5

[U-Boot-Users] Regarding ARM7TDMI Boot Loader
by kushal ahuja 18 Jul '06

18 Jul '06

Respected Sir, We are building Boot Loader for ARM7TDMI,for that we have successfully compile AP7 (INTEGRATOR ARM CORE MODULE FOR ARM7TDMI) while loading that image on our board we are getting some error as : *ERROR: DATA ABORT EXCEPTION:01009ABC* *ERROR:DATA ABORT EXCEPTION: 01009ABC* As our Board is specific to ARM7TDMI only, kindly help us in finding solution. Thank You in Advance -- Best Regards, Kushal Ahuja

1 0

[U-Boot-Users] dual boot by modifying u-boot environment variables from linux/vxWorks
by Robert Steele 17 Jul '06

17 Jul '06

I want to be able to 'dual' boot my system, a Gespac PCIPPC-2 (power pc), by the using the method of modifying the u-boot environment variables from either vxWorks or Linux. This way if I boot up in the 'wrong' system I can change the boot parameters and reboot. It looks like the fw* utilities require /dev/mds1 and /dev/mds1 to reference the flash memory and this does not appear to be set up in my kernel. Does this require a re-build of my kernel or are there other less dramatic options available. If anyone can point me to any solutions I would very much appreciate the references. If someone knows of a better solutions for rebooting that would be nice to hear about. One of my requirements is that the final system will have a monitor or keyboard on it for this rebooting process. Thanks, Rob Steele

3 6

[U-Boot-Users] marvell db64460
by Engin KARABULUT 17 Jul '06

17 Jul '06

hi, we have purchased marvell DB64460 development board by means of NDA agreement. the board comes with patch for uboot 1.0.0, I successfully compiled it with gcc 2.95 recommended by marvell and board boots well. But my aim is to debug uboot,because I wanna learn uboot programming and debugging for our custom board that hasn't any bootloader. our debugger is bdi2000 and I prepare configuration file via looking configuration files of other boards.now I can successfully write flash with jtag. But I have some problems with gdb, gdb that I'm using is version 5.2.1. When I set breakpoints somewhere in code, code doesn't stop in breakpoints. sometimes it stop but gives "Program received signal SIGSTOP" error. sometimes I'm able to execute code step by step,sometimes not. I'm confused too much. In your opinion Is there any combination between gdb and gcc version? For gcc 2.95, what is right gdb version? thanks a lot.

2 4

[U-Boot-Users] If I-cache enabled, the system gets unstable
by Andrew Zhukov 17 Jul '06

17 Jul '06

I use 2.6.10 with at91rm9200dk, root via NFS - to use ARM native compiler. If I enable I-cache, the system can't compile anything non-trivial. Does anybody know - if it's error in old kernel and I should use 2.6.16, or it's hardware (I have the same kind of errors - segmentation fault - on all identical boards)? I attached the kernel's config, kernel boot messages in start.log, and errors while native-compile some kernel via NFS. With I-cache disables any native-compilation works well. (I could do without native compilre, but with I-cache enabled I can't run any non-trivial program)

2 1

[U-Boot-Users] Regarding MPC7448
by Balakumar 17 Jul '06

17 Jul '06

Hi Group, Is there any patch available for MPC7448 Boot code. Regards Bala

2 1

[U-Boot-Users] Building U-Boot without cmd_ide.c?
by Timur Tabi 17 Jul '06

17 Jul '06

I'm working on providing support for the 8349e-mITX, which is another 8349 variant. I'm taking some old patches for U-Boot 1.1.3 and porting them to the latest code. Unfortunately, I don't know a whole lot about the 8349 or U-Boot, so this is a difficult task. What I have so far won't boot at all. I power on the system and nothing happens. Obviously, I have no idea what's wrong. So right now, I'm trying to compile U-Boot with as many options turned off as possible. The problem is that the build process tries to compile some files even if support for them has been turned off. For instance, if I undefine CONFIG_COMPACT_FLASH, then I get these errors when I build: powerpc-unknown-linux-gnu-gcc -g -Os -fPIC -ffixed-r14 -meabi -D__KERNEL__ -DTEXT_BASE=0xFEF00000 -I/temp/u-boot/include -fno-builtin -ffreestanding -nostdinc -isystem /_TOOLS_/dist/gnu-gcc-3.4.3-binutils-2.15-powerpc-unknown-linux-gnu/i686-pc-linux2.4/lib/gcc/powerpc-unknown-linux-gnu/3.4.3/include -pipe -DCONFIG_PPC -D__powerpc__ -DCONFIG_MPC83XX -DCONFIG_E300 -ffixed-r2 -ffixed-r29 -msoft-float -Wall -Wstrict-prototypes -c -o cmd_ide.o cmd_ide.c cmd_ide.c:128: error: `CFG_IDE_MAXBUS' undeclared here (not in a function) cmd_ide.c:135: error: empty scalar initializer cmd_ide.c:135: error: (near initialization for `ide_bus_offset') cmd_ide.c:146: error: `CFG_IDE_MAXDEVICE' undeclared here (not in a function) cmd_ide.c: In function `ide_init': cmd_ide.c:618: error: `CFG_ATA_REG_OFFSET' undeclared (first use in this function) cmd_ide.c:618: error: (Each undeclared identifier is reported only once cmd_ide.c:618: error: for each function it appears in.) cmd_ide.c: In function `ide_outb': cmd_ide.c:810: error: `CFG_ATA_BASE_ADDR' undeclared (first use in this function) cmd_ide.c: In function `ide_inb': cmd_ide.c:828: error: `CFG_ATA_BASE_ADDR' undeclared (first use in this function) cmd_ide.c: In function `input_swap_data': cmd_ide.c:885: error: `CFG_ATA_BASE_ADDR' undeclared (first use in this function) cmd_ide.c:885: error: `CFG_ATA_DATA_OFFSET' undeclared (first use in this function) cmd_ide.c: In function `output_data': cmd_ide.c:930: error: `CFG_ATA_BASE_ADDR' undeclared (first use in this function) cmd_ide.c:930: error: `CFG_ATA_DATA_OFFSET' undeclared (first use in this function) cmd_ide.c: In function `input_data': cmd_ide.c:978: error: `CFG_ATA_BASE_ADDR' undeclared (first use in this function) cmd_ide.c:978: error: `CFG_ATA_DATA_OFFSET' undeclared (first use in this function) cmd_ide.c: In function `ide_ident': cmd_ide.c:1062: error: `CFG_ATA_REG_OFFSET' undeclared (first use in this function) cmd_ide.c: In function `ide_read': cmd_ide.c:1270: error: `CFG_ATA_REG_OFFSET' undeclared (first use in this function) cmd_ide.c: In function `ide_write': cmd_ide.c:1398: error: `CFG_ATA_REG_OFFSET' undeclared (first use in this function) cmd_ide.c: In function `ide_wait': cmd_ide.c:1505: error: `CFG_ATA_REG_OFFSET' undeclared (first use in this function) cmd_ide.c: At top level: cmd_ide.c:128: error: storage size of `ide_bus_offset' isn't known cmd_ide.c:141: error: storage size of `ide_bus_ok' isn't known cmd_ide.c:146: error: storage size of `ide_dev_desc' isn't known cmd_ide.c:141: warning: 'ide_bus_ok' defined but not used make[1]: *** [cmd_ide.o] Error 1 make[1]: Leaving directory `/temp/u-boot/common' make: *** [common/libcommon.a] Error 2 The reason these CFG_ATA_xxxx options are undefined is because they only get defined if CONFIG_COMPACT_FLASH is defined. Is there a reason why cmd_ide.c is being compiled? Do I really need it? Should I modified the Makefile to not build cmd_ide.c if CONFIG_COMPACT_FLASH is undefined? -- Timur Tabi Linux Kernel Developer @ Freescale

3 3

[U-Boot-Users] Help_Regaring ARM7TDMI
by ashis dash 17 Jul '06

17 Jul '06

Hi Friends, I am making a boot loader for a ARM7TDMI-S target board.I have taken the U-boot 1.1.4 as the reference boot loader and compiling it for ARM7TDMI. I have compiled the u-boot for Arm720t , which uses arm7tdmi core. Then i have written the image in the flash by using MultiIce Arm debugger. But the board is not booting up and i am getting a message like this. " Erasing the flash, Please wait ! Programming the flash, Please wait ! Loader successfully burned into the flash ! New loader booting Now ...! If cant bootup,please remove power and turn on again" After switching off and making in on , the loader is not booting up. please help me . Thanks in advance, Ashis --------------------------------- The all-new Yahoo! Mail goes wherever you go - free your email address from your Internet provider.

2 1