November 2004 - U-Boot

RE: [U-Boot-Users] CONFIG_BOOTBINFUNC for AT91RM9200
by Friedrich, Lars 24 Nov '04

24 Nov '04

> > Why is that so bad? It makes it possible to debug U-Boot > > with just loading the image into RAM using BDI2000. > Which is an unsupported mode of operation which works for a handfull > of experts and causes confusion with many, many newbees. How do these lines of code confuse a newbie more than any other code in the file? > > But how would you debug U-Boot? > As I always do. Attach the BDI, burn to flash, start in GDB. There are 53982 other hardware debuggers out there and only the minority (is there actually one besides the BDI?) support the burn to flash feature you rely on. So if you need to start U-Boot to flash U-Boot, you either do those few lines of code to skip the relocation or get/use another piece of actually redundant software to get the image on the flash. I don't know why the latter should save me time. > You can do this if you know exactly what you're doing, Isn't this what is assumed here anyway? Best regards, Lars Friedrich --

3 2

[U-Boot-Users] boot problem
by Alessio Raccis 24 Nov '04

24 Nov '04

Hi all, I have a problem and now, it's weird to say, I pray to a kernel panic ! I'm working on smdk2410 and I have a u-boot version modified to boot from smartmedia. I want, or better I hope to boot 2.6.7 linux kernel. So, I create a zImage by compiling kernel and then I make: gzip -9 zImage ./mkimage -A arm -O linux -T kernel -C gzip -a 0x30008000 -e 0x30008000 -n "Linux Kernel Image" -d zImage.gz uimage At this point, I download uimage at 33000000 address from a tftp server. SMDK2410 # tftpboot 33000000 uimage TFTP from server 10.124.7.161; our IP address is 10.124.7.50 Filename 'uimage'. Load address: 0x33000000 Loading: ################################################################# ################################################################# ######################################################## done Bytes transferred = 950519 (e80f7 hex) Now I boot the kernel and I have: SMDK2410 # bootm ## Booting image at 33000000 ... Image Name: Linux Kernel Image Created: 2004-11-17 13:50:56 UTC Image Type: ARM Linux Kernel Image (gzip compressed) Data Size: 950455 Bytes = 928.2 kB Load Address: 30008000 Entry Point: 30008000 Verifying Checksum ... OK Uncompressing Kernel Image ... OK Starting kernel ... Uncompressing Linux................................................................... done, booting the kernel. Nothing else !! Can anyone help me, please ? Which is my mistake ? thanks in advance. Best regards, Alex

1 0

[U-Boot-Users] copy from RAM to Flash
by Cliff Brake 23 Nov '04

23 Nov '04

Can someone tell me how to copy an image from RAM to flash? I am trying to load images from a MMC card to flash. I did figure out the fatload command, but I am stuck after that. Target hardware is Accelent/Vibren PXA255 IDP -- yes I'll submit a patch when everything is working. Is anyone else working on this platform? Thanks, Cliff -- ======================= Cliff Brake http://bec-systems.com

2 1

[U-Boot-Users] General Flash Programming Error
by CG 23 Nov '04

23 Nov '04

Hello, I got this error when I execute "cp.b 20000000 10000000 X". X = # of bytes to copy ******** results : > Copy to Flash... General Flash Programming Error ******** It's a custom board based on the at91rm9200-dk. I can execute flinfo, erase sectors, display memory "md 10000000". Any hints to see if my flash works well for programming with u-boot... The flash chip is AT49BV162A Regards Christian

2 1

Re: Re: [U-Boot-Users] start.S code snippet doubt
by Nishant Galange 23 Nov '04

23 Nov '04

Thank you Yuli. Thank you very much :-) Regards, Nishant -----Original Message----- From: Yuli Barcohen <yuli(a)arabellasw.com> To: "Nishant Galange" <nishant.subs(a)myrealbox.com> Date: Tue, 23 Nov 2004 11:58:55 +0200 Subject: Re: [U-Boot-Users] start.S code snippet doubt This code fixes incorrect BADDR configuration caused by wrong HRCW supplied by BCSR (FPGA) on some MPC8260ADS boards. It's configured as if there was L2 cache so that lower address lines are not functional (64-bit data bus). Without this code, such a board can't start from 32-bit flash SIMM. This code is for MPC8260ADS only so there should not be any problem. -- ======================================================================== Yuli Barcohen | Phone +972-9-765-1788 | Software Project Leader yuli(a)arabellasw.com | Fax +972-9-765-7494 | Arabella Software, Israel ========================================================================

1 0

[U-Boot-Users] PATCH: Introducing CONFIG_BOARD_PRE_RESET
by Steven Scholz 23 Nov '04

23 Nov '04

Hi there, as mentioned on list before some boards/cpus might need some extra code to be executed _before_ rebooting. E.g. the i.MX hangs when doing "reset" while the LCD controller is still enabled. That is why I'd like to introduce CONFIG_BOARD_PRE_RESET and board_pre_reset(). The attached does exactly that. But only for cpu/arm920t. If wished I'll add the appropriate code to all cpu/$(CPU)/cpu.c files. * Patch by Steven Scholz, 25 Oct 2004: - Introducing CONFIG_BOARD_PRE_RESET and board_pre_reset() to allow code to be executed before actually rebooting -- Steven Scholz Index: board/mx1fs2/mx1fs2.c =================================================================== RCS file: /cvsroot/u-boot/u-boot/board/mx1fs2/mx1fs2.c,v retrieving revision 1.2 diff -p -u -r1.2 mx1fs2.c --- board/mx1fs2/mx1fs2.c 10 Oct 2004 18:41:10 -0000 1.2 +++ board/mx1fs2/mx1fs2.c 25 Oct 2004 13:32:14 -0000 @@ -123,3 +123,14 @@ show_boot_progress(int status) #endif return; } + +#ifdef CONFIG_BOARD_PRE_RESET +void board_pre_reset(void) +{ + /* we must disable the LCD controller before reboot + * otherwise the processor might hang */ + DR(3) &= ~0x00001000; + LCDC_RMCR &= ~RMCR_LCDC_EN; +} +#endif + Index: cpu/arm920t/cpu.c =================================================================== RCS file: /cvsroot/u-boot/u-boot/cpu/arm920t/cpu.c,v retrieving revision 1.3 diff -p -u -r1.3 cpu.c --- cpu/arm920t/cpu.c 8 Feb 2004 19:38:43 -0000 1.3 +++ cpu/arm920t/cpu.c 25 Oct 2004 13:32:14 -0000 @@ -127,6 +127,9 @@ int do_reset (cmd_tbl_t *cmdtp, int flag { extern void reset_cpu (ulong addr); disable_interrupts (); +#ifdef CONFIG_BOARD_PRE_RESET + board_pre_reset(); +#endif reset_cpu (0); /*NOTREACHED*/ Index: include/common.h =================================================================== RCS file: /cvsroot/u-boot/u-boot/include/common.h,v retrieving revision 1.46 diff -p -u -r1.46 common.h --- include/common.h 10 Oct 2004 18:41:14 -0000 1.46 +++ include/common.h 25 Oct 2004 13:32:15 -0000 @@ -318,6 +318,7 @@ int board_late_init (void); int board_postclk_init (void); /* after clocks/timebase, before env/serial */ int board_early_init_r (void); void board_poweroff (void); +void board_pre_reset (void); #if defined(CFG_DRAM_TEST) int testdram(void); Index: include/configs/mx1fs2.h =================================================================== RCS file: /cvsroot/u-boot/u-boot/include/configs/mx1fs2.h,v retrieving revision 1.1 diff -p -u -r1.1 mx1fs2.h --- include/configs/mx1fs2.h 1 Aug 2004 22:48:22 -0000 1.1 +++ include/configs/mx1fs2.h 25 Oct 2004 13:32:15 -0000 @@ -114,6 +114,7 @@ #define AUTOREFRESH_CMD 0xa10a8300 #define CONFIG_INIT_CRITICAL +#define CONFIG_BOARD_PRE_RESET 1 #define BUS32BIT_VERSION /*

3 12

[U-Boot-Users] Error in saving the Env varibles on the flash
by pankaj goyal 23 Nov '04

23 Nov '04

Hi all, I m using the U-boot for the arm 926ej processor.For that I want to save enviorment variable on the flash and I m using the flash of 16MB with the base address of the flash is 0x30000000 and the flash is bottom boot type with the sector size is 128KB and the parameter sector size 32KB.When I use the second parameter sector to store the envoirnment variable in the flash I m getting the error "Error: start and/or end address not on sector boundary".When I explore the code I found out the in the function flash_fill_sector_range in the file 'u-boot/comman/cmd_flash.c' the range is checked for the each sector and the parameter sector is not being considered to determine the number of sector to unprotect them .Can some one please tell me it this an error in the code or i m specifying the parameter sector in the wrong way. bye

2 1

[U-Boot-Users] start.S code snippet doubt
by Nishant Galange 23 Nov '04

23 Nov '04

Hi all, What does the following code achieve: (Code snippet from start.S) boot_cold: #if defined(CONFIG_MPC8260ADS) lis r3, CFG_DEFAULT_IMMR@h nop lwz r4, 0(r3) nop rlwinm r4, r4, 0, 8, 5 nop oris r4, r4, 0x0200 nop stw r4, 0(r3) nop #endif /* CONFIG_MPC8260ADS */ Can anyone please tell me the importance/significance of this code? Regards, Nishant

5 5

[U-Boot-Users] Bug in cpu/arm920t/imx/speed.c
by Steven Scholz 23 Nov '04

23 Nov '04

Hi there, there are two bugs in cpu/arm920t/imx/speed.c preventing us from determine the correct speed of the processor. CHANGELOG: * Patch by Steven Scholz, 22 Oct 2004: - Fix masking bug in cpu/arm920t/imx/speed.c -- Steven Scholz imc Measurement & Control Index: u-boot.SoC/cpu/arm920t/imx/speed.c =================================================================== RCS file: /cvsroot/u-boot/u-boot/cpu/arm920t/imx/speed.c,v retrieving revision 1.1 diff -u -r1.1 speed.c --- u-boot.SoC/cpu/arm920t/imx/speed.c 9 Oct 2004 22:21:30 -0000 1.1 +++ u-boot.SoC/cpu/arm920t/imx/speed.c 22 Oct 2004 06:59:53 -0000 @@ -41,9 +41,9 @@ { /* FIXME: We assume System_SEL = 0 here */ u32 spctl0 = SPCTL0; + u32 mfn = (spctl0 >> 0) & 0x3ff; u32 mfi = (spctl0 >> 10) & 0xf; - u32 mfn = spctl0 & 0x3f; - u32 mfd = (spctl0 >> 16) & 0x3f; + u32 mfd = (spctl0 >> 16) & 0x3ff; u32 pd = (spctl0 >> 26) & 0xf; mfi = mfi<=5 ? 5 : mfi; @@ -55,9 +55,9 @@ { /* FIXME: We assume System_SEL = 0 here */ u32 mpctl0 = MPCTL0; + u32 mfn = (mpctl0 >> 0) & 0x3ff; u32 mfi = (mpctl0 >> 10) & 0xf; - u32 mfn = mpctl0 & 0x3f; - u32 mfd = (mpctl0 >> 16) & 0x3f; + u32 mfd = (mpctl0 >> 16) & 0x3ff; u32 pd = (mpctl0 >> 26) & 0xf; mfi = mfi<=5 ? 5 : mfi; @@ -74,7 +74,6 @@ ulong get_HCLK(void) { u32 bclkdiv = (( CSCR >> 10 ) & 0xf) + 1; - printf("bclkdiv: %d\n", bclkdiv); return get_systemPLLCLK() / bclkdiv; }

1 1

Re: [U-Boot-Users] Problems with udelay() on AT91RM9200 !
by Steven Scholz 22 Nov '04

22 Nov '04

Wolfgang Denk wrote: > In message <41A1F410.6040107(a)imc-berlin.de> you wrote: > >>I just noticed that the implementation of udelay() - i.e udelay_masked() - >>resets the timer. Why is that? > > Because of a bug? Why didn't I think of this! ;-) But seriously what's the point in letting the timer run with 40 or 80 MHz? In my case udelay won't work for delays > 107ms anyway.... :-( -- Steven

2 1