March 2004 - U-Boot

[U-Boot-Users] PATCH: PCMCIA and PXA250
by Leon KUKOVEC 15 Apr '04

15 Apr '04

Hi Woflgang, Note that I had to change TEXT_ADDRESS for board/wepep250, otherwise the image does not fit in (with IDE, PCMCIA, FAT support). The TEXT_ADDRESS change is as minimal as it can get. Note that before applying PXA PCMCIA, you need to apply Pierre's patch for IDE. Can anyone that worked on ARM memory layout confirm that please? Changes: o PCMCIA support for PXA Build Report: U-BOOT TOP BUILD REPORT: builds that FAILED to build: AR405 ERIC EVAL5200 MPC8540ADS MPC8560ADS U-BOOT PXA PCMCIA BUILD REPORT: (applied top-pxa-pcmcia patch) builds that FAILED to build: AR405 ERIC EVAL5200 MPC8540ADS MPC8560ADS U-BOOT PXA PCMCIA BUILD REPORT: (applied top-pxa-pcmcia patch + configuration changes) builds that FAILED to build: AR405 ERIC EVAL5200 MPC8540ADS MPC8560ADS Any issues, let me know. -- Best Regards, Leon.

2 4

[U-Boot-Users] Hush 'elif' operation?
by Robin Gilks 12 Apr '04

12 Apr '04

Greetings I'm trying to get as close to a switch statement in hush shell language as possible and thought that a set of if/then/elif/else would be tidier than nested if/then/else. It doesn't work how I expected however. The example below illustrates the problem - I was expecting execution after a true test (in the first 'if' in this case) to continue after the 'fi' but instead it ploughs through all the other executable (echo) statements having skipped the elif statements. It does correctly skip the else clause however :-)) Could someone assist in the navigation of this module to point out where would be a good place to start (i.e. some basic principles and assumptions of how the parser works). [apologies for the word wrapping] =>if iml; then echo "one"; elif iml; then echo "two"; elif iml; then echo "three"; else echo "four"; fi Image at 40340000: Image Name: Linux-2.4.22-20040129-12:57 Image Type: PowerPC Linux Kernel Image (gzip compressed) Data Size: 630253 Bytes = 615.5 kB Load Address: 00000000 Entry Point: 00000000 one two three => Many thanks -- Robin Gilks Senior Design Engineer Phone: (+64)(3) 357 1569 Tait Electronics Fax : (+64)(3) 359 4632 PO Box 1645 Christchurch Email : robin.gilks(a)tait.co.nz New Zealand

2 2

[U-Boot-Users] 8260 PCI
by Rune Torgersen 01 Apr '04

01 Apr '04

Something subtle broke in the 826x PCI support from 1.0.0 to top of CVS (1.0.2?) I havent figured out what yet, but if I use the cpu/mpc8260/pci.c from 1.0.0, everything works. If I use the new one, the Silicon Image IDE disk driver in Linux (2.6.3) causes a kernel exception. Rune Torgersen System Developer Innovative Systems LLC 1000 Innovative Drive Mitchell, SD 57301 Ph: 605-995-6120 www.innovsys.com

2 1

[U-Boot-Users] Re: boot commands
by Stephen Williams 31 Mar '04

31 Mar '04

Stephen Williams wrote: > > I am setting up my device to autoboot, and I want to compile > in a default boot command (a la CONFIG_BOOTCOMMAND) and only > use the bootcmd variable if it is defined. I want to use the > other environment for other things, though. > > It appears to me, though, that in main.c autoboot will be > prevented if getenv("bootcmd") returns 0, no matter what > the CONFIG_BOOTCOMMAND value is set. > > Am I mis-reading this? Is this an error in need of fixing? > Arg! It seems like *all* the autoboot related variables, including BOOTARGS and AUTOBOOT_PROMPT are ignored if there is *any* environment. This makes no sense to me. Surely people want to set per-board stuff in environment and keep common autoboot settings compiled in. My BOOTCOMMAND and BOOTARGS together are too large to fit into my 256byte EEPROM. -- Steve Williams "The woods are lovely, dark and deep. steve at XXXXXXXXXX But I have promises to keep, http://www.XXXXXXXXXX and lines to code before I sleep, http://www.picturel.com And lines to code before I sleep."

2 2

[U-Boot-Users] boot commands
by Stephen Williams 31 Mar '04

31 Mar '04

I am setting up my device to autoboot, and I want to compile in a default boot command (a la CONFIG_BOOTCOMMAND) and only use the bootcmd variable if it is defined. I want to use the other environment for other things, though. It appears to me, though, that in main.c autoboot will be prevented if getenv("bootcmd") returns 0, no matter what the CONFIG_BOOTCOMMAND value is set. Am I mis-reading this? Is this an error in need of fixing? -- Steve Williams "The woods are lovely, dark and deep. steve at XXXXXXXXXX But I have promises to keep, http://www.XXXXXXXXXX and lines to code before I sleep, http://www.picturel.com And lines to code before I sleep."

1 0

[U-Boot-Users] FCC & Linux
by uboot＠sundmangroup.com 31 Mar '04

31 Mar '04

Hello, I have U-boot-1.0.2 working on our custom board and the FCC1 is working fine, I can TFTP the linux Kernel and start it. Now, here comes a problem I do not really understand. U-boot has configured the FCC fine. Why does Linux need to re-configure it, when it tries to do that by means of MII & MDIO it does not find any PHY...and I lose the ethernet port. I am using MontaVista 3.1 and the 8260ADS LSP as a base for our Linux port. Thanks in advance for your help! Cheers // Matias

2 1

[U-Boot-Users] Support POST on MPC824x_based board
by Birkin Shen 31 Mar '04

31 Mar '04

Dear all: I make our MPC824x_based board support POST, but not support watchdog POST. I add a item "unsigned long post_word" in struct global_data in include/asm-ppc/global_data.h instead of the macro CPM_POST_WORD_ADDR such as in function post_word_load() in file cpu/mpc8260/commproc.c. Since all items in global_data struct gd_t will be initiated to "0" after reboot, the post_word can't remember whether the watchdog POST has been executed. So, watchdog POST can't be supported. I try to find a free 32 bits register in MPC824x to place the post_word, but I can't find the gold register. :( Now, my poor POST can support Memory,CPU,I2C,RTC. I also added the Cache,UART to the POST list, but I still have no chance to debug my codes of Cache and UART POST in our board (our board is being used to debug linux drivers now.). I'm not sure of my method. Any advice will be welcome, really. Regards, Birkin

1 0

[U-Boot-Users] [path #2.2/12] Intracom boards patch
by Pantelis Antoniou 31 Mar '04

31 Mar '04

Hello The following patch adds two new Intracom boards in u-boot and fixes to one more. [part 2] Regards Pantelis board/netta/Makefile | 40 + board/netta/config.mk | 28 + board/netta/dsp.c | 1031 +++++++++++++++++++++++++++++++++++++++++++ board/netta/flash.c | 511 +++++++++++++++++++++ board/netta/netta.c | 575 +++++++++++++++++++++++ board/netta/u-boot.lds | 138 +++++ board/netta/u-boot.lds.debug | 135 +++++ common/cmd_pcmcia.c | 364 +++++++++++++++ include/configs/NETPHONE.h | 694 ++++++++++++++++++++++++++++ 9 files changed, 3516 insertions(+)

1 0

[U-Boot-Users] [path #2.3/12] Intracom boards patch
by Pantelis Antoniou 31 Mar '04

31 Mar '04

Hello The following patch adds two new Intracom boards in u-boot and fixes to one more. [part 3] Regards Pantelis configs/NETTA.h | 745 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ configs/NETVIA.h | 35 +- pcmcia.h | 2 status_led.h | 3 4 files changed, 770 insertions(+), 15 deletions(-)

1 0

[U-Boot-Users] [path #2.1/12] Intracom boards patch
by Pantelis Antoniou 31 Mar '04

31 Mar '04

Hello The following patch adds two new Intracom boards in u-boot and fixes to one more. [part 1] Regards Pantelis CREDITS | 2 MAKEALL | 2 Makefile | 16 README | 3 board/netphone/Makefile | 40 + board/netphone/config.mk | 28 board/netphone/flash.c | 511 ++++++++++++++++++ board/netphone/netphone.c | 621 ++++++++++++++++++++++ board/netphone/phone_console.c | 1123 ++++++++++++++++++++++++++++++++++++++++ board/netphone/u-boot.lds | 138 ++++ board/netphone/u-boot.lds.debug | 135 ++++ 11 files changed, 2616 insertions(+), 3 deletions(-)

1 0