June 2007 - U-Boot

[U-Boot-Users] Linux kernel hang (infinite loop)
by Madhu Saravana Sibi Govindan 02 Jan '08

02 Jan '08

Hello all, I'm using U-boot to bring up a custom 440gp-based board. We are able to transfer the kernel image (uImage) to the target board's memory and jump to it. But, the kernel seems to hang after that. We have verified that the board info struct and the clock frequencies are ok. We also have a BDI2000 hardware debugger with us, but we are unable to set any breakpoints using this (both using telnet or using gdb). However, I can use the telnet interface and single-step the 440gp after it seems to hang. I found that 440gp is infinitely looping through the "DataTLBError" routine. It is unclear to me what might have caused this in the first place. Has anyone seen this behaviour before? We are currently working on getting breakpoints to work using the BDI2000. Any help is greatly appreciated. Regards, G.Sibi

4 11

Re: [U-Boot-Users] boot up u-boot from RAM
by Nghiem, Tien 26 Nov '07

26 Nov '07

Lei, You have to rebuild your code to put your code in the RAM area (other than the FLASH area). Don't use the address 0x00-0x2fff because its reserved for the exception vectors usually. I have a Lite5200 EVB. I use the address 0x20000 to put my u-boot in RAM and it works. I don't know if it could be done for the mpc8270. Tien -----Original Message----- From: u-boot-users-bounces(a)lists.sourceforge.net [mailto:u-boot-users-bounces@lists.sourceforge.net]On Behalf Of Lei Sun Sent: Thursday, October 26, 2006 4:15 PM To: u-boot-users(a)lists.sourceforge.net Subject: [U-Boot-Users] boot up u-boot from RAM Hi all: I am debugging a costomized board based on mpc8270, due to the design flaw, we can't boot up the system using flash, so I was trying to debug u-boot by loading it into the RAM. I used JTAG debugger to initialize the SDRAM and load the u-boot to 0x0, after reset, the PC is at 0x100. the disassembled code looks correct on that location, however, i can't single step start from 0x100. Is that because 0x100 is the reset vector ? I tried to put some breakpoint at some places not far from start point, but it gave me "unknown halt at 0x00000000" error after "run" command. after that the memory seems corrupted. Too bad I was using a bad debugger(Guardian-SE), I have a evaluation copy of the codewarrior plus USB+TAP from freescale. I tried that on our PQ2FADS-VR SDK. After reset, and CPU halt, I can't see any disassembed code on 0x100 (the uboot is already in the flash, and I assume it should read data from CS0 upon reset). Although the code does execute after hitting "run" button. I wonder why can't I see the disassembed code after reset? Thanks lei ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ U-Boot-Users mailing list U-Boot-Users(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/u-boot-users

4 3

[U-Boot-Users] AMCC440SPe & PCI Express
by Patil, Pankaj P. 26 Nov '07

26 Nov '07

Hi, I have been using u-boot-1.1.5 on my custom 440SPe based board. I am able to establish a link on PCIE0. However, I am having trouble accessing outbound Memory on PCIE0. My memory map seetings for Outbound PCIE memory are the default U-boot Settings. I was curious if anybody's tried accessing PCIE Outbound memory with the default settings. I seem to get a bus error when I try to access that region. Thanks Pankaj

3 3

Re: [U-Boot-Users] adding support to new board
by Jeff Mann 26 Sep '07

26 Sep '07

See http://www.freescale.com/files/32bit/doc/app_note/AN3220.pdf -JM -----Original Message----- From: u-boot-users-bounces(a)lists.sourceforge.net [mailto:u-boot-users-bounces@lists.sourceforge.net] On Behalf Of Maximus Sent: Saturday, July 15, 2006 5:31 AM To: U-Boot-Users(a)lists.sourceforge.net Subject: [U-Boot-Users] adding support to new board Hi, Is there any document (step by step) approach on how to add u-boot support to a new board? any advices, Regards, jo ------------------------------------------------------------------------ - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ U-Boot-Users mailing list U-Boot-Users(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/u-boot-users

4 3

[U-Boot-Users] helloworld app from flash?
by Ram 08 Sep '07

08 Sep '07

Hi, im trying to run the helloworld application 1st Option: download the hello_world.bin to some location on RAM and execute it OMAP5912OSK: tftpboot 0x10000000 hello_world.bin OMAP5912OSK: go 0x10000004 This works fine. no problem 2nd Option: I want to burn the hello_world.bin into flash. During startup i want u-boot to launch it after booting - just like the way it boots the kernel. It does not work for me, this is what i did. $ ../tools/mkimage -A arm -O linux -T kernel -C gzip -a 0x10000000 -e 0x10000004 -n 'Hello Image' -d hello_world.bin.gz hello.img I downloaded the hello.img and burnt it to 0x40000 I did: setenv bootcmd bootm 0x40000 Now when i restarted the boot i get the following:- Example expects ABI version 2 Actual U-Boot ABI version 2 Hello World argc = 0 data abort pc : [<10000058>] lr : [<10000048>] sp : 1103fc10 ip : ffffffff fp : 00000001 r10: 110996d8 r9 : 1103fca4 r8 : 1103ffdc r7 : 110996d8 r6 : 00000203 r5 : 00000000 r4 : 00000000 r3 : 00000020 r2 : 00000001 r1 : 0000000a r0 : 00000000 Flags: nZCv IRQs off FIQs off Mode SVC_32 Resetting CPU ... All i am doing in the hello world example is to read a line and print the line entered. I am using readline function for doing that. The programs works fine if i download it to RAM and say go. but if i flash it it does not work?. Am i missing something?. Regards, sriram

4 6

[U-Boot-Users] CFG_MONITOR_BASE < CFG_FLASH_BASE
by Timur Tabi 16 Aug '07

16 Aug '07

In several board config files (namely 8xxx boards), I see code that looks like this: #if (CFG_MONITOR_BASE < CFG_FLASH_BASE) #define CFG_RAMBOOT #else #undef CFG_RAMBOOT #endif ... #ifndef CFG_RAMBOOT #define CFG_ENV_IS_IN_FLASH 1 #define CFG_ENV_ADDR (CFG_MONITOR_BASE + CFG_MONITOR_LEN) #define CFG_ENV_SECT_SIZE 0x40000 /* 256K(one sector) for env */ #define CFG_ENV_SIZE 0x2000 #else #define CFG_NO_FLASH 1 /* Flash is not usable now */ #define CFG_ENV_IS_NOWHERE 1 /* Store ENV in memory only */ #define CFG_ENV_ADDR (CFG_MONITOR_BASE - 0x1000) #define CFG_ENV_SIZE 0x2000 #endif So CFG_RAMBOOT is defined if U-Boot is located in RAM instead of in Flash. On a system that has flash, why would we ever support this configuration? It doesn't even make any sense, because a whole bunch of DDR and LAWBAR setup needs to happen before RAM can even be accessed, so you can't copy u-boot to RAM and reset the system. -- Timur Tabi Linux Kernel Developer @ Freescale

6 16

[U-Boot-Users] USB custodian repo updates
by Markus Klotzbücher 10 Aug '07

10 Aug '07

Dear List, The following changes since commit 09444143670c9c2243cb7aba9f70b3713d33bed1: Markus Klotzbuecher (1): Change duplicate usb_cpu_init_fail to usb_board_init_fail are found in the git repository at: git://www.denx.de/git/u-boot-usb.git Markus Klotzbuecher (4): USB/OHCI: endianness cleanup in the generic ohci driver USB: ohci fixes and cleanup for mpc5xxx and IceCube board config USB: ohci fixes and cleanup for ppc4xx and yosemite board. TRAB, USB: update trab board configuration for use of generic ohci driver Rodolfo Giometti (2): ISP116x: delay for crappy USB keys Files include/linux/byteorder/{big,little}_endian.h define Zhang Wei (3): USB event poll support USB PCI-OHCI, interrupt pipe and usb event poll support Add USB PCI-OHCI, USB keyboard and event poll support to the README | 3 + common/usb_kbd.c | 6 +- cpu/mpc5xxx/usb.c | 54 ++++++ cpu/ppc4xx/Makefile | 2 +- cpu/ppc4xx/usb.c | 50 ++++++ doc/README.generic_usb_ohci | 49 ++++-- drivers/isp116x-hcd.c | 1 + drivers/usb_ohci.c | 386 ++++++++++++++++++++++++++++++----------- drivers/usb_ohci.h | 12 +- drivers/usbtty.c | 6 +- include/configs/IceCube.h | 9 +- include/configs/MPC8641HPCN.h | 13 ++ include/configs/trab.h | 3 +- include/configs/yosemite.h | 1 + include/usb.h | 1 + 15 files changed, 462 insertions(+), 134 deletions(-) create mode 100644 cpu/mpc5xxx/usb.c create mode 100644 cpu/ppc4xx/usb.c This also contains a big endianness cleanup which introduces two new config options: CFG_OHCI_BE_CONTROLLER: The ohci controller is big endian CFG_OHCI_SWAP_REG_ACCESS: For PCI, do byte swapping for register accesses. I successfully tested these changes on arm (trab board), ppc4xx (yosemite) and mpc5xxx (lite5200b). If no problems show up, i'll ask Wolfgang to pull in about two weeks (this time really). Viele Grüße / Best regards Markus Klotzbücher -- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: +49-8142-66989-0 Fax: +49-8142-66989-80 Email: office(a)denx.de

4 5

[U-Boot-Users] [PATCH] m68k fix for Ethernet Buffers at address 0
by Wilson Callan 10 Aug '07

10 Aug '07

Hi, Attached is a patch that fixes a problem i saw with a coldfire 5235 port. i'll include the patch here in the email as well since it describes the problem: diff -purN u-boot-1.1.6/lib_m68k/board.c u-boot-1.1.6-pktfix/lib_m68k/ board.c --- u-boot-1.1.6/lib_m68k/board.c 2006-11-02 09:15:01.000000000 -0500 +++ u-boot-1.1.6-pktfix/lib_m68k/board.c 2007-06-28 11:07:19.000000000 -0400 @@ -642,6 +642,22 @@ void board_init_r (gd_t *id, ulong dest_ #if (CONFIG_COMMANDS & CFG_CMD_NET) && defined(FEC_ENET) WATCHDOG_RESET(); + + /* init NetRxPackets here otherwise eth_init() will cause + * buffer descriptors to use 0 as their buffer and overwrite lower memory. + * NetLoop() does this properly later. + * Cannot remove eth_init() call because linux driver assumes this + * happened, for MAC Address to be loaded into microcontroller + */ + { + int i; + static uchar pkt[(PKTBUFSRX+1) * PKTSIZE_ALIGN + PKTALIGN]; + uchar *tx = pkt + (PKTALIGN - 1); + tx -= (ulong) tx % PKTALIGN; + for (i = 0; i < PKTBUFSRX; i++) { + NetRxPackets[i] = tx + (i+1)*PKTSIZE_ALIGN; + } + } eth_init(bd); #endif

6 22

[U-Boot-Users] ROMFS
by Michal Simek 05 Aug '07

05 Aug '07

ROMFS: Add support for ROMFS filesystem based on MTD system. signed-off-by: Michal Simek <monstr(a)monstr.eu> Patch is in attachement. Best regards, Michal Simek

6 13

[U-Boot-Users] New architecture -- TI DaVinci
by Peter Pearse 18 Jul '07

18 Jul '07

Sergey Apologies - you fooled me with the "New Architecture". I did a quick google & found the TMS320DM6443 was a DSP, so assumed it would be a new DSP architecture, rather than being part of arm. I don't always know where ARM cores have been used - it helps if the email subject mentions ARM. I'll add your patch to my list. Peter -- IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.

5 8