March 2004 - U-Boot

[U-Boot-Users] [PATCH] make some targets more related to latest NFS patch
by Stephan Linz 14 Mar '04

14 Mar '04

Hi all, ... smal bug fix for some targets without CFG_CMD_NFS support (esp. Nios) ... fixes broken compilation run ... Best Regards, Stephan Linz

2 1

[U-Boot-Users] [PATCH] lba48 support & 64bit vsprintf
by Rune Torgersen 14 Mar '04

14 Mar '04

Added some CONFIG and CFG variables to for LBA48 support and for 64bit printing. Now you can use LBA48 and not use 64bit in vsprintf. LBA48 now works without 64bit. Changelog Rune Torgersen - Added LBA48 support (CONFIG_LBA48 & CFG_64BIT_LBA) - Added support for 64bit printing in vsprintf (CFG_64BIT_VSPRINTF) - Added support for 64bit strtoul (CFG_64BIT_STRTOUL) README is also updated for these variables 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] patch to add autoload via NFS for rarpboot command
by Masami Komiya 13 Mar '04

13 Mar '04

I send the patch against net/rarp.c to fix and add autoload by NFS. Best regards Masami Komiya

2 1

[U-Boot-Users] Patch: flash_write return value in saveenv
by Eisenhut, Daniel (MED) 13 Mar '04

13 Mar '04

In the CFG_ENV_ADDR_REDUND version of the function saveenv in common/env_flash.c... If there is a failure with the flash_write function it will always display an error message for a Flash Timeout regardless of what the actual error is. The variable "rc" is set to 1 at declaration and flash_perror(rc) is called if there is an error, but the actual return value of flash_write is not stored in rc, therefore the error code displayed will always be 1. The below patch corrects this... Dan diff -uNr clean/common/env_flash.c u-boot-1.0.2/common/env_flash.c --- clean/common/env_flash.c 2004-02-25 16:36:28.000000000 -0600 +++ u-boot-1.0.2/common/env_flash.c 2004-02-26 12:55:38.000000000 -0600 @@ -207,21 +207,21 @@ debug (" %08lX ... %08lX ...", (ulong)&(flash_addr_new->data), sizeof(env_ptr->data)+(ulong)&(flash_addr_new->data)); - if (flash_write(env_ptr->data, + if ((rc = flash_write(env_ptr->data, (ulong)&(flash_addr_new->data), - sizeof(env_ptr->data)) || + sizeof(env_ptr->data))) || - flash_write((char *)&(env_ptr->crc), + (rc = flash_write((char *)&(env_ptr->crc), (ulong)&(flash_addr_new->crc), - sizeof(env_ptr->crc)) || + sizeof(env_ptr->crc))) || - flash_write((char *)&obsolete_flag, + (rc = flash_write((char *)&obsolete_flag, (ulong)&(flash_addr->flags), - sizeof(flash_addr->flags)) || + sizeof(flash_addr->flags))) || - flash_write((char *)&active_flag, + (rc = flash_write((char *)&active_flag, (ulong)&(flash_addr_new->flags), - sizeof(flash_addr_new->flags))) + sizeof(flash_addr_new->flags)))) { flash_perror (rc); goto Done; diff -uNr clean/CHANGELOG u-boot-1.0.2/CHANGELOG --- clean/CHANGELOG 2004-02-26 13:02:21.000000000 -0600 +++ u-boot-1.0.2/CHANGELOG 2004-02-26 13:03:33.000000000 -0600 @@ -2,6 +2,9 @@ Changes for U-Boot 1.0.2: ====================================================================== +* Patch by Dan Eisenhut, 26 Feb 2004: + fix flash_write return value in saveenv + * Patch by Masami Komiya, 24 Feb 2004: Update NetBootFileXferSize in NFS code

2 1

[U-Boot-Users] 16C552 parallel port initialization problems in u-boot
by Sam Song 13 Mar '04

13 Mar '04

Hi, I am working adding support of parallel port of ST16C552 in u-boot but have some weird initialization problems.So I list my related programs and questions as follows for any further help. ***Problems*** 1. How come board_parallel_init became invalid when I put it in lib_ppc/board.c board_init_f() for debugging? 2. Why could I only get 0x00 or 0xff initialization results for the parallel port-PD0-PD7 when I changed par_port->pr1 in ns16552.c as different values like 0x55 or 0xAA?It seemed that the PR register only accpected 0x00 or 0xff as it's first value.But I could change it's value correctly by u-boot command. 3. What's meaning of "__attribute__ ((packed))" in NS16552 structure definition?I copyed it from ns16550.h but don't understand it's function. ***16C552 parallel port register content*** u-boot>md fa010000 fa010000: ff67e0e0 ff67e0e0 ff67e0e0 ff67e0e0 .g fa010010: ff67e0e0 ff67e0e0 ff67e0e0 ff67e0e0 .g u-boot>mm.b fa010000 fa010000: ff ? 55 fa010001: 67 ? 55 fa010002: e0 ? e4 fa010003: e4 ? e4 fa010004: 55 ? . u-boot>md fa010000 fa010000: 5567e4e4 5567e4e4 5567e4e4 5567e4e4 Ug. fa010010: 5567e4e4 5567e4e4 5567e4e4 5567e4e4 Ug. u-boot> When setting 'par_port->pr1 = 0xAA' ,it displayed: fa010000: ff67e0e0 ff67e0e0 ff67e0e0 ff67e0e0 ***Program*** [MYBOARD.h] /*------------------------------------- * NS16552 PRINTER PORT Configuration */ #define CFG_PORT_BASE 0xFA010000 #define CFG_NS16552 #define CONFIG_PARALLEL #define CFG_NS16552_PORT CFG_PORT_BASE #ifdef CONFIG_PARALLEL #define CFG_BR6_PRELIM 0xFA010401 /* REGISTER CS6 */ #define CFG_OR6_PRELIM 0xFFFF0040 #endif [include/ns16552.h] #ifdef CONFIG_PARALLEL /* Sam */ struct NS16552 { unsigned short pr0; /* 0 */ unsigned short sr; /* 1 */ unsigned short com; /* 2 */ } __attribute__ ((packed)); #endif #define pr1 pr0 #define iosel sr /* Sam */ #define con com void NS16552_init (NS16552_t par_port); [driver/ns16552.c] #include <config.h> #ifdef CFG_NS16552 #include <ns16552.h> void NS16552_init (NS16552_t par_port) { par_port->iosel = 0x55; par_port->pr1 = 0xFF; par_port->con = (CON_LOGIC3 | CON_LOGIC2 | ~CON_INTP | CON_INIT ); } #endif [driver/parallel.c] #include <ns16552.h> static NS16552_t port = (NS16552_t) CFG_NS16552_PORT; int parallel_init (void) { NS16552_init (port); return (0); } [lib_ppc/board.c] #include <config.h> --In board_init_r()--- #if defined(CONFIG_MYBOARD) board_parallel_init(); #endif Sorry for disturbing your nice weekend by many questions!Thanks in advance! Sam _________________________________________________________ Do You Yahoo!? 完全免费的雅虎电邮，马上注册获赠额外60兆网络存储空间 http://cn.rd.yahoo.com/mail_cn/tag/?http://cn.mail.yahoo.com

1 0

[U-Boot-Users] ML300 NFS rfs mount problem
by Mike Wellington 12 Mar '04

12 Mar '04

My kernel command line is set as follows in include/configs/ml300.h > #define CONFIG_BOOTARGS "console=ttyS0,9600 root=/dev/nfs rw \ > nfsroot=135.4.21.60:/opt2/timesys/linux/4.0/amxap100/rfs \ > rootpath=/opt2/timesys/linux/4.0/amxap100/rfs \ > ip=135.4.21.155:135.4.21.60:135.4.21.254:255.255.254.0:prolab6::off" However when the kernel outputs its command line during the boot process I see: "Kernel command line: console=ttyS0,9600 root=/dev/nfs rw nfsroot=135.4.21.60:/opt2/timesys/linux/4.0/amxap100/rfs rootpath=/opt2/timesys/linux/4.0/amxap100/rfs ip=135.4.21.155:135.4.21.60:135.4.21.254:255.255.254.0:prolab6::off" Then, later I see: "IP-Config: Complete: device=eth0, addr=135.4.21.155, mask=255.255.254.0, gw=135.4.21.254, host=prolab6, domain=, nis-domain=(none) bootserver=135.4.21.60 rootserver=135.4.21.60 rootpath=" Note that rootpath is now empty, even though I set it on the command line. Can somebody give me a clue? -- =mike wellington wellington(a)lucent.com 303.920.6412 Desk 720.434.7559 Cell

2 1

Re: [U-Boot-Users] Newbie needs help with RAMDISK
by Luke Fimmerzane 12 Mar '04

12 Mar '04

Sorry for the wrong posting. Because I was slavely following the set-up in the book I referred to in my earlier email for configuring u-boot and running the root file system from RAM disk, I thought someone else might also have followed the same path and might have indicated whether it worked for them or not. The info in the book suggested that as long as you did a bootm <kernel address> <RAM disk address> it should work. I've also read this in the README but I think its only in the context of PPC. I do have ext2 configured in my kernel but maybe I need other options also enabled. (What kind of message should I get if its not enabled) Also what kernel command line should I use? Is root =/dev/ram correct? Do I have to set-up a u-boot envionment variable for a RAM disk? I don't have access to the hardware for the next few days so can't send the full error message list you mentioned. But I want to prepare as much as possible. Thanks again, Luke I know I am present at the alter of the gods of booting and linux and I apologise if my requests fall short...... :( >From: Wolfgang Denk <wd(a)denx.de> >To: "Luke Fimmerzane" <lukefimmerzane(a)hotmail.com> >CC: u-boot-users(a)lists.sourceforge.net >Subject: Re: [U-Boot-Users] Newbie needs help with RAMDISK Date: Wed, 10 >Mar 2004 19:02:42 +0100 > >Dear Luke, > >in message <BAY10-F113QOTQWKySS0001f3e9(a)hotmail.com> you wrote: > > > > I've managed to create the ramdisk, download it to my target but when I >boot > > I get this message (only relevant sections included): > > > > Kernel command line: root=/dev/ram > >Actually this is a Linux problem. Strictly speaking, this is off >topic on the U-Boot list. > > > ...... > > RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize > > ...... > > ..... > > cramfs: wrong magic > > FAT: bogus logical sector size 65535 > > Kernel panic: VFS: Unable to mount root fs on 01:00 > > > > AAgh!!! Where could I have gone wrong!! > >One of your mistakes is not providing enough information. It would >have been VERY useful to see the COMPLETE log of the boot messages. >So we can only speculate. Do you have ext2 support enabled in your >kernel configuration? > > > I know this is a voluntary site but if someone could give me some give >me a > > few hints then thanks. > >This is what these lists are for. > >Best regards, > >Wolfgang Denk > >-- >Software Engineering: Embedded and Realtime Systems, Embedded Linux >Phone: (+49)-8142-4596-87 Fax: (+49)-8142-4596-88 Email: wd(a)denx.de >"They that can give up essential liberty to obtain a little temporary >saftey deserve neither liberty not saftey." - Benjamin Franklin, 1759 _________________________________________________________________ Express yourself with cool emoticons - download MSN Messenger today! http://www.msn.co.uk/messenger

2 1

[U-Boot-Users] Flash partions ...
by Wolfgang Kanovsky 12 Mar '04

12 Mar '04

Hi all does any one know where i have to fill an mtd structure in uboot when i wonna have partitions on my flash ? i patched the linux kernel ...with mtd struct ..but where can i do the same for uboot ? /* partition_info gives details on the logical partitions that the split the * single flash device into. If the size if zero we use up to the end of the * device. */ static struct mtd_partition partition_info[]={ { .name = "HopfSC520 boot kernel", .offset = 0, .size = 0xc0000 }, { .name = "HopfSC520", .offset = 0xc0000, .size = 0x40000 }, { .name = "HopfSC520 file system 15MB", .offset = 0x100000, .size = 0xe80000 }, { .name = "HopfSC520 U-boot 512K", .offset = 0xf80000, .size = 0x80000 }, }; #define NUM_PARTITIONS (sizeof(partition_info)/sizeof(partition_info[0])) best regards Wolfgang Kanovsky

2 1

[U-Boot-Users] IPMI support
by Figini Enzo 12 Mar '04

12 Mar '04

Looking u-boot tree, I see that that u-boot doesn't support IPMI. For my board MPC8280 based it's necessary introduce it in order to be used in ATCA shelf for boot and pre-boot phases. At the end of work I'll submit a patch with support for IPMI using SMBUS (i2c). Does anybody is interested to this ???? thanks Enzo ---------------------------------------------------------------------------- -- Figini Enzo senior software engineer Italtel S.p.A. via G.Reiss Romoli Castelletto di Settimo M. - Milano - 20019 - Italy phone : +39 2 4388 8774 mobile : +39 2 4388 9793 www.italtel.it

2 1

[U-Boot-Users] MPC8245 help needed.....
by Jian-Bang Lai 12 Mar '04

12 Mar '04

Dear all, I am fighting with a custom MPC8245 board and u-boot. However, I can't get any message from the on chip serail port. I notice the u-boot initialization serquence from the source code. It contain a lot of initail sub-routine, get_clocks, init_timebase, env_init, init_baudrate, serial_init, console_init_f, display_options, checkcpu, etc.. My question is that does the on chip serial port show the U-Boot_version message after U-boot finish the console initialization, if my configuration is correct (even some initalization sub-routines behind the console initialization is going wrong)? I focus on my memory initializtion first, because I assume the serial port will output some messages after the serial port initializtion finish. Any advice? Ban _________________________________________________________________ 下載最新版本的 MSN Messenger：和朋友在線上聊天，用有趣的表情符號傳達心意! http://messenger.msn.com.tw/

1 0