June 2004 - U-Boot

[U-Boot-Users] compiling u-boot
by Brock Denson 05 Jun '04

05 Jun '04

i am trying to compile u-boot 1.1.1 for the atmel at91rm9200dk under suse linux professional 9.1. i have downloaded u-boot-1.1.1.tar.bz2 and extracted the source code to /usr/src/arm/uboot and tried to complie it with 'make at91rm200_config'. i get the following error: Configuring for at91rm9200dk board.... ln: creating symbolic link 'asm' to 'asm-arm' : Permission denied make: *** [at91rm9200dk_config] Error 1 is there some modifications i need to make to the makefile? do i need to install in a different directory? Can someone help me? Brock Denson WAIV bdenson(a)wa-ventures.com 210 849-1988

2 1

[U-Boot-Users] bootm image problem
by wyu hotmail 04 Jun '04

04 Jun '04

Hi, I'm trying to boot linux image, I do as follow: 1. set env and saved into flash, passed 2. tftpboot, image was loaded into 0x400000 (SDRAM on board 32MB), passed 3. bootm I got the following information: ## Booting image at 00400000 ... Image Name: Linux Kernel Image Image Type: PowerPC Linux Kernel Image (gzip compressed) Data Size: 603024 Bytes = 588.9 kB Load Address: 00000000 Entry Point: 00000000 Verifying Checksum ... OK Uncompressing Kernel Image ... OK ## Current stack ends at 0x01FA2598 => set upper limit to 0x00800000 ## cmdline at 0x007FFC00 ... 0x007FFC50 bd address = 0x01FA2F84 memstart = 0x00000000 memsize = 0x02000000 flashstart = 0x10000000 flashsize = 0x00200000 flashoffset = 0x00000000 sramstart = 0x00000000 sramsize = 0x00000000 bootflags = 0x0000A000 procfreq = 266.666 MHz plb_busfreq = 133.333 MHz pci_busfreq = 33.333 MHz ethaddr = 00:E5:00:08:0F:E5 IP addr = 128.127.18.3 baudrate = 9600 bps No initrd ## Transferring control to Linux (at address 00000000) ... then system seems halt: nothing in terminal anymore. what's the possible problem? Thanks, yuwp

1 0

[U-Boot-Users] problems with BDI2000 / u-boot / Linux on IBM 440GX Ocotea refere nce board
by Bill Hargen 04 Jun '04

04 Jun '04

I have been trying to get u-boot and Linux running on the IBM 440GX "Ocotea" reference board and I have been hitting some weird problems. I'm hoping that someone might have been down this path or have a suggestion on something else to try or check. I got u-boot and Linux working on a 405EP reference board in December, so I've got some experience with this. I'm using the ELDK 3.0 package for the tools and Linux kernel, and am using u-boot version 1.1.1 from the FTP site. I am able to build, install, and run u-boot from flash. And I am able to load a kernel via TFTP and run it (sort of). Right now, I don't have a root file system provided, so I'm getting a panic as expected at that point. My questions are about a few of the quirks that I've hit along the way. The first problem is if I try to run u-boot from flash with a BDI2000 connected (i.e. do "reset" and "go" commands), then u-boot starts running but hangs in the SDRAM init. Specifically, it hangs waiting for the MRS Command Complete status bit to be set. If I power-up without the BDI2000 connected, it starts just fine. With the BDI2000 connected, I can confirm that the bit isn't going true. I've also tried clearing and setting the DCEN enable bit, but the status bit still never goes true. The BDI config file that I am using is a minor modification of the one included with the BDI2000 for the 440GX eval board. And also, with the BDI2000 disconnected, the system reset switch doesn't work. It kills the system, but it doesn't restart - I have to power cycle. It did work when IBM's PIBS code was in flash. The second problem is with Linux startup. I get a number of pages of output on the serial port, including seeing the IDE drives connected to an IDE controller that I have plugged into the PCI bus. It finally stops with a panic due to no rootfs, as expected. But there were a number of other problems along the way: - It reports that the RTC "seems stuck". - It reports about 37 BogoMIPS, which seems way to low for a 440GX running at 533MHz. - It gets read timeouts trying to find the Ethernet PHYs. (And this works from u-boot - I'm booting from emac0!) - It isn't getting the boot command line from u-boot. And this is where it gets very strange. It seems that the kernel is not expecting the register set and data structure that is passed by u-boot. No problem - I've hit that before. So I added the appropriate logic to platform_init() to check the registers and do a strcpy(cmd_line, (char *)(r6 + KERNELBASE)). But that seems to kill the system. Even adding strcpy(cmd_line, "foo") kills it! Yet the calling function copied the default command line into that buffer. But if I change that code to copy it twice then it dies. I have absolutely no clue as to what could cause such behavior! (I'd poke around with the BDI2000, but I can't get past u-boot's SDRAM init when it is connected.) I don't know if the BDI2000 quirk, the reset switch quirk, and the Linux strcpy(cmd_line... quirks are connected or not. Improper SDRAM initialization is usually blamed for such things, but I'm using the existing OCOTEA port which gets the parameters from the SPD chip on the memory DIMM. And yes, I tried two different DIMMs, one of which was the 256MB DIMM that came with the board. Any ideas, suggestions, sample BDI config files, or pointers to a patch that I missed would be welcome! It's probably something simple, just not obvious. Thanks. --Bill

2 1

[U-Boot-Users] RE: Where is the CPU clock speed definition
by wyu hotmail 04 Jun '04

04 Jun '04

Hi, CONFIG_SYS_CLK_FREQ is the input frequency of 405GP's clocking system. The internal PLL uses this frequency and the division parameters strapped from IO pins at power on reset to generate the different clocks needed in 405GP. You can find the detail description about 405GP's clock system in 405GP's user manual, which can be downloaded from IBM's website. yuwp >From: "Frank Young" <young726(a)hotmail.com> >To: u-boot-users(a)lists.sourceforge.net >Date: Thu, 03 Jun 2004 14:19:44 -0400 >Subject: [U-Boot-Users] Where is the CPU clock speed definition > >Hi, > >I am using a WALNUT with PPC405 200MHz CPU. I am trying to install the >u-boot on this board. I have built the image. But before I put the image to >the flash, I want to confirm one thing: do I have to configure the CPU speed >for this board? I checked the file include/configs/WALNUT405.h. I only found >a definition for CONFIG_SYS_CLK_FREQ, which is for external clock. Where is >the CPU clock frequency defined? > >Thanks! > >Frank >

1 0

[U-Boot-Users] Sandpoint8245 questions
by Matthew McClintock 03 Jun '04

03 Jun '04

Hello all, I have some questions for anyone who might know the answer. All the questions apply to the Sandpoint8245 configuration, and might apply to the Sandpoint8240 as well. 1) Who is the maintainer for the Sandpoint boards? 2) Has anyone tested the boards with different memory bus speeds? U-boot appears (it is only an educated guess) to not initialize memory correctly when the memory bus is 66mhz, and it works correctly for a memory bus of 133mhz. I have not tested 100mhz. I am basing this assumption of the fact that when I boot Linux at 66mhz I get weird errors at random times and at 133mhz the board works perfectly. 3) Is anyone aware of the fact that the serial ports on the Sandpoint motherboard itself and the serial ports on the Unity PMC card run at at different clocks? I've done some simple changes to select the correct clock but I just wanted to ask how others might be dealing with this situation. -- Matthew McClintock <mcclintock(a)freescale.com>

2 1

[U-Boot-Users] Where is the CPU clock speed definition
by Frank Young 03 Jun '04

03 Jun '04

Hi, I am using a WALNUT with PPC405 200MHz CPU. I am trying to install the u-boot on this board. I have built the image. But before I put the image to the flash, I want to confirm one thing: do I have to configure the CPU speed for this board? I checked the file include/configs/WALNUT405.h. I only found a definition for CONFIG_SYS_CLK_FREQ, which is for external clock. Where is the CPU clock frequency defined? Thanks! Frank _________________________________________________________________ Stop worrying about overloading your inbox - get MSN Hotmail Extra Storage! http://join.msn.click-url.com/go/onm00200362ave/direct/01/

1 0

RE: [U-Boot-Users] NFS problem
by Woodruff, Richard 03 Jun '04

03 Jun '04

Change your kernel command line to the following and it is likely your NFS mount errors will go away. I also use the 'tcp' option at times, as it further reduced communication problems, you do have to have compiled the server to support the tcp option. root=/dev/nfs rw nfsroot=192.168.55.1:/opt/eldk_3_0/ppc_82xx,rsize=1024,wsize=1024,nolock ip=192.168.55.3:192.168.55.1::255.255.255.0:::off If you read the NFS FAQ you will understand why these can help. They certainly help underpowered ARM's work vastly better over NFS. Regards, Richard W. > -----Original Message----- > From: u-boot-users-admin(a)lists.sourceforge.net [mailto:u-boot-users- > admin(a)lists.sourceforge.net] On Behalf Of Marco Schramel > Sent: Thursday, June 03, 2004 4:04 AM > To: U-BOOT > Subject: [U-Boot-Users] NFS problem > > hi all, > > my 8270 board (2.4.26, u-boot 1.0.1, ELDK3.0) can't boot over NFS. > > The messages are .. > > => bootm 100000 > ## Booting image at 00100000 ... > Image Name: Linux-2.4.26-rc2 > Image Type: PowerPC Linux Kernel Image (gzip compressed) > Data Size: 723855 Bytes = 706.9 kB > Load Address: 00000000 > Entry Point: 00000000 > Verifying Checksum ... OK > Uncompressing Kernel Image ... OK > Memory BAT mapping: BAT2=32Mb, BAT3=0Mb, residual: 0Mb > Linux version 2.4.26-rc2 (marco@linux) (gcc-Version 3.2.2 20030217 (Yellow > Dog Linux 3.0 3.2.2-2a_1)) #2 Do Mai 27 08:48:15 CEST 2004 > On node 0 totalpages: 8192 > zone(0): 8192 pages. > zone(1): 0 pages. > zone(2): 0 pages. > Kernel command line: root=/dev/nfs rw > nfsroot=192.168.55.1:/opt/eldk_3_0/ppc_82xx > ip=192.168.55.3:192.168.55.1::255.255.255.0:::off > Warning: real time clock seems stuck! > Calibrating delay loop... 131.89 BogoMIPS > Memory: 30624k available (1260k kernel code, 432k data, 56k init, 0k > highmem) > Dentry cache hash table entries: 4096 (order: 3, 32768 bytes) > Inode cache hash table entries: 2048 (order: 2, 16384 bytes) > Mount cache hash table entries: 512 (order: 0, 4096 bytes) > Buffer cache hash table entries: 1024 (order: 0, 4096 bytes) > Page-cache hash table entries: 8192 (order: 3, 32768 bytes) > POSIX conformance testing by UNIFIX > Linux NET4.0 for Linux 2.4 > Based upon Swansea University Computer Society NET3.039 > Initializing RT netlink socket > Starting kswapd > Installing knfsd (copyright (C) 1996 okir(a)monad.swb.de). > JFFS2 version 2.1. (C) 2001 Red Hat, Inc., designed by Axis Communications > AB. > CPM UART driver version 0.01 > ttyS0 on SCC1 at 0x8000, BRG1 > ttyS1 on SCC2 at 0x8100, BRG2 > pty: 256 Unix98 ptys configured > eth0: FCC ENET Version 0.3 , 00:11:22:73:44:55 > RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize > loop: loaded (max 8 devices) > NET4: Linux TCP/IP 1.0 for NET4.0 > IP Protocols: ICMP, UDP, TCP > IP: routing cache hash table of 512 buckets, 4Kbytes > TCP: Hash tables configured (established 2048 bind 4096) > IP-Config: Complete: > device=eth0, addr=192.168.55.3, mask=255.255.255.0, > gw=255.255.255.255, > host=192.168.55.3, domain=, nis-domain=(none), > bootserver=192.168.55.1, rootserver=192.168.55.1, rootpath= > NET4: Unix domain sockets 1.0/SMP for Linux NET4.0. > Looking up port of RPC 100003/2 on 192.168.55.1 > Looking up port of RPC 100005/1 on 192.168.55.1 > VFS: Mounted root (nfs filesystem). > Freeing unused kernel memory: 56k init > INIT: version 2.84 booting > Welcome to DENX Embedded Linux Environment > Press 'I' to enter interactive startup. > Building the cache nfs: server 192.168.55.1 not responding, still trying > nfs: server 192.168.55.1 OK > nfs: server 192.168.55.1 not responding, still trying > nfs: server 192.168.55.1 OK > [ OK ] > Mounting proc filesystem: > '''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' '' > '''''''''''''''''' > than it hangs ... > sometimes it gets on a little bit, sometimes not... > Most time it hangs on "INIT: version 2.84 booting" > > Any ideas ?? > > Why the "rootpath" isnt passed to the kernel? > Are the domain and nis-domain necessary for booting ?? > With an initrd it works fine and i can ping the target and reverse. > > Thanks for your ideas > > Marco > > > > --------- > Marco Schramel > R&D > Bartec GmbH > Schulstr. 30 > 94239 Gotteszell, Germany > www.bartec.de > Marco.Schramel(a)go.bartec.de > Phone: +49 (0)9929/301332 > Fax: +49 (0)9929/301112 > > > > ------------------------------------------------------- > This SF.Net email is sponsored by the new InstallShield X. > From Windows to Linux, servers to mobile, InstallShield X is the one > installation-authoring solution that does it all. Learn more and > evaluate today! http://www.installshield.com/Dev2Dev/0504 > _______________________________________________ > U-Boot-Users mailing list > U-Boot-Users(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/u-boot-users

1 0

RE: [U-Boot-Users] NFS problem
by Woodruff, Richard 03 Jun '04

03 Jun '04

Change your kernel command line to the following and it is likely your NFS mount errors will go away. I also use the 'tcp' option at times, as it further reduced communication problems, you do have to have compiled the server to support the tcp option. root=/dev/nfs rw nfsroot=192.168.55.1:/opt/eldk_3_0/ppc_82xx,rsize=1024,wsize=1024,nolock ip=192.168.55.3:192.168.55.1::255.255.255.0:::off If you read the NFS FAQ you will understand why these can help. They certainly help underpowered ARM's work vastly better over NFS. Regards, Richard W. > -----Original Message----- > From: u-boot-users-admin(a)lists.sourceforge.net [mailto:u-boot-users- > admin(a)lists.sourceforge.net] On Behalf Of Marco Schramel > Sent: Thursday, June 03, 2004 4:04 AM > To: U-BOOT > Subject: [U-Boot-Users] NFS problem > > hi all, > > my 8270 board (2.4.26, u-boot 1.0.1, ELDK3.0) can't boot over NFS. > > The messages are .. > > => bootm 100000 > ## Booting image at 00100000 ... > Image Name: Linux-2.4.26-rc2 > Image Type: PowerPC Linux Kernel Image (gzip compressed) > Data Size: 723855 Bytes = 706.9 kB > Load Address: 00000000 > Entry Point: 00000000 > Verifying Checksum ... OK > Uncompressing Kernel Image ... OK > Memory BAT mapping: BAT2=32Mb, BAT3=0Mb, residual: 0Mb > Linux version 2.4.26-rc2 (marco@linux) (gcc-Version 3.2.2 20030217 (Yellow > Dog Linux 3.0 3.2.2-2a_1)) #2 Do Mai 27 08:48:15 CEST 2004 > On node 0 totalpages: 8192 > zone(0): 8192 pages. > zone(1): 0 pages. > zone(2): 0 pages. > Kernel command line: root=/dev/nfs rw > nfsroot=192.168.55.1:/opt/eldk_3_0/ppc_82xx > ip=192.168.55.3:192.168.55.1::255.255.255.0:::off > Warning: real time clock seems stuck! > Calibrating delay loop... 131.89 BogoMIPS > Memory: 30624k available (1260k kernel code, 432k data, 56k init, 0k > highmem) > Dentry cache hash table entries: 4096 (order: 3, 32768 bytes) > Inode cache hash table entries: 2048 (order: 2, 16384 bytes) > Mount cache hash table entries: 512 (order: 0, 4096 bytes) > Buffer cache hash table entries: 1024 (order: 0, 4096 bytes) > Page-cache hash table entries: 8192 (order: 3, 32768 bytes) > POSIX conformance testing by UNIFIX > Linux NET4.0 for Linux 2.4 > Based upon Swansea University Computer Society NET3.039 > Initializing RT netlink socket > Starting kswapd > Installing knfsd (copyright (C) 1996 okir(a)monad.swb.de). > JFFS2 version 2.1. (C) 2001 Red Hat, Inc., designed by Axis Communications > AB. > CPM UART driver version 0.01 > ttyS0 on SCC1 at 0x8000, BRG1 > ttyS1 on SCC2 at 0x8100, BRG2 > pty: 256 Unix98 ptys configured > eth0: FCC ENET Version 0.3 , 00:11:22:73:44:55 > RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize > loop: loaded (max 8 devices) > NET4: Linux TCP/IP 1.0 for NET4.0 > IP Protocols: ICMP, UDP, TCP > IP: routing cache hash table of 512 buckets, 4Kbytes > TCP: Hash tables configured (established 2048 bind 4096) > IP-Config: Complete: > device=eth0, addr=192.168.55.3, mask=255.255.255.0, > gw=255.255.255.255, > host=192.168.55.3, domain=, nis-domain=(none), > bootserver=192.168.55.1, rootserver=192.168.55.1, rootpath= > NET4: Unix domain sockets 1.0/SMP for Linux NET4.0. > Looking up port of RPC 100003/2 on 192.168.55.1 > Looking up port of RPC 100005/1 on 192.168.55.1 > VFS: Mounted root (nfs filesystem). > Freeing unused kernel memory: 56k init > INIT: version 2.84 booting > Welcome to DENX Embedded Linux Environment > Press 'I' to enter interactive startup. > Building the cache nfs: server 192.168.55.1 not responding, still trying > nfs: server 192.168.55.1 OK > nfs: server 192.168.55.1 not responding, still trying > nfs: server 192.168.55.1 OK > [ OK ] > Mounting proc filesystem: > '''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' '' > '''''''''''''''''' > than it hangs ... > sometimes it gets on a little bit, sometimes not... > Most time it hangs on "INIT: version 2.84 booting" > > Any ideas ?? > > Why the "rootpath" isnt passed to the kernel? > Are the domain and nis-domain necessary for booting ?? > With an initrd it works fine and i can ping the target and reverse. > > Thanks for your ideas > > Marco > > > > --------- > Marco Schramel > R&D > Bartec GmbH > Schulstr. 30 > 94239 Gotteszell, Germany > www.bartec.de > Marco.Schramel(a)go.bartec.de > Phone: +49 (0)9929/301332 > Fax: +49 (0)9929/301112 > > > > ------------------------------------------------------- > This SF.Net email is sponsored by the new InstallShield X. > From Windows to Linux, servers to mobile, InstallShield X is the one > installation-authoring solution that does it all. Learn more and > evaluate today! http://www.installshield.com/Dev2Dev/0504 > _______________________________________________ > U-Boot-Users mailing list > U-Boot-Users(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/u-boot-users

1 0

[U-Boot-Users] setup_ramdisk_tag omitted in armlinux.c
by Annamalai Prakash 03 Jun '04

03 Jun '04

Hi there... I found that in U-Boot 1.1.1. version , setup_ramdisk_tag() function is not defined. so that we couldnt pass RAMDISK_TAG to Kernel. I have added the following code with armlinux.c. Now it passing the RAMDISK_TAG. #ifdef CONFIG_RAMDISK_TAG static void setup_ramdisk_tag (bd_t *bd) { /* patch added by PRAKASH */ params->hdr.tag = ATAG_RAMDISK; params->hdr.size = tag_size (tag_ramdisk); params->u.ramdisk.flags = 1; params->u.ramdisk.size = 8192; params->u.ramdisk.start = 0; params = tag_next (params); } #endif /* CONFIG_RAMDISK_TAG */ I dont know is it correct or not .. I have one more query also, should the bootloader load the RAM disk into memory or kernel will take care of ?. My ATAG_INITRD2 node passing my Flash's physical address of initrd. is it also correct or not. becoz my DRAM Starts with 0x00. My flash starts at 0x50000000. So my kernel displays "diabling initrd". plz tell me how to fix it. Thanks Prakash

2 1

Re: [U-Boot-Users] NFS problem
by Laurent Mohin 03 Jun '04

03 Jun '04

Marco, I've encountered the same behaviour and the problem was due to the network interface which misses some packets from time to time. I was using a National DP83846 as the PHY and the clock accuracy (+/- 25 ppm needed) wasn't good enough. Are you using a PHY from national? Laurent P.S : By the way, Wolfgang is right. It's not an u-boot topic. Laurent MOHIN ACTERNA

1 0