October 2004 - U-Boot

[U-Boot-Users] u-boot initialize PCI in ixdp425 platform ?
by Annamalai Prakash 04 Nov '04

04 Nov '04

Hi all , I m using u-boot 1.1.1 bootloader on my ixdp425. Everything is working fine.. i have connected miniPCI card in my board. I want know whether u-boot initialize PCI or not ?. coz in include/asm/arch/ixdp425.h, i seen the following lines. #if 0 #ifndef __ASSEMBLY__ extern int ixp425_pci_read(u32 addr, u32 cmd, u32* data); extern int ixp425_pci_write(u32 addr, u32 cmd, u32 data); extern void ixp425_pci_init(void *); #endif #endif My kernel doesnt identify that card also. #define IXP425_PERIPHERAL_BUS_CLOCK (66) /* 66Mhzi APB BUS */ The above line refers the PCI speed ?. Please clear my doubts ?.. Thanks Prakash

3 2

[U-Boot-Users] Another DHCP/BOOTP problem? (Was: "Re: U-Boot DHCP request does n ot time out." and "[PATCH] fix dhcp command retry")
by Larsson Roger 02 Nov '04

02 Nov '04

(resent - I can't find it on archieve) Hi, I see a similar problem to the one described in "Re: U-Boot DHCP request does not time out." with U-Boot 1.1.1.1 But I have not seen any patch from Pantelis. The only related message is "[PATCH] fix dhcp command retry" from Roger Blofeld Does that patch fix the problem? In my case I think the counter restarts at 1 but the reboot BOOTP does not work. If I interrupt (CTRL-C) and issue a 'reset' the BOOTP broad cast after reset works. How to reproduce: * Power on -> BOOTP boot into Linux * In Linux issue 'reboot' command * Warm booted U-BOOT runs into this problem. - - - * U-BOOT CTRL-C and 'reset' -> BOOTP boot into Linux /RogerL >From: Pantelis Antoniou <panto@in...> > Re: U-Boot DHCP request does not time out. > >2004-08-30 23:49 Wolfgang Denk wrote: >> In message <31ADFA827355984B9E2A161514595B561C33A9@lp...> you wrote: >> >>>OMAP1510 Innovator # dhcp >>>Using MAC Address 00:0B:36:00:03:FB >>>BOOTP broadcast 1 >>>BOOTP broadcast 2 >>>BOOTP broadcast 3 >>>BOOTP broadcast 4 >>>BOOTP broadcast 5 >>> >>>Retry count exceeded; starting again >>>BOOTP broadcast 6 >>>... >>>Retry count exceeded; starting again >>>BOOTP broadcast 13 >>>... >> >> >> I see. Same here. Something is really broken. >> >> Best regards, >> >> Wolfgang Denk >> > > I"ve had the same problems. > > Please bear it till I send a patch that fixes that > and a number of other issues in DHCP. > > Next week perhaps since I"m swamped right now. > > Regards > > Pantelis ########################################### This message has been scanned by F-Secure Anti-Virus for Microsoft Exchange. For more information, connect to http://www.F-Secure.com/

2 1

[U-Boot-Users] Hello world application running problem
by Jhavk 01 Nov '04

01 Nov '04

> I have read README.standalone. My target board is > smdk2440(Arm 9). > > So i transfer hello_world using tftp to address > 0x0c100000 and then call "go 0x0c100000 hello world" > > But after that the application seems to get stuck > and > nothing happens. Eventually i have to reset the > board. > > I have some doubts: > 1. Which file should i send. I think it should be > hello_world and not the srec or bin format. I ran > file > command on all 3 and found that hello_world is the > one meant for ARM. > > 2.Is the load and start address correct? May be its > different for ARM9? > > If all these are correct then what may be the > problem? > > > > > > > __________________________________ > Do you Yahoo!? > Take Yahoo! Mail with you! Get it on your mobile > phone. > http://mobile.yahoo.com/maildemo > __________________________________ Do you Yahoo!? Yahoo! Mail Address AutoComplete - You start. We finish. http://promotions.yahoo.com/new_mail

3 4

[U-Boot-Users] hello world application
by Jhavk 31 Oct '04

31 Oct '04

I want to make gzipped hello_world application and later execute it using bootm command. I figured that mkimage is the utility for making compressed STANDALONE image. But it needs an option mkimage -d <datafile> Now what is this datafile? Is it hello_world or hello_world.bin or hello_world.srec I tried all 3, but they give different sizes of gzipped image. So i am confused. My target architecture is ARM. Thanks. __________________________________ Do you Yahoo!? Yahoo! Mail Address AutoComplete - You start. We finish. http://promotions.yahoo.com/new_mail

3 3

[U-Boot-Users] Loading U-Boot SREC file
by Devraj Mukherjee 31 Oct '04

31 Oct '04

Hi everyone, Thanks to everyone who responded to my email. I managed to get a version of UBoot compiled for our target board and have ended up with an SREC file. The boot loader that already exists on the board allows me to load the SREC file via TFTP, and I have succeeded doing that. I am still unsure if we should go ahead burn the image and restart the board to see UBoot come up. If there is anyone out there using a LH7A400 based board then please let me know if I should look out for anything else. Any other suggestions are very welcome. Regards, Devraj -- ---------------------------------------------------------------------- Devraj Mukherjee, Eternity Technologies Pty. Ltd. Australia Host: Debian (Sarge) 3.1 Kernel 2.4.27-1-686 / GCC version 3.3.5 Target: LPD7A400 (ARM9) LogicPD eval. board / ARM GCC 3.3 GlibC 2.3.2 ----------------------------------------------------------------------

3 2

[U-Boot-Users] Re: [ARMboot-users] Asking the concept for a newbie
by Wolfgang Denk 31 Oct '04

31 Oct '04

In message <4183729E.40196727(a)ms10.hinet.net> you wrote: > > I'm a newbie in this field, and deal with an ARM946E based borad. Note that ARMBoot is a dead project; deveoplment continues in the environment of the U-Boot project; followup set to the u-boot-users mailing list. > In my guess, there should have armboot, bzImage, & romfs in the flash. No. bzImage does not work. You will have to use the "mkimage" tool to build a special U-Boot / ARMBoot image. > How are they arranged by address in the flash memory? U-Boot must be locatewd such that it gets started when the CPU comes out of reset. On ARM this usually means that you will have to put it in flash such that it starts at physical address 0x0000. For the other images you can chose which addresses you want - it makes sense to align the images with the sectors of your flash memory, though. > Thanks for any information You are welcome. Best regards, Wolfgang Denk -- See us @ Embedded/Electronica Munich, Nov 09 - 12, Hall A.6 Booth 513 Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd(a)denx.de Lady Nancy Astor: "Winston, if you were my husband, I'd put poison in your coffee." Winston Churchill: "Nancy, if you were my wife, I'd drink it."

1 0

[U-Boot-Users] Re: emulate u-boot in gdb
by Detlev Zundel 30 Oct '04

30 Oct '04

Hi Luc, > I am looking into SkyEye and busy in creating an extension for > potentially ipaq (cheap finished hardware to play with for embedded > software/ui development). > > How can I debug a boot loader such as u-boot in a emulator/simulator > environment ? Good question. I have only thought for a short while about this question, but I think in such an environment, you need only small pieces of U-Boot. Of course you don't have real hardware to setup in a debugger (so I guess, but I have never used one) things like setting up a memory controller etc. will probably not work without emulating all of the hardware of a "real system" - and I guess this is simply not possible to have emulators for all hardware components. So what you would need for such an environment would be U-Boot stripped from all the hardware inits which is probably a major surgery on the sources. Try asking on the U-Boot mailing list (I cc'ed the list) as there are _very probably_ more competent people in this field. > Is it possible to debug u-boot in a arm-linux-gdb environment ? If I > assign this to memory 0x000000 and put 'set $pc =0' should that work > ? ... it didn't when I tried It is of course possible to debug U-Boot on ARM systems - we did ports for some ARM based boards - but again putting U-Boot at address 0 is a much bigger question. What board config did you use? What address does this image think it runs from? And what address was U-Boot linked for? You have to load the U-Boot image there - but again I think the U-Boot hardware inits will cause you grief. > I searched the web but could not find any good result. Can give me > some direction ? I hope I did ;) Cheers Detlev -- ;; Self-replicator in ELisp ((lambda (l) (prin1-to-string (list l (list (quote quote) l)))) (quote (lambda (l) (prin1-to-string (list l (list (quote quote) l))))))

1 0

[U-Boot-Users] Re: u-boot error message
by Detlev Zundel 30 Oct '04

30 Oct '04

Hi Bruce, > Hello Detlev Zundel , > > > I used u-boot-1.1.1 to boot linux.My target board is ALTERA DK1C20 and cpu > is nios.I used tools/mkimage to encapsulate a linux kernel image with > header information.But it display some error messages,it like that: First off - I am not the best person to answer these questions, as I have never used a NIOS system. So please try to get help on the mailing list that I cc'ed my last reply to. Someone there should be able to have better answers to your questions. One more thing though - from your description I am not sure what version of U-Boot you are using and whether you were able _at all_ to boot Linux on your system. The error messages below strongly leave me to not believe this. > Index: tools/mkimage.c > > =================================================================== > > RCS file: /cvsroot/u-boot/u-boot/tools/mkimage.c,v > > retrieving revision 1.13 > > diff -p -u -r1.13 mkimage.c > > --- tools/mkimage.c 10 Jul 2004 22:36:01 -0000 1.13 > > +++ tools/mkimage.c 29 Oct 2004 00:01:12 -0000 > > @@ -86,6 +86,8 @@ table_entry_t arch_name[] = { > > { IH_CPU_MICROBLAZE, "microblaze", "MicroBlaze", }, > > { IH_CPU_MIPS, "mips", "MIPS", }, > > { IH_CPU_MIPS64, "mips64", "MIPS 64 Bit", }, > > + { IH_CPU_NIOS, "nios", "NIOS", }, > > { IH_CPU_PPC, "ppc", "PowerPC", }, > > { IH_CPU_S390, "s390", "IBM S390", }, > > { IH_CPU_SH, "sh", "SuperH", }, Why did you remove the other line from my patch? Obviously this is the patch I suggested and the second line was in there because I have seen that there is still another CPU type not implemented yet in mkimage so one should try to fix this in one go. It does _not make sense_ to throw out such things after the work it has taken to put them in. > When I boot linux image to my target board, it have error messages. > > Like that: > > > > DK1C20 > imi 200000 > > > > ## Checking Image at 00200000 ... > > Image Name: linux kernel > > Image Type: Unknown Architecture Linux Kernel Image (uncompressed) > > Data Size: 1015552 Bytes = 991.8 kB > > Load Address: 01010000 > > Entry Point: 01010000 > > Verifying Checksum ... OK > As you know from studying cmd_bootm.c the imi command _does not_ check the compression type. The image type shows uncompressed so mkimage should have put a 0 in the compression type field. > DK1C20 > bootm 200000 > > ## Booting image at 00200000 ... > > Image Name: linux kernel > > Image Type: Unknown Architecture Linux Kernel Image (uncompressed) > > Data Size: 1015552 Bytes = 991.8 kB > > Load Address: 01010000 > > Entry Point: 01010000 > > Verifying Checksum ... OK > > Unimplemented compression type 39 > > DK1C20 > Yep, there are compression type 0, 1 and 2 but nothing else. According to what I wrote above the compression field is boguos. Of course you noticed reading the source for mkimage that it should have put a 0 in it and now theres a 39 in there. It looks like the image is corrupt for whatever reason - try using the debugger and a hexdump program to find out at what stage this goes wrong. And of course use U-Boot from CVS and _not_ U-Boot 1.1.1 Cheers Detlev PS: Mail f'up-to set _again_ to u-boot-users(a)lists.sourceforge.net. You know of course from the netiquette, that it is considered rude behaviour to ignore such hints repeatedly, don't you? -- (define Y (lambda (f) ((lambda (g) (f (lambda (x) ((g g) x)))) (lambda (g) (f (lambda (x) ((g g) x)))))))

1 0

[U-Boot-Users] McAfee GroupShield Alert
by groupshield＠eucorp.westell.com 30 Oct '04

30 Oct '04

McAfee GroupShield Alert McAfee GroupShield discovered a problem with the following email. See your system administrator for further information. Date/Time sent: 30 Oct 2004 05:19:27 Subject line: U-Boot-Users digest, Vol 1 #1065 - 2 msgs From: u-boot-users-admin(a)lists.sourceforge.net To: u-boot-users(a)lists.sourceforge.net Action taken: Replaced Virus found: W32/Bagle!eml.gen Quarantined item: Filename: Ticket number: 11fc-4183-164f-0002 Copyright © 1993-2003, Networks Associates Technology, Inc. All Rights Reserved. http://www.mcafeesecurity.com <http://www.mcafeesecurity.com/>

1 0

[U-Boot-Users] Delivery Status Notification (Failure)
by postmaster＠doubleclick.net 30 Oct '04

30 Oct '04

This is an automatically generated Delivery Status Notification. Delivery to the following recipients failed. 179-1(a)flonetwork.com

1 0