October 2005 - U-Boot

[U-Boot-Users] PATCH: Point e500 IVPR at memory during code relocation
by Kylo Ginsberg 05 Oct '05

05 Oct '05

Hello, Please consider the following e500-specific patch, which reloads the IVPR to point at ram when code is relocated to ram. The IVPR assumes 64KB alignment, so code is copied to a 64K aligned region. * Patch by Kylo Ginsberg, 13 Apr 2005: - repoint IVPR at ram when code is relocated Thanks, Kylo

2 1

[U-Boot-Users] fix patch u-boot/lib_generic string.c,1.4,1.5
by Andrew Dyer 04 Oct '05

04 Oct '05

The patch for lib_generic/string.c:strswab() from cvs rev. 1.4 to 1.5 is a bit different than what I sent in. Attached is an update. CHANGELOG *fix loop end test in lib_generic/string.c:strswab() Patch by Andrew Dyer, October 10, 2005 Signed-off-by: Andrew Dyer <amdyer(a)gmail.com> Index: lib_generic/string.c =================================================================== RCS file: /home/cvsroot/Projects/u-boot/lib_generic/string.c,v retrieving revision 1.1.1.2 retrieving revision 1.3 diff -p -u -r1.1.1.2 -r1.3 --- lib_generic/string.c 4 Oct 2005 14:15:53 -0000 1.1.1.2 +++ lib_generic/string.c 4 Oct 2005 15:00:44 -0000 1.3 @@ -380,7 +380,7 @@ char *strswab(const char *s) return (NULL); } - for (p=(char *)s, q=p+1; (*p != '\0') && (*p != '\0'); p+=2, q+=2) { + for (p=(char *)s, q=p+1; (*p != '\0') && (*q != '\0'); p+=2, q+=2) { char tmp; tmp = *p; -- Hardware, n.: The parts of a computer system that can be kicked.

2 1

[U-Boot-Users] [PATCH] ARM supplied & supported development boards. Errata & rationalization - corrected patch
by Peter Pearse 04 Oct '05

04 Oct '05

Attached (diff -purN) patch against the GIT head code of 4th October 2005. This replaces yesterday's patch which broke CONFIG_SMC91111_BASE. Apologies for any inconvenience caused. CHANGELOG * Patch 005 by Peter Pearse <peter.pearse(a)arm.com> 4th Oct 2005 Correct addessing errors in platform files. Split off common core module data from Integrator header files to include/armcoremodule.h. Valid patches are now: --------------------- 000 Patch submitted 30th June 2005 Import cc-option shell function from kernel Use it to get the correct ARM gcc options for individual cpus 002 Patch submitted 29th Sept 2005 Ensure Integrator boards call platformsetup() rather than lowlevel_init() 003 Patch submitted 29th Sept 2005 Provide Makefile targets for all ARM supported development boards 005 Patch submitted 4th Oct 2005 Split off common core module data Regards Peter Pearse

2 2

[U-Boot-Users] [PATCH] Makefile targets for ARM supported development boards.
by Peter Pearse 04 Oct '05

04 Oct '05

This gzipped patch shortens the configuration names of the Makefile targets for ARM supported developments boards and ensures all currently supported combinations of board and core module are supported. The patch was produced using cg-diff against branch based on commit fb34a9a20132216a47c9c91551cec3d7b4ebc0a3 of http://source.denx.net/git/u-boot.git CHANGELOG * Provide Makefile targets for all ARM supported development boards * Patch by Peter Pearse 29 Sept 2005 Regards Peter Pearse

2 1

[U-Boot-Users] [PATCH] Correction to previous patch of 26th Sept which broke other boards
by Peter Pearse 04 Oct '05

04 Oct '05

Correction to previous patch of 26th Sept which broke other boards - apologies to anyone affected. I was so embarassed I forgot to attach the patch - here it is. This patch replaces the arm920t.patch submitted on 26th Sept. The patch was produced using cg-diff against branch based on commit fb34a9a20132216a47c9c91551cec3d7b4ebc0a3 of http://source.denx.net/git/u-boot.git CHANGELOG * Ensure Integrator boards call platformsetup() rather than lowlevel_init() * Patch by Peter Pearse 29 Sept 2005 Regards Peter Pearse

2 2

[U-Boot-Users] [PATCH] Differing gcc version ARM options
by Peter Pearse 04 Oct '05

04 Oct '05

Hi Attached a bzipped patch file to fix the problems with different ARM options in different gcc versions. Patched against CVS head 30th June 2005. Tested with MAKEALL arm. CHANGELOG: * Patch by Peter Pearse, 30th June 2005 - Import cc-option shell function from kernel - Use it to get the correct ARM gcc options for individual cpus Replaces my earlier "[PATCH] Differing gcc version ARM options". ------------------------------------------------------------------------- Peter Pearse ARM Ltd

2 1

[U-Boot-Users] [Patch] CFI flash for ARM supplied & supported development boards
by Peter Pearse 04 Oct '05

04 Oct '05

The attached patch has been tested for building with MAKEALL arm. Patched code has been tested to boot linux & save environment with - Integrator/CP CM926EJ-S - Versatile PB926EJ-S - Versatile AB926EJ-S. It provides CFI support for all ARM supplied ARM supported development boards i.e. http://www.arm.com/products/DevTools/VPB926EJ-S.html http://www.arm.com/products/DevTools/Applicationbaseboard.html http://www.arm.com/products/DevTools/IntegratorCP.html This protects against possible future flash component changes. Replaces the patch submitted on 2005-03-15. Patched against CVS head of 12 Apr 2005. > CHANGELOG: > * Patch by Peter Pearse, 12 Apr 2005 > Changes all ARM supplied development boards, supported by ARM, to use CFI flash. > Adds CFI support for the Versatile/AB926EJ-S. Please report any errors and/or omissions to ------------------------ Peter.Pearse(a)arm.com -------------------------

3 2

[U-Boot-Users] [Patch] Multicore patch for ARM supplied development boards.
by Peter Pearse 04 Oct '05

04 Oct '05

Hi Attached corrected multicore patch - somehow I lost the fact that the omap boards not yet migrated to using lowlevel_init. Please inform me of any defects found. CHANGELOG: * Patch by Peter Pearse, 12 Apr 2005 - add core module targets for integratorap, integratorcp - add ARM946E cpu support - remap SSRAM memory to 0x00000000 Notes ----- a) Provides targets for all supported Core Modules - targets are of the form integrator<type>_CM<id>_config Examples make integratorap_CM946E_S make integratorcp_CM920T_config b) Previous targets integegratorcp, integratorap retained. c) Tested to build with MAKEALL arm d) Tested to boot linux with - Integrator/CP CM920T - Integrator/CP CM926EJ-S - Versatile PB926EJ-S - both Integrator combinations boot in all three settings - Manually run U-Boot in ARM Boot Monitor - Auto boot U-Boot through ARM Boot Monitor - Auto boot direct into U-Boot e) Patched against head at 12 Apr 2005 f) Replaces the patch submitted on 11 Apr 2005 ------------------------ Peter.Pearse(a)arm.com -------------------------

3 2

[U-Boot-Users] [Patch]
by Peter Pearse 04 Oct '05

04 Oct '05

Hi Attached remade multicore patch to allow users to specify the core module fitted to their ARM supplied Integrator development board. Please inform me of any defects found. CHANGELOG: * Patch by Peter Pearse, 11 Apr 2005 - add core module targets for integratorap, integratorcp - add ARM946E cpu support - remap SSRAM memory to 0x00000000 Notes ----- a) Provides targets for all supported Core Modules - targets are of the form integrator<type>_CM<id>_config Examples make integratorap_CM946E_S make integratorcp_CM920T_config b) Previous targets integegratorcp, integratorap retained. c) Tested to build with MAKEALL arm d) Tested to boot linux with - Integrator/CP CM920T - Integrator/CP CM926EJ-S - Versatile PB926EJ-S - both Integrator combinations boot in all three settings - Manually run U-Boot in ARM Boot Monitor - Auto boot U-Boot through ARM Boot Monitor - Auto boot direct into U-Boot e) Patched against head at 11 Apr 2005 f) Replaces the patch submitted on 02 Feb 2005 ------------------------ Peter.Pearse(a)arm.com -------------------------

3 2

[U-Boot-Users] Autobooting limitation on .3 ? : kernel 2.4 detect mac address unset after rebooting
by Mathieu Deschamps 04 Oct '05

04 Oct '05

It faults, I guess, because of bootd behavior. Explanation: (I've set bootcmd with my bootm command, and everything that's fine in my environnement (serverip, ethaddr, ..) and had a saveenv. ) * When I've finished updating my rootfs via tftpboot, I do a boot which does what bootcmd variable stores, that's to say, a bootm on a flash adr. Kernel is runing all right, see : . . eth0: Link now 100-FullDuplex eth0: AT91 ethernet at 0xfefbc000 int=24 100-FullDuplex (12:34:56:78:99:AA) eth0: Davicom 9196 PHY (Copper) . Then I can execute this command, all right : # ifconfig eth0 192.168.0.5 up && echo Ok || echo Ko Ok # * After this, on rebooting my board via a reboot command (or even with a pushbutton reset) to test autoboot functionnality on reboot, kernel starts but it seems it can't retrieve my mac addr. : . . eth0: Link now 100-FullDuplex eth0: AT91 ethernet at 0xfefbc000 int=24 100-FullDuplex (00:00:00:00:00:00) eth0: Davicom 9196 PHY (Copper) . So naturally I can't execute my ifconfig : # ifconfig eth0 192.168.0.5 up && echo Ok || echo Ko SIOCSIFFLAGS: Cannot assign requested address SIOCSIFFLAGS: Cannot assign requested address Ko # This is always reproductible, but never happens after my rootfs update. I've found a temp. solution : run a ping command on serverip or simply adding it to bootcmd like this : bootcmd=ping $(serverip);bootm 0x10020000 It seems "net commands" update environnement or something like that. If no net commands is executed before the boot/bootd command, then it could be a reason why kernel is fed with a left blank, wrong mac addr. value. Any ideas on more reliable way-to or better way to make it work ? Thanks in advance Mathieu Deschamps.

2 1

[U-Boot-Users] Replacing AT91 u-boot
by kentropy 04 Oct '05

04 Oct '05

Hello, I'm sorry but I am still facing to problems trying to replace the existing u-boot-0.3.2 with the latest 1.1.3 on a at91rm9200dk board. I built u.boot make at91rm9200dk_config make then I reomved R159 from the board and booted, then I sent loader.bin (found on Atmel website) then I send the just builded u-boot.bin but at this point the AT91 system freezes showing few random characters on the serial console. If I use the original u-boot.bin 0.3.2 everything is ok, what am I missing ? Thank you for your time ---- loader 1.0 (Aug 8 2003 - 12:01:07) XMODEM: Download U-BOOT C U-Boot downloaded successfully U-Boot 0.3.2 (Jun 24 2003 - 17:04:08) U-Boot code: 21F00000 -> 21F14160 BSS: -> 21F1FAEC DRAM Configuration: Bank #0: 20000000 32 MB Atmel: AT49BV1614A (16Mbit) Flash: 2 MB DataFlash:AT45DB642 Nb pages: 8192 Page Size: 1056 Size= 8650752 bytes Logical address: 0xC0000000 Uboot>

4 3

[U-Boot-Users] u-boot and gcc 4.0
by Dmytro Bablinyuk 04 Oct '05

04 Oct '05

I have noticed today that u-boot uses some techniques that were deprecated in 3.4.x and now removed in 4.0 (so I cannot just get source and compile it under 4.0) For example in u-boot/fs/jffs2/compr_rubin.c temp = *(++((u32 *) in)); compr_rubin.c:89: error: invalid lvalue in increment ... According to the gcc 4.0.0 changelog: --8<-- The cast-as-lvalue, conditional-expression-as-lvalue and compound-expression-as-lvalue extensions, which were deprecated in 3.3.4 and 3.4, have been removed. --8<-- Is there are any plans to support 4.0 compiler or shall I downgrade my compiler to 3.4.x version. I may try to fix such places but since I have another task such as get our board running I am quite afraid to introduce a new problems. Thank you Dimitry

2 1

[U-Boot-Users] how to support the lcd logo in u-boot?
by zhuzhenhua 04 Oct '05

04 Oct '05

if i want to support lcd logo, need i write a lcd.c in my board directory? is there some codes or structure can be reused?

2 3

Re: [U-Boot-Users] AT91 and u-boot tests
by kentropy 04 Oct '05

04 Oct '05

> > u-boot always configured and compiled "as is" for at91rm9200dk > > You are aware that you cannot use the same configuration for your > tests? I am doing the same, what is the difference between these command sequence to generate u-boot using 1.1.2 and 1.1.3 without doing any further setting? # make at91rm9200dk_config # make Should be the exactly the same > > u-boot-1.1.3 > > ------------ > > - running in ram : DOES NOT BOOT > > - stored in flash with boot.bin as pre-loader > > and then u-boot as bootloader : DOES NOT BOOT > > - stored in flash as unique bootloader : OK BUT FAILS KERNEL UNCOMPRESSION > > > Your message is not helpful as it is totally unclear if you change > the configuration (i. e. with or without CONFIG_SKIP_LOWLEVEL_INIT > and/or CONFIG_SKIP_RELOCATE_UBOOT #defined in your cofiguration which > is *essential*. Please make sure to read the documentation (README). A quick watch to lowlevel_init.S shows that something low level is done here too ;-) However CONFIG_SKIP_LOWLEVEL_INIT and/or CONFIG_SKIP_RELOCATE_UBOOT are not set as default for at91-dk, so why version 1.1.3 (and only this version) doesn't run in ram? P.S. Manual read ;-) http://www.denx.de/twiki/bin/view/DULG/CanUBootBeConfiguredSuchThatItCanBeS…

2 1

[U-Boot-Users] AT91 and u-boot tests
by Marco Cavallini 03 Oct '05

03 Oct '05

Reading the AT91 related thread I digged a bit and maybe I found something strange between u-boot-1.1.2 and 1.1.3 ? I usually use u-boot-1.1.2 using the "usual Atmel procedure"(TM) I removed T159 and booted, loading loader.bin in ram with xmodem, then it asked for the next file u-boot.bin that has been succesfully loaded. If I use u-boot-1.1.2 all is OK as with preinstalled 0.3.2 This is the result of my tests: Board AT91RM9200DK Kernel 2.6.13.2 u-boot always configured and compiled "as is" for at91rm9200dk u-boot-1.1.2 ------------ - running in ram : OK - stored in flash with boot.bin as pre-loader and then u-boot as bootloader : OK - stored in flash as unique bootloader : DOES NOT BOOT u-boot-1.1.3 ------------ - running in ram : DOES NOT BOOT - stored in flash with boot.bin as pre-loader and then u-boot as bootloader : DOES NOT BOOT - stored in flash as unique bootloader : OK BUT FAILS KERNEL UNCOMPRESSION P.S. Looking into boot.bin sources seems that it required for Master clock, PLL frequency and SDRAM very low level settings. Hoping this could be useful :-) Marco Cavallini ===================================================== Koan s.a.s. - Software Engineering Linux and WinCE solutions for Embedded and Real-Time Software Klinux : the embedded distribution for industrial applications - Atmel AT91 ARM Third Party Consultant - Intel PCA Developer Network Member - Microsoft Windows Embedded Partner Via Pascoli, 3 - 24121 Bergamo - ITALIA Tel. (++39) 035-255.235 - Fax (++39) 178-223.9748 http://www.koansoftware.com - http://www.klinux.org =====================================================

2 1

[U-Boot-Users] broken rsync to U-Boot git archive
by Bryan O'Donoghue 03 Oct '05

03 Oct '05

git clone rsync://source.denx.net/git/u-boot.git somedir appears to be broken, at least when I tried it just now, I got errors ! I had an archive locally which dereferenced the above rsync path. I did a cg-update inside of that directory tree and got a similar error. I rm rf'd the archive assuming I'd made some noobish git/cogito error and tried the git clone command, as with the cg-update command which wanted to talk to the rsync above, I got errors like the below. timeout in data send/receive connection unexpectedly closed etc. rsync --version == 2.6.5 git core == 0.99.4 cogito == 0.13 OS == Ubuntu Breezy Badger Preview Release http cloning and presumably updating, appears to be working. Either rsync is broken or I've made an error in synchronising to it. Possibly I've hosed rsync somehow, as I've compiled git from source using an earlier version of Ubuntu "Hoary Hedgehog" and have not as yet, recompiled against my current OS libraries. Possibly that _is_ in fact the case, but, it would probably be worth the while of others to confirm/deny this. Best, Bryan -- "Beware of bugs in the above code; I have only proved it correct, not tried it." -- Donald Knuth

1 0

[U-Boot-Users] lastest 1.1.3 CONFIG_SKIP problems with arm920t on rm9200.
by Mathieu Deschamps 03 Oct '05

03 Oct '05

Hello, I have upgraded U-Boot 1.1.2 to lastest 1.1.3. I still uses the boot.bin program to set up the flash and the dram and such while having defined SKIP_RELOCATE_UBOOT and SKIP_LOWLEVEL_INIT. Indeed kept undefined I have seen that it breaks everything reaching SoC specific lowlevel_init.S in cpu/.../start.S. Imagine reinitializing flash while reading init values from it, or reinit the memory controller while using it ! It seems that these two were compulsory in my case. But even defined, my cpu goes in ABORT. Any ideas ? Regards, Mathieu Deschamps

3 4

[U-Boot-Users] [PATCH] ARM supplied & supported development boards. Errata & rationalization.
by Peter Pearse 03 Oct '05

03 Oct '05

Attached (diff -purN) patch against the GIT head code corrects some errors in the existing code and adds a new file, include/armcoremodule.h which contains the common Core Module data used by both Integrator/AP & Integrator/CP boards. This patch should be applied after my previously submitted patches. For future reference my patches are now numbered, numbers for the previously submitted patches assigned below. CHANGELOG * Patch 004 by Peter Pearse <peter.pearse(a)arm.com> Correct addessing errors in platform files. Split off common core module data from Integrator header files to include/armcoremodule.h. Previous patches have now been assigned numbers as follows 000 Patch submitted 30th June 2005 Import cc-option shell function from kernel Use it to get the correct ARM gcc options for individual cpus 001 Patch submitted 26th Sept 2005 **INCORRECT** withdrawn Ensure only Integrator boards call platformsetup 002 Patch submitted 29th Sept 2005 Ensure Integrator boards call platformsetup() rather than lowlevel_init() 003 Patch submitted 29th Sept 2005 Provide Makefile targets for all ARM supported development boards Regards Peter Pearse

1 0

Re: [U-Boot-Users] Replacing AT91 u-boot
by kentropy 03 Oct '05

03 Oct '05

> AFAIK loader.bin is used to download boot.bin to RAM location. Correct > boot.bin itself is an u-boot binary image. > It is straight executed after > downloading. Now it's possible to install a new u-boot version. Definitely not. BTW binary sizes are too different to say that boot.bin is 10KB loader.bin is 7KB u-boot is 92KB Referring to Atmel manuals boot.bin is a pre-loader ususlly stored at the beginning of flash (0x10000000) which calls u-boot (usually stored at 0x10010000). You can found sources for loader.bin and boot.bin on Atmel CDROM and until last saturday on this website. http://www.at91.com/Pages/products/EvaluationBoard/AT91RM3400DK/at91rm3400d… Today these sources have been removed (I don't know why ?)

2 1

Re: [U-Boot-Users] Replacing AT91 u-boot
by kentropy 03 Oct '05

03 Oct '05

> > May be into boot.bin there were some setting necessary for the system ? > > Nargh. Please forget boot.bin. You will not need it any more. > > > AFAIK nobody (either Atmel) uses u-boot stored at the beginning of the flash > > (as many others ARM board producers do), but AT91 uses a pre-bootloader. > > I don't know where you got this "knowledge" from, but it is plain wrong. Confusing news from AT91 website about boot.bin / loader.bin [ You need to download a "boot.bin" into the first sector. The boot.bin may need to know about which flash you are using. loader.bin has the same problem. .snip. Ulf Samuelsson ]

2 1

Re: [U-Boot-Users] Replacing AT91 u-boot
by kentropy 03 Oct '05

03 Oct '05

>You should > attach a debugger and actually try to *devbug* this > insetad of guessing (and blaming U-Boot). Ok, I won't guess anymore ;-) But I don't know how to debug it, I have no debuggers. > One of the things that did change is the mach-types.h file; > if your > kernel uses some invalid ID this could explain such a hang. > But this > is wild guessing - please debug the problem. I verified that entries MACH_TYPE_AT91RM9200 and MACH_TYPE_AT91RM9200DK are the same you found in kernel 2.6.13 and tried replacing MACH_TYPE_AT91RM9200DK MACHINE_START(MACH_TYPE_AT91RM9200DK with MACHINE_START(MACH_TYPE_AT91RM9200 in board-dk.c without success.

2 1

Re: [U-Boot-Users] Replacing AT91 u-boot
by kentropy 03 Oct '05

03 Oct '05

Thank you again for your patience answering. > The messages you showed make it clear that U-Boot *is* running fine > on your board. The thing I can't understand due to my inexperience is why using u-boot-0.3.2 with my kernel 2.6.13.2 everything worked and simply changing u-boot-1.1.3 not. So I suppose that the kernel is not the problem, but only the changed things: loader and u-boot. Trying to guess: can the problem be caused by a different kind of SDRAM or Flash between my DK and your compatible one ? TIA

3 2

[U-Boot-Users] why I don't see my msgs?
by kentropy 02 Oct '05

02 Oct '05

Hi, I wonder why I can't get back my messages from Mailing List since last days of Sept. Is it a server problem ? Cheers :-)

2 1

Re: [U-Boot-Users] Replacing AT91 u-boot
by kentropy 02 Oct '05

02 Oct '05

> Sorry, I missed this. > Remove the initrd= and mem= part. U-Boot passes correct ATAGS > to the kernel. Thank you again. I followed your suggestions: U-Boot> printenv bootdelay=3 baudrate=115200 ethaddr=00:02:a5:6e:3f:fd filesize=3638c0 fileaddr=21140000 netmask=255.255.255.0 ipaddr=192.168.0.184 serverip=192.168.0.11 bootargs=root=/dev/ram rw ramdisk_size=15360 console=ttyS0,115200 stdin=serial stdout=serial stderr=serial U-Boot> tftp 21100000 ramdisk TFTP from server 192.168.0.11; our IP address is 192.168.0.184 Filename 'ramdisk'. Load address: 0x21100000 Loading: ### .snip. ######## done Bytes transferred = 3553472 (3638c0 hex) U-Boot> tftp 21000000 uImage TFTP from server 192.168.0.11; our IP address is 192.168.0.184 Filename 'uImage'. Load address: 0x21000000 Loading: ############################################## ####################################### done Bytes transferred = 646841 (9deb9 hex) U-Boot> bootm ## Booting image at 21000000 ... Image Name: Image Type: ARM Linux Kernel Image (gzip compressed) Data Size: 646777 Bytes = 631.6 kB Load Address: 20008000 Entry Point: 20008000 Verifying Checksum ... OK Uncompressing Kernel Image ... [here is still freezed] Every test I do I'm more confused. Please apologize my question, but did you ever used an AT91RM9200DK board with u-boot (reading help and your site seems that you are a PPC expert but there is no examples using AT91) ? Why Atmel give out and say to use boot.bin as pre-bootloader and you say not ? Why the u-boot-1.1.3 default configuration for this board don't work ? Maybe there is something missing or wrong ? TIA

2 1

[U-Boot-Users] Testing rewrite of NAND code?
by Wolfgang Denk 02 Oct '05

02 Oct '05

Maybe this slipped through unnoticed, so here again: A rewrite of the NAND flash code by Ladislav Michl is available in the " "testing-NAND" branch of our git repository; you can download it using git or cogito like this: git clone rsync://source.denx.net/git/u-boot.git\#testing-NAND u-boot-testing-NAND resp. cg-clone rsync://source.denx.net/git/u-boot.git\#testing-NAND u-boot-testing-NAND Please help testing this code as thoroughly as possible. I would like to merge this into the main branch ASAP. Please send your Comments, complaints, bug reports etc. to this mailing list + Ladislav Michl + me. Thanks. Best regards, Wolfgang Denk -- Software Engineering: Embedded and Realtime Systems, Embedded Linux Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd(a)denx.de Hello! I'm from outer space, and I've made myself look like a signa- ture. While you are reading this, I'm having sex with your eyes. I know it feels good to you, because you're smiling. I'm very horny, so send me to someone else when you've had enough. Thanks! Sincerely, A Stranger in a Strange Land

4 6