January 2007 - U-Boot

[U-Boot-Users] ROTALIGN Alpha support
by Krzysztof Helt 30 Jan '07

30 Jan '07

Hello, I ported the U-boot to Pruftechnik ROTALIGN Alpha board (PXA based). I added the patch to patch tracker. Should I repost the patch to the list as well? Regards, Krzysztof ---------------------------------------------------- Tajemnica - i to niewątpliwa! Poszukiwaczom przygód polecamy "Dom luster" Ulyssesa Mora. Kim jest Tajemniczy Ulysses? Zobacz: http://klik.wp.pl/?adr=http%3A%2F%2Fadv.reklama.wp.pl%2Fas%2Fd62.html&sid=1…

2 1

[U-Boot-Users] [PATCH] mpc83xx: don't hang if watchdog configured on 8360, 832x
by Kim Phillips 30 Jan '07

30 Jan '07

don't hang if watchdog configured on 8360, 832x The watchdog programming model is the same across all 83xx devices; make the code reflect that. --- cpu/mpc83xx/cpu.c | 4 ---- 1 files changed, 0 insertions(+), 4 deletions(-) diff --git a/cpu/mpc83xx/cpu.c b/cpu/mpc83xx/cpu.c index f1b2749..817f422 100644 --- a/cpu/mpc83xx/cpu.c +++ b/cpu/mpc83xx/cpu.c @@ -266,7 +266,6 @@ unsigned long get_tbclk(void) #if defined(CONFIG_WATCHDOG) void watchdog_reset (void) { -#ifdef CONFIG_MPC834X int re_enable = disable_interrupts(); /* Reset the 83xx watchdog */ @@ -276,9 +275,6 @@ void watchdog_reset (void) if (re_enable) enable_interrupts (); -#else - hang(); -#endif } #endif -- 1.4.4

1 0

[U-Boot-Users] mpc83xx: make 8360 default environment fdt be 8360 (not 8349)
by Kim Phillips 30 Jan '07

30 Jan '07

make 8360 default environment fdt be 8360 (not 8349) --- include/configs/MPC8360EMDS.h | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/include/configs/MPC8360EMDS.h b/include/configs/MPC8360EMDS.h index 8ad6551..fd61ea5 100644 --- a/include/configs/MPC8360EMDS.h +++ b/include/configs/MPC8360EMDS.h @@ -609,7 +609,7 @@ "ramdiskaddr=1000000\0" \ "ramdiskfile=ramfs.83xx\0" \ "fdtaddr=400000\0" \ - "fdtfile=mpc8349emds.dtb\0" \ + "fdtfile=mpc8360emds.dtb\0" \ "" #define CONFIG_NFSBOOTCOMMAND \ -- 1.4.4

1 0

[U-Boot-Users] [PATCH] Fix alternating tx error / tx buffer not ready bug in QE UEC
by Kim Phillips 30 Jan '07

30 Jan '07

From: Emilian Medve <Emilian.Medve(a)freescale.com> Fix alternating tx error / tx buffer not ready bug in QE UEC The problem is not gcc4 but the code itself. The BD_STATUS() macro can't be used for busy-waiting since it strips the 'volatile' property from the bd variable. gcc3 was working by pure luck. This is a follow on patch to "Fix the UEC driver bug of QE" (commit daa593ca78c63aadf77aa35c3065d42c535e5ddf). --- drivers/qe/uec.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/qe/uec.c b/drivers/qe/uec.c index aea455b..c416a67 100644 --- a/drivers/qe/uec.c +++ b/drivers/qe/uec.c @@ -1148,7 +1148,7 @@ static int uec_send(struct eth_device* d /* Init TxBD */ BD_DATA_SET(bd, buf); BD_LENGTH_SET(bd, len); - status = BD_STATUS(bd); + status = bd->status; status &= BD_WRAP; status |= (TxBD_READY | TxBD_LAST); BD_STATUS_SET(bd, status); -- 1.4.4

1 0

[U-Boot-Users] nand write.jffs2 without clean markers
by Markus Pietrek 30 Jan '07

30 Jan '07

Hi, newer U-Boot versions (1.1.4ff) have "nand erase clean remove". But linux expects to have clean markers in OOB when mounting JFFS2 partitions. So, is it possible to update a JFFS2 partition by U-Boot correctly? Bye, Markus Pietrek

1 0

Re: [U-Boot-Users] u-boot on blackfin problems loading linux kernel
by matti h 30 Jan '07

30 Jan '07

Aubrey, I think I accidently replied to ur last posting, could u please forward it to the list ? /cheers Matty >From: "Aubrey Li" <aubrey.adi(a)gmail.com> >To: "matti h" <sinking_wasa(a)hotmail.com> >Subject: Re: [U-Boot-Users] u-boot on blackfin problems loading linux >kernel >Date: Tue, 30 Jan 2007 18:03:06 +0800 > >On 1/30/07, matti h <sinking_wasa(a)hotmail.com> wrote: >>Ok, I shall try the other forum from now on.. >>Anyhow, the problem that I experineced with your version was that >>typing a single character at the prompt resulted in multiple characters... >>so typing commands was somewhat impossible. >>I got the feeling that some sort of timing is corrupt.. > >That shouldn't happen. We have regression test running all day and >night, we didn't see any problems. It may be caused by your hardware >or the silicon version of the blackfin chip on your board is old. > >>So u have modified the function for load_elf_image ? >> >Yes, Please compare the one in our svn server with the other in the >upstream repo. > >-Aubrey _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/

1 0

[U-Boot-Users] u-boot on AT91SAM9260
by Chris Sharman 30 Jan '07

30 Jan '07

Hello, Does u-boot support the nandflash as found on the at91sam9260ek board? Also, does u-boot automatically detect the MAC address on the at9260ek board? I have been using the u-boot that came from Atmel as part of a linux demo. This one has an example booting from nandflash, which works fine. I owuldn't be changing from this except the MAC address in linux is 00:00:00:00:00:00. There is also u-boot in Ulf's buildroot with patches. But this one doesn't seem to detect the nandflash. "NAND bus width 16 instead 8 bit" Where do i change the bus width? Any pointers would be great. Regards Chris

1 0

Re: [U-Boot-Users] u-boot on blackfin problems loading linux kernel
by Aubrey Li 30 Jan '07

30 Jan '07

On 1/30/07, matti h <sinking_wasa(a)hotmail.com> wrote: > Hello Aubrey, > Well I did actually tryout the one from u r svn. > but I had no luck with it either... > I dont have an tftpserver here at the moment, but I'll try to load kernel > over tftp aswell to see if it makes any diffrence (or run the memcmp as u > suggest). > Anyhow, I tried to load linux to 0x800000 and added a breakpoint to > do_bootelf > in vdsp, the typed > bootelf 0x800000 and started single stepping... > and in load_elf_image I get some exception with excause 0x2A, could it be > that my > memory mappings are wrong somehow ? Im sort of new to these linker scripts > and I get > the feeling something might be missconfigured in there...? > Or could there be some clash because Im loading u-boot using vdsp to ram ? > We have pieces of coding hack on load_elf_image for special L1 section handling. You may need to port it to your own package. Cmdline editing and history have been enabled on u-boot-1.1.6 in our svn server. What's your mean by "console works very poorly with their source"? BTW, posting your problem on our forum is the best way to get fastest response. -Aubrey

1 0

[U-Boot-Users] CSB637 configuration maps code wrong?
by Patrick Fallberg 29 Jan '07

29 Jan '07

Hello! I am trying to get u-boot to work on a Cogent CSB637, but it seem that the default configuration maps the code to the wrong place. In config.mk, TEXT_BASE is configured to 0x23fc0000 which seem to make the ELF file locate the code to that location instead of FLASH (0x10000000). I use a Lauterbach to flash the board using JTAG and if I write the binary file to 0x10000000 it seems to start executing OK, but when it comes to the relocation part, it looks like it cannot map the memory correctly. When it attempts to transfer to SDRAM it copies to addr. 0x0, but it has not remapped the memory so 0x0 still points at FLASH, resulting in a definite stop. Is there any additional configuration necessary to set up the address-ranges apart from a "make csb637_config" followed by a "make all"? I have tried this on u-boot-1.1.6. Best regards, Patrick Fallberg

1 0

[U-Boot-Users] u-boot on blackfin problems loading linux kernel
by matti h 29 Jan '07

29 Jan '07

Hi, Ive downloaded u-boot 1.1.6 sources and modified them to the extend that I can figure out to fit our custom blackfin board. Im able to start u-boot by loading it to target using visualdsp++ and in the u-boot prompt I can load hello world example from a connected cf disk. But the linux kernel does not run... Im sure the kernel is ok, aswell as the hardware (tested with other bootloader and works sound). Now how do I debug this ?, is it possible that I have missed out some needed change in the source ? I have access to a jtag device to trace...but im not sure how I should use it to find out the problem. The hardware features the following bf533 cpu @ 400 Mhz core 100 mhz ram . 16mb ram 2mb flash Not sure about the following option TEXT_BASE = 0x00600000 I grabbed it from the older bootloader...how do I tell it its "correct" ? I have followed this guide for porting the source: http://docs.blackfin.uclinux.org/doku.php?id=customizing_u-boot_for_your_ow… However I have not modified cpu/bf533/cpu.c , should I ? and what should I do in there? I should note that I also tried the u-boot source from blackfin.uclinux.org but I had even less luck running it, console works very poorly with their source (double characters apearing etc, ide not working ). //Any help is appriciated - Matty And this is how it all looks: boot>ide res Reset IDE: Bus 0: OK Device 0: Model: STI Flash 7.2.0 Firm: 12/15/04 Ser#: STI1J80505285182323 Type: Hard Disk Capacity: 122.5 MB = 0.1 GB (250880 x 512) boot>fatload ide 0:1 0x00800000 a/helloworld reading a/helloworld 7813 bytes read boot>bootelf 0x00800000 Loading .text @ 0x00001000 (312 bytes) Loading .rodata @ 0x00001138 (151 bytes) ## Starting application at 0x00001000 ... Example expects ABI version 3 Actual U-Boot ABI version 3 Hello World argc = 1 argv[0] = "0x00800000" argv[1] = "<NULL>" Hit any key to exit ... ## Application terminated, rc = 0x0 boot>fatload ide 0:1 0x00800000 b/linux reading b/linux 1576768 bytes read boot>bootelf 0x00800000 Loading .text @ 0x00001000 (1076368 bytes) Exception! Exception! _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/

2 1