u-boot and bdi2000
Is there quick hint why u-boot does not start up when bdi2000 is connected to mpc5200 board.
problem is simple. I load my flash programming code ( rload ) to mpc5200 internal sram
with bdi2000 and program u-boot to flash. So, loading and running code works.
Then when i reset system and try start u-boot it hangs in initialization code but
if i remove bdi2000 it boots u-boot without problems.
The bdi2000 initializtion file is based one from denx ftp server. I just ask, if this
is commonly known problem before spending any bit more time for it.
Kate
Quoted 3 lines · In message <[email protected]> you wrote:
Is there quick hint why u-boot does not start up when bdi2000 is connected to mpc5200 board.
Because you did not enter a "go" command, or used a bad address with the "go" ?
Quoted 3 lines
problem is simple. I load my flash programming code ( rload ) to mpc5200 internal sram
with bdi2000 and program u-boot to flash. So, loading and running code
Why that? Why don't you just use the BDI2000 to load and program the image to flash?
Quoted 3 lines
Then when i reset system and try start u-boot it hangs in initialization code but
if i remove bdi2000 it boots u-boot without problems.
What is PC pointing to after reset, and what is your correct entry point?
Quoted 3 lines
The bdi2000 initializtion file is based one from denx ftp server. I just ask, if this
is commonly known problem before spending any bit more time for it.
I guess that reading the docs would help. See for example "doc/README.IceCube"
Best regards,
Wolfgang Denk
Signature · 4 lines
Software Engineering: Embedded and Realtime Systems, Embedded Linux Phone: (+49)-8142-4596-87 Fax: (+49)-8142-4596-88 Email: wd at denx.de In accord with UNIX philosophy, Perl gives you enough rope to hang yourself. - L. Wall & R. L. Schwartz, _Programming Perl_
Quoted 13 lines · Wolfgang Denk
Quoted 4 lines · In message <[email protected]> you wrote:
Is there quick hint why u-boot does not start up when bdi2000 is connected to mpc5200 board.
Because you did not enter a "go" command, or used a bad address with the "go" ?
Of cource i entered go command and and user right address
Quoted 10 lines
Quoted 5 lines
problem is simple. I load my flash programming code ( rload ) to mpc5200 internal sram
with bdi2000 and program u-boot to flash. So, loading and running codeWhy that? Why don't you just use the BDI2000 to load and program the image to flash?
Few reasons, at first, the board is not Icecube, it is my own and it has diferent flash etc.
It is easiest to do initial testing to load code to internal sram . With this code i can
test ethernet, serial, flash and sdram. And it is fast to program flash directly over ethernet.
Quoted 11 lines
Quoted 5 lines
Then when i reset system and try start u-boot it hangs in initialization code but
if i remove bdi2000 it boots u-boot without problems.What is PC pointing to after reset, and what is your correct entry point?
It is pointing reset vector and there is right reset code in this address. I traced execution
with bdi2000 ti command and it executes something 10 instructions ultill it hangs
allways in same place.
- Target MBAR is 0xF0000000
- Target XLBA is 0x80000006
BDI> Target CPU : MPC8280/MGT5200 (Zeppo)
Target state : debug mode
Debug entry cause : trace
Current PC : 0xfff00130
Current CR : 0x00000000
Current MSR : 0x00001002
Current LR : 0x00000000BDI>ti
- Target MBAR is 0xF0000000
- Target XLBA is 0x00000000
BDI> Target CPU : MPC8280/MGT5200 (Zeppo)
Target state : debug mode
Debug entry cause : COP freeze (startup)
Current PC : 0xfff00134
Current CR : 0x00000000
Current MSR : 0x00001002
Current LR : 0x00000000# Step timeout detected
BDI>
Or, if i just run from reset:
- TARGET: processing target startup passed
BDI>info
Target CPU : MPC8280/MGT5200 (Zeppo)
Target state : debug mode
Debug entry cause : COP freeze (startup)
Current PC : 0xfff00100
Current CR : 0x00000000
Current MSR : 0x00001002
Current LR : 0x00000000BDI>go
BDI>info
Target CPU : MPC8280/MGT5200 (Zeppo)
Target state : running
BDI>halt
- Target MBAR is 0xF0000000
- Target XLBA is 0x00000000
BDI> Target CPU : MPC8280/MGT5200 (Zeppo)
Target state : debug mode
Debug entry cause : COP freeze (startup)
Current PC : 0xfff03178
Current CR : 0x00000000
Current MSR : 0x00001002
Current LR : 0xfff00138BDI>ti
- Target MBAR is 0xF0000000
- Target XLBA is 0x00000000
BDI> Target CPU : MPC8280/MGT5200 (Zeppo)
Target state : debug mode
Debug entry cause : COP freeze (startup)
Current PC : 0xfff03178
Current CR : 0x00000000
Current MSR : 0x00001002
Current LR : 0xfff00138# Step timeout detected
BDI>
In either case nothing appears in to the console. Also, allways the address is same
and i got the step timeout.
Quoted 10 lines
Quoted 5 lines
The bdi2000 initializtion file is based one from denx ftp server. I just ask, if this
is commonly known problem before spending any bit more time for it.I guess that reading the docs would help. See for example "doc/README.IceCube"
I did not find anything about this issue. The u-boot works, it has right addresses,
vectors etc. It just things that i already know. The problem is to get it working
with bdi2000. There is something that i did wrong with bdi2000 configuration
and i try figure out what.
Kate
Quoted 4 lines · Kate Alhola
........The problem is to get it working
with bdi2000. There is something that i did wrong with bdi2000 configuration
and i try figure out what.
Check to make sure the start up code isn't writing over some of the processor debug registers. Debuggers get upset when they set a particular mode and then you come behind them in software and mess it up :-) I've seen similar things happen with other processors, and this has been the cause. It is proper and necessary to have a boot rom initialize these registers for normal operation, but not when you are trying to use a debugger. If you take out such initialization for debugging, don't forget to put it back when you ship the system! :-)
-- Dan
Quoted 5 lines · In message <[email protected]> you wrote:
Quoted 4 lines
Because you did not enter a "go" command, or used a bad address with the "go" ?
Of cource i entered go command and and user right address
This is not obvious. You didn;t mention it in your report, and it didn't work, so there was a chance ther eis an error here.
Quoted 5 lines
Quoted 3 lines
Why that? Why don't you just use the BDI2000 to load and program the image to flash?
Few reasons, at first, the board is not Icecube, it is my own and it has diferent flash etc.
This is not a reason at all. Of course the BDi2000 can program your flash as well.
Quoted 1 line
It is easiest to do initial testing to load code to internal sram . With
This is something between a matter of taste and an urban legend.
Quoted 2 lines
test ethernet, serial, flash and sdram. And it is fast to program flash directly over ethernet.
It's also fast to program flash with the BDI2000.
Quoted 3 lines
In either case nothing appears in to the console. Also, allways the address is same
and i got the step timeout.
Did you contact Abatron support? What did they say (except for: check for hardware problems - which I'd say, too).
Quoted 3 lines
with bdi2000. There is something that i did wrong with bdi2000 configuration
and i try figure out what.
Which differences are between your board and the LITE5200?
Best regards,
Wolfgang Denk
Signature · 3 lines
Software Engineering: Embedded and Realtime Systems, Embedded Linux Phone: (+49)-8142-4596-87 Fax: (+49)-8142-4596-88 Email: wd at denx.de Any technology distinguishable from magic is insufficiently advanced.
Quoted 10 lines · Wolfgang Denk
Quoted 3 lines · In message <[email protected]> you wrote:
It is easiest to do initial testing to load code to internal sram . With
This is something between a matter of taste and an urban legend.
I don't think so. If you have brand new board to test, you can't be sure that everything is working
at all. Things inside of cpu are most propably ones that work. That means internal SRAM, UART
and with little external support ethernet.
Quoted 11 lines
Quoted 5 lines
In either case nothing appears in to the console. Also, allways the address is same
and i got the step timeout.Did you contact Abatron support? What did they say (except for: check for hardware problems - which I'd say, too).
The reason why i don't think that it is not hw problem in abatron or board but rather
some incompatibility with bdi2000 initialization and u-boot. The abatron works
without problemt to load my own code to internal sram, i can start and run it etc.
I was expecting something same than Dan Malek said that there is something with debug register write etc.
For some reason the both cases, trace and starting with go fails in same place ( before or after bl )
fff00118 <boot_cold>:
fff00118: 7c a0 00 a6 mfmsr r5 fff0011c: 3c 60 f0 00 lis r3,-4096 fff00120: 60 63 00 00 ori r3,r3,0 fff00124: 7c 77 4b a6 mtspr 311,r3 fff00128: 54 63 84 3e rlwinm r3,r3,16,16,31 fff0012c: 3c 80 80 00 lis r4,-32768 fff00130: 90 64 00 00 stw r3,0(r4) fff00134: 48 00 30 45 bl fff03178 <init_5xxx_core>
Also the board works, u-boot works, the board boots and runs linux etc. The only
thing that don't work is u-boot with bdi2000 connected.
Quoted 9 lines
Quoted 5 lines
with bdi2000. There is something that i did wrong with bdi2000 configuration
and i try figure out what.Which differences are between your board and the LITE5200?
There is lot ot abot nothing diferencies. In basic boot level it is wery compaible with LITE5200.
Only meaningfull diference is that boot flash is diferent. I just needed to change u-boot
to support intel-compatible flash algorithm and change flash size to get it working.
The most diferencies goes in expansions. My board has NAND flash for filesystem, MMC for
filesystem, TI2301 audio codec/touch screen/system ADC, two banks of SDRAM and
FPGA for LCD display controller and i-o expansion use. The board is
http://www.iti.fi/iti5200.php
I was just working to make u-boot to boot kernel from nand flash or mmc.
Kate