[U-Boot-Users] Board hangs during relocate_code..
Hi All,
During porting of U-boot on mpc8280 (U-boot port of MPC8260ads as base) I am facing following problems. I was able to burn the u-boot onto the flash. The first part of intialization was fine ie board_init_f () including the SDRAM (256 MB) initialization. The board gets hung during the second part of initialization. But the same piece of code works fine when I run from RAM using the JTAG. Is there any specific way to solve this problem? I have configured the FLASH_BASE as 0xFE000000; Flash Chip is 2MB On Board Flash; TEXT_BASE is FFF00000; IMMR is F0000000. Do I miss any specific things in my configuration?
One more thing is, the board I am using has two flashes.
The 2MB On Board Flash: "AMD 29F160xB (2048 x 8) 1 Device" AND The 64MB SIMM Flash: "INTEL 28F128Jx (16384 x 8) 4 Devices"
I checked almost all the flash driver modules that is there with U-boot version of 1.0.0, but I was not able to get the driver for these flashes. Where can I get the same?
regards, Sudhakar.
__________________________________ Do you Yahoo!? Yahoo! Hotjobs: Enter the "Signing Bonus" Sweepstakes http://hotjobs.sweepstakes.yahoo.com/signingbonus
In message 20040116144155.27139.qmail@web10605.mail.yahoo.com you wrote:
During porting of U-boot on mpc8280 (U-boot port of MPC8260ads as base) I am facing following problems. I was able to burn the u-boot onto the flash. The first part of intialization was fine ie board_init_f () including the SDRAM (256 MB) initialization. The board gets hung during the second part of initialization. But the same piece of code works fine when I run from RAM using the JTAG. Is there any specific way to solve this problem?
Aaaaaaarrrrrrrgggghhhhh!!!!
Is nobody readuing mailing ist archives or FAQ's any more?
Obviously your JTAG debugger performs the correct SDRAM initiali- zation sequence which is missing in your code. And no, this has NOTHING to do with the UPM tables.
Wolfgang Denk
Aaaaaaarrrrrrrgggghhhhh!!!!
Is nobody readuing mailing ist archives or FAQ's any more?
Pardon my ignorance, but can you point me to the mailing list archives or the FAQ?
I see you flaming people on this list quite frequently for asking questions that have been asked before; I think it's fair to expect people to do their homework before asking a question, however seing as
- u-boot.sourceforge.net is just a blank page,
- http://lists.sourceforge.net/lists/listinfo/u-boot-users/ (as referenced in the README file) contains no link to the list archive, and the "Mailing Lists" link on www.sourceforge.net/projects/u-boot suggests that the u-boot-users list is "not yet archived",
- the "Project Documentation" (Docmanager) on sourceforge again only links to a blank page...
... i find it not trivial to find those archives/FAQs you are referring to. Google searches for "u-boot-users" only seem to return cross-posts to other lists, no archive of this list itself. Suggestions?
-Carsten
Obviously your JTAG debugger performs the correct SDRAM initiali- zation sequence which is missing in your code. And no, this has NOTHING to do with the UPM tables.
Wolfgang Denk
- u-boot.sourceforge.net is just a blank page,
Yes, it would be nice if the home page for the project had some basic links. I believe at the minimum the following information should be linked:
1) Link to mailing list archives. Some people needs to be pointed to the search function in the "Lists" page.
BTW, I just checked: http://sourceforge.net/mail/?group_id=65938
It seems sourceforge.net took the archives offline again...
2) Link to U-Boot README file. It is expected people check it fist after getting U-Boot tarball or from CVS but not always.
3) Link to Twiki FAQ: http://www.denx.de/twiki/bin/view/DULG
(It would be great if a PDF or Tarball HTML version was available for offline reading)
Optionally the following information:
- Link to sample BDI configuration files.
- Link to related lists (most people involved with u-boot is also involved with embedded linux. So, link to some embedded linux lists
- For supported CPU platforms, a link to CPU handbook and other supplimentary infomation.
In message Pine.GSO.4.58.0401161637170.9331@bru-cse-065.cisco.com you wrote:
Is nobody readuing mailing ist archives or FAQ's any more?
Pardon my ignorance, but can you point me to the mailing list archives or the FAQ?
The link to the mailing list is in the README, see section "Where to get help".
The link to the manual (including the FAQ seqction) is in the README, too; see section "U-Boot Porting Guide".
I see you flaming people on this list quite frequently for asking questions that have been asked before; I think it's fair to expect people to do their homework before asking a question, however seing as
- u-boot.sourceforge.net is just a blank page,
So what? All relevant information can be found at http://sourceforge.net/projects/u-boot
- http://lists.sourceforge.net/lists/listinfo/u-boot-users/ (as referenced
in the README file) contains no link to the list archive, and the "Mailing Lists" link on www.sourceforge.net/projects/u-boot suggests that the u-boot-users list is "not yet archived",
Well, you got a point here. Normally this should be working. I don't know what's going on at SF again. What you see now is NOT the normal situation. The list _is_ archived, and normally you can access it following the link above.
- the "Project Documentation" (Docmanager) on sourceforge again only links
to a blank page...
OK, I added a link to the DULG.
... i find it not trivial to find those archives/FAQs you are referring to. Google searches for "u-boot-users" only seem to return cross-posts to other lists, no archive of this list itself. Suggestions?
I have no idea what needs to be done to get list added to google...
Best regards,
Wolfgang Denk
Hello,
I have been porting das u-boot to run directly out of RAM. So far so good, boots, serial works, ethernet works..
After I tftpboot a known good Linux kernel and try to bootm it it fails while inflating the kernel in inflate_blocks_new due to a malloc failure. I reserved 1MB of RAM for malloc in the config file, but when malloc is called there is not enough memory.
The board is an 8275 with 32MB of RAM. das u-boot is downloaded with a BDI to 0.
Any ideas?
Thank you Kevin
In message 40082CE1.40905@catapult.com you wrote:
I have been porting das u-boot to run directly out of RAM. So far so good, boots, serial works, ethernet works..
Running U-Boot out of RAM requires some understanding of the underlying principles.
The board is an 8275 with 32MB of RAM. das u-boot is downloaded with a BDI to 0.
This will never work. You cannot boot Linux that way, as the Linux kernel will overwrite U-Boot when you try to boot it.
Best regards,
Wolfgang Denk
Wolfgang,
Thank you for the response. My target will have 256M of RAM and I can locate das u-boot anywhere in this range so that it is not overwritten during the Linux boot process.
Would the top of RAM be a better choice then? If das u-boot is not viable after Linux is started that is fine, the system will re-download das u-boot when required.
Thanks Kevin
Wolfgang Denk wrote:
In message 40082CE1.40905@catapult.com you wrote:
I have been porting das u-boot to run directly out of RAM. So far so good, boots, serial works, ethernet works..
Running U-Boot out of RAM requires some understanding of the underlying principles.
The board is an 8275 with 32MB of RAM. das u-boot is downloaded with a BDI to 0.
This will never work. You cannot boot Linux that way, as the Linux kernel will overwrite U-Boot when you try to boot it.
Best regards,
Wolfgang Denk
In message 400ABE6E.2090404@catapult.com you wrote:
Would the top of RAM be a better choice then?
At least this is what the original implementation does (except for special features like PRAM, log buffers, video buffers, etc.)
--------------090906070306060007010501 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
And please, stop posting HTML.
Wolfgang Denk
participants (5)
-
Carsten Koester -
Kevin A. Sapp -
listmember＠orkun.us -
sudhakar rajashekhara -
Wolfgang Denk