Skip to contentSkip to conversation

do_run_[os]. why restart if there is a minor problem?

U-Boot-Users Ara Avanesyan 2 participants 5 messages Jan 28, 2005 – Feb 1, 2005 1 message not in the archive

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

#1 lore raw

Hi,

in function do_run_linux if it finds ramdisk's crc is bad or it's not "u-boot-ized" (mkimage) it resets the board. and many other similar minor things. so why restart? I propose making do_run_linux return something to indicate there is no need to reboot the board and bootm checks this value and returns.

if you find this may cause backwards compatibility problems, let's make it depending on some preprocessor define.

I can submit a patch if you say ok.

__
Thanks,
Ara

#2 lore raw
Quoted 6 lines · In message <005501c50539$af259e00$1000000a@araavanesyan> you wrote:

in function do_run_linux if it finds ramdisk's crc is bad or it's not "u-boot-ized" (mkimage) it resets the board. and many other similar minor things. so why restart? I propose making do_run_linux return something to indicate there is no need to reboot the board and bootm checks this value and returns.

Your proposal indicates that you did not follow the boot process. What is being done right before the ramdisk is being testet? Right, we uncompressed and loaded the Linux kernel, writing it to RAM starting at address 0x0000, thus overwriting the exception vertors installed by U-Boot.

Quoted 4 lines

if you find this may cause backwards compatibility problems, let's make it depending on some preprocessor define.

I can submit a patch if you say ok.

I'd be really, really surprised if you could do that. Actually U-Boot will crash at the next timer interrupt, when it junps into completely unexpected handler code.

I strongly recommend to check your ideas before posting.

I also recommend to read the code, ideally before posting. Pay special attention to comments. Sometimes these are actually helpful, like here:

"common/cmd_bootm.c":

284 /*
285 * We have reached the point of no return: we are going to
286 * overwrite all exception vector code, so we cannot easily
287 * recover from any failures any more...
288 */

Best regards,

Wolfgang Denk

Signature · 4 lines
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
The human mind ordinarily operates at only ten percent of its capaci-
ty - the rest is overhead for the operating system.
#3 lore raw

Hi Wolfgang,

Subject: Re: [U-Boot-Users] do_run_[os]. why restart if there is a minor problem?

Quoted 4 lines
Quoted 3 lines · In message <005501c50539$af259e00$1000000a@araavanesyan> you wrote:

in function do_run_linux if it finds ramdisk's crc is bad or it's not "u-boot-ized" (mkimage) it resets the board. and many other similar

minor

Quoted 1 line
Quoted 1 line

things. so why restart? I propose making do_run_linux return something

to

Quoted 1 line
Quoted 1 line

indicate there is no need to reboot the board and bootm checks this

value

Quoted 8 lines
Quoted 1 line

and returns.

Your proposal indicates that you did not follow the boot process. What is being done right before the ramdisk is being testet? Right, we uncompressed and loaded the Linux kernel, writing it to RAM starting at address 0x0000, thus overwriting the exception vertors installed by U-Boot.

Oh, my changes were working for my board (seems because my hdr->ih_load == 0x8000).
Anyways, I'm sure u-boot should not enter into un-recoverable state before making sure there are no Minor errors, such as ramdisk's checksum etc. How about calling a function like can_run_[os] before entering into un-recoverable state?
I think this would make u-boot more user friendly.

__
Thanks,
Ara

#4 lore raw
Quoted 3 lines · In message <002101c505df$adfd6d40$1000000a@araavanesyan> you wrote:

Oh, my changes were working for my board (seems because my hdr->ih_load == 0x8000).

This will not work on PowerPC systems where the kernel must be loaded to physical address 0x000.

Quoted 1 line

I think this would make u-boot more user friendly.

Maybe.

Best regards,

Wolfgang Denk

Signature · 4 lines
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
The nice thing about  standards  is that there are  so many to choose
from.                                           - Andrew S. Tanenbaum
#5 lore raw

Hi Wolfgang,

----- Original Message -----
From: "Wolfgang Denk" <[email protected]>
To: "Ara Avanesyan" <[email protected]>
Cc: <[email protected]>
Sent: Saturday, January 29, 2005 9:10 PM
Subject: Re: [U-Boot-Users] do_run_[os]. why restart if there is a minor problem?

Quoted 3 lines
Quoted 2 lines · In message <002101c505df$adfd6d40$1000000a@araavanesyan> you wrote:

Oh, my changes were working for my board (seems because my hdr->ih_load

==

Quoted 8 lines
Quoted 1 line

0x8000).

This will not work on PowerPC systems where the kernel must be loaded to physical address 0x000.

Quoted 1 line

I think this would make u-boot more user friendly.

Maybe.

So what's next?

btw, a small typo in tools/env/fw_env.c:line 729 DEVNAME (2) should be DEVNAME (1) :)

__
Thanks,
Ara