How to make a GTA02 pine for the fjords

Andy Green andy at
Wed Jul 23 21:21:17 CEST 2008

Hash: SHA1

Somebody in the thread at some point said:
| On Wednesday 23 July 2008 19:13:58 Andy Green wrote:
|> This used to work, I have done it myself... I think it can be related to
|> the not coming back from GSM wake issue maybe.
| I'm not certain that it goes to sleep.

But it doesn't "know" to fail until after you pull the USB plug if I
understood you?  So it should have been asleep OK?

| inserting/removing usb cable makes usb0 come and works...
| openocd/gdb say (assuming I'm not sitting in the bootloader):
| (gdb) bt
| #0  s3c24xx_default_idle () at include/asm/arch/system.h:41
| #1  0xc002980c in default_idle () at include/asm/arch/system.h:56
| #2  0xc00295ec in cpu_idle () at arch/arm/kernel/process.c:170
| #3  0xc02c0714 in rest_init () at init/main.c:453
| #4  0xc0008c18 in start_kernel () at init/main.c:648
| #5  0x30008034 in ?? ()
| So user space has been halted, we sleep somewhere... :)
| no idea if that is the state I initially reported... let me continue
to dig...
| I will probably not end up in the office tomorrow..

Wah I didn't know you were in .tw.

What annoys me about these resume problems is I put special code in to
dump nicely on serial console if it ever OOPSed, it never OOPSes.

This code is handy because it does the serial traffic without needing
serial devices out of resume.;a=commitdiff;h=8e18a9db5f895f654b06000d7bd10528e6b795f3

If you get to thinking it can help, dropping this somewhere should get
you a dump of syslog so far and a stack trace, you can maybe move it
around and see how far it is resuming...

	extern int global_inside_suspend;
	extern int oops_in_progress;

	oops_in_progress = 1;
	global_inside_suspend = 1;


Another idea is stick that in the AUX key interrupt if that is
operational by then, and / or disable the disable of it on suspend.

- -Andy
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora -


More information about the openmoko-kernel mailing list