#9702 BLOC 1.5-har: Linux (OS42) crashes on resume on an XO-1.5 B3
Zarro Boogs per Child
bugtracker at laptop.org
Fri Nov 20 20:55:05 EST 2009
#9702: Linux (OS42) crashes on resume on an XO-1.5 B3
--------------------------------+-------------------------------------------
Reporter: wad | Owner: dsaxena
Type: defect | Status: new
Priority: blocker | Milestone: 1.5-hardware-C
Component: kernel | Version: Development build as of this date
Resolution: | Keywords:
Next_action: diagnose | Verified: 1
Deployment_affected: | Blockedby:
Blocking: |
--------------------------------+-------------------------------------------
Comment(by cjb):
The two necessary printks are the ones in _set_ios and in _pci_resume.
Without both of them, resuming the powered-over-suspend card fails and we
hang.
I've switched the printks over to being mdelay(20)s. Approximately
anything less than 20ms seems insufficient to avoid the hang.
So! Conclusions from the limited testing done here are:
* /if/ we're going to do a resume with a continually-powered SD card
* we need to be more permissive about timing than the code currently is
There are other possibilities, such as a change in SD power sequencing
between B2 and B3. But the largest change between B2 and B3 is, of
course, that power is maintained to the SD card by default across suspend.
I'm not using pgf's EC that downs SD power when MAIN_ON goes low.
I'll attach the kernel patch that makes things work for me to this bug.
It should be tried on a machine with pgf's EC change too, since that may
be an entirely different problem. (Note that this patch is, of course, a
temporary workaround while we figure out what's really going on.)
--
Ticket URL: <http://dev.laptop.org/ticket/9702#comment:15>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system
More information about the Bugs
mailing list