#9707 BLOC 1.5-har: occasional oops in sdio_remove_func() in S/R (was: occasional oops in sdio_remove_func() in suspend)

[Zarro Boogs per Child]

#9707: occasional oops in sdio_remove_func() in S/R
--------------------------------+-------------------------------------------
           Reporter:  wad       |       Owner:  dsaxena       
               Type:  defect    |      Status:  new           
           Priority:  blocker   |   Milestone:  1.5-hardware-B
          Component:  kernel    |     Version:  1.5-B2        
         Resolution:            |    Keywords:                
        Next_action:  diagnose  |    Verified:  0             
Deployment_affected:            |   Blockedby:                
           Blocking:            |  
--------------------------------+-------------------------------------------
Changes (by wad):

 * cc: dsd (removed)
  * keywords:  q3a16 os42 b2 =>
  * milestone:  1.5-hardware-C => 1.5-hardware-B


Comment:

 Somewhat anticipating a development like this, we did run a WLAN reset
 signal to the southbridge.   In XO-1, we never got the WLAN module to
 reliably reset itself on power-up.  On XO-1.5, a similar reset signal
 (WLAN_RESET#, active low) goes to pin GPIO0 (powered in suspend,
 guaranteed to power up high or deasserted for an active low signal).

 I suggest that we fix the bug in the mmc_rescan error handling code, and
 add a quirk to the libertas driver that asserts a side-channel reset
 signal (in our case through a GPIO) whenever the device is powered up.

-- 
Ticket URL: <http://dev.laptop.org/ticket/9707#comment:23>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system