Bug 113342

Summary: suspend to ram pc-card
Product: [openSUSE] SUSE LINUX 10.0 Reporter: Martin Lasarsch <martin.lasarsch>
Component: Mobile DevicesAssignee: Forgotten User ZhJd0F0L3x <forgotten_ZhJd0F0L3x>
Status: RESOLVED WONTFIX QA Contact: E-mail List <qa-bugs>
Severity: Minor    
Priority: P5 - None    
Version: Beta 3   
Target Milestone: ---   
Hardware: Other   
OS: All   
Whiteboard:
Found By: Other Services Priority:
Business Priority: Blocker: ---
Marketing QA Status: --- IT Deployment: ---
Attachments: screenshot of panic

Description Martin Lasarsch 2005-08-26 15:02:16 UTC
on a t41p when i suspend to ram with ethernet pc-card inserted it will wake up 
proper, but when i remove the card after that i will get kernel panic. Only 
when card is inserted when i suspend. Booted with acpi_sleep=s3_bios. 
  
I tried it also with another (16bit) card, works. 9.3 with the pc-card works  
too. suspend to disk works also. 
 
I have the card here, let me know if you want it to test. 
  
0000:03:00.0 Ethernet controller: Abocom Systems Inc 21x4x DEC-Tulip  
compatible 10/100 Ethernet (rev 11)  
  
26: PCI 300.0: 0200 Ethernet controller  
  [Created at pci.277]  
  UDI: /org/freedesktop/Hal/devices/pci_13d1_ab08  
  Unique ID: JNkJ.5l_yCol3Ui8  
  Parent ID: B35A.3eeZp8MpqF4  
  SysFS ID: /devices/pci0000:00/0000:00:1e.0/0000:02:00.0/0000:03:00.0  
  SysFS BusID: 0000:03:00.0  
  Hardware Class: network  
  Model: "Abocom 21x4x DEC-Tulip compatible 10/100 Ethernet"  
  Vendor: pci 0x13d1 "Abocom Systems Inc"  
  Device: pci 0xab08 "21x4x DEC-Tulip compatible 10/100 Ethernet"  
  SubVendor: pci 0x13d1 "Abocom Systems Inc"  
  SubDevice: pci 0xab08  
  Revision: 0x11  
  Driver: "tulip"  
  Device File: eth1  
  I/O Ports: 0x4000-0x8fff (rw)  
  Memory Range: 0xc2000000-0xc20003ff (rw,non-prefetchable)  
  Memory Range: 0xe8000000-0xe801ffff (ro,prefetchable,disabled)  
  IRQ: 11 (11339 events)  
  HW Address: 00:e0:98:b8:d2:15  
  Module Alias: "pci:v000013D1d0000AB08sv000013D1sd0000AB08bc02sc00i00"  
  Driver Info #0:  
    Driver Status: tulip is active  
    Driver Activation Cmd: "modprobe tulip"  
  Config Status: cfg=new, avail=yes, need=no, active=unknown  
  Attached to: #22 (CardBus bridge)
Comment 1 Stefan Behlert 2005-08-26 17:49:06 UTC
Pavel, any idea about that? 
Comment 2 Forgotten User ZhJd0F0L3x 2005-08-26 18:41:43 UTC
1.) this is not a pc-card but a cardbus card.

2.) We'd need to see the panic to do anything about this. (serial console?,
digital camera?)
Comment 3 Pavel Machek 2005-08-26 22:21:23 UTC
Well, something wrong with PCMCIA drivers...  they are in pretty bad shape :-(.
Comment 4 Martin Lasarsch 2005-08-29 12:57:25 UTC
#2: 1) yes 
2) i have no serial, just a bad camera ... see attachment. 
 
if you have a laptop which is usually capable of suspend to ram with a serial 
port, say something i will bring you the card ... 
Comment 5 Martin Lasarsch 2005-08-29 12:58:50 UTC
Created attachment 47989 [details]
screenshot of panic
Comment 6 Forgotten User ZhJd0F0L3x 2005-08-29 14:37:58 UTC
Tainted: P

aka INVALID.
You can bring the card and we can test it on an untainted machine.
Comment 7 Forgotten User ZhJd0F0L3x 2005-08-30 16:07:38 UTC
Martin brought the Card over to my office and i will try to reproduce
Comment 8 Holger Macht 2005-08-31 10:12:29 UTC
Ok, great. So reassigning bug to you.
Comment 9 Forgotten User ZhJd0F0L3x 2005-12-17 11:19:32 UTC
i see lockups after suspend to ram and then removing or inserting cardbus cards on my systems too. will investigate.
Comment 11 Pavel Machek 2005-12-19 22:03:16 UTC
I see some problems even with regular PCMCIA cards: bluetooth is completely broken after suspend/resume (and it may be design problem with bluetooth subsystem :-().

Is it broken with more than one cardbus card?
Comment 14 Forgotten User ZhJd0F0L3x 2006-11-11 09:44:42 UTC
we are  now ejecting pc-cards per default before suspend.
I´ll keep this bug open to verify the bug (worked around is not a solution
Comment 15 Stefan Behlert 2007-02-15 12:04:15 UTC
Ok, please verify for 10.3 then.
Comment 16 Stefan Behlert 2007-05-07 10:16:25 UTC
Seife, had you time to check this?
Comment 17 Forgotten User ZhJd0F0L3x 2007-06-20 17:09:37 UTC
.
Comment 18 Forgotten User ZhJd0F0L3x 2007-06-21 21:00:02 UTC
sorry, i closed the wrong bug.
Comment 21 Forgotten User ZhJd0F0L3x 2007-11-27 11:07:13 UTC
SUSE 10.0 is short before EOL, i don't think we still want to fix this one.