bugThe GNU Hurd - Bugs: bug #25054, Kernel panic with eth-multiplexer

 
 

bug #25054: Kernel panic with eth-multiplexer

Submitted by:  None
Submitted on:  Tue 09 Dec 2008 10:31:05 PM UTC  
 
Category: GNU MachSeverity: 3 - Normal
Priority: 5 - NormalItem Group: None
Status: NonePrivacy: Public
Assigned to: NoneOriginator Name: antrik
Originator Email: -unavailable-Open/Closed: Open
Reproducibility: IntermittentSize (loc): None
Planned Release: NoneEffort: 0.00
Wiki-like text discussion box: 

Add a New Comment (Rich MarkupRich Markup):
   

You are not logged in

Please log in, so followups can be emailed to you.

 

Mon 06 Jul 2009 08:00:18 AM UTC, comment #2:

Err, even a bogus user-land tool is not supposed to make a kernel
crash. The zalloc panic can be understood as we do not have quotas &
such, but the reference counter shouldn't have issues.

Samuel Thibault <sthibaul>
Project Member
Mon 06 Jul 2009 06:51:18 AM UTC, comment #1:

The kernel panic was caused by a bug in eth-multiplexer, which didn't deallocate the out-of-line data in ds_device_write().
The problem doesn't exist any more.

Zheng Da <zhengda>
Project Member
Tue 09 Dec 2008 10:31:05 PM UTC, original submission:

Transferring large files (some hundred MB) with NFS or HTTP, when using the
ethernet multiplexer (from zhengda-soc2008-virt-branch in Hurd CVS), regularily
produces:

Assertion `object->ref_count > 0' failed in file "../vm/vm_object.c", line 346

vm_object_reference
/build/buildd/gnumach-1.3.99.dfsg.cvs20080708/build-dbg/../vm/vm_object.c:346
vm_object_collapse
/build/buildd/gnumach-1.3.99.dfsg.cvs20080708/build-dbg/../vm/vm_object.c:2689
vm_fault_page
/build/buildd/gnumach-1.3.99.dfsg.cvs20080708/build-dbg/../vm/vm_fault.c:882
vm_fault
/build/buildd/gnumach-1.3.99.dfsg.cvs20080708/build-dbg/../vm/vm_fault.c:1279
user_trap
/build/buildd/gnumach-1.3.99.dfsg.cvs20080708/build-dbg/../i386/i386/trap.c:514

This is with gnumach-dbg=2:1.3.99.dfsg.cvs20080708-1

Interestingly, doing the same with a locally built gnumach that contains the
patch enabling promiscuous mode for the network device, produces a zalloc panic
instead. (A quick check with vmstat and ps showed growing memory use by the
multiplexer process, but haven't investigated further.) I wonder whether the
difference is because it's locally built (different toolchain?), or because of
the patch.

Anonymous

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach File(s):
   
   
Comment:
   

No files currently attached

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -unavailable- added by sthibaul (Posted a comment)
  • -unavailable- added by zhengda (Posted a comment)
  •  

    Do you think this task is very important?
    If so, you can click here to add your encouragement to it.
    This task has 0 encouragements so far.

    Only logged-in users can vote.

     

    Please enter the title of George Orwell's famous dystopian book (it's a date):

     

     

    No Changes Have Been Made to This Item

    Back to the top


    Powered by Savane 3.1-cleanup