[Barrelfish-users] Kernel user page faults

Baumann Andrew andrewb at inf.ethz.ch
Mon Dec 5 17:10:22 CET 2011


Looks like the program 'bfib1' on core 1 (which is probably a spanned domain, i.e. the dispatcher is for the bfib program) has taken a page-fault in low-level dispatcher code. This is fatal to the dispatcher, and generally only the first fault is meaningful. It might be triggered by corrupting some dispatcher structures from your program, or it might be our bug, but from just this it's hard to tell. The first step in debugging it would be to disassemble the code at 48f8d8.

Andrew

-----Original Message-----
From: Georgios Varisteas [mailto:yorgos at kth.se] 
Sent: Monday, 05 December, 2011 2:08
To: barrelfish-users at lists.inf.ethz.ch
Subject: [Barrelfish-users] Kernel user page faults

Hi,

When I run an app for the second time on any core (the same or different) I always get this

kernel 1: user page fault WHILE DISABLED in 'bfib1': addr 0 IP 48f8d8, error 0x4
kernel 1: user page fault WHILE DISABLED in 'bfib1': addr 48f8f6 IP 48f8f6, error 0x7
kernel 1: user page fault WHILE DISABLED in 'bfib1': addr 48f8f6 IP 48f8f6, error 0x7
kernel 1: generic_handle_user_exception: too many faults, making domain unrunnable

It does not seem to affect the application but I'm wondering which domain it refers to and if an other app/service is failing at that point.

cheers,
Georgios

_______________________________________________
Barrelfish-users mailing list
Barrelfish-users at lists.inf.ethz.ch
https://lists.inf.ethz.ch/mailman/listinfo/barrelfish-users



More information about the Barrelfish-users mailing list