Cardbus panics

[M. Warner Losh]

In message: <20051212051813.GA19499 at mooseriver.com>
            Josef Grosch <jgrosch at MooseRiver.com> writes:
: 
: I have an IBM T22 thinkpad running FreeBSD 5.4-RELEASE-p8. I have the
: CardBus devices in the current kernel. When I insert a CardBus device, A
: Xircom RealPort CardBus Ethernet I get a kernel panic like so;
: 
: Dec  9 18:42:32 paris kernel: cbb alloc res fail 
: Dec  9 18:42:32 paris kernel: cardbus1: Can't get memory for IO ports 
: Dec  9 18:42:32 paris kernel: dc0: <Xircom X3201 10/100BaseTX> port 0-0x7f mem \
: 0x88000000-0x880007ff,0x88000800-0x88000fff at device 0.0 on cardbus1 
: Dec  9 18:42:32 paris kernel: cbb alloc res fail 
: Dec  9 18:42:32 paris kernel: dc0: couldn't map ports/memory 
: Dec  9 18:42:32 paris kernel:  
: Dec  9 18:42:32 paris kernel:  
: Dec  9 18:42:32 paris kernel: Fatal trap 12: page fault while in kernel mode 
: Dec  9 18:42:32 paris kernel: fault virtual address     = 0x30 
: Dec  9 18:42:32 paris kernel: fault code                = supervisor write, pag\
: e not present 
: Dec  9 18:42:32 paris kernel: instruction pointer       = 0x8:0xc07b01f2 
: Dec  9 18:42:32 paris kernel: stack pointer             = 0x10:0xd5428c10 
: Dec  9 18:42:32 paris kernel: frame pointer             = 0x10:0xd5428c14 
: Dec  9 18:42:32 paris kernel: code segment              = base 0x0, limit 0xfff\
: ff, type 0x1b 
: Dec  9 18:42:32 paris kernel: = DPL 0, pres 1, def32 1, gran 1 
: Dec  9 18:42:32 paris kernel: processor eflags  = interrupt enabled, resume, IO\
: PL = 0 
: Dec  9 18:42:32 paris kernel: current process           = 38 (cbb1) 
: Dec  9 18:42:32 paris kernel: trap number               = 12 
: Dec  9 18:42:32 paris kernel: panic: page fault 
: Dec  9 18:42:32 paris kernel: Uptime: 35s 
: Dec  9 18:42:32 paris kernel: Cannot dump. No dump device defined. 
: Dec  9 18:42:32 paris kernel: Automatic reboot in 15 seconds - press a key on t\
: he console to abort 
: 
: 
: Any ideas How I can fix this? The suggested sysctl, 
: hw.pci.allow_unsupported_io_range, does not seem to exist in the kernel.

Try 6.0 or a recent 7.0 snapshot.  A bug similar to this has been
fixed.

Warner