Problem with C4B on FreeBSD-Stable

[Cian Hughes]

Ok, well i've done lots of messing around since i last wrote  
anything, I have cvsup'ed my src to FreeBSD-Current, and made a new  
patchset for c4b, with only some minor changes to the Makefiles it  
all compiled, and I have asterisk talking to my AVM B1 through  
chan_capi.so, which is very nice, I have got one vmcore from a panic  
(my old kernel did not have debug flags enabled), unfortunately it  
weighs in a 132MB so I haven't attached it, I think it's usefulness  
may be limited, someone else with ssh access to the box thinks they  
may have accidentally cvsuped src-all since I compiled the kernel  
last week, so I think that the panic may have been due to source  
synchronisation issue, to be safe I actually dumped /usr/src & /usr/ 
obj and started again, my reason for not posting again is that the  
patch & compile of everything (world included) only finished at  
11.30pm last night.  For the moment I'll put the Sun compile on Hold,  
and concentrate on making a new patchset for FreeBSD-Current. How  
hard do you think it would be to get c4b incorporated into FreeBSD- 
Current, this patching lark is a bit of a pain.

Regards, Cian.

On 18 Meith 2005, at 09:48, Thomas Wintergerst wrote:

> Hi Cian,
>
> Could you provide a stack backtrace for the panics? Maybe this  
> could shed some light on where the crash occurs.
>
> Is this still 5.4-RELEASE or is it newer? Single- or multi-CPU?
>
> Cian Hughes wrote:
>
>> (I know I shouldn't reply to my own posts)
>>
> [...]
>
>> ______
>> Now we kernel panic when running capitest:
>> # capitest -s -tx 014900248
>> capidev: T capidev_open: CAPI device is beeing opened
>> capidev: D capidev_open: new temp fd 4 allocated
>> kernel trap 12 with interrupts disabled
>> Fatal trap 12: page fault while in kernel mode
>> fault virtual address   = 0x6c
>> fault code              = supervisor read, page not present
>> instruction pointer     = 0x8:0xc050b6d1
>> stack pointer           = 0x10:0xc7dddbd8
>> frame pointer           = 0x10:0xc7dddbdc
>> code segment            = base 0x0, limit 0xfffff, type 0x1b
>>                         = DPL 0, pres 1, def32 1, gran 1
>> processor eflags        = resume, IOPL = 0
>> current process         = 558 (capitest)
>> trap number             = 12
>> panic: page fault
>> Uptime: 54s
>> Cannot dump. No dump device defined.
>> Automatic reboot in 15 seconds - press a key on the console to abort
>> and again when starting asterisk:
>> oppy:root:/usr/local/etc/asterisk:0# capidev: T capidev_open:  
>> CAPI  device is beeing opened
>> capidev: D capidev_open: new temp fd 25 allocated
>> capidev: D capidev_ioctl: CAPI_GET_PROFILE: Result 0x0000,  
>> controller 0
>> capidev: D capidev_close: Application not registered, only close  
>> file  descriptor
>> capidev: T capidev_open: CAPI device is beeing opened
>> capidev: D capidev_open: new temp fd 25 allocated
>> kernel trap 12 with interrupts disabled
>> Fatal trap 12: page fault while in kernel mode
>> fault virtual address   = 0x6c
>> fault code              = supervisor read, page not present
>> instruction pointer     = 0x8:0xc050b6d1
>> stack pointer           = 0x10:0xcc445bd8
>> frame pointer           = 0x10:0xcc445bdc
>> code segment            = base 0x0, limit 0xfffff, type 0x1b
>>                         = DPL 0, pres 1, def32 1, gran 1
>> processor eflags        = resume, IOPL = 0
>> current process         = 599 (asterisk)
>> trap number             = 12
>> panic: page fault
>> Uptime: 5m14s
>> Cannot dump. No dump device defined.
>> Regards, Cian
>>
> [...]
>
> -- 
>
> Gruss,
>
> Thomas Wintergerst
>