recent 8-current panics if load device modules from userspace under vmware

[John Baldwin]

On Friday 05 June 2009 5:10:43 am Vladimir Grebenschikov wrote:
> Hi
> 
> intsmb0: <Intel PIIX4 SMBUS Interface> port 0x1040-0x104f at device 7.3 on 
pci0
> panic: resource_list_alloc: resource entry is busy
> 
> If load modules from startup scripts it panics with such message.
> 
> (kgdb) bt
> #0  doadump () at pcpu.h:246
> #1  0xc0557413 in boot (howto=260) at /usr/src/sys/kern/kern_shutdown.c:420
> #2  0xc055764d in panic (fmt=Variable "fmt" is not available.) 
at /usr/src/sys/kern/kern_shutdown.c:576
> #3  0xc057d176 in resource_list_alloc (rl=0xc38b2304, bus=0xc38f2c00, 
>     child=0xc38f2e00, type=4, rid=0xc3742934, start=0, end=4294967295, 
>     count=1, flags=2) at /usr/src/sys/kern/subr_bus.c:2836
> #4  0xc04d6f47 in pci_alloc_resource (dev=0xc38f2c00, child=0xc38f2e00, 
>     type=4, rid=0xc3742934, start=0, end=4294967295, count=1, flags=2)
>     at /usr/src/sys/dev/pci/pci.c:3599
> #5  0xc057d08c in bus_alloc_resource (dev=0xc38f2e00, type=4, 
rid=0xc3742934, 
>     start=0, end=4294967295, count=1, flags=2) at bus_if.h:263
> #6  0xc0639a3d in intsmb_attach (dev=0xc38f2e00) at bus.h:379
> #7  0xc057bdff in device_attach (dev=0xc38f2e00) at device_if.h:178
> #8  0xc057cd6c in device_probe_and_attach (dev=0xc38f2e00)
>     at /usr/src/sys/kern/subr_bus.c:2473
> #9  0xc04d63e5 in pci_driver_added (dev=0xc38f2c00, driver=0xc4053440)
>     at /usr/src/sys/dev/pci/pci.c:2833
> #10 0xc057a1a8 in devclass_driver_added (dc=0xc3840e00, driver=0xc4053440)
>     at bus_if.h:183
> #11 0xc057ab49 in devclass_add_driver (dc=0xc3840e00, driver=0xc4053440)
>     at /usr/src/sys/kern/subr_bus.c:942
> #12 0xc057b9b9 in driver_module_handler (mod=0xc3bafe40, what=0, 
>     arg=0xc40534fc) at /usr/src/sys/kern/subr_bus.c:3952
> #13 0xc05480f5 in module_register_init (arg=0xc40534e4)
>     at /usr/src/sys/kern/kern_module.c:124
> #14 0xc05407bd in linker_load_module (kldname=Variable "kldname" is not 
available.)
>     at /usr/src/sys/kern/kern_linker.c:234
> #15 0xc0540cdc in kern_kldload (td=0xc3a4caf0, file=0xc3924000 "if_le", 
>     fileid=0xc3742c70) at /usr/src/sys/kern/kern_linker.c:1022
> #16 0xc0540db4 in kldload (td=0xc3a4caf0, uap=0xc3742cf8)
>     at /usr/src/sys/kern/kern_linker.c:1050
> #17 0xc06c9157 in syscall (frame=0xc3742d38)
>     at /usr/src/sys/i386/i386/trap.c:1073
> #18 0xc06af7b0 in Xint0x80_syscall () 
at /usr/src/sys/i386/i386/exception.s:261
> #19 0x00000033 in ?? ()
> (kgdb) 
> 
> If I add same modules into loader.conf everything goes as expected.
> Modules tried: if_le, snd_es137x.

Can you get devinfo -r output for 1) prior to loading the module that causes a 
panic and 2) when you load the modules from the loader.

-- 
John Baldwin