On Wed, Dec 23, 2009 at 08:42:47PM -0800, Matthew Dillon wrote:
> I'm thinking possibly something in the pmap unwiring code, possibly
> in _pmap_unwire_pte_hold(), could be racing.
>
> Here is a patch to try. It adds a bunch of assertions in an attempt
> to catch the potential race.
>
> -Matt
> Matthew Dillon
> <dillon@backplane.com>
>
> diff --git a/sys/platform/pc32/i386/pmap.c b/sys/platform/pc32/i386/pmap.c
> index 53cc386..b07f2cf 100644
I tried changing the number of CPUs on vkernel from 16 to 4 and reduced
the -j number given to make command, then I get a panic in a different place:
(kgdb) bt
:
#4 0xc01a14fc in panic (fmt=0xc030329e "assertion: %s in %s")
at /usr/src/sys/kern/kern_shutdown.c:743
#5 0xc02d88c5 in pmap_remove_pages (pmap=0xd6ea6fb4, sva=0, eva=3217031168)
at /usr/src/sys/platform/pc32/i386/pmap.c:2856
#6 0xc02a5ee8 in vmspace_entry_delete (ve=0xdae62b40,
vkp=<value optimized out>) at /usr/src/sys/vm/vm_vmspace.c:514
#7 0xc02a639a in sys_vmspace_destroy (uap=0xdbadecf0)
at /usr/src/sys/vm/vm_vmspace.c:147
#8 0xc02dcd8c in syscall2 (frame=0xdbaded40)
at /usr/src/sys/platform/pc32/i386/trap.c:1359
#9 0xc02c83f6 in Xint0x80_syscall ()
at /usr/src/sys/platform/pc32/i386/exception.s:876
:
(kgdb) fr
#5 0xc02d88c5 in pmap_remove_pages (pmap=0xd6ea6fb4, sva=0, eva=3217031168)
at /usr/src/sys/platform/pc32/i386/pmap.c:2856
2856 KKASSERT(*pte);
(kgdb) info locals
lp = <value optimized out>
pte = 0xff80cff8
tpte = <value optimized out>
pv = 0xc2ebca9c
npv = 0xdbadec8c
m = 0xdaebbc30
info = {pir_flags = 0, pir_cpusync = {cs_run_func = 0xdaec4d00,
cs_fin1_func = 0x1d, cs_fin2_func = 0, cs_data = 0x0,
cs_maxcount = -609358712, cs_count = -1070966064, cs_mask = 3273535440}}
iscurrentpmap = 0
save_generation = <value optimized out>
__func__ = "pmap_remove_pages"
The kernel and vmcore are in ~y0netan1/crash/{kern,vmcore}.22 .
The kernel is compiled from source as of a307f7025 and your patch applied.