hi,

crashes in pmap_clear_attrs() are invariably due flaky hardware.
have you run a memory test lately?

-Chuck


On Fri, Oct 28, 2005 at 10:49:44AM -0700, Matt SF wrote:
> Hi all--
> 
> had a crash today on a machine on which we do a lot of file access via 
> samba.  it is running 2.0.2 vanilla/generic:
> NetBSD 2.0.2 (INSTALL) #0: Wed Mar 23 09:13:49 UTC 2005
> jmc@faith.netbsd.org:/home/builds/ab/netbsd-2-0-2-RELEASE/i386/200503220
> 140Z-obj/home/builds/ab/netbsd-2-0-2-RELEASE/src/sys/arch/i386/compile/INSTALL
> 
> 
> was able to get this out of the db prompt:
> 
> db> cont
> uvm_fault(0xc079c660, 0xffefe000, 0, 1) -> 0xe
> kernel: page fault trap, code=0
> Stopped in pid 12.1 (ioflush) at        netbsd:pmap_clear_attrs+0x5d:   movl    0
> (%eax,%ecx,4),%edx
> db> trace
> pmap_clear_attrs(c0ba2400,40,1,c12bd0ca,c0dd1fc0) at 
> netbsd:pmap_clear_attrs+0x5
> d
> genfs_putpages(cb52ed34,0,cb52ed6c,c03a7650,c05bce20) at 
> netbsd:genfs_putpages+0
> xa68
> VOP_PUTPAGES(cdfc7d1c,0,0,0,0) at netbsd:VOP_PUTPAGES+0x40
> ffs_full_fsync(cb52ee54,ffffffff,0,cb52edfc,cdfc7d8c) at 
> netbsd:ffs_full_fsync+0
> x274
> ffs_fsync(cb52ee54,cdfc7d1c,10012,c03a7314,c05bc660) at 
> netbsd:ffs_fsync+0x4c
> VOP_FSYNC(cdfc7d1c,c1110e80,0,0,0) at netbsd:VOP_FSYNC+0x4c
> ffs_sync(c1340200,3,c1110e80,cb52a19c,c07673f0) at netbsd:ffs_sync+0x1ea
> sync_fsync(cb52ef14,0,cb52ef3c,c03a736c,c05bc660) at 
> netbsd:sync_fsync+0x8e
> VOP_FSYNC(cb90e398,c1110e80,8,0,0) at netbsd:VOP_FSYNC+0x4c
> sched_sync(cb4fb5ac,893000,89c000,0,c0100321) at netbsd:sched_sync+0x1bb
> db> 
> 
> 
> I have no problem upgrading to something like 2.1 but is there a way to 
> find if a problem like the above has been addressed?  thanks!
> -- 
> -------------------------------------------------------------------------