diskless VFS lock issue

Poul-Henning Kamp phk at phk.freebsd.dk
Fri Dec 23 02:18:21 PST 2005

fxp0: link state changed to UP
KDB: stack backtrace:
kdb_backtrace(d448a9b4,c06c18b6,c088e359,c08a84c5,c45fca50) at 0xc067f70d = kdb_backtrace+0x29
vfs_badlock(c088e359,c08a84c5,c45fca50) at 0xc06c17bd = vfs_badlock+0x11
assert_vop_locked(c45fca50,c08a84c5,c45fca50,c08a84c5) at 0xc06c18b6 = assert_vop_locked+0x4a
VOP_GETPAGES_APV(c091c9e0,d448a9e8) at 0xc0830e9e = VOP_GETPAGES_APV+0x52
vnode_pager_getpages(c1054c1c,d448aa34,1,0) at 0xc07c3917 = vnode_pager_getpages+0xd3
vm_imgact_hold_page(c1054c1c,6d104,0,d448aa8c,c0634851) at 0xc07b3051 = vm_imgact_hold_page+0x81
vm_imgact_map_page(c1054c1c,6d104,0,d448aaa4,c0649c03) at 0xc07b31cd = vm_imgact_map_page+0x11
elf32_load_section(c4365000,c1054c1c,6a580,80b3580,225a8) at 0xc0634851 = elf32_load_section+0x169
exec_elf32_imgact(d448abc4,c45fd000,0,1,0) at 0xc0634e1f = exec_elf32_imgact+0x24b
do_execve(c435e480,d448ac90,0,0,d448ac90) at 0xc0648cbe = do_execve+0x262
kern_execve(c435e480,d448ac90,0,d35d8000,d35d8000) at 0xc06489d0 = kern_execve+0xc8
execve(c435e480,d448acf4,bfbfffe4,bfbffff2,bfbfffe8,bfbffffd,bfbfffec,0) at 0xc06488e2 = execve+0x32
start_init(0,d448ad38) at 0xc063653f = start_init+0x20f
fork_exit(c0636330,0,d448ad38) at 0xc064d81d = fork_exit+0x75
fork_trampoline() at 0xc0809d6c = fork_trampoline+0x8
--- trap 0x1, eip = 0, esp = 0xd448ad6c, ebp = 0 ---
VOP_GETPAGES: 0xc45fca50 is not locked but should be
KDB: enter: lock violation
[thread pid 1 tid 100007 ]
Stopped at      0xc067f78f = kdb_enter+0x2b:    nop     

Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

More information about the freebsd-current mailing list