Re: Three new LORs

Nikolai Nespor wrote:
> Hi list!
> 
> The following three LORs are from BETA-3 as of three days ago. As far as 
> I know, they haven't been reported yet.
> 
> lock order reversal:
> 1st 0xc086dc58 allprison (allprison) _at_ /usr/src/sys/kern/kern_jail.c:2449
> 2nd 0xc3308058 ufs (ufs) _at_ /usr/src/sys/kern/vfs_subr.c:2188
> KDB: stack backtrace:
> db_trace_self_wrapper(c07c8e78,c2ef8aac,c0569675,c055acab,c07cbb5c,...) 
> at db_trace_self_wrapper+0x26
> kdb_backtrace(c055acab,c07cbb5c,c3028138,c302e590,c2ef8b08,...) at 
> kdb_backtrace+0x29
> _witness_debugger(c07cbb5c,c3308058,c07bee83,c302e590,c07d2e8d,...) at 
> _witness_debugger+0x25
> witness_checkorder(c3308058,9,c07d2e8d,88c,0,...) at 
> witness_checkorder+0x839
> __lockmgr_args(c3308058,80100,c3308074,0,0,...) at __lockmgr_args+0x7a7
> ffs_lock(c2ef8c24,c2ef8c18,c05694bc,c3028138,c07d2e8d,...) at ffs_lock+0x8a
> VOP_LOCK1_APV(c083d6a0,c2ef8c24,c3308074,c08530e0,c3308000,...) at 
> VOP_LOCK1_APV+0xf3
> _vn_lock(c3308000,80100,c07d2e8d,88c,991,...) at _vn_lock+0x5e
> vrele(c3308000,c07c1905,998,991,0,...) at vrele+0x137
> prison_deref(c2ef8cd4,c0562823,c534d000,1,c07ca4e6,...) at 
> prison_deref+0x49d
> prison_complete(c534d000,1,c07ca4e6,51,c30cda18,...) at 
> prison_complete+0x10
> taskqueue_run(c30cda00,c30cda18,0,c07ba2a8,0,...) at taskqueue_run+0x103
> taskqueue_thread_loop(c08746ac,c2ef8d38,c07c1448,33e,c086d300,...) at 
> taskqueue_thread_loop+0x68
> fork_exit(c0562910,c08746ac,c2ef8d38) at fork_exit+0xb8
> fork_trampoline() at fork_trampoline+0x8
> --- trap 0, eip = 0, esp = 0xc2ef8d70, ebp = 0 ---
> lock order reversal:
> 1st 0xc54c1df4 ufs (ufs) _at_ /usr/src/sys/kern/vfs_mount.c:1200
> 2nd 0xc3c198b8 syncer (syncer) _at_ /usr/src/sys/kern/vfs_subr.c:2188
> KDB: stack backtrace:
> db_trace_self_wrapper(c07c8e78,d6225a54,c0569675,c055acab,c07cbb5c,...) 
> at db_trace_self_wrapper+0x26
> kdb_backtrace(c055acab,c07cbb5c,c302e590,c302e660,d6225ab0,...) at 
> kdb_backtrace+0x29
> _witness_debugger(c07cbb5c,c3c198b8,c07d300c,c302e660,c07d2e8d,...) at 
> _witness_debugger+0x25
> witness_checkorder(c3c198b8,9,c07d2e8d,88c,0,...) at 
> witness_checkorder+0x839
> __lockmgr_args(c3c198b8,80100,c3c198d4,0,0,...) at __lockmgr_args+0x7a7
> vop_stdlock(d6225bc4,d6225bb8,c05694bc,c09a0af8,c07d2e8d,...) at 
> vop_stdlock+0x62
> VOP_LOCK1_APV(c081fae0,d6225bc4,c3c198d4,c08530e0,c3c19860,...) at 
> VOP_LOCK1_APV+0xf3
> _vn_lock(c3c19860,80100,c07d2e8d,88c,df,...) at _vn_lock+0x5e
> vrele(c3c19860,0,c07d26a2,4f5,c08130c0,...) at vrele+0x137
> dounmount(c3272508,8080000,c32986c0,47a,3,...) at dounmount+0x3ce
> unmount(c32986c0,d6225cf8,8,d6225d38,c080cc08,...) at unmount+0x2ff
> syscall(d6225d38) at syscall+0x2d3
> Xint0x80_syscall() at Xint0x80_syscall+0x20
> --- syscall (22, FreeBSD ELF32, unmount), eip = 0x280d9c2f, esp = 
> 0xbfbfe56c, ebp = 0xbfbfe638 ---
> 
> lock order reversal:
> 1st 0xc4f07e00 so_snd_sx (so_snd_sx) _at_ /usr/src/sys/kern/uipc_sockbuf.c:145
> 2nd 0xc38d2058 ufs (ufs) _at_ /usr/src/sys/kern/uipc_syscalls.c:2090
> KDB: stack backtrace:
> db_trace_self_wrapper(c07c8e78,d62eb9ac,c0569675,c055acab,c07cbb5c,...) 
> at db_trace_self_wrapper+0x26
> kdb_backtrace(c055acab,c07cbb5c,c302e6c8,c302e590,d62eba08,...) at 
> kdb_backtrace+0x29
> _witness_debugger(c07cbb5c,c38d2058,c07bee83,c302e590,c07d01bd,...) at 
> _witness_debugger+0x25
> witness_checkorder(c38d2058,1,c07d01bd,82a,c38d2074,...) at 
> witness_checkorder+0x839
> __lockmgr_args(c38d2058,200400,c38d2074,0,0,...) at __lockmgr_args+0x237
> ffs_lock(d62ebb24,d62ebb18,c09a0720,c3939764,c09a0710,...) at ffs_lock+0x8a
> VOP_LOCK1_APV(c083d6a0,d62ebb24,c46e0ee0,c08530e0,c38d2000,...) at 
> VOP_LOCK1_APV+0xf3
> _vn_lock(c38d2000,200400,c07d01bd,82a,0,...) at _vn_lock+0x5e
> kern_sendfile(c39396c0,d62ebcf8,c358dc40,0,0,...) at kern_sendfile+0x9a0
> sendfile(c39396c0,d62ebcf8,20,d62ebd38,c080f49c,...) at sendfile+0xa8
> syscall(d62ebd38) at syscall+0x2d3
> Xint0x80_syscall() at Xint0x80_syscall+0x20
> --- syscall (393, FreeBSD ELF32, sendfile), eip = 0x282f062b, esp = 
> 0xbfbfc52c, ebp = 0xbfbfc588 ---
> 
> Sorry if the lines are messed up, I'm stuck with webmail here.
> 
> Nikolai

The first (allprison_lock) has been reported in a PR is a fix is 
forthcoming.

- Jamie