em JumboFrame improovement and PCIe addon-card regression [Was: Re: em regression, UDP LOR followed by ssh stall]

[pluknet]

On 17 April 2010 00:53, Jack Vogel <jfvogel at gmail.com> wrote:
> Why are you using ZERO_COPY_SOCKETS?  And is this LOR happening on STABLE or
> CURRENT?
>

I got exactly this and another similar LORs with GENERIC on head.

The second:
login: lock order reversal:
 1st 0xffffff0002539a18 em0:rx(0) (em0:rx(0)) @
/home/svn/freebsd/head/sys/dev/e1000/if_em.c:1452
 2nd 0xffffffff80e3a280 in_multi_mtx (in_multi_mtx) @
/home/svn/freebsd/head/sys/netinet/igmp.c:823
KDB: stack backtrace:
db_trace_self_wrapper() at db_trace_self_wrapper+0x2a
_witness_debugger() at _witness_debugger+0x2e
witness_checkorder() at witness_checkorder+0x81e
_mtx_lock_flags() at _mtx_lock_flags+0x78
igmp_input() at igmp_input+0x4ce
ip_input() at ip_input+0xbc
netisr_dispatch_src() at netisr_dispatch_src+0xb8
ether_demux() at ether_demux+0x17d
ether_input() at ether_input+0x175
em_rxeof() at em_rxeof+0x165
em_handle_que() at em_handle_que+0x68
taskqueue_run() at taskqueue_run+0x91
taskqueue_thread_loop() at taskqueue_thread_loop+0x3f
fork_exit() at fork_exit+0x12a
fork_trampoline() at fork_trampoline+0xe
--- trap 0, rip = 0, rsp = 0xffffff80000bed30, rbp = 0 ---

-- 
wbr,
pluknet