CURRENT r296381 panic in vn_sendfile (/usr/src/sys/kern/kern_sendfile.c:833)

Hello.

I get kernel panic on high loaded server with messages 

savecore: reboot after panic:
   vn_sendfile: mlen 326 space -20 hdrlen 326


# kgdb kernel.debug /var/crash/vmcore.0

Unread portion of the kernel message buffer:
panic: vn_sendfile: mlen 326 space -20 hdrlen 326
cpuid = 5
KDB: stack backtrace:
db_trace_self_wrapper() at db_trace_self_wrapper+0x2b/frame 0xfffffe20206314f0
vpanic() at vpanic+0x182/frame 0xfffffe2020631570
kassert_panic() at kassert_panic+0x126/frame 0xfffffe20206315e0
vn_sendfile() at vn_sendfile+0x14ca/frame 0xfffffe2020631900
sys_sendfile() at sys_sendfile+0x11e/frame 0xfffffe20206319a0
amd64_syscall() at amd64_syscall+0x2db/frame 0xfffffe2020631ab0
Xfast_syscall() at Xfast_syscall+0xfb/frame 0xfffffe2020631ab0
--- syscall (393, FreeBSD ELF64, sys_sendfile), rip = 0x801ef062a, rsp = 0x7fffffffd8d8, rbp = 0x7fffffffe1d0 ---
KDB: enter: panic

Reading symbols from /boot/kernel/zfs.ko...Reading symbols from /usr/lib/debug//boot/kernel/zfs.ko.debug...done.
done.
Loaded symbols for /boot/kernel/zfs.ko
Reading symbols from /boot/kernel/opensolaris.ko...Reading symbols from /usr/lib/debug//boot/kernel/opensolaris.ko.debug...done.
done.
Loaded symbols for /boot/kernel/opensolaris.ko
Reading symbols from /boot/kernel/carp.ko...Reading symbols from /usr/lib/debug//boot/kernel/carp.ko.debug...done.
done.
Loaded symbols for /boot/kernel/carp.ko
Reading symbols from /boot/kernel/ums.ko...Reading symbols from /usr/lib/debug//boot/kernel/ums.ko.debug...done.
done.
Loaded symbols for /boot/kernel/ums.ko
Reading symbols from /boot/kernel/tmpfs.ko...Reading symbols from /usr/lib/debug//boot/kernel/tmpfs.ko.debug...done.
done.
Loaded symbols for /boot/kernel/tmpfs.ko
#0  doadump (textdump=0) at pcpu.h:221
221             __asm("movq %%gs:%1,%0" : "=r" (td)
(kgdb) bt
#0  doadump (textdump=0) at pcpu.h:221
#1  0xffffffff80384a0b in db_dump (dummy=<value optimized out>, dummy2=false, dummy3=0, dummy4=0x0) at /usr/src/sys/ddb/db_command.c:533
#2  0xffffffff803847fe in db_command (cmd_table=0x0) at /usr/src/sys/ddb/db_command.c:440
#3  0xffffffff80384594 in db_command_loop () at /usr/src/sys/ddb/db_command.c:493
#4  0xffffffff8038702b in db_trap (type=<value optimized out>, code=0) at /usr/src/sys/ddb/db_main.c:251
#5  0xffffffff80a656e3 in kdb_trap (type=3, code=0, tf=<value optimized out>) at /usr/src/sys/kern/subr_kdb.c:654
#6  0xffffffff80ea1298 in trap (frame=0xfffffe2020631420) at /usr/src/sys/amd64/amd64/trap.c:556
#7  0xffffffff80e81a77 in calltrap () at /usr/src/sys/amd64/amd64/exception.S:234
#8  0xffffffff80a64dcb in kdb_enter (why=0xffffffff813b6c2f "panic", msg=0x80 <Address 0x80 out of bounds>) at cpufunc.h:63
#9  0xffffffff80a27b5f in vpanic (fmt=<value optimized out>, ap=<value optimized out>) at /usr/src/sys/kern/kern_shutdown.c:750
#10 0xffffffff80a279b6 in kassert_panic (fmt=<value optimized out>) at /usr/src/sys/kern/kern_shutdown.c:647
#11 0xffffffff80a25efa in vn_sendfile (fp=<value optimized out>, sockfd=1619, hdr_uio=<value optimized out>, trl_uio=0x0, offset=0, 
    nbytes=<value optimized out>, sent=<value optimized out>, flags=<value optimized out>, kflags=<value optimized out>, td=0xa8)
    at /usr/src/sys/kern/kern_sendfile.c:833
#12 0xffffffff80a2641e in sys_sendfile (td=0xfffff80253593000, uap=0xfffffe2020631a40) at file.h:382
#13 0xffffffff80ea214b in amd64_syscall (td=0xfffff80253593000, traced=0) at subr_syscall.c:135
#14 0xffffffff80e81d5b in Xfast_syscall () at /usr/src/sys/amd64/amd64/exception.S:394
#15 0x0000000801ef062a in ?? ()
Previous frame inner to this frame (corrupt stack?)
Current language:  auto; currently minimal
(kgdb) list *0xffffffff80a25efa
0xffffffff80a25efa is in vn_sendfile (/usr/src/sys/kern/kern_sendfile.c:833).
828                             free(sfio, M_TEMP);
829                             goto done;
830                     }
831
832                     /* Add the buffer chain to the socket buffer. */
833                     KASSERT(m_length(m, NULL) == space + hdrlen,
834                         ("%s: mlen %u space %d hdrlen %d",
835                         __func__, m_length(m, NULL), space, hdrlen));
836
837                     CURVNET_SET(so->so_vnet);


System have 128Gb memory
zfs as FS
DB's worked on it and web pages served by this server.

core saved. 
panic periodicaly repeted (few hours -- up to few days) 

Before this, old current (about two year old CURRENT ) work on this server without crashes.

Can anybody point me to way of more complex problem diagnostic or any other useful things

Thank you.