help with latest version of ZFS

When I try to copy large amounts of data using very recent sources (cvsup'd
this morning) I see this panic:

root:peedub:crash:bash:1> kgdb /boot/test/kernel vmcore.3 
[GDB will not be able to debug user-mode threads: /usr/lib/libthread_db.so: Undefined symbol "ps_pglobal_lookup"]
GNU gdb 6.1.1 [FreeBSD]
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i386-marcel-freebsd".

Unread portion of the kernel message buffer:
panic: kmem_malloc(126976): kmem_map too small: 535797760 total allocated
cpuid = 0
KDB: stack backtrace:
db_trace_self_wrapper(c07bf256) at db_trace_self_wrapper+0x25
kdb_backtrace(100,c57091b0,c4d2e3c0,0,4,...) at kdb_backtrace+0x29
panic(c07d7608,1f000,1fefa000,c109d08c,0,...) at panic+0x109
kmem_malloc(c109d08c,1f000,2,f4592bc0,c0704497,...) at kmem_malloc+0x185
page_alloc(0,1f000,f4592bb3,2) at page_alloc+0x1a
uma_large_malloc(1f000,2,190,0,e45578a0,...) at uma_large_malloc+0x3b
malloc(1e800,c5411060,2,f4592c04,c53f2a31,...) at malloc+0x99
zfs_kmem_alloc(1e800,2,f4592c40,c53e5748,1e800,...) at zfs_kmem_alloc+0x13
zio_buf_alloc(1e800,f4592c30,c53f5014,c800c678,200,...) at zio_buf_alloc+0xd
vdev_queue_io_to_issue(c5920ee4,23,0,f4592c5c,c5920f34,...) at vdev_queue_io_to_issue+0x190
vdev_queue_io_done(e48c2450,12d,f4592c88,c53e19b7,e48c2450,...) at vdev_queue_io_done+0x66
vdev_geom_io_done(e48c2450,f4592c94,c53f503e,e48c2450,f4592d00,...) at vdev_geom_io_done+0xd
vdev_io_done(e48c2450,f4592d00,c53adaa7,e48c2450,c5456a94,...) at vdev_io_done+0x13
zio_vdev_io_done(e48c2450,c5456a94,c5456a7c,c5456a94,c5456a7c,...) at zio_vdev_io_done+0x22
taskq_thread(c5456a5c,f4592d38) at taskq_thread+0x183
fork_exit(c53ad924,c5456a5c,f4592d38) at fork_exit+0x7b
fork_trampoline() at fork_trampoline+0x8
--- trap 0, eip = 0, esp = 0xf4592d70, ebp = 0 ---
Uptime: 21m33s
Physical memory: 2035 MB
Dumping 579 MB: 564 548 532 516 500 484 468 452 436 420 404 388 372 356 340 324 308 292 276 260 244 228 212 196 180 164 148 132 116 100 84 68 52 36 20 4

#0  doadump () at pcpu.h:172
172     pcpu.h: No such file or directory.
        in pcpu.h
(kgdb) quit

I set the following values in loader.conf:
vfs.zfs.zil_disable="1"
vm.kmem_size="536870912"
vfs.zfs.arc_max="67108865"

The box has 2GB of real memory.

With an older version from

FreeBSD peedub.jennejohn.org 7.0-CURRENT FreeBSD 7.0-CURRENT #11: Mon Apr
23 09:45:36 CEST 2007

I can copy 10s of gigbaytes (really 67GB) without any errors.

Any suggestions how I can work around this problem?

-- 
Gary Jennejohn / garyjATjennejohnDOTorg gjATfreebsdDOTorg garyjATdenxDOTde