panic: UMA: Increase vm.boot_pages on Dell R920 r279210

I (temporarily) have a Dell R920 with 1T RAM, and 4 CPUs with 15
cores.  I get a kernel panic when attempting to boot
FreeBSD-11.0-CURRENT-amd64-20150223-r279210-memstick.img:

...
Dell System PowerEdge R920
www.dell.com
...
Four 2.30 GHz Fifteen-core Processors, L2/L3 Cache:3.75 MB/30 MB
System running at 2.30 GHz
System Bus Speed: 8.00 GT/s
System Memory Size: 1024.0 GB, System Memory Speed: 1333 MHz, Voltage: 1.35V
...
Booting...
GDB: no debug ports present
KDB: debugger backends: ddb
KDB: current backend: ddb
Copyright (c) 1992-2015 The FreeBSD Project.
Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
         The Regents of the University of California. All rights reserved.
FreeBSD is a registered trademark of The FreeBSD Foundation.
FreeBSD 11.0-CURRENT #0 r279210: Mon Feb 23 19:14:12 UTC 2015
     root_at_releng2.nyi.freebsd.org:/usr/obj/usr/src/sys/GENERIC amd64
FreeBSD clang version 3.5.1 (tags/RELEASE_351/final 225668) 20150115
WARNING: WITNESS option enabled, expect reduced performance.
panic: UMA: Increase vm.boot_pages
cpuid = 0
KDB: stack backtrace:
db_trace_self_wrapper() at db_trace_self_wrapper+0x2b/frame 0xffffffff81bfe7a0
vpanic() at vpanic+0x189/frame 0xffffffff81bfe820
panic() at panic+0x43/frame 0xffffffff81bfe880
startup_alloc() at startup_alloc+0x1da/frame 0xffffffff81bfe8d0
keg_alloc_slab() at keg_alloc_slab+0xce/frame 0xffffffff81bfe930
keg_fetch_slab() at keg_fetch_slab+0x171/frame 0xffffffff81bfe980
zone_fetch_slab() at zone_fetch_slab+0x6e/frame 0xffffffff81bfe9c0
zone_import() at zone_import+0x40/frame 0xffffffff81bfea30
zone_alloc_item() at zone_alloc_item+0x36/frame 0xffffffff81bfea70
uma_zcreate() at uma_zcreate+0x8a/frame 0xffffffff81bfeb10
vm_map_startup() at vm_map_startup+0x52/frame 0xffffffff81bfeb30
vm_mem_init() at vm_mem_init+0x31/frame 0xffffffff81bfeb50
mi_startup() at mi_startup+0x118/frame 0xffffffff81bfeb70
btext() at btext+0x2c
KDB: enter: panic
[ thread pid 0 tid 0 ]
Stopped at      kdb_enter+0x3e: movq    $0,kdb_why
db>

I tried the suggestion "Increase vm.boot_pages" but am unsure how
much.  I tried doubling to 128, and even excessive(?) values like
102400; but got the same panic.

If, however, I reduce the number of active cores to 10 in the BIOS,
I am able to boot with the original value of vm.boot_pages.

I have serial console and sneaker net access for the next couple
of days.  Any suggestions?

...keith
-- 
Keith White, genie.uottawa.ca engineering.uottawa.ca
kwhite_at_uottawa.ca [+1 613 562 5800 x6681]