failures with if_em on txcsum/rxcsum enabled

I've been working on a number of problems related to the txcsum/ 
rxcsum options on if_em and if_mxge and with that stuff enabled while  
talking to a large range of HTTP hosts, it's causing crashes.  It  
seems pretty consistent across the board with varying symptoms.  This  
was built from today's tree:


em0: discard frame w/o packet header


Fatal trap 12: page fault while in kernel mode
cpuid = 0; apic id = 00
fault virtual address   = 0xc
fault code              = supervisor read, page not present
instruction pointer     = 0x20:0xc06fd90c
stack pointer           = 0x28:0xe38bc9b8
frame pointer           = 0x28:0xe38bc9e0
code segment            = base 0x0, limit 0xfffff, type 0x1b
                         = DPL 0, pres 1, def32 1, gran 1
processor eflags        = interrupt enabled, resume, IOPL = 0
current process         = 26 (em0 taskq)
trap number             = 12
panic: page fault
cpuid = 0
Uptime: 6m41s
Physical memory: 2039 MB
Dumping 245 MB:

Fatal trap 12: page fault while in kernel mode
cpuid = 0; apic id = 00
fault virtual address   = 0x200
fault code              = supervisor read, page not present
instruction pointer     = 0x20:0x200
stack pointer           = 0x28:0xe38bc500
frame pointer           = 0x28:0xe38bc510
code segment            = base 0x0, limit 0xfffff, type 0x1b
                         = DPL 0, pres 1, def32 1, gran 1
processor eflags        = interrupt enabled, resume, IOPL = 0
current process         = 26 (em0 taskq)
trap number             = 12
panic: page fault
cpuid = 0
Uptime: 6m42s
Automatic reboot in 15 seconds - press a key on the console to abort
Rebooting...
cpu_reset: Stopping other CPUs