amd64 MP (was Re: pae MP on cherry-xenmp)

[Jeff Rizzo <riz%netbsd.org@localhost>]

On 9/9/11 4:57 AM, Cherry G. Mathew wrote:
> Just to say I've checked in amd64 support. Do give it a spin, and let
> us know (on list) how it goes.
>
> I haven't tested this on dom0.

I haven't tried dom0 yet (hopefully in a couple hours, though).

But, I just built an MP amd64 XEN3_DOMU kernel, and got through building 
libc on a release build of src before panic:

xpq_flush_queue: 3 entries (0 successful)
0x00000001c2cba000: 0x00000001a9731007
0x00000001c625d000: 0x00000001a9731007
0x00000001a712c000: 0x00000001a9731007
panic: HYPERVISOR_mmu_update failed, ret: -22

fatal breakpoint trap in supervisor mode
trap type 1 code 0 rip ffffffff801322cd cs e030 rflags 246 cr2  40357f 
cpl 6 rsp ffffa0002a2a7620
Stopped in pid 14556.1 (groff) at       netbsd:breakpoint+0x5:  leave
breakpoint() at netbsd:breakpoint+0x5
panic() at netbsd:panic+0x2b7
xpq_update_foreign() at netbsd:xpq_update_foreign
pmap_enter_ma() at netbsd:pmap_enter_ma+0x8e3
pmap_enter() at netbsd:pmap_enter+0x5a
uvm_fault_lower() at netbsd:uvm_fault_lower+0x81d
uvm_fault_internal() at netbsd:uvm_fault_internal+0xf12
trap() at netbsd:trap+0x85d
--- trap (number 0) ---
0:
ds          7630
es          44c7
fs          7600
gs          6e81
rdi         0
rsi         d
rbp         ffffa0002a2a7620
rbx         ffffa0002a2a7630
rdx         ffffa000236c0500
rcx         8
rax         1
r8          ffffa000236c0500
r9          0
r10         ffffa0002a2a7540
r11         1
r12         104
r13         ffffffff80481138    copyright+0x410b8
r14         3
r15         0
rip         ffffffff801322cd    breakpoint+0x5
cs          e030
rflags      246
rsp         ffffa0002a2a7620
ss          e02b
netbsd:breakpoint+0x5:  leave