K 10
svn:author
V 10
nwhitehorn
K 8
svn:date
V 27
2012-03-03T02:19:33.190407Z
K 7
svn:log
V 730
MFC r230123,230139:

Rework SLB trap handling so that double-faults into an SLB trap handler are
possible, and double faults within an SLB trap handler are not. The result
is that it possible to take an SLB fault at any time, on any address, for
any reason, at any point in the kernel.

This lets us do two important things. First, it removes the (soft) 16 GB RAM
ceiling on PPC64 as well as any architectural limitations on KVA space.
Second, it lets the kernel tolerate poorly designed hypervisors that
have a tendency to fail to restore the SLB properly after a hypervisor
context switch.

Now that we can tolerate LPAR context switches on the PS3 hypervisor, going
to hypervisor-idle on both threads will not hang the kernel.

END