4G/4G split on x86, 64 GB RAM TLB flush

embeddedlwp

http://lwn.net/Articles/39283/

Performance impact of the 4G/4G feature:

There's a runtime cost with the 4G/4G patch: to implement separate address
spaces for the kernel and userspace VM, the entry/exit code has to switch
between the kernel pagetables and the user pagetables. This causes TLB
flushes, which are quite expensive, not so much in terms of TLB misses
(which are quite fast on Intel CPUs if they come from caches), but in
terms of the direct TLB flushing cost (%cr3 manipulation) done on
system-entry.

这里提到TLB flush，那么2G/2G split呢？是不是跟之间 3G/1G 差不多了？