Comment 10 for bug 1838575

Now to gain a good result, lets use 1G Huge Pages.
Kernel cmdline:
  default_hugepagesz=1G hugepagesz=1G hugepages=1210

Gives:
HugePages_Total: 1210
HugePages_Free: 1210
HugePages_Rsvd: 0
HugePages_Surp: 0
Hugepagesize: 1048576 kB

Guest config extra:
 <memoryBacking>
   <hugepages/>
 </memoryBacking>

Slightly changed profile:

Samples: 28K of event 'cycles:ppp', Event count (approx.): 19134377327
Overhead Shared Object Symbol
  27.02% [kernel] [k] follow_hugetlb_page
  11.23% [kernel] [k] __get_user_pages
   9.10% [kernel] [k] _raw_spin_lock
   7.58% [kernel] [k] down_read
   7.53% [kernel] [k] up_read
   6.14% [kernel] [k] huge_pte_offset

T4: use 1.2 TB with one PT device, 1G sized explicit HugePages
  #1: 59 seconds
  #2: 21 seconds
  #3: 18 seconds

This is what you'd expect from such a huge guest, and it works just as expected.