Could not the system just release zero'd pages altogether until someone writes something in them, rather sharing them amongst KVM instances? Seems easier and less costly to do than the SHA1 thing and seems (in Windows' case) to yield some memory economies?
Especially if it's freed memory...
Or did I miss something?
Copyright © 2017, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds