----- Original Message ----- From: "Steve Reinhardt" <ste...@gmail.com>
To: "Gabriel Michael Black" <gbl...@eecs.umich.edu> Cc: "M5 Developer List" <m5-dev@m5sim.org> Sent: 2. јул 2010 0:45 Subject: Re: [m5-dev] X86_FS vtophys implementation
I can't think of a legitimate situation where the TLB would differ from the page tables with the possible exception of permissions (which I just argued shouldn't matter) or perhaps the unlucky situation of doing a functional access in the middle of a TLB shootdown (though even there, I think the shootdown has to happen before the page table change, so you're probably OK). Steve
I'm not shure if this situation can arise in x86 (it can on MIPS), but what about case when page tabels are changed in memory and TLB has old copy? I guess TLB doesn't snoop writes to memory where page tables reside. So, some of (faster case) or all (simpler case) entries in TLB should be invalidated to get fresh copy from memory.
_______________________________________________ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev