linux-kernel - Re: [PATCH 04/22] KVM: x86/mmu: Skip emulation on page fault iff 1+ SPs were unprotected

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <Zszlab0BMPKmnOsy@google.com>
Date: Mon, 26 Aug 2024 13:28:25 -0700
From: Sean Christopherson <seanjc@...gle.com>
To: Yao Yuan <yaoyuan0329os@...il.com>
Cc: Yuan Yao <yuan.yao@...ux.intel.com>, Paolo Bonzini <pbonzini@...hat.com>, kvm@...r.kernel.org, 
	linux-kernel@...r.kernel.org, Peter Gonda <pgonda@...gle.com>, 
	Michael Roth <michael.roth@....com>, Vishal Annapurve <vannapurve@...gle.com>, 
	Ackerly Tng <ackerleytng@...gle.com>
Subject: Re: [PATCH 04/22] KVM: x86/mmu: Skip emulation on page fault iff 1+
 SPs were unprotected

On Fri, Aug 23, 2024, Sean Christopherson wrote:
> On Thu, Aug 22, 2024, Sean Christopherson wrote:
> Intel support is especially misleading, because sharing page tables between EPT
> and IA32 is rather nonsensical due to them having different formats.  I.e. I doubt
> Paolo had a use case for the VMX changes, and was just providing parity with SVM.
> Of course, reusing L1's page tables as the NPT tables for L2 is quite crazy too,

Actually, it's not _that_ crazy, e.g. KVM s390 does this for last-level page tables
so that changes to the host userspace mappings don't require mmu_notifier-induced
changes in KVM.

> but at least the PTE formats are identical.