linux-kernel - Re: [PATCH] KVM: SVM: Clear the CR4 register on reset

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]

Date:   Tue, 2 Mar 2021 11:20:17 -0800
From:   Sean Christopherson <seanjc@...gle.com>
To:     Babu Moger <babu.moger@....com>
Cc:     pbonzini@...hat.com, wanpengli@...cent.com, kvm@...r.kernel.org,
        joro@...tes.org, x86@...nel.org, linux-kernel@...r.kernel.org,
        mingo@...hat.com, bp@...en8.de, hpa@...or.com, vkuznets@...hat.com,
        tglx@...utronix.de, jmattson@...gle.com
Subject: Re: [PATCH] KVM: SVM: Clear the CR4 register on reset

On Tue, Mar 02, 2021, Babu Moger wrote:
> This problem was reported on a SVM guest while executing kexec.
> Kexec fails to load the new kernel when the PCID feature is enabled.
> 
> When kexec starts loading the new kernel, it starts the process by
> resetting the vCPU's and then bringing each vCPU online one by one.
> The vCPU reset is supposed to reset all the register states before the
> vCPUs are brought online. However, the CR4 register is not reset during
> this process. If this register is already setup during the last boot,
> all the flags can remain intact. The X86_CR4_PCIDE bit can only be
> enabled in long mode. So, it must be enabled much later in SMP
> initialization.  Having the X86_CR4_PCIDE bit set during SMP boot can
> cause a boot failures.
> 
> Fix the issue by resetting the CR4 register in init_vmcb().
> 
> Signed-off-by: Babu Moger <babu.moger@....com>

Cc: stable@...r.kernel.org

The bug goes back too far to have a meaningful Fixes.

Reviewed-by: Sean Christopherson <seanjc@...gle.com>


On a related topic, I think we can clean up the RESET/INIT flows by hoisting the
common code into kvm_vcpu_reset().  That would also provide good motivation for
removing the init_vmcb() call in svm_create_vcpu(), which is fully redundant
with the call in svm_vcpu_reset().  I'll put that on the todo list.