[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <57155925.2080009@amd.com>
Date: Mon, 18 Apr 2016 17:01:09 -0500
From: Suravee Suthikulpanit <Suravee.Suthikulpanit@....com>
To: Paolo Bonzini <pbonzini@...hat.com>,
Radim Krčmář <rkrcmar@...hat.com>
CC: <joro@...tes.org>, <bp@...en8.de>, <gleb@...nel.org>,
<alex.williamson@...hat.com>, <kvm@...r.kernel.org>,
<linux-kernel@...r.kernel.org>, <wei@...hat.com>,
<sherry.hurwitz@....com>
Subject: Re: [PART1 RFC v4 02/11] KVM: x86: Introducing kvm_x86_ops VM
init/uninit hooks
Paolo,
On 04/12/2016 04:55 PM, Paolo Bonzini wrote:
>
>
> On 11/04/2016 22:49, Radim Krčmář wrote:
>>>> @@ -7781,6 +7784,9 @@ static void kvm_free_vcpus(struct kvm *kvm)
>>>> kvm_for_each_vcpu(i, vcpu, kvm)
>>>> kvm_arch_vcpu_free(vcpu);
>>>>
>>>> + if (kvm_x86_ops->vm_uninit)
>>>> + kvm_x86_ops->vm_uninit(kvm);
>> vm_uninit() doesn't seem to have much to do with kvm_free_vcpus(),
>> please call it from kvm_arch_destroy_vm().
>>
>> (kvm_x86_ops.vm_destroy would be a better name then.)
>
> Especially, you're calling it with struct kvm full of dangling pointer,
> so please call it early, right after the "if (current->mm == kvm->mm)"
> block.
>
> Paolo
Good point.
Thanks,
Suravee
Powered by blists - more mailing lists