linux-kernel - Re: [PATCH v9 04/19] KVM: selftests: Expose function to allocate guest vCPU stack

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: <9ef0d1f4-3257-4821-8241-aedae0957c6a@linux.intel.com>
Date: Tue, 26 Aug 2025 13:39:20 +0800
From: Binbin Wu <binbin.wu@...ux.intel.com>
To: Sagi Shahar <sagis@...gle.com>
Cc: linux-kselftest@...r.kernel.org, Paolo Bonzini <pbonzini@...hat.com>,
 Shuah Khan <shuah@...nel.org>, Sean Christopherson <seanjc@...gle.com>,
 Ackerley Tng <ackerleytng@...gle.com>, Ryan Afranji <afranji@...gle.com>,
 Andrew Jones <ajones@...tanamicro.com>,
 Isaku Yamahata <isaku.yamahata@...el.com>,
 Erdem Aktas <erdemaktas@...gle.com>,
 Rick Edgecombe <rick.p.edgecombe@...el.com>,
 Roger Wang <runanwang@...gle.com>, Oliver Upton <oliver.upton@...ux.dev>,
 "Pratik R. Sampat" <pratikrajesh.sampat@....com>,
 Reinette Chatre <reinette.chatre@...el.com>, Ira Weiny
 <ira.weiny@...el.com>, Chao Gao <chao.gao@...el.com>,
 Chenyi Qiang <chenyi.qiang@...el.com>, linux-kernel@...r.kernel.org,
 kvm@...r.kernel.org
Subject: Re: [PATCH v9 04/19] KVM: selftests: Expose function to allocate
 guest vCPU stack



On 8/21/2025 12:28 PM, Sagi Shahar wrote:
> TDX guests' registers cannot be initialized directly using
> vcpu_regs_set(), hence the stack pointer needs to be initialized by
> the guest itself, running boot code beginning at the reset vector.
>
> Expose the function to allocate the guest stack so that TDX
> initialization code can allocate it itself and skip the allocation in
> vm_arch_vcpu_add() in that case.
>
> Signed-off-by: Sagi Shahar <sagis@...gle.com>
> ---
>   .../selftests/kvm/include/x86/processor.h       |  2 ++
>   tools/testing/selftests/kvm/lib/x86/processor.c | 17 ++++++++++++-----
>   2 files changed, 14 insertions(+), 5 deletions(-)
>
> diff --git a/tools/testing/selftests/kvm/include/x86/processor.h b/tools/testing/selftests/kvm/include/x86/processor.h
> index 5c16507f9b2d..8fcc5118683e 100644
> --- a/tools/testing/selftests/kvm/include/x86/processor.h
> +++ b/tools/testing/selftests/kvm/include/x86/processor.h
> @@ -1111,6 +1111,8 @@ static inline void vcpu_clear_cpuid_feature(struct kvm_vcpu *vcpu,
>   	vcpu_set_or_clear_cpuid_feature(vcpu, feature, false);
>   }
>   
> +vm_vaddr_t kvm_allocate_vcpu_stack(struct kvm_vm *vm);
> +
>   uint64_t vcpu_get_msr(struct kvm_vcpu *vcpu, uint64_t msr_index);
>   int _vcpu_set_msr(struct kvm_vcpu *vcpu, uint64_t msr_index, uint64_t msr_value);
>   
> diff --git a/tools/testing/selftests/kvm/lib/x86/processor.c b/tools/testing/selftests/kvm/lib/x86/processor.c
> index b2a4b11ac8c0..1eae92957456 100644
> --- a/tools/testing/selftests/kvm/lib/x86/processor.c
> +++ b/tools/testing/selftests/kvm/lib/x86/processor.c
> @@ -687,12 +687,9 @@ void vcpu_arch_set_entry_point(struct kvm_vcpu *vcpu, void *guest_code)
>   	vcpu_regs_set(vcpu, &regs);
>   }
>   
> -struct kvm_vcpu *vm_arch_vcpu_add(struct kvm_vm *vm, uint32_t vcpu_id)
> +vm_vaddr_t kvm_allocate_vcpu_stack(struct kvm_vm *vm)
>   {
> -	struct kvm_mp_state mp_state;
> -	struct kvm_regs regs;
>   	vm_vaddr_t stack_vaddr;
> -	struct kvm_vcpu *vcpu;
>   
>   	stack_vaddr = __vm_vaddr_alloc(vm, DEFAULT_STACK_PGS * getpagesize(),
>   				       DEFAULT_GUEST_STACK_VADDR_MIN,
> @@ -713,6 +710,15 @@ struct kvm_vcpu *vm_arch_vcpu_add(struct kvm_vm *vm, uint32_t vcpu_id)
>   		    "__vm_vaddr_alloc() did not provide a page-aligned address");
>   	stack_vaddr -= 8;
>   
> +	return stack_vaddr;
> +}
> +
> +struct kvm_vcpu *vm_arch_vcpu_add(struct kvm_vm *vm, uint32_t vcpu_id)
> +{
> +	struct kvm_mp_state mp_state;
> +	struct kvm_regs regs;
> +	struct kvm_vcpu *vcpu;
> +
>   	vcpu = __vm_vcpu_add(vm, vcpu_id);
>   	vcpu_init_cpuid(vcpu, kvm_get_supported_cpuid());
>   	vcpu_init_sregs(vm, vcpu);
> @@ -721,7 +727,8 @@ struct kvm_vcpu *vm_arch_vcpu_add(struct kvm_vm *vm, uint32_t vcpu_id)
>   	/* Setup guest general purpose registers */
>   	vcpu_regs_get(vcpu, &regs);
>   	regs.rflags = regs.rflags | 0x2;
> -	regs.rsp = stack_vaddr;
> +	if (vm->type != KVM_X86_TDX_VM)
> +		regs.rsp = kvm_allocate_vcpu_stack(vm);

I am wondering if this could be more generic.
I.e, make vcpu_regs_get() return the error code.
If vcpu_regs_get() failed (for TDX, since it's guest state is protected, the
ioctl will return -EINVAL), the vcpu_regs_set(), including the allocation for
the vcpu stack, could be skipped.

>   	vcpu_regs_set(vcpu, &regs);
>   
>   	/* Setup the MP state */