linux-kernel - Re: [PATCH v2 3/8] KVM: TDX: Add a place holder for handler of TDX hypercalls (TDG.VP.VMCALL)

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: <175a8914-67ec-4b14-972d-98e9a1122daf@linux.intel.com>
Date: Tue, 11 Feb 2025 17:08:07 +0800
From: Binbin Wu <binbin.wu@...ux.intel.com>
To: Chao Gao <chao.gao@...el.com>
Cc: pbonzini@...hat.com, seanjc@...gle.com, kvm@...r.kernel.org,
 rick.p.edgecombe@...el.com, kai.huang@...el.com, adrian.hunter@...el.com,
 reinette.chatre@...el.com, xiaoyao.li@...el.com, tony.lindgren@...el.com,
 isaku.yamahata@...el.com, yan.y.zhao@...el.com, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 3/8] KVM: TDX: Add a place holder for handler of TDX
 hypercalls (TDG.VP.VMCALL)



On 2/11/2025 4:41 PM, Chao Gao wrote:
>> +static __always_inline unsigned long tdvmcall_exit_type(struct kvm_vcpu *vcpu)
>> +{
>> +	return to_tdx(vcpu)->vp_enter_args.r10;
>> +}
> please add a newline here.
>
>> +static __always_inline unsigned long tdvmcall_leaf(struct kvm_vcpu *vcpu)
>> +{
>> +	return to_tdx(vcpu)->vp_enter_args.r11;
>> +}
> ..
>
>> +static __always_inline void tdvmcall_set_return_code(struct kvm_vcpu *vcpu,
>> +						     long val)
>> +{
>> +	to_tdx(vcpu)->vp_enter_args.r10 = val;
>> +}
> ditto.
>
>> +static __always_inline void tdvmcall_set_return_val(struct kvm_vcpu *vcpu,
>> +						    unsigned long val)
>> +{
>> +	to_tdx(vcpu)->vp_enter_args.r11 = val;
>> +}
>> +
>> static inline void tdx_hkid_free(struct kvm_tdx *kvm_tdx)
>> {
>> 	tdx_guest_keyid_free(kvm_tdx->hkid);
>> @@ -810,6 +829,7 @@ static bool tdx_guest_state_is_invalid(struct kvm_vcpu *vcpu)
>> static __always_inline u32 tdx_to_vmx_exit_reason(struct kvm_vcpu *vcpu)
>> {
>> 	struct vcpu_tdx *tdx = to_tdx(vcpu);
>> +	u32 exit_reason;
>>
>> 	switch (tdx->vp_enter_ret & TDX_SEAMCALL_STATUS_MASK) {
>> 	case TDX_SUCCESS:
>> @@ -822,7 +842,21 @@ static __always_inline u32 tdx_to_vmx_exit_reason(struct kvm_vcpu *vcpu)
>> 		return -1u;
>> 	}
>>
>> -	return tdx->vp_enter_ret;
>> +	exit_reason = tdx->vp_enter_ret;
>> +
>> +	switch (exit_reason) {
>> +	case EXIT_REASON_TDCALL:
>> +		if (tdvmcall_exit_type(vcpu))
>> +			return EXIT_REASON_VMCALL;
>> +
>> +		if (tdvmcall_leaf(vcpu) < 0x10000)
> Can you add a comment for the hard-coded 0x10000?
>
> I am wondering what would happen if the guest tries to make a tdvmcall with
> leaf=0 or leaf=1 to mislead KVM into calling the NMI/interrupt handling
> routine. Would it trigger the unknown NMI warning or effectively inject an
> interrupt into the host?
Oh, yes, it's possible.

>
> I think we should do the conversion for leafs that are defined in the current
> GHCI spec.
Yes, it should be limited to the supported leaves defined in the GHCI.
Thanks for pointing it out!

>
>> +			return tdvmcall_leaf(vcpu);
>> +		break;
>> +	default:
>> +		break;
>> +	}
>> +
>> +	return exit_reason;
>> }