linux-kernel - Re: [PATCH next V2] KVM: VMX: use __always_inline for is_td_vcpu and is

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: <20250527110752.GB20019@noisy.programming.kicks-ass.net>
Date: Tue, 27 May 2025 13:07:52 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Edward Adam Davis <eadavis@...com>
Cc: binbin.wu@...ux.intel.com, bp@...en8.de, dave.hansen@...ux.intel.com,
	hpa@...or.com, kvm@...r.kernel.org, linux-kernel@...r.kernel.org,
	mingo@...hat.com, pbonzini@...hat.com, seanjc@...gle.com,
	tglx@...utronix.de, x86@...nel.org
Subject: Re: [PATCH next V2] KVM: VMX: use __always_inline for is_td_vcpu and
 is_td

On Tue, May 27, 2025 at 04:44:37PM +0800, Edward Adam Davis wrote:
> is_td() and is_td_vcpu() run in no instrumentation, so use __always_inline
> to replace inline.
> 
> [1]
> vmlinux.o: error: objtool: vmx_handle_nmi+0x47:
>         call to is_td_vcpu.isra.0() leaves .noinstr.text section
> 
> Fixes: 7172c753c26a ("KVM: VMX: Move common fields of struct vcpu_{vmx,tdx} to a struct")
> Signed-off-by: Edward Adam Davis <eadavis@...com>
> ---
> V1 -> V2: using __always_inline to replace noinstr
> 
>  arch/x86/kvm/vmx/common.h | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/x86/kvm/vmx/common.h b/arch/x86/kvm/vmx/common.h
> index 8f46a06e2c44..a0c5e8781c33 100644
> --- a/arch/x86/kvm/vmx/common.h
> +++ b/arch/x86/kvm/vmx/common.h
> @@ -71,8 +71,8 @@ static __always_inline bool is_td_vcpu(struct kvm_vcpu *vcpu)
>  
>  #else
>  
> -static inline bool is_td(struct kvm *kvm) { return false; }
> -static inline bool is_td_vcpu(struct kvm_vcpu *vcpu) { return false; }
> +static __always_inline bool is_td(struct kvm *kvm) { return false; }
> +static __always_inline bool is_td_vcpu(struct kvm_vcpu *vcpu) { return false; }
>  
>  #endif

Right; this is the 'right' fix. Although the better fix would be for the
compiler to not be stupid :-)