2022-11-10 01:32:22

by Sean Christopherson

Subject: [PATCH] KVM: x86: Remove unnecessary export of kvm_cpu_has_pending_timer()

Don't export x86's kvm_cpu_has_pending_timer(), the helper isn't used by
KVM x86's vendor modules.

No functional change intended.

Signed-off-by: Sean Christopherson <[email protected]>
---
arch/x86/kvm/irq.c | 1 -
1 file changed, 1 deletion(-)

diff --git a/arch/x86/kvm/irq.c b/arch/x86/kvm/irq.c
index d8d50558f165..04a009fbdf8a 100644
--- a/arch/x86/kvm/irq.c
+++ b/arch/x86/kvm/irq.c
@@ -31,7 +31,6 @@ int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)

return r;
}
-EXPORT_SYMBOL(kvm_cpu_has_pending_timer);

/*
* check if there is a pending userspace external interrupt

base-commit: d663b8a285986072428a6a145e5994bc275df994
--
2.38.1.431.g37b22c650d-goog

2022-11-14 07:18:29

by Like Xu

Subject: Re: [PATCH] KVM: x86: Remove unnecessary export of kvm_cpu_has_pending_timer()

On 10/11/2022 9:03 am, Sean Christopherson wrote:
> Don't export x86's kvm_cpu_has_pending_timer(), the helper isn't used by
> KVM x86's vendor modules.
>
> No functional change intended.

The callers of EXPORT_SYMBOL (considering that this line of code has been around
for a long time) will not be public like the callers of EXPORT_SYMBOL_GPL.

Whether we close this door in the future or not, this description is inappropriate.

>
> Signed-off-by: Sean Christopherson <[email protected]>
> ---
> arch/x86/kvm/irq.c | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git a/arch/x86/kvm/irq.c b/arch/x86/kvm/irq.c
> index d8d50558f165..04a009fbdf8a 100644
> --- a/arch/x86/kvm/irq.c
> +++ b/arch/x86/kvm/irq.c
> @@ -31,7 +31,6 @@ int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
>
> return r;
> }
> -EXPORT_SYMBOL(kvm_cpu_has_pending_timer);

Not sure if the kernel's symbol or kbuild sub-system has tools to
automatically warn or turn off similar interface exposures.

>
> /*
> * check if there is a pending userspace external interrupt
>
> base-commit: d663b8a285986072428a6a145e5994bc275df994