Re: [PATCH v1 1/1] x86/kvm: Fix compilation warning in non-x86_64 builds

[Paolo Bonzini]

On 2/18/22 04:41, Leonardo Bras wrote:
> On non-x86_64 builds, the helper gtod_is_based_on_tsc() is defined but
> never used, which results in an warning with -Wunused-function, and
> becomes an error if -Werror is present.
>
> Add #ifdef so gtod_is_based_on_tsc() is only defined in x86_64 builds.
>
> Reported-by: kernel test robot <lkp@xxxxxxxxx>
> Signed-off-by: Leonardo Bras <leobras@xxxxxxxxxx>
> ---
>   arch/x86/kvm/x86.c | 2 ++
>   1 file changed, 2 insertions(+)
>
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index ca0fae020961..b389517aa6ed 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -2363,10 +2363,12 @@ static u64 compute_guest_tsc(struct kvm_vcpu *vcpu, s64 kernel_ns)
>   	return tsc;
>   }
>   
> +#ifdef CONFIG_X86_64
>   static inline int gtod_is_based_on_tsc(int mode)
>   {
>   	return mode == VDSO_CLOCKMODE_TSC || mode == VDSO_CLOCKMODE_HVCLOCK;
>   }
> +#endif
>   
>   static void kvm_track_tsc_matching(struct kvm_vcpu *vcpu)
>   {

More precisely, this is an "inline" in a .c (main compilation unit) 
file.  clang warns on them even though it doesn't do that on a .h 
(included) file.

I tend not to rewind kvm/master and kvm/next unless absolutely 
necessary, so I pushed an incremental fix for kvm_guest_supported_xfd() 
as well.

Paolo