KVM: x86/mmu: Remove unused "type" of split_page_type()

[PATCH] KVM: x86/mmu: Remove unused "type" of split_page_type()

Posted by sunliming 3 years, 10 months ago

The variable 'type' in split_page_type() is set but not used, so remove
it.

Fixes the following w1 warning:

arch/x86/kvm/mmu/mmu.c:982:28: warning: variable 'type' set but not used [-Wunused-but-set-variable]

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: sunliming <sunliming@kylinos.cn>
---
 arch/x86/kvm/mmu/mmu.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index 7b3df91a93cf..f4d577335f94 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -979,14 +979,12 @@ static void split_page_type(gfn_t gfn, struct kvm_memory_slot *slot,
 			    enum pg_level level)
 {
 	struct kvm_page_attr *page_attr = page_attr_slot(gfn, slot, level);
-	enum kvm_page_type type;
 	gfn_t base_gfn;
 
 	if (WARN_ON_ONCE(!kvm_page_type_valid(page_attr) || level <= PG_LEVEL_4K))
 		return;
 
 	base_gfn = gfn & ~(KVM_PAGES_PER_HPAGE(level) - 1);
-	type = page_attr->type;
 
 	/*
 	 * Set the type to KVM_PAGE_TYPE_MIXED in advance since when a large
-- 
2.25.1

Re: [PATCH] KVM: x86/mmu: Remove unused "type" of split_page_type()

Posted by Xiaoyao Li 3 years, 10 months ago

On 6/12/2022 11:56 AM, sunliming wrote:
> The variable 'type' in split_page_type() is set but not used, so remove
> it.
> 
> Fixes the following w1 warning:
> 
> arch/x86/kvm/mmu/mmu.c:982:28: warning: variable 'type' set but not used [-Wunused-but-set-variable]

Please note, the code doesn't get into upstream yet.

The fix shouldn't be sent to upstream maillist.

> Reported-by: kernel test robot <lkp@intel.com>
> Signed-off-by: sunliming <sunliming@kylinos.cn>
> ---
>   arch/x86/kvm/mmu/mmu.c | 2 --
>   1 file changed, 2 deletions(-)
> 
> diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
> index 7b3df91a93cf..f4d577335f94 100644
> --- a/arch/x86/kvm/mmu/mmu.c
> +++ b/arch/x86/kvm/mmu/mmu.c
> @@ -979,14 +979,12 @@ static void split_page_type(gfn_t gfn, struct kvm_memory_slot *slot,
>   			    enum pg_level level)
>   {
>   	struct kvm_page_attr *page_attr = page_attr_slot(gfn, slot, level);
> -	enum kvm_page_type type;
>   	gfn_t base_gfn;
>   
>   	if (WARN_ON_ONCE(!kvm_page_type_valid(page_attr) || level <= PG_LEVEL_4K))
>   		return;
>   
>   	base_gfn = gfn & ~(KVM_PAGES_PER_HPAGE(level) - 1);
> -	type = page_attr->type;
>   
>   	/*
>   	 * Set the type to KVM_PAGE_TYPE_MIXED in advance since when a large

Re: [PATCH] KVM: x86/mmu: Remove unused "type" of split_page_type()

Posted by sunliming 3 years, 10 months ago

I got it, I am sorry aboot it , thanks.

Xiaoyao Li <xiaoyao.li@intel.com> 于2022年6月13日周一 10:48写道：

>
> On 6/12/2022 11:56 AM, sunliming wrote:
> > The variable 'type' in split_page_type() is set but not used, so remove
> > it.
> >
> > Fixes the following w1 warning:
> >
> > arch/x86/kvm/mmu/mmu.c:982:28: warning: variable 'type' set but not used [-Wunused-but-set-variable]
>
> Please note, the code doesn't get into upstream yet.
>
> The fix shouldn't be sent to upstream maillist.
>
> > Reported-by: kernel test robot <lkp@intel.com>
> > Signed-off-by: sunliming <sunliming@kylinos.cn>
> > ---
> >   arch/x86/kvm/mmu/mmu.c | 2 --
> >   1 file changed, 2 deletions(-)
> >
> > diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
> > index 7b3df91a93cf..f4d577335f94 100644
> > --- a/arch/x86/kvm/mmu/mmu.c
> > +++ b/arch/x86/kvm/mmu/mmu.c
> > @@ -979,14 +979,12 @@ static void split_page_type(gfn_t gfn, struct kvm_memory_slot *slot,
> >                           enum pg_level level)
> >   {
> >       struct kvm_page_attr *page_attr = page_attr_slot(gfn, slot, level);
> > -     enum kvm_page_type type;
> >       gfn_t base_gfn;
> >
> >       if (WARN_ON_ONCE(!kvm_page_type_valid(page_attr) || level <= PG_LEVEL_4K))
> >               return;
> >
> >       base_gfn = gfn & ~(KVM_PAGES_PER_HPAGE(level) - 1);
> > -     type = page_attr->type;
> >
> >       /*
> >        * Set the type to KVM_PAGE_TYPE_MIXED in advance since when a large
>