| Message ID | 20200201185218.24473-27-sean.j.christopherson@intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | KVM: x86: Introduce KVM cpu caps | expand |
On 2/2/2020 2:51 AM, Sean Christopherson wrote: > Introduce accessors to retrieve feature bits from CPUID entries and use > the new accessors where applicable. Using the accessors eliminates the > need to manually specify the register to be queried at no extra cost > (binary output is identical) and will allow adding runtime consistency > checks on the function and index in a future patch. > > No functional change intended. > > Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> > --- > arch/x86/kvm/cpuid.c | 9 +++++---- > arch/x86/kvm/cpuid.h | 46 +++++++++++++++++++++++++++++++++++--------- > 2 files changed, 42 insertions(+), 13 deletions(-) > > diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c > index e3026fe638aa..3316963dad3d 100644 > --- a/arch/x86/kvm/cpuid.c > +++ b/arch/x86/kvm/cpuid.c > @@ -68,7 +68,7 @@ int kvm_update_cpuid(struct kvm_vcpu *vcpu) > best->edx |= F(APIC); > > if (apic) { > - if (best->ecx & F(TSC_DEADLINE_TIMER)) > + if (cpuid_entry_has(best, X86_FEATURE_TSC_DEADLINE_TIMER)) > apic->lapic_timer.timer_mode_mask = 3 << 17; > else > apic->lapic_timer.timer_mode_mask = 1 << 17; > @@ -96,7 +96,8 @@ int kvm_update_cpuid(struct kvm_vcpu *vcpu) > } > > best = kvm_find_cpuid_entry(vcpu, 0xD, 1); > - if (best && (best->eax & (F(XSAVES) | F(XSAVEC)))) > + if (best && (cpuid_entry_has(best, X86_FEATURE_XSAVES) || > + cpuid_entry_has(best, X86_FEATURE_XSAVEC))) > best->ebx = xstate_required_size(vcpu->arch.xcr0, true); > > /* > @@ -155,7 +156,7 @@ static void cpuid_fix_nx_cap(struct kvm_vcpu *vcpu) > break; > } > } > - if (entry && (entry->edx & F(NX)) && !is_efer_nx()) { > + if (entry && cpuid_entry_has(entry, X86_FEATURE_NX) && !is_efer_nx()) { > entry->edx &= ~F(NX); > printk(KERN_INFO "kvm: guest NX capability removed\n"); > } > @@ -387,7 +388,7 @@ static inline void do_cpuid_7_mask(struct kvm_cpuid_entry2 *entry) > entry->ebx |= F(TSC_ADJUST); > > entry->ecx &= kvm_cpuid_7_0_ecx_x86_features; > - f_la57 = entry->ecx & F(LA57); > + f_la57 = cpuid_entry_get(entry, X86_FEATURE_LA57); > cpuid_mask(&entry->ecx, CPUID_7_ECX); > /* Set LA57 based on hardware capability. */ > entry->ecx |= f_la57; > diff --git a/arch/x86/kvm/cpuid.h b/arch/x86/kvm/cpuid.h > index 72a79bdfed6b..64e96e4086e2 100644 > --- a/arch/x86/kvm/cpuid.h > +++ b/arch/x86/kvm/cpuid.h > @@ -95,16 +95,10 @@ static __always_inline struct cpuid_reg x86_feature_cpuid(unsigned x86_feature) > return reverse_cpuid[x86_leaf]; > } > > -static __always_inline u32 *guest_cpuid_get_register(struct kvm_vcpu *vcpu, unsigned x86_feature) > +static __always_inline u32 *__cpuid_entry_get_reg(struct kvm_cpuid_entry2 *entry, > + const struct cpuid_reg *cpuid) > { > - struct kvm_cpuid_entry2 *entry; > - const struct cpuid_reg cpuid = x86_feature_cpuid(x86_feature); > - > - entry = kvm_find_cpuid_entry(vcpu, cpuid.function, cpuid.index); > - if (!entry) > - return NULL; > - > - switch (cpuid.reg) { > + switch (cpuid->reg) { > case CPUID_EAX: > return &entry->eax; > case CPUID_EBX: > @@ -119,6 +113,40 @@ static __always_inline u32 *guest_cpuid_get_register(struct kvm_vcpu *vcpu, unsi > } > } > > +static __always_inline u32 *cpuid_entry_get_reg(struct kvm_cpuid_entry2 *entry, > + unsigned x86_feature) > +{ > + const struct cpuid_reg cpuid = x86_feature_cpuid(x86_feature); > + > + return __cpuid_entry_get_reg(entry, &cpuid); > +} > + > +static __always_inline u32 cpuid_entry_get(struct kvm_cpuid_entry2 *entry, > + unsigned x86_feature) > +{ > + u32 *reg = cpuid_entry_get_reg(entry, x86_feature); > + > + return *reg & __feature_bit(x86_feature); > +} > + This helper function is unnecessary. There is only one user throughout this series, i.e., cpuid_entry_has() below. And I cannot image other possible use case of it. > +static __always_inline bool cpuid_entry_has(struct kvm_cpuid_entry2 *entry, > + unsigned x86_feature) > +{ > + return cpuid_entry_get(entry, x86_feature); > +} > + > +static __always_inline int *guest_cpuid_get_register(struct kvm_vcpu *vcpu, unsigned x86_feature) ^ Should be u32 otherwise, previous patch will be unhappy. :) > +{ > + struct kvm_cpuid_entry2 *entry; > + const struct cpuid_reg cpuid = x86_feature_cpuid(x86_feature); > + > + entry = kvm_find_cpuid_entry(vcpu, cpuid.function, cpuid.index); > + if (!entry) > + return NULL; > + > + return __cpuid_entry_get_reg(entry, &cpuid); > +} > + > static __always_inline bool guest_cpuid_has(struct kvm_vcpu *vcpu, unsigned x86_feature) > { > u32 *reg; >
On Fri, Feb 14, 2020 at 05:44:41PM +0800, Xiaoyao Li wrote: > On 2/2/2020 2:51 AM, Sean Christopherson wrote: > >@@ -387,7 +388,7 @@ static inline void do_cpuid_7_mask(struct kvm_cpuid_entry2 *entry) > > entry->ebx |= F(TSC_ADJUST); > > entry->ecx &= kvm_cpuid_7_0_ecx_x86_features; > >- f_la57 = entry->ecx & F(LA57); > >+ f_la57 = cpuid_entry_get(entry, X86_FEATURE_LA57); Note, cpuid_entry_get() is used here. > > cpuid_mask(&entry->ecx, CPUID_7_ECX); > > /* Set LA57 based on hardware capability. */ > > entry->ecx |= f_la57; > >diff --git a/arch/x86/kvm/cpuid.h b/arch/x86/kvm/cpuid.h > >index 72a79bdfed6b..64e96e4086e2 100644 > >--- a/arch/x86/kvm/cpuid.h > >+++ b/arch/x86/kvm/cpuid.h > >@@ -95,16 +95,10 @@ static __always_inline struct cpuid_reg x86_feature_cpuid(unsigned x86_feature) > > return reverse_cpuid[x86_leaf]; > > } > >-static __always_inline u32 *guest_cpuid_get_register(struct kvm_vcpu *vcpu, unsigned x86_feature) > >+static __always_inline u32 *__cpuid_entry_get_reg(struct kvm_cpuid_entry2 *entry, > >+ const struct cpuid_reg *cpuid) > > { > >- struct kvm_cpuid_entry2 *entry; > >- const struct cpuid_reg cpuid = x86_feature_cpuid(x86_feature); > >- > >- entry = kvm_find_cpuid_entry(vcpu, cpuid.function, cpuid.index); > >- if (!entry) > >- return NULL; > >- > >- switch (cpuid.reg) { > >+ switch (cpuid->reg) { > > case CPUID_EAX: > > return &entry->eax; > > case CPUID_EBX: > >@@ -119,6 +113,40 @@ static __always_inline u32 *guest_cpuid_get_register(struct kvm_vcpu *vcpu, unsi > > } > > } > >+static __always_inline u32 *cpuid_entry_get_reg(struct kvm_cpuid_entry2 *entry, > >+ unsigned x86_feature) > >+{ > >+ const struct cpuid_reg cpuid = x86_feature_cpuid(x86_feature); > >+ > >+ return __cpuid_entry_get_reg(entry, &cpuid); > >+} > >+ > >+static __always_inline u32 cpuid_entry_get(struct kvm_cpuid_entry2 *entry, > >+ unsigned x86_feature) > >+{ > >+ u32 *reg = cpuid_entry_get_reg(entry, x86_feature); > >+ > >+ return *reg & __feature_bit(x86_feature); > >+} > >+ > > This helper function is unnecessary. There is only one user throughout this > series, i.e., cpuid_entry_has() below. And the LA57 case above. > And I cannot image other possible use case of it. The LA57 case, which admittedly goes away soon, was subtle enough (OR in the flag instead of querying yes/no) that I wanted keep the accessor around in case a similar case popped up in the future. > >+static __always_inline bool cpuid_entry_has(struct kvm_cpuid_entry2 *entry, > >+ unsigned x86_feature) > >+{ > >+ return cpuid_entry_get(entry, x86_feature); > >+} > >+ > >+static __always_inline int *guest_cpuid_get_register(struct kvm_vcpu *vcpu, unsigned x86_feature) > ^ > Should be u32 > otherwise, previous patch will be unhappy. :) Doh, thanks! > >+{ > >+ struct kvm_cpuid_entry2 *entry; > >+ const struct cpuid_reg cpuid = x86_feature_cpuid(x86_feature); > >+ > >+ entry = kvm_find_cpuid_entry(vcpu, cpuid.function, cpuid.index); > >+ if (!entry) > >+ return NULL; > >+ > >+ return __cpuid_entry_get_reg(entry, &cpuid); > >+} > >+ > > static __always_inline bool guest_cpuid_has(struct kvm_vcpu *vcpu, unsigned x86_feature) > > { > > u32 *reg; > > >
Sean Christopherson <sean.j.christopherson@intel.com> writes: > Introduce accessors to retrieve feature bits from CPUID entries and use > the new accessors where applicable. Using the accessors eliminates the > need to manually specify the register to be queried at no extra cost > (binary output is identical) and will allow adding runtime consistency > checks on the function and index in a future patch. > > No functional change intended. > > Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> > --- > arch/x86/kvm/cpuid.c | 9 +++++---- > arch/x86/kvm/cpuid.h | 46 +++++++++++++++++++++++++++++++++++--------- > 2 files changed, 42 insertions(+), 13 deletions(-) > > diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c > index e3026fe638aa..3316963dad3d 100644 > --- a/arch/x86/kvm/cpuid.c > +++ b/arch/x86/kvm/cpuid.c > @@ -68,7 +68,7 @@ int kvm_update_cpuid(struct kvm_vcpu *vcpu) > best->edx |= F(APIC); > > if (apic) { > - if (best->ecx & F(TSC_DEADLINE_TIMER)) > + if (cpuid_entry_has(best, X86_FEATURE_TSC_DEADLINE_TIMER)) > apic->lapic_timer.timer_mode_mask = 3 << 17; > else > apic->lapic_timer.timer_mode_mask = 1 << 17; > @@ -96,7 +96,8 @@ int kvm_update_cpuid(struct kvm_vcpu *vcpu) > } > > best = kvm_find_cpuid_entry(vcpu, 0xD, 1); > - if (best && (best->eax & (F(XSAVES) | F(XSAVEC)))) > + if (best && (cpuid_entry_has(best, X86_FEATURE_XSAVES) || > + cpuid_entry_has(best, X86_FEATURE_XSAVEC))) > best->ebx = xstate_required_size(vcpu->arch.xcr0, true); > > /* > @@ -155,7 +156,7 @@ static void cpuid_fix_nx_cap(struct kvm_vcpu *vcpu) > break; > } > } > - if (entry && (entry->edx & F(NX)) && !is_efer_nx()) { > + if (entry && cpuid_entry_has(entry, X86_FEATURE_NX) && !is_efer_nx()) { > entry->edx &= ~F(NX); > printk(KERN_INFO "kvm: guest NX capability removed\n"); > } > @@ -387,7 +388,7 @@ static inline void do_cpuid_7_mask(struct kvm_cpuid_entry2 *entry) > entry->ebx |= F(TSC_ADJUST); > > entry->ecx &= kvm_cpuid_7_0_ecx_x86_features; > - f_la57 = entry->ecx & F(LA57); > + f_la57 = cpuid_entry_get(entry, X86_FEATURE_LA57); > cpuid_mask(&entry->ecx, CPUID_7_ECX); > /* Set LA57 based on hardware capability. */ > entry->ecx |= f_la57; > diff --git a/arch/x86/kvm/cpuid.h b/arch/x86/kvm/cpuid.h > index 72a79bdfed6b..64e96e4086e2 100644 > --- a/arch/x86/kvm/cpuid.h > +++ b/arch/x86/kvm/cpuid.h > @@ -95,16 +95,10 @@ static __always_inline struct cpuid_reg x86_feature_cpuid(unsigned x86_feature) > return reverse_cpuid[x86_leaf]; > } > > -static __always_inline u32 *guest_cpuid_get_register(struct kvm_vcpu *vcpu, unsigned x86_feature) > +static __always_inline u32 *__cpuid_entry_get_reg(struct kvm_cpuid_entry2 *entry, > + const struct cpuid_reg *cpuid) > { > - struct kvm_cpuid_entry2 *entry; > - const struct cpuid_reg cpuid = x86_feature_cpuid(x86_feature); > - > - entry = kvm_find_cpuid_entry(vcpu, cpuid.function, cpuid.index); > - if (!entry) > - return NULL; > - > - switch (cpuid.reg) { > + switch (cpuid->reg) { > case CPUID_EAX: > return &entry->eax; > case CPUID_EBX: > @@ -119,6 +113,40 @@ static __always_inline u32 *guest_cpuid_get_register(struct kvm_vcpu *vcpu, unsi > } > } > > +static __always_inline u32 *cpuid_entry_get_reg(struct kvm_cpuid_entry2 *entry, > + unsigned x86_feature) It is just me who dislikes bare 'unsigned'? > +{ > + const struct cpuid_reg cpuid = x86_feature_cpuid(x86_feature); > + > + return __cpuid_entry_get_reg(entry, &cpuid); > +} > + > +static __always_inline u32 cpuid_entry_get(struct kvm_cpuid_entry2 *entry, > + unsigned x86_feature) > +{ > + u32 *reg = cpuid_entry_get_reg(entry, x86_feature); > + > + return *reg & __feature_bit(x86_feature); > +} > + > +static __always_inline bool cpuid_entry_has(struct kvm_cpuid_entry2 *entry, > + unsigned x86_feature) > +{ > + return cpuid_entry_get(entry, x86_feature); > +} > + > +static __always_inline int *guest_cpuid_get_register(struct kvm_vcpu *vcpu, unsigned x86_feature) > +{ > + struct kvm_cpuid_entry2 *entry; > + const struct cpuid_reg cpuid = x86_feature_cpuid(x86_feature); > + > + entry = kvm_find_cpuid_entry(vcpu, cpuid.function, cpuid.index); > + if (!entry) > + return NULL; > + > + return __cpuid_entry_get_reg(entry, &cpuid); > +} > + > static __always_inline bool guest_cpuid_has(struct kvm_vcpu *vcpu, unsigned x86_feature) > { > u32 *reg; Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
On Fri, Feb 21, 2020 at 04:57:52PM +0100, Vitaly Kuznetsov wrote: > Sean Christopherson <sean.j.christopherson@intel.com> writes: > > > @@ -119,6 +113,40 @@ static __always_inline u32 *guest_cpuid_get_register(struct kvm_vcpu *vcpu, unsi > > } > > } > > > > +static __always_inline u32 *cpuid_entry_get_reg(struct kvm_cpuid_entry2 *entry, > > + unsigned x86_feature) > > It is just me who dislikes bare 'unsigned'? I don't like it either. I also don't like get yelled at by checkpatch. I used "unsigned" here and throughout to be consistent with the existing guest_cpuid_*() and x86_feature_cpuid() helpers in cpuid.h. I will happily add a patch to change those to use "unsigned int" and then also use "unsigned int" for the cpuid_entry_*() code.
diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c index e3026fe638aa..3316963dad3d 100644 --- a/arch/x86/kvm/cpuid.c +++ b/arch/x86/kvm/cpuid.c @@ -68,7 +68,7 @@ int kvm_update_cpuid(struct kvm_vcpu *vcpu) best->edx |= F(APIC); if (apic) { - if (best->ecx & F(TSC_DEADLINE_TIMER)) + if (cpuid_entry_has(best, X86_FEATURE_TSC_DEADLINE_TIMER)) apic->lapic_timer.timer_mode_mask = 3 << 17; else apic->lapic_timer.timer_mode_mask = 1 << 17; @@ -96,7 +96,8 @@ int kvm_update_cpuid(struct kvm_vcpu *vcpu) } best = kvm_find_cpuid_entry(vcpu, 0xD, 1); - if (best && (best->eax & (F(XSAVES) | F(XSAVEC)))) + if (best && (cpuid_entry_has(best, X86_FEATURE_XSAVES) || + cpuid_entry_has(best, X86_FEATURE_XSAVEC))) best->ebx = xstate_required_size(vcpu->arch.xcr0, true); /* @@ -155,7 +156,7 @@ static void cpuid_fix_nx_cap(struct kvm_vcpu *vcpu) break; } } - if (entry && (entry->edx & F(NX)) && !is_efer_nx()) { + if (entry && cpuid_entry_has(entry, X86_FEATURE_NX) && !is_efer_nx()) { entry->edx &= ~F(NX); printk(KERN_INFO "kvm: guest NX capability removed\n"); } @@ -387,7 +388,7 @@ static inline void do_cpuid_7_mask(struct kvm_cpuid_entry2 *entry) entry->ebx |= F(TSC_ADJUST); entry->ecx &= kvm_cpuid_7_0_ecx_x86_features; - f_la57 = entry->ecx & F(LA57); + f_la57 = cpuid_entry_get(entry, X86_FEATURE_LA57); cpuid_mask(&entry->ecx, CPUID_7_ECX); /* Set LA57 based on hardware capability. */ entry->ecx |= f_la57; diff --git a/arch/x86/kvm/cpuid.h b/arch/x86/kvm/cpuid.h index 72a79bdfed6b..64e96e4086e2 100644 --- a/arch/x86/kvm/cpuid.h +++ b/arch/x86/kvm/cpuid.h @@ -95,16 +95,10 @@ static __always_inline struct cpuid_reg x86_feature_cpuid(unsigned x86_feature) return reverse_cpuid[x86_leaf]; } -static __always_inline u32 *guest_cpuid_get_register(struct kvm_vcpu *vcpu, unsigned x86_feature) +static __always_inline u32 *__cpuid_entry_get_reg(struct kvm_cpuid_entry2 *entry, + const struct cpuid_reg *cpuid) { - struct kvm_cpuid_entry2 *entry; - const struct cpuid_reg cpuid = x86_feature_cpuid(x86_feature); - - entry = kvm_find_cpuid_entry(vcpu, cpuid.function, cpuid.index); - if (!entry) - return NULL; - - switch (cpuid.reg) { + switch (cpuid->reg) { case CPUID_EAX: return &entry->eax; case CPUID_EBX: @@ -119,6 +113,40 @@ static __always_inline u32 *guest_cpuid_get_register(struct kvm_vcpu *vcpu, unsi } } +static __always_inline u32 *cpuid_entry_get_reg(struct kvm_cpuid_entry2 *entry, + unsigned x86_feature) +{ + const struct cpuid_reg cpuid = x86_feature_cpuid(x86_feature); + + return __cpuid_entry_get_reg(entry, &cpuid); +} + +static __always_inline u32 cpuid_entry_get(struct kvm_cpuid_entry2 *entry, + unsigned x86_feature) +{ + u32 *reg = cpuid_entry_get_reg(entry, x86_feature); + + return *reg & __feature_bit(x86_feature); +} + +static __always_inline bool cpuid_entry_has(struct kvm_cpuid_entry2 *entry, + unsigned x86_feature) +{ + return cpuid_entry_get(entry, x86_feature); +} + +static __always_inline int *guest_cpuid_get_register(struct kvm_vcpu *vcpu, unsigned x86_feature) +{ + struct kvm_cpuid_entry2 *entry; + const struct cpuid_reg cpuid = x86_feature_cpuid(x86_feature); + + entry = kvm_find_cpuid_entry(vcpu, cpuid.function, cpuid.index); + if (!entry) + return NULL; + + return __cpuid_entry_get_reg(entry, &cpuid); +} + static __always_inline bool guest_cpuid_has(struct kvm_vcpu *vcpu, unsigned x86_feature) { u32 *reg;
Introduce accessors to retrieve feature bits from CPUID entries and use the new accessors where applicable. Using the accessors eliminates the need to manually specify the register to be queried at no extra cost (binary output is identical) and will allow adding runtime consistency checks on the function and index in a future patch. No functional change intended. Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> --- arch/x86/kvm/cpuid.c | 9 +++++---- arch/x86/kvm/cpuid.h | 46 +++++++++++++++++++++++++++++++++++--------- 2 files changed, 42 insertions(+), 13 deletions(-)