Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1631709 > unrolled thread

[PATCH v2 4/9] KVM: mark requests that do not need a wakeup

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  [PATCH v2 4/9] KVM: mark requests that do not need a wakeup Radim Krčmář <rkrcmar@redhat.com> - 2017-04-26 22:40 +0200
    Re: [PATCH v2 4/9] KVM: mark requests that do not need a wakeup Andrew Jones <drjones@redhat.com> - 2017-04-27 13:40 +0200
    Re: [PATCH v2 4/9] KVM: mark requests that do not need a wakeup Cornelia Huck <cornelia.huck@de.ibm.com> - 2017-04-27 14:10 +0200

#1631709 — [PATCH v2 4/9] KVM: mark requests that do not need a wakeup

Some operations must ensure that the guest is not running with stale
data, but if the guest is halted, then the update can wait until another
event happens.  kvm_make_all_requests() currently doesn't wake up, so we
can mark all requests used with it.

First 8 bits were arbitrarily reserved for request numbers.

Most uses of requests have the request type as a constant, so a compiler
will optimize the '&'.

An alternative would be to have an inline function that would return
whether the request needs a wake-up or not, but I like this one better
even though it might produce worse assembly.

Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
---
 v2: use GENMASK [Marc]
---
 arch/arm/include/asm/kvm_host.h   |  2 +-
 arch/arm64/include/asm/kvm_host.h |  2 +-
 arch/x86/include/asm/kvm_host.h   |  6 +++---
 include/linux/kvm_host.h          | 12 +++++++-----
 4 files changed, 12 insertions(+), 10 deletions(-)

diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
index de67ce647501..49358f20d36f 100644
--- a/arch/arm/include/asm/kvm_host.h
+++ b/arch/arm/include/asm/kvm_host.h
@@ -44,7 +44,7 @@
 #define KVM_MAX_VCPUS VGIC_V2_MAX_CPUS
 #endif
 
-#define KVM_REQ_VCPU_EXIT	8
+#define KVM_REQ_VCPU_EXIT	(8 | KVM_REQUEST_NO_WAKEUP)
 
 u32 *kvm_vcpu_reg(struct kvm_vcpu *vcpu, u8 reg_num, u32 mode);
 int __attribute_const__ kvm_target_cpu(void);
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index 522e4f60976e..1c9458a7ec92 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -41,7 +41,7 @@
 
 #define KVM_VCPU_MAX_FEATURES 4
 
-#define KVM_REQ_VCPU_EXIT	8
+#define KVM_REQ_VCPU_EXIT	(8 | KVM_REQUEST_NO_WAKEUP)
 
 int __attribute_const__ kvm_target_cpu(void);
 int kvm_reset_vcpu(struct kvm_vcpu *vcpu);
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index f5c942edbc86..19219826bed6 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -61,10 +61,10 @@
 #define KVM_REQ_PMI               19
 #define KVM_REQ_SMI               20
 #define KVM_REQ_MASTERCLOCK_UPDATE 21
-#define KVM_REQ_MCLOCK_INPROGRESS 22
-#define KVM_REQ_SCAN_IOAPIC       23
+#define KVM_REQ_MCLOCK_INPROGRESS (22 | KVM_REQUEST_NO_WAKEUP)
+#define KVM_REQ_SCAN_IOAPIC       (23 | KVM_REQUEST_NO_WAKEUP)
 #define KVM_REQ_GLOBAL_CLOCK_UPDATE 24
-#define KVM_REQ_APIC_PAGE_RELOAD  25
+#define KVM_REQ_APIC_PAGE_RELOAD  (25 | KVM_REQUEST_NO_WAKEUP)
 #define KVM_REQ_HV_CRASH          26
 #define KVM_REQ_IOAPIC_EOI_EXIT   27
 #define KVM_REQ_HV_RESET          28
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 374fa92c7657..a805ddcb7eb0 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -115,12 +115,14 @@ static inline bool is_error_page(struct page *page)
 	return IS_ERR(page);
 }
 
+#define KVM_REQUEST_MASK           GENMASK(7,0)
+#define KVM_REQUEST_NO_WAKEUP      BIT(8)
 /*
  * Architecture-independent vcpu->requests bit members
  * Bits 4-7 are reserved for more arch-independent bits.
  */
-#define KVM_REQ_TLB_FLUSH          0
-#define KVM_REQ_MMU_RELOAD         1
+#define KVM_REQ_TLB_FLUSH          (0 | KVM_REQUEST_NO_WAKEUP)
+#define KVM_REQ_MMU_RELOAD         (1 | KVM_REQUEST_NO_WAKEUP)
 #define KVM_REQ_PENDING_TIMER      2
 #define KVM_REQ_UNHALT             3
 
@@ -1076,17 +1078,17 @@ static inline void kvm_make_request(int req, struct kvm_vcpu *vcpu)
 	 * caller.  Paired with the smp_mb__after_atomic in kvm_check_request.
 	 */
 	smp_wmb();
-	set_bit(req, &vcpu->requests);
+	set_bit(req & KVM_REQUEST_MASK, &vcpu->requests);
 }
 
 static inline bool kvm_test_request(int req, struct kvm_vcpu *vcpu)
 {
-	return test_bit(req, &vcpu->requests);
+	return test_bit(req & KVM_REQUEST_MASK, &vcpu->requests);
 }
 
 static inline void kvm_clear_request(int req, struct kvm_vcpu *vcpu)
 {
-	clear_bit(req, &vcpu->requests);
+	clear_bit(req & KVM_REQUEST_MASK, &vcpu->requests);
 }
 
 static inline bool kvm_check_request(int req, struct kvm_vcpu *vcpu)
-- 
2.12.2

[toc] | [next] | [standalone]

#1632041

On Wed, Apr 26, 2017 at 10:32:22PM +0200, Radim Krčmář wrote:
> Some operations must ensure that the guest is not running with stale
> data, but if the guest is halted, then the update can wait until another
> event happens.  kvm_make_all_requests() currently doesn't wake up, so we
> can mark all requests used with it.
> 
> First 8 bits were arbitrarily reserved for request numbers.
> 
> Most uses of requests have the request type as a constant, so a compiler
> will optimize the '&'.
> 
> An alternative would be to have an inline function that would return
> whether the request needs a wake-up or not, but I like this one better
> even though it might produce worse assembly.
> 
> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
> ---
>  v2: use GENMASK [Marc]
> ---
>  arch/arm/include/asm/kvm_host.h   |  2 +-
>  arch/arm64/include/asm/kvm_host.h |  2 +-
>  arch/x86/include/asm/kvm_host.h   |  6 +++---
>  include/linux/kvm_host.h          | 12 +++++++-----
>  4 files changed, 12 insertions(+), 10 deletions(-)

Reviewed-by: Andrew Jones <drjones@redhat.com>

[toc] | [prev] | [next] | [standalone]

#1632060

On Wed, 26 Apr 2017 22:32:22 +0200
Radim Krčmář <rkrcmar@redhat.com> wrote:

> Some operations must ensure that the guest is not running with stale
> data, but if the guest is halted, then the update can wait until another
> event happens.  kvm_make_all_requests() currently doesn't wake up, so we
> can mark all requests used with it.
> 
> First 8 bits were arbitrarily reserved for request numbers.
> 
> Most uses of requests have the request type as a constant, so a compiler
> will optimize the '&'.
> 
> An alternative would be to have an inline function that would return
> whether the request needs a wake-up or not, but I like this one better
> even though it might produce worse assembly.
> 
> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
> ---
>  v2: use GENMASK [Marc]
> ---
>  arch/arm/include/asm/kvm_host.h   |  2 +-
>  arch/arm64/include/asm/kvm_host.h |  2 +-
>  arch/x86/include/asm/kvm_host.h   |  6 +++---
>  include/linux/kvm_host.h          | 12 +++++++-----
>  4 files changed, 12 insertions(+), 10 deletions(-)

Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web