From d35268da66870d733ae763fd7f9b06a1f63f395e Mon Sep 17 00:00:00 2001 From: Christoffer Dall Date: Tue, 25 Aug 2015 19:48:21 +0200 Subject: arm/arm64: KVM: arch_timer: Only schedule soft timer on vcpu_block We currently schedule a soft timer every time we exit the guest if the timer did not expire while running the guest. This is really not necessary, because the only work we do in the timer work function is to kick the vcpu. Kicking the vcpu does two things: (1) If the vpcu thread is on a waitqueue, make it runnable and remove it from the waitqueue. (2) If the vcpu is running on a different physical CPU from the one doing the kick, it sends a reschedule IPI. The second case cannot happen, because the soft timer is only ever scheduled when the vcpu is not running. The first case is only relevant when the vcpu thread is on a waitqueue, which is only the case when the vcpu thread has called kvm_vcpu_block(). Therefore, we only need to make sure a timer is scheduled for kvm_vcpu_block(), which we do by encapsulating all calls to kvm_vcpu_block() with kvm_timer_{un}schedule calls. Additionally, we only schedule a soft timer if the timer is enabled and unmasked, since it is useless otherwise. Note that theoretically userspace can use the SET_ONE_REG interface to change registers that should cause the timer to fire, even if the vcpu is blocked without a scheduled timer, but this case was not supported before this patch and we leave it for future work for now. Signed-off-by: Christoffer Dall --- include/kvm/arm_arch_timer.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/kvm') diff --git a/include/kvm/arm_arch_timer.h b/include/kvm/arm_arch_timer.h index e1e4d7c38dda..ef14cc1f1f26 100644 --- a/include/kvm/arm_arch_timer.h +++ b/include/kvm/arm_arch_timer.h @@ -71,5 +71,7 @@ u64 kvm_arm_timer_get_reg(struct kvm_vcpu *, u64 regid); int kvm_arm_timer_set_reg(struct kvm_vcpu *, u64 regid, u64 value); bool kvm_timer_should_fire(struct kvm_vcpu *vcpu); +void kvm_timer_schedule(struct kvm_vcpu *vcpu); +void kvm_timer_unschedule(struct kvm_vcpu *vcpu); #endif -- cgit v1.2.3 From 4b4b4512da2a844b8da2585609b67fae1ce4f4db Mon Sep 17 00:00:00 2001 From: Christoffer Dall Date: Sun, 30 Aug 2015 15:01:27 +0200 Subject: arm/arm64: KVM: Rework the arch timer to use level-triggered semantics The arch timer currently uses edge-triggered semantics in the sense that the line is never sampled by the vgic and lowering the line from the timer to the vgic doesn't have any effect on the pending state of virtual interrupts in the vgic. This means that we do not support a guest with the otherwise valid behavior of (1) disable interrupts (2) enable the timer (3) disable the timer (4) enable interrupts. Such a guest would validly not expect to see any interrupts on real hardware, but will see interrupts on KVM. This patch fixes this shortcoming through the following series of changes. First, we change the flow of the timer/vgic sync/flush operations. Now the timer is always flushed/synced before the vgic, because the vgic samples the state of the timer output. This has the implication that we move the timer operations in to non-preempible sections, but that is fine after the previous commit getting rid of hrtimer schedules on every entry/exit. Second, we change the internal behavior of the timer, letting the timer keep track of its previous output state, and only lower/raise the line to the vgic when the state changes. Note that in theory this could have been accomplished more simply by signalling the vgic every time the state *potentially* changed, but we don't want to be hitting the vgic more often than necessary. Third, we get rid of the use of the map->active field in the vgic and instead simply set the interrupt as active on the physical distributor whenever the input to the GIC is asserted and conversely clear the physical active state when the input to the GIC is deasserted. Fourth, and finally, we now initialize the timer PPIs (and all the other unused PPIs for now), to be level-triggered, and modify the sync code to sample the line state on HW sync and re-inject a new interrupt if it is still pending at that time. Signed-off-by: Christoffer Dall --- include/kvm/arm_arch_timer.h | 2 +- include/kvm/arm_vgic.h | 3 --- 2 files changed, 1 insertion(+), 4 deletions(-) (limited to 'include/kvm') diff --git a/include/kvm/arm_arch_timer.h b/include/kvm/arm_arch_timer.h index ef14cc1f1f26..1800227af9d6 100644 --- a/include/kvm/arm_arch_timer.h +++ b/include/kvm/arm_arch_timer.h @@ -51,7 +51,7 @@ struct arch_timer_cpu { bool armed; /* Timer IRQ */ - const struct kvm_irq_level *irq; + struct kvm_irq_level irq; /* VGIC mapping */ struct irq_phys_map *map; diff --git a/include/kvm/arm_vgic.h b/include/kvm/arm_vgic.h index 4e14dac282bb..7bc5d0224ab0 100644 --- a/include/kvm/arm_vgic.h +++ b/include/kvm/arm_vgic.h @@ -159,7 +159,6 @@ struct irq_phys_map { u32 virt_irq; u32 phys_irq; u32 irq; - bool active; }; struct irq_phys_map_entry { @@ -354,8 +353,6 @@ int kvm_vgic_vcpu_active_irq(struct kvm_vcpu *vcpu); struct irq_phys_map *kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu, int virt_irq, int irq); int kvm_vgic_unmap_phys_irq(struct kvm_vcpu *vcpu, struct irq_phys_map *map); -bool kvm_vgic_get_phys_irq_active(struct irq_phys_map *map); -void kvm_vgic_set_phys_irq_active(struct irq_phys_map *map, bool active); #define irqchip_in_kernel(k) (!!((k)->arch.vgic.in_kernel)) #define vgic_initialized(k) (!!((k)->arch.vgic.nr_cpus)) -- cgit v1.2.3 From 5fdf876d30ce7c9b63045b3db8374912ab9b262c Mon Sep 17 00:00:00 2001 From: Michal Marek Date: Thu, 15 Oct 2015 22:16:28 +0200 Subject: KVM: arm: Do not indent the arguments of DECLARE_BITMAP Besides being a coding style issue, it confuses make tags: ctags: Warning: include/kvm/arm_vgic.h:307: null expansion of name pattern "\1" ctags: Warning: include/kvm/arm_vgic.h:308: null expansion of name pattern "\1" ctags: Warning: include/kvm/arm_vgic.h:309: null expansion of name pattern "\1" ctags: Warning: include/kvm/arm_vgic.h:317: null expansion of name pattern "\1" Cc: kvmarm@lists.cs.columbia.edu Signed-off-by: Michal Marek Signed-off-by: Christoffer Dall --- include/kvm/arm_vgic.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include/kvm') diff --git a/include/kvm/arm_vgic.h b/include/kvm/arm_vgic.h index 7bc5d0224ab0..8065801a1847 100644 --- a/include/kvm/arm_vgic.h +++ b/include/kvm/arm_vgic.h @@ -299,9 +299,9 @@ struct vgic_cpu { u8 *vgic_irq_lr_map; /* Pending/active/both interrupts on this VCPU */ - DECLARE_BITMAP( pending_percpu, VGIC_NR_PRIVATE_IRQS); - DECLARE_BITMAP( active_percpu, VGIC_NR_PRIVATE_IRQS); - DECLARE_BITMAP( pend_act_percpu, VGIC_NR_PRIVATE_IRQS); + DECLARE_BITMAP(pending_percpu, VGIC_NR_PRIVATE_IRQS); + DECLARE_BITMAP(active_percpu, VGIC_NR_PRIVATE_IRQS); + DECLARE_BITMAP(pend_act_percpu, VGIC_NR_PRIVATE_IRQS); /* Pending/active/both shared interrupts, dynamically sized */ unsigned long *pending_shared; @@ -309,7 +309,7 @@ struct vgic_cpu { unsigned long *pend_act_shared; /* Bitmap of used/free list registers */ - DECLARE_BITMAP( lr_used, VGIC_V2_MAX_LRS); + DECLARE_BITMAP(lr_used, VGIC_V2_MAX_LRS); /* Number of list registers on this CPU */ int nr_lr; -- cgit v1.2.3 From c4cd4c168b81dad53e659d18cdae653bc0ec2384 Mon Sep 17 00:00:00 2001 From: Pavel Fedin Date: Tue, 27 Oct 2015 11:37:29 +0300 Subject: KVM: arm/arm64: Optimize away redundant LR tracking Currently we use vgic_irq_lr_map in order to track which LRs hold which IRQs, and lr_used bitmap in order to track which LRs are used or free. vgic_irq_lr_map is actually used only for piggy-back optimization, and can be easily replaced by iteration over lr_used. This is good because in future, when LPI support is introduced, number of IRQs will grow up to at least 16384, while numbers from 1024 to 8192 are never going to be used. This would be a huge memory waste. In its turn, lr_used is also completely redundant since ae705930fca6322600690df9dc1c7d0516145a93 ("arm/arm64: KVM: Keep elrsr/aisr in sync with software model"), because together with lr_used we also update elrsr. This allows to easily replace lr_used with elrsr, inverting all conditions (because in elrsr '1' means 'free'). Signed-off-by: Pavel Fedin Signed-off-by: Christoffer Dall --- include/kvm/arm_vgic.h | 6 ------ 1 file changed, 6 deletions(-) (limited to 'include/kvm') diff --git a/include/kvm/arm_vgic.h b/include/kvm/arm_vgic.h index 8065801a1847..3936bf802e1d 100644 --- a/include/kvm/arm_vgic.h +++ b/include/kvm/arm_vgic.h @@ -295,9 +295,6 @@ struct vgic_v3_cpu_if { }; struct vgic_cpu { - /* per IRQ to LR mapping */ - u8 *vgic_irq_lr_map; - /* Pending/active/both interrupts on this VCPU */ DECLARE_BITMAP(pending_percpu, VGIC_NR_PRIVATE_IRQS); DECLARE_BITMAP(active_percpu, VGIC_NR_PRIVATE_IRQS); @@ -308,9 +305,6 @@ struct vgic_cpu { unsigned long *active_shared; unsigned long *pend_act_shared; - /* Bitmap of used/free list registers */ - DECLARE_BITMAP(lr_used, VGIC_V2_MAX_LRS); - /* Number of list registers on this CPU */ int nr_lr; -- cgit v1.2.3 From 26caea7693cb99833fe4ecc544c842289d6b3f69 Mon Sep 17 00:00:00 2001 From: Pavel Fedin Date: Tue, 27 Oct 2015 11:37:31 +0300 Subject: KVM: arm/arm64: Merge vgic_set_lr() and vgic_sync_lr_elrsr() Now we see that vgic_set_lr() and vgic_sync_lr_elrsr() are always used together. Merge them into one function, saving from second vgic_ops dereferencing every time. Signed-off-by: Pavel Fedin Signed-off-by: Christoffer Dall --- include/kvm/arm_vgic.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/kvm') diff --git a/include/kvm/arm_vgic.h b/include/kvm/arm_vgic.h index 3936bf802e1d..f62addc17dcf 100644 --- a/include/kvm/arm_vgic.h +++ b/include/kvm/arm_vgic.h @@ -112,7 +112,6 @@ struct vgic_vmcr { struct vgic_ops { struct vgic_lr (*get_lr)(const struct kvm_vcpu *, int); void (*set_lr)(struct kvm_vcpu *, int, struct vgic_lr); - void (*sync_lr_elrsr)(struct kvm_vcpu *, int, struct vgic_lr); u64 (*get_elrsr)(const struct kvm_vcpu *vcpu); u64 (*get_eisr)(const struct kvm_vcpu *vcpu); void (*clear_eisr)(struct kvm_vcpu *vcpu); -- cgit v1.2.3