From 68c3b4d1676d870f0453c31d5a52e7e65c7448ae Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 31 Mar 2014 21:50:44 +0300 Subject: KVM: VMX: speed up wildcard MMIO EVENTFD With KVM, MMIO is much slower than PIO, due to the need to do page walk and emulation. But with EPT, it does not have to be: we know the address from the VMCS so if the address is unique, we can look up the eventfd directly, bypassing emulation. Unfortunately, this only works if userspace does not need to match on access length and data. The implementation adds a separate FAST_MMIO bus internally. This serves two purposes: - minimize overhead for old userspace that does not use eventfd with lengtth = 0 - minimize disruption in other code (since we don't know the length, devices on the MMIO bus only get a valid address in write, this way we don't need to touch all devices to teach them to handle an invalid length) At the moment, this optimization only has effect for EPT on x86. It will be possible to speed up MMIO for NPT and MMU using the same idea in the future. With this patch applied, on VMX MMIO EVENTFD is essentially as fast as PIO. I was unable to detect any measureable slowdown to non-eventfd MMIO. Making MMIO faster is important for the upcoming virtio 1.0 which includes an MMIO signalling capability. The idea was suggested by Peter Anvin. Lots of thanks to Gleb for pre-review and suggestions. Signed-off-by: Michael S. Tsirkin Signed-off-by: Marcelo Tosatti --- include/linux/kvm_host.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 7d21cf9f4380..6c3c2eb96d06 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -163,6 +163,7 @@ enum kvm_bus { KVM_MMIO_BUS, KVM_PIO_BUS, KVM_VIRTIO_CCW_NOTIFY_BUS, + KVM_FAST_MMIO_BUS, KVM_NR_BUSES }; -- cgit v1.2.3 From dfeec843fb237d73947e818f961e8d6f0df22b01 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 1 Jan 2014 16:09:21 +0100 Subject: KVM: add kvm_is_error_gpa() helper It's quite common (in the s390 guest access code) to test if a guest physical address points to a valid guest memory area or not. So add a simple helper function in common code, since this might be of interest for other architectures as well. Signed-off-by: Heiko Carstens Reviewed-by: Thomas Huth Reviewed-by: Cornelia Huck Signed-off-by: Christian Borntraeger --- include/linux/kvm_host.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include/linux') diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 7d21cf9f4380..471d1400c4ac 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -879,6 +879,13 @@ static inline hpa_t pfn_to_hpa(pfn_t pfn) return (hpa_t)pfn << PAGE_SHIFT; } +static inline bool kvm_is_error_gpa(struct kvm *kvm, gpa_t gpa) +{ + unsigned long hva = gfn_to_hva(kvm, gpa_to_gfn(gpa)); + + return kvm_is_error_hva(hva); +} + static inline void kvm_migrate_timers(struct kvm_vcpu *vcpu) { set_bit(KVM_REQ_MIGRATE_TIMER, &vcpu->requests); -- cgit v1.2.3 From a086f6a1ebc9d8d2d028b99e779ce0dbd9691dea Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Thu, 17 Apr 2014 17:06:12 +0800 Subject: Revert "KVM: Simplify kvm->tlbs_dirty handling" This reverts commit 5befdc385ddb2d5ae8995ad89004529a3acf58fc. Since we will allow flush tlb out of mmu-lock in the later patch Signed-off-by: Xiao Guangrong Signed-off-by: Marcelo Tosatti --- include/linux/kvm_host.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 32d263f683dc..820fc2e1d9df 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -411,9 +411,7 @@ struct kvm { unsigned long mmu_notifier_seq; long mmu_notifier_count; #endif - /* Protected by mmu_lock */ - bool tlbs_dirty; - + long tlbs_dirty; struct list_head devices; }; -- cgit v1.2.3 From 8ad357551797b1edc184fb9f6a4f80a6fa626459 Mon Sep 17 00:00:00 2001 From: David Hildenbrand Date: Fri, 14 Mar 2014 11:00:21 +0100 Subject: KVM: s390: enable IBS for single running VCPUs This patch enables the IBS facility when a single VCPU is running. The facility is dynamically turned on/off as soon as other VCPUs enter/leave the stopped state. When this facility is operating, some instructions can be executed faster for single-cpu guests. Signed-off-by: David Hildenbrand Reviewed-by: Dominik Dingel Reviewed-by: Cornelia Huck Signed-off-by: Christian Borntraeger --- include/linux/kvm_host.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 820fc2e1d9df..1e125b055327 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -134,6 +134,8 @@ static inline bool is_error_page(struct page *page) #define KVM_REQ_EPR_EXIT 20 #define KVM_REQ_SCAN_IOAPIC 21 #define KVM_REQ_GLOBAL_CLOCK_UPDATE 22 +#define KVM_REQ_ENABLE_IBS 23 +#define KVM_REQ_DISABLE_IBS 24 #define KVM_USERSPACE_IRQ_SOURCE_ID 0 #define KVM_IRQFD_RESAMPLE_IRQ_SOURCE_ID 1 -- cgit v1.2.3 From 719d93cd5f5c5c8775b7a38192069e8e1d1ac46e Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Thu, 16 Jan 2014 13:44:20 +0100 Subject: kvm/irqchip: Speed up KVM_SET_GSI_ROUTING When starting lots of dataplane devices the bootup takes very long on Christian's s390 with irqfd patches. With larger setups he is even able to trigger some timeouts in some components. Turns out that the KVM_SET_GSI_ROUTING ioctl takes very long (strace claims up to 0.1 sec) when having multiple CPUs. This is caused by the synchronize_rcu and the HZ=100 of s390. By changing the code to use a private srcu we can speed things up. This patch reduces the boot time till mounting root from 8 to 2 seconds on my s390 guest with 100 disks. Uses of hlist_for_each_entry_rcu, hlist_add_head_rcu, hlist_del_init_rcu are fine because they do not have lockdep checks (hlist_for_each_entry_rcu uses rcu_dereference_raw rather than rcu_dereference, and write-sides do not do rcu lockdep at all). Note that we're hardly relying on the "sleepable" part of srcu. We just want SRCU's faster detection of grace periods. Testing was done by Andrew Theurer using netperf tests STREAM, MAERTS and RR. The difference between results "before" and "after" the patch has mean -0.2% and standard deviation 0.6%. Using a paired t-test on the data points says that there is a 2.5% probability that the patch is the cause of the performance difference (rather than a random fluctuation). (Restricting the t-test to RR, which is the most likely to be affected, changes the numbers to respectively -0.3% mean, 0.7% stdev, and 8% probability that the numbers actually say something about the patch. The probability increases mostly because there are fewer data points). Cc: Marcelo Tosatti Cc: Michael S. Tsirkin Tested-by: Christian Borntraeger # s390 Reviewed-by: Christian Borntraeger Signed-off-by: Christian Borntraeger Signed-off-by: Paolo Bonzini --- include/linux/kvm_host.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 1e125b055327..970c68197c69 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -370,6 +370,7 @@ struct kvm { struct mm_struct *mm; /* userspace tied to this vm */ struct kvm_memslots *memslots; struct srcu_struct srcu; + struct srcu_struct irq_srcu; #ifdef CONFIG_KVM_APIC_ARCHITECTURE u32 bsp_vcpu_id; #endif -- cgit v1.2.3