| Message ID | 20200326152438.6218-23-alexandru.elisei@arm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | Add reassignable BARs and PCIE 1.1 support | expand |
On 26/03/2020 15:24, Alexandru Elisei wrote: > When we want to map a device region into the guest address space, first we > perform an mmap on the device fd. The resulting VMA is a mapping between > host userspace addresses and physical addresses associated with the device. > Next, we create a memslot, which populates the stage 2 table with the > mappings between guest physical addresses and the device physical adresses. > > However, when we want to unmap the device from the guest address space, we > only call munmap, which destroys the VMA and the stage 2 mappings, but > doesn't destroy the memslot and kvmtool's internal mem_bank structure > associated with the memslot. > > This has been perfectly fine so far, because we only unmap a device region > when we exit kvmtool. This is will change when we add support for > reassignable BARs, and we will have to unmap vfio regions as the guest > kernel writes new addresses in the BARs. This can lead to two possible > problems: > > - We refuse to create a valid BAR mapping because of a stale mem_bank > structure which belonged to a previously unmapped region. > > - It is possible that the mmap in vfio_map_region returns the same address > that was used to create a memslot, but was unmapped by vfio_unmap_region. > Guest accesses to the device memory will fault because the stage 2 > mappings are missing, and this can lead to performance degradation. > > Let's do the right thing and destroy the memslot and the mem_bank struct > associated with it when we unmap a vfio region. Set host_addr to NULL after > the munmap call so we won't try to unmap an address which is currently used > by the process for something else if vfio_unmap_region gets called twice. > Thanks for adding the locking here, looks alright now. > Signed-off-by: Alexandru Elisei <alexandru.elisei@arm.com> Reviewed-by: Andre Przywara <andre.przywara@arm.com> Cheers, Andre > --- > include/kvm/kvm.h | 4 ++ > kvm.c | 101 ++++++++++++++++++++++++++++++++++++++++------ > vfio/core.c | 6 +++ > 3 files changed, 99 insertions(+), 12 deletions(-) > > diff --git a/include/kvm/kvm.h b/include/kvm/kvm.h > index 50119a8672eb..9428f57a1c0c 100644 > --- a/include/kvm/kvm.h > +++ b/include/kvm/kvm.h > @@ -1,6 +1,7 @@ > #ifndef KVM__KVM_H > #define KVM__KVM_H > > +#include "kvm/mutex.h" > #include "kvm/kvm-arch.h" > #include "kvm/kvm-config.h" > #include "kvm/util-init.h" > @@ -56,6 +57,7 @@ struct kvm_mem_bank { > void *host_addr; > u64 size; > enum kvm_mem_type type; > + u32 slot; > }; > > struct kvm { > @@ -72,6 +74,7 @@ struct kvm { > u64 ram_size; > void *ram_start; > u64 ram_pagesize; > + struct mutex mem_banks_lock; > struct list_head mem_banks; > > bool nmi_disabled; > @@ -106,6 +109,7 @@ void kvm__irq_line(struct kvm *kvm, int irq, int level); > void kvm__irq_trigger(struct kvm *kvm, int irq); > bool kvm__emulate_io(struct kvm_cpu *vcpu, u16 port, void *data, int direction, int size, u32 count); > bool kvm__emulate_mmio(struct kvm_cpu *vcpu, u64 phys_addr, u8 *data, u32 len, u8 is_write); > +int kvm__destroy_mem(struct kvm *kvm, u64 guest_phys, u64 size, void *userspace_addr); > int kvm__register_mem(struct kvm *kvm, u64 guest_phys, u64 size, void *userspace_addr, > enum kvm_mem_type type); > static inline int kvm__register_ram(struct kvm *kvm, u64 guest_phys, u64 size, > diff --git a/kvm.c b/kvm.c > index 57c4ff98ec4c..26f6b9bc58a3 100644 > --- a/kvm.c > +++ b/kvm.c > @@ -157,6 +157,7 @@ struct kvm *kvm__new(void) > if (!kvm) > return ERR_PTR(-ENOMEM); > > + mutex_init(&kvm->mem_banks_lock); > kvm->sys_fd = -1; > kvm->vm_fd = -1; > > @@ -183,20 +184,84 @@ int kvm__exit(struct kvm *kvm) > } > core_exit(kvm__exit); > > +int kvm__destroy_mem(struct kvm *kvm, u64 guest_phys, u64 size, > + void *userspace_addr) > +{ > + struct kvm_userspace_memory_region mem; > + struct kvm_mem_bank *bank; > + int ret; > + > + mutex_lock(&kvm->mem_banks_lock); > + list_for_each_entry(bank, &kvm->mem_banks, list) > + if (bank->guest_phys_addr == guest_phys && > + bank->size == size && bank->host_addr == userspace_addr) > + break; > + > + if (&bank->list == &kvm->mem_banks) { > + pr_err("Region [%llx-%llx] not found", guest_phys, > + guest_phys + size - 1); > + ret = -EINVAL; > + goto out; > + } > + > + if (bank->type == KVM_MEM_TYPE_RESERVED) { > + pr_err("Cannot delete reserved region [%llx-%llx]", > + guest_phys, guest_phys + size - 1); > + ret = -EINVAL; > + goto out; > + } > + > + mem = (struct kvm_userspace_memory_region) { > + .slot = bank->slot, > + .guest_phys_addr = guest_phys, > + .memory_size = 0, > + .userspace_addr = (unsigned long)userspace_addr, > + }; > + > + ret = ioctl(kvm->vm_fd, KVM_SET_USER_MEMORY_REGION, &mem); > + if (ret < 0) { > + ret = -errno; > + goto out; > + } > + > + list_del(&bank->list); > + free(bank); > + kvm->mem_slots--; > + ret = 0; > + > +out: > + mutex_unlock(&kvm->mem_banks_lock); > + return ret; > +} > + > int kvm__register_mem(struct kvm *kvm, u64 guest_phys, u64 size, > void *userspace_addr, enum kvm_mem_type type) > { > struct kvm_userspace_memory_region mem; > struct kvm_mem_bank *merged = NULL; > struct kvm_mem_bank *bank; > + struct list_head *prev_entry; > + u32 slot; > int ret; > > - /* Check for overlap */ > + mutex_lock(&kvm->mem_banks_lock); > + /* Check for overlap and find first empty slot. */ > + slot = 0; > + prev_entry = &kvm->mem_banks; > list_for_each_entry(bank, &kvm->mem_banks, list) { > u64 bank_end = bank->guest_phys_addr + bank->size - 1; > u64 end = guest_phys + size - 1; > - if (guest_phys > bank_end || end < bank->guest_phys_addr) > + if (guest_phys > bank_end || end < bank->guest_phys_addr) { > + /* > + * Keep the banks sorted ascending by slot, so it's > + * easier for us to find a free slot. > + */ > + if (bank->slot == slot) { > + slot++; > + prev_entry = &bank->list; > + } > continue; > + } > > /* Merge overlapping reserved regions */ > if (bank->type == KVM_MEM_TYPE_RESERVED && > @@ -226,38 +291,50 @@ int kvm__register_mem(struct kvm *kvm, u64 guest_phys, u64 size, > kvm_mem_type_to_string(bank->type), bank->guest_phys_addr, > bank->guest_phys_addr + bank->size - 1); > > - return -EINVAL; > + ret = -EINVAL; > + goto out; > } > > - if (merged) > - return 0; > + if (merged) { > + ret = 0; > + goto out; > + } > > bank = malloc(sizeof(*bank)); > - if (!bank) > - return -ENOMEM; > + if (!bank) { > + ret = -ENOMEM; > + goto out; > + } > > INIT_LIST_HEAD(&bank->list); > bank->guest_phys_addr = guest_phys; > bank->host_addr = userspace_addr; > bank->size = size; > bank->type = type; > + bank->slot = slot; > > if (type != KVM_MEM_TYPE_RESERVED) { > mem = (struct kvm_userspace_memory_region) { > - .slot = kvm->mem_slots++, > + .slot = slot, > .guest_phys_addr = guest_phys, > .memory_size = size, > .userspace_addr = (unsigned long)userspace_addr, > }; > > ret = ioctl(kvm->vm_fd, KVM_SET_USER_MEMORY_REGION, &mem); > - if (ret < 0) > - return -errno; > + if (ret < 0) { > + ret = -errno; > + goto out; > + } > } > > - list_add(&bank->list, &kvm->mem_banks); > + list_add(&bank->list, prev_entry); > + kvm->mem_slots++; > + ret = 0; > > - return 0; > +out: > + mutex_unlock(&kvm->mem_banks_lock); > + return ret; > } > > void *guest_flat_to_host(struct kvm *kvm, u64 offset) > diff --git a/vfio/core.c b/vfio/core.c > index 0ed1e6fee6bf..b80ebf3bb913 100644 > --- a/vfio/core.c > +++ b/vfio/core.c > @@ -256,8 +256,14 @@ int vfio_map_region(struct kvm *kvm, struct vfio_device *vdev, > > void vfio_unmap_region(struct kvm *kvm, struct vfio_region *region) > { > + u64 map_size; > + > if (region->host_addr) { > + map_size = ALIGN(region->info.size, PAGE_SIZE); > + kvm__destroy_mem(kvm, region->guest_phys_addr, map_size, > + region->host_addr); > munmap(region->host_addr, region->info.size); > + region->host_addr = NULL; > } else if (region->is_ioport) { > ioport__unregister(kvm, region->port_base); > } else { >
diff --git a/include/kvm/kvm.h b/include/kvm/kvm.h index 50119a8672eb..9428f57a1c0c 100644 --- a/include/kvm/kvm.h +++ b/include/kvm/kvm.h @@ -1,6 +1,7 @@ #ifndef KVM__KVM_H #define KVM__KVM_H +#include "kvm/mutex.h" #include "kvm/kvm-arch.h" #include "kvm/kvm-config.h" #include "kvm/util-init.h" @@ -56,6 +57,7 @@ struct kvm_mem_bank { void *host_addr; u64 size; enum kvm_mem_type type; + u32 slot; }; struct kvm { @@ -72,6 +74,7 @@ struct kvm { u64 ram_size; void *ram_start; u64 ram_pagesize; + struct mutex mem_banks_lock; struct list_head mem_banks; bool nmi_disabled; @@ -106,6 +109,7 @@ void kvm__irq_line(struct kvm *kvm, int irq, int level); void kvm__irq_trigger(struct kvm *kvm, int irq); bool kvm__emulate_io(struct kvm_cpu *vcpu, u16 port, void *data, int direction, int size, u32 count); bool kvm__emulate_mmio(struct kvm_cpu *vcpu, u64 phys_addr, u8 *data, u32 len, u8 is_write); +int kvm__destroy_mem(struct kvm *kvm, u64 guest_phys, u64 size, void *userspace_addr); int kvm__register_mem(struct kvm *kvm, u64 guest_phys, u64 size, void *userspace_addr, enum kvm_mem_type type); static inline int kvm__register_ram(struct kvm *kvm, u64 guest_phys, u64 size, diff --git a/kvm.c b/kvm.c index 57c4ff98ec4c..26f6b9bc58a3 100644 --- a/kvm.c +++ b/kvm.c @@ -157,6 +157,7 @@ struct kvm *kvm__new(void) if (!kvm) return ERR_PTR(-ENOMEM); + mutex_init(&kvm->mem_banks_lock); kvm->sys_fd = -1; kvm->vm_fd = -1; @@ -183,20 +184,84 @@ int kvm__exit(struct kvm *kvm) } core_exit(kvm__exit); +int kvm__destroy_mem(struct kvm *kvm, u64 guest_phys, u64 size, + void *userspace_addr) +{ + struct kvm_userspace_memory_region mem; + struct kvm_mem_bank *bank; + int ret; + + mutex_lock(&kvm->mem_banks_lock); + list_for_each_entry(bank, &kvm->mem_banks, list) + if (bank->guest_phys_addr == guest_phys && + bank->size == size && bank->host_addr == userspace_addr) + break; + + if (&bank->list == &kvm->mem_banks) { + pr_err("Region [%llx-%llx] not found", guest_phys, + guest_phys + size - 1); + ret = -EINVAL; + goto out; + } + + if (bank->type == KVM_MEM_TYPE_RESERVED) { + pr_err("Cannot delete reserved region [%llx-%llx]", + guest_phys, guest_phys + size - 1); + ret = -EINVAL; + goto out; + } + + mem = (struct kvm_userspace_memory_region) { + .slot = bank->slot, + .guest_phys_addr = guest_phys, + .memory_size = 0, + .userspace_addr = (unsigned long)userspace_addr, + }; + + ret = ioctl(kvm->vm_fd, KVM_SET_USER_MEMORY_REGION, &mem); + if (ret < 0) { + ret = -errno; + goto out; + } + + list_del(&bank->list); + free(bank); + kvm->mem_slots--; + ret = 0; + +out: + mutex_unlock(&kvm->mem_banks_lock); + return ret; +} + int kvm__register_mem(struct kvm *kvm, u64 guest_phys, u64 size, void *userspace_addr, enum kvm_mem_type type) { struct kvm_userspace_memory_region mem; struct kvm_mem_bank *merged = NULL; struct kvm_mem_bank *bank; + struct list_head *prev_entry; + u32 slot; int ret; - /* Check for overlap */ + mutex_lock(&kvm->mem_banks_lock); + /* Check for overlap and find first empty slot. */ + slot = 0; + prev_entry = &kvm->mem_banks; list_for_each_entry(bank, &kvm->mem_banks, list) { u64 bank_end = bank->guest_phys_addr + bank->size - 1; u64 end = guest_phys + size - 1; - if (guest_phys > bank_end || end < bank->guest_phys_addr) + if (guest_phys > bank_end || end < bank->guest_phys_addr) { + /* + * Keep the banks sorted ascending by slot, so it's + * easier for us to find a free slot. + */ + if (bank->slot == slot) { + slot++; + prev_entry = &bank->list; + } continue; + } /* Merge overlapping reserved regions */ if (bank->type == KVM_MEM_TYPE_RESERVED && @@ -226,38 +291,50 @@ int kvm__register_mem(struct kvm *kvm, u64 guest_phys, u64 size, kvm_mem_type_to_string(bank->type), bank->guest_phys_addr, bank->guest_phys_addr + bank->size - 1); - return -EINVAL; + ret = -EINVAL; + goto out; } - if (merged) - return 0; + if (merged) { + ret = 0; + goto out; + } bank = malloc(sizeof(*bank)); - if (!bank) - return -ENOMEM; + if (!bank) { + ret = -ENOMEM; + goto out; + } INIT_LIST_HEAD(&bank->list); bank->guest_phys_addr = guest_phys; bank->host_addr = userspace_addr; bank->size = size; bank->type = type; + bank->slot = slot; if (type != KVM_MEM_TYPE_RESERVED) { mem = (struct kvm_userspace_memory_region) { - .slot = kvm->mem_slots++, + .slot = slot, .guest_phys_addr = guest_phys, .memory_size = size, .userspace_addr = (unsigned long)userspace_addr, }; ret = ioctl(kvm->vm_fd, KVM_SET_USER_MEMORY_REGION, &mem); - if (ret < 0) - return -errno; + if (ret < 0) { + ret = -errno; + goto out; + } } - list_add(&bank->list, &kvm->mem_banks); + list_add(&bank->list, prev_entry); + kvm->mem_slots++; + ret = 0; - return 0; +out: + mutex_unlock(&kvm->mem_banks_lock); + return ret; } void *guest_flat_to_host(struct kvm *kvm, u64 offset) diff --git a/vfio/core.c b/vfio/core.c index 0ed1e6fee6bf..b80ebf3bb913 100644 --- a/vfio/core.c +++ b/vfio/core.c @@ -256,8 +256,14 @@ int vfio_map_region(struct kvm *kvm, struct vfio_device *vdev, void vfio_unmap_region(struct kvm *kvm, struct vfio_region *region) { + u64 map_size; + if (region->host_addr) { + map_size = ALIGN(region->info.size, PAGE_SIZE); + kvm__destroy_mem(kvm, region->guest_phys_addr, map_size, + region->host_addr); munmap(region->host_addr, region->info.size); + region->host_addr = NULL; } else if (region->is_ioport) { ioport__unregister(kvm, region->port_base); } else {
When we want to map a device region into the guest address space, first we perform an mmap on the device fd. The resulting VMA is a mapping between host userspace addresses and physical addresses associated with the device. Next, we create a memslot, which populates the stage 2 table with the mappings between guest physical addresses and the device physical adresses. However, when we want to unmap the device from the guest address space, we only call munmap, which destroys the VMA and the stage 2 mappings, but doesn't destroy the memslot and kvmtool's internal mem_bank structure associated with the memslot. This has been perfectly fine so far, because we only unmap a device region when we exit kvmtool. This is will change when we add support for reassignable BARs, and we will have to unmap vfio regions as the guest kernel writes new addresses in the BARs. This can lead to two possible problems: - We refuse to create a valid BAR mapping because of a stale mem_bank structure which belonged to a previously unmapped region. - It is possible that the mmap in vfio_map_region returns the same address that was used to create a memslot, but was unmapped by vfio_unmap_region. Guest accesses to the device memory will fault because the stage 2 mappings are missing, and this can lead to performance degradation. Let's do the right thing and destroy the memslot and the mem_bank struct associated with it when we unmap a vfio region. Set host_addr to NULL after the munmap call so we won't try to unmap an address which is currently used by the process for something else if vfio_unmap_region gets called twice. Signed-off-by: Alexandru Elisei <alexandru.elisei@arm.com> --- include/kvm/kvm.h | 4 ++ kvm.c | 101 ++++++++++++++++++++++++++++++++++++++++------ vfio/core.c | 6 +++ 3 files changed, 99 insertions(+), 12 deletions(-)