user/sven/linux.git/virt, branch v4.4.148

KVM/Eventfd: Avoid crash when assign and deassign specific eventfd in parallel.

2018-07-25T08:18:15Z

commit b5020a8e6b54d2ece80b1e7dedb33c79a40ebd47 upstream. Syzbot reports crashes in kvm_irqfd_assign(), caused by use-after-free when kvm_irqfd_assign() and kvm_irqfd_deassign() run in parallel for one specific eventfd. When the assign path hasn't finished but irqfd has been added to kvm->irqfds.items list, another thead may deassign the eventfd and free struct kvm_kernel_irqfd(). The assign path then uses the struct kvm_kernel_irqfd that has been freed by deassign path. To avoid such issue, keep irqfd under kvm->irq_srcu protection after the irqfd has been added to kvm->irqfds.items list, and call synchronize_srcu() in irq_shutdown() to make sure that irqfd has been fully initialized in the assign path. Reported-by: Dmitry Vyukov Cc: Paolo Bonzini Cc: Radim Krčmář Cc: Dmitry Vyukov Signed-off-by: Tianyu Lan Cc: stable@vger.kernel.org Signed-off-by: Paolo Bonzini Signed-off-by: Greg Kroah-Hartman

KVM: mmu: Fix overlap between public and private memslots

2018-03-11T15:19:45Z

commit b28676bb8ae4569cced423dc2a88f7cb319d5379 upstream. Reported by syzkaller: pte_list_remove: ffff9714eb1f8078 0->BUG ------------[ cut here ]------------ kernel BUG at arch/x86/kvm/mmu.c:1157! invalid opcode: 0000 [#1] SMP RIP: 0010:pte_list_remove+0x11b/0x120 [kvm] Call Trace: drop_spte+0x83/0xb0 [kvm] mmu_page_zap_pte+0xcc/0xe0 [kvm] kvm_mmu_prepare_zap_page+0x81/0x4a0 [kvm] kvm_mmu_invalidate_zap_all_pages+0x159/0x220 [kvm] kvm_arch_flush_shadow_all+0xe/0x10 [kvm] kvm_mmu_notifier_release+0x6c/0xa0 [kvm] ? kvm_mmu_notifier_release+0x5/0xa0 [kvm] __mmu_notifier_release+0x79/0x110 ? __mmu_notifier_release+0x5/0x110 exit_mmap+0x15a/0x170 ? do_exit+0x281/0xcb0 mmput+0x66/0x160 do_exit+0x2c9/0xcb0 ? __context_tracking_exit.part.5+0x4a/0x150 do_group_exit+0x50/0xd0 SyS_exit_group+0x14/0x20 do_syscall_64+0x73/0x1f0 entry_SYSCALL64_slow_path+0x25/0x25 The reason is that when creates new memslot, there is no guarantee for new memslot not overlap with private memslots. This can be triggered by the following program: #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include long r[16]; int main() { void *p = valloc(0x4000); r[2] = open("/dev/kvm", 0); r[3] = ioctl(r[2], KVM_CREATE_VM, 0x0ul); uint64_t addr = 0xf000; ioctl(r[3], KVM_SET_IDENTITY_MAP_ADDR, &addr); r[6] = ioctl(r[3], KVM_CREATE_VCPU, 0x0ul); ioctl(r[3], KVM_SET_TSS_ADDR, 0x0ul); ioctl(r[6], KVM_RUN, 0); ioctl(r[6], KVM_RUN, 0); struct kvm_userspace_memory_region mr = { .slot = 0, .flags = KVM_MEM_LOG_DIRTY_PAGES, .guest_phys_addr = 0xf000, .memory_size = 0x4000, .userspace_addr = (uintptr_t) p }; ioctl(r[3], KVM_SET_USER_MEMORY_REGION, &mr); return 0; } This patch fixes the bug by not adding a new memslot even if it overlaps with private memslots. Reported-by: Dmitry Vyukov Cc: Paolo Bonzini Cc: Radim Krčmář Cc: Dmitry Vyukov Cc: Eric Biggers Cc: stable@vger.kernel.org Signed-off-by: Wanpeng Li

KVM: pci-assign: do not map smm memory slot pages in vt-d page tables

2017-12-25T13:22:13Z

[ Upstream commit 0292e169b2d9c8377a168778f0b16eadb1f578fd ] or VM memory are not put thus leaked in kvm_iommu_unmap_memslots() when destroy VM. This is consistent with current vfio implementation. Signed-off-by: herongguang Signed-off-by: Paolo Bonzini Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

KVM: arm/arm64: Fix occasional warning from the timer work function

2017-12-09T17:42:42Z

[ Upstream commit 63e41226afc3f7a044b70325566fa86ac3142538 ] When a VCPU blocks (WFI) and has programmed the vtimer, we program a soft timer to expire in the future to wake up the vcpu thread when appropriate. Because such as wake up involves a vcpu kick, and the timer expire function can get called from interrupt context, and the kick may sleep, we have to schedule the kick in the work function. The work function currently has a warning that gets raised if it turns out that the timer shouldn't fire when it's run, which was added because the idea was that in that case the work should never have been cancelled. However, it turns out that this whole thing is racy and we can get spurious warnings. The problem is that we clear the armed flag in the work function, which may run in parallel with the kvm_timer_unschedule->timer_disarm() call. This results in a possible situation where the timer_disarm() call does not call cancel_work_sync(), which effectively synchronizes the completion of the work function with running the VCPU. As a result, the VCPU thread proceeds before the work function completees, causing changes to the timer state such that kvm_timer_should_fire(vcpu) returns false in the work function. All we do in the work function is to kick the VCPU, and an occasional rare extra kick never harmed anyone. Since the race above is extremely rare, we don't bother checking if the race happens but simply remove the check and the clearing of the armed flag from the work function. Reported-by: Matthias Brugger Reviewed-by: Marc Zyngier Signed-off-by: Christoffer Dall Signed-off-by: Marc Zyngier Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

vfio: New external user group/file match

2017-07-27T22:06:07Z

commit 5d6dee80a1e94cc284d03e06d930e60e8d3ecf7d upstream. At the point where the kvm-vfio pseudo device wants to release its vfio group reference, we can't always acquire a new reference to make that happen. The group can be in a state where we wouldn't allow a new reference to be added. This new helper function allows a caller to match a file to a group to facilitate this. Given a file and group, report if they match. Thus the caller needs to already have a group reference to match to the file. This allows the deletion of a group without acquiring a new reference. Signed-off-by: Alex Williamson Reviewed-by: Eric Auger Reviewed-by: Paolo Bonzini Tested-by: Eric Auger Signed-off-by: Greg Kroah-Hartman

KVM: kvm_io_bus_unregister_dev() should never fail

2017-04-08T07:53:32Z

commit 90db10434b163e46da413d34db8d0e77404cc645 upstream. No caller currently checks the return value of kvm_io_bus_unregister_dev(). This is evil, as all callers silently go on freeing their device. A stale reference will remain in the io_bus, getting at least used again, when the iobus gets teared down on kvm_destroy_vm() - leading to use after free errors. There is nothing the callers could do, except retrying over and over again. So let's simply remove the bus altogether, print an error and make sure no one can access this broken bus again (returning -ENOMEM on any attempt to access it). Fixes: e93f8a0f821e ("KVM: convert io_bus to SRCU") Reported-by: Dmitry Vyukov Reviewed-by: Cornelia Huck Signed-off-by: David Hildenbrand Signed-off-by: Paolo Bonzini Signed-off-by: Greg Kroah-Hartman

KVM: x86: clear bus pointer when destroyed

2017-04-08T07:53:31Z

commit df630b8c1e851b5e265dc2ca9c87222e342c093b upstream. When releasing the bus, let's clear the bus pointers to mark it out. If any further device unregister happens on this bus, we know that we're done if we found the bus being released already. Signed-off-by: Peter Xu Signed-off-by: Radim Krčmář Signed-off-by: Greg Kroah-Hartman

KVM: eventfd: fix NULL deref irqbypass consumer

2017-01-19T19:17:19Z

commit 4f3dbdf47e150016aacd734e663347fcaa768303 upstream. Reported syzkaller: BUG: unable to handle kernel NULL pointer dereference at 0000000000000008 IP: irq_bypass_unregister_consumer+0x9d/0xb70 [irqbypass] PGD 0 Oops: 0002 [#1] SMP CPU: 1 PID: 125 Comm: kworker/1:1 Not tainted 4.9.0+ #1 Workqueue: kvm-irqfd-cleanup irqfd_shutdown [kvm] task: ffff9bbe0dfbb900 task.stack: ffffb61802014000 RIP: 0010:irq_bypass_unregister_consumer+0x9d/0xb70 [irqbypass] Call Trace: irqfd_shutdown+0x66/0xa0 [kvm] process_one_work+0x16b/0x480 worker_thread+0x4b/0x500 kthread+0x101/0x140 ? process_one_work+0x480/0x480 ? kthread_create_on_node+0x60/0x60 ret_from_fork+0x25/0x30 RIP: irq_bypass_unregister_consumer+0x9d/0xb70 [irqbypass] RSP: ffffb61802017e20 CR2: 0000000000000008 The syzkaller folks reported a NULL pointer dereference that due to unregister an consumer which fails registration before. The syzkaller creates two VMs w/ an equal eventfd occasionally. So the second VM fails to register an irqbypass consumer. It will make irqfd as inactive and queue an workqueue work to shutdown irqfd and unregister the irqbypass consumer when eventfd is closed. However, the second consumer has been initialized though it fails registration. So the token(same as the first VM's) is taken to unregister the consumer through the workqueue, the consumer of the first VM is found and unregistered, then NULL deref incurred in the path of deleting consumer from the consumers list. This patch fixes it by making irq_bypass_register/unregister_consumer() looks for the consumer entry based on consumer pointer itself instead of token matching. Reported-by: Dmitry Vyukov Suggested-by: Alex Williamson Cc: Paolo Bonzini Cc: Radim Krčmář Cc: Dmitry Vyukov Cc: Alex Williamson Signed-off-by: Wanpeng Li Signed-off-by: Paolo Bonzini Signed-off-by: Greg Kroah-Hartman

KVM: nVMX: Fix memory corruption when using VMCS shadowing

2016-08-20T16:09:18Z

commit 2f1fe81123f59271bddda673b60116bde9660385 upstream. When freeing the nested resources of a vcpu, there is an assumption that the vcpu's vmcs01 is the current VMCS on the CPU that executes nested_release_vmcs12(). If this assumption is violated, the vcpu's vmcs01 may be made active on multiple CPUs at the same time, in violation of Intel's specification. Moreover, since the vcpu's vmcs01 is not VMCLEARed on every CPU on which it is active, it can linger in a CPU's VMCS cache after it has been freed and potentially repurposed. Subsequent eviction from the CPU's VMCS cache on a capacity miss can result in memory corruption. It is not sufficient for vmx_free_vcpu() to call vmx_load_vmcs01(). If the vcpu in question was last loaded on a different CPU, it must be migrated to the current CPU before calling vmx_load_vmcs01(). Signed-off-by: Jim Mattson Signed-off-by: Paolo Bonzini Signed-off-by: Greg Kroah-Hartman

kvm: Fix irq route entries exceeding KVM_MAX_IRQ_ROUTES

2016-07-27T16:47:31Z

commit caf1ff26e1aa178133df68ac3d40815fed2187d9 upstream. These days, we experienced one guest crash with 8 cores and 3 disks, with qemu error logs as bellow: qemu-system-x86_64: /build/qemu-2.0.0/kvm-all.c:984: kvm_irqchip_commit_routes: Assertion `ret == 0' failed. And then we found one patch(bdf026317d) in qemu tree, which said could fix this bug. Execute the following script will reproduce the BUG quickly: irq_affinity.sh ======================================================================== vda_irq_num=25 vdb_irq_num=27 while [ 1 ] do for irq in {1,2,4,8,10,20,40,80} do echo $irq > /proc/irq/$vda_irq_num/smp_affinity echo $irq > /proc/irq/$vdb_irq_num/smp_affinity dd if=/dev/vda of=/dev/zero bs=4K count=100 iflag=direct dd if=/dev/vdb of=/dev/zero bs=4K count=100 iflag=direct done done ======================================================================== The following qemu log is added in the qemu code and is displayed when this bug reproduced: kvm_irqchip_commit_routes: max gsi: 1008, nr_allocated_irq_routes: 1024, irq_routes->nr: 1024, gsi_count: 1024. That's to say when irq_routes->nr == 1024, there are 1024 routing entries, but in the kernel code when routes->nr >= 1024, will just return -EINVAL; The nr is the number of the routing entries which is in of [1 ~ KVM_MAX_IRQ_ROUTES], not the index in [0 ~ KVM_MAX_IRQ_ROUTES - 1]. This patch fix the BUG above. Signed-off-by: Xiubo Li Signed-off-by: Wei Tang Signed-off-by: Zhang Zhuoyu Signed-off-by: Paolo Bonzini Signed-off-by: Greg Kroah-Hartman