KVM generic changes for 6.11
- Enable halt poll shrinking by default, as Intel found it to be a clear win.
- Setup empty IRQ routing when creating a VM to avoid having to synchronize
SRCU when creating a split IRQCHIP on x86.
- Rework the sched_in/out() paths to replace kvm_arch_sched_in() with a flag
that arch code can use for hooking both sched_in() and sched_out().
- Take the vCPU @id as an "unsigned long" instead of "u32" to avoid
truncating a bogus value from userspace, e.g. to help userspace detect bugs.
- Mark a vCPU as preempted if and only if it's scheduled out while in the
KVM_RUN loop, e.g. to avoid marking it preempted and thus writing guest
memory when retrieving guest state during live migration blackout.
- A few minor cleanups
KVM Xen:
Fix a bug where KVM fails to check the validity of an incoming userspace
virtual address and tries to activate a gfn_to_pfn_cache with a kernel address.
Pre-population has been requested several times to mitigate KVM page faults
during guest boot or after live migration. It is also required by TDX
before filling in the initial guest memory with measured contents.
Introduce it as a generic API.
The flags AS_UNMOVABLE and AS_INACCESSIBLE were both added just for guest_memfd;
AS_UNMOVABLE is already in existing versions of Linux, while AS_INACCESSIBLE was
acked for inclusion in 6.11.
But really, they are the same thing: only guest_memfd uses them, at least for
now, and guest_memfd pages are unmovable because they should not be
accessed by the CPU.
So merge them into one; use the AS_INACCESSIBLE name which is more comprehensive.
At the same time, this fixes an embarrassing bug where AS_INACCESSIBLE was used
as a bit mask, despite it being just a bit index.
The bug was mostly benign, because AS_INACCESSIBLE's bit representation (1010)
corresponded to setting AS_UNEVICTABLE (which is already set) and AS_ENOSPC
(except no async writes can happen on the guest_memfd). So the AS_INACCESSIBLE
flag simply had no effect.
Fixes: 1d23040caa ("KVM: guest_memfd: Use AS_INACCESSIBLE when creating guest_memfd inode")
Fixes: c72ceafbd1 ("mm: Introduce AS_INACCESSIBLE for encrypted/confidential memory")
Cc: linux-mm@kvack.org
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: David Hildenbrand <david@redhat.com>
Tested-by: Michael Roth <michael.roth@amd.com>
Reviewed-by: Michael Roth <michael.roth@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add a module description for kvm.ko to fix a 'make W=1' warning:
WARNING: modpost: missing MODULE_DESCRIPTION() in arch/x86/kvm/kvm.o
Opportunistically update kvm_main.c's comically stale file comment to
match the module description.
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Link: https://lore.kernel.org/r/20240622-md-kvm-v2-1-29a60f7c48b1@quicinc.com
[sean: split x86 changes to a separate commit, remove stale VT-x comment]
Signed-off-by: Sean Christopherson <seanjc@google.com>
KVM fixes for 6.10
- Fix a "shift too big" goof in the KVM_SEV_INIT2 selftest.
- Compute the max mappable gfn for KVM selftests on x86 using GuestMaxPhyAddr
from KVM's supported CPUID (if it's available).
- Fix a race in kvm_vcpu_on_spin() by ensuring loads and stores are atomic.
- Fix technically benign bug in __kvm_handle_hva_range() where KVM consumes
the return from a void-returning function as if it were a boolean.
kvm_gmem_populate() is a potentially lengthy operation that can involve
multiple calls to the firmware. Interrupt it if a signal arrives.
Fixes: 1f6c06b177 ("KVM: guest_memfd: Add interface for populating gmem pages with user data")
Cc: Isaku Yamahata <isaku.yamahata@intel.com>
Cc: Michael Roth <michael.roth@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Function kvm_reset_dirty_gfn may be called with parameters cur_slot /
cur_offset / mask are all zero, it does not represent real dirty page.
It is not necessary to clear dirty page in this condition. Also return
value of macro __fls() is undefined if mask is zero which is called in
funciton kvm_reset_dirty_gfn(). Here just return.
Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Message-ID: <20240613122803.1031511-1-maobibo@loongson.cn>
[Move the conditional inside kvm_reset_dirty_gfn; suggested by
Sean Christopherson. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If kvm_gmem_get_pfn() detects an hwpoisoned page, it returns -EHWPOISON
but it does not put back the reference that kvm_gmem_get_folio() had
grabbed. Add the forgotten folio_put().
Fixes: a7800aa80e ("KVM: Add KVM_CREATE_GUEST_MEMFD ioctl() for guest-specific backing memory")
Cc: stable@vger.kernel.org
Reviewed-by: Liam Merwick <liam.merwick@oracle.com>
Reviewed-by: Isaku Yamahata <isaku.yamahata@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Some allocations done by KVM are temporary, they are created as result
of program actions, but can't exists for arbitrary long times.
They should have been GFP_TEMPORARY (rip!).
OTOH, kvm-nx-lpage-recovery and kvm-pit kernel threads exist for as long
as VM exists but their task_struct memory is not accounted.
This is story for another day.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Message-ID: <c0122f66-f428-417e-a360-b25fc0f154a0@p183>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Mark a vCPU as preempted/ready if-and-only-if it's scheduled out while
running. i.e. Do not mark a vCPU preempted/ready if it's scheduled out
during a non-KVM_RUN ioctl() or when userspace is doing KVM_RUN with
immediate_exit.
Commit 54aa83c901 ("KVM: x86: do not set st->preempted when going back
to user space") stopped marking a vCPU as preempted when returning to
userspace, but if userspace then invokes a KVM vCPU ioctl() that gets
preempted, the vCPU will be marked preempted/ready. This is arguably
incorrect behavior since the vCPU was not actually preempted while the
guest was running, it was preempted while doing something on behalf of
userspace.
Marking a vCPU preempted iff its running also avoids KVM dirtying guest
memory after userspace has paused vCPUs, e.g. for live migration, which
allows userspace to collect the final dirty bitmap before or in parallel
with saving vCPU state, without having to worry about saving vCPU state
triggering writes to guest memory.
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: David Matlack <dmatlack@google.com>
Link: https://lore.kernel.org/r/20240503181734.1467938-4-dmatlack@google.com
[sean: massage changelog]
Signed-off-by: Sean Christopherson <seanjc@google.com>
Ensure that any new KVM code that references immediate_exit gets extra
scrutiny by renaming it to immediate_exit__unsafe in kernel code.
All fields in struct kvm_run are subject to TOCTOU races since they are
mapped into userspace, which may be malicious or buggy. To protect KVM,
introduces a new macro that appends __unsafe to select field names in
struct kvm_run, hinting to developers and reviewers that accessing such
fields must be done carefully.
Apply the new macro to immediate_exit, since userspace can make
immediate_exit inconsistent with vcpu->wants_to_run, i.e. accessing
immediate_exit directly could lead to unexpected bugs in the future.
Signed-off-by: David Matlack <dmatlack@google.com>
Link: https://lore.kernel.org/r/20240503181734.1467938-3-dmatlack@google.com
[sean: massage changelog]
Signed-off-by: Sean Christopherson <seanjc@google.com>
Introduce vcpu->wants_to_run to indicate when a vCPU is in its core run
loop, i.e. when the vCPU is running the KVM_RUN ioctl and immediate_exit
was not set.
Replace all references to vcpu->run->immediate_exit with
!vcpu->wants_to_run to avoid TOCTOU races with userspace. For example, a
malicious userspace could invoked KVM_RUN with immediate_exit=true and
then after KVM reads it to set wants_to_run=false, flip it to false.
This would result in the vCPU running in KVM_RUN with
wants_to_run=false. This wouldn't cause any real bugs today but is a
dangerous landmine.
Signed-off-by: David Matlack <dmatlack@google.com>
Link: https://lore.kernel.org/r/20240503181734.1467938-2-dmatlack@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
If, on a 64 bit system, a vCPU ID is provided that has the upper 32 bits
set to a non-zero value, it may get accepted if the truncated to 32 bits
integer value is below KVM_MAX_VCPU_IDS and 'max_vcpus'. This feels very
wrong and triggered the reporting logic of PaX's SIZE_OVERFLOW plugin.
Instead of silently truncating and accepting such values, pass the full
value to kvm_vm_ioctl_create_vcpu() and make the existing limit checks
return an error.
Even if this is a userland ABI breaking change, no sane userland could
have ever relied on that behaviour.
Reported-by: PaX's SIZE_OVERFLOW plugin running on grsecurity's syzkaller
Fixes: 6aa8b732ca ("[PATCH] kvm: userspace interface")
Cc: Emese Revfy <re.emese@gmail.com>
Cc: PaX Team <pageexec@freemail.hu>
Signed-off-by: Mathias Krause <minipli@grsecurity.net>
Link: https://lore.kernel.org/r/20240614202859.3597745-2-minipli@grsecurity.net
[sean: tweak comment about INT_MAX assertion]
Signed-off-by: Sean Christopherson <seanjc@google.com>
Bail from outer address space loop, not just the inner memslot loop, when
a "null" handler is encountered by __kvm_handle_hva_range(), which is the
intended behavior. On x86, which has multiple address spaces thanks to
SMM emulation, breaking from just the memslot loop results in undefined
behavior due to assigning the non-existent return value from kvm_null_fn()
to a bool.
In practice, the bug is benign as kvm_mmu_notifier_invalidate_range_end()
is the only caller that passes handler=kvm_null_fn, and it doesn't set
flush_on_ret, i.e. assigning garbage to r.ret is ultimately ignored. And
for most configuration the compiler elides the entire sequence, i.e. there
is no undefined behavior at runtime.
------------[ cut here ]------------
UBSAN: invalid-load in arch/x86/kvm/../../../virt/kvm/kvm_main.c:655:10
load of value 160 is not a valid value for type '_Bool'
CPU: 370 PID: 8246 Comm: CPU 0/KVM Not tainted 6.8.2-amdsos-build58-ubuntu-22.04+ #1
Hardware name: AMD Corporation Sh54p/Sh54p, BIOS WPC4429N 04/25/2024
Call Trace:
<TASK>
dump_stack_lvl+0x48/0x60
ubsan_epilogue+0x5/0x30
__ubsan_handle_load_invalid_value+0x79/0x80
kvm_mmu_notifier_invalidate_range_end.cold+0x18/0x4f [kvm]
__mmu_notifier_invalidate_range_end+0x63/0xe0
__split_huge_pmd+0x367/0xfc0
do_huge_pmd_wp_page+0x1cc/0x380
__handle_mm_fault+0x8ee/0xe50
handle_mm_fault+0xe4/0x4a0
__get_user_pages+0x190/0x840
get_user_pages_unlocked+0xe0/0x590
hva_to_pfn+0x114/0x550 [kvm]
kvm_faultin_pfn+0xed/0x5b0 [kvm]
kvm_tdp_page_fault+0x123/0x170 [kvm]
kvm_mmu_page_fault+0x244/0xaa0 [kvm]
vcpu_enter_guest+0x592/0x1070 [kvm]
kvm_arch_vcpu_ioctl_run+0x145/0x8a0 [kvm]
kvm_vcpu_ioctl+0x288/0x6d0 [kvm]
__x64_sys_ioctl+0x8f/0xd0
do_syscall_64+0x77/0x120
entry_SYSCALL_64_after_hwframe+0x6e/0x76
</TASK>
---[ end trace ]---
Fixes: 071064f14d ("KVM: Don't take mmu_lock for range invalidation unless necessary")
Signed-off-by: Babu Moger <babu.moger@amd.com>
Link: https://lore.kernel.org/r/b8723d39903b64c241c50f5513f804390c7b5eec.1718203311.git.babu.moger@amd.com
[sean: massage changelog]
Signed-off-by: Sean Christopherson <seanjc@google.com>
The error path for OOM when allocating buses used to return -ENOMEM using
the local variable 'r', where 'r' was initialized at the top of the
function. But a new "r = kvm_init_irq_routing(kvm);" was introduced in
the middle of the function, so now the error code is not set and it
eventually leads to a NULL dereference due to kvm_dev_ioctl_create_vm()
thinking kvm_create_vm() succeeded. Set the error code back to -ENOMEM.
Opportunistically tweak the logic to pre-set "r = -ENOMEM" immediately
before the flows that can fail due to memory allocation failure to make
it less likely that the bug recurs in the future.
Fixes: fbe4a7e881 ("KVM: Setup empty IRQ routing when creating a VM")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/02051e0a-09d8-49a2-917f-7c2f278a1ba1@moroto.mountain
[sean: tweak all of the "r = -ENOMEM" sites, massage changelog]
Signed-off-by: Sean Christopherson <seanjc@google.com>
Add a kvm_vcpu.scheduled_out flag to track if a vCPU is in the process of
being scheduled out (vCPU put path), or if the vCPU is being reloaded
after being scheduled out (vCPU load path). In the short term, this will
allow dropping kvm_arch_sched_in(), as arch code can query scheduled_out
during kvm_arch_vcpu_load().
Longer term, scheduled_out opens up other potential optimizations, without
creating subtle/brittle dependencies. E.g. it allows KVM to keep guest
state (that is managed via kvm_arch_vcpu_{load,put}()) loaded across
kvm_sched_{out,in}(), if KVM knows the state isn't accessed by the host
kernel. Forcing arch code to coordinate between kvm_arch_sched_{in,out}()
and kvm_arch_vcpu_{load,put}() is awkward, not reusable, and relies on the
exact ordering of calls into arch code.
Adding scheduled_out also obviates the need for a kvm_arch_sched_out()
hook, e.g. if arch code needs to do something novel when putting vCPU
state.
And even if KVM never uses scheduled_out for anything beyond dropping
kvm_arch_sched_in(), just being able to remove all of the arch stubs makes
it worth adding the flag.
Link: https://lore.kernel.org/all/20240430224431.490139-1-seanjc@google.com
Cc: Oliver Upton <oliver.upton@linux.dev>
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Acked-by: Kai Huang <kai.huang@intel.com>
Link: https://lore.kernel.org/r/20240522014013.1672962-2-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Setup empty IRQ routing during VM creation so that x86 and s390 don't need
to set empty/dummy IRQ routing during KVM_CREATE_IRQCHIP (in future
patches). Initializing IRQ routing before there are any potential readers
allows KVM to avoid the synchronize_srcu() in kvm_set_irq_routing(), which
can introduces 20+ milliseconds of latency in the VM creation path.
Ensuring that all VMs have non-NULL IRQ routing also hardens KVM against
misbehaving userspace VMMs, e.g. RISC-V dynamically instantiates its
interrupt controller, but doesn't override kvm_arch_intc_initialized() or
kvm_arch_irqfd_allowed(), and so can likely reach kvm_irq_map_gsi()
without fully initialized IRQ routing.
Signed-off-by: Yi Wang <foxywang@tencent.com>
Acked-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Link: https://lore.kernel.org/r/20240506101751.3145407-2-foxywang@tencent.com
[sean: init refcount after IRQ routing, fix stub, massage changelog]
Signed-off-by: Sean Christopherson <seanjc@google.com>
Use {READ,WRITE}_ONCE() to access kvm->last_boosted_vcpu to ensure the
loads and stores are atomic. In the extremely unlikely scenario the
compiler tears the stores, it's theoretically possible for KVM to attempt
to get a vCPU using an out-of-bounds index, e.g. if the write is split
into multiple 8-bit stores, and is paired with a 32-bit load on a VM with
257 vCPUs:
CPU0 CPU1
last_boosted_vcpu = 0xff;
(last_boosted_vcpu = 0x100)
last_boosted_vcpu[15:8] = 0x01;
i = (last_boosted_vcpu = 0x1ff)
last_boosted_vcpu[7:0] = 0x00;
vcpu = kvm->vcpu_array[0x1ff];
As detected by KCSAN:
BUG: KCSAN: data-race in kvm_vcpu_on_spin [kvm] / kvm_vcpu_on_spin [kvm]
write to 0xffffc90025a92344 of 4 bytes by task 4340 on cpu 16:
kvm_vcpu_on_spin (arch/x86/kvm/../../../virt/kvm/kvm_main.c:4112) kvm
handle_pause (arch/x86/kvm/vmx/vmx.c:5929) kvm_intel
vmx_handle_exit (arch/x86/kvm/vmx/vmx.c:?
arch/x86/kvm/vmx/vmx.c:6606) kvm_intel
vcpu_run (arch/x86/kvm/x86.c:11107 arch/x86/kvm/x86.c:11211) kvm
kvm_arch_vcpu_ioctl_run (arch/x86/kvm/x86.c:?) kvm
kvm_vcpu_ioctl (arch/x86/kvm/../../../virt/kvm/kvm_main.c:?) kvm
__se_sys_ioctl (fs/ioctl.c:52 fs/ioctl.c:904 fs/ioctl.c:890)
__x64_sys_ioctl (fs/ioctl.c:890)
x64_sys_call (arch/x86/entry/syscall_64.c:33)
do_syscall_64 (arch/x86/entry/common.c:?)
entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130)
read to 0xffffc90025a92344 of 4 bytes by task 4342 on cpu 4:
kvm_vcpu_on_spin (arch/x86/kvm/../../../virt/kvm/kvm_main.c:4069) kvm
handle_pause (arch/x86/kvm/vmx/vmx.c:5929) kvm_intel
vmx_handle_exit (arch/x86/kvm/vmx/vmx.c:?
arch/x86/kvm/vmx/vmx.c:6606) kvm_intel
vcpu_run (arch/x86/kvm/x86.c:11107 arch/x86/kvm/x86.c:11211) kvm
kvm_arch_vcpu_ioctl_run (arch/x86/kvm/x86.c:?) kvm
kvm_vcpu_ioctl (arch/x86/kvm/../../../virt/kvm/kvm_main.c:?) kvm
__se_sys_ioctl (fs/ioctl.c:52 fs/ioctl.c:904 fs/ioctl.c:890)
__x64_sys_ioctl (fs/ioctl.c:890)
x64_sys_call (arch/x86/entry/syscall_64.c:33)
do_syscall_64 (arch/x86/entry/common.c:?)
entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130)
value changed: 0x00000012 -> 0x00000000
Fixes: 217ece6129 ("KVM: use yield_to instead of sleep in kvm_vcpu_on_spin")
Cc: stable@vger.kernel.org
Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240510092353.2261824-1-leitao@debian.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Pull base x86 KVM support for running SEV-SNP guests from Michael Roth:
* add some basic infrastructure and introduces a new KVM_X86_SNP_VM
vm_type to handle differences versus the existing KVM_X86_SEV_VM and
KVM_X86_SEV_ES_VM types.
* implement the KVM API to handle the creation of a cryptographic
launch context, encrypt/measure the initial image into guest memory,
and finalize it before launching it.
* implement handling for various guest-generated events such as page
state changes, onlining of additional vCPUs, etc.
* implement the gmem/mmu hooks needed to prepare gmem-allocated pages
before mapping them into guest private memory ranges as well as
cleaning them up prior to returning them to the host for use as
normal memory. Because those cleanup hooks supplant certain
activities like issuing WBINVDs during KVM MMU invalidations, avoid
duplicating that work to avoid unecessary overhead.
This merge leaves out support support for attestation guest requests
and for loading the signing keys to be used for attestation requests.
