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.
Now that KVM does NOT gift async #PF workers a "struct kvm" reference,
don't bother skipping "done" workers when flushing/canceling queued
workers, as the deadlock that was being fudged around can no longer occur.
When workers, i.e. async_pf_execute(), were gifted a referenced, it was
possible for a worker to put the last reference and trigger VM destruction,
i.e. trigger flushing of a workqueue from a worker in said workqueue.
Note, there is no actual lock, the deadlock was that a worker will be
stuck waiting for itself (the workqueue code simulates a lock/unlock via
lock_map_{acquire,release}()).
Skipping "done" workers isn't problematic per se, but using work->vcpu as
a "done" flag is confusing, e.g. it's not clear that async_pf.lock is
acquired to protect the work->vcpu, NOT the processing of async_pf.queue
(which is protected by vcpu->mutex).
This reverts commit 22583f0d9c.
Suggested-by: Xu Yilun <yilun.xu@linux.intel.com>
Link: https://lore.kernel.org/r/20240423191649.2885257-1-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Default halt_poll_ns_shrink value of 0 always resets polling interval
to 0 on an un-successful poll where vcpu wakeup is not received. This is
mostly to avoid pointless polling for more number of shorter intervals. But
disabled shrink assumes vcpu wakeup is less likely to be received in
subsequent shorter polling intervals. Another side effect of 0 shrink value
is that, even on a successful poll if total block time was greater than
current polling interval, the polling interval starts over from 0 instead
of shrinking by a factor.
Enabling shrink with value of 2 allows the polling interval to gradually
decrement in case of un-successful poll events as well. This gives a fair
chance for successful polling events in subsequent polling intervals rather
than resetting it to 0 and starting over from grow_start.
Below kvm stat log snippet shows interleaved growth and shrinking of
polling interval:
87162647182125: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 10000 (grow 0)
87162647637763: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 20000 (grow 10000)
87162649627943: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 40000 (grow 20000)
87162650892407: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 20000 (shrink 40000)
87162651540378: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 40000 (grow 20000)
87162652276768: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 20000 (shrink 40000)
87162652515037: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 40000 (grow 20000)
87162653383787: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 20000 (shrink 40000)
87162653627670: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 10000 (shrink 20000)
87162653796321: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 20000 (grow 10000)
87162656171645: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 10000 (shrink 20000)
87162661607487: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 0 (shrink 10000)
Having both grow and shrink enabled creates a balance in polling interval
growth and shrink behavior. Tests show improved successful polling attempt
ratio which contribute to VM performance. Power penalty is quite negligible
as shrunk polling intervals create bursts of very short durations.
Performance assessment results show 3-6% improvements in CPU+GPU, Memory
and Storage Android VM workloads whereas 5-9% improvement in average FPS of
gaming VM workloads.
Power penalty is below 1% where host OS is either idle or running a
native workload having 2 VMs enabled. CPU/GPU intensive gaming workloads
as well do not show any increased power overhead with shrink enabled.
Co-developed-by: Rajendran Jaishankar <jaishankar.rajendran@intel.com>
Signed-off-by: Rajendran Jaishankar <jaishankar.rajendran@intel.com>
Signed-off-by: Parshuram Sangle <parshuram.sangle@intel.com>
Link: https://lore.kernel.org/r/20231102154628.2120-2-parshuram.sangle@intel.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Pull mm updates from Andrew Morton:
"The usual shower of singleton fixes and minor series all over MM,
documented (hopefully adequately) in the respective changelogs.
Notable series include:
- Lucas Stach has provided some page-mapping cleanup/consolidation/
maintainability work in the series "mm/treewide: Remove pXd_huge()
API".
- In the series "Allow migrate on protnone reference with
MPOL_PREFERRED_MANY policy", Donet Tom has optimized mempolicy's
MPOL_PREFERRED_MANY mode, yielding almost doubled performance in
one test.
- In their series "Memory allocation profiling" Kent Overstreet and
Suren Baghdasaryan have contributed a means of determining (via
/proc/allocinfo) whereabouts in the kernel memory is being
allocated: number of calls and amount of memory.
- Matthew Wilcox has provided the series "Various significant MM
patches" which does a number of rather unrelated things, but in
largely similar code sites.
- In his series "mm: page_alloc: freelist migratetype hygiene"
Johannes Weiner has fixed the page allocator's handling of
migratetype requests, with resulting improvements in compaction
efficiency.
- In the series "make the hugetlb migration strategy consistent"
Baolin Wang has fixed a hugetlb migration issue, which should
improve hugetlb allocation reliability.
- Liu Shixin has hit an I/O meltdown caused by readahead in a
memory-tight memcg. Addressed in the series "Fix I/O high when
memory almost met memcg limit".
- In the series "mm/filemap: optimize folio adding and splitting"
Kairui Song has optimized pagecache insertion, yielding ~10%
performance improvement in one test.
- Baoquan He has cleaned up and consolidated the early zone
initialization code in the series "mm/mm_init.c: refactor
free_area_init_core()".
- Baoquan has also redone some MM initializatio code in the series
"mm/init: minor clean up and improvement".
- MM helper cleanups from Christoph Hellwig in his series "remove
follow_pfn".
- More cleanups from Matthew Wilcox in the series "Various
page->flags cleanups".
- Vlastimil Babka has contributed maintainability improvements in the
series "memcg_kmem hooks refactoring".
- More folio conversions and cleanups in Matthew Wilcox's series:
"Convert huge_zero_page to huge_zero_folio"
"khugepaged folio conversions"
"Remove page_idle and page_young wrappers"
"Use folio APIs in procfs"
"Clean up __folio_put()"
"Some cleanups for memory-failure"
"Remove page_mapping()"
"More folio compat code removal"
- David Hildenbrand chipped in with "fs/proc/task_mmu: convert
hugetlb functions to work on folis".
- Code consolidation and cleanup work related to GUP's handling of
hugetlbs in Peter Xu's series "mm/gup: Unify hugetlb, part 2".
- Rick Edgecombe has developed some fixes to stack guard gaps in the
series "Cover a guard gap corner case".
- Jinjiang Tu has fixed KSM's behaviour after a fork+exec in the
series "mm/ksm: fix ksm exec support for prctl".
- Baolin Wang has implemented NUMA balancing for multi-size THPs.
This is a simple first-cut implementation for now. The series is
"support multi-size THP numa balancing".
- Cleanups to vma handling helper functions from Matthew Wilcox in
the series "Unify vma_address and vma_pgoff_address".
- Some selftests maintenance work from Dev Jain in the series
"selftests/mm: mremap_test: Optimizations and style fixes".
- Improvements to the swapping of multi-size THPs from Ryan Roberts
in the series "Swap-out mTHP without splitting".
- Kefeng Wang has significantly optimized the handling of arm64's
permission page faults in the series
"arch/mm/fault: accelerate pagefault when badaccess"
"mm: remove arch's private VM_FAULT_BADMAP/BADACCESS"
- GUP cleanups from David Hildenbrand in "mm/gup: consistently call
it GUP-fast".
- hugetlb fault code cleanups from Vishal Moola in "Hugetlb fault
path to use struct vm_fault".
- selftests build fixes from John Hubbard in the series "Fix
selftests/mm build without requiring "make headers"".
- Memory tiering fixes/improvements from Ho-Ren (Jack) Chuang in the
series "Improved Memory Tier Creation for CPUless NUMA Nodes".
Fixes the initialization code so that migration between different
memory types works as intended.
- David Hildenbrand has improved follow_pte() and fixed an errant
driver in the series "mm: follow_pte() improvements and acrn
follow_pte() fixes".
- David also did some cleanup work on large folio mapcounts in his
series "mm: mapcount for large folios + page_mapcount() cleanups".
- Folio conversions in KSM in Alex Shi's series "transfer page to
folio in KSM".
- Barry Song has added some sysfs stats for monitoring multi-size
THP's in the series "mm: add per-order mTHP alloc and swpout
counters".
- Some zswap cleanups from Yosry Ahmed in the series "zswap
same-filled and limit checking cleanups".
- Matthew Wilcox has been looking at buffer_head code and found the
documentation to be lacking. The series is "Improve buffer head
documentation".
- Multi-size THPs get more work, this time from Lance Yang. His
series "mm/madvise: enhance lazyfreeing with mTHP in madvise_free"
optimizes the freeing of these things.
- Kemeng Shi has added more userspace-visible writeback
instrumentation in the series "Improve visibility of writeback".
- Kemeng Shi then sent some maintenance work on top in the series
"Fix and cleanups to page-writeback".
- Matthew Wilcox reduces mmap_lock traffic in the anon vma code in
the series "Improve anon_vma scalability for anon VMAs". Intel's
test bot reported an improbable 3x improvement in one test.
- SeongJae Park adds some DAMON feature work in the series
"mm/damon: add a DAMOS filter type for page granularity access recheck"
"selftests/damon: add DAMOS quota goal test"
- Also some maintenance work in the series
"mm/damon/paddr: simplify page level access re-check for pageout"
"mm/damon: misc fixes and improvements"
- David Hildenbrand has disabled some known-to-fail selftests ni the
series "selftests: mm: cow: flag vmsplice() hugetlb tests as
XFAIL".
- memcg metadata storage optimizations from Shakeel Butt in "memcg:
reduce memory consumption by memcg stats".
- DAX fixes and maintenance work from Vishal Verma in the series
"dax/bus.c: Fixups for dax-bus locking""
* tag 'mm-stable-2024-05-17-19-19' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (426 commits)
memcg, oom: cleanup unused memcg_oom_gfp_mask and memcg_oom_order
selftests/mm: hugetlb_madv_vs_map: avoid test skipping by querying hugepage size at runtime
mm/hugetlb: add missing VM_FAULT_SET_HINDEX in hugetlb_wp
mm/hugetlb: add missing VM_FAULT_SET_HINDEX in hugetlb_fault
selftests: cgroup: add tests to verify the zswap writeback path
mm: memcg: make alloc_mem_cgroup_per_node_info() return bool
mm/damon/core: fix return value from damos_wmark_metric_value
mm: do not update memcg stats for NR_{FILE/SHMEM}_PMDMAPPED
selftests: cgroup: remove redundant enabling of memory controller
Docs/mm/damon/maintainer-profile: allow posting patches based on damon/next tree
Docs/mm/damon/maintainer-profile: change the maintainer's timezone from PST to PT
Docs/mm/damon/design: use a list for supported filters
Docs/admin-guide/mm/damon/usage: fix wrong schemes effective quota update command
Docs/admin-guide/mm/damon/usage: fix wrong example of DAMOS filter matching sysfs file
selftests/damon: classify tests for functionalities and regressions
selftests/damon/_damon_sysfs: use 'is' instead of '==' for 'None'
selftests/damon/_damon_sysfs: find sysfs mount point from /proc/mounts
selftests/damon/_damon_sysfs: check errors from nr_schemes file reads
mm/damon/core: initialize ->esz_bp from damos_quota_init_priv()
selftests/damon: add a test for DAMOS quota goal
...
KVM x86 misc changes for 6.10:
- Advertise the max mappable GPA in the "guest MAXPHYADDR" CPUID field, which
is unused by hardware, so that KVM can communicate its inability to map GPAs
that set bits 51:48 due to lack of 5-level paging. Guest firmware is
expected to use the information to safely remap BARs in the uppermost GPA
space, i.e to avoid placing a BAR at a legal, but unmappable, GPA.
- Use vfree() instead of kvfree() for allocations that always use vcalloc()
or __vcalloc().
- Don't completely ignore same-value writes to immutable feature MSRs, as
doing so results in KVM failing to reject accesses to MSR that aren't
supposed to exist given the vCPU model and/or KVM configuration.
- Don't mark APICv as being inhibited due to ABSENT if APICv is disabled
KVM-wide to avoid confusing debuggers (KVM will never bother clearing the
ABSENT inhibit, even if userspace enables in-kernel local APIC).
KVM cleanups for 6.10:
- Misc cleanups extracted from the "exit on missing userspace mapping" series,
which has been put on hold in anticipation of a "KVM Userfault" approach,
which should provide a superset of functionality.
- Remove kvm_make_all_cpus_request_except(), which got added to hack around an
AVIC bug, and then became dead code when a more robust fix came along.
- Fix a goof in the KVM_CREATE_GUEST_MEMFD documentation.
KVM/arm64 updates for Linux 6.10
- Move a lot of state that was previously stored on a per vcpu
basis into a per-CPU area, because it is only pertinent to the
host while the vcpu is loaded. This results in better state
tracking, and a smaller vcpu structure.
- Add full handling of the ERET/ERETAA/ERETAB instructions in
nested virtualisation. The last two instructions also require
emulating part of the pointer authentication extension.
As a result, the trap handling of pointer authentication has
been greattly simplified.
- Turn the global (and not very scalable) LPI translation cache
into a per-ITS, scalable cache, making non directly injected
LPIs much cheaper to make visible to the vcpu.
- A batch of pKVM patches, mostly fixes and cleanups, as the
upstreaming process seems to be resuming. Fingers crossed!
- Allocate PPIs and SGIs outside of the vcpu structure, allowing
for smaller EL2 mapping and some flexibility in implementing
more or less than 32 private IRQs.
- Purge stale mpidr_data if a vcpu is created after the MPIDR
map has been created.
- Preserve vcpu-specific ID registers across a vcpu reset.
- Various minor cleanups and improvements.
In some cases, like with SEV-SNP, guest memory needs to be updated in a
platform-specific manner before it can be safely freed back to the host.
Wire up arch-defined hooks to the .free_folio kvm_gmem_aops callback to
allow for special handling of this sort when freeing memory in response
to FALLOC_FL_PUNCH_HOLE operations and when releasing the inode, and go
ahead and define an arch-specific hook for x86 since it will be needed
for handling memory used for SEV-SNP guests.
Signed-off-by: Michael Roth <michael.roth@amd.com>
Message-Id: <20231230172351.574091-6-michael.roth@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
During guest run-time, kvm_arch_gmem_prepare() is issued as needed to
prepare newly-allocated gmem pages prior to mapping them into the guest.
In the case of SEV-SNP, this mainly involves setting the pages to
private in the RMP table.
However, for the GPA ranges comprising the initial guest payload, which
are encrypted/measured prior to starting the guest, the gmem pages need
to be accessed prior to setting them to private in the RMP table so they
can be initialized with the userspace-provided data. Additionally, an
SNP firmware call is needed afterward to encrypt them in-place and
measure the contents into the guest's launch digest.
While it is possible to bypass the kvm_arch_gmem_prepare() hooks so that
this handling can be done in an open-coded/vendor-specific manner, this
may expose more gmem-internal state/dependencies to external callers
than necessary. Try to avoid this by implementing an interface that
tries to handle as much of the common functionality inside gmem as
possible, while also making it generic enough to potentially be
usable/extensible for TDX as well.
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Michael Roth <michael.roth@amd.com>
Co-developed-by: Michael Roth <michael.roth@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In preparation for adding a function that walks a set of pages
provided by userspace and populates them in a guest_memfd,
add a version of kvm_gmem_get_pfn() that has a "bool prepare"
argument and passes it down to kvm_gmem_get_folio().
Populating guest memory has to call repeatedly __kvm_gmem_get_pfn()
on the same file, so make the new function take struct file*.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
guest_memfd pages are generally expected to be in some arch-defined
initial state prior to using them for guest memory. For SEV-SNP this
initial state is 'private', or 'guest-owned', and requires additional
operations to move these pages into a 'private' state by updating the
corresponding entries the RMP table.
Allow for an arch-defined hook to handle updates of this sort, and go
ahead and implement one for x86 so KVM implementations like AMD SVM can
register a kvm_x86_ops callback to handle these updates for SEV-SNP
guests.
The preparation callback is always called when allocating/grabbing
folios via gmem, and it is up to the architecture to keep track of
whether or not the pages are already in the expected state (e.g. the RMP
table in the case of SEV-SNP).
In some cases, it is necessary to defer the preparation of the pages to
handle things like in-place encryption of initial guest memory payloads
before marking these pages as 'private'/'guest-owned'. Add an argument
(always true for now) to kvm_gmem_get_folio() that allows for the
preparation callback to be bypassed. To detect possible issues in
the way userspace initializes memory, it is only possible to add an
unprepared page if it is not already included in the filemap.
Link: https://lore.kernel.org/lkml/ZLqVdvsF11Ddo7Dq@google.com/
Co-developed-by: Michael Roth <michael.roth@amd.com>
Signed-off-by: Michael Roth <michael.roth@amd.com>
Message-Id: <20231230172351.574091-5-michael.roth@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Because kvm_gmem_get_pfn() is called from the page fault path without
any of the slots_lock, filemap lock or mmu_lock taken, it is
possible for it to race with kvm_gmem_unbind(). This is not a
problem, as any PTE that is installed temporarily will be zapped
before the guest has the occasion to run.
However, it is not possible to have a complete unbind+bind
racing with the page fault, because deleting the memslot
will call synchronize_srcu_expedited() and wait for the
page fault to be resolved. Thus, we can still warn if
the file is there and is not the one we expect.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Some SNP ioctls will require the page not to be in the pagecache, and as such they
will want to return EEXIST to userspace. Start by passing the error up from
filemap_grab_folio.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
truncate_inode_pages_range() may attempt to zero pages before truncating
them, and this will occur before arch-specific invalidations can be
triggered via .invalidate_folio/.free_folio hooks via kvm_gmem_aops. For
AMD SEV-SNP this would result in an RMP #PF being generated by the
hardware, which is currently treated as fatal (and even if specifically
allowed for, would not result in anything other than garbage being
written to guest pages due to encryption). On Intel TDX this would also
result in undesirable behavior.
Set the AS_INACCESSIBLE flag to prevent the MM from attempting
unexpected accesses of this sort during operations like truncation.
This may also in some cases yield a decent performance improvement for
guest_memfd userspace implementations that hole-punch ranges immediately
after private->shared conversions via KVM_SET_MEMORY_ATTRIBUTES, since
the current implementation of truncate_inode_pages_range() always ends
up zero'ing an entire 4K range if it is backing by a 2M folio.
Link: https://lore.kernel.org/lkml/ZR9LYhpxTaTk6PJX@google.com/
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Michael Roth <michael.roth@amd.com>
Message-ID: <20240329212444.395559-6-michael.roth@amd.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Remove kvm_make_all_cpus_request_except() as it effectively has no users,
and arguably should never have been added in the first place.
Commit 54163a346d ("KVM: Introduce kvm_make_all_cpus_request_except()")
added the "except" variation for use in SVM's AVIC update path, which used
it to skip sending a request to the current vCPU (commit 7d611233b0
("KVM: SVM: Disable AVIC before setting V_IRQ")).
But the AVIC usage of kvm_make_all_cpus_request_except() was essentially a
hack-a-fix that simply squashed the most likely scenario of a racy WARN
without addressing the underlying problem(s). Commit f1577ab214 ("KVM:
SVM: svm_set_vintr don't warn if AVIC is active but is about to be
deactivated") eventually fixed the WARN itself, and the "except" usage was
subsequently dropped by df63202fe5 ("KVM: x86: APICv: drop immediate
APICv disablement on current vCPU").
That kvm_make_all_cpus_request_except() hasn't gained any users in the
last ~3 years isn't a coincidence. If a VM-wide broadcast *needs* to skip
the current vCPU, then odds are very good that there is underlying bug
that could be better fixed elsewhere.
Signed-off-by: Venkatesh Srinivas <venkateshs@chromium.org>
Link: https://lore.kernel.org/r/20240404232651.1645176-1-venkateshs@chromium.org
[sean: rewrite changelog with --verbose]
Signed-off-by: Sean Christopherson <seanjc@google.com>
A subsequent change to KVM/arm64 will necessitate walking the device
list outside of the kvm->lock. Prepare by converting to an rculist. This
has zero effect on the VM destruction path, as it is expected every
reader is backed by a reference on the kvm struct.
On the other hand, ensure a given device is completely destroyed before
dropping the kvm->lock in the release() path, as certain devices expect
to be a singleton (e.g. the vfio-kvm device).
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Sean Christopherson <seanjc@google.com>
Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Reviewed-by: Sean Christopherson <seanjc@google.com>
Link: https://lore.kernel.org/r/20240422200158.2606761-2-oliver.upton@linux.dev
Signed-off-by: Marc Zyngier <maz@kernel.org>
Add support to MMU caches for initializing a page with a custom 64-bit
value, e.g. to pre-fill an entire page table with non-zero PTE values.
The functionality will be used by x86 to support Intel's TDX, which needs
to set bit 63 in all non-present PTEs in order to prevent !PRESENT page
faults from getting reflected into the guest (Intel's EPT Violation #VE
architecture made the less than brilliant decision of having the per-PTE
behavior be opt-out instead of opt-in).
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Message-Id: <5919f685f109a1b0ebc6bd8fc4536ee94bcc172d.1705965635.git.isaku.yamahata@intel.com>
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Reviewed-by: Binbin Wu <binbin.wu@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The .change_pte() MMU notifier callback was intended as an
optimization. The original point of it was that KSM could tell KVM to flip
its secondary PTE to a new location without having to first zap it. At
the time there was also an .invalidate_page() callback; both of them were
*not* bracketed by calls to mmu_notifier_invalidate_range_{start,end}(),
and .invalidate_page() also doubled as a fallback implementation of
.change_pte().
Later on, however, both callbacks were changed to occur within an
invalidate_range_start/end() block.
In the case of .change_pte(), commit 6bdb913f0a ("mm: wrap calls to
set_pte_at_notify with invalidate_range_start and invalidate_range_end",
2012-10-09) did so to remove the fallback from .invalidate_page() to
.change_pte() and allow sleepable .invalidate_page() hooks.
This however made KVM's usage of the .change_pte() callback completely
moot, because KVM unmaps the sPTEs during .invalidate_range_start()
and therefore .change_pte() has no hope of finding a sPTE to change.
Drop the generic KVM code that dispatches to kvm_set_spte_gfn(), as
well as all the architecture specific implementations.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Acked-by: Anup Patel <anup@brainfault.org>
Acked-by: Michael Ellerman <mpe@ellerman.id.au> (powerpc)
Reviewed-by: Bibo Mao <maobibo@loongson.cn>
Message-ID: <20240405115815.3226315-2-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Explicit disallow activating a gfn_to_pfn_cache with an error gpa, i.e.
INVALID_GPA, to ensure that KVM doesn't mistake a GPA-based cache for an
HVA-based cache (KVM uses INVALID_GPA as a magic value to differentiate
between GPA-based and HVA-based caches).
WARN if KVM attempts to activate a cache with INVALID_GPA, purely so that
new caches need to at least consider what to do with a "bad" GPA, as all
existing usage of kvm_gpc_activate() guarantees gpa != INVALID_GPA. I.e.
removing the WARN in the future is completely reasonable if doing so would
yield cleaner/better code overall.
Reviewed-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
Link: https://lore.kernel.org/r/20240320001542.3203871-4-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Add a helper to check that the incoming length for a gfn_to_pfn_cache is
valid with respect to the cache's GPA and/or HVA. To avoid activating a
cache with a bogus GPA, a future fix will fork the page split check in
the inner refresh path into activate() and the public rerfresh() APIs, at
which point KVM will check the length in three separate places.
Deliberately keep the "page offset" logic open coded, as the only other
path that consumes the offset, __kvm_gpc_refresh(), already needs to
differentiate between GPA-based and HVA-based caches, and it's not obvious
that using a helper is a net positive in overall code readability.
Note, for GPA-based caches, this has a subtle side effect of using the GPA
instead of the resolved HVA in the check() path, but that should be a nop
as the HVA offset is derived from the GPA, i.e. the two offsets are
identical, barring a KVM bug.
Reviewed-by: Paul Durrant <paul@xen.org>
Reviewed-by: David Woodhouse <dwmw@amazon.co.uk>
Link: https://lore.kernel.org/r/20240320001542.3203871-2-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
KVM Xen and pfncache changes for 6.9:
- Rip out the half-baked support for using gfn_to_pfn caches to manage pages
that are "mapped" into guests via physical addresses.
- Add support for using gfn_to_pfn caches with only a host virtual address,
i.e. to bypass the "gfn" stage of the cache. The primary use case is
overlay pages, where the guest may change the gfn used to reference the
overlay page, but the backing hva+pfn remains the same.
- Add an ioctl() to allow mapping Xen's shared_info page using an hva instead
of a gpa, so that userspace doesn't need to reconfigure and invalidate the
cache/mapping if the guest changes the gpa (but userspace keeps the resolved
hva the same).
- When possible, use a single host TSC value when computing the deadline for
Xen timers in order to improve the accuracy of the timer emulation.
- Inject pending upcall events when the vCPU software-enables its APIC to fix
a bug where an upcall can be lost (and to follow Xen's behavior).
- Fall back to the slow path instead of warning if "fast" IRQ delivery of Xen
events fails, e.g. if the guest has aliased xAPIC IDs.
- Extend gfn_to_pfn_cache's mutex to cover (de)activation (in addition to
refresh), and drop a now-redundant acquisition of xen_lock (that was
protecting the shared_info cache) to fix a deadlock due to recursively
acquiring xen_lock.
KVM x86 misc changes for 6.9:
- Explicitly initialize a variety of on-stack variables in the emulator that
triggered KMSAN false positives (though in fairness in KMSAN, it's comically
difficult to see that the uninitialized memory is never truly consumed).
- Fix the deubgregs ABI for 32-bit KVM, and clean up code related to reading
DR6 and DR7.
- Rework the "force immediate exit" code so that vendor code ultimately
decides how and when to force the exit. This allows VMX to further optimize
handling preemption timer exits, and allows SVM to avoid sending a duplicate
IPI (SVM also has a need to force an exit).
- Fix a long-standing bug where kvm_has_noapic_vcpu could be left elevated if
vCPU creation ultimately failed, and add WARN to guard against similar bugs.
- Provide a dedicated arch hook for checking if a different vCPU was in-kernel
(for directed yield), and simplify the logic for checking if the currently
loaded vCPU is in-kernel.
- Misc cleanups and fixes.
KVM common MMU changes for 6.9:
- Harden KVM against underflowing the active mmu_notifier invalidation
count, so that "bad" invalidations (usually due to bugs elsehwere in the
kernel) are detected earlier and are less likely to hang the kernel.
- Fix a benign bug in __kvm_mmu_topup_memory_cache() where the object size
and number of objects parameters to kvmalloc_array() were swapped.
KVM async page fault changes for 6.9:
- Always flush the async page fault workqueue when a work item is being
removed, especially during vCPU destruction, to ensure that there are no
workers running in KVM code when all references to KVM-the-module are gone,
i.e. to prevent a use-after-free if kvm.ko is unloaded.
- Grab a reference to the VM's mm_struct in the async #PF worker itself instead
of gifting the worker a reference, e.g. so that there's no need to remember
to *conditionally* clean up after the worker.
KVM/arm64 updates for 6.9
- Infrastructure for building KVM's trap configuration based on the
architectural features (or lack thereof) advertised in the VM's ID
registers
- Support for mapping vfio-pci BARs as Normal-NC (vaguely similar to
x86's WC) at stage-2, improving the performance of interacting with
assigned devices that can tolerate it
- Conversion of KVM's representation of LPIs to an xarray, utilized to
address serialization some of the serialization on the LPI injection
path
- Support for _architectural_ VHE-only systems, advertised through the
absence of FEAT_E2H0 in the CPU's ID register
- Miscellaneous cleanups, fixes, and spelling corrections to KVM and
selftests
KVM GUEST_MEMFD fixes for 6.8:
- Make KVM_MEM_GUEST_MEMFD mutually exclusive with KVM_MEM_READONLY to
avoid creating ABI that KVM can't sanely support.
- Update documentation for KVM_SW_PROTECTED_VM to make it abundantly
clear that such VMs are purely a development and testing vehicle, and
come with zero guarantees.
- Limit KVM_SW_PROTECTED_VM guests to the TDP MMU, as the long term plan
is to support confidential VMs with deterministic private memory (SNP
and TDX) only in the TDP MMU.
- Fix a bug in a GUEST_MEMFD negative test that resulted in false passes
when verifying that KVM_MEM_GUEST_MEMFD memslots can't be dirty logged.
The locking on the gfn_to_pfn_cache is... interesting. And awful.
There is a rwlock in ->lock which readers take to ensure protection
against concurrent changes. But __kvm_gpc_refresh() makes assumptions
that certain fields will not change even while it drops the write lock
and performs MM operations to revalidate the target PFN and kernel
mapping.
Commit 93984f19e7 ("KVM: Fully serialize gfn=>pfn cache refresh via
mutex") partly addressed that — not by fixing it, but by adding a new
mutex, ->refresh_lock. This prevented concurrent __kvm_gpc_refresh()
calls on a given gfn_to_pfn_cache, but is still only a partial solution.
There is still a theoretical race where __kvm_gpc_refresh() runs in
parallel with kvm_gpc_deactivate(). While __kvm_gpc_refresh() has
dropped the write lock, kvm_gpc_deactivate() clears the ->active flag
and unmaps ->khva. Then __kvm_gpc_refresh() determines that the previous
->pfn and ->khva are still valid, and reinstalls those values into the
structure. This leaves the gfn_to_pfn_cache with the ->valid bit set,
but ->active clear. And a ->khva which looks like a reasonable kernel
address but is actually unmapped.
All it takes is a subsequent reactivation to cause that ->khva to be
dereferenced. This would theoretically cause an oops which would look
something like this:
[1724749.564994] BUG: unable to handle page fault for address: ffffaa3540ace0e0
[1724749.565039] RIP: 0010:__kvm_xen_has_interrupt+0x8b/0xb0
I say "theoretically" because theoretically, that oops that was seen in
production cannot happen. The code which uses the gfn_to_pfn_cache is
supposed to have its *own* locking, to further paper over the fact that
the gfn_to_pfn_cache's own papering-over (->refresh_lock) of its own
rwlock abuse is not sufficient.
For the Xen vcpu_info that external lock is the vcpu->mutex, and for the
shared info it's kvm->arch.xen.xen_lock. Those locks ought to protect
the gfn_to_pfn_cache against concurrent deactivation vs. refresh in all
but the cases where the vcpu or kvm object is being *destroyed*, in
which case the subsequent reactivation should never happen.
Theoretically.
Nevertheless, this locking abuse is awful and should be fixed, even if
no clear explanation can be found for how the oops happened. So expand
the use of the ->refresh_lock mutex to ensure serialization of
activate/deactivate vs. refresh and make the pfncache locking entirely
self-sufficient.
This means that a future commit can simplify the locking in the callers,
such as the Xen emulation code which has an outstanding problem with
recursive locking of kvm->arch.xen.xen_lock, which will no longer be
necessary.
The rwlock abuse described above is still not best practice, although
it's harmless now that the ->refresh_lock is held for the entire duration
while the offending code drops the write lock, does some other stuff,
then takes the write lock again and assumes nothing changed. That can
also be fixed^W cleaned up in a subsequent commit, but this commit is
a simpler basis for the Xen deadlock fix mentioned above.
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
Link: https://lore.kernel.org/r/20240227115648.3104-5-dwmw2@infradead.org
[sean: use guard(mutex) to fix a missed unlock]
Signed-off-by: Sean Christopherson <seanjc@google.com>
The general expectation with debugfs is that any initialization failure
is nonfatal. Nevertheless, kvm_arch_create_vm_debugfs() allows
implementations to return an error and kvm_create_vm_debugfs() allows
that to fail VM creation.
Change to a void return to discourage architectures from making debugfs
failures fatal for the VM. Seems like everyone already had the right
idea, as all implementations already return 0 unconditionally.
Acked-by: Marc Zyngier <maz@kernel.org>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Link: https://lore.kernel.org/r/20240216155941.2029458-1-oliver.upton@linux.dev
Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Disallow creating read-only memslots that support GUEST_MEMFD, as
GUEST_MEMFD is fundamentally incompatible with KVM's semantics for
read-only memslots. Read-only memslots allow the userspace VMM to emulate
option ROMs by filling the backing memory with readable, executable code
and data, while triggering emulated MMIO on writes. GUEST_MEMFD doesn't
currently support writes from userspace and KVM doesn't support emulated
MMIO on private accesses, i.e. the guest can only ever read zeros, and
writes will always be treated as errors.
Cc: Fuad Tabba <tabba@google.com>
Cc: Michael Roth <michael.roth@amd.com>
Cc: Isaku Yamahata <isaku.yamahata@gmail.com>
Cc: Yu Zhang <yu.c.zhang@linux.intel.com>
Cc: Chao Peng <chao.p.peng@linux.intel.com>
Fixes: a7800aa80e ("KVM: Add KVM_CREATE_GUEST_MEMFD ioctl() for guest-specific backing memory")
Link: https://lore.kernel.org/r/20240222190612.2942589-2-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
gcc-14 notices that the arguments to kvmalloc_array() are mixed up:
arch/x86/kvm/../../../virt/kvm/kvm_main.c: In function '__kvm_mmu_topup_memory_cache':
arch/x86/kvm/../../../virt/kvm/kvm_main.c:424:53: error: 'kvmalloc_array' sizes specified with 'sizeof' in the earlier argument and not in the later argument [-Werror=calloc-transposed-args]
424 | mc->objects = kvmalloc_array(sizeof(void *), capacity, gfp);
| ^~~~
arch/x86/kvm/../../../virt/kvm/kvm_main.c:424:53: note: earlier argument should specify number of elements, later size of each element
The code still works correctly, but the incorrect order prevents the compiler
from properly tracking the object sizes.
Fixes: 837f66c712 ("KVM: Allow for different capacities in kvm_mmu_memory_cache structs")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20240212112419.1186065-1-arnd@kernel.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Plumb in a dedicated hook for querying whether or not a vCPU was preempted
in-kernel. Unlike literally every other architecture, x86's VMX can check
if a vCPU is in kernel context if and only if the vCPU is loaded on the
current pCPU.
x86's kvm_arch_vcpu_in_kernel() works around the limitation by querying
kvm_get_running_vcpu() and redirecting to vcpu->arch.preempted_in_kernel
as needed. But that's unnecessary, confusing, and fragile, e.g. x86 has
had at least one bug where KVM incorrectly used a stale
preempted_in_kernel.
No functional change intended.
Reviewed-by: Yuan Yao <yuan.yao@intel.com>
Link: https://lore.kernel.org/r/20240110003938.490206-2-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
When processing mmu_notifier invalidations for gpc caches, pre-check for
overlap with the invalidation event while holding gpc->lock for read, and
only take gpc->lock for write if the cache needs to be invalidated. Doing
a pre-check without taking gpc->lock for write avoids unnecessarily
contending the lock for unrelated invalidations, which is very beneficial
for caches that are heavily used (but rarely subjected to mmu_notifier
invalidations).
Signed-off-by: Paul Durrant <pdurrant@amazon.com>
Reviewed-by: David Woodhouse <dwmw@amazon.co.uk>
Link: https://lore.kernel.org/r/20240215152916.1158-20-paul@xen.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Some pfncache pages may actually be overlays on guest memory that have a
fixed HVA within the VMM. It's pointless to invalidate such cached
mappings if the overlay is moved so allow a cache to be activated directly
with the HVA to cater for such cases. A subsequent patch will make use
of this facility.
Signed-off-by: Paul Durrant <pdurrant@amazon.com>
Reviewed-by: David Woodhouse <dwmw@amazon.co.uk>
Link: https://lore.kernel.org/r/20240215152916.1158-10-paul@xen.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Currently the pfncache page offset is sometimes determined using the gpa
and sometimes the khva, whilst the uhva is always page-aligned. After a
subsequent patch is applied the gpa will not always be valid so adjust
the code to include the page offset in the uhva and use it consistently
as the source of truth.
Also, where a page-aligned address is required, use PAGE_ALIGN_DOWN()
for clarity.
No functional change intended.
Signed-off-by: Paul Durrant <pdurrant@amazon.com>
Reviewed-by: David Woodhouse <dwmw@amazon.co.uk>
Link: https://lore.kernel.org/r/20240215152916.1158-8-paul@xen.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
