Currently, every time a VCPU is scheduled out, the host kernel will
first save the guest FPU/xstate context, then load the qemu userspace
FPU context, only to then immediately save the qemu userspace FPU
context back to memory. When scheduling in a VCPU, the same extraneous
FPU loads and saves are done.
This could be avoided by moving from a model where the guest FPU is
loaded and stored with preemption disabled, to a model where the
qemu userspace FPU is swapped out for the guest FPU context for
the duration of the KVM_RUN ioctl.
This is done under the VCPU mutex, which is also taken when other
tasks inspect the VCPU FPU context, so the code should already be
safe for this change. That should come as no surprise, given that
s390 already has this optimization.
This can fix a bug where KVM calls get_user_pages while owning the
FPU, and the file system ends up requesting the FPU again:
[258270.527947] __warn+0xcb/0xf0
[258270.527948] warn_slowpath_null+0x1d/0x20
[258270.527951] kernel_fpu_disable+0x3f/0x50
[258270.527953] __kernel_fpu_begin+0x49/0x100
[258270.527955] kernel_fpu_begin+0xe/0x10
[258270.527958] crc32c_pcl_intel_update+0x84/0xb0
[258270.527961] crypto_shash_update+0x3f/0x110
[258270.527968] crc32c+0x63/0x8a [libcrc32c]
[258270.527975] dm_bm_checksum+0x1b/0x20 [dm_persistent_data]
[258270.527978] node_prepare_for_write+0x44/0x70 [dm_persistent_data]
[258270.527985] dm_block_manager_write_callback+0x41/0x50 [dm_persistent_data]
[258270.527988] submit_io+0x170/0x1b0 [dm_bufio]
[258270.527992] __write_dirty_buffer+0x89/0x90 [dm_bufio]
[258270.527994] __make_buffer_clean+0x4f/0x80 [dm_bufio]
[258270.527996] __try_evict_buffer+0x42/0x60 [dm_bufio]
[258270.527998] dm_bufio_shrink_scan+0xc0/0x130 [dm_bufio]
[258270.528002] shrink_slab.part.40+0x1f5/0x420
[258270.528004] shrink_node+0x22c/0x320
[258270.528006] do_try_to_free_pages+0xf5/0x330
[258270.528008] try_to_free_pages+0xe9/0x190
[258270.528009] __alloc_pages_slowpath+0x40f/0xba0
[258270.528011] __alloc_pages_nodemask+0x209/0x260
[258270.528014] alloc_pages_vma+0x1f1/0x250
[258270.528017] do_huge_pmd_anonymous_page+0x123/0x660
[258270.528021] handle_mm_fault+0xfd3/0x1330
[258270.528025] __get_user_pages+0x113/0x640
[258270.528027] get_user_pages+0x4f/0x60
[258270.528063] __gfn_to_pfn_memslot+0x120/0x3f0 [kvm]
[258270.528108] try_async_pf+0x66/0x230 [kvm]
[258270.528135] tdp_page_fault+0x130/0x280 [kvm]
[258270.528149] kvm_mmu_page_fault+0x60/0x120 [kvm]
[258270.528158] handle_ept_violation+0x91/0x170 [kvm_intel]
[258270.528162] vmx_handle_exit+0x1ca/0x1400 [kvm_intel]
No performance changes were detected in quick ping-pong tests on
my 4 socket system, which is expected since an FPU+xstate load is
on the order of 0.1us, while ping-ponging between CPUs is on the
order of 20us, and somewhat noisy.
Cc: stable@vger.kernel.org
Signed-off-by: Rik van Riel <riel@redhat.com>
Suggested-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
[Fixed a bug where reset_vcpu called put_fpu without preceding load_fpu,
which happened inside from KVM_CREATE_VCPU ioctl. - Radim]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
KVM API says for the signal mask you set via KVM_SET_SIGNAL_MASK, that
"any unblocked signal received [...] will cause KVM_RUN to return with
-EINTR" and that "the signal will only be delivered if not blocked by
the original signal mask".
This, however, is only true, when the calling task has a signal handler
registered for a signal. If not, signal evaluation is short-circuited for
SIG_IGN and SIG_DFL, and the signal is either ignored without KVM_RUN
returning or the whole process is terminated.
Make KVM_SET_SIGNAL_MASK behave as advertised by utilizing logic similar
to that in do_sigtimedwait() to avoid short-circuiting of signals.
Signed-off-by: Jan H. Schönherr <jschoenh@amazon.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We will not see -ENOMEM (gfn_to_hva() will return KVM_ERR_PTR_BAD_PAGE
for all errors). So we can also get rid of special handling in the
callers of pin_guest_page() and always assume that it is a g2 error.
As also kvm_s390_inject_program_int() should never fail, we can
simplify pin_scb(), too.
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20170901151143.22714-1-david@redhat.com>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
If a vcpu exits due to request a user mode spinlock, then
the spinlock-holder may be preempted in user mode or kernel mode.
(Note that not all architectures trap spin loops in user mode,
only AMD x86 and ARM/ARM64 currently do).
But if a vcpu exits in kernel mode, then the holder must be
preempted in kernel mode, so we should choose a vcpu in kernel mode
as a more likely candidate for the lock holder.
This introduces kvm_arch_vcpu_in_kernel() to decide whether the
vcpu is in kernel-mode when it's preempted. kvm_vcpu_on_spin's
new argument says the same of the spinning VCPU.
Signed-off-by: Longpeng(Mike) <longpeng2@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
nested_get_page() just sounds confusing. All we want is a page from G1.
This is even unrelated to nested.
Let's introduce kvm_vcpu_gpa_to_page() so we don't get too lengthy
lines.
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
[Squash pasto fix from Wanpeng Li. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
During teardown, accesses to memslots and buses are using
rcu_dereference_protected with an always-true condition because
these accesses are done outside the usual mutexes. This
is because the last reference is gone and there cannot be any
concurrent modifications, but rcu_dereference_protected is
ugly and unobvious.
Instead, check the refcount in kvm_get_bus and __kvm_memslots.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Simplify and improve the code so that the PID is always available in
the uevent even when debugfs is not available.
This adds a userspace_pid field to struct kvm, as per Radim's
suggestion, so that the PID can be retrieved on destruction too.
Acked-by: Janosch Frank <frankja@linux.vnet.ibm.com>
Fixes: 286de8f6ac ("KVM: trigger uevents when creating or destroying a VM")
Signed-off-by: Claudio Imbrenda <imbrenda@linux.vnet.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
kvm memslots are protected by srcu and not by rcu. We must use
srcu_dereference_check instead of rcu_dereference_check.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
we access the memslots array via srcu. Mark it as such and
use the right access functions also for the freeing of
memory slots.
Found by sparse:
./include/linux/kvm_host.h:565:16: error: incompatible types in
comparison expression (different address spaces)
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
mark kvm->busses as rcu protected and use the correct access
function everywhere.
found by sparse
virt/kvm/kvm_main.c:3490:15: error: incompatible types in comparison expression (different address spaces)
virt/kvm/kvm_main.c:3509:15: error: incompatible types in comparison expression (different address spaces)
virt/kvm/kvm_main.c:3561:15: error: incompatible types in comparison expression (different address spaces)
virt/kvm/kvm_main.c:3644:15: error: incompatible types in comparison expression (different address spaces)
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
We do use rcu to protect the pid pointer. Mark it as such and
adopt all code to use the proper access methods.
This was detected by sparse.
"virt/kvm/kvm_main.c:2248:15: error: incompatible types in comparison
expression (different address spaces)"
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
A first step in vcpu->requests encapsulation. Additionally, we now
use READ_ONCE() when accessing vcpu->requests, which ensures we
always load vcpu->requests when it's accessed. This is important as
other threads can change it any time. Also, READ_ONCE() documents
that vcpu->requests is used with other threads, likely requiring
memory barriers, which it does.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
[ Documented the new use of READ_ONCE() and converted another check
in arch/mips/kvm/vz.c ]
Signed-off-by: Andrew Jones <drjones@redhat.com>
Acked-by: Christoffer Dall <cdall@linaro.org>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
Marc Zyngier suggested that we define the arch specific VCPU request
base, rather than requiring each arch to remember to start from 8.
That suggestion, along with Radim Krcmar's recent VCPU request flag
addition, snowballed into defining something of an arch VCPU request
defining API.
No functional change.
(Looks like x86 is running out of arch VCPU request bits. Maybe
someday we'll need to extend to 64.)
Signed-off-by: Andrew Jones <drjones@redhat.com>
Acked-by: Christoffer Dall <cdall@linaro.org>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
Pull more KVM updates from Paolo Bonzini:
"ARM:
- bugfixes
- moved shared 32-bit/64-bit files to virt/kvm/arm
- support for saving/restoring virtual ITS state to userspace
PPC:
- XIVE (eXternal Interrupt Virtualization Engine) support
x86:
- nVMX improvements, including emulated page modification logging
(PML) which brings nice performance improvements on some workloads"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (45 commits)
KVM: arm/arm64: vgic-its: Cleanup after failed ITT restore
KVM: arm/arm64: Don't call map_resources when restoring ITS tables
KVM: arm/arm64: Register ITS iodev when setting base address
KVM: arm/arm64: Get rid of its->initialized field
KVM: arm/arm64: Register iodevs when setting redist base and creating VCPUs
KVM: arm/arm64: Slightly rework kvm_vgic_addr
KVM: arm/arm64: Make vgic_v3_check_base more broadly usable
KVM: arm/arm64: Refactor vgic_register_redist_iodevs
KVM: Add kvm_vcpu_get_idx to get vcpu index in kvm->vcpus
nVMX: Advertise PML to L1 hypervisor
nVMX: Implement emulated Page Modification Logging
kvm: x86: Add a hook for arch specific dirty logging emulation
kvm: nVMX: Validate CR3 target count on nested VM-entry
KVM: set no_llseek in stat_fops_per_vm
KVM: arm/arm64: vgic: Rename kvm_vgic_vcpu_init to kvm_vgic_vcpu_enable
KVM: arm/arm64: Clarification and relaxation to ITS save/restore ABI
KVM: arm64: vgic-v3: KVM_DEV_ARM_VGIC_SAVE_PENDING_TABLES
KVM: arm64: vgic-its: Fix pending table sync
KVM: arm64: vgic-its: ITT save and restore
KVM: arm64: vgic-its: Device table save/restore
...
Pull RCU updates from Ingo Molnar:
"The main changes are:
- Debloat RCU headers
- Parallelize SRCU callback handling (plus overlapping patches)
- Improve the performance of Tree SRCU on a CPU-hotplug stress test
- Documentation updates
- Miscellaneous fixes"
* 'core-rcu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (74 commits)
rcu: Open-code the rcu_cblist_n_lazy_cbs() function
rcu: Open-code the rcu_cblist_n_cbs() function
rcu: Open-code the rcu_cblist_empty() function
rcu: Separately compile large rcu_segcblist functions
srcu: Debloat the <linux/rcu_segcblist.h> header
srcu: Adjust default auto-expediting holdoff
srcu: Specify auto-expedite holdoff time
srcu: Expedite first synchronize_srcu() when idle
srcu: Expedited grace periods with reduced memory contention
srcu: Make rcutorture writer stalls print SRCU GP state
srcu: Exact tracking of srcu_data structures containing callbacks
srcu: Make SRCU be built by default
srcu: Fix Kconfig botch when SRCU not selected
rcu: Make non-preemptive schedule be Tasks RCU quiescent state
srcu: Expedite srcu_schedule_cbs_snp() callback invocation
srcu: Parallelize callback handling
kvm: Move srcu_struct fields to end of struct kvm
rcu: Fix typo in PER_RCU_NODE_PERIOD header comment
rcu: Use true/false in assignment to bool
rcu: Use bool value directly
...
Second round of KVM/ARM Changes for v4.12.
Changes include:
- A fix related to the 32-bit idmap stub
- A fix to the bitmask used to deode the operands of an AArch32 CP
instruction
- We have moved the files shared between arch/arm/kvm and
arch/arm64/kvm to virt/kvm/arm
- We add support for saving/restoring the virtual ITS state to
userspace
There are occasional needs to use the index of vcpu in the kvm->vcpus
array to map something related to a VCPU. For example, unlike the
vcpu->vcpu_id, the vcpu index is guaranteed to not be sparse across all
vcpus which is useful when allocating a memory area for each vcpu.
Signed-off-by: Christoffer Dall <cdall@linaro.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
The main thing here is a new implementation of the in-kernel
XICS interrupt controller emulation for POWER9 machines, from Ben
Herrenschmidt.
POWER9 has a new interrupt controller called XIVE (eXternal Interrupt
Virtualization Engine) which is able to deliver interrupts directly
to guest virtual CPUs in hardware without hypervisor intervention.
With this new code, the guest still sees the old XICS interface but
performance is better because the XICS emulation in the host uses the
XIVE directly rather than going through a XICS emulation in firmware.
Conflicts:
arch/powerpc/kernel/cpu_setup_power.S [cherry-picked fix]
arch/powerpc/kvm/book3s_xive.c [include asm/debugfs.h]
Merge more updates from Andrew Morton:
- the rest of MM
- various misc things
- procfs updates
- lib/ updates
- checkpatch updates
- kdump/kexec updates
- add kvmalloc helpers, use them
- time helper updates for Y2038 issues. We're almost ready to remove
current_fs_time() but that awaits a btrfs merge.
- add tracepoints to DAX
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (114 commits)
drivers/staging/ccree/ssi_hash.c: fix build with gcc-4.4.4
selftests/vm: add a test for virtual address range mapping
dax: add tracepoint to dax_insert_mapping()
dax: add tracepoint to dax_writeback_one()
dax: add tracepoints to dax_writeback_mapping_range()
dax: add tracepoints to dax_load_hole()
dax: add tracepoints to dax_pfn_mkwrite()
dax: add tracepoints to dax_iomap_pte_fault()
mtd: nand: nandsim: convert to memalloc_noreclaim_*()
treewide: convert PF_MEMALLOC manipulations to new helpers
mm: introduce memalloc_noreclaim_{save,restore}
mm: prevent potential recursive reclaim due to clearing PF_MEMALLOC
mm/huge_memory.c: deposit a pgtable for DAX PMD faults when required
mm/huge_memory.c: use zap_deposited_table() more
time: delete CURRENT_TIME_SEC and CURRENT_TIME
gfs2: replace CURRENT_TIME with current_time
apparmorfs: replace CURRENT_TIME with current_time()
lustre: replace CURRENT_TIME macro
fs: ubifs: replace CURRENT_TIME_SEC with current_time
fs: ufs: use ktime_get_real_ts64() for birthtime
...
Patch series "kvmalloc", v5.
There are many open coded kmalloc with vmalloc fallback instances in the
tree. Most of them are not careful enough or simply do not care about
the underlying semantic of the kmalloc/page allocator which means that
a) some vmalloc fallbacks are basically unreachable because the kmalloc
part will keep retrying until it succeeds b) the page allocator can
invoke a really disruptive steps like the OOM killer to move forward
which doesn't sound appropriate when we consider that the vmalloc
fallback is available.
As it can be seen implementing kvmalloc requires quite an intimate
knowledge if the page allocator and the memory reclaim internals which
strongly suggests that a helper should be implemented in the memory
subsystem proper.
Most callers, I could find, have been converted to use the helper
instead. This is patch 6. There are some more relying on __GFP_REPEAT
in the networking stack which I have converted as well and Eric Dumazet
was not opposed [2] to convert them as well.
[1] http://lkml.kernel.org/r/20170130094940.13546-1-mhocko@kernel.org
[2] http://lkml.kernel.org/r/1485273626.16328.301.camel@edumazet-glaptop3.roam.corp.google.com
This patch (of 9):
Using kmalloc with the vmalloc fallback for larger allocations is a
common pattern in the kernel code. Yet we do not have any common helper
for that and so users have invented their own helpers. Some of them are
really creative when doing so. Let's just add kv[mz]alloc and make sure
it is implemented properly. This implementation makes sure to not make
a large memory pressure for > PAGE_SZE requests (__GFP_NORETRY) and also
to not warn about allocation failures. This also rules out the OOM
killer as the vmalloc is a more approapriate fallback than a disruptive
user visible action.
This patch also changes some existing users and removes helpers which
are specific for them. In some cases this is not possible (e.g.
ext4_kvmalloc, libcfs_kvzalloc) because those seems to be broken and
require GFP_NO{FS,IO} context which is not vmalloc compatible in general
(note that the page table allocation is GFP_KERNEL). Those need to be
fixed separately.
While we are at it, document that __vmalloc{_node} about unsupported gfp
mask because there seems to be a lot of confusion out there.
kvmalloc_node will warn about GFP_KERNEL incompatible (which are not
superset) flags to catch new abusers. Existing ones would have to die
slowly.
[sfr@canb.auug.org.au: f2fs fixup]
Link: http://lkml.kernel.org/r/20170320163735.332e64b7@canb.auug.org.au
Link: http://lkml.kernel.org/r/20170306103032.2540-2-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Reviewed-by: Andreas Dilger <adilger@dilger.ca> [ext4 part]
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This reverts commit bbd6411513.
I've been sitting on this revert for too long and it unfortunately
missed 4.11. It's also the reason why I haven't merged ring-based
dirty tracking for 4.12.
Using kvm_vcpu_memslots in kvm_gfn_to_hva_cache_init and
kvm_vcpu_write_guest_offset_cached means that the MSR value can
now be used to access SMRAM, simply by making it point to an SMRAM
physical address. This is problematic because it lets the guest
OS overwrite memory that it shouldn't be able to touch.
Cc: stable@vger.kernel.org
Fixes: bbd6411513
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We needed the lock to avoid racing with creation of the irqchip on x86. As
kvm_set_irq_routing() calls srcu_synchronize_expedited(), this lock
might be held for a longer time.
Let's introduce an arch specific callback to check if we can actually
add irq routes. For x86, all we have to do is check if we have an
irqchip in the kernel. We don't need kvm->lock at that point as the
irqchip is marked as inititalized only when actually fully created.
Reported-by: Steve Rutherford <srutherford@google.com>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Fixes: 1df6ddede1 ("KVM: x86: race between KVM_SET_GSI_ROUTING and KVM_CREATE_IRQCHIP")
Signed-off-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This merges in the powerpc topic/xive branch to bring in the code for
the in-kernel XICS interrupt controller emulation to use the new XIVE
(eXternal Interrupt Virtualization Engine) hardware in the POWER9 chip
directly, rather than via a XICS emulation in firmware.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
kvm_make_all_requests() provides a synchronization that waits until all
kicked VCPUs have acknowledged the kick. This is important for
KVM_REQ_MMU_RELOAD as it prevents freeing while lockless paging is
underway.
This patch adds the synchronization property into all requests that are
currently being used with kvm_make_all_requests() in order to preserve
the current behavior and only introduce a new framework. Removing it
from requests where it is not necessary is left for future patches.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
