| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
When halting a guest, QEMU flushes the virtual ITS caches, which
amounts to writing to the various tables that the guest has allocated.
When doing this, we fail to take the srcu lock, and the kernel
shouts loudly if running a lockdep kernel:
[ 69.680416] =============================
[ 69.680819] WARNING: suspicious RCU usage
[ 69.681526] 5.1.0-rc1-00008-g600025238f51-dirty #18 Not tainted
[ 69.682096] -----------------------------
[ 69.682501] ./include/linux/kvm_host.h:605 suspicious rcu_dereference_check() usage!
[ 69.683225]
[ 69.683225] other info that might help us debug this:
[ 69.683225]
[ 69.683975]
[ 69.683975] rcu_scheduler_active = 2, debug_locks = 1
[ 69.684598] 6 locks held by qemu-system-aar/4097:
[ 69.685059] #0: 0000000034196013 (&kvm->lock){+.+.}, at: vgic_its_set_attr+0x244/0x3a0
[ 69.686087] #1: 00000000f2ed935e (&its->its_lock){+.+.}, at: vgic_its_set_attr+0x250/0x3a0
[ 69.686919] #2: 000000005e71ea54 (&vcpu->mutex){+.+.}, at: lock_all_vcpus+0x64/0xd0
[ 69.687698] #3: 00000000c17e548d (&vcpu->mutex){+.+.}, at: lock_all_vcpus+0x64/0xd0
[ 69.688475] #4: 00000000ba386017 (&vcpu->mutex){+.+.}, at: lock_all_vcpus+0x64/0xd0
[ 69.689978] #5: 00000000c2c3c335 (&vcpu->mutex){+.+.}, at: lock_all_vcpus+0x64/0xd0
[ 69.690729]
[ 69.690729] stack backtrace:
[ 69.691151] CPU: 2 PID: 4097 Comm: qemu-system-aar Not tainted 5.1.0-rc1-00008-g600025238f51-dirty #18
[ 69.691984] Hardware name: rockchip evb_rk3399/evb_rk3399, BIOS 2019.04-rc3-00124-g2feec69fb1 03/15/2019
[ 69.692831] Call trace:
[ 69.694072] lockdep_rcu_suspicious+0xcc/0x110
[ 69.694490] gfn_to_memslot+0x174/0x190
[ 69.694853] kvm_write_guest+0x50/0xb0
[ 69.695209] vgic_its_save_tables_v0+0x248/0x330
[ 69.695639] vgic_its_set_attr+0x298/0x3a0
[ 69.696024] kvm_device_ioctl_attr+0x9c/0xd8
[ 69.696424] kvm_device_ioctl+0x8c/0xf8
[ 69.696788] do_vfs_ioctl+0xc8/0x960
[ 69.697128] ksys_ioctl+0x8c/0xa0
[ 69.697445] __arm64_sys_ioctl+0x28/0x38
[ 69.697817] el0_svc_common+0xd8/0x138
[ 69.698173] el0_svc_handler+0x38/0x78
[ 69.698528] el0_svc+0x8/0xc
The fix is to obviously take the srcu lock, just like we do on the
read side of things since bf308242ab98. One wonders why this wasn't
fixed at the same time, but hey...
Fixes: bf308242ab98 ("KVM: arm/arm64: VGIC/ITS: protect kvm_read_guest() calls with SRCU lock")
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
Pull KVM updates from Paolo Bonzini:
"ARM:
- some cleanups
- direct physical timer assignment
- cache sanitization for 32-bit guests
s390:
- interrupt cleanup
- introduction of the Guest Information Block
- preparation for processor subfunctions in cpu models
PPC:
- bug fixes and improvements, especially related to machine checks
and protection keys
x86:
- many, many cleanups, including removing a bunch of MMU code for
unnecessary optimizations
- AVIC fixes
Generic:
- memcg accounting"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (147 commits)
kvm: vmx: fix formatting of a comment
KVM: doc: Document the life cycle of a VM and its resources
MAINTAINERS: Add KVM selftests to existing KVM entry
Revert "KVM/MMU: Flush tlb directly in the kvm_zap_gfn_range()"
KVM: PPC: Book3S: Add count cache flush parameters to kvmppc_get_cpu_char()
KVM: PPC: Fix compilation when KVM is not enabled
KVM: Minor cleanups for kvm_main.c
KVM: s390: add debug logging for cpu model subfunctions
KVM: s390: implement subfunction processor calls
arm64: KVM: Fix architecturally invalid reset value for FPEXC32_EL2
KVM: arm/arm64: Remove unused timer variable
KVM: PPC: Book3S: Improve KVM reference counting
KVM: PPC: Book3S HV: Fix build failure without IOMMU support
Revert "KVM: Eliminate extra function calls in kvm_get_dirty_log_protect()"
x86: kvmguest: use TSC clocksource if invariant TSC is exposed
KVM: Never start grow vCPU halt_poll_ns from value below halt_poll_ns_grow_start
KVM: Expose the initial start value in grow_halt_poll_ns() as a module parameter
KVM: grow_halt_poll_ns() should never shrink vCPU halt_poll_ns
KVM: x86/mmu: Consolidate kvm_mmu_zap_all() and kvm_mmu_zap_mmio_sptes()
KVM: x86/mmu: WARN if zapping a MMIO spte results in zapping children
...
|
|
Until now, we haven't differentiated between HYP calls that
have a return value and those who don't. As we're about to
change this, introduce kvm_call_hyp_ret(), and change all
call sites that actually make use of a return value.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
|
|
vgic_irq->irq_lock must always be taken with interrupts disabled as
it is used in interrupt context.
For configurations such as PREEMPT_RT_FULL, this means that it should
be a raw_spinlock since RT spinlocks are interruptible.
Signed-off-by: Julien Thierry <julien.thierry@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
|
|
kvm_vgic_sync_hwstate is only called with IRQ being disabled.
There is thus no need to call spin_lock_irqsave/restore in
vgic_fold_lr_state and vgic_prune_ap_list.
This patch replace them with the non irq-safe version.
Signed-off-by: Jia He <jia.he@hxt-semitech.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
[maz: commit message tidy-up]
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
Now when we have a group configuration on the struct IRQ, use this state
when populating the LR and signaling interrupts as either group 0 or
group 1 to the VM. Depending on the model of the emulated GIC, and the
guest's configuration of the VMCR, interrupts may be signaled as IRQs or
FIQs to the VM.
Reviewed-by: Andrew Jones <drjones@redhat.com>
Signed-off-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
When booting a 64 KB pages kernel on a ACPI GICv3 system that
implements support for v2 emulation, the following warning is
produced
GICV size 0x2000 not a multiple of page size 0x10000
and support for v2 emulation is disabled, preventing GICv2 VMs
from being able to run on such hosts.
The reason is that vgic_v3_probe() performs a sanity check on the
size of the window (it should be a multiple of the page size),
while the ACPI MADT parsing code hardcodes the size of the window
to 8 KB. This makes sense, considering that ACPI does not bother
to describe the size in the first place, under the assumption that
platforms implementing ACPI will follow the architecture and not
put anything else in the same 64 KB window.
So let's just drop the sanity check altogether, and assume that
the window is at least 64 KB in size.
Fixes: 909777324588 ("KVM: arm/arm64: vgic-new: vgic_init: implement kvm_vgic_hyp_init")
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm into HEAD
KVM/ARM updates for 4.18
- Lazy context-switching of FPSIMD registers on arm64
- Allow virtual redistributors to be part of two or more MMIO ranges
|
|
Now all the internals are ready to handle multiple redistributor
regions, let's allow the userspace to register them.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
On vcpu first run, we eventually know the actual number of vcpus.
This is a synchronization point to check all redistributors
were assigned. On kvm_vgic_map_resources() we check both dist and
redist were set, eventually check potential base address inconsistencies.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
vgic_v3_check_base() currently only handles the case of a unique
legacy redistributor region whose size is not explicitly set but
inferred, instead, from the number of online vcpus.
We adapt it to handle the case of multiple redistributor regions
with explicitly defined size. We rely on two new helpers:
- vgic_v3_rdist_overlap() is used to detect overlap with the dist
region if defined
- vgic_v3_rd_region_size computes the size of the redist region,
would it be a legacy unique region or a new explicitly sized
region.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
We introduce vgic_v3_rdist_free_slot to help identifying
where we can place a new 2x64KB redistributor.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
At the moment KVM supports a single rdist region. We want to
support several separate rdist regions so let's introduce a list
of them. This patch currently only cares about a single
entry in this list as the functionality to register several redist
regions is not yet there. So this only translates the existing code
into something functionally similar using that new data struct.
The redistributor region handle is stored in the vgic_cpu structure
to allow later computation of the TYPER last bit.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
kvm_read_guest() will eventually look up in kvm_memslots(), which requires
either to hold the kvm->slots_lock or to be inside a kvm->srcu critical
section.
In contrast to x86 and s390 we don't take the SRCU lock on every guest
exit, so we have to do it individually for each kvm_read_guest() call.
Use the newly introduced wrapper for that.
Cc: Stable <stable@vger.kernel.org> # 4.12+
Reported-by: Jan Glauber <jan.glauber@caviumnetworks.com>
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Now that we make sure we don't inject multiple instances of the
same GICv2 SGI at the same time, we've made another bug more
obvious:
If we exit with an active SGI, we completely lose track of which
vcpu it came from. On the next entry, we restore it with 0 as a
source, and if that wasn't the right one, too bad. While this
doesn't seem to trouble GIC-400, the architectural model gets
offended and doesn't deactivate the interrupt on EOI.
Another connected issue is that we will happilly make pending
an interrupt from another vcpu, overriding the above zero with
something that is just as inconsistent. Don't do that.
The final issue is that we signal a maintenance interrupt when
no pending interrupts are present in the LR. Assuming we've fixed
the two issues above, we end-up in a situation where we keep
exiting as soon as we've reached the active state, and not be
able to inject the following pending.
The fix comes in 3 parts:
- GICv2 SGIs have their source vcpu saved if they are active on
exit, and restored on entry
- Multi-SGIs cannot go via the Pending+Active state, as this would
corrupt the source field
- Multi-SGIs are converted to using MI on EOI instead of NPIE
Fixes: 16ca6a607d84bef0 ("KVM: arm/arm64: vgic: Don't populate multiple LRs with the same vintid")
Reported-by: Mark Rutland <mark.rutland@arm.com>
Tested-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: Christoffer Dall <christoffer.dall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
It was recently reported that VFIO mediated devices, and anything
that VFIO exposes as level interrupts, do no strictly follow the
expected logic of such interrupts as it only lowers the input
line when the guest has EOId the interrupt at the GIC level, rather
than when it Acked the interrupt at the device level.
THe GIC's Active+Pending state is fundamentally incompatible with
this behaviour, as it prevents KVM from observing the EOI, and in
turn results in VFIO never dropping the line. This results in an
interrupt storm in the guest, which it really never expected.
As we cannot really change VFIO to follow the strict rules of level
signalling, let's forbid the A+P state altogether, as it is in the
end only an optimization. It ensures that we will transition via
an invalid state, which we can use to notify VFIO of the EOI.
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Shunyong Yang <shunyong.yang@hxt-semitech.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
Resolve conflicts with current mainline
|
|
We can finally get completely rid of any calls to the VGICv3
save/restore functions when the AP lists are empty on VHE systems. This
requires carefully factoring out trap configuration from saving and
restoring state, and carefully choosing what to do on the VHE and
non-VHE path.
One of the challenges is that we cannot save/restore the VMCR lazily
because we can only write the VMCR when ICC_SRE_EL1.SRE is cleared when
emulating a GICv2-on-GICv3, since otherwise all Group-0 interrupts end
up being delivered as FIQ.
To solve this problem, and still provide fast performance in the fast
path of exiting a VM when no interrupts are pending (which also
optimized the latency for actually delivering virtual interrupts coming
from physical interrupts), we orchestrate a dance of only doing the
activate/deactivate traps in vgic load/put for VHE systems (which can
have ICC_SRE_EL1.SRE cleared when running in the host), and doing the
configuration on every round-trip on non-VHE systems.
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
The APRs can only have bits set when the guest acknowledges an interrupt
in the LR and can only have a bit cleared when the guest EOIs an
interrupt in the LR. Therefore, if we have no LRs with any
pending/active interrupts, the APR cannot change value and there is no
need to clear it on every exit from the VM (hint: it will have already
been cleared when we exited the guest the last time with the LRs all
EOIed).
The only case we need to take care of is when we migrate the VCPU away
from a CPU or migrate a new VCPU onto a CPU, or when we return to
userspace to capture the state of the VCPU for migration. To make sure
this works, factor out the APR save/restore functionality into separate
functions called from the VCPU (and by extension VGIC) put/load hooks.
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
There is really no need to store the vgic_elrsr on the VGIC data
structures as the only need we have for the elrsr is to figure out if an
LR is inactive when we save the VGIC state upon returning from the
guest. We can might as well store this in a temporary local variable.
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
The vgic code is trying to be clever when injecting GICv2 SGIs,
and will happily populate LRs with the same interrupt number if
they come from multiple vcpus (after all, they are distinct
interrupt sources).
Unfortunately, this is against the letter of the architecture,
and the GICv2 architecture spec says "Each valid interrupt stored
in the List registers must have a unique VirtualID for that
virtual CPU interface.". GICv3 has similar (although slightly
ambiguous) restrictions.
This results in guests locking up when using GICv2-on-GICv3, for
example. The obvious fix is to stop trying so hard, and inject
a single vcpu per SGI per guest entry. After all, pending SGIs
with multiple source vcpus are pretty rare, and are mostly seen
in scenario where the physical CPUs are severely overcomitted.
But as we now only inject a single instance of a multi-source SGI per
vcpu entry, we may delay those interrupts for longer than strictly
necessary, and run the risk of injecting lower priority interrupts
in the meantime.
In order to address this, we adopt a three stage strategy:
- If we encounter a multi-source SGI in the AP list while computing
its depth, we force the list to be sorted
- When populating the LRs, we prevent the injection of any interrupt
of lower priority than that of the first multi-source SGI we've
injected.
- Finally, the injection of a multi-source SGI triggers the request
of a maintenance interrupt when there will be no pending interrupt
in the LRs (HCR_NPIE).
At the point where the last pending interrupt in the LRs switches
from Pending to Active, the maintenance interrupt will be delivered,
allowing us to add the remaining SGIs using the same process.
Cc: stable@vger.kernel.org
Fixes: 0919e84c0fc1 ("KVM: arm/arm64: vgic-new: Add IRQ sync/flush framework")
Acked-by: Christoffer Dall <cdall@kernel.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
Level-triggered mapped IRQs are special because we only observe rising
edges as input to the VGIC, and we don't set the EOI flag and therefore
are not told when the level goes down, so that we can re-queue a new
interrupt when the level goes up.
One way to solve this problem is to side-step the logic of the VGIC and
special case the validation in the injection path, but it has the
unfortunate drawback of having to peak into the physical GIC state
whenever we want to know if the interrupt is pending on the virtual
distributor.
Instead, we can maintain the current semantics of a level triggered
interrupt by sort of treating it as an edge-triggered interrupt,
following from the fact that we only observe an asserting edge. This
requires us to be a bit careful when populating the LRs and when folding
the state back in though:
* We lower the line level when populating the LR, so that when
subsequently observing an asserting edge, the VGIC will do the right
thing.
* If the guest never acked the interrupt while running (for example if
it had masked interrupts at the CPU level while running), we have
to preserve the pending state of the LR and move it back to the
line_level field of the struct irq when folding LR state.
If the guest never acked the interrupt while running, but changed the
device state and lowered the line (again with interrupts masked) then
we need to observe this change in the line_level.
Both of the above situations are solved by sampling the physical line
and set the line level when folding the LR back.
* Finally, if the guest never acked the interrupt while running and
sampling the line reveals that the device state has changed and the
line has been lowered, we must clear the physical active state, since
we will otherwise never be told when the interrupt becomes asserted
again.
This has the added benefit of making the timer optimization patches
(https://lists.cs.columbia.edu/pipermail/kvmarm/2017-July/026343.html) a
bit simpler, because the timer code doesn't have to clear the active
state on the sync anymore. It also potentially improves the performance
of the timer implementation because the GIC knows the state or the LR
and only needs to clear the
active state when the pending bit in the LR is still set, where the
timer has to always clear it when returning from running the guest with
an injected timer interrupt.
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
|
|
The current pending table parsing code assumes that we keep the
previous read of the pending bits, but keep that variable in
the current block, making sure it is discarded on each loop.
We end-up using whatever is on the stack. Who knows, it might
just be the right thing...
Fixes: 280771252c1ba ("KVM: arm64: vgic-v3: KVM_DEV_ARM_VGIC_SAVE_PENDING_TABLES")
Cc: stable@vger.kernel.org # 4.12
Reported-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
|
|
All it takes is the has_v4 flag to be set in gic_kvm_info
as well as "kvm-arm.vgic_v4_enable=1" being passed on the
command line for GICv4 to be enabled in KVM.
Acked-by: Christoffer Dall <cdall@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
|
|
We are about to optimize our timer handling logic which involves
injecting irqs to the vgic directly from the irq handler.
Unfortunately, the injection path can take any AP list lock and irq lock
and we must therefore make sure to use spin_lock_irqsave where ever
interrupts are enabled and we are taking any of those locks, to avoid
deadlocking between process context and the ISR.
This changes a lot of the VGIC code, but the good news are that the
changes are mostly mechanical.
Acked-by: Marc Zyngier <marc,zyngier@arm.com>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
|
|
In order to facilitate debug, let's log which class of GICv3 system
registers are trapped.
Tested-by: Alexander Graf <agraf@suse.de>
Acked-by: David Daney <david.daney@cavium.com>
Acked-by: Christoffer Dall <cdall@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
|
|
Now that we're able to safely handle common sysreg access, let's
give the user the opportunity to enable it by passing a specific
command-line option (vgic_v3.common_trap).
Tested-by: Alexander Graf <agraf@suse.de>
Acked-by: David Daney <david.daney@cavium.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <cdall@linaro.org>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
|
|
Some Cavium Thunder CPUs suffer a problem where a KVM guest may
inadvertently cause the host kernel to quit receiving interrupts.
Use the Group-0/1 trapping in order to deal with it.
[maz]: Adapted patch to the Group-0/1 trapping, reworked commit log
Tested-by: Alexander Graf <agraf@suse.de>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: David Daney <david.daney@cavium.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
|
|
Now that we're able to safely handle Group-0 sysreg access, let's
give the user the opportunity to enable it by passing a specific
command-line option (vgic_v3.group0_trap).
Tested-by: Alexander Graf <agraf@suse.de>
Acked-by: David Daney <david.daney@cavium.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
|
|
In order to be able to trap Group-0 GICv3 system registers, we need to
set ICH_HCR_EL2.TALL0 begore entering the guest. This is conditionnaly
done after having restored the guest's state, and cleared on exit.
Tested-by: Alexander Graf <agraf@suse.de>
Acked-by: David Daney <david.daney@cavium.com>
Acked-by: Christoffer Dall <cdall@linaro.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
|
|
Now that we're able to safely handle Group-1 sysreg access, let's
give the user the opportunity to enable it by passing a specific
command-line option (vgic_v3.group1_trap).
Tested-by: Alexander Graf <agraf@suse.de>
Acked-by: David Daney <david.daney@cavium.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <cdall@linaro.org>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
|
|
In order to be able to trap Group-1 GICv3 system registers, we need to
set ICH_HCR_EL2.TALL1 before entering the guest. This is conditionally
done after having restored the guest's state, and cleared on exit.
Tested-by: Alexander Graf <agraf@suse.de>
Acked-by: David Daney <david.daney@cavium.com>
Acked-by: Christoffer Dall <cdall@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
|
|
In order to start handling guest access to GICv3 system registers,
let's add a hook that will get called when we trap a system register
access. This is gated by a new static key (vgic_v3_cpuif_trap).
Tested-by: Alexander Graf <agraf@suse.de>
Acked-by: David Daney <david.daney@cavium.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Christoffer Dall <cdall@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
|
|
We have been a little loose with our intermediate VMCR representation
where we had a 'ctlr' field, but we failed to differentiate between the
GICv2 GICC_CTLR and ICC_CTLR_EL1 layouts, and therefore ended up mapping
the wrong bits into the individual fields of the ICH_VMCR_EL2 when
emulating a GICv2 on a GICv3 system.
Fix this by using explicit fields for the VMCR bits instead.
Cc: Eric Auger <eric.auger@redhat.com>
Reported-by: wanghaibin <wanghaibin.wang@huawei.com>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Tested-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
When an interrupt is injected with the HW bit set (indicating that
deactivation should be propagated to the physical distributor),
special care must be taken so that we never mark the corresponding
LR with the Active+Pending state (as the pending state is kept in
the physycal distributor).
Cc: stable@vger.kernel.org
Fixes: 59529f69f504 ("KVM: arm/arm64: vgic-new: Add GICv3 world switch backend")
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Christoffer Dall <cdall@linaro.org>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
|
|
We have to register the ITS iodevice before running the VM, because in
migration scenarios, we may be restoring a live device that wishes to
inject MSIs before the VCPUs have started.
All we need to register the ITS io device is the base address of the
ITS, so we can simply register that when the base address of the ITS is
set.
[ Code to fix concurrency issues when setting the ITS base address and
to fix the undef base address check written by Marc Zyngier ]
Signed-off-by: Christoffer Dall <cdall@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
|
|
Instead of waiting with registering KVM iodevs until the first VCPU is
run, we can actually create the iodevs when the redist base address is
set. The only downside is that we must now also check if we need to do
this for VCPUs which are created after creating the VGIC, because there
is no enforced ordering between creating the VGIC (and setting its base
addresses) and creating the VCPUs.
Signed-off-by: Christoffer Dall <cdall@linaro.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
|
|
As we are about to fiddle with the IO device registration mechanism,
let's be a little more careful when setting base addresses as early as
possible. When setting a base address, we can check that there's
address space enough for its scope and when the last of the two
base addresses (dist and redist) get set, we can also check if the
regions overlap at that time.
This allows us to provide error messages to the user at time when trying
to set the base address, as opposed to later when trying to run the VM.
To do this, we make vgic_v3_check_base available in the core vgic-v3
code as well as in the other parts of the GICv3 code, namely the MMIO
config code.
We also return true for undefined base addresses so that the function
can be used before all base addresses are set; all callers already check
for uninitialized addresses before calling this function.
Signed-off-by: Christoffer Dall <cdall@linaro.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
|
|
Split out the function to register all the redistributor iodevs into a
function that handles a single redistributor at a time in preparation
for being able to call this per VCPU as these get created.
Signed-off-by: Christoffer Dall <cdall@linaro.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
|
|
This patch adds a new attribute to GICV3 KVM device
KVM_DEV_ARM_VGIC_GRP_CTRL group. This allows userspace to
flush all GICR pending tables into guest RAM.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Christoffer Dall <cdall@linaro.org>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
this new helper synchronizes the irq pending_latch
with the LPI pending bit status found in rdist pending table.
As the status is consumed, we reset the bit in pending table.
As we need the PENDBASER_ADDRESS() in vgic-v3, let's move its
definition in the irqchip header. We restore the full length
of the field, ie [51:16]. Same for PROPBASER_ADDRESS with full
field length of [51:12].
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Christoffer Dall <cdall@linaro.org>
|
|
When emulating a GICv2-on-GICv3, special care must be taken to only
save/restore VMCR_EL2 when ICC_SRE_EL1.SRE is cleared. Otherwise,
all Group-0 interrupts end-up being delivered as FIQ, which is
probably not what the guest expects, as demonstrated here with
an unhappy EFI:
FIQ Exception at 0x000000013BD21CC4
This means that we cannot perform the load/put trick when dealing
with VMCR_EL2 (because the host has SRE set), and we have to deal
with it in the world-switch.
Fortunately, this is not the most common case (modern guests should
be able to deal with GICv3 directly), and the performance is not worse
than what it was before the VMCR optimization.
Reviewed-by: Christoffer Dall <cdall@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
|
|
There is no need to call any functions to fold LRs when we don't use any
LRs and we don't need to mess with overflow flags, take spinlocks, or
prune the AP list if the AP list is empty.
Note: list_empty is a single atomic read (uses READ_ONCE) and can
therefore check if a list is empty or not without the need to take the
spinlock protecting the list.
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
|
|
Since we always read back the LRs that we wrote to the guest and the
MISR and EISR registers simply provide a summary of the configuration of
the bits in the LRs, there is really no need to read back those status
registers and process them. We might as well just signal the
notifyfd when folding the LR state and save some cycles in the process.
We now clear the underflow bit in the fold_lr_state functions as we only
need to clear this bit if we had used all the LRs, so this is as good a
place as any to do that work.
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
|
|
We don't have to save/restore the VMCR on every entry to/from the guest,
since on GICv2 we can access the control interface from EL1 and on VHE
systems with GICv3 we can access the control interface from KVM running
in EL2.
GICv3 systems without VHE becomes the rare case, which has to
save/restore the register on each round trip.
Note that userspace accesses may see out-of-date values if the VCPU is
running while accessing the VGIC state via the KVM device API, but this
is already the case and it is up to userspace to quiesce the CPUs before
reading the CPU registers from the GIC for an up-to-date view.
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <cdall@cs.columbia.edu>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
|
|
Our GICv3 emulation always presents ICC_SRE_EL1 with DIB/DFB set to
zero, which implies that there is a way to bypass the GIC and
inject raw IRQ/FIQ by driving the CPU pins.
Of course, we don't allow that when the GIC is configured, but
we fail to indicate that to the guest. The obvious fix is to
set these bits (and never let them being changed again).
Reported-by: Peter Maydell <peter.maydell@linaro.org>
Acked-by: Christoffer Dall <cdall@linaro.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
VGICv3 CPU interface registers are accessed using
KVM_DEV_ARM_VGIC_CPU_SYSREGS ioctl. These registers are accessed
as 64-bit. The cpu MPIDR value is passed along with register id.
It is used to identify the cpu for registers access.
The VM that supports SEIs expect it on destination machine to handle
guest aborts and hence checked for ICC_CTLR_EL1.SEIS compatibility.
Similarly, VM that supports Affinity Level 3 that is required for AArch64
mode, is required to be supported on destination machine. Hence checked
for ICC_CTLR_EL1.A3V compatibility.
The arch/arm64/kvm/vgic-sys-reg-v3.c handles read and write of VGIC
CPU registers for AArch64.
For AArch32 mode, arch/arm/kvm/vgic-v3-coproc.c file is created but
APIs are not implemented.
Updated arch/arm/include/uapi/asm/kvm.h with new definitions
required to compile for AArch32.
The version of VGIC v3 specification is defined here
Documentation/virtual/kvm/devices/arm-vgic-v3.txt
Acked-by: Christoffer Dall <christoffer.dall@linaro.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Pavel Fedin <p.fedin@samsung.com>
Signed-off-by: Vijaya Kumar K <Vijaya.Kumar@cavium.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
ICC_VMCR_EL2 supports virtual access to ICC_IGRPEN1_EL1.Enable
and ICC_IGRPEN0_EL1.Enable fields. Add grpen0 and grpen1 member
variables to struct vmcr to support read and write of these fields.
Also refactor vgic_set_vmcr and vgic_get_vmcr() code.
Drop ICH_VMCR_CTLR_SHIFT and ICH_VMCR_CTLR_MASK macros and instead
use ICH_VMCR_EOI* and ICH_VMCR_CBPR* macros.
Signed-off-by: Vijaya Kumar K <Vijaya.Kumar@cavium.com>
Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
One of the goals behind the VGIC redesign was to get rid of cached or
intermediate state in the data structures, but we decided to allow
ourselves to precompute the pending value of an IRQ based on the line
level and pending latch state. However, this has now become difficult
to base proper GICv3 save/restore on, because there is a potential to
modify the pending state without knowing if an interrupt is edge or
level configured.
See the following post and related message for more background:
https://lists.cs.columbia.edu/pipermail/kvmarm/2017-January/023195.html
This commit gets rid of the precomputed pending field in favor of a
function that calculates the value when needed, irq_is_pending().
The soft_pending field is renamed to pending_latch to represent that
this latch is the equivalent hardware latch which gets manipulated by
the input signal for edge-triggered interrupts and when writing to the
SPENDR/CPENDR registers.
After this commit save/restore code should be able to simply restore the
pending_latch state, line_level state, and config state in any order and
get the desired result.
Reviewed-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Tested-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
|
|
Dmitry Vyukov reported that the syzkaller fuzzer triggered a
deadlock in the vgic setup code when an error was detected, as
the cleanup code tries to take a lock that is already held by
the setup code.
The fix is to avoid retaking the lock when cleaning up, by
telling the cleanup function that we already hold it.
Cc: stable@vger.kernel.org
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
