When a vPE is made resident, the GIC starts parsing the virtual pending
table to deliver pending interrupts. This takes place asynchronously,
and can at times take a long while. Long enough that the vcpu enters
the guest and hits WFI before any interrupt has been signaled yet.
The vcpu then exits, blocks, and now gets a doorbell. Rince, repeat.
In order to avoid the above, a (optional on GICv4, mandatory on v4.1)
feature allows the GIC to feedback to the hypervisor whether it is
done parsing the VPT by clearing the GICR_VPENDBASER.Dirty bit.
The hypervisor can then wait until the GIC is ready before actually
running the vPE.
Plug the detection code as well as polling on vPE schedule. While
at it, tidy-up the kernel message that displays the GICv4 optional
features.
Reviewed-by: Zenghui Yu <yuzenghui@huawei.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
In order to hide some of the differences between v4.0 and v4.1, move
the doorbell management out of the KVM code, and into the GICv4-specific
layer. This allows the calling code to ask for the doorbell when blocking,
and otherwise to leave the doorbell permanently disabled.
This matches the v4.1 code perfectly, and only results in a minor
refactoring of the v4.0 code.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Zenghui Yu <yuzenghui@huawei.com>
Link: https://lore.kernel.org/r/20200304203330.4967-14-maz@kernel.org
Just like for vLPIs, there is some configuration information that cannot
be directly communicated through the normal irqchip API, and we have to
use our good old friend set_vcpu_affinity as a side-band communication
mechanism.
This is used to configure group and priority for a given vSGI.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Zenghui Yu <yuzenghui@huawei.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Link: https://lore.kernel.org/r/20200304203330.4967-13-maz@kernel.org
To implement the get/set_irqchip_state callbacks (limited to the
PENDING state), we have to use a particular set of hacks:
- Reading the pending state is done by using a pair of new redistributor
registers (GICR_VSGIR, GICR_VSGIPENDR), which allow the 16 interrupts
state to be retrieved.
- Setting the pending state is done by generating it as we'd otherwise do
for a guest (writing to GITS_SGIR).
- Clearing the pending state is done by emitting a VSGI command with the
"clear" bit set.
This requires some interesting locking though:
- When talking to the redistributor, we must make sure that the VPE
affinity doesn't change, hence taking the VPE lock.
- At the same time, we must ensure that nobody accesses the same
redistributor's GICR_VSGIR registers for a different VPE, which
would corrupt the reading of the pending bits. We thus take the
per-RD spinlock. Much fun.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Zenghui Yu <yuzenghui@huawei.com>
Link: https://lore.kernel.org/r/20200304203330.4967-12-maz@kernel.org
The GICv4.1 ITS has yet another new command (VSGI) which allows
a VPE-targeted SGI to be configured (or have its pending state
cleared). Add support for this command and plumb it into the
activate irqdomain callback so that it is ready to be used.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Zenghui Yu <yuzenghui@huawei.com>
Link: https://lore.kernel.org/r/20200304203330.4967-10-maz@kernel.org
Since GICv4.1 has the capability to inject 16 SGIs into each VPE,
and that I'm keen not to invent too many specific interfaces to
manipulate these interrupts, let's pretend that each of these SGIs
is an actual Linux interrupt.
For that matter, let's introduce a minimal irqchip and irqdomain
setup that will get fleshed up in the following patches.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Zenghui Yu <yuzenghui@huawei.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Link: https://lore.kernel.org/r/20200304203330.4967-9-maz@kernel.org
There is no special reason to set virtual LPI pending table as
non-shareable. If we choose to hard code the shareability without
probing, Inner-Shareable is likely to be a better choice, as the
VPEs can move around and benefit from having the redistributors
snooping each other's cache, if that's something they can do.
Furthermore, Hisilicon hip08 ends up with unspecified errors when
mixing shareability attributes. So let's move to IS attributes for
the VPT. This has also been tested on D05 and didn't show any
regression.
Signed-off-by: Heyi Guo <guoheyi@huawei.com>
[maz: rewrote commit message]
Signed-off-by: Marc Zyngier <maz@kernel.org>
Tested-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20191130073849.38378-1-guoheyi@huawei.com
The GICv4.1 spec says that it is CONTRAINED UNPREDICTABLE to write to
any of the GICR_INV{LPI,ALL}R registers if GICR_SYNCR.Busy == 1.
To deal with it, we must ensure that only a single invalidation can
happen at a time for a given redistributor. Add a per-RD lock to that
effect and take it around the invalidation/syncr-read to deal with this.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Zenghui Yu <yuzenghui@huawei.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Link: https://lore.kernel.org/r/20200304203330.4967-6-maz@kernel.org
Before GICv4.1, all operations would be serialized with the affinity
changes by virtue of using the same ITS command queue. With v4.1, things
change, as invalidations (and a number of other operations) are issued
using the redistributor MMIO frame.
We must thus make sure that these redistributor accesses cannot race
against aginst the affinity change, or we may end-up talking to the
wrong redistributor.
To ensure this, we expand the irq_to_cpuid() helper to take a spinlock
when the LPI is mapped to a vLPI (a new per-VPE lock) on each operation
that requires mutual exclusion.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Zenghui Yu <yuzenghui@huawei.com>
Link: https://lore.kernel.org/r/20200304203330.4967-4-maz@kernel.org
To allow the direct injection of SGIs into a guest, the GICv4.1
architecture has to sacrifice the Active state so that SGIs look
a lot like LPIs (they are injected by the same mechanism).
In order not to break existing software, the architecture gives
offers guests OSs the choice: SGIs with or without an active
state. It is the hypervisors duty to honor the guest's choice.
For this, the architecture offers a discovery bit indicating whether
the GIC supports GICv4.1 SGIs (GICD_TYPER2.nASSGIcap), and another
bit indicating whether the guest wants Active-less SGIs or not
(controlled by GICD_CTLR.nASSGIreq).
A hypervisor not supporting GICv4.1 SGIs would leave nASSGIcap
clear, and a guest not knowing about GICv4.1 SGIs (or definitely
wanting an Active state) would leave nASSGIreq clear (both being
thankfully backward compatible with older revisions of the GIC).
Since Linux is perfectly happy without an active state on SGIs,
inform the hypervisor that we'll use that if offered.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Zenghui Yu <yuzenghui@huawei.com>
Link: https://lore.kernel.org/r/20200304203330.4967-2-maz@kernel.org
Currently, we will not set vpe_l1_page for the current RD if we can
inherit the vPE configuration table from another RD (or ITS), which
results in an inconsistency between RDs within the same CommonLPIAff
group.
Let's rename it to vpe_l1_base to indicate the base address of the
vPE configuration table of this RD, and set it properly for *all*
v4.1 redistributors.
Signed-off-by: Zenghui Yu <yuzenghui@huawei.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20200206075711.1275-3-yuzenghui@huawei.com
The ITS VMAPP command gains some new fields with GICv4.1:
- a default doorbell, which allows a single doorbell to be used for
all the VLPIs routed to a given VPE
- a pointer to the configuration table (instead of having it in a register
that gets context switched)
- a flag indicating whether this is the first map or the last unmap for
this particular VPE
- a flag indicating whether the pending table is known to be zeroed, or not
Plumb in the new fields in the VMAPP builder, and add the map/unmap
refcounting so that the ITS can do the right thing.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Zenghui Yu <yuzenghui@huawei.com>
Link: https://lore.kernel.org/r/20191224111055.11836-7-maz@kernel.org
GICv4.1 defines a new VPE table that is potentially shared between
both the ITSs and the redistributors, following complicated affinity
rules.
To make things more confusing, the programming of this table at
the redistributor level is reusing the GICv4.0 GICR_VPROPBASER register
for something completely different.
The code flow is somewhat complexified by the need to respect the
affinities required by the HW, meaning that tables can either be
inherited from a previously discovered ITS or redistributor.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Zenghui Yu <yuzenghui@huawei.com>
Link: https://lore.kernel.org/r/20191224111055.11836-6-maz@kernel.org
GICv4.1 supports the RVPEID ("Residency per vPE ID"), which allows for
a much efficient way of making virtual CPUs resident (to allow direct
injection of interrupts).
The functionnality needs to be discovered on each and every redistributor
in the system, and disabled if the settings are inconsistent.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Zenghui Yu <yuzenghui@huawei.com>
Link: https://lore.kernel.org/r/20191224111055.11836-2-maz@kernel.org
Pull irq updates from Ingo Molnar:
"Most of the IRQ subsystem changes in this cycle were irq-chip driver
updates:
- Qualcomm PDC wakeup interrupt support
- Layerscape external IRQ support
- Broadcom bcm7038 PM and wakeup support
- Ingenic driver cleanup and modernization
- GICv3 ITS preparation for GICv4.1 updates
- GICv4 fixes
There's also the series from Frederic Weisbecker that fixes memory
ordering bugs for the irq-work logic, whose primary fix is to turn
work->irq_work.flags into an atomic variable and then convert the
complex (and buggy) atomic_cmpxchg() loop in irq_work_claim() into a
much simpler atomic_fetch_or() call.
There are also various smaller cleanups"
* 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (44 commits)
pinctrl/sdm845: Add PDC wakeup interrupt map for GPIOs
pinctrl/msm: Setup GPIO chip in hierarchy
irqchip/qcom-pdc: Add irqchip set/get state calls
irqchip/qcom-pdc: Add irqdomain for wakeup capable GPIOs
irqchip/qcom-pdc: Do not toggle IRQ_ENABLE during mask/unmask
irqchip/qcom-pdc: Update max PDC interrupts
of/irq: Document properties for wakeup interrupt parent
genirq: Introduce irq_chip_get/set_parent_state calls
irqdomain: Add bus token DOMAIN_BUS_WAKEUP
genirq: Fix function documentation of __irq_alloc_descs()
irq_work: Fix IRQ_WORK_BUSY bit clearing
irqchip/ti-sci-inta: Use ERR_CAST inlined function instead of ERR_PTR(PTR_ERR(...))
irq_work: Slightly simplify IRQ_WORK_PENDING clearing
irq_work: Fix irq_work_claim() memory ordering
irq_work: Convert flags to atomic_t
irqchip: Ingenic: Add process for more than one irq at the same time.
irqchip: ingenic: Alloc generic chips from IRQ domain
irqchip: ingenic: Get virq number from IRQ domain
irqchip: ingenic: Error out if IRQ domain creation failed
irqchip: ingenic: Drop redundant irq_suspend / irq_resume functions
...
Pull KVM updates from Paolo Bonzini:
"ARM:
- data abort report and injection
- steal time support
- GICv4 performance improvements
- vgic ITS emulation fixes
- simplify FWB handling
- enable halt polling counters
- make the emulated timer PREEMPT_RT compliant
s390:
- small fixes and cleanups
- selftest improvements
- yield improvements
PPC:
- add capability to tell userspace whether we can single-step the
guest
- improve the allocation of XIVE virtual processor IDs
- rewrite interrupt synthesis code to deliver interrupts in virtual
mode when appropriate.
- minor cleanups and improvements.
x86:
- XSAVES support for AMD
- more accurate report of nested guest TSC to the nested hypervisor
- retpoline optimizations
- support for nested 5-level page tables
- PMU virtualization optimizations, and improved support for nested
PMU virtualization
- correct latching of INITs for nested virtualization
- IOAPIC optimization
- TSX_CTRL virtualization for more TAA happiness
- improved allocation and flushing of SEV ASIDs
- many bugfixes and cleanups"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (127 commits)
kvm: nVMX: Relax guest IA32_FEATURE_CONTROL constraints
KVM: x86: Grab KVM's srcu lock when setting nested state
KVM: x86: Open code shared_msr_update() in its only caller
KVM: Fix jump label out_free_* in kvm_init()
KVM: x86: Remove a spurious export of a static function
KVM: x86: create mmu/ subdirectory
KVM: nVMX: Remove unnecessary TLB flushes on L1<->L2 switches when L1 use apic-access-page
KVM: x86: remove set but not used variable 'called'
KVM: nVMX: Do not mark vmcs02->apic_access_page as dirty when unpinning
KVM: vmx: use MSR_IA32_TSX_CTRL to hard-disable TSX on guest that lack it
KVM: vmx: implement MSR_IA32_TSX_CTRL disable RTM functionality
KVM: x86: implement MSR_IA32_TSX_CTRL effect on CPUID
KVM: x86: do not modify masked bits of shared MSRs
KVM: x86: fix presentation of TSX feature in ARCH_CAPABILITIES
KVM: PPC: Book3S HV: XIVE: Fix potential page leak on error path
KVM: PPC: Book3S HV: XIVE: Free previous EQ page when setting up a new one
KVM: nVMX: Assume TLB entries of L1 and L2 are tagged differently if L0 use EPT
KVM: x86: Unexport kvm_vcpu_reload_apic_access_page()
KVM: nVMX: add CR4_LA57 bit to nested CR4_FIXED1
KVM: nVMX: Use semi-colon instead of comma for exit-handlers initialization
...
Marc Zyngier
Heyi Guo
Zenghui Yu
Linus Torvalds