In our KVM GICv2 realize function, we try to cope with old kernels
that don't provide the device control API (KVM_CAP_DEVICE_CTRL): we
try to use the device control, and if that fails we fall back to
assuming that the kernel has the old style KVM_CREATE_IRQCHIP and
that it will provide a GICv2.
This doesn't cater for the possibility of a kernel and hardware which
only provide a GICv3, which is very common now. On that setup we
will abort() later on in kvm_arm_pmu_set_irq() when we try to wire up
an interrupt to the GIC we failed to create:
qemu-system-aarch64: PMU: KVM_SET_DEVICE_ATTR: Invalid argument
qemu-system-aarch64: failed to set irq for PMU
Aborted
If the kernel advertises KVM_CAP_DEVICE_CTRL we should trust it if it
says it can't create a GICv2, rather than assuming it has one. We
can then produce a more helpful error message including a hint about
the most probable reason for the failure.
If the kernel doesn't advertise KVM_CAP_DEVICE_CTRL then it is truly
ancient by this point but we might as well still fall back to a
KVM_CREATE_IRQCHIP GICv2.
With this patch then the user misconfiguration which previously
caused an abort now prints:
qemu-system-aarch64: Initialization of device kvm-arm-gic failed: error creating in-kernel VGIC: No such device
Perhaps the host CPU does not support GICv2?
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Tested-by: Andrew Jones <drjones@redhat.com>
Message-id: 20200225182435.1131-1-peter.maydell@linaro.org
Host kernels that expose the KVM_CAP_ARM_IRQ_LINE_LAYOUT_2 capability
allow injection of interrupts along with vcpu ids larger than 255.
Let's encode the vpcu id on 12 bits according to the upgraded KVM_IRQ_LINE
ABI when needed.
Given that we have two callsites that need to assemble
the value for kvm_set_irq(), a new helper routine, kvm_arm_set_irq
is introduced.
Without that patch qemu exits with "kvm_set_irq: Invalid argument"
message.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reported-by: Zenghui Yu <yuzenghui@huawei.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Acked-by: Marc Zyngier <maz@kernel.org>
Message-id: 20191003154640.22451-3-eric.auger@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Add the necessary parts of the virtualization extensions state to the
GIC state. We choose to increase the size of the CPU interfaces state to
add space for the vCPU interfaces (the GIC_NCPU_VCPU macro). This way,
we'll be able to reuse most of the CPU interface code for the vCPUs.
The only exception is the APR value, which is stored in h_apr in the
virtual interface state for vCPUs. This is due to some complications
with the GIC VMState, for which we don't want to break backward
compatibility. APRs being stored in 2D arrays, increasing the second
dimension would lead to some ugly VMState description. To avoid
that, we keep it in h_apr for vCPUs.
The vCPUs are numbered from GIC_NCPU to (GIC_NCPU * 2) - 1. The
`gic_is_vcpu` function help to determine if a given CPU id correspond to
a physical CPU or a virtual one.
For the in-kernel KVM VGIC, since the exposed VGIC does not implement
the virtualization extensions, we report an error if the corresponding
property is set to true.
Signed-off-by: Luc Michel <luc.michel@greensocs.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20180727095421.386-6-luc.michel@greensocs.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
In preparation for the virtualization extensions implementation,
refactor the name of the functions and macros that act on the GIC
distributor to make that fact explicit. It will be useful to
differentiate them from the ones that will act on the virtual
interfaces.
Signed-off-by: Luc Michel <luc.michel@greensocs.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Sai Pavan Boddu <sai.pavan.boddu@xilinx.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20180727095421.386-2-luc.michel@greensocs.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
for KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION attribute, the attribute
data pointed to by kvm_device_attr.addr is a OR of the
redistributor region address and other fields such as the index
of the redistributor region and the number of redistributors the
region can contain.
The existing machine init done notifier framework sets the address
field to the actual address of the device and does not allow to OR
this value with other fields.
This patch extends the KVMDevice struct with a new kda_addr_ormask
member. Its value is passed at registration time and OR'ed with the
resolved address on kvm_arm_set_device_addr().
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1529072910-16156-3-git-send-email-eric.auger@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
kvm_irqchip_create called by kvm_init will call kvm_init_irq_routing to
initialize global capability variables. If we call kvm_init_irq_routing in
GIC realize function, previous allocated memory will leak.
Fix this by deleting the unnecessary call.
Signed-off-by: Shannon Zhao <zhaoshenglong@huawei.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Message-id: 1527750994-14360-1-git-send-email-zhaoshenglong@huawei.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
In some circumstances, we don't want to abort if the
kvm_device_access fails. This will be the case during ITS
migration, in case the ITS table save/restore fails because
the guest did not program the vITS correctly. So let's pass an
error object to the function and return the ioctl value. New
callers will be able to make a decision upon this returned
value.
Existing callers pass &error_abort which will cause the
function to abort on failure.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Message-id: 1497023553-18411-2-git-send-email-eric.auger@redhat.com
[PMM: wrapped long line]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
This allows us to remove lots of includes of migration/migration.h
Signed-off-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
If a migration is already in progress and somebody attempts
to add a migration blocker, this should rightly fail.
Add an errp parameter and a retcode return value to migrate_add_blocker.
Signed-off-by: John Snow <jsnow@redhat.com>
Signed-off-by: Ashijeet Acharya <ashijeetacharya@gmail.com>
Message-Id: <1484566314-3987-5-git-send-email-ashijeetacharya@gmail.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Acked-by: Greg Kurz <groug@kaod.org>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Merged with recent 'Allow invtsc migration' change
Remove unused debugging code to fix native building on aarch64. Without
this change, the following -Werr output inhibits make from completing.
qemu/hw/intc/arm_gic_kvm.c:38:18: error: debug_gic_kvm defined but not used [-Werror=unused-const-variable=]
static const int debug_gic_kvm = 0;
^~~~~~~~~~~~~
cc1: all warnings being treated as errors
qemu/rules.mak:60: recipe for target 'hw/intc/arm_gic_kvm.o' failed
make[1]: *** [hw/intc/arm_gic_kvm.o] Error 1
Makefile:205: recipe for target 'subdir-aarch64-softmmu' failed
Signed-off-by: Christopher Covington <cov@codeaurora.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20161011163202.19720-1-cov@codeaurora.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Advertise gsi routing and set up irqchip routing entries for
GIC SPIs.
This is not mandated as long as MSI routing is not used
(because the kernel sets a default irqchip routing table).
However once MSI routing gets used (for VIRTIO-PCI vhost for
example), the first call to KVM_SET_GSI_ROUTING overrides the
kernel default irqchip table.
If no routing entry exists for the GSI, any IRQFD signaling for
this GSI will fail.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1474616617-366-2-git-send-email-eric.auger@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Commit 57cb38b included qapi/error.h into qemu/osdep.h to get the
Error typedef. Since then, we've moved to include qemu/osdep.h
everywhere. Its file comment explains: "To avoid getting into
possible circular include dependencies, this file should not include
any other QEMU headers, with the exceptions of config-host.h,
compiler.h, os-posix.h and os-win32.h, all of which are doing a
similar job to this file and are under similar constraints."
qapi/error.h doesn't do a similar job, and it doesn't adhere to
similar constraints: it includes qapi-types.h. That's in excess of
100KiB of crap most .c files don't actually need.
Add the typedef to qemu/typedefs.h, and include that instead of
qapi/error.h. Include qapi/error.h in .c files that need it and don't
get it now. Include qapi-types.h in qom/object.h for uint16List.
Update scripts/clean-includes accordingly. Update it further to match
reality: replace config.h by config-target.h, add sysemu/os-posix.h,
sysemu/os-win32.h. Update the list of includes in the qemu/osdep.h
comment quoted above similarly.
This reduces the number of objects depending on qapi/error.h from "all
of them" to less than a third. Unfortunately, the number depending on
qapi-types.h shrinks only a little. More work is needed for that one.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
[Fix compilation without the spice devel packages. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Clean up includes so that osdep.h is included first and headers
which it implies are not included manually.
This commit was created with scripts/clean-includes.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1453832250-766-13-git-send-email-peter.maydell@linaro.org
Currently, if the kernel does not have live migration API, the migration
will still be attempted, but vGIC save/restore functions will just not do
anything. This will result in a broken machine state.
This patch fixes the problem by adding migration blocker if kernel API is
not supported.
Signed-off-by: Pavel Fedin <p.fedin@samsung.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Some functions previously used only by vGICv2 are useful also for vGICv3
implementation. Untie them from GICState and make accessible from within
other modules:
- kvm_arm_gic_set_irq()
- kvm_gic_supports_attr() - moved to common code and renamed to
kvm_device_check_attr()
- kvm_gic_access() - turned into GIC-independent kvm_device_access().
Data pointer changed to void * because some GICv3 registers are
64-bit wide
Some of these changes are not used right now, but they will be helpful for
implementing live migration.
Actually kvm_dist_get() and kvm_dist_put() could also be made reusable, but
they would require two extra parameters (s->dev_fd and s->num_cpu) as well as
lots of typecasts of 's' to DeviceState * and back to GICState *. This makes
the code very ugly so i decided to stop at this point. I tried also an
approach with making a base class for all possible GICs, but it would contain
only three variables (dev_fd, cpu_num and irq_num), and accessing them through
the rest of the code would be again tedious (either ugly casts or qemu-style
separate object pointer). So i disliked it too.
Signed-off-by: Pavel Fedin <p.fedin@samsung.com>
Tested-by: Ashok kumar <ashoks@broadcom.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 2ef56d1dd64ffb75ed02a10dcdaf605e5b8ff4f8.1441784344.git.p.fedin@samsung.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Extracted duplicated initialization code from SW-emulated and KVM GIC
implementations and put into gic_init_irqs_and_mmio()
Signed-off-by: Pavel Fedin <p.fedin@samsung.com>
Message-id: 8ea5b2781ef39cb5989420987fc73c70e377687d.1438758065.git.p.fedin@samsung.com
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
The arm_gic_kvm now calls kvm_irqchip_set_qemuirq_gsi to build
the hash table storing qemu_irq/gsi mappings. From that point on
irqfd can be setup directly from the qemu_irq using
kvm_irqchip_add_irqfd_notifier.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Tested-by: Vikram Sethi <vikrams@codeaurora.org>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
GICs with Security Extensions restrict the non-secure view of the
interrupt priority and priority mask registers.
Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-11-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-15-git-send-email-greg.bellows@linaro.org
[PMM: minor code tweaks; fixed missing masking in gic_set_priority_mask
and gic_set_priority]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
ICCICR/GICC_CTLR is banked in GICv1 implementations with Security
Extensions or in GICv2 in independent from Security Extensions.
This makes it possible to enable forwarding of interrupts from
the CPU interfaces to the connected processors for Group0 and Group1.
We also allow to set additional bits like AckCtl and FIQEn by changing
the type from bool to uint32. Since the field does not only store the
enable bit anymore and since we are touching the vmstate, we use the
opportunity to rename the field to cpu_ctlr.
Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-9-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-9-git-send-email-greg.bellows@linaro.org
[PMM: rewrote to store state in a single uint32_t rather than
keeping the NS and S banked variants separate; this considerably
simplifies the get/set functions]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
ICDDCR/GICD_CTLR is banked if the GIC has the security extensions,
and the S (or only) copy has separate enable bits for Group0 and
Group1 enable if the GIC implements interrupt groups.
EnableGroup0 (Bit [1]) in GICv1 is architecturally IMPDEF. Since this
bit (Enable Non-secure) is present in the integrated GIC of the Cortex-A9
MPCore, we support this bit in our GICv1 implementation too.
Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-7-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-8-git-send-email-greg.bellows@linaro.org
[PMM: rewritten to store the state in a single s->ctlr uint32,
with the NS register handled as an alias of bit 1 in that value;
added vmstate version bump]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Now that the GIC base class has state fields for the GICD_IGROUPRn
registers, make kvm_arm_gic_get() and kvm_arm_gic_put() write and
read them. This allows us to remove the check that made us
fail migration if the guest had set any of the group register bits.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-6-git-send-email-peter.maydell@linaro.org
Add a QOM property which allows the GIC Security Extensions to be
enabled. These are an optional part of the GICv1 and GICv2 architecture.
This commit just adds the property and some sanity checks that it
is only enabled on GIC revisions that support it.
Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-3-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-5-git-send-email-greg.bellows@linaro.org
[PMM: changed property name, added checks that it isn't set for
older GIC revisions or if using the KVM VGIC; reworded commit message]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Create the outbound FIQ lines from the GIC to the CPUs; these are
used if the GIC has security extensions or grouping support.
Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-2-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-2-git-send-email-greg.bellows@linaro.org
[PMM: added FIQ lines to kvm-arm-gic so its interface is the same;
tweaked commit message]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
As there is logic to deal with the difference between edge and level
triggered interrupts in the kernel we must ensure it knows the
configuration of the IRQs before we restore the pending state.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Acked-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
This patch forces vgic initialization in the vgic realize function.
It uses a new group/attribute that allows such operation:
KVM_DEV_ARM_VGIC_GRP_CTRL/KVM_DEV_ARM_VGIC_CTRL_INIT
This earlier initialization allows, for example, to setup VFIO
signaling and irqfd after vgic initialization, on a reset notifier.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
Message-id: 1426094226-8515-1-git-send-email-eric.auger@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Newer kernels support a device attribute on the GIC which allows us to
tell it how many IRQs this GIC instance is configured with; use it, if
it exists.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Christoffer Dall <christoffer.dall@linaro.org>
Message-id: 1417718679-1071-1-git-send-email-peter.maydell@linaro.org
Using error_is_set(ERRP) to find out whether a function failed is
either wrong, fragile, or unnecessarily opaque. It's wrong when ERRP
may be null, because errors go undetected when it is. It's fragile
when proving ERRP non-null involves a non-local argument. Else, it's
unnecessarily opaque (see commit 84d18f0).
I guess the error_is_set(errp) in the DeviceClass realize() methods
are merely fragile right now, because I can't find a call chain that
passes a null errp argument.
Make the code more robust and more obviously correct: receive the
error in a local variable, then propagate it through the parameter.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Acked-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Andreas Färber <afaerber@suse.de>
Codespell found and fixed these new typos:
* doesnt -> doesn't
* funtion -> function
* perfomance -> performance
* remaing -> remaining
A coding style issue (line too long) was fixed manually.
Signed-off-by: Stefan Weil <sw@weilnetz.de>
Reviewed-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
Save and restore the ARM KVM VGIC state from the kernel. We rely on
QEMU to marshal the GICState data structure and therefore simply
synchronize the kernel state with the QEMU emulated state in both
directions.
We take some care on the restore path to check the VGIC has been
configured with enough IRQs and CPU interfaces that we can properly
restore the state, and for separate set/clear registers we first fully
clear the registers and then set the required bits.
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
Message-id: 1392687921-26921-1-git-send-email-christoffer.dall@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Support creating the ARM vgic device through the device control API and
setting the base address for the distributor and cpu interfaces in KVM
VMs using this API.
Because the older KVM_CREATE_IRQCHIP interface needs the irq chip to be
created prior to creating the VCPUs, we first test if we can use the
device control API in kvm_arch_irqchip_create (using the test flag from
the device control API). If we cannot, it means we have to fall back to
KVM_CREATE_IRQCHIP and use the older ioctl at this point in time. If
however, we can use the device control API, we don't do anything and
wait until the arm_gic_kvm driver initializes and let that use the
device control API.
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
Message-id: 1392687720-26806-5-git-send-email-christoffer.dall@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
device_add plugs devices into suitable bus. For "real" buses, that
actually connects the device. For sysbus, the connections need to be
made separately, and device_add can't do that. The device would be
left unconnected, and could not possibly work.
Quite a few, but not all sysbus devices already set
cannot_instantiate_with_device_add_yet in their class init function.
Set it in their abstract base's class init function
sysbus_device_class_init(), and remove the now redundant assignments
from device class init functions.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel.a@redhat.com>
Signed-off-by: Andreas Färber <afaerber@suse.de>
In an ideal world, machines can be built by wiring devices together
with configuration, not code. Unfortunately, that's not the world we
live in right now. We still have quite a few devices that need to be
wired up by code. If you try to device_add such a device, it'll fail
in sometimes mysterious ways. If you're lucky, you get an
unmysterious immediate crash.
To protect users from such badness, DeviceClass member no_user used to
make device models unavailable with -device / device_add, but that
regressed in commit 18b6dad. The device model is still omitted from
help, but is available anyway.
Attempts to fix the regression have been rejected with the argument
that the purpose of no_user isn't clear, and it's prone to misuse.
This commit clarifies no_user's purpose. Anthony suggested to rename
it cannot_instantiate_with_device_add_yet_due_to_internal_bugs, which
I shorten somewhat to keep checkpatch happy. While there, make it
bool.
Every use of cannot_instantiate_with_device_add_yet gets a FIXME
comment asking for rationale. The next few commits will clean them
all up, either by providing a rationale, or by getting rid of the use.
With that done, the regression fix is hopefully acceptable.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel.a@redhat.com>
Signed-off-by: Andreas Färber <afaerber@suse.de>
