| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
This field is set to APIC_DELIVERY_MODE_FIXED in all cases, and is read
exactly once. Fold the constant in uv_program_mmr() and drop the field.
Searching for the origin of the stale HyperV comment reveals commit
a31e58e129f7 ("x86/apic: Switch all APICs to Fixed delivery mode") which
notes:
As a consequence of this change, the apic::irq_delivery_mode field is
now pointless, but this needs to be cleaned up in a separate patch.
6 years is long enough for this technical debt to have survived.
[ bp: Fold in
https://lore.kernel.org/r/20231121123034.1442059-1-andrew.cooper3@citrix.com
]
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Steve Wahl <steve.wahl@hpe.com>
Link: https://lore.kernel.org/r/20231102-x86-apic-v1-1-bf049a2a0ed6@citrix.com
|
|
Prepare for the coming implementation by GCC and Clang of the __counted_by
attribute. Flexible array members annotated with __counted_by can have
their accesses bounds-checked at run-time checking via CONFIG_UBSAN_BOUNDS
(for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family
functions).
As found with Coccinelle[1], add __counted_by for struct hv_dr_state.
[1] https://github.com/kees/kernel-tools/blob/trunk/coccinelle/examples/counted_by.cocci
Link: https://lore.kernel.org/linux-pci/20230922175257.work.900-kees@kernel.org
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Krzysztof Wilczyński <kwilczynski@kernel.org>
Reviewed-by: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Acked-by: Wei Liu <wei.liu@kernel.org>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Dexuan Cui <decui@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Cc: "K. Y. Srinivasan" <kys@microsoft.com>
Cc: Krzysztof Wilczyński <kw@linux.com>
Cc: Lorenzo Pieralisi <lpieralisi@kernel.org>
Cc: Rob Herring <robh@kernel.org>
Cc: Wei Liu <wei.liu@kernel.org>
Cc: linux-hyperv@vger.kernel.org
Cc: linux-pci@vger.kernel.org
|
|
When a Linux VM with an assigned PCI device runs on Hyper-V, if the PCI
device driver is not loaded yet (i.e. MSI-X/MSI is not enabled on the
device yet), doing a VM hibernation triggers a panic in
hv_pci_restore_msi_msg() -> msi_lock_descs(&pdev->dev), because
pdev->dev.msi.data is still NULL.
Avoid the panic by checking if MSI-X/MSI is enabled.
Link: https://lore.kernel.org/r/20230816175939.21566-1-decui@microsoft.com
Fixes: dc2b453290c4 ("PCI: hv: Rework MSI handling")
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Reviewed-by: sathyanarayanan.kuppuswamy@linux.intel.com
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Cc: stable@vger.kernel.org
|
|
In the case of fast device addition/removal, it's possible that
hv_eject_device_work() can start to run before create_root_hv_pci_bus()
starts to run; as a result, the pci_get_domain_bus_and_slot() in
hv_eject_device_work() can return a 'pdev' of NULL, and
hv_eject_device_work() can remove the 'hpdev', and immediately send a
message PCI_EJECTION_COMPLETE to the host, and the host immediately
unassigns the PCI device from the guest; meanwhile,
create_root_hv_pci_bus() and the PCI device driver can be probing the
dead PCI device and reporting timeout errors.
Fix the issue by adding a per-bus mutex 'state_lock' and grabbing the
mutex before powering on the PCI bus in hv_pci_enter_d0(): when
hv_eject_device_work() starts to run, it's able to find the 'pdev' and call
pci_stop_and_remove_bus_device(pdev): if the PCI device driver has
loaded, the PCI device driver's probe() function is already called in
create_root_hv_pci_bus() -> pci_bus_add_devices(), and now
hv_eject_device_work() -> pci_stop_and_remove_bus_device() is able
to call the PCI device driver's remove() function and remove the device
reliably; if the PCI device driver hasn't loaded yet, the function call
hv_eject_device_work() -> pci_stop_and_remove_bus_device() is able to
remove the PCI device reliably and the PCI device driver's probe()
function won't be called; if the PCI device driver's probe() is already
running (e.g., systemd-udev is loading the PCI device driver), it must
be holding the per-device lock, and after the probe() finishes and releases
the lock, hv_eject_device_work() -> pci_stop_and_remove_bus_device() is
able to proceed to remove the device reliably.
Fixes: 4daace0d8ce8 ("PCI: hv: Add paravirtual PCI front-end for Microsoft Hyper-V VMs")
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Acked-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230615044451.5580-6-decui@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
This reverts commit d6af2ed29c7c1c311b96dac989dcb991e90ee195.
The statement "the hv_pci_bus_exit() call releases structures of all its
child devices" in commit d6af2ed29c7c is not true: in the path
hv_pci_probe() -> hv_pci_enter_d0() -> hv_pci_bus_exit(hdev, true): the
parameter "keep_devs" is true, so hv_pci_bus_exit() does *not* release the
child "struct hv_pci_dev *hpdev" that is created earlier in
pci_devices_present_work() -> new_pcichild_device().
The commit d6af2ed29c7c was originally made in July 2020 for RHEL 7.7,
where the old version of hv_pci_bus_exit() was used; when the commit was
rebased and merged into the upstream, people didn't notice that it's
not really necessary. The commit itself doesn't cause any issue, but it
makes hv_pci_probe() more complicated. Revert it to facilitate some
upcoming changes to hv_pci_probe().
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Acked-by: Wei Hu <weh@microsoft.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230615044451.5580-5-decui@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
The hpdev->state is never really useful. The only use in
hv_pci_eject_device() and hv_eject_device_work() is not really necessary.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Acked-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230615044451.5580-4-decui@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
When the host tries to remove a PCI device, the host first sends a
PCI_EJECT message to the guest, and the guest is supposed to gracefully
remove the PCI device and send a PCI_EJECTION_COMPLETE message to the host;
the host then sends a VMBus message CHANNELMSG_RESCIND_CHANNELOFFER to
the guest (when the guest receives this message, the device is already
unassigned from the guest) and the guest can do some final cleanup work;
if the guest fails to respond to the PCI_EJECT message within one minute,
the host sends the VMBus message CHANNELMSG_RESCIND_CHANNELOFFER and
removes the PCI device forcibly.
In the case of fast device addition/removal, it's possible that the PCI
device driver is still configuring MSI-X interrupts when the guest receives
the PCI_EJECT message; the channel callback calls hv_pci_eject_device(),
which sets hpdev->state to hv_pcichild_ejecting, and schedules a work
hv_eject_device_work(); if the PCI device driver is calling
pci_alloc_irq_vectors() -> ... -> hv_compose_msi_msg(), we can break the
while loop in hv_compose_msi_msg() due to the updated hpdev->state, and
leave data->chip_data with its default value of NULL; later, when the PCI
device driver calls request_irq() -> ... -> hv_irq_unmask(), the guest
crashes in hv_arch_irq_unmask() due to data->chip_data being NULL.
Fix the issue by not testing hpdev->state in the while loop: when the
guest receives PCI_EJECT, the device is still assigned to the guest, and
the guest has one minute to finish the device removal gracefully. We don't
really need to (and we should not) test hpdev->state in the loop.
Fixes: de0aa7b2f97d ("PCI: hv: Fix 2 hang issues in hv_compose_msi_msg()")
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230615044451.5580-3-decui@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
Since day 1 of the driver, there has been a race between
hv_pci_query_relations() and survey_child_resources(): during fast
device hotplug, hv_pci_query_relations() may error out due to
device-remove and the stack variable 'comp' is no longer valid;
however, pci_devices_present_work() -> survey_child_resources() ->
complete() may be running on another CPU and accessing the no-longer-valid
'comp'. Fix the race by flushing the workqueue before we exit from
hv_pci_query_relations().
Fixes: 4daace0d8ce8 ("PCI: hv: Add paravirtual PCI front-end for Microsoft Hyper-V VMs")
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Acked-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230615044451.5580-2-decui@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
4 commits are involved here:
A (2016): commit 0de8ce3ee8e3 ("PCI: hv: Allocate physically contiguous hypercall params buffer")
B (2017): commit be66b6736591 ("PCI: hv: Use page allocation for hbus structure")
C (2019): commit 877b911a5ba0 ("PCI: hv: Avoid a kmemleak false positive caused by the hbus buffer")
D (2018): commit 68bb7bfb7985 ("X86/Hyper-V: Enable IPI enlightenments")
Patch D introduced the per-CPU hypercall input page "hyperv_pcpu_input_arg"
in 2018. With patch D, we no longer need the per-Hyper-V-PCI-bus hypercall
input page "hbus->retarget_msi_interrupt_params" that was added in patch A,
and the issue addressed by patch B is no longer an issue, and we can also
get rid of patch C.
The change here is required for PCI device assignment to work for
Confidential VMs (CVMs) running without a paravisor, because otherwise we
would have to call set_memory_decrypted() for
"hbus->retarget_msi_interrupt_params" before calling the hypercall
HVCALL_RETARGET_INTERRUPT.
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Acked-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Link: https://lore.kernel.org/r/20230421013025.17152-1-decui@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
For PCI pass-thru devices in a Confidential VM, Hyper-V requires
that PCI config space be accessed via hypercalls. In normal VMs,
config space accesses are trapped to the Hyper-V host and emulated.
But in a confidential VM, the host can't access guest memory to
decode the instruction for emulation, so an explicit hypercall must
be used.
Add functions to make the new MMIO read and MMIO write hypercalls.
Update the PCI config space access functions to use the hypercalls
when such use is indicated by Hyper-V flags. Also, set the flag to
allow the Hyper-V PCI driver to be loaded and used in a Confidential
VM (a.k.a., "Isolation VM"). The driver has previously been hardened
against a malicious Hyper-V host[1].
[1] https://lore.kernel.org/all/20220511223207.3386-2-parri.andrea@gmail.com/
Co-developed-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: Michael Kelley <mikelley@microsoft.com>
Reviewed-by: Boqun Feng <boqun.feng@gmail.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Link: https://lore.kernel.org/r/1679838727-87310-13-git-send-email-mikelley@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
Since commit fc7a6209d571 ("bus: Make remove callback return
void") forces bus_type::remove be void-returned, it doesn't
make much sense for any bus based driver implementing remove
callbalk to return non-void to its caller.
As such, change the remove function for Hyper-V VMBus based
drivers to return void.
Signed-off-by: Dawei Li <set_pte_at@outlook.com>
Link: https://lore.kernel.org/r/TYCP286MB2323A93C55526E4DF239D3ACCAFA9@TYCP286MB2323.JPNP286.PROD.OUTLOOK.COM
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull irq updates from Thomas Gleixner:
"Updates for the interrupt core and driver subsystem:
The bulk is the rework of the MSI subsystem to support per device MSI
interrupt domains. This solves conceptual problems of the current
PCI/MSI design which are in the way of providing support for
PCI/MSI[-X] and the upcoming PCI/IMS mechanism on the same device.
IMS (Interrupt Message Store] is a new specification which allows
device manufactures to provide implementation defined storage for MSI
messages (as opposed to PCI/MSI and PCI/MSI-X that has a specified
message store which is uniform accross all devices). The PCI/MSI[-X]
uniformity allowed us to get away with "global" PCI/MSI domains.
IMS not only allows to overcome the size limitations of the MSI-X
table, but also gives the device manufacturer the freedom to store the
message in arbitrary places, even in host memory which is shared with
the device.
There have been several attempts to glue this into the current MSI
code, but after lengthy discussions it turned out that there is a
fundamental design problem in the current PCI/MSI-X implementation.
This needs some historical background.
When PCI/MSI[-X] support was added around 2003, interrupt management
was completely different from what we have today in the actively
developed architectures. Interrupt management was completely
architecture specific and while there were attempts to create common
infrastructure the commonalities were rudimentary and just providing
shared data structures and interfaces so that drivers could be written
in an architecture agnostic way.
The initial PCI/MSI[-X] support obviously plugged into this model
which resulted in some basic shared infrastructure in the PCI core
code for setting up MSI descriptors, which are a pure software
construct for holding data relevant for a particular MSI interrupt,
but the actual association to Linux interrupts was completely
architecture specific. This model is still supported today to keep
museum architectures and notorious stragglers alive.
In 2013 Intel tried to add support for hot-pluggable IO/APICs to the
kernel, which was creating yet another architecture specific mechanism
and resulted in an unholy mess on top of the existing horrors of x86
interrupt handling. The x86 interrupt management code was already an
incomprehensible maze of indirections between the CPU vector
management, interrupt remapping and the actual IO/APIC and PCI/MSI[-X]
implementation.
At roughly the same time ARM struggled with the ever growing SoC
specific extensions which were glued on top of the architected GIC
interrupt controller.
This resulted in a fundamental redesign of interrupt management and
provided the today prevailing concept of hierarchical interrupt
domains. This allowed to disentangle the interactions between x86
vector domain and interrupt remapping and also allowed ARM to handle
the zoo of SoC specific interrupt components in a sane way.
The concept of hierarchical interrupt domains aims to encapsulate the
functionality of particular IP blocks which are involved in interrupt
delivery so that they become extensible and pluggable. The X86
encapsulation looks like this:
|--- device 1
[Vector]---[Remapping]---[PCI/MSI]--|...
|--- device N
where the remapping domain is an optional component and in case that
it is not available the PCI/MSI[-X] domains have the vector domain as
their parent. This reduced the required interaction between the
domains pretty much to the initialization phase where it is obviously
required to establish the proper parent relation ship in the
components of the hierarchy.
While in most cases the model is strictly representing the chain of IP
blocks and abstracting them so they can be plugged together to form a
hierarchy, the design stopped short on PCI/MSI[-X]. Looking at the
hardware it's clear that the actual PCI/MSI[-X] interrupt controller
is not a global entity, but strict a per PCI device entity.
Here we took a short cut on the hierarchical model and went for the
easy solution of providing "global" PCI/MSI domains which was possible
because the PCI/MSI[-X] handling is uniform across the devices. This
also allowed to keep the existing PCI/MSI[-X] infrastructure mostly
unchanged which in turn made it simple to keep the existing
architecture specific management alive.
A similar problem was created in the ARM world with support for IP
block specific message storage. Instead of going all the way to stack
a IP block specific domain on top of the generic MSI domain this ended
in a construct which provides a "global" platform MSI domain which
allows overriding the irq_write_msi_msg() callback per allocation.
In course of the lengthy discussions we identified other abuse of the
MSI infrastructure in wireless drivers, NTB etc. where support for
implementation specific message storage was just mindlessly glued into
the existing infrastructure. Some of this just works by chance on
particular platforms but will fail in hard to diagnose ways when the
driver is used on platforms where the underlying MSI interrupt
management code does not expect the creative abuse.
Another shortcoming of today's PCI/MSI-X support is the inability to
allocate or free individual vectors after the initial enablement of
MSI-X. This results in an works by chance implementation of VFIO (PCI
pass-through) where interrupts on the host side are not set up upfront
to avoid resource exhaustion. They are expanded at run-time when the
guest actually tries to use them. The way how this is implemented is
that the host disables MSI-X and then re-enables it with a larger
number of vectors again. That works by chance because most device
drivers set up all interrupts before the device actually will utilize
them. But that's not universally true because some drivers allocate a
large enough number of vectors but do not utilize them until it's
actually required, e.g. for acceleration support. But at that point
other interrupts of the device might be in active use and the MSI-X
disable/enable dance can just result in losing interrupts and
therefore hard to diagnose subtle problems.
Last but not least the "global" PCI/MSI-X domain approach prevents to
utilize PCI/MSI[-X] and PCI/IMS on the same device due to the fact
that IMS is not longer providing a uniform storage and configuration
model.
The solution to this is to implement the missing step and switch from
global PCI/MSI domains to per device PCI/MSI domains. The resulting
hierarchy then looks like this:
|--- [PCI/MSI] device 1
[Vector]---[Remapping]---|...
|--- [PCI/MSI] device N
which in turn allows to provide support for multiple domains per
device:
|--- [PCI/MSI] device 1
|--- [PCI/IMS] device 1
[Vector]---[Remapping]---|...
|--- [PCI/MSI] device N
|--- [PCI/IMS] device N
This work converts the MSI and PCI/MSI core and the x86 interrupt
domains to the new model, provides new interfaces for post-enable
allocation/free of MSI-X interrupts and the base framework for
PCI/IMS. PCI/IMS has been verified with the work in progress IDXD
driver.
There is work in progress to convert ARM over which will replace the
platform MSI train-wreck. The cleanup of VFIO, NTB and other creative
"solutions" are in the works as well.
Drivers:
- Updates for the LoongArch interrupt chip drivers
- Support for MTK CIRQv2
- The usual small fixes and updates all over the place"
* tag 'irq-core-2022-12-10' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (134 commits)
irqchip/ti-sci-inta: Fix kernel doc
irqchip/gic-v2m: Mark a few functions __init
irqchip/gic-v2m: Include arm-gic-common.h
irqchip/irq-mvebu-icu: Fix works by chance pointer assignment
iommu/amd: Enable PCI/IMS
iommu/vt-d: Enable PCI/IMS
x86/apic/msi: Enable PCI/IMS
PCI/MSI: Provide pci_ims_alloc/free_irq()
PCI/MSI: Provide IMS (Interrupt Message Store) support
genirq/msi: Provide constants for PCI/IMS support
x86/apic/msi: Enable MSI_FLAG_PCI_MSIX_ALLOC_DYN
PCI/MSI: Provide post-enable dynamic allocation interfaces for MSI-X
PCI/MSI: Provide prepare_desc() MSI domain op
PCI/MSI: Split MSI-X descriptor setup
genirq/msi: Provide MSI_FLAG_MSIX_ALLOC_DYN
genirq/msi: Provide msi_domain_alloc_irq_at()
genirq/msi: Provide msi_domain_ops:: Prepare_desc()
genirq/msi: Provide msi_desc:: Msi_data
genirq/msi: Provide struct msi_map
x86/apic/msi: Remove arch_create_remap_msi_irq_domain()
...
|
|
The function hv_set_affinity was removed in commit 831c1ae7 ("PCI: hv:
Make the code arch neutral by adding arch specific interfaces").
Signed-off-by: Olaf Hering <olaf@aepfle.de>
Link: https://lore.kernel.org/r/20221107171831.25283-1-olaf@aepfle.de
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
Now that the PCI/MSI core code does early checking for multi-MSI support
X86_IRQ_ALLOC_CONTIGUOUS_VECTORS is not required anymore.
Remove the flag and rely on MSI_FLAG_MULTI_PCI_MSI.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20221111122015.865042356@linutronix.de
|
|
Jeffrey added Multi-MSI support to the pci-hyperv driver by the 4 patches:
08e61e861a0e ("PCI: hv: Fix multi-MSI to allow more than one MSI vector")
455880dfe292 ("PCI: hv: Fix hv_arch_irq_unmask() for multi-MSI")
b4b77778ecc5 ("PCI: hv: Reuse existing IRTE allocation in compose_msi_msg()")
a2bad844a67b ("PCI: hv: Fix interrupt mapping for multi-MSI")
It turns out that the third patch (b4b77778ecc5) causes a performance
regression because all the interrupts now happen on 1 physical CPU (or two
pCPUs, if one pCPU doesn't have enough vectors). When a guest has many PCI
devices, it may suffer from soft lockups if the workload is heavy, e.g.,
see https://lwn.net/ml/linux-kernel/20220804025104.15673-1-decui@microsoft.com/
Commit b4b77778ecc5 itself is good. The real issue is that the hypercall in
hv_irq_unmask() -> hv_arch_irq_unmask() ->
hv_do_hypercall(HVCALL_RETARGET_INTERRUPT...) only changes the target
virtual CPU rather than physical CPU; with b4b77778ecc5, the pCPU is
determined only once in hv_compose_msi_msg() where only vCPU0 is specified;
consequently the hypervisor only uses 1 target pCPU for all the interrupts.
Note: before b4b77778ecc5, the pCPU is determined twice, and when the pCPU
is determined the second time, the vCPU in the effective affinity mask is
used (i.e., it isn't always vCPU0), so the hypervisor chooses different
pCPU for each interrupt.
The hypercall will be fixed in future to update the pCPU as well, but
that will take quite a while, so let's restore the old behavior in
hv_compose_msi_msg(), i.e., don't reuse the existing IRTE allocation for
single-MSI and MSI-X; for multi-MSI, we choose the vCPU in a round-robin
manner for each PCI device, so the interrupts of different devices can
happen on different pCPUs, though the interrupts of each device happen on
some single pCPU.
The hypercall fix may not be backported to all old versions of Hyper-V, so
we want to have this guest side change forever (or at least till we're sure
the old affected versions of Hyper-V are no longer supported).
Fixes: b4b77778ecc5 ("PCI: hv: Reuse existing IRTE allocation in compose_msi_msg()")
Co-developed-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Co-developed-by: Carl Vanderlip <quic_carlv@quicinc.com>
Signed-off-by: Carl Vanderlip <quic_carlv@quicinc.com>
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/20221104222953.11356-1-decui@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
The local variable 'vector' must be u32 rather than u8: see the
struct hv_msi_desc3.
'vector_count' should be u16 rather than u8: see struct hv_msi_desc,
hv_msi_desc2 and hv_msi_desc3.
Fixes: a2bad844a67b ("PCI: hv: Fix interrupt mapping for multi-MSI")
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Cc: Jeffrey Hugo <quic_jhugo@quicinc.com>
Cc: Carl Vanderlip <quic_carlv@quicinc.com>
Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Link: https://lore.kernel.org/r/20221027205256.17678-1-decui@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
The cpumask that is passed to this function ultimately comes from
irq_data_get_effective_affinity_mask(), which was recently changed to
return a const cpumask pointer. The first level of functions handling
the affinity mask were updated, but not this helper function.
Fixes: 4d0b8298818b ("genirq: Return a const cpumask from irq_data_get_affinity_mask")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Samuel Holland <samuel@sholland.org>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20220708004931.1672-1-samuel@sholland.org
|
|
Now that the irq_data_update_affinity helper exists, enforce its use
by returning a a const cpumask from irq_data_get_affinity_mask.
Since the previous commit already updated places that needed to call
irq_data_update_affinity, this commit updates the remaining code that
either did not modify the cpumask or immediately passed the modified
mask to irq_set_affinity.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20220701200056.46555-8-samuel@sholland.org
|
|
[ Similarly to commit a765ed47e4516 ("PCI: hv: Fix synchronization
between channel callback and hv_compose_msi_msg()"): ]
The (on-stack) teardown packet becomes invalid once the completion
timeout in hv_pci_bus_exit() has expired and hv_pci_bus_exit() has
returned. Prevent the channel callback from accessing the invalid
packet by removing the ID associated to such packet from the VMbus
requestor in hv_pci_bus_exit().
Signed-off-by: Andrea Parri (Microsoft) <parri.andrea@gmail.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Acked-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Link: https://lore.kernel.org/r/20220511223207.3386-3-parri.andrea@gmail.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
For additional robustness in the face of Hyper-V errors or malicious
behavior, validate all values that originate from packets that Hyper-V
has sent to the guest in the host-to-guest ring buffer. Ensure that
invalid values cannot cause data being copied out of the bounds of the
source buffer in hv_pci_onchannelcallback().
While at it, remove a redundant validation in hv_pci_generic_compl():
hv_pci_onchannelcallback() already ensures that all processed incoming
packets are "at least as large as [in fact larger than] a response".
Signed-off-by: Andrea Parri (Microsoft) <parri.andrea@gmail.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Acked-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Link: https://lore.kernel.org/r/20220511223207.3386-2-parri.andrea@gmail.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
According to Dexuan, the hypervisor folks beleive that multi-msi
allocations are not correct. compose_msi_msg() will allocate multi-msi
one by one. However, multi-msi is a block of related MSIs, with alignment
requirements. In order for the hypervisor to allocate properly aligned
and consecutive entries in the IOMMU Interrupt Remapping Table, there
should be a single mapping request that requests all of the multi-msi
vectors in one shot.
Dexuan suggests detecting the multi-msi case and composing a single
request related to the first MSI. Then for the other MSIs in the same
block, use the cached information. This appears to be viable, so do it.
Suggested-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Dexuan Cui <decui@microsoft.com>
Tested-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/1652282599-21643-1-git-send-email-quic_jhugo@quicinc.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
Currently if compose_msi_msg() is called multiple times, it will free any
previous IRTE allocation, and generate a new allocation. While nothing
prevents this from occurring, it is extraneous when Linux could just reuse
the existing allocation and avoid a bunch of overhead.
However, when future IRTE allocations operate on blocks of MSIs instead of
a single line, freeing the allocation will impact all of the lines. This
could cause an issue where an allocation of N MSIs occurs, then some of
the lines are retargeted, and finally the allocation is freed/reallocated.
The freeing of the allocation removes all of the configuration for the
entire block, which requires all the lines to be retargeted, which might
not happen since some lines might already be unmasked/active.
Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Dexuan Cui <decui@microsoft.com>
Tested-by: Dexuan Cui <decui@microsoft.com>
Tested-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/1652282582-21595-1-git-send-email-quic_jhugo@quicinc.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
Currently when the pci-hyperv driver finishes probing and initializing the
PCI device, it sets the PCI_COMMAND_MEMORY bit; later when the PCI device
is registered to the core PCI subsystem, the core PCI driver's BAR detection
and initialization code toggles the bit multiple times, and each toggling of
the bit causes the hypervisor to unmap/map the virtual BARs from/to the
physical BARs, which can be slow if the BAR sizes are huge, e.g., a Linux VM
with 14 GPU devices has to spend more than 3 minutes on BAR detection and
initialization, causing a long boot time.
Reduce the boot time by not setting the PCI_COMMAND_MEMORY bit when we
register the PCI device (there is no need to have it set in the first place).
The bit stays off till the PCI device driver calls pci_enable_device().
With this change, the boot time of such a 14-GPU VM is reduced by almost
3 minutes.
Link: https://lore.kernel.org/lkml/20220419220007.26550-1-decui@microsoft.com/
Tested-by: Boqun Feng (Microsoft) <boqun.feng@gmail.com>
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Acked-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Cc: Jake Oshins <jakeo@microsoft.com>
Link: https://lore.kernel.org/r/20220502074255.16901-1-decui@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
In the multi-MSI case, hv_arch_irq_unmask() will only operate on the first
MSI of the N allocated. This is because only the first msi_desc is cached
and it is shared by all the MSIs of the multi-MSI block. This means that
hv_arch_irq_unmask() gets the correct address, but the wrong data (always
0).
This can break MSIs.
Lets assume MSI0 is vector 34 on CPU0, and MSI1 is vector 33 on CPU0.
hv_arch_irq_unmask() is called on MSI0. It uses a hypercall to configure
the MSI address and data (0) to vector 34 of CPU0. This is correct. Then
hv_arch_irq_unmask is called on MSI1. It uses another hypercall to
configure the MSI address and data (0) to vector 33 of CPU0. This is
wrong, and results in both MSI0 and MSI1 being routed to vector 33. Linux
will observe extra instances of MSI1 and no instances of MSI0 despite the
endpoint device behaving correctly.
For the multi-MSI case, we need unique address and data info for each MSI,
but the cached msi_desc does not provide that. However, that information
can be gotten from the int_desc cached in the chip_data by
compose_msi_msg(). Fix the multi-MSI case to use that cached information
instead. Since hv_set_msi_entry_from_desc() is no longer applicable,
remove it.
Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/1651068453-29588-1-git-send-email-quic_jhugo@quicinc.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
Dexuan wrote:
"[...] when we disable AccelNet, the host PCI VSP driver sends a
PCI_EJECT message first, and the channel callback may set
hpdev->state to hv_pcichild_ejecting on a different CPU. This can
cause hv_compose_msi_msg() to exit from the loop and 'return', and
the on-stack variable 'ctxt' is invalid. Now, if the response
message from the host arrives, the channel callback will try to
access the invalid 'ctxt' variable, and this may cause a crash."
Schematically:
Hyper-V sends PCI_EJECT msg
hv_pci_onchannelcallback()
state = hv_pcichild_ejecting
hv_compose_msi_msg()
alloc and init comp_pkt
state == hv_pcichild_ejecting
Hyper-V sends VM_PKT_COMP msg
hv_pci_onchannelcallback()
retrieve address of comp_pkt
'free' comp_pkt and return
comp_pkt->completion_func()
Dexuan also showed how the crash can be triggered after introducing
suitable delays in the driver code, thus validating the 'assumption'
that the host can still normally respond to the guest's compose_msi
request after the host has started to eject the PCI device.
Fix the synchronization by leveraging the requestor lock as follows:
- Before 'return'-ing in hv_compose_msi_msg(), remove the ID (while
holding the requestor lock) associated to the completion packet.
- Retrieve the address *and call ->completion_func() within a same
(requestor) critical section in hv_pci_onchannelcallback().
Reported-by: Wei Hu <weh@microsoft.com>
Reported-by: Dexuan Cui <decui@microsoft.com>
Suggested-by: Michael Kelley <mikelley@microsoft.com>
Signed-off-by: Andrea Parri (Microsoft) <parri.andrea@gmail.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/20220419122325.10078-7-parri.andrea@gmail.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
Currently, pointers to guest memory are passed to Hyper-V as transaction
IDs in hv_pci. In the face of errors or malicious behavior in Hyper-V,
hv_pci should not expose or trust the transaction IDs returned by
Hyper-V to be valid guest memory addresses. Instead, use small integers
generated by vmbus_requestor as request (transaction) IDs.
Suggested-by: Michael Kelley <mikelley@microsoft.com>
Signed-off-by: Andrea Parri (Microsoft) <parri.andrea@gmail.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/20220419122325.10078-3-parri.andrea@gmail.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
If the allocation of multiple MSI vectors for multi-MSI fails in the core
PCI framework, the framework will retry the allocation as a single MSI
vector, assuming that meets the min_vecs specified by the requesting
driver.
Hyper-V advertises that multi-MSI is supported, but reuses the VECTOR
domain to implement that for x86. The VECTOR domain does not support
multi-MSI, so the alloc will always fail and fallback to a single MSI
allocation.
In short, Hyper-V advertises a capability it does not implement.
Hyper-V can support multi-MSI because it coordinates with the hypervisor
to map the MSIs in the IOMMU's interrupt remapper, which is something the
VECTOR domain does not have. Therefore the fix is simple - copy what the
x86 IOMMU drivers (AMD/Intel-IR) do by removing
X86_IRQ_ALLOC_CONTIGUOUS_VECTORS after calling the VECTOR domain's
pci_msi_prepare().
Fixes: 4daace0d8ce8 ("PCI: hv: Add paravirtual PCI front-end for Microsoft Hyper-V VMs")
Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Dexuan Cui <decui@microsoft.com>
Link: https://lore.kernel.org/r/1649856981-14649-1-git-send-email-quic_jhugo@quicinc.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux
Pull hyperv fixes from Wei Liu:
- Correctly propagate coherence information for VMbus devices (Michael
Kelley)
- Disable balloon and memory hot-add on ARM64 temporarily (Boqun Feng)
- Use barrier to prevent reording when reading ring buffer (Michael
Kelley)
- Use virt_store_mb in favour of smp_store_mb (Andrea Parri)
- Fix VMbus device object initialization (Andrea Parri)
- Deactivate sysctl_record_panic_msg on isolated guest (Andrea Parri)
- Fix a crash when unloading VMbus module (Guilherme G. Piccoli)
* tag 'hyperv-fixes-signed-20220407' of git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux:
Drivers: hv: vmbus: Replace smp_store_mb() with virt_store_mb()
Drivers: hv: balloon: Disable balloon and hot-add accordingly
Drivers: hv: balloon: Support status report for larger page sizes
Drivers: hv: vmbus: Prevent load re-ordering when reading ring buffer
PCI: hv: Propagate coherence from VMbus device to PCI device
Drivers: hv: vmbus: Propagate VMbus coherence to each VMbus device
Drivers: hv: vmbus: Fix potential crash on module unload
Drivers: hv: vmbus: Fix initialization of device object in vmbus_device_register()
Drivers: hv: vmbus: Deactivate sysctl_record_panic_msg by default in isolated guests
|
|
Fix the following build error:
drivers/pci/controller/pci-hyperv.c:769:13: error: ‘hv_set_msi_entry_from_desc’ defined but not used [-Werror=unused-function]
769 | static void hv_set_msi_entry_from_desc(union hv_msi_entry *msi_entry,
The arm64 implementation of hv_set_msi_entry_from_desc() is not used after
d06957d7a692 ("PCI: hv: Avoid the retarget interrupt hypercall in
irq_unmask() on ARM64"), so remove it.
Fixes: d06957d7a692 ("PCI: hv: Avoid the retarget interrupt hypercall in irq_unmask() on ARM64")
Link: https://lore.kernel.org/r/20220317085130.36388-1-yuehaibing@huawei.com
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
Acked-by: Boqun Feng <boqun.feng@gmail.com>
|
|
PCI pass-thru devices in a Hyper-V VM are represented as a VMBus
device and as a PCI device. The coherence of the VMbus device is
set based on the VMbus node in ACPI, but the PCI device has no
ACPI node and defaults to not hardware coherent. This results
in extra software coherence management overhead on ARM64 when
devices are hardware coherent.
Fix this by setting up the PCI host bus so that normal
PCI mechanisms will propagate the coherence of the VMbus
device to the PCI device. There's no effect on x86/x64 where
devices are always hardware coherent.
Signed-off-by: Michael Kelley <mikelley@microsoft.com>
Acked-by: Boqun Feng <boqun.feng@gmail.com>
Acked-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/1648138492-2191-3-git-send-email-mikelley@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
Pull pci updates from Bjorn Helgaas:
"Enumeration:
- Move the VGA arbiter from drivers/gpu to drivers/pci because it's
PCI-specific, not GPU-specific (Bjorn Helgaas)
- Select the default VGA device consistently whether it's enumerated
before or after VGA arbiter init, which fixes arches that enumerate
PCI devices late (Huacai Chen)
Resource management:
- Support BAR sizes up to 8TB (Dongdong Liu)
PCIe native device hotplug:
- Fix "Command Completed" tracking to avoid spurious timouts when
powering off empty slots (Liguang Zhang)
- Quirk Qualcomm devices that don't implement Command Completed
correctly, again to avoid spurious timeouts (Manivannan Sadhasivam)
Peer-to-peer DMA:
- Add Intel 3rd Gen Intel Xeon Scalable Processors to whitelist
(Michael J. Ruhl)
APM X-Gene PCIe controller driver:
- Revert generic DT parsing changes that broke some machines in the
field (Marc Zyngier)
Freescale i.MX6 PCIe controller driver:
- Allow controller probe to succeed even when no devices currently
present to allow hot-add later (Fabio Estevam)
- Enable power management on i.MX6QP (Richard Zhu)
- Assert CLKREQ# on i.MX8MM so enumeration doesn't hang when no
device is connected (Richard Zhu)
Marvell Aardvark PCIe controller driver:
- Fix MSI and MSI-X support (Marek Behún, Pali Rohár)
- Add support for ERR and PME interrupts (Pali Rohár)
Marvell MVEBU PCIe controller driver:
- Add DT binding and support for "num-lanes" (Pali Rohár)
- Add support for INTx interrupts (Pali Rohár)
Microsoft Hyper-V host bridge driver:
- Avoid unnecessary hypercalls when unmasking IRQs on ARM64 (Boqun
Feng)
Qualcomm PCIe controller driver:
- Add SM8450 DT binding and driver support (Dmitry Baryshkov)
Renesas R-Car PCIe controller driver:
- Help the controller get to the L1 state since the hardware can't do
it on its own (Marek Vasut)
- Return PCI_ERROR_RESPONSE (~0) for reads that fail on PCIe (Marek
Vasut)
SiFive FU740 PCIe controller driver:
- Drop redundant '-gpios' from DT GPIO lookup (Ben Dooks)
- Force 2.5GT/s for initial device probe (Ben Dooks)
Socionext UniPhier Pro5 controller driver:
- Add NX1 DT binding and driver support (Kunihiko Hayashi)
Synopsys DesignWare PCIe controller driver:
- Restore MSI configuration so MSI works after resume (Jisheng
Zhang)"
* tag 'pci-v5.18-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci: (94 commits)
x86/PCI: Add #includes to asm/pci_x86.h
PCI: ibmphp: Remove unused assignments
PCI: cpqphp: Remove unused assignments
PCI: fu740: Remove unused assignments
PCI: kirin: Remove unused assignments
PCI: Remove unused assignments
PCI: Declare pci_filp_private only when HAVE_PCI_MMAP
PCI: Avoid broken MSI on SB600 USB devices
PCI: fu740: Force 2.5GT/s for initial device probe
PCI: xgene: Revert "PCI: xgene: Fix IB window setup"
PCI: xgene: Revert "PCI: xgene: Use inbound resources for setup"
PCI: imx6: Assert i.MX8MM CLKREQ# even if no device present
PCI: imx6: Invoke the PHY exit function after PHY power off
PCI: rcar: Use PCI_SET_ERROR_RESPONSE after read which triggered an exception
PCI: rcar: Finish transition to L1 state in rcar_pcie_config_access()
PCI: dwc: Restore MSI Receiver mask during resume
PCI: fu740: Drop redundant '-gpios' from DT GPIO lookup
PCI/VGA: Replace full MIT license text with SPDX identifier
PCI/VGA: Use unsigned format string to print lock counts
PCI/VGA: Log bridge control messages when adding devices
...
|
|
On ARM64 Hyper-V guests, SPIs are used for the interrupts of virtual PCI
devices, and SPIs can be managed directly via GICD registers. Therefore
the retarget interrupt hypercall is not needed on ARM64.
An arch-specific interface hv_arch_irq_unmask() is introduced to handle
the architecture level differences on this. For x86, the behavior
remains unchanged, while for ARM64 no hypercall is invoked when
unmasking an irq for virtual PCI devices.
Link: https://lore.kernel.org/r/20220217034525.1687678-1-boqun.feng@gmail.com
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
|
