Alex Williamson [Fri, 8 Mar 2024 23:05:28 +0000 (16:05 -0700)]
vfio/fsl-mc: Block calling interrupt handler without trigger
The eventfd_ctx trigger pointer of the vfio_fsl_mc_irq object is
initially NULL and may become NULL if the user sets the trigger
eventfd to -1. The interrupt handler itself is guaranteed that
trigger is always valid between request_irq() and free_irq(), but
the loopback testing mechanisms to invoke the handler function
need to test the trigger. The triggering and setting ioctl paths
both make use of igate and are therefore mutually exclusive.
The vfio-fsl-mc driver does not make use of irqfds, nor does it
support any sort of masking operations, therefore unlike vfio-pci
and vfio-platform, the flow can remain essentially unchanged.
Cc: Diana Craciun <diana.craciun@oss.nxp.com> Cc: <stable@vger.kernel.org> Fixes: cc0ee20bd969 ("vfio/fsl-mc: trigger an interrupt via eventfd") Reviewed-by: Kevin Tian <kevin.tian@intel.com> Reviewed-by: Eric Auger <eric.auger@redhat.com> Link: https://lore.kernel.org/r/20240308230557.805580-8-alex.williamson@redhat.com Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Alex Williamson [Fri, 8 Mar 2024 23:05:27 +0000 (16:05 -0700)]
vfio/platform: Create persistent IRQ handlers
The vfio-platform SET_IRQS ioctl currently allows loopback triggering of
an interrupt before a signaling eventfd has been configured by the user,
which thereby allows a NULL pointer dereference.
Rather than register the IRQ relative to a valid trigger, register all
IRQs in a disabled state in the device open path. This allows mask
operations on the IRQ to nest within the overall enable state governed
by a valid eventfd signal. This decouples @masked, protected by the
@locked spinlock from @trigger, protected via the @igate mutex.
In doing so, it's guaranteed that changes to @trigger cannot race the
IRQ handlers because the IRQ handler is synchronously disabled before
modifying the trigger, and loopback triggering of the IRQ via ioctl is
safe due to serialization with trigger changes via igate.
For compatibility, request_irq() failures are maintained to be local to
the SET_IRQS ioctl rather than a fatal error in the open device path.
This allows, for example, a userspace driver with polling mode support
to continue to work regardless of moving the request_irq() call site.
This necessarily blocks all SET_IRQS access to the failed index.
Cc: Eric Auger <eric.auger@redhat.com> Cc: <stable@vger.kernel.org> Fixes: 57f972e2b341 ("vfio/platform: trigger an interrupt via eventfd") Reviewed-by: Kevin Tian <kevin.tian@intel.com> Reviewed-by: Eric Auger <eric.auger@redhat.com> Link: https://lore.kernel.org/r/20240308230557.805580-7-alex.williamson@redhat.com Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Alex Williamson [Fri, 8 Mar 2024 23:05:26 +0000 (16:05 -0700)]
vfio/platform: Disable virqfds on cleanup
irqfds for mask and unmask that are not specifically disabled by the
user are leaked. Remove any irqfds during cleanup
Cc: Eric Auger <eric.auger@redhat.com> Cc: <stable@vger.kernel.org> Fixes: a7fa7c77cf15 ("vfio/platform: implement IRQ masking/unmasking via an eventfd") Reviewed-by: Kevin Tian <kevin.tian@intel.com> Reviewed-by: Eric Auger <eric.auger@redhat.com> Link: https://lore.kernel.org/r/20240308230557.805580-6-alex.williamson@redhat.com Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Alex Williamson [Fri, 8 Mar 2024 23:05:25 +0000 (16:05 -0700)]
vfio/pci: Create persistent INTx handler
A vulnerability exists where the eventfd for INTx signaling can be
deconfigured, which unregisters the IRQ handler but still allows
eventfds to be signaled with a NULL context through the SET_IRQS ioctl
or through unmask irqfd if the device interrupt is pending.
Ideally this could be solved with some additional locking; the igate
mutex serializes the ioctl and config space accesses, and the interrupt
handler is unregistered relative to the trigger, but the irqfd path
runs asynchronous to those. The igate mutex cannot be acquired from the
atomic context of the eventfd wake function. Disabling the irqfd
relative to the eventfd registration is potentially incompatible with
existing userspace.
As a result, the solution implemented here moves configuration of the
INTx interrupt handler to track the lifetime of the INTx context object
and irq_type configuration, rather than registration of a particular
trigger eventfd. Synchronization is added between the ioctl path and
eventfd_signal() wrapper such that the eventfd trigger can be
dynamically updated relative to in-flight interrupts or irqfd callbacks.
Alex Williamson [Fri, 8 Mar 2024 23:05:24 +0000 (16:05 -0700)]
vfio: Introduce interface to flush virqfd inject workqueue
In order to synchronize changes that can affect the thread callback,
introduce an interface to force a flush of the inject workqueue. The
irqfd pointer is only valid under spinlock, but the workqueue cannot
be flushed under spinlock. Therefore the flush work for the irqfd is
queued under spinlock. The vfio_irqfd_cleanup_wq workqueue is re-used
for queuing this work such that flushing the workqueue is also ordered
relative to shutdown.
Alex Williamson [Fri, 8 Mar 2024 23:05:23 +0000 (16:05 -0700)]
vfio/pci: Lock external INTx masking ops
Mask operations through config space changes to DisINTx may race INTx
configuration changes via ioctl. Create wrappers that add locking for
paths outside of the core interrupt code.
In particular, irq_type is updated holding igate, therefore testing
is_intx() requires holding igate. For example clearing DisINTx from
config space can otherwise race changes of the interrupt configuration.
This aligns interfaces which may trigger the INTx eventfd into two
camps, one side serialized by igate and the other only enabled while
INTx is configured. A subsequent patch introduces synchronization for
the latter flows.
Alex Williamson [Fri, 8 Mar 2024 23:05:22 +0000 (16:05 -0700)]
vfio/pci: Disable auto-enable of exclusive INTx IRQ
Currently for devices requiring masking at the irqchip for INTx, ie.
devices without DisINTx support, the IRQ is enabled in request_irq()
and subsequently disabled as necessary to align with the masked status
flag. This presents a window where the interrupt could fire between
these events, resulting in the IRQ incrementing the disable depth twice.
This would be unrecoverable for a user since the masked flag prevents
nested enables through vfio.
Instead, invert the logic using IRQF_NO_AUTOEN such that exclusive INTx
is never auto-enabled, then unmask as required.
Cc: <stable@vger.kernel.org> Fixes: 89e1f7d4c66d ("vfio: Add PCI device driver") Reviewed-by: Kevin Tian <kevin.tian@intel.com> Reviewed-by: Eric Auger <eric.auger@redhat.com> Link: https://lore.kernel.org/r/20240308230557.805580-2-alex.williamson@redhat.com Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Brett Creeley [Fri, 8 Mar 2024 18:21:49 +0000 (10:21 -0800)]
vfio/pds: Refactor/simplify reset logic
The current logic for handling resets is more complicated than it needs
to be. The deferred_reset flag is used to indicate a reset is needed
and the deferred_reset_state is the requested, post-reset, state.
Also, the deferred_reset logic was added to vfio migration drivers to
prevent a circular locking dependency with respect to mm_lock and state
mutex. This is mainly because of the copy_to/from_user() functions(which
takes mm_lock) invoked under state mutex.
Remove all of the deferred reset logic and just pass the requested
next state to pds_vfio_reset() so it can be used for VMM and DSC
initiated resets.
This removes the need for pds_vfio_state_mutex_lock(), so remove that
and replace its use with a simple mutex_unlock().
Also, remove the reset_mutex as it's no longer needed since the
state_mutex can be the driver's primary protector.
Suggested-by: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com> Reviewed-by: Shannon Nelson <shannon.nelson@amd.com> Signed-off-by: Brett Creeley <brett.creeley@amd.com> Link: https://lore.kernel.org/r/20240308182149.22036-3-brett.creeley@amd.com Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Brett Creeley [Fri, 8 Mar 2024 18:21:48 +0000 (10:21 -0800)]
vfio/pds: Make sure migration file isn't accessed after reset
It's possible the migration file is accessed after reset when it has
been cleaned up, especially when it's initiated by the device. This is
because the driver doesn't rip out the filep when cleaning up it only
frees the related page structures and sets its local struct
pds_vfio_lm_file pointer to NULL. This can cause a NULL pointer
dereference, which is shown in the example below during a restore after
a device initiated reset:
Add a disabled flag to the driver's struct pds_vfio_lm_file that gets
set during cleanup. Then make sure to check the flag when the migration
file is accessed via its file_operations. By default this flag will be
false as the memory for struct pds_vfio_lm_file is kzalloc'd, which means
the struct pds_vfio_lm_file is enabled and accessible. Also, since the
file_operations and driver's migration file cleanup happen under the
protection of the same pds_vfio_lm_file.lock, using this flag is thread
safe.
Fixes: 8512ed256334 ("vfio/pds: Always clear the save/restore FDs on reset") Reviewed-by: Shannon Nelson <shannon.nelson@amd.com> Signed-off-by: Brett Creeley <brett.creeley@amd.com> Link: https://lore.kernel.org/r/20240308182149.22036-2-brett.creeley@amd.com Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Uwe Kleine-König [Fri, 8 Mar 2024 08:51:19 +0000 (09:51 +0100)]
vfio/platform: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
Ricardo B. Marliere [Fri, 1 Mar 2024 17:51:48 +0000 (14:51 -0300)]
vfio/mbochs: make mbochs_class constant
Since commit 43a7206b0963 ("driver core: class: make class_register() take
a const *"), the driver core allows for struct class to be in read-only
memory, so move the mbochs_class structure to be declared at build time
placing it into read-only memory, instead of having to be dynamically
allocated at boot time.
Ricardo B. Marliere [Fri, 1 Mar 2024 17:51:47 +0000 (14:51 -0300)]
vfio/mdpy: make mdpy_class constant
Since commit 43a7206b0963 ("driver core: class: make class_register() take
a const *"), the driver core allows for struct class to be in read-only
memory, so move the mdpy_class structure to be declared at build time
placing it into read-only memory, instead of having to be dynamically
allocated at boot time.
Shameer Kolothum [Thu, 29 Feb 2024 09:11:52 +0000 (09:11 +0000)]
hisi_acc_vfio_pci: Remove the deferred_reset logic
The deferred_reset logic was added to vfio migration drivers to prevent
a circular locking dependency with respect to mm_lock and state mutex.
This is mainly because of the copy_to/from_user() functions(which takes
mm_lock) invoked under state mutex. But for HiSilicon driver, the only
place where we now hold the state mutex for copy_to_user is during the
PRE_COPY IOCTL. So for pre_copy, release the lock as soon as we have
updated the data and perform copy_to_user without state mutex. By this,
we can get rid of the deferred_reset logic.
This was a heinous workaround and it turns out it's been fixed in mm
twice since it was introduced. Most recently, commit c8070b787519
("mm: Don't pin ZERO_PAGE in pin_user_pages()") would have prevented
running up the zeropage refcount, but even before that commit 84209e87c696 ("mm/gup: reliable R/O long-term pinning in COW mappings")
avoids the vfio use case from pinning the zeropage at all, instead
replacing it with exclusive anonymous pages.
Ankit Agrawal [Thu, 29 Feb 2024 19:39:34 +0000 (19:39 +0000)]
vfio/nvgrace-gpu: Convey kvm to map device memory region as noncached
The NVIDIA Grace Hopper GPUs have device memory that is supposed to be
used as a regular RAM. It is accessible through CPU-GPU chip-to-chip
cache coherent interconnect and is present in the system physical
address space. The device memory is split into two regions - termed
as usemem and resmem - in the system physical address space,
with each region mapped and exposed to the VM as a separate fake
device BAR [1].
Owing to a hardware defect for Multi-Instance GPU (MIG) feature [2],
there is a requirement - as a workaround - for the resmem BAR to
display uncached memory characteristics. Based on [3], on system with
FWB enabled such as Grace Hopper, the requisite properties
(uncached, unaligned access) can be achieved through a VM mapping (S1)
of NORMAL_NC and host mapping (S2) of MT_S2_FWB_NORMAL_NC.
KVM currently maps the MMIO region in S2 as MT_S2_FWB_DEVICE_nGnRE by
default. The fake device BARs thus displays DEVICE_nGnRE behavior in the
VM.
The following table summarizes the behavior for the various S1 and S2
mapping combinations for systems with FWB enabled [3].
S1 | S2 | Result
NORMAL_NC | NORMAL_NC | NORMAL_NC
NORMAL_NC | DEVICE_nGnRE | DEVICE_nGnRE
Recently a change was added that modifies this default behavior and
make KVM map MMIO as MT_S2_FWB_NORMAL_NC when a VMA flag
VM_ALLOW_ANY_UNCACHED is set [4]. Setting S2 as MT_S2_FWB_NORMAL_NC
provides the desired behavior (uncached, unaligned access) for resmem.
To use VM_ALLOW_ANY_UNCACHED flag, the platform must guarantee that
no action taken on the MMIO mapping can trigger an uncontained
failure. The Grace Hopper satisfies this requirement. So set
the VM_ALLOW_ANY_UNCACHED flag in the VMA.
Brett Creeley [Wed, 28 Feb 2024 00:32:04 +0000 (16:32 -0800)]
vfio/pds: Always clear the save/restore FDs on reset
After reset the VFIO device state will always be put in
VFIO_DEVICE_STATE_RUNNING, but the save/restore files will only be
cleared if the previous state was VFIO_DEVICE_STATE_ERROR. This
can/will cause the restore/save files to be leaked if/when the
migration state machine transitions through the states that
re-allocates these files. Fix this by always clearing the
restore/save files for resets.
Fixes: 7dabb1bcd177 ("vfio/pds: Add support for firmware recovery") Cc: stable@vger.kernel.org Signed-off-by: Brett Creeley <brett.creeley@amd.com> Reviewed-by: Shannon Nelson <shannon.nelson@amd.com> Reviewed-by: Kevin Tian <kevin.tian@intel.com> Link: https://lore.kernel.org/r/20240228003205.47311-2-brett.creeley@amd.com Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Ankit Agrawal [Sat, 24 Feb 2024 15:05:46 +0000 (20:35 +0530)]
vfio: Convey kvm that the vfio-pci device is wc safe
The VM_ALLOW_ANY_UNCACHED flag is implemented for ARM64,
allowing KVM stage 2 device mapping attributes to use Normal-NC
rather than DEVICE_nGnRE, which allows guest mappings supporting
write-combining attributes (WC). ARM does not architecturally
guarantee this is safe, and indeed some MMIO regions like the GICv2
VCPU interface can trigger uncontained faults if Normal-NC is used.
To safely use VFIO in KVM the platform must guarantee full safety
in the guest where no action taken against a MMIO mapping can
trigger an uncontained failure. The expectation is that most VFIO PCI
platforms support this for both mapping types, at least in common
flows, based on some expectations of how PCI IP is integrated. So
make vfio-pci set the VM_ALLOW_ANY_UNCACHED flag.
Suggested-by: Catalin Marinas <catalin.marinas@arm.com> Acked-by: Jason Gunthorpe <jgg@nvidia.com> Acked-by: Catalin Marinas <catalin.marinas@arm.com> Acked-by: Alex Williamson <alex.williamson@redhat.com> Reviewed-by: David Hildenbrand <david@redhat.com> Reviewed-by: Marc Zyngier <maz@kernel.org> Signed-off-by: Ankit Agrawal <ankita@nvidia.com> Link: https://lore.kernel.org/r/20240224150546.368-5-ankita@nvidia.com Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Ankit Agrawal [Sat, 24 Feb 2024 15:05:45 +0000 (20:35 +0530)]
KVM: arm64: Set io memory s2 pte as normalnc for vfio pci device
To provide VM with the ability to get device IO memory with NormalNC
property, map device MMIO in KVM for ARM64 at stage2 as NormalNC.
Having NormalNC S2 default puts guests in control (based on [1],
"Combining stage 1 and stage 2 memory type attributes") of device
MMIO regions memory mappings. The rules are summarized below:
([(S1) - stage1], [(S2) - stage 2])
Still this cannot be generalized to non PCI devices such as GICv2.
There is insufficient information and uncertainity in the behavior
of non PCI driver. A driver must indicate support using the
new flag VM_ALLOW_ANY_UNCACHED.
Adapt KVM to make use of the flag VM_ALLOW_ANY_UNCACHED as indicator to
activate the S2 setting to NormalNc.
[1] section D8.5.5 of DDI0487J_a_a-profile_architecture_reference_manual.pdf
Suggested-by: Catalin Marinas <catalin.marinas@arm.com> Acked-by: Jason Gunthorpe <jgg@nvidia.com> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> Reviewed-by: Marc Zyngier <maz@kernel.org> Signed-off-by: Ankit Agrawal <ankita@nvidia.com> Link: https://lore.kernel.org/r/20240224150546.368-4-ankita@nvidia.com Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Ankit Agrawal [Sat, 24 Feb 2024 15:05:44 +0000 (20:35 +0530)]
mm: Introduce new flag to indicate wc safe
The VM_ALLOW_ANY_UNCACHED flag is implemented for ARM64, allowing KVM
stage 2 device mapping attributes to use NormalNC rather than
DEVICE_nGnRE, which allows guest mappings supporting write-combining
attributes (WC). ARM does not architecturally guarantee this is safe,
and indeed some MMIO regions like the GICv2 VCPU interface can trigger
uncontained faults if NormalNC is used.
Even worse, the expectation is that there are platforms where even
DEVICE_nGnRE can allow uncontained faults in corner cases. Unfortunately
existing ARM IP requires platform integration to take responsibility to
prevent this.
To safely use VFIO in KVM the platform must guarantee full safety in the
guest where no action taken against a MMIO mapping can trigger an
uncontained failure. The assumption is that most VFIO PCI platforms
support this for both mapping types, at least in common flows, based
on some expectations of how PCI IP is integrated. This can be enabled
more broadly, for instance into vfio-platform drivers, but only after
the platform vendor completes auditing for safety.
The VMA flag VM_ALLOW_ANY_UNCACHED was found to be the simplest and
cleanest way to communicate the information from VFIO to KVM that
mapping the region in S2 as NormalNC is safe. KVM consumes it to
activate the code that does the S2 mapping as NormalNC.
Suggested-by: Catalin Marinas <catalin.marinas@arm.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Reviewed-by: Marc Zyngier <maz@kernel.org> Acked-by: David Hildenbrand <david@redhat.com> Signed-off-by: Ankit Agrawal <ankita@nvidia.com> Link: https://lore.kernel.org/r/20240224150546.368-3-ankita@nvidia.com Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Ankit Agrawal [Sat, 24 Feb 2024 15:05:43 +0000 (20:35 +0530)]
KVM: arm64: Introduce new flag for non-cacheable IO memory
Currently, KVM for ARM64 maps at stage 2 memory that is considered device
(i.e. it is not RAM) with DEVICE_nGnRE memory attributes; this setting
overrides (as per the ARM architecture [1]) any device MMIO mapping
present at stage 1, resulting in a set-up whereby a guest operating
system cannot determine device MMIO mapping memory attributes on its
own but it is always overridden by the KVM stage 2 default.
This set-up does not allow guest operating systems to select device
memory attributes independently from KVM stage-2 mappings
(refer to [1], "Combining stage 1 and stage 2 memory type attributes"),
which turns out to be an issue in that guest operating systems
(e.g. Linux) may request to map devices MMIO regions with memory
attributes that guarantee better performance (e.g. gathering
attribute - that for some devices can generate larger PCIe memory
writes TLPs) and specific operations (e.g. unaligned transactions)
such as the NormalNC memory type.
The default device stage 2 mapping was chosen in KVM for ARM64 since
it was considered safer (i.e. it would not allow guests to trigger
uncontained failures ultimately crashing the machine) but this
turned out to be asynchronous (SError) defeating the purpose.
Failures containability is a property of the platform and is independent
from the memory type used for MMIO device memory mappings.
Actually, DEVICE_nGnRE memory type is even more problematic than
Normal-NC memory type in terms of faults containability in that e.g.
aborts triggered on DEVICE_nGnRE loads cannot be made, architecturally,
synchronous (i.e. that would imply that the processor should issue at
most 1 load transaction at a time - it cannot pipeline them - otherwise
the synchronous abort semantics would break the no-speculation attribute
attached to DEVICE_XXX memory).
This means that regardless of the combined stage1+stage2 mappings a
platform is safe if and only if device transactions cannot trigger
uncontained failures and that in turn relies on platform capabilities
and the device type being assigned (i.e. PCIe AER/DPC error containment
and RAS architecture[3]); therefore the default KVM device stage 2
memory attributes play no role in making device assignment safer
for a given platform (if the platform design adheres to design
guidelines outlined in [3]) and therefore can be relaxed.
For all these reasons, relax the KVM stage 2 device memory attributes
from DEVICE_nGnRE to Normal-NC.
The NormalNC was chosen over a different Normal memory type default
at stage-2 (e.g. Normal Write-through) to avoid cache allocation/snooping.
Relaxing S2 KVM device MMIO mappings to Normal-NC is not expected to
trigger any issue on guest device reclaim use cases either (i.e. device
MMIO unmap followed by a device reset) at least for PCIe devices, in that
in PCIe a device reset is architected and carried out through PCI config
space transactions that are naturally ordered with respect to MMIO
transactions according to the PCI ordering rules.
Having Normal-NC S2 default puts guests in control (thanks to
stage1+stage2 combined memory attributes rules [1]) of device MMIO
regions memory mappings, according to the rules described in [1]
and summarized here ([(S1) - stage1], [(S2) - stage 2]):
It is worth noting that currently, to map devices MMIO space to user
space in a device pass-through use case the VFIO framework applies memory
attributes derived from pgprot_noncached() settings applied to VMAs, which
result in device-nGnRnE memory attributes for the stage-1 VMM mappings.
This means that a userspace mapping for device MMIO space carried
out with the current VFIO framework and a guest OS mapping for the same
MMIO space may result in a mismatched alias as described in [2].
Defaulting KVM device stage-2 mappings to Normal-NC attributes does not
change anything in this respect, in that the mismatched aliases would
only affect (refer to [2] for a detailed explanation) ordering between
the userspace and GuestOS mappings resulting stream of transactions
(i.e. it does not cause loss of property for either stream of
transactions on its own), which is harmless given that the userspace
and GuestOS access to the device is carried out through independent
transactions streams.
A Normal-NC flag is not present today. So add a new kvm_pgtable_prot
(KVM_PGTABLE_PROT_NORMAL_NC) flag for it, along with its
corresponding PTE value 0x5 (0b101) determined from [1].
Lastly, adapt the stage2 PTE property setter function
(stage2_set_prot_attr) to handle the NormalNC attribute.
Ankit Agrawal [Tue, 20 Feb 2024 11:50:55 +0000 (17:20 +0530)]
vfio/nvgrace-gpu: Add vfio pci variant module for grace hopper
NVIDIA's upcoming Grace Hopper Superchip provides a PCI-like device
for the on-chip GPU that is the logical OS representation of the
internal proprietary chip-to-chip cache coherent interconnect.
The device is peculiar compared to a real PCI device in that whilst
there is a real 64b PCI BAR1 (comprising region 2 & region 3) on the
device, it is not used to access device memory once the faster
chip-to-chip interconnect is initialized (occurs at the time of host
system boot). The device memory is accessed instead using the chip-to-chip
interconnect that is exposed as a contiguous physically addressable
region on the host. This device memory aperture can be obtained from host
ACPI table using device_property_read_u64(), according to the FW
specification. Since the device memory is cache coherent with the CPU,
it can be mmap into the user VMA with a cacheable mapping using
remap_pfn_range() and used like a regular RAM. The device memory
is not added to the host kernel, but mapped directly as this reduces
memory wastage due to struct pages.
There is also a requirement of a minimum reserved 1G uncached region
(termed as resmem) to support the Multi-Instance GPU (MIG) feature [1].
This is to work around a HW defect. Based on [2], the requisite properties
(uncached, unaligned access) can be achieved through a VM mapping (S1)
of NORMAL_NC and host (S2) mapping with MemAttr[2:0]=0b101. To provide
a different non-cached property to the reserved 1G region, it needs to
be carved out from the device memory and mapped as a separate region
in Qemu VMA with pgprot_writecombine(). pgprot_writecombine() sets the
Qemu VMA page properties (pgprot) as NORMAL_NC.
Provide a VFIO PCI variant driver that adapts the unique device memory
representation into a more standard PCI representation facing userspace.
The variant driver exposes these two regions - the non-cached reserved
(resmem) and the cached rest of the device memory (termed as usemem) as
separate VFIO 64b BAR regions. This is divergent from the baremetal
approach, where the device memory is exposed as a device memory region.
The decision for a different approach was taken in view of the fact that
it would necessiate additional code in Qemu to discover and insert those
regions in the VM IPA, along with the additional VM ACPI DSDT changes to
communicate the device memory region IPA to the VM workloads. Moreover,
this behavior would have to be added to a variety of emulators (beyond
top of tree Qemu) out there desiring grace hopper support.
Since the device implements 64-bit BAR0, the VFIO PCI variant driver
maps the uncached carved out region to the next available PCI BAR (i.e.
comprising of region 2 and 3). The cached device memory aperture is
assigned BAR region 4 and 5. Qemu will then naturally generate a PCI
device in the VM with the uncached aperture reported as BAR2 region,
the cacheable as BAR4. The variant driver provides emulation for these
fake BARs' PCI config space offset registers.
The hardware ensures that the system does not crash when the memory
is accessed with the memory enable turned off. It synthesis ~0 reads
and dropped writes on such access. So there is no need to support the
disablement/enablement of BAR through PCI_COMMAND config space register.
The memory layout on the host looks like the following:
devmem (memlength)
|--------------------------------------------------|
|-------------cached------------------------|--NC--|
| |
usemem.memphys resmem.memphys
PCI BARs need to be aligned to the power-of-2, but the actual memory on the
device may not. A read or write access to the physical address from the
last device PFN up to the next power-of-2 aligned physical address
results in reading ~0 and dropped writes. Note that the GPU device
driver [6] is capable of knowing the exact device memory size through
separate means. The device memory size is primarily kept in the system
ACPI tables for use by the VFIO PCI variant module.
Note that the usemem memory is added by the VM Nvidia device driver [5]
to the VM kernel as memblocks. Hence make the usable memory size memblock
(MEMBLK_SIZE) aligned. This is a hardwired ABI value between the GPU FW and
VFIO driver. The VM device driver make use of the same value for its
calculation to determine USEMEM size.
Currently there is no provision in KVM for a S2 mapping with
MemAttr[2:0]=0b101, but there is an ongoing effort to provide the same [3].
As previously mentioned, resmem is mapped pgprot_writecombine(), that
sets the Qemu VMA page properties (pgprot) as NORMAL_NC. Using the
proposed changes in [3] and [4], KVM marks the region with
MemAttr[2:0]=0b101 in S2.
If the device memory properties are not present, the driver registers the
vfio-pci-core function pointers. Since there are no ACPI memory properties
generated for the VM, the variant driver inside the VM will only use
the vfio-pci-core ops and hence try to map the BARs as non cached. This
is not a problem as the CPUs have FWB enabled which blocks the VM
mapping's ability to override the cacheability set by the host mapping.
This goes along with a qemu series [6] to provides the necessary
implementation of the Grace Hopper Superchip firmware specification so
that the guest operating system can see the correct ACPI modeling for
the coherent GPU device. Verified with the CUDA workload in the VM.
Ankit Agrawal [Tue, 20 Feb 2024 11:50:54 +0000 (17:20 +0530)]
vfio/pci: rename and export range_intersect_range
range_intersect_range determines an overlap between two ranges. If an
overlap, the helper function returns the overlapping offset and size.
The VFIO PCI variant driver emulates the PCI config space BAR offset
registers. These offset may be accessed for read/write with a variety
of lengths including sub-word sizes from sub-word offsets. The driver
makes use of this helper function to read/write the targeted part of
the emulated register.
Make this a vfio_pci_core function, rename and export as GPL. Also
update references in virtio driver.
Reviewed-by: Kevin Tian <kevin.tian@intel.com> Reviewed-by: Yishai Hadas <yishaih@nvidia.com> Signed-off-by: Ankit Agrawal <ankita@nvidia.com> Link: https://lore.kernel.org/r/20240220115055.23546-3-ankita@nvidia.com Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Ricardo B. Marliere [Thu, 8 Feb 2024 20:02:04 +0000 (17:02 -0300)]
vfio: mdev: make mdev_bus_type const
Now that the driver core can properly handle constant struct bus_type,
move the mdev_bus_type variable to be a constant structure as well,
placing it into read-only memory which can not be modified at runtime.
Linus Torvalds [Sun, 18 Feb 2024 18:09:25 +0000 (10:09 -0800)]
Merge tag 'kbuild-fixes-v6.8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild
Pull Kbuild fixes from Masahiro Yamada:
- Reformat nested if-conditionals in Makefiles with 4 spaces
- Fix CONFIG_DEBUG_INFO_BTF builds for big endian
- Fix modpost for module srcversion
- Fix an escape sequence warning in gen_compile_commands.py
- Fix kallsyms to ignore ARMv4 thunk symbols
* tag 'kbuild-fixes-v6.8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
kallsyms: ignore ARMv4 thunks along with others
modpost: trim leading spaces when processing source files list
gen_compile_commands: fix invalid escape sequence warning
kbuild: Fix changing ELF file type for output of gen_btf for big endian
docs: kconfig: Fix grammar and formatting
kbuild: use 4-space indentation when followed by conditionals
Linus Torvalds [Sun, 18 Feb 2024 17:22:48 +0000 (09:22 -0800)]
Merge tag 'x86_urgent_for_v6.8_rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fix from Borislav Petkov:
- Use a GB page for identity mapping only when memory of this size is
requested so that mapping of reserved regions is prevented which
would otherwise lead to system crashes on UV machines
* tag 'x86_urgent_for_v6.8_rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/mm/ident_map: Use gbpages only where full GB page should be mapped.
- Prevent spurious interrupts on Broadcom devices using GIC v3
architecture
- Other minor fixes
* tag 'irq_urgent_for_v6.8_rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
irqchip/gic-v3-its: Fix GICv4.1 VPE affinity update
irqchip/gic-v3-its: Restore quirk probing for ACPI-based systems
irqchip/gic-v3-its: Handle non-coherent GICv4 redistributors
irqchip/qcom-mpm: Fix IS_ERR() vs NULL check in qcom_mpm_init()
irqchip/loongson-eiointc: Use correct struct type in eiointc_domain_alloc()
irqchip/irq-brcmstb-l2: Add write memory barrier before exit
Linus Torvalds [Sun, 18 Feb 2024 17:08:57 +0000 (09:08 -0800)]
Merge tag 'i2c-for-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux
Pull i2c fixes from Wolfram Sang:
"Two fixes for i801 and qcom-geni devices. Meanwhile, a fix from Arnd
addresses a compilation error encountered during compile test on
powerpc"
* tag 'i2c-for-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
i2c: i801: Fix block process call transactions
i2c: pasemi: split driver into two separate modules
i2c: qcom-geni: Correct I2C TRE sequence
Linus Torvalds [Sun, 18 Feb 2024 00:59:31 +0000 (16:59 -0800)]
Merge tag 'powerpc-6.8-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
Pull powerpc fixes from Michael Ellerman:
"This is a bit of a big batch for rc4, but just due to holiday hangover
and because I didn't send any fixes last week due to a late revert
request. I think next week should be back to normal.
- Fix ftrace bug on boot caused by exit text sections with
'-fpatchable-function-entry'
- Fix accuracy of stolen time on pseries since the switch to
VIRT_CPU_ACCOUNTING_GEN
- Fix a crash in the IOMMU code when doing DLPAR remove
- Set pt_regs->link on scv entry to fix BPF stack unwinding
- Add missing PPC_FEATURE_BOOKE on 64-bit e5500/e6500, which broke
gdb
- Fix boot on some 6xx platforms with STRICT_KERNEL_RWX enabled
- Fix build failures with KASAN enabled and 32KB stack size
- Some other minor fixes
Thanks to Arnd Bergmann, Benjamin Gray, Christophe Leroy, David
Engraf, Gaurav Batra, Jason Gunthorpe, Jiangfeng Xiao, Matthias
Schiffer, Nathan Lynch, Naveen N Rao, Nicholas Piggin, Nysal Jan K.A,
R Nageswara Sastry, Shivaprasad G Bhat, Shrikanth Hegde, Spoorthy,
Srikar Dronamraju, and Venkat Rao Bagalkote"
* tag 'powerpc-6.8-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
powerpc/iommu: Fix the missing iommu_group_put() during platform domain attach
powerpc/pseries: fix accuracy of stolen time
powerpc/ftrace: Ignore ftrace locations in exit text sections
powerpc/cputable: Add missing PPC_FEATURE_BOOKE on PPC64 Book-E
powerpc/kasan: Limit KASAN thread size increase to 32KB
Revert "powerpc/pseries/iommu: Fix iommu initialisation during DLPAR add"
powerpc: 85xx: mark local functions static
powerpc: udbg_memcons: mark functions static
powerpc/kasan: Fix addr error caused by page alignment
powerpc/6xx: set High BAT Enable flag on G2_LE cores
selftests/powerpc/papr_vpd: Check devfd before get_system_loc_code()
powerpc/64: Set task pt_regs->link to the LR value on scv entry
powerpc/pseries/iommu: Fix iommu initialisation during DLPAR add
powerpc/pseries/papr-sysparm: use u8 arrays for payloads
Linus Torvalds [Sat, 17 Feb 2024 16:56:41 +0000 (08:56 -0800)]
Merge tag 'driver-core-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
Pull driver core fixes from Greg KH:
"Here are some driver core fixes, a kobject fix, and a documentation
update for 6.8-rc5. In detail these changes are:
- devlink fixes for reported issues with 6.8-rc1
- topology scheduling regression fix that has been reported by many
- kobject loosening of checks change in -rc1 is now reverted as some
codepaths seemed to need the checks
- documentation update for the CVE process. Has been reviewed by
many, the last minute change to the document was to bring the .rst
format back into the the new style rules, the contents did not
change.
All of these, except for the documentation update, have been in
linux-next for over a week. The documentation update has been reviewed
for weeks by a group of developers, and in public for a week and the
wording has stabilized for now. If future changes are needed, we can
do so before 6.8-final is out (or anytime after that)"
* tag 'driver-core-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
Documentation: Document the Linux Kernel CVE process
Revert "kobject: Remove redundant checks for whether ktype is NULL"
driver core: fw_devlink: Improve logs for cycle detection
driver core: fw_devlink: Improve detection of overlapping cycles
driver core: Fix device_link_flag_is_sync_state_only()
topology: Set capacity_freq_ref in all cases
Linus Torvalds [Sat, 17 Feb 2024 16:52:38 +0000 (08:52 -0800)]
Merge tag 'char-misc-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
Pull char / miscdriver fixes from Greg KH:
"Here is a small set of char/misc and IIO driver fixes for 6.8-rc5.
Included in here are:
- lots of iio driver fixes for reported issues
- nvmem device naming fixup for reported problem
- interconnect driver fixes for reported issues
All of these have been in linux-next for a while with no reported the
issues (the nvmem patch was included in a different branch in
linux-next before sent to me for inclusion here)"
* tag 'char-misc-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: (21 commits)
nvmem: include bit index in cell sysfs file name
iio: adc: ad4130: only set GPIO_CTRL if pin is unused
iio: adc: ad4130: zero-initialize clock init data
interconnect: qcom: x1e80100: Add missing ACV enable_mask
interconnect: qcom: sm8650: Use correct ACV enable_mask
iio: accel: bma400: Fix a compilation problem
iio: commom: st_sensors: ensure proper DMA alignment
iio: hid-sensor-als: Return 0 for HID_USAGE_SENSOR_TIME_TIMESTAMP
iio: move LIGHT_UVA and LIGHT_UVB to the end of iio_modifier
staging: iio: ad5933: fix type mismatch regression
iio: humidity: hdc3020: fix temperature offset
iio: adc: ad7091r8: Fix error code in ad7091r8_gpio_setup()
iio: adc: ad_sigma_delta: ensure proper DMA alignment
iio: imu: adis: ensure proper DMA alignment
iio: humidity: hdc3020: Add Makefile, Kconfig and MAINTAINERS entry
iio: imu: bno055: serdev requires REGMAP
iio: magnetometer: rm3100: add boundary check for the value read from RM3100_REG_TMRC
iio: pressure: bmp280: Add missing bmp085 to SPI id table
iio: core: fix memleak in iio_device_register_sysfs
interconnect: qcom: sm8550: Enable sync_state
...
Linus Torvalds [Sat, 17 Feb 2024 16:46:57 +0000 (08:46 -0800)]
Merge tag 'tty-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
Pull tty / serial fixes from Greg KH:
"Here are three small tty and serial driver fixes for 6.8-rc5:
- revert a 8250_pci1xxxx off-by-one change that was incorrect
- two changes to fix the transmit path of the mxs-auart driver,
fixing a regression in the 6.2 release
All of these have been in linux-next for over a week with no reported
issues"
* tag 'tty-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
serial: mxs-auart: fix tx
serial: core: introduce uart_port_tx_flags()
serial: 8250_pci1xxxx: partially revert off by one patch
Linus Torvalds [Sat, 17 Feb 2024 16:44:55 +0000 (08:44 -0800)]
Merge tag 'usb-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Pull USB / Thunderbolt fixes from Greg KH:
"Here are two small fixes for 6.8-rc5:
- thunderbolt to fix a reported issue on many platforms
- dwc3 driver revert of a commit that caused problems in -rc1
Both of these changes have been in linux-next for over a week with no
reported issues"
* tag 'usb-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
Revert "usb: dwc3: Support EBC feature of DWC_usb31"
thunderbolt: Fix setting the CNS bit in ROUTER_CS_5
Linus Torvalds [Sat, 17 Feb 2024 16:13:32 +0000 (08:13 -0800)]
Merge tag 'media/v6.8-4' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
Pull media fixes from Mauro Carvalho Chehab:
- regression fix for rkisp1 shared IRQ logic
- fix atomisp breakage due to a kAPI change
- permission fix for remote controller BPF support
- memleak fix in ir_toy driver
- Kconfig dependency fix for pwm-ir-rx
* tag 'media/v6.8-4' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media:
media: pwm-ir-tx: Depend on CONFIG_HIGH_RES_TIMERS
media: ir_toy: fix a memleak in irtoy_tx
media: rc: bpf attach/detach requires write permission
media: atomisp: Adjust for v4l2_subdev_state handling changes in 6.8
media: rkisp1: Fix IRQ handling due to shared interrupts
media: Revert "media: rkisp1: Drop IRQF_SHARED"
Linus Torvalds [Sat, 17 Feb 2024 16:06:20 +0000 (08:06 -0800)]
Merge tag 'pci-v6.8-fixes-3' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci
Pull pci fixes from Bjorn Helgaas:
- Keep bridges in D0 if we need to poll downstream devices for PME to
resolve a v6.6 regression where we failed to enumerate devices below
bridges put in D3hot by runtime PM, e.g., NVMe drives connected via
Thunderbolt or USB4 docks (Alex Williamson)
- Add Siddharth Vadapalli as PCI TI DRA7XX/J721E reviewer
* tag 'pci-v6.8-fixes-3' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci:
MAINTAINERS: Add Siddharth Vadapalli as PCI TI DRA7XX/J721E reviewer
PCI: Fix active state requirement in PME polling
Linus Torvalds [Sat, 17 Feb 2024 15:59:47 +0000 (07:59 -0800)]
Merge tag 'probes-fixes-v6.8-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull probes fix from Masami Hiramatsu:
- tracing/probes: Fix BTF structure member finder to find the members
which are placed after any anonymous union member correctly.
* tag 'probes-fixes-v6.8-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
tracing/probes: Fix to search structure fields correctly
Linus Torvalds [Sat, 17 Feb 2024 15:56:10 +0000 (07:56 -0800)]
Merge tag '6.8-rc4-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6
Pull smb client fixes from Steve French:
"Five smb3 client fixes, most also for stable:
- Two multichannel fixes (one to fix potential handle leak on retry)
- Work around possible serious data corruption (due to change in
folios in 6.3, for cases when non standard maximum write size
negotiated)
- Symlink creation fix
- Multiuser automount fix"
* tag '6.8-rc4-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
smb: Fix regression in writes when non-standard maximum write size negotiated
smb: client: handle path separator of created SMB symlinks
smb: client: set correct id, uid and cruid for multiuser automounts
cifs: update the same create_guid on replay
cifs: fix underflow in parse_server_interfaces()
Greg Kroah-Hartman [Sat, 17 Feb 2024 12:55:31 +0000 (13:55 +0100)]
Documentation: Document the Linux Kernel CVE process
The Linux kernel project now has the ability to assign CVEs to fixed
issues, so document the process and how individual developers can get a
CVE if one is not automatically assigned for their fixes.
Reviewed-by: Kees Cook <keescook@chromium.org> Reviewed-by: Konstantin Ryabitsev <konstantin@linuxfoundation.org> Reviewed-by: Krzysztof Kozlowski <krzk@kernel.org> Reviewed-by: Lukas Bulwahn <lukas.bulwahn@gmail.com> Signed-off-by: Sasha Levin <sashal@kernel.org> Signed-off-by: Lee Jones <lee@kernel.org> Link: https://lore.kernel.org/r/2024021731-essence-sadness-28fd@gregkh Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
tracing/probes: Fix to search structure fields correctly
Fix to search a field from the structure which has anonymous union
correctly.
Since the reference `type` pointer was updated in the loop, the search
loop suddenly aborted where it hits an anonymous union. Thus it can not
find the field after the anonymous union. This avoids updating the
cursor `type` pointer in the loop.
Wolfram Sang [Sat, 17 Feb 2024 12:13:33 +0000 (13:13 +0100)]
Merge tag 'i2c-host-fixes-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/andi.shyti/linux into i2c/for-current
Three fixes are included here. Two are strictly hardware-related
for the i801 and qcom-geni devices. Meanwhile, a fix from Arnd
addresses a compilation error encountered during compile test on
powerpc.
Linus Torvalds [Fri, 16 Feb 2024 22:05:02 +0000 (14:05 -0800)]
Merge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI fixes from James Bottomley:
"Three fixes: the two fnic ones are a revert and a refix, which is why
the diffstat is a bit big. The target one also extracts a function to
add a check for configuration and so looks bigger than it is"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: fnic: Move fnic_fnic_flush_tx() to a work queue
scsi: Revert "scsi: fcoe: Fix potential deadlock on &fip->ctlr_lock"
scsi: target: Fix unmap setup during configuration
Linus Torvalds [Fri, 16 Feb 2024 22:00:19 +0000 (14:00 -0800)]
Merge tag 'wq-for-6.8-rc4-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq
Pull workqueue fix from Tejun Heo:
"Just one patch to revert commit ca10d851b9ad ("workqueue: Override
implicit ordered attribute in workqueue_apply_unbound_cpumask()").
This commit could break ordering guarantees for ordered workqueues.
The problem that the commit tried to resolve partially - making
ordered workqueues follow unbound cpumask - is fully solved in
wq/for-6.9 branch"
* tag 'wq-for-6.8-rc4-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
Revert "workqueue: Override implicit ordered attribute in workqueue_apply_unbound_cpumask()"
Linus Torvalds [Fri, 16 Feb 2024 21:37:15 +0000 (13:37 -0800)]
Merge tag 'ceph-for-6.8-rc5' of https://github.com/ceph/ceph-client
Pull ceph fixes from Ilya Dryomov:
"Additional cap handling fixes from Xiubo to avoid "client isn't
responding to mclientcaps(revoke)" stalls on the MDS side"
* tag 'ceph-for-6.8-rc5' of https://github.com/ceph/ceph-client:
ceph: add ceph_cap_unlink_work to fire check_caps() immediately
ceph: always queue a writeback when revoking the Fb caps
Linus Torvalds [Fri, 16 Feb 2024 18:48:14 +0000 (10:48 -0800)]
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM fixes from Paolo Bonzini:
"ARM:
- Avoid dropping the page refcount twice when freeing an unlinked
page-table subtree.
- Don't source the VFIO Kconfig twice
- Fix protected-mode locking order between kvm and vcpus
RISC-V:
- Fix steal-time related sparse warnings
x86:
- Cleanup gtod_is_based_on_tsc() to return "bool" instead of an "int"
- Make a KVM_REQ_NMI request while handling KVM_SET_VCPU_EVENTS if
and only if the incoming events->nmi.pending is non-zero. If the
target vCPU is in the UNITIALIZED state, the spurious request will
result in KVM exiting to userspace, which in turn causes QEMU to
constantly acquire and release QEMU's global mutex, to the point
where the BSP is unable to make forward progress.
- Fix a type (u8 versus u64) goof that results in pmu->fixed_ctr_ctrl
being incorrectly truncated, and ultimately causes KVM to think a
fixed counter has already been disabled (KVM thinks the old value
is '0').
- Fix a stack leak in KVM_GET_MSRS where a failed MSR read from
userspace that is ultimately ignored due to ignore_msrs=true
doesn't zero the output as intended.
Selftests cleanups and fixes:
- Remove redundant newlines from error messages.
- Delete an unused variable in the AMX test (which causes build
failures when compiling with -Werror).
- Fail instead of skipping tests if open(), e.g. of /dev/kvm, fails
with an error code other than ENOENT (a Hyper-V selftest bug
resulted in an EMFILE, and the test eventually got skipped).
- Fix TSC related bugs in several Hyper-V selftests.
- Fix a bug in the dirty ring logging test where a sem_post() could
be left pending across multiple runs, resulting in incorrect
synchronization between the main thread and the vCPU worker thread.
- Relax the dirty log split test's assertions on 4KiB mappings to fix
false positives due to the number of mappings for memslot 0 (used
for code and data that is NOT being dirty logged) changing, e.g.
due to NUMA balancing"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (25 commits)
KVM: arm64: Fix double-free following kvm_pgtable_stage2_free_unlinked()
RISC-V: KVM: Use correct restricted types
RISC-V: paravirt: Use correct restricted types
RISC-V: paravirt: steal_time should be static
KVM: selftests: Don't assert on exact number of 4KiB in dirty log split test
KVM: selftests: Fix a semaphore imbalance in the dirty ring logging test
KVM: x86: Fix KVM_GET_MSRS stack info leak
KVM: arm64: Do not source virt/lib/Kconfig twice
KVM: x86/pmu: Fix type length error when reading pmu->fixed_ctr_ctrl
KVM: x86: Make gtod_is_based_on_tsc() return 'bool'
KVM: selftests: Make hyperv_clock require TSC based system clocksource
KVM: selftests: Run clocksource dependent tests with hyperv_clocksource_tsc_page too
KVM: selftests: Use generic sys_clocksource_is_tsc() in vmx_nested_tsc_scaling_test
KVM: selftests: Generalize check_clocksource() from kvm_clock_test
KVM: x86: make KVM_REQ_NMI request iff NMI pending for vcpu
KVM: arm64: Fix circular locking dependency
KVM: selftests: Fail tests when open() fails with !ENOENT
KVM: selftests: Avoid infinite loop in hyperv_features when invtsc is missing
KVM: selftests: Delete superfluous, unused "stage" variable in AMX test
KVM: selftests: x86_64: Remove redundant newlines
...
Linus Torvalds [Fri, 16 Feb 2024 18:33:51 +0000 (10:33 -0800)]
Merge tag 'trace-v6.8-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull tracing fixes from Steven Rostedt:
- Fix the #ifndef that didn't have the 'CONFIG_' prefix on
HAVE_DYNAMIC_FTRACE_WITH_REGS
The fix to have dynamic trampolines work with x86 broke arm64 as the
config used in the #ifdef was HAVE_DYNAMIC_FTRACE_WITH_REGS and not
CONFIG_HAVE_DYNAMIC_FTRACE_WITH_REGS which removed the fix that the
previous fix was to fix.
- Fix tracing_on state
The code to test if "tracing_on" is set incorrectly used
ring_buffer_record_is_on() which returns false if the ring buffer
isn't able to be written to.
But the ring buffer disable has several bits that disable it. One is
internal disabling which is used for resizing and other modifications
of the ring buffer. But the "tracing_on" user space visible flag
should only report if tracing is actually on and not internally
disabled, as this can cause confusion as writing "1" when it is
disabled will not enable it.
Instead use ring_buffer_record_is_set_on() which shows the user space
visible settings.
- Fix a false positive kmemleak on saved cmdlines
Now that the saved_cmdlines structure is allocated via alloc_page()
and not via kmalloc() it has become invisible to kmemleak. The
allocation done to one of its pointers was flagged as a dangling
allocation leak. Make kmemleak aware of this allocation and free.
- Fix synthetic event dynamic strings
An update that cleaned up the synthetic event code removed the return
value of trace_string(), and had it return zero instead of the
length, causing dynamic strings in the synthetic event to always have
zero size.
- Clean up documentation and header files for seq_buf
* tag 'trace-v6.8-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
seq_buf: Fix kernel documentation
seq_buf: Don't use "proxy" headers
tracing/synthetic: Fix trace_string() return value
tracing: Inform kmemleak of saved_cmdlines allocation
tracing: Use ring_buffer_record_is_set_on() in tracer_tracing_is_on()
tracing: Fix HAVE_DYNAMIC_FTRACE_WITH_REGS ifdef
Linus Torvalds [Fri, 16 Feb 2024 18:28:29 +0000 (10:28 -0800)]
Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fixes from Will Deacon:
"It's a little busier than normal, but it's still not a lot of code and
things seem fairly quiet in general:
- Fix allocation failure during SVE coredumps
- Fix handling of SVE context on signal delivery
- Enable Neoverse N2 CPU errata workarounds for Microsoft's "Azure
Cobalt 100" clone
- Work around CMN PMU erratum in AmpereOneX implementation
- Fix typo in CXL PMU event definition
- Fix jump label asm constraints"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64/sve: Lower the maximum allocation for the SVE ptrace regset
arm64: Subscribe Microsoft Azure Cobalt 100 to ARM Neoverse N2 errata
perf/arm-cmn: Workaround AmpereOneX errata AC04_MESH_1 (incorrect child count)
arm64: jump_label: use constraints "Si" instead of "i"
arm64: fix typo in comments
perf: CXL: fix mismatched cpmu event opcode
arm64/signal: Don't assume that TIF_SVE means we saved SVE state
Linus Torvalds [Fri, 16 Feb 2024 17:29:26 +0000 (09:29 -0800)]
Merge tag 'zonefs-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal/zonefs
Pull zonefs fix from Damien Le Moal:
- Fix direct write error handling to avoid a race between failed IO
completion and the submission path itself which can result in an
invalid file size exposed to the user after the failed IO.
* tag 'zonefs-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal/zonefs:
zonefs: Improve error handling
Linus Torvalds [Fri, 16 Feb 2024 17:02:19 +0000 (09:02 -0800)]
Merge tag 'sound-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull sound fixes from Takashi Iwai:
"A collection of device-specific fixes. It became a bit bigger than
wished, but all look reasonably small and safe to apply.
- A few Cirrus Logic CS35L56 and CS42L43 driver fixes
- ASoC SOF fixes and workarounds
- Various ASoC Intel fixes
- Lots of HD-, USB-audio and AMD ACP quirks"
* tag 'sound-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (33 commits)
ALSA: usb-audio: More relaxed check of MIDI jack names
ALSA: hda/realtek: fix mute/micmute LED For HP mt645
ALSA: hda/realtek: cs35l41: Fix order and duplicates in quirks table
ALSA: hda/realtek: cs35l41: Fix device ID / model name
ALSA: hda/realtek: cs35l41: Add internal speaker support for ASUS UM3402 with missing DSD
ASoC: cs35l56: Workaround for ACPI with broken spk-id-gpios property
ALSA: hda: Add Lenovo Legion 7i gen7 sound quirk
ASoC: SOF: IPC3: fix message bounds on ipc ops
ASoC: SOF: ipc4-pcm: Workaround for crashed firmware on system suspend
ASoC: q6dsp: fix event handler prototype
ASoC: SOF: Intel: pci-lnl: Change the topology path to intel/sof-ipc4-tplg
ASoC: SOF: Intel: pci-tgl: Change the default paths and firmware names
ASoC: amd: yc: Fix non-functional mic on Lenovo 82UU
ASoC: rt5645: Add DMI quirk for inverted jack-detect on MeeGoPad T8
ASoC: rt5645: Make LattePanda board DMI match more precise
ASoC: SOF: amd: Fix locking in ACP IRQ handler
ASoC: rt5645: Fix deadlock in rt5645_jack_detect_work()
ASoC: Intel: cht_bsw_rt5645: Cleanup codec_name handling
ASoC: Intel: Boards: Fix NULL pointer deref in BYT/CHT boards
ASoC: cs35l56: Remove default from IRQ1_CFG register
...
Linus Torvalds [Fri, 16 Feb 2024 16:55:46 +0000 (08:55 -0800)]
Merge tag 'gpio-fixes-for-v6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux
Pull gpio fixes from Bartosz Golaszewski:
- add missing stubs for functions that are not built with GPIOLIB
disabled
* tag 'gpio-fixes-for-v6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:
gpiolib: add gpio_device_get_label() stub for !GPIOLIB
gpiolib: add gpio_device_get_base() stub for !GPIOLIB
gpiolib: add gpiod_to_gpio_device() stub for !GPIOLIB
Linus Torvalds [Fri, 16 Feb 2024 16:05:30 +0000 (08:05 -0800)]
Merge tag 'drm-fixes-2024-02-16' of git://anongit.freedesktop.org/drm/drm
Pull drm fixes from Dave Airlie:
"Regular weekly fixes, nothing too major, mostly amdgpu, then i915, xe,
msm and nouveau with some scattered bits elsewhere.
crtc:
- fix uninit variable
prime:
- support > 4GB page arrays
buddy:
- fix error handling in allocations
i915:
- fix blankscreen on JSL chromebooks
- stable fix to limit DP sst link rates
xe:
- Fix an out-of-bounds shift.
- Fix the display code thinking xe uses shmem
- Fix a warning about index out-of-bound
- Fix a clang-16 compilation warning
amdgpu:
- PSR fixes
- Suspend/resume fixes
- Link training fix
- Aspect ratio fix
- DCN 3.5 fixes
- VCN 4.x fix
- GFX 11 fix
- Misc display fixes
- Misc small fixes
amdkfd:
- Cache size reporting fix
- SIMD distribution fix
msm:
- GPU:
- dmabuf vmap fix
- a610 UBWC corruption fix (incorrect hbb)
- revert a commit that was making GPU recovery unreliable
- tlb invalidation fix
* tag 'drm-fixes-2024-02-16' of git://anongit.freedesktop.org/drm/drm: (38 commits)
drm/amdgpu: Fix implicit assumtion in gfx11 debug flags
drm/amdkfd: update SIMD distribution algo for GFXIP 9.4.2 onwards
drm/amd/display: Increase ips2_eval delay for DCN35
drm/amdgpu/display: Initialize gamma correction mode variable in dcn30_get_gamcor_current()
drm/amdgpu/soc21: update VCN 4 max HEVC encoding resolution
drm/amd/display: fixed integer types and null check locations
drm/amd/display: Fix array-index-out-of-bounds in dcn35_clkmgr
drm/amd/display: Preserve original aspect ratio in create stream
drm/amd/display: Fix possible NULL dereference on device remove/driver unload
Revert "drm/amd/display: increased min_dcfclk_mhz and min_fclk_mhz"
drm/amd/display: Add align done check
Revert "drm/amd: flush any delayed gfxoff on suspend entry"
drm/amd: Stop evicting resources on APUs in suspend
drm/amd/display: Fix possible buffer overflow in 'find_dcfclk_for_voltage()'
drm/amd/display: Fix possible use of uninitialized 'max_chunks_fbc_mode' in 'calculate_bandwidth()'
drm/amd/display: Initialize 'wait_time_microsec' variable in link_dp_training_dpia.c
drm/amd/display: Fix && vs || typos
drm/amdkfd: Fix L2 cache size reporting in GFX9.4.3
drm/amdgpu: make damage clips support configurable
drm/msm: Wire up tlb ops
...
Dave Airlie [Fri, 16 Feb 2024 05:45:03 +0000 (15:45 +1000)]
Merge tag 'drm-xe-fixes-2024-02-15' of https://gitlab.freedesktop.org/drm/xe/kernel into drm-fixes
Driver Changes:
- Fix an out-of-bounds shift.
- Fix the display code thinking xe uses shmem
- Fix a warning about index out-of-bound
- Fix a clang-16 compilation warning
Dave Airlie [Fri, 16 Feb 2024 04:33:09 +0000 (14:33 +1000)]
Merge tag 'drm-misc-fixes-2024-02-15' of git://anongit.freedesktop.org/drm/drm-misc into drm-fixes
A suspend/resume error fix for ivpu, a couple of scheduler fixes for
nouveau, a patch to support large page arrays in prime, a uninitialized
variable fix in crtc, a locking fix in rockchip/vop2 and a buddy
allocator error reporting fix.
Steve French [Tue, 6 Feb 2024 22:34:22 +0000 (16:34 -0600)]
smb: Fix regression in writes when non-standard maximum write size negotiated
The conversion to netfs in the 6.3 kernel caused a regression when
maximum write size is set by the server to an unexpected value which is
not a multiple of 4096 (similarly if the user overrides the maximum
write size by setting mount parm "wsize", but sets it to a value that
is not a multiple of 4096). When negotiated write size is not a
multiple of 4096 the netfs code can skip the end of the final
page when doing large sequential writes, causing data corruption.
This section of code is being rewritten/removed due to a large
netfs change, but until that point (ie for the 6.3 kernel until now)
we can not support non-standard maximum write sizes.
Add a warning if a user specifies a wsize on mount that is not
a multiple of 4096 (and round down), also add a change where we
round down the maximum write size if the server negotiates a value
that is not a multiple of 4096 (we also have to check to make sure that
we do not round it down to zero).
Reported-by: R. Diez" <rdiez-2006@rd10.de> Fixes: d08089f649a0 ("cifs: Change the I/O paths to use an iterator rather than a page list") Suggested-by: Ronnie Sahlberg <ronniesahlberg@gmail.com> Acked-by: Ronnie Sahlberg <ronniesahlberg@gmail.com> Tested-by: Matthew Ruffell <matthew.ruffell@canonical.com> Reviewed-by: Shyam Prasad N <sprasad@microsoft.com> Cc: stable@vger.kernel.org # v6.3+ Cc: David Howells <dhowells@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
Damien Le Moal [Thu, 8 Feb 2024 08:26:59 +0000 (17:26 +0900)]
zonefs: Improve error handling
Write error handling is racy and can sometime lead to the error recovery
path wrongly changing the inode size of a sequential zone file to an
incorrect value which results in garbage data being readable at the end
of a file. There are 2 problems:
1) zonefs_file_dio_write() updates a zone file write pointer offset
after issuing a direct IO with iomap_dio_rw(). This update is done
only if the IO succeed for synchronous direct writes. However, for
asynchronous direct writes, the update is done without waiting for
the IO completion so that the next asynchronous IO can be
immediately issued. However, if an asynchronous IO completes with a
failure right before the i_truncate_mutex lock protecting the update,
the update may change the value of the inode write pointer offset
that was corrected by the error path (zonefs_io_error() function).
2) zonefs_io_error() is called when a read or write error occurs. This
function executes a report zone operation using the callback function
zonefs_io_error_cb(), which does all the error recovery handling
based on the current zone condition, write pointer position and
according to the mount options being used. However, depending on the
zoned device being used, a report zone callback may be executed in a
context that is different from the context of __zonefs_io_error(). As
a result, zonefs_io_error_cb() may be executed without the inode
truncate mutex lock held, which can lead to invalid error processing.
Fix both problems as follows:
- Problem 1: Perform the inode write pointer offset update before a
direct write is issued with iomap_dio_rw(). This is safe to do as
partial direct writes are not supported (IOMAP_DIO_PARTIAL is not
set) and any failed IO will trigger the execution of zonefs_io_error()
which will correct the inode write pointer offset to reflect the
current state of the one on the device.
- Problem 2: Change zonefs_io_error_cb() into zonefs_handle_io_error()
and call this function directly from __zonefs_io_error() after
obtaining the zone information using blkdev_report_zones() with a
simple callback function that copies to a local stack variable the
struct blk_zone obtained from the device. This ensures that error
handling is performed holding the inode truncate mutex.
This change also simplifies error handling for conventional zone files
by bypassing the execution of report zones entirely. This is safe to
do because the condition of conventional zones cannot be read-only or
offline and conventional zone files are always fully mapped with a
constant file size.
Reported-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com> Fixes: 8dcc1a9d90c1 ("fs: New zonefs file system") Cc: stable@vger.kernel.org Signed-off-by: Damien Le Moal <dlemoal@kernel.org> Tested-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com> Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Linus Torvalds [Thu, 15 Feb 2024 19:39:27 +0000 (11:39 -0800)]
Merge tag 'net-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Pull networking fixes from Jakub Kicinski:
"Including fixes from can, wireless and netfilter.
Current release - regressions:
- af_unix: fix task hung while purging oob_skb in GC
- pds_core: do not try to run health-thread in VF path
Current release - new code bugs:
- sched: act_mirred: don't zero blockid when net device is being
deleted
Previous releases - regressions:
- netfilter:
- nat: restore default DNAT behavior
- nf_tables: fix bidirectional offload, broken when unidirectional
offload support was added
- openvswitch: limit the number of recursions from action sets
- eth: i40e: do not allow untrusted VF to remove administratively set
MAC address
Previous releases - always broken:
- tls: fix races and bugs in use of async crypto
- mptcp: prevent data races on some of the main socket fields, fix
races in fastopen handling
- dpll: fix possible deadlock during netlink dump operation
- dsa: lan966x: fix crash when adding interface under a lag when some
of the ports are disabled
- can: j1939: prevent deadlock by changing j1939_socks_lock to rwlock
Misc:
- a handful of fixes and reliability improvements for selftests
- fix sysfs documentation missing net/ in paths
- finish the work of squashing the missing MODULE_DESCRIPTION()
warnings in networking"
* tag 'net-6.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (92 commits)
net: fill in MODULE_DESCRIPTION()s for missing arcnet
net: fill in MODULE_DESCRIPTION()s for mdio_devres
net: fill in MODULE_DESCRIPTION()s for ppp
net: fill in MODULE_DESCRIPTION()s for fddik/skfp
net: fill in MODULE_DESCRIPTION()s for plip
net: fill in MODULE_DESCRIPTION()s for ieee802154/fakelb
net: fill in MODULE_DESCRIPTION()s for xen-netback
net: ravb: Count packets instead of descriptors in GbEth RX path
pppoe: Fix memory leak in pppoe_sendmsg()
net: sctp: fix skb leak in sctp_inq_free()
net: bcmasp: Handle RX buffer allocation failure
net-timestamp: make sk_tskey more predictable in error path
selftests: tls: increase the wait in poll_partial_rec_async
ice: Add check for lport extraction to LAG init
netfilter: nf_tables: fix bidirectional offload regression
netfilter: nat: restore default DNAT behavior
netfilter: nft_set_pipapo: fix missing : in kdoc
igc: Remove temporary workaround
igb: Fix string truncation warnings in igb_set_fw_version
can: netlink: Fix TDCO calculation using the old data bittiming
...
Linus Torvalds [Thu, 15 Feb 2024 19:33:35 +0000 (11:33 -0800)]
Merge tag 'for-linus-6.8a-rc5-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip
Pull xen fixes from Juergen Gross:
"Fixes and simple cleanups:
- use a proper flexible array instead of a one-element array in order
to avoid array-bounds sanitizer errors
- add NULL pointer checks after allocating memory
- use memdup_array_user() instead of open-coding it
- fix a rare race condition in Xen event channel allocation code
- make struct bus_type instances const
- make kerneldoc inline comments match reality"
* tag 'for-linus-6.8a-rc5-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
xen/events: close evtchn after mapping cleanup
xen/gntalloc: Replace UAPI 1-element array
xen: balloon: make balloon_subsys const
xen: pcpu: make xen_pcpu_subsys const
xen/privcmd: Use memdup_array_user() in alloc_ioreq()
x86/xen: Add some null pointer checking to smp.c
xen/xenbus: document will_handle argument for xenbus_watch_path()
Rajneesh Bhardwaj [Sat, 10 Feb 2024 01:23:19 +0000 (20:23 -0500)]
drm/amdgpu: Fix implicit assumtion in gfx11 debug flags
Gfx11 debug flags mask is currently set with an implicit assumption that
no other mqd update flags exist. This needs to be fixed with newly
introduced flag UPDATE_FLAG_IS_GWS by the previous patch.
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Rajneesh Bhardwaj [Thu, 1 Feb 2024 00:33:49 +0000 (19:33 -0500)]
drm/amdkfd: update SIMD distribution algo for GFXIP 9.4.2 onwards
In certain cooperative group dispatch scenarios the default SPI resource
allocation may cause reduced per-CU workgroup occupancy. Set
COMPUTE_RESOURCE_LIMITS.FORCE_SIMD_DIST=1 to mitigate soft hang
scenarions.
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Suggested-by: Joseph Greathouse <Joseph.Greathouse@amd.com> Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Srinivasan Shanmugam [Sun, 11 Feb 2024 03:34:19 +0000 (09:04 +0530)]
drm/amdgpu/display: Initialize gamma correction mode variable in dcn30_get_gamcor_current()
The dcn30_get_gamcor_current() function is responsible for determining
the current gamma correction mode used by the display controller.
However, the 'mode' variable, which stores the gamma correction mode,
was not initialized before its first usage, leading to an uninitialized
symbol error.
Thus initializes the 'mode' variable with a default value of LUT_BYPASS
before the conditional statements in the function, improves code clarity
and stability, ensuring correct behavior of the
dcn30_get_gamcor_current() function in determining the gamma correction
mode.
Fixes the below:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn30/dcn30_dpp_cm.c:77 dpp30_get_gamcor_current() error: uninitialized symbol 'mode'.
Fixes: 03f54d7d3448 ("drm/amd/display: Add DCN3 DPP") Cc: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com> Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Cc: Aurabindo Pillai <aurabindo.pillai@amd.com> Cc: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Suggested-by: Roman Li <roman.li@amd.com> Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Roman Li [Tue, 30 Jan 2024 23:07:24 +0000 (18:07 -0500)]
drm/amd/display: Fix array-index-out-of-bounds in dcn35_clkmgr
[Why]
There is a potential memory access violation while
iterating through array of dcn35 clks.
[How]
Limit iteration per array size.
Cc: Mario Limonciello <mario.limonciello@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Roman Li <roman.li@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Tom Chung [Tue, 30 Jan 2024 07:34:08 +0000 (15:34 +0800)]
drm/amd/display: Preserve original aspect ratio in create stream
[Why]
The original picture aspect ratio in mode struct may have chance be
overwritten with wrong aspect ratio data in create_stream_for_sink().
It will create a different VIC output and cause HDMI compliance test
failed.
[How]
Preserve the original picture aspect ratio data during create the
stream.
Cc: Mario Limonciello <mario.limonciello@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Tom Chung <chiahsuan.chung@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Srinivasan Shanmugam [Tue, 6 Feb 2024 04:04:25 +0000 (09:34 +0530)]
drm/amd/display: Fix possible NULL dereference on device remove/driver unload
As part of a cleanup amdgpu_dm_fini() function, which is typically
called when a device is being shut down or a driver is being unloaded
The below error message suggests that there is a potential null pointer
dereference issue with adev->dm.dc.
In the below, line of code where adev->dm.dc is used without a preceding
null check:
for (i = 0; i < adev->dm.dc->caps.max_links; i++) {
To fix this issue, add a null check for adev->dm.dc before this line.
Reported by smatch:
drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm.c:1959 amdgpu_dm_fini() error: we previously assumed 'adev->dm.dc' could be null (see line 1943)
Fixes: 006c26a0f1c8 ("drm/amd/display: Fix crash on device remove/driver unload") Cc: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Cc: Harry Wentland <harry.wentland@amd.com> Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Cc: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Roman Li <roman.li@amd.com> Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Mario Limonciello [Thu, 8 Feb 2024 05:52:54 +0000 (23:52 -0600)]
Revert "drm/amd: flush any delayed gfxoff on suspend entry"
commit ab4750332dbe ("drm/amdgpu/sdma5.2: add begin/end_use ring
callbacks") caused GFXOFF control to be used more heavily and the
codepath that was removed from commit 0dee72639533 ("drm/amd: flush any
delayed gfxoff on suspend entry") now can be exercised at suspend again.
Users report that by using GNOME to suspend the lockscreen trigger will
cause SDMA traffic and the system can deadlock.
Acked-by: Alex Deucher <alexander.deucher@amd.com> Fixes: ab4750332dbe ("drm/amdgpu/sdma5.2: add begin/end_use ring callbacks") Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Mario Limonciello [Thu, 8 Feb 2024 05:52:55 +0000 (23:52 -0600)]
drm/amd: Stop evicting resources on APUs in suspend
commit 5095d5418193 ("drm/amd: Evict resources during PM ops prepare()
callback") intentionally moved the eviction of resources to earlier in
the suspend process, but this introduced a subtle change that it occurs
before adev->in_s0ix or adev->in_s3 are set. This meant that APUs
actually started to evict resources at suspend time as well.
Explicitly set s0ix or s3 in the prepare() stage, and unset them if the
prepare() stage failed.
v2: squash in warning fix from Stephen Rothwell
Reported-by: Jürg Billeter <j@bitron.ch> Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3132#note_2271038 Fixes: 5095d5418193 ("drm/amd: Evict resources during PM ops prepare() callback") Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Srinivasan Shanmugam [Mon, 5 Feb 2024 11:24:10 +0000 (16:54 +0530)]
drm/amd/display: Fix possible buffer overflow in 'find_dcfclk_for_voltage()'
when 'find_dcfclk_for_voltage()' function is looping over
VG_NUM_SOC_VOLTAGE_LEVELS (which is 8), but the size of the DcfClocks
array is VG_NUM_DCFCLK_DPM_LEVELS (which is 7).
When the loop variable i reaches 7, the function tries to access
clock_table->DcfClocks[7]. However, since the size of the DcfClocks
array is 7, the valid indices are 0 to 6. Index 7 is beyond the size of
the array, leading to a buffer overflow.
Reported by smatch & thus fixing the below:
drivers/gpu/drm/amd/amdgpu/../display/dc/clk_mgr/dcn301/vg_clk_mgr.c:550 find_dcfclk_for_voltage() error: buffer overflow 'clock_table->DcfClocks' 7 <= 7
Fixes: 3a83e4e64bb1 ("drm/amd/display: Add dcn3.01 support to DC (v2)") Cc: Roman Li <Roman.Li@amd.com> Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Cc: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Roman Li <roman.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
If 'data->fbc_en[i]' is not equal to 1 for any i, max_chunks_fbc_mode
will not be initialized if it's used outside of this for loop.
Ensure that 'max_chunks_fbc_mode' is properly initialized before it's
used. Initialize it to a default value right after its declaration to
ensure that it gets a value assigned under all possible control flow
paths.
Thus fixing the below:
drivers/gpu/drm/amd/amdgpu/../display/dc/basics/dce_calcs.c:914 calculate_bandwidth() error: uninitialized symbol 'max_chunks_fbc_mode'.
drivers/gpu/drm/amd/amdgpu/../display/dc/basics/dce_calcs.c:917 calculate_bandwidth() error: uninitialized symbol 'max_chunks_fbc_mode'.
Fixes: 4562236b3bc0 ("drm/amd/dc: Add dc display driver (v2)") Cc: Harry Wentland <harry.wentland@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Cc: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Roman Li <roman.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Above line is trying to assign the maximum value between
'wait_time_microsec' and 'DPIA_CLK_SYNC_DELAY' to wait_time_microsec.
However, 'wait_time_microsec' has not been assigned a value before this
line, initialize 'wait_time_microsec' at the point of declaration.
Fixes the below:
drivers/gpu/drm/amd/amdgpu/../display/dc/link/protocols/link_dp_training_dpia.c:697 dpia_training_eq_non_transparent() error: uninitialized symbol 'wait_time_microsec'.
Fixes: 630168a97314 ("drm/amd/display: move dp link training logic to link_dp_training") Cc: Wenjing Liu <wenjing.liu@amd.com> Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Cc: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Roman Li <roman.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Dan Carpenter [Fri, 9 Feb 2024 13:02:42 +0000 (16:02 +0300)]
drm/amd/display: Fix && vs || typos
These ANDs should be ORs or it will lead to a NULL dereference.
Fixes: fb5a3d037082 ("drm/amd/display: Add NULL test for 'timing generator' in 'dcn21_set_pipe()'") Fixes: 886571d217d7 ("drm/amd/display: Fix 'panel_cntl' could be null in 'dcn21_set_backlight_level()'") Reviewed-by: Anthony Koo <anthony.koo@amd.com> Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Kent Russell [Tue, 6 Feb 2024 17:45:44 +0000 (12:45 -0500)]
drm/amdkfd: Fix L2 cache size reporting in GFX9.4.3
Its currently incorrectly multiplied by number of XCCs in the partition
Fixes: be457b2252b6 ("drm/amdkfd: Update cache info for GFX 9.4.3") Signed-off-by: Kent Russell <kent.russell@amd.com> Reviewed-by: Mukul Joshi <mukul.joshi@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Hamza Mahfooz [Thu, 8 Feb 2024 21:23:29 +0000 (16:23 -0500)]
drm/amdgpu: make damage clips support configurable
We have observed that there are quite a number of PSR-SU panels on the
market that are unable to keep up with what user space throws at them,
resulting in hangs and random black screens. So, make damage clips
support configurable and disable it by default for PSR-SU displays.
Cc: stable@vger.kernel.org Reviewed-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Linus Torvalds [Thu, 15 Feb 2024 19:14:33 +0000 (11:14 -0800)]
update workarounds for gcc "asm goto" issue
In commit 4356e9f841f7 ("work around gcc bugs with 'asm goto' with
outputs") I did the gcc workaround unconditionally, because the cause of
the bad code generation wasn't entirely clear.
In the meantime, Jakub Jelinek debugged the issue, and has come up with
a fix in gcc [2], which also got backported to the still maintained
branches of gcc-11, gcc-12 and gcc-13.
Note that while the fix technically wasn't in the original gcc-14
branch, Jakub says:
"while it is true that no GCC 14 snapshots until today (or whenever the
fix will be committed) have the fix, for GCC trunk it is up to the
distros to use the latest snapshot if they use it at all and would
allow better testing of the kernel code without the workaround, so
that if there are other issues they won't be discovered years later.
Most userland code doesn't actually use asm goto with outputs..."
so we will consider gcc-14 to be fixed - if somebody is using gcc
snapshots of the gcc-14 before the fix, they should upgrade.
Note that while the bug goes back to gcc-11, in practice other gcc
changes seem to have effectively hidden it since gcc-12.1 as per a
bisect by Jakub. So even a gcc-14 snapshot without the fix likely
doesn't show actual problems.
Also, make the default 'asm_goto_output()' macro mark the asm as
volatile by hand, because of an unrelated gcc issue [1] where it doesn't
match the documented behavior ("asm goto is always volatile").
Linus Torvalds [Thu, 15 Feb 2024 18:19:55 +0000 (10:19 -0800)]
Merge tag 'devicetree-fixes-for-6.8-1' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux
Pull devicetree fixes from Rob Herring:
- Improve devlink dependency parsing for DT graphs
- Fix devlink handling of io-channels dependencies
- Fix PCI addressing in marvell,prestera example
- A few schema fixes for property constraints
- Improve performance of DT unprobed devices kselftest
- Fix regression in DT_SCHEMA_FILES handling
- Fix compile error in unittest for !OF_DYNAMIC
* tag 'devicetree-fixes-for-6.8-1' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux:
dt-bindings: ufs: samsung,exynos-ufs: Add size constraints on "samsung,sysreg"
of: property: Add in-ports/out-ports support to of_graph_get_port_parent()
of: property: Improve finding the supplier of a remote-endpoint property
of: property: Improve finding the consumer of a remote-endpoint property
net: marvell,prestera: Fix example PCI bus addressing
of: unittest: Fix compile in the non-dynamic case
of: property: fix typo in io-channels
dt-bindings: tpm: Drop type from "resets"
dt-bindings: display: nxp,tda998x: Fix 'audio-ports' constraints
dt-bindings: xilinx: replace Piyush Mehta maintainership
kselftest: dt: Stop relying on dirname to improve performance
dt-bindings: don't anchor DT_SCHEMA_FILES to bindings directory
Andy Shevchenko [Thu, 15 Feb 2024 15:25:06 +0000 (17:25 +0200)]
seq_buf: Fix kernel documentation
There are plenty of issues with the kernel documentation here:
- misspelled word "sequence"
- different style of returned value descriptions
- missed Return sections
- unaligned style of ASCII / NUL-terminated / etc
- wrong function references
projects / users / willy / xarray.git / log