Resolve all kernel-doc warnings in thermal_core.h:
Warning: drivers/thermal/thermal_core.h:99 bad line: trip point.
Warning: drivers/thermal/thermal_core.h:101 bad line: passive trip point.
Warning: drivers/thermal/thermal_core.h:152 struct member 'trips_attribute_group' not described in 'thermal_zone_device'
Warning: drivers/thermal/thermal_core.h:152 struct member 'debugfs' not described in 'thermal_zone_device'
Warning: drivers/thermal/thermal_core.h:152 struct member 'user_thresholds' not described in 'thermal_zone_device'
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Link: https://patch.msgid.link/20260128062446.402175-1-rdunlap@infradead.org
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
The device object rescan in acpi_scan_clear_dep_fn() is scheduled on a
system workqueue which is not guaranteed to be finished before entering
userspace. This may cause some key devices to be missing when userspace
init task tries to find them. Two issues observed on RISCV platforms:
- Kernel panic due to userspace init cannot have an opened
console.
The console device scanning is queued by acpi_scan_clear_dep_queue()
and not finished by the time userspace init process running, thus by
the time userspace init runs, no console is present.
- Entering rescue shell due to the lack of root devices (PCIe nvme in
our case).
Same reason as above, the PCIe host bridge scanning is queued on
a system workqueue and finished after init process runs.
The reason is because both devices (console, PCIe host bridge) depend on
riscv-aplic irqchip to serve their interrupts (console's wired interrupt
and PCI's INTx interrupts). In order to keep the dependency, these
devices are scanned and created after initializing riscv-aplic. The
riscv-aplic is initialized in device_initcall() and a device scan work
is queued via acpi_scan_clear_dep_queue(), which is close to the time
userspace init process is run. Since system_dfl_wq is used in
acpi_scan_clear_dep_queue() with no synchronization, the issues will
happen if userspace init runs before these devices are ready.
The solution is to wait for the queued work to complete before entering
userspace init. One possible way would be to use a dedicated workqueue
instead of system_dfl_wq, and explicitly flush it somewhere in the
initcall stage before entering userspace. Another way is to use
async_schedule_dev_nocall() for scanning these devices. It's designed
for asynchronous initialization and will work in the same way as before
because it's using a dedicated unbound workqueue as well, but the kernel
init code calls async_synchronize_full() right before entering userspace
init which will wait for the work to complete.
Compared to a dedicated workqueue, the second approach is simpler
because the async schedule framework takes care of all of the details.
The ACPI code only needs to focus on its job. A dedicated workqueue for
this could also be redundant because some platforms don't need
acpi_scan_clear_dep_queue() for their device scanning.
Signed-off-by: Yicong Yang <yang.yicong@picoheart.com>
[ rjw: Subject adjustment, changelog edits ]
Link: https://patch.msgid.link/20260128132848.93638-1-yang.yicong@picoheart.com
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Windows 11, version 22H2 introduced a new function index (Function 9) to
the Microsoft LPS0 _DSM, titled "Turn On Display Notification".
According to Microsoft documentation, this function signals to the system
firmware that the OS intends to turn on the display when exiting Modern
Standby. This allows the firmware to release Power Limits (PLx) earlier.
Crucially, this patch fixes a functional issue observed on the Lenovo Yoga
Slim 7i Aura (15ILL9), where system fans and keyboard backlights fail to
resume after suspend. Investigation linked shows the EC on this device
turns off these components during sleep but requires the Function 9
notification to wake them up again.
This patch defines the new function index (ACPI_MS_TURN_ON_DISPLAY) and
invokes it in acpi_s2idle_restore_early_lps0(). The execution order is
updated to match the logic of an "intent" signal:
1. LPS0 Exit (Function 6)
2. Turn On Display Intent (Function 9)
3. Modern Standby Exit (Function 8)
4. Screen On (Function 4)
Invoking Function 9 before the Modern Standby Exit ensures the firmware
has time to restore power rails and functionality (like fans) before the
software fully exits the sleep state.
Link: https://learn.microsoft.com/en-us/windows-hardware/design/device-experiences/modern-standby-firmware-notifications#turn-on-display-notification-function-9
Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220505
Suggested-by: Antheas Kapenekakis <antheas@antheas.dev>
Signed-off-by: Jakob Riemenschneider <riemenschneiderjakob@gmail.com>
Link: https://patch.msgid.link/20260127200121.1292216-1-riemenschneiderjakob@gmail.com
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Bindings expect 4 to be the default value for cdns,fifo-width. Said
otherwise, if the property (which is not mandatory) is not provided, the
OS, in order to comply with the bindings, should not error out and take
4 as default value.
Comply with the bindings. This would have slighlty simplyfied my testing
if it had been implemented correctly in the first place, but in practice
it should have no impact on the existing boards using this controller, as
they all set cdns,fifo-width to 4 explicitly in their upstream DTS.
Tested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Miquel Raynal (Schneider Electric) <miquel.raynal@bootlin.com>
Tested-by: Santhosh Kumar K <s-k6@ti.com>
Link: https://patch.msgid.link/20260122-schneider-6-19-rc1-qspi-v4-13-f9c21419a3e6@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Renesas RZ/N1 QSPI controllers embed the Cadence IP with some
modifications. For instance, they feature a write protection of the
direct mapping at the controller level, with this feature all data
writes to the AHB region are aborted.
Despite the fact that the flag setting write protection is disabled by
default, Bootloaders may (and actually do) set it, so mark this feature
as being available with a specific flag to, if applicable, make sure it
is disabled.
Tested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Miquel Raynal (Schneider Electric) <miquel.raynal@bootlin.com>
Tested-by: Santhosh Kumar K <s-k6@ti.com>
Link: https://patch.msgid.link/20260122-schneider-6-19-rc1-qspi-v4-12-f9c21419a3e6@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Renesas RZ/N1 QSPI controllers embed the Cadence IP with some
limitations/simplifications. One of the is that only direct access is
supported, none of the registers related to indirect writes are
populated, so create a flag to avoid these accesses and make sure only
direct accessors are called.
Tested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Miquel Raynal (Schneider Electric) <miquel.raynal@bootlin.com>
Tested-by: Santhosh Kumar K <s-k6@ti.com>
Link: https://patch.msgid.link/20260122-schneider-6-19-rc1-qspi-v4-11-f9c21419a3e6@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
In the remove path, we should try hard to perform all steps as we simply
cannot fail.
The "no runtime PM" quirk must only alter the state of the RPM core, but
the clocks should still be disabled if that is possible. Move the
disable call outside of the RPM quirk.
Tested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Miquel Raynal (Schneider Electric) <miquel.raynal@bootlin.com>
Tested-by: Santhosh Kumar K <s-k6@ti.com>
Link: https://patch.msgid.link/20260122-schneider-6-19-rc1-qspi-v4-9-f9c21419a3e6@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The probe has been modified by many different users, it is hard to track
history, but for sure its current state is partially broken. One easy
rule to follow is to drop/free/release the resources in the opposite
order they have been queried.
Fix the labels, the order for freeing the resources, and add the
missing DMA channel step. Replicate these changes in the remove path as
well.
Tested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Miquel Raynal (Schneider Electric) <miquel.raynal@bootlin.com>
Tested-by: Santhosh Kumar K <s-k6@ti.com>
Link: https://patch.msgid.link/20260122-schneider-6-19-rc1-qspi-v4-8-f9c21419a3e6@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The Cadence driver does not support anything else than repeating the
command opcode twice while in octal DTR mode. Make this clear by
checking for this in the ->supports_op() hook.
Reviewed-by: Pratyush Yadav <pratyush@kernel.org>
Tested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Miquel Raynal (Schneider Electric) <miquel.raynal@bootlin.com>
Tested-by: Santhosh Kumar K <s-k6@ti.com>
Link: https://patch.msgid.link/20260122-schneider-6-19-rc1-qspi-v4-7-f9c21419a3e6@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Right above writing the register value back based on 'div' value, there
is the following check:
if (div > CQSPI_REG_CONFIG_BAUD_MASK)
div = CQSPI_REG_CONFIG_BAUD_MASK;
which means div does not need to be AND'ed against the bitfield mask.
Remove this redundant operation.
Reviewed-by: Pratyush Yadav <pratyush@kernel.org>
Tested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Miquel Raynal (Schneider Electric) <miquel.raynal@bootlin.com>
Tested-by: Santhosh Kumar K <s-k6@ti.com>
Link: https://patch.msgid.link/20260122-schneider-6-19-rc1-qspi-v4-6-f9c21419a3e6@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
These definitions do not follow the standard patterns. Alignments are
incoherent and the logical OR symbols '|' are misplaced. Reorganize
these definitions.
There is no functional change.
Acked-by: Pratyush Yadav <pratyush@kernel.org>
Tested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Miquel Raynal (Schneider Electric) <miquel.raynal@bootlin.com>
Tested-by: Santhosh Kumar K <s-k6@ti.com>
Link: https://patch.msgid.link/20260122-schneider-6-19-rc1-qspi-v4-5-f9c21419a3e6@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
It took me several seconds to correctly understand this block. I
understand the goal: showing that we are in the if, or in one of the two
other cases. Improve the organization of the code to both improve
readability and fix the style.
Suggested-by: Pratyush Yadav <pratyush@kernel.org>
Tested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Miquel Raynal (Schneider Electric) <miquel.raynal@bootlin.com>
Tested-by: Santhosh Kumar K <s-k6@ti.com>
Link: https://patch.msgid.link/20260122-schneider-6-19-rc1-qspi-v4-4-f9c21419a3e6@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Fix alignment on the #defines.
Reviewed-by: Pratyush Yadav <pratyush@kernel.org>
Tested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Miquel Raynal (Schneider Electric) <miquel.raynal@bootlin.com>
Tested-by: Santhosh Kumar K <s-k6@ti.com>
Link: https://patch.msgid.link/20260122-schneider-6-19-rc1-qspi-v4-3-f9c21419a3e6@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The label is useless here. Plus, if there are several examples with the
same label, we'll get very useless yet annoying warnings.
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Miquel Raynal (Schneider Electric) <miquel.raynal@bootlin.com>
Tested-by: Santhosh Kumar K <s-k6@ti.com>
Link: https://patch.msgid.link/20260122-schneider-6-19-rc1-qspi-v4-1-f9c21419a3e6@bootlin.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Let ghes_edac be the preferred driver to load on __ZX__ and _BYO_
systems by extending the platform detection list in ghes.c
Signed-off-by: Tony W Wang-oc <TonyWWang-oc@zhaoxin.com>
Tested-by: Lyle Li <LyleLi@zhaoxin.com>
Acked-by: Borislav Petkov (AMD) <bp@alien8.de>
[ rjw: Subject and changelog edits ]
Link: https://patch.msgid.link/20260128025216.12564-1-TonyWWang-oc@zhaoxin.com
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
After a recent innocuous change to drivers/acpi/apei/ghes.c, building
ARCH=arm64 allmodconfig with clang-17 or older (which has both
CONFIG_KASAN=y and CONFIG_WERROR=y) fails with:
drivers/acpi/apei/ghes.c:902:13: error: stack frame size (2768) exceeds limit (2048) in 'ghes_do_proc' [-Werror,-Wframe-larger-than]
902 | static void ghes_do_proc(struct ghes *ghes,
| ^
A KASAN pass that removes unneeded stack instrumentation, enabled by
default in clang-18 [1], drastically improves stack usage in this case.
To avoid the warning in the common allmodconfig case when it can break
the build, disable KASAN for ghes.o when compile testing with clang-17
and older. Disabling KASAN outright may hide legitimate runtime issues,
so live with the warning in that case; the user can either increase the
frame warning limit or disable -Werror, which they should probably do
when debugging with KASAN anyways.
Closes: https://github.com/ClangBuiltLinux/linux/issues/2148
Link: 51fbab1345 [1]
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Cc: All applicable <stable@vger.kernel.org>
Link: https://patch.msgid.link/20260114-ghes-avoid-wflt-clang-older-than-18-v1-1-9c8248bfe4f4@kernel.org
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
With multiple virtio crypto devices supported with different NUMA
nodes, when crypto session is created, it will search virtio crypto
device with the same numa node of current CPU.
Here API topology_physical_package_id() is replaced with cpu_to_node()
since package id is physical concept, and one package id have multiple
memory numa id.
Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260113030556.3522533-4-maobibo@loongson.cn>
With function virtio_crypto_skcipher_crypt_req(), there is already
virtqueue_kick() call with spinlock held in function
__virtio_crypto_skcipher_do_req(). Remove duplicated virtqueue_kick()
function call here.
Fixes: d79b5d0bbf ("crypto: virtio - support crypto engine framework")
Cc: stable@vger.kernel.org
Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260113030556.3522533-3-maobibo@loongson.cn>
When VM boots with one virtio-crypto PCI device and builtin backend,
run openssl benchmark command with multiple processes, such as
openssl speed -evp aes-128-cbc -engine afalg -seconds 10 -multi 32
openssl processes will hangup and there is error reported like this:
virtio_crypto virtio0: dataq.0:id 3 is not a head!
It seems that the data virtqueue need protection when it is handled
for virtio done notification. If the spinlock protection is added
in virtcrypto_done_task(), openssl benchmark with multiple processes
works well.
Fixes: fed93fb62e ("crypto: virtio - Handle dataq logic with tasklet")
Cc: stable@vger.kernel.org
Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260113030556.3522533-2-maobibo@loongson.cn>
Address Space IDs allows the VDUSE framework to support devices able to
expose different virtqueues to different part of the drivers. For
example, to let QEMU handle the net device control virtqueue, so QEMU
always knows the state of the device like mac address or number of
queues enabled, while leaving the dataplane passthrough to the guest
intact. This enables live migration.
Expands the VDUSE documentation to explain how to use the new ioctls or
the new struct members of old ioctls.
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260119143306.1818855-14-eperezma@redhat.com>
Finalize the series by advertising VDUSE API v1 support to userspace.
Now that all required infrastructure for v1 (ASIDs, VQ groups,
update_iotlb_v2) is in place, VDUSE devices can opt in to the new
features.
Assume API version 0 if the VDUSE instance does not call
VDUSE_GET_API_VERSION to maintain compatibility.
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260119143306.1818855-13-eperezma@redhat.com>
Add support for assigning Address Space Identifiers (ASIDs) to each VQ
group. This enables mapping each group into a distinct memory space.
The vq group to ASID association is protected by a rwlock now. But the
mutex domain_lock keeps protecting the domains of all ASIDs, as some
operations like the one related with the bounce buffer size still
requires to lock all the ASIDs.
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260119143306.1818855-12-eperezma@redhat.com>
The next patch adds new ioctl with the ASID member per entry. Abstract
these two so it can be build on top easily.
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260119143306.1818855-11-eperezma@redhat.com>
The function vduse_dev_alloc_coherent will be called under rwlock in
next patches. Make it out of the lock to avoid increasing its fail
rate.
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260119143306.1818855-10-eperezma@redhat.com>
We will modify the function in next patches so let's clean it first.
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260119143306.1818855-9-eperezma@redhat.com>
Next patches introduce more error paths in this function. Refactor it
so they can be accommodated through gotos.
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Xie Yongji <xieyongji@bytedance.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260119143306.1818855-8-eperezma@redhat.com>
Only vdpa_sim support it. Forbid this behaviour as there is no use for
it right now, we can always enable it in the future with a feature flag.
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260119143306.1818855-7-eperezma@redhat.com>
Document that the function races with the check of DRIVER_OK.
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260119143306.1818855-6-eperezma@redhat.com>
Return the internal struct that represents the vq group as virtqueue map
token, instead of the device. This allows the map functions to access
the information per group.
At this moment all the virtqueues share the same vq group, that only
can point to ASID 0. This change prepares the infrastructure for actual
per-group address space handling
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260119143306.1818855-5-eperezma@redhat.com>
This allows separate the different virtqueues in groups that shares the
same address space. Asking the VDUSE device for the groups of the vq at
the beginning as they're needed for the DMA API.
Allocating 3 vq groups as net is the device that need the most groups:
* Dataplane (guest passthrough)
* CVQ
* Shadowed vrings.
Future versions of the series can include dynamic allocation of the
groups array so VDUSE can declare more groups.
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Xie Yongji <xieyongji@bytedance.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260119143306.1818855-4-eperezma@redhat.com>
This allows the kernel to detect whether the userspace VDUSE device
supports the VQ group and ASID features. VDUSE devices that don't set
the V1 API will not receive the new messages, and vdpa device will be
created with only one vq group and asid.
The next patches implement the new feature incrementally, only enabling
the VDUSE device to set the V1 API version by the end of the series.
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Xie Yongji <xieyongji@bytedance.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260119143306.1818855-3-eperezma@redhat.com>
Remove duplication by consolidating these here. This reduces the
posibility of a parent driver missing them.
While we're at it, fix a bug in vdpa_sim where a valid ASID can be
assigned to a group equal to ngroups, causing an out of bound write.
Cc: stable@vger.kernel.org
Fixes: bda324fd03 ("vdpasim: control virtqueue support")
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20260119143306.1818855-2-eperezma@redhat.com>
Currently, cacheline group macros trigger checkpatch warnings.
For example:
$ ./scripts/checkpatch.pl -g ba7e025a6c84aed012421468d83639e5dae982b0
WARNING: Missing a blank line after declarations
#58: FILE: drivers/gpio/gpio-virtio.c:32:
+ struct virtio_gpio_response res;
+ __dma_from_device_group_end();
$ ./scripts/checkpatch.pl -g 5d4cc87414
WARNING: Missing a blank line after declarations
#267: FILE: include/net/sock.h:431:
+ int sk_rcvlowat;
+ __cacheline_group_end(sock_read_rx);
But these are not actually statements - the following macros
all expand to zero-length fields:
__cacheline_group_begin()
__cacheline_group_end()
__cacheline_group_begin_aligned()
__cacheline_group_end_aligned()
__dma_from_device_group_begin()
__dma_from_device_group_end()
Add them to $declaration_macros so checkpatch recognizes this fact.
Message-ID: <b345bb7e2d4e23672e3e5d1b283754dc11c7d8cd.1767647872.git.mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reorder struct virtio_gpio_line fields to place the DMA buffers
(req/res) last.
This eliminates the padding from aligning struct size on
ARCH_DMA_MINALIGN.
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Message-ID: <f1221bbc120df6adaba9006710a517f1e84a10b2.1767601130.git.mst@redhat.com>
Acked-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The res and ires buffers in struct virtio_gpio_line and struct
vgpio_irq_line respectively are used for DMA_FROM_DEVICE via
virtqueue_add_sgs(). However, within these structs, even though these
elements are tagged as ____cacheline_aligned, adjacent struct elements
can share DMA cachelines on platforms where ARCH_DMA_MINALIGN >
L1_CACHE_BYTES (e.g., arm64 with 128-byte DMA alignment but 64-byte
cache lines).
The existing ____cacheline_aligned annotation aligns to L1_CACHE_BYTES
which is not always sufficient for DMA alignment. For example, with
L1_CACHE_BYTES = 32 and ARCH_DMA_MINALIGN = 128
- irq_lines[0].ires at offset 128
- irq_lines[1].type at offset 192
both in same 128-byte DMA cacheline [128-256)
When the device writes to irq_lines[0].ires and the CPU concurrently
modifies one of irq_lines[1].type/disabled/masked/queued flags,
corruption can occur on non-cache-coherent platforms.
Fix by using __dma_from_device_group_begin()/end() annotations on the
DMA buffers. Drop ____cacheline_aligned - it's not required to isolate
request and response, and keeping them would increase the memory cost.
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Message-ID: <ba7e025a6c84aed012421468d83639e5dae982b0.1767601130.git.mst@redhat.com>
Acked-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reorder struct virtio_vsock fields to place the DMA buffer (event_list)
last. This eliminates the padding from aligning the struct size on
ARCH_DMA_MINALIGN.
Message-ID: <ce44f61af415521e00ab7492aa16d3d19f00bd5e.1769632071.git.mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
'struct qcom_cc_desc' is passed to qcom_cc_map() and
qcom_cc_really_probe() only as pointer to const, so make the memory
const for safety.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Reviewed-by: Taniya Das <taniya.das@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20260128154306.133047-2-krzysztof.kozlowski@oss.qualcomm.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
[Why&How]
On amdgpu_dm_connector_destroy(), the driver attempts to cancel pending
HDMI HPD work without checking if the HDMI HPD is enabled.
Added a check that it is enabled before clearing it.
Fixes: 6a681cd903 ("drm/amd/display: Add an hdmi_hpd_debounce_delay_ms module")
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 17b2c526fd)
Remove __assume_ctx_lock() from lock initializers.
Implicitly asserting an active context during initialization caused
false-positive double-lock errors when acquiring a lock immediately after its
initialization. Moving forward, guarded member initialization must either:
1. Use guard(type_init)(&lock) or scoped_guard(type_init, ...).
2. Use context_unsafe() for simple initialization.
Reported-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/all/57062131-e79e-42c2-aa0b-8f931cb8cac2@acm.org/
Link: https://patch.msgid.link/20260119094029.1344361-7-elver@google.com
Convert lock initialization to scoped guarded initialization where
lock-guarded members are initialized in the same scope.
This ensures the context analysis treats the context as active during member
initialization. This is required to avoid errors once implicit context
assertion is removed.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260119094029.1344361-6-elver@google.com
Convert lock initialization to scoped guarded initialization where
lock-guarded members are initialized in the same scope.
This ensures the context analysis treats the context as active during member
initialization. This is required to avoid errors once implicit context
assertion is removed.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260119094029.1344361-5-elver@google.com
Convert lock initialization to scoped guarded initialization where
lock-guarded members are initialized in the same scope.
This ensures the context analysis treats the context as active during
member initialization. This is required to avoid errors once implicit
context assertion is removed.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260119094029.1344361-4-elver@google.com
Add scoped init guard definitions for common synchronization primitives
supported by context analysis.
The scoped init guards treat the context as active within initialization
scope of the underlying context lock, given initialization implies
exclusive access to the underlying object. This allows initialization of
guarded members without disabling context analysis, while documenting
initialization from subsequent usage.
The documentation is updated with the new recommendation. Where scoped
init guards are not provided or cannot be implemented (ww_mutex omitted
for lack of multi-arg guard initializers), the alternative is to just
disable context analysis where guarded members are initialized.
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/all/20251212095943.GM3911114@noisy.programming.kicks-ass.net/
Link: https://patch.msgid.link/20260119094029.1344361-3-elver@google.com
Jiri Olsa says:
====================
ftrace,bpf: Use single direct ops for bpf trampolines
hi,
while poking the multi-tracing interface I ended up with just one ftrace_ops
object to attach all trampolines.
This change allows to use less direct API calls during the attachment changes
in the future code, so in effect speeding up the attachment.
In current code we get a speed up from using just a single ftrace_ops object.
- with current code:
Performance counter stats for 'bpftrace -e fentry:vmlinux:ksys_* {} -c true':
6,364,157,902 cycles:k
828,728,902 cycles:u
1,064,803,824 instructions:u # 1.28 insn per cycle
23,797,500,067 instructions:k # 3.74 insn per cycle
4.416004987 seconds time elapsed
0.164121000 seconds user
1.289550000 seconds sys
- with the fix:
Performance counter stats for 'bpftrace -e fentry:vmlinux:ksys_* {} -c true':
6,535,857,905 cycles:k
810,809,429 cycles:u
1,064,594,027 instructions:u # 1.31 insn per cycle
23,962,552,894 instructions:k # 3.67 insn per cycle
1.666961239 seconds time elapsed
0.157412000 seconds user
1.283396000 seconds sys
The speedup seems to be related to the fact that with single ftrace_ops object
we don't call ftrace_shutdown anymore (we use ftrace_update_ops instead) and
we skip the synchronize rcu calls (each ~100ms) at the end of that function.
rfc: https://lore.kernel.org/bpf/20250729102813.1531457-1-jolsa@kernel.org/
v1: https://lore.kernel.org/bpf/20250923215147.1571952-1-jolsa@kernel.org/
v2: https://lore.kernel.org/bpf/20251113123750.2507435-1-jolsa@kernel.org/
v3: https://lore.kernel.org/bpf/20251120212402.466524-1-jolsa@kernel.org/
v4: https://lore.kernel.org/bpf/20251203082402.78816-1-jolsa@kernel.org/
v5: https://lore.kernel.org/bpf/20251215211402.353056-10-jolsa@kernel.org/
v6 changes:
- rename add_hash_entry_direct to add_ftrace_hash_entry_direct [Steven]
- factor hash_add/hash_sub [Steven]
- add kerneldoc header for update_ftrace_direct_* functions [Steven]
- few assorted smaller fixes [Steven]
- added missing direct_ops wrappers for !CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS
case [Steven]
v5 changes:
- do not export ftrace_hash object [Steven]
- fix update_ftrace_direct_add new_filter_hash leak [ci]
v4 changes:
- rebased on top of bpf-next/master (with jmp attach changes)
added patch 1 to deal with that
- added extra checks for update_ftrace_direct_del/mod to address
the ci bot review
v3 changes:
- rebased on top of bpf-next/master
- fixed update_ftrace_direct_del cleanup path
- added missing inline to update_ftrace_direct_* stubs
v2 changes:
- rebased on top fo bpf-next/master plus Song's livepatch fixes [1]
- renamed the API functions [2] [Steven]
- do not export the new api [Steven]
- kept the original direct interface:
I'm not sure if we want to melt both *_ftrace_direct and the new interface
into single one. It's bit different in semantic (hence the name change as
Steven suggested [2]) and I don't think the changes are not that big so
we could easily keep both APIs.
v1 changes:
- make the change x86 specific, after discussing with Mark options for
arm64 [Mark]
thanks,
jirka
[1] https://lore.kernel.org/bpf/20251027175023.1521602-1-song@kernel.org/
[2] https://lore.kernel.org/bpf/20250924050415.4aefcb91@batman.local.home/
---
====================
Link: https://patch.msgid.link/20251230145010.103439-1-jolsa@kernel.org
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Using single ftrace_ops for direct calls update instead of allocating
ftrace_ops object for each trampoline.
With single ftrace_ops object we can use update_ftrace_direct_* api
that allows multiple ip sites updates on single ftrace_ops object.
Adding HAVE_SINGLE_FTRACE_DIRECT_OPS config option to be enabled on
each arch that supports this.
At the moment we can enable this only on x86 arch, because arm relies
on ftrace_ops object representing just single trampoline image (stored
in ftrace_ops::direct_call). Archs that do not support this will continue
to use *_ftrace_direct api.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Link: https://lore.kernel.org/bpf/20251230145010.103439-10-jolsa@kernel.org
We are going to remove "ftrace_ops->private == bpf_trampoline" setup
in following changes.
Adding ip argument to ftrace_ops_func_t callback function, so we can
use it to look up the trampoline.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Link: https://lore.kernel.org/bpf/20251230145010.103439-9-jolsa@kernel.org
Adding update_ftrace_direct_mod function that modifies all entries
(ip -> direct) provided in hash argument to direct ftrace ops and
updates its attachments.
The difference to current modify_ftrace_direct is:
- hash argument that allows to modify multiple ip -> direct
entries at once
This change will allow us to have simple ftrace_ops for all bpf
direct interface users in following changes.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Link: https://lore.kernel.org/bpf/20251230145010.103439-7-jolsa@kernel.org
Adding update_ftrace_direct_del function that removes all entries
(ip -> addr) provided in hash argument to direct ftrace ops and
updates its attachments.
The difference to current unregister_ftrace_direct is
- hash argument that allows to unregister multiple ip -> direct
entries at once
- we can call update_ftrace_direct_del multiple times on the
same ftrace_ops object, becase we do not need to unregister
all entries at once, we can do it gradualy with the help of
ftrace_update_ops function
This change will allow us to have simple ftrace_ops for all bpf
direct interface users in following changes.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Link: https://lore.kernel.org/bpf/20251230145010.103439-6-jolsa@kernel.org
