No need to bother the higher level pinning code with the
FORCE_WC assert. Move it into the lower level function.
This also introduces the check to intel_fb_pin_to_ggtt(), which
(for the moment) is still used for the fbdev bo setup. But this
is all display stuff we're talking about here so having the check
is correct everywhere.
v2: Pimp the commit message a bit (Jani)
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patch.msgid.link/20260508143426.26504-8-ville.syrjala@linux.intel.com
Extract the inner DPT parts of intel_plane_(un)pin() into the
i915_fb_pin_ggtt_(un)pin(). These will become part of the new
fb_pin parent interface.
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patch.msgid.link/20260508143426.26504-7-ville.syrjala@linux.intel.com
Extract the inner DPT parts of intel_plane_(un)pin() into the
i915_fb_pin_dpt_(un)pin(). These will become part of the new
fb_pin parent interface.
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patch.msgid.link/20260508143426.26504-6-ville.syrjala@linux.intel.com
Move most of the plane state stuff out from the inner parts
of intel_plane_pin_fb(). The plan is to take those inner parts and
abstract them into the new fb_pin parent interface, and we don't
want any plane_state stuff there.
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patch.msgid.link/20260508143426.26504-5-ville.syrjala@linux.intel.com
Move the i915_dpt_offset() check into the lower level
intel_fb_pin_to_dpt() function. Clears out some of the unnecessary
junk from the higher level code, making it easier to introduce
the new fb_pin parent interface.
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patch.msgid.link/20260508143426.26504-4-ville.syrjala@linux.intel.com
strut intel_fb_pin_params will be an important part of the fb_pin
interface, so move the definition to the parent interface file.
Or maybe we should have a separate header for this kind of stuff
since the users of the parent interface will need the struct
definition but not the parent interface vfunc struct definitions?
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patch.msgid.link/20260508143426.26504-3-ville.syrjala@linux.intel.com
Introduce the "fb_pin" parent interface, as the first trivial step
move the *_get_map() stuff there.
The whole "fb_pin" as an interface might not really make sense,
and perhaps this (and other stuff) should just be collected into
some kind of "bo" interface. But let's go with "fb_pin" for now
to match where things are implemented, and possibly restructure
it later.
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patch.msgid.link/20260508143426.26504-2-ville.syrjala@linux.intel.com
Problem determination using s390dbf logging sometimes requires changing
the default logging level or log area size. While this is possible
using sysfs interfaces, there is no easy way to adjust these parameters
for early boot code that emits logs before userspace is available.
Add an s390dbf kernel parameter to address this shortcoming. The
parameter can be used to specify log level and area size (in units of
pages). A level of '-' turns logging off for an area. Logs can be
identified by name or a shell-style pattern.
Parameter format:
s390dbf=<name|pattern>:[<level>|-]:[<pages>][,...]
Example:
s390dbf=cio*:6:128,sclp_err::2
Specified parameters are applied immediately during debug area
registration for regular log areas. For early, static debug areas,
log levels are changed during early_param() parsing, while size
changes are applied at arch_initcall-time.
Signed-off-by: Peter Oberparleiter <oberpar@linux.ibm.com>
Tested-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Notify userspace about two important events on AP queues
when run within Secure Execution (SE) environment:
- Send AP CHANGE uevent with "SE_BIND=1" on successful bind
operation on this AP queue device.
- Send AP CHANGE uevent with "SE_ASSOC=<association_index>"
on successful association operation with the secret of the
reported index on this AP queue device.
Note there is no SE unbind/unassociate event. Unbind/unassociate
can have different triggers and technically there is no signaling
done which the AP code could catch. A user space application can,
if this information is crucial, query the sysfs attribute se_bind
on the AP queue which runs a synchronous TAPQ. If the attribute
returns with "unbound" a reset took place and SE bind and associate
states are unbound and unassociated.
Suggested-by: Marc Hartmayer mhartmay@linux.ibm.com
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Reviewed-by: Holger Dengler <dengler@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
/proc/irq/<n>/ directory names are built in `char name[10]` buffers
with `sprintf(name, "%u", irq)`.
Ten-digit IRQ numbers already need 11 bytes including the trailing NUL, and
current sparse-IRQ configurations allow interrupt numbers in that range.
Size the temporary name buffer for the current decimal form and switch
to bounded formatting when creating or removing the proc entry.
Signed-off-by: Pengpeng Hou <pengpeng@iscas.ac.cn>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260404101001.1-genirq-proc-pengpeng@iscas.ac.cn
check_mft_mirror() still computes the number of bytes to validate in each
mirrored MFT record, but the actual comparison against $MFTMirr was dropped
when the superblock code was updated.
As a result, mount misses a stale or inconsistent $MFTMirr as long as both
records pass the structural baad-record checks. Restore the comparison and
log an error when the primary $MFT record differs from its mirror copy.
Returning false lets the existing mount error handling mark the volume as
having NTFS errors and, with on_errors=remount-ro, continue read-only. The
default on_errors=continue mount policy still allows the mount to proceed.
Fixes: 6251f0b0de ("ntfs: update super block operations")
Signed-off-by: DaeMyung Kang <charsyam@gmail.com>
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
On PREEMPT_RT, non-HARD irq_work runs in per-CPU kthreads via
run_irq_workd(), so irq_work_sync() uses rcuwait() to wait for BUSY==0.
After irq_work_single() clears BUSY via atomic_cmpxchg(), it still
dereferences @work for irq_work_is_hard() and rcuwait_wake_up().
An irq_work_sync() caller on another CPU that enters after BUSY is cleared
can observe BUSY==0 immediately, return, and free the work before those
accesses complete — causing a use-after-free.
Fix this by wrapping run_irq_workd() in guard(rcu)() so that the entire
irq_work_single() execution is within an RCU read-side critical
section. Then add synchronize_rcu() in irq_work_sync() after
rcuwait_wait_event() to ensure the caller waits for the RCU grace period
before returning, preventing premature frees.
Fixes: 810979682c ("irq_work: Allow irq_work_sync() to sleep if irq_work() no IRQ support.")
Suggested-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Suggested-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Jiayuan Chen <jiayuan.chen@linux.dev>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Reviewed-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20260330073234.303732-1-jiayuan.chen@linux.dev
Exynos 7870 compatible controllers, such as SDIO ones are not able to
perform DMA transfers for small sizes of data (~16 to ~512 bytes),
resulting in cache issues in subsequent transfers. Increase the DMA
transfer threshold to 512 to allow the shorter transfers to take place,
bypassing DMA.
Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Reviewed-by: Shawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Some controllers, such as certain Exynos SDIO ones, are unable to
perform DMA transfers of small amount of bytes properly. Following the
device tree schema, implement the property to define the DMA transfer
threshold (from a hard coded value of 16 bytes) so that lesser number of
bytes can be transferred safely skipping DMA in such controllers. The
value of 16 bytes stays as the default for controllers which do not
define it. This value can be overridden by implementation-specific init
sequences.
Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Reviewed-by: Shawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
While being more verbose using a named initializer yields easier to
understand code and doesn't rely on the two hidden zeros in the
PCI_VDEVICE macro.
This doesn't introduce any changes to the compiled result of the array,
which was confirmed with an ARCH=x86 build.
Signed-off-by: Uwe Kleine-König (The Capable Hub) <u.kleine-koenig@baylibre.com>
Link: https://patch.msgid.link/20260507155025.3186956-2-u.kleine-koenig@baylibre.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Use the existing zenbook duo keyboard quirk for the UX8407AA model too.
Signed-off-by: Paolo Pisati <p.pisati@gmail.com>
Reviewed-by: Denis Benato <denis.benato@linux.dev>
Link: https://patch.msgid.link/20260508070956.62201-1-p.pisati@gmail.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Since Qualcomm inline-crypto engine (ICE) is now a dedicated driver
de-coupled from the QCOM UFS driver, it explicitly votes for its required
clocks during probe. For scenarios where the 'clk_ignore_unused' flag is
not passed on the kernel command line, to avoid potential unclocked ICE
hardware register access during probe the ICE driver should additionally
vote on the 'iface' clock.
Also update the suspend and resume callbacks to handle un-voting and voting
on the 'iface' clock.
Fixes: 2afbf43a4a ("soc: qcom: Make the Qualcomm UFS/SDCC ICE a dedicated driver")
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
Reviewed-by: Kuldeep Singh <kuldeep.singh@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Harshal Dev <harshal.dev@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20260416-qcom_ice_power_and_clk_vote-v5-2-5ccf5d7e2846@oss.qualcomm.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
The watchdog on the Apple silicon t8122 (M3) SoC is compatible with the
existing driver. Add "apple,t8122-wdt" as SoC specific compatible under
"apple,t8103-wdt" used by the driver.
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Reviewed-by: Joshua Peisach <jpeisach@ubuntu.com>
Reviewed-by: Neal Gompa <neal@gompa.dev>
Signed-off-by: Janne Grunau <j@jannau.net>
Link: https://lore.kernel.org/r/20260507-apple-m3-initial-devicetrees-v3-2-ca07c81b5dc7@jannau.net
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Introduce a CPUID parser for populating the system's CPUID tables.
Since accessing a leaf within the CPUID table requires compile time
tokenization, split the parser into two stages:
(a) Compile-time macros for tokenizing the leaf/subleaf offsets within
the CPUID table.
(b) Generic runtime code to fill the CPUID data, using a parsing table
which collects these compile-time offsets.
For actual CPUID output parsing, support both generic and leaf-specific
read functions.
To ensure CPUID data early availability, invoke the parser during early
boot, early Xen boot, and at early secondary CPUs bring up.
Provide call site APIs to refresh a single leaf, or a leaf range, within
the CPUID tables. This is for sites issuing MSR writes that partially
change the CPU's CPUID layout. Doing full CPUID table rescans in such
cases will be destructive since the CPUID tables will host all of the
kernel's X86_FEATURE flags at a later stage.
Suggested-by: Thomas Gleixner <tglx@kernel.org>
Signed-off-by: Ahmed S. Darwish <darwi@linutronix.de>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/all/20260327021645.555257-1-darwi@linutronix.de
The DT bindings for inline-crypto engine do not specify the UFS_PHY_GDSC
power-domain and iface clock. Without enabling the iface clock and the
associated power-domain the ICE hardware cannot function correctly and
leads to unclocked hardware accesses being observed during probe.
Fix the DT bindings for inline-crypto engine to require the UFS_PHY_GDSC
power-domain and iface clock for new devices (Eliza and Milos) introduced
in the current release (7.1) with yet-to-stabilize ABI, while preserving
backward compatibility for older devices.
Fixes: 618195a7ac ("dt-bindings: crypto: qcom,inline-crypto-engine: Document the Eliza ICE")
Fixes: 85faec1e85 ("dt-bindings: crypto: qcom,inline-crypto-engine: document the Milos ICE")
Reviewed-by: Kuldeep Singh <kuldeep.singh@oss.qualcomm.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Signed-off-by: Harshal Dev <harshal.dev@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20260416-qcom_ice_power_and_clk_vote-v5-1-5ccf5d7e2846@oss.qualcomm.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
Machines with a larger number of CPUs and under heavy load sometimes
loose PAI counter increments during recording using events
-e CRYPTO_ÂLL or -e NNPA_ALL. Counting is not affected.
This happens when several PAI crypto counters are incremented during
the same cryptographic operation.
During schedule out the functions
paiXXX_sched_task() (with XXX either crypt or ext)
+--> pai_have_samples()
+--> pai_have_sample()
+--> pai_copy()
+--> pai_push_sample()
are called to read out PAI counter values.
In pai_copy() the current values of PAI counters are read from the
PMU memory mapped page and compared to the values read during last
schedule out operation, which have been saved in a backup page
named PAI_SAVE_AREA(event). For each PAI counter a delta is calculated
and when the delta is positive, that PAI counter was incremented by
hardware. This positve delta is reported as raw data record attached
to a sample.
After all deltas have been calculated, the new PAI counter values
are saved in the backup page PAI_SAVE_AREA(event). However this is
done in pai_push_sample(), leaving a small window for missing hardware
triggered updates. Here is one scenario:
PAI counter idx: 0 1 2 3 4 5 6 7 .... N
+---+---+---+---+---+---+---+---+ +---+
PAI counter page:| | | X | | | | | |....| Y |
+---+---+---+---+---+---+---+---+ +---+
In pai_copy() each PAI counter value is read and compared
to its old value. This is done in a loop. When PAI counter indexed
N is read, the hardware might increment PAI counter indexed 2 again,
updating its value from X to X+1.
Later pai_push_sample() simply mem-copies the complete PAI counter
page to a backup page and the increment of X+1 is lost, because the
backup page now contains the new value.
Read each PAI counter and save this value in the backup page when
there is a positive delta. This omits any time window between read
and store. This also reduced the work load as only modified PAI
counters are saved.
Cc: stable@vger.kernel.org
Fixes: fe861b0c8d ("s390/pai: save PAI counter value page in event structure")
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Reviewed-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
When executing NS_GET_PID_FROM_PIDNS (or similar pidns ioctls), if the
target task cannot be found in the corresponding pid_ns, the error code
should be ESRCH instead of ENOTTY.
This bug was introduced when the extensible ioctl handling was added.
Without proper return, ret would be overwritten by the default case in
the extensible ioctl switch statement.
Fixes: a1d220d9da ("nsfs: iterate through mount namespaces")
Signed-off-by: Zhihao Cheng <chengzhihao1@huawei.com>
Link: https://patch.msgid.link/20260507112301.1042757-1-chengzhihao1@huawei.com
Reviewed-by: Yang Erkun <yangerkun@huawei.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Add Embedded controller driver support for Hamoa/Purwa/Glymur qualcomm
reference boards. It handles fan control, temperature sensors, access
to EC state changes and supports reporting suspend entry/exit to the
EC.
Co-developed-by: Maya Matuszczyk <maccraft123mc@gmail.com>
Signed-off-by: Maya Matuszczyk <maccraft123mc@gmail.com>
Signed-off-by: Sibi Sankar <sibi.sankar@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Acked-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Tested-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Co-developed-by: Anvesh Jain P <anvesh.p@oss.qualcomm.com>
Signed-off-by: Anvesh Jain P <anvesh.p@oss.qualcomm.com>
Link: https://patch.msgid.link/20260511-add-driver-for-ec-v9-2-e5437c39b7f8@oss.qualcomm.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
blk_validate_limits() requires max_hw_sectors >= PAGE_SECTORS and fires
a WARN_ON_ONCE if this invariant is violated. ublk_validate_params()
only checked the upper bound of max_sectors against max_io_buf_bytes,
allowing userspace to pass small values (including zero) that trigger
the warning when blk_mq_alloc_disk() is called from
ublk_ctrl_start_dev().
Before 494ea040bc, ublk used blk_queue_max_hw_sectors() which silently
clamped small values up to PAGE_SECTORS. The conversion to passing
queue_limits directly to blk_mq_alloc_disk() lost that clamping and now
hits blk_validate_limits()'s WARN_ON_ONCE instead.
Validate that max_sectors is at least PAGE_SECTORS in
ublk_validate_params() so invalid values are rejected early with
-EINVAL instead of reaching the block layer.
Fixes: 494ea040bc ("ublk: pass queue_limits to blk_mq_alloc_disk")
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Link: https://patch.msgid.link/20260510144843.769031-1-tom.leiming@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
SQPOLL stores current->pid (init_pid_ns view) in sqd->task_pid
at thread creation. fdinfo prints it raw via
seq_printf("SqThread:\t%d\n", sq_pid). A reader inside a
non-initial pid_ns sees the host PID, not the kthread's PID in
the reader's own pid_ns.
The SQPOLL kthread is created with CLONE_THREAD and no
CLONE_NEW*, so it lives in the submitter's pid_ns. An
unprivileged user_ns + pid_ns submitter can read fdinfo and
learn the host PID of a kthread whose in-namespace PID is
different.
Reproducer (mainline 7.0, KASAN): unshare CLONE_NEWUSER |
CLONE_NEWPID | CLONE_NEWNS, mount a private /proc, then have a
grandchild that is pid 1 in the new pid_ns open an io_uring
ring with IORING_SETUP_SQPOLL. /proc/self/task lists {1, 2};
the SQPOLL kthread is pid 2. Before: fdinfo prints
SqThread = <host pid>. After: SqThread = 2.
Use task_pid_nr_ns() against the proc inode's pid_ns to compute
sq_pid, instead of reading the stored sq->task_pid (which holds
the init_pid_ns view). pidfd_show_fdinfo() in kernel/pid.c
follows the same pattern.
Signed-off-by: Maoyi Xie <maoyi.xie@ntu.edu.sg>
Link: https://patch.msgid.link/20260510084119.457578-1-maoyi.xie@ntu.edu.sg
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This patch prepares the iomap framework for exFAT's upcoming migration to
iomap. During testing of the exFAT iomap branch with xfstests generic/299 on
a VM with 8GB RAM and a 40GB disk, system unresponsiveness was observed.
iomap_zero_iter() lacked dirty page throttling, which could cause memory
pressure when exFAT's valid_size mechanism triggers large-scale zeroing
operations during writes beyond valid_size.
Align iomap_zero_iter() with iomap_write_iter() by adding
balance_dirty_pages_ratelimited() to throttle dirty page generation during
large zeroing operations
Signed-off-by: Chi Zhiling <chizhiling@kylinos.cn>
Link: https://patch.msgid.link/20260511094007.728011-1-chizhiling@163.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
When iomap_iter() finishes its iteration (returns <= 0), it is no longer
necessary to memset the entire iomap and srcmap structures.
In high-IOPS scenarios (like 4k randread NVMe polling with io_uring),
where the majority of I/Os complete in a single extent map, this wasted
memory write bandwidth, as the caller will just discard the iterator.
Use this command to test:
taskset -c 30 ./t/io_uring -p1 -d512 -b4096 -s32 -c32 -F1 -B1 -R1 -X1
-n1 -P1 /mnt/testfile
IOPS improve about 5% on ext4 and XFS.
However, we MUST still call iomap_iter_reset_iomap() to release the
folio_batch if IOMAP_F_FOLIO_BATCH is set, otherwise we leak page
references. Therefore, split the cleanup logic: always release the
folio_batch, but skip the memset() when ret <= 0.
Signed-off-by: Fengnan Chang <changfengnan@bytedance.com>
Link: https://patch.msgid.link/20260420061630.62077-1-changfengnan@bytedance.com
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christian Brauner <brauner@kernel.org>
When updating timestamps with lazytime enabled, if only I_DIRTY_TIME is
set (pure lazytime update), inode_maybe_inc_iversion() should not be
forced to increment i_version. The force parameter should only be true
when actual data or metadata changes require an iversion bump.
The current code uses "!!dirty" which evaluates to true whenever dirty
has any bits set, including the I_DIRTY_TIME bit alone. This forces an
iversion increment on every lazytime timestamp update, which then sets
I_DIRTY_SYNC, triggering expensive log flushes on subsequent fdatasync
calls. Andres reported this issue when he noticed a perf regression[1].
Fix this by using "dirty != I_DIRTY_TIME" as the force parameter. This
passes false for pure lazytime updates (allowing the I_VERSION_QUERIED
optimization to work), while still forcing the increment when dirty
contains other flags indicating real changes that require iversion
updates.
[1] https://lore.kernel.org/linux-xfs/7ys6erh3nnyeerv2nybyfvp7dmaknuxrlxv74wx56ocdothkc6@ekfiadtkfn2r/
Fixes: 85c871a02b ("fs: add support for non-blocking timestamp updates")
Signed-off-by: Pankaj Raghav <p.raghav@samsung.com>
Link: https://patch.msgid.link/20260511111918.1793689-1-p.raghav@samsung.com
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
The Vectored External Interrupt Controller mode present in the MIPS 34Kc
and 1004Kc variants causes the CPU to stop dispatching interrupts by the
normal code path and instead it sends those interrupts to the external
interrupt controller to be prioritized, renumbered, and sent back. When
they come back, they are handled through a different path using a dispatch
table, so plat_irq_dispatch() never sees action.
This of course subverts the traditional intc hierarchy, and on the 1004Kc
the interrupt controller is standardized (IRQ_GIC) so it can be reasonably
considered part of the CPU itself - and tighter coupling between IRQ_GIC
and arch/mips/* is tolerable. However on the 34Kc the intc is defined by
each SoC vendor, so it's required to have a modular driver - but for a
device which in fact ends up taking over the entire interrupt system.
Let the DT describe which IRQs which come from the CPU and should be
routed back and handled by the CPU intc. These particularly include the
two IPI interrupts which would otherwise necessitate duplication of all
the IPI supporting infrastructure from the CPU intc.
Signed-off-by: Caleb James DeLisle <cjd@cjdns.fr>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260430164157.6026-3-cjd@cjdns.fr
In MIPS VEIC mode (Vectored External Interrupt Controller), the
hardware stops directly dispatching CPU interrupts such as IPIs or CPU
performance counters, and instead it communicates them to the external
interrupt controller (the hardware described here) which prioritizes,
renumbers, and integrates them with its own hardware interrupt pins.
Interrupts from the external controller are then dispatched through a
different method via a dispatch table. In effect, the external
controller subsumes the CPU controller and becomes the root.
34K Manual (MD00534) Section 6.3.1.3 rev 1.13 page 136
Since there are interrupts which ought to be controlled by the CPU
controller driver - particularly the IPI interrupts - we create a
reverse mapping where those interrupts may be sent back to the CPU
intc when they are received. This maintains the fiction that there is
still a hierarchy, and keeps the DT the same no matter whether the
processor is in VEIC mode or not. The econet,cpu-interrupt-map is
optional and if omitted, it's assumed that no interrupts need to be
mapped.
Signed-off-by: Caleb James DeLisle <cjd@cjdns.fr>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://patch.msgid.link/20260430164157.6026-2-cjd@cjdns.fr
It was originally part of the @cmd_status field description but became
separated and now appears between @ring_size and @dms without proper context.
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
The max_segs field is of type unsigned short, and if a host driver
sets an excessively large value, it may be truncated to zero. This
can cause mmc_alloc_sg() to call kmalloc_objs() with a zero size
allocation request, which leads to undefined behavior.
Under the SLUB allocator, kmalloc(0) returns a special pointer
(ZERO_SIZE_PTR). The subsequent 'if (sg)' check will evaluate to
true, and sg_init_table() will then attempt to access invalid memory,
resulting in a crash:
dwmmc_rockchip 2a310000.mmc: Successfully tuned phase to 133
mmc1: new UHS-I speed SDR104 SDHC card at address aaaa
Unable to handle kernel paging request at virtual address 0000001ffffffff0
Mem abort info:
ESR = 0x0000000096000004
EC = 0x25: DABT (current EL), IL = 32 bits
SET = 0, FnV = 0
EA = 0, S1PTW = 0
FSC = 0x04: level 0 translation fault
Data abort info:
ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
CM = 0, WnR = 0, TnD = 0, TagAccess = 0
GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
user pgtable: 4k pages, 48-bit VAs, pgdp=0000000102c88000
[0000001ffffffff0] pgd=0000000000000000, p4d=0000000000000000
Internal error: Oops: 0000000096000004 [#1] SMP
Modules linked in:
CPU: 2 UID: 0 PID: 102 Comm: kworker/2:1 Not tainted 7.0.0-rc6-next-20260331-00013-g4d93c25963c5-dirty #80 PREEMPT
Hardware name: Rockchip RK3576 EVB V10 Board (DT)
Workqueue: events_freezable mmc_rescan
pstate: 80000005 (Nzcv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
pc : sg_init_table+0x2c/0x50
lr : sg_init_table+0x24/0x50
sp : ffff8000837db710
x29: ffff8000837db710 x28: 000000000000c000 x27: 0000000000000300
x26: 0000000000000000 x25: 0000000000000040 x24: ffff0000c46a0000
x23: 0000000000000000 x22: ffff0000c0c73c00 x21: 0000000000000010
x20: 0000000000000010 x19: 0000000000000000 x18: 000000000000002c
x17: 0000000000000000 x16: 0000000000000001 x15: 0000000000000000
x14: 0000000000000400 x13: ffff8000837dc000 x12: 0000000000000000
x11: ffff0000c0c73ca0 x10: 0000000000000040 x9 : 459ec1f0abbdbb00
x8 : 0000001fffffffe0 x7 : 0000000000000000 x6 : 000000000000003f
x5 : 0000000000035579 x4 : 0000000000000901 x3 : 0000000000000000
x2 : 0000000000000000 x1 : 0000000000000000 x0 : 0000000000000010
Call trace:
sg_init_table+0x2c/0x50 (P)
mmc_mq_init_request+0x64/0x90
blk_mq_alloc_map_and_rqs+0x3ac/0x480
blk_mq_alloc_set_map_and_rqs+0x98/0x1e0
blk_mq_alloc_tag_set+0x1c0/0x290
mmc_init_queue+0x120/0x370
mmc_blk_alloc_req+0x150/0x420
To prevent this, add a validation check in mmc_mq_init_request() to
detect when sg_len (derived from max_segs) is zero. If sg_len is zero,
we return an error and print an error message, allowing host driver
developers to identify and fix incorrect max_segs configuration.
This is a defensive measure that ensures the MMC core fails gracefully
when host drivers provide invalid max_segs values, rather than crashing
with a page fault.
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
When drvdata() was introduced in commit 6f61a2637a ("rust: device:
introduce Device::drvdata()"), its commit message already noted that a
direct accessor to the driver's bus device private data is not commonly
required -- bus callbacks provide access through &self, and other entry
points (IRQs, workqueues, IOCTLs, etc.) carry their own private data.
The sole motivation for drvdata() was inter-driver interaction -- an
auxiliary driver deriving the parent's bus device private data from the
parent device.
However, drvdata() exposes the driver's bus device private data beyond
the driver's own scope. This creates ordering constraints; for instance
drvdata may not be set yet when the first caller of drvdata() can
appear. It also forces the driver's bus device private data to outlive
all registrations that access it, which causes unnecessary
complications.
Private data should be private to the entity that issues it, i.e. bus
device private data belongs to bus callbacks, class device private data
to class callbacks, IRQ private data to the IRQ handler, etc.
With registration-private data now available through the auxiliary bus,
there is no remaining user of drvdata(), thus remove it.
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Link: https://patch.msgid.link/20260505152400.3905096-4-dakr@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Add a registration_data pointer to struct auxiliary_device, allowing the
registering (parent) driver to attach private data to the device at
registration time and retrieve it later when called back by the
auxiliary (child) driver.
By tying the data to the device's registration, Rust drivers can bind
the lifetime of device resources to it, since the auxiliary bus
guarantees that the parent driver remains bound while the auxiliary
device is bound.
On the Rust side, Registration<T> takes ownership of the data via
ForeignOwnable. A TypeId is stored alongside the data for runtime type
checking, making Device::registration_data<T>() a safe method.
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Link: https://patch.msgid.link/20260505152400.3905096-3-dakr@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Add Box::zeroed() for T: Zeroable types.
This allocates with __GFP_ZERO directly, letting the underlying
allocator deal with zeroing out the memory compared to
Box::new(T::zeroed(), flags).
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Link: https://patch.msgid.link/20260505152400.3905096-2-dakr@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
The SPEAr SPI chip-select GPIO driver only depends on generic platform,
OF, and MMIO interfaces, so it can be built outside SPEAr platform
configurations.
Enable compile-test coverage to catch build regressions on other
architectures.
Assisted-by: Codex:GPT-5.5
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Link: https://patch.msgid.link/20260510195531.10561-1-rosenp@gmail.com
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>
Affinity changes of IMSIC interrupts have to be careful to not lose an
interrupt in the process. Each vector keeps track of an affinity change in
progress with two pointers in struct imsic_vector.
imsic_vector::move_prev points to the previous CPU target data and
imsic_vector::move_next to the designated new CPU target data.
imsic_vector::move_prev on the new CPU can only be cleared after the
previous CPU has cleared imsic_vector::move_next, which ususally happens in
__imsic_remote_sync().
In case of CPU hot-unplug __imsic_remote_sync() is not invoked because the
CPU is already marked offline. That means imsic_vector::move_prev becomes
stale until the CPU is onlined again.
The stale pointer prevents further affinity changes for the affected
interrupts.
Solve this by clearing the imsic_vector::move_prev pointers in the CPU
hotplug offline path.
[ tglx: Replace word salad in change log ]
Fixes: 0f67911e82 ("irqchip/riscv-imsic: Separate next and previous pointers in IMSIC vector")
Signed-off-by: Yong-Xuan Wang <yongxuan.wang@sifive.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20260508-imsic-v2-1-e9f08dd46cf5@sifive.com
meson_s4_gpio_irq_set_type() uses the both-edge trigger register for
configuring level type and single edge mode interrupts, which is not
correct.
Use REG_EDGE_POL instead.
Fixes: bbd6fcc76b ("irqchip: Add support for Amlogic A4 and A5 SoCs")
Signed-off-by: Xianwei Zhao <xianwei.zhao@amlogic.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20260508-a9-gpio-irqchip-v1-1-9dc5f3e022e0@amlogic.com
The Amlogic A9 SoCs supports the following GPIO interrupt lines:
A9 IRQ Number:
- 95:86 10 pins on bank Y
- 85:84 2 pins on bank CC
- 83:64 20 pins on bank A
- 63:48 16 pins on bank Z
- 47:30 18 pins on bank X
- 29:22 8 pins on bank H
- 21:14 8 pins on bank M
- 13:0 14 pins on bank B
A9 AO IRQ Number:
- 38 1 pins on bank TESTN
- 37:31 7 pins on bank C
- 30:13 18 pins on bank D
- 12:0 13 pins on bank AO
Update the driver to handle these variants.
Signed-off-by: Xianwei Zhao <xianwei.zhao@amlogic.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260508-a9-gpio-irqchip-v1-3-9dc5f3e022e0@amlogic.com
meson_s4_gpio_irq_set_type() uses the both-edge trigger register for
configuring level type and single edge mode interrupts, which is not
correct.
Use REG_EDGE_POL instead.
Fixes: bbd6fcc76b ("irqchip: Add support for Amlogic A4 and A5 SoCs")
Signed-off-by: Xianwei Zhao <xianwei.zhao@amlogic.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260508-a9-gpio-irqchip-v1-1-9dc5f3e022e0@amlogic.com
Recently handle_percpu_devid_irq() was changed to call
add_interrupt_randomness(). This introduced a potential deadlock when
handle_percpu_devid_irq() is used to handle an NMI, which can be
detected with lockdep, e.g.
================================
WARNING: inconsistent lock state
7.1.0-rc2-pnmi #465 Not tainted
--------------------------------
inconsistent {INITIAL USE} -> {IN-NMI} usage.
perf/695 [HC1[1]:SC0[0]:HE0:SE1] takes:
ffff00837dfd3a18 (&base->lock){-.-.}-{2:2}, at: lock_timer_base+0x6c/0xac
{INITIAL USE} state was registered at:
_raw_spin_lock_irqsave+0x68/0xb0
lock_timer_base+0x6c/0xac
__mod_timer+0x100/0x32c
add_timer_global+0x2c/0x40
__queue_delayed_work+0xf0/0x140
queue_delayed_work_on+0x134/0x138
mem_cgroup_css_online+0x30c/0x310
online_css+0x34/0x10c
cgroup_init_subsys+0x158/0x1c8
cgroup_init+0x440/0x524
start_kernel+0x888/0x998
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(&base->lock);
<Interrupt>
lock(&base->lock);
*** DEADLOCK ***
Call trace:
_raw_spin_lock_irqsave+0x68/0xb0
lock_timer_base+0x6c/0xac
add_timer_on+0x78/0x16c
add_interrupt_randomness+0x124/0x134
handle_percpu_devid_irq+0xd4/0x16c
handle_irq_desc+0x40/0x58
generic_handle_domain_nmi+0x28/0x50
__gic_handle_nmi.isra.0+0x4c/0xa0
gic_handle_irq+0x38/0x2bc
call_on_irq_stack+0x30/0x48
do_interrupt_handler+0x80/0x98
el1_interrupt+0x90/0xac
el1h_64_irq_handler+0x18/0x24
el1h_64_irq+0x80/0x84
[...]
During review, Thomas pointed out it wouldn't be safe for
handle_percpu_devid_irq() to call add_interrupt_randomness() if it was
used to handle NMIs:
https://lore.kernel.org/lkml/87bjgik042.ffs@tglx/
... but evidently people missed that handle_percpu_devid_irq() *is* used
for NMIs.
While it might seem that NMIs should be handled with a separate
handle_percpu_devid_nmi() function, for various structural reasons this was
impractical, and handle_percpu_devid_irq() has been expected to be used for
NMIs since commits:
21bbbc50f3 ("irqchip/gic-v3: Switch high priority PPIs over to handle_percpu_devid_irq()")
5ff78c8de9 ("genirq: Kill handle_percpu_devid_fasteoi_nmi()")
Taking the above into account, avoid the deadlock by not calling
add_interrupt_randomness() when handle_percpu_devid_irq() is called in an
NMI context. This is consistent with other NNI handling flows, which do not
call add_interrupt_randomness().
At the same time, update the kernel-doc comment to make it clear that
handle_percpu_devid_irq() can be called in NMI context. The rest of
handle_percpu_devid_irq() is currently NMI safe and doesn't need to change.
Fixes: fd7400cfcb ("genirq/chip: Invoke add_interrupt_randomness() in handle_percpu_devid_irq()")
Reported-by: Ada Couprie Diaz <ada.coupriediaz@arm.com>
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Reviewed-by: Jinjie Ruan <ruanjinjie@huawei.com>
Reviewed-by: Marc Zyngier <maz@kernel.org>
Link: https://patch.msgid.link/20260507110518.3128248-1-mark.rutland@arm.com
