Kunkun Jiang reports that for a workload involving the simultaneous startup
of a large number of VMs (for a total of about 200 vcpus), a lot of CPU
time gets spent on spinning on the tmp_mask_lock that exists as a static
raw spinlock in irq_do_set_affinity(). This lock protects a global cpumask
(tmp_mask) that is used as a temporary variable to compute the resulting
affinity.
While this is triggered by KVM issuing a irq_set_affinity() call each time
a vcpu is about to execute, it is obvious that having a single global
resource is not very scalable.
Since a cpumask can be a fairly large structure on systems with a high core
count, a stack allocation is not really appropriate. Instead, turn the
global cpumask into a per-CPU variable, removing the need for locking
altogether as the code is executed with preemption and interrupts disabled.
[ tglx: Moved the per CPU variable declaration outside of the function ]
Reported-by: Kunkun Jiang <jiangkunkun@huawei.com>
Suggested-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Kunkun Jiang <jiangkunkun@huawei.com>
Link: https://lore.kernel.org/all/20240826080618.3886694-1-maz@kernel.org
Link: https://lore.kernel.org/all/a7fc58e4-64c2-77fc-c1dc-f5eb78dbbb01@huawei.com
A proper task_work_cancel() API that actually cancels a callback and not
*any* callback pointing to a given function is going to be needed for
perf events event freeing. Do the appropriate rename to prepare for
that.
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20240621091601.18227-2-frederic@kernel.org
- Core code:
- Interrupt storm detection for the lockup watchdog:
Lockups which are caused by interrupt storms are not easy to debug
because there is no information about the events which make the lockup
detector trigger.
To make this more user friendly, provide an extenstion to interrupt
statistics which allows to take snapshots and an interface to retrieve
the delta to the snapshot. Use this new mechanism in the watchdog code
to do a two stage lockup analysis by taking the snapshot and printing
the deltas for the topmost active interrupts on the second trigger.
Note: This contains both the interrupt and the watchdog changes as
the latter depend on the former obviously.
- Avoid summation loops in the /proc/interrupts output and use the global
counter when possible
- Skip suspended interrupts on CPU hotplug operations to ensure that they
are not delivered before the system resumes the device drivers when
coming out of suspend.
- On CPU hot-unplug interrupts which are affine to the outgoing CPU are
migrated to a different CPU in the affinity mask. This can fail when
the CPUs have no vectors left. Instead of giving up try to migrate it
to any online CPU and thereby breaking the affinity setting in order to
prevent a stale device interrupt which targets an offline CPU
- The usual small cleanups
- Driver code:
- Support for the RISCV AIA MSI controller
- Make the interrupt allocation for the Loongson PCH controller more
flexible to prevent vector exhaustion
- The usual set of cleanups and fixes all over the place
-----BEGIN PGP SIGNATURE-----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=jFLf
-----END PGP SIGNATURE-----
Merge tag 'irq-core-2024-05-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull interrupt subsystem updates from Thomas Gleixner:
"Core code:
- Interrupt storm detection for the lockup watchdog:
Lockups which are caused by interrupt storms are not easy to debug
because there is no information about the events which make the
lockup detector trigger.
To make this more user friendly, provide an extenstion to interrupt
statistics which allows to take snapshots and an interface to
retrieve the delta to the snapshot. Use this new mechanism in the
watchdog code to do a two stage lockup analysis by taking the
snapshot and printing the deltas for the topmost active interrupts
on the second trigger.
Note: This contains both the interrupt and the watchdog changes as
the latter depend on the former obviously.
- Avoid summation loops in the /proc/interrupts output and use the
global counter when possible
- Skip suspended interrupts on CPU hotplug operations to ensure that
they are not delivered before the system resumes the device drivers
when coming out of suspend.
- On CPU hot-unplug interrupts which are affine to the outgoing CPU
are migrated to a different CPU in the affinity mask. This can fail
when the CPUs have no vectors left. Instead of giving up try to
migrate it to any online CPU and thereby breaking the affinity
setting in order to prevent a stale device interrupt which targets
an offline CPU
- The usual small cleanups
Driver code:
- Support for the RISCV AIA MSI controller
- Make the interrupt allocation for the Loongson PCH controller more
flexible to prevent vector exhaustion
- The usual set of cleanups and fixes all over the place"
* tag 'irq-core-2024-05-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (51 commits)
irqchip/gic-v3-its: Remove BUG_ON in its_vpe_irq_domain_alloc
cpuidle: Avoid explicit cpumask allocation on stack
irqchip/sifive-plic: Avoid explicit cpumask allocation on stack
irqchip/riscv-aplic-direct: Avoid explicit cpumask allocation on stack
irqchip/loongson-eiointc: Avoid explicit cpumask allocation on stack
irqchip/gic-v3-its: Avoid explicit cpumask allocation on stack
irqchip/irq-bcm6345-l1: Avoid explicit cpumask allocation on stack
cpumask: Introduce cpumask_first_and_and()
irqchip/irq-brcmstb-l2: Avoid saving mask on shutdown
genirq: Reuse irq_is_nmi()
genirq/cpuhotplug: Retry with cpu_online_mask when migration fails
genirq/cpuhotplug: Skip suspended interrupts when restoring affinity
arm64: dts: st: Add interrupt parent to pinctrl on stm32mp251
arm64: dts: st: Add exti1 and exti2 nodes on stm32mp251
ARM: dts: stm32: List exti parent interrupts on stm32mp131
ARM: dts: stm32: List exti parent interrupts on stm32mp151
arm64: Kconfig.platforms: Enable STM32_EXTI for ARCH_STM32
irqchip/stm32-exti: Mark events reserved with RIF configuration check
irqchip/stm32-exti: Skip secure events
irqchip/stm32-exti: Convert driver to standard PM
...
irq_restore_affinity_of_irq() restarts managed interrupts unconditionally
when the first CPU in the affinity mask comes online. That's correct during
normal hotplug operations, but not when resuming from S3 because the
drivers are not resumed yet and interrupt delivery is not expected by them.
Skip the startup of suspended interrupts and let resume_device_irqs() deal
with restoring them. This ensures that irqs are not delivered to drivers
during the noirq phase of resuming from S3, after non-boot CPUs are brought
back online.
Signed-off-by: David Stevens <stevensd@chromium.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20240424090341.72236-1-stevensd@chromium.org
There is a problem when a driver requests a shared interrupt line to run a
threaded handler on it without IRQF_ONESHOT set if that flag has been set
already for the IRQ in question by somebody else. Namely, the request
fails which usually leads to a probe failure even though the driver might
have worked just fine with IRQF_ONESHOT, but it does not want to use it by
default. Currently, the only way to handle this is to try to request the
IRQ without IRQF_ONESHOT, but with IRQF_PROBE_SHARED set and if this fails,
try again with IRQF_ONESHOT set. However, this is a bit cumbersome and not
very clean.
When commit 7a36b901a6 ("ACPI: OSL: Use a threaded interrupt handler for
SCI") switched the ACPI subsystem over to using a threaded interrupt
handler for the SCI, it had to use IRQF_ONESHOT for it because that's
required due to the way the SCI handler works (it needs to walk all of the
enabled GPEs before the interrupt line can be unmasked). The SCI interrupt
line is not shared with other users very often due to the SCI handling
overhead, but on sone systems it is shared and when the other user of it
attempts to install a threaded handler, a flags mismatch related to
IRQF_ONESHOT may occur.
As it turned out, that happened to the pinctrl-amd driver and so commit
4451e8e841 ("pinctrl: amd: Add IRQF_ONESHOT to the interrupt request")
attempted to address the issue by adding IRQF_ONESHOT to the interrupt
flags in that driver, but this is now causing an IRQF_ONESHOT-related
mismatch to occur on another system which cannot boot as a result of it.
Clearly, pinctrl-amd can work with IRQF_ONESHOT if need be, but it should
not set that flag by default, so it needs a way to indicate that to the
interrupt subsystem.
To that end, introdcuce a new interrupt flag, IRQF_COND_ONESHOT, which will
only have effect when the IRQ line is shared and IRQF_ONESHOT has been set
for it already, in which case it will be promoted to the latter.
This is sufficient for drivers sharing the interrupt line with the SCI as
it is requested by the ACPI subsystem before any drivers are probed, so
they will always see IRQF_ONESHOT set for the interrupt in question.
Fixes: 4451e8e841 ("pinctrl: amd: Add IRQF_ONESHOT to the interrupt request")
Reported-by: Francisco Ayala Le Brun <francisco@videowindow.eu>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Cc: 6.8+ <stable@vger.kernel.org> # 6.8+
Closes: https://lore.kernel.org/lkml/CAN-StX1HqWqi+YW=t+V52-38Mfp5fAz7YHx4aH-CQjgyNiKx3g@mail.gmail.com/
Link: https://lore.kernel.org/r/12417336.O9o76ZdvQC@kreacher
The affinity setting of interrupt threads happens in the context of the
thread when the thread is woken up by an hard interrupt. As this can be an
arbitrary after changing the affinity, the thread can become runnable on an
isolated CPU and cause isolation disruption.
Avoid this by checking the set affinity request in wait_for_interrupt() and
waking the threads immediately when the affinity is modified.
Note that this is of the most benefit on systems where the interrupt
affinity itself does not need to be deferred to the interrupt handler, but
even where that's not the case, the total dirsuption will be less.
Signed-off-by: Crystal Wood <crwood@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20240122235353.15235-1-crwood@redhat.com
Add a kthread_stop_put() helper that stops a thread and puts its task
struct. Use it to replace the various instances of kthread_stop()
followed by put_task_struct().
Remove the kthread_stop_put() macro in usbip that is similar but doesn't
return the result of kthread_stop().
[agruenba@redhat.com: fix kerneldoc comment]
Link: https://lkml.kernel.org/r/20230911111730.2565537-1-agruenba@redhat.com
[akpm@linux-foundation.org: document kthread_stop_put()'s argument]
Link: https://lkml.kernel.org/r/20230907234048.2499820-1-agruenba@redhat.com
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
There is a possibility of deadlock if synchronize_hardirq() is called
when the nested threaded interrupt is active. The following scenario
was observed on a uniprocessor PREEMPT_NONE system:
Thread 1 Thread 2
handle_nested_thread()
Set INPROGRESS
Call ->thread_fn()
thread_fn goes to sleep
free_irq()
__synchronize_hardirq()
Busy-loop forever waiting for INPROGRESS
to be cleared
The INPROGRESS flag is only supposed to be used for hard interrupt
handlers. Remove the incorrect usage in the nested threaded interrupt
case and instead re-use the threads_active / wait_for_threads mechanism
to wait for nested threaded interrupts to complete.
Signed-off-by: Vincent Whitchurch <vincent.whitchurch@axis.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20230613-genirq-nested-v3-1-ae58221143eb@axis.com
For interrupts with secondary threads, the affinity is applied when the
thread is created but if the interrupts affinity is changed later only
the primary thread is updated.
Update the secondary thread's affinity as well to keep all the interrupts
activity on the assigned CPUs.
Signed-off-by: John Keeping <john@metanate.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20230406180857.588682-1-john@metanate.com
With the introduction of threaded interrupt handlers, it is virtually
never safe to call disable_irq() from non-premptible context.
Thus: Update the documentation, add an explicit might_sleep() to catch any
offenders. This is more obvious and straight forward than the implicit
might_sleep() check deeper down in the disable_irq() call chain.
Fixes: 3aa551c9b4 ("genirq: add threaded interrupt handler support")
Signed-off-by: Manfred Spraul <manfred@colorfullife.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20221216150441.200533-3-manfred@colorfullife.com
Core code:
- Make the managed interrupts more robust by shutting them down in the
core code when the assigned affinity mask does not contain online
CPUs.
- Make the irq simulator chip work on RT
- A small set of cpumask and power manageent cleanups
Drivers:
- A set of changes which mark GPIO interrupt chips immutable to prevent
the GPIO subsystem from modifying it under the hood. This provides
the necessary infrastructure and converts a set of GPIO and pinctrl
drivers over.
- A set of changes to make the pseudo-NMI handling for GICv3 more
robust: a missing barrier and consistent handling of the priority
mask.
- Another set of GICv3 improvements and fixes, but nothing outstanding
- The usual set of improvements and cleanups all over the place
- No new irqchip drivers and not even a new device tree binding!
100+ interrupt chips are truly enough.
-----BEGIN PGP SIGNATURE-----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=xcXv
-----END PGP SIGNATURE-----
Merge tag 'irq-core-2022-05-23' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull interrupt handling updates from Thomas Gleixner:
"Core code:
- Make the managed interrupts more robust by shutting them down in
the core code when the assigned affinity mask does not contain
online CPUs.
- Make the irq simulator chip work on RT
- A small set of cpumask and power manageent cleanups
Drivers:
- A set of changes which mark GPIO interrupt chips immutable to
prevent the GPIO subsystem from modifying it under the hood. This
provides the necessary infrastructure and converts a set of GPIO
and pinctrl drivers over.
- A set of changes to make the pseudo-NMI handling for GICv3 more
robust: a missing barrier and consistent handling of the priority
mask.
- Another set of GICv3 improvements and fixes, but nothing
outstanding
- The usual set of improvements and cleanups all over the place
- No new irqchip drivers and not even a new device tree binding!
100+ interrupt chips are truly enough"
* tag 'irq-core-2022-05-23' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (39 commits)
irqchip: Add Kconfig symbols for sunxi drivers
irqchip/gic-v3: Fix priority mask handling
irqchip/gic-v3: Refactor ISB + EOIR at ack time
irqchip/gic-v3: Ensure pseudo-NMIs have an ISB between ack and handling
genirq/irq_sim: Make the irq_work always run in hard irq context
irqchip/armada-370-xp: Do not touch Performance Counter Overflow on A375, A38x, A39x
irqchip/gic: Improved warning about incorrect type
irqchip/csky: Return true/false (not 1/0) from bool functions
irqchip/imx-irqsteer: Add runtime PM support
irqchip/imx-irqsteer: Constify irq_chip struct
irqchip/armada-370-xp: Enable MSI affinity configuration
irqchip/aspeed-scu-ic: Fix irq_of_parse_and_map() return value
irqchip/aspeed-i2c-ic: Fix irq_of_parse_and_map() return value
irqchip/sun6i-r: Use NULL for chip_data
irqchip/xtensa-mx: Fix initial IRQ affinity in non-SMP setup
irqchip/exiu: Fix acknowledgment of edge triggered interrupts
irqchip/gic-v3: Claim iomem resources
dt-bindings: interrupt-controller: arm,gic-v3: Make the v2 compat requirements explicit
irqchip/gic-v3: Relax polling of GIC{R,D}_CTLR.RWP
irqchip/gic-v3: Detect LPI invalidation MMIO registers
...
A kernel hang can be observed when running setserial in a loop on a kernel
with force threaded interrupts. The sequence of events is:
setserial
open("/dev/ttyXXX")
request_irq()
do_stuff()
-> serial interrupt
-> wake(irq_thread)
desc->threads_active++;
close()
free_irq()
kthread_stop(irq_thread)
synchronize_irq() <- hangs because desc->threads_active != 0
The thread is created in request_irq() and woken up, but does not get on a
CPU to reach the actual thread function, which would handle the pending
wake-up. kthread_stop() sets the should stop condition which makes the
thread immediately exit, which in turn leaves the stale threads_active
count around.
This problem was introduced with commit 519cc8652b, which addressed a
interrupt sharing issue in the PCIe code.
Before that commit free_irq() invoked synchronize_irq(), which waits for
the hard interrupt handler and also for associated threads to complete.
To address the PCIe issue synchronize_irq() was replaced with
__synchronize_hardirq(), which only waits for the hard interrupt handler to
complete, but not for threaded handlers.
This was done under the assumption, that the interrupt thread already
reached the thread function and waits for a wake-up, which is guaranteed to
be handled before acting on the stop condition. The problematic case, that
the thread would not reach the thread function, was obviously overlooked.
Make sure that the interrupt thread is really started and reaches
thread_fn() before returning from __setup_irq().
This utilizes the existing wait queue in the interrupt descriptor. The
wait queue is unused for non-shared interrupts. For shared interrupts the
usage might cause a spurious wake-up of a waiter in synchronize_irq() or the
completion of a threaded handler might cause a spurious wake-up of the
waiter for the ready flag. Both are harmless and have no functional impact.
[ tglx: Amended changelog ]
Fixes: 519cc8652b ("genirq: Synchronize only with single thread on free_irq()")
Signed-off-by: Thomas Pfaff <tpfaff@pcs.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Marc Zyngier <maz@kernel.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/552fe7b4-9224-b183-bb87-a8f36d335690@pcs.com
Although setting the affinity of an interrupt to a set of CPUs that doesn't
have any online CPU is generally frowned apon, there are a few limited
cases where such affinity is set from a CPUHP notifier, setting the
affinity to a CPU that isn't online yet.
The saving grace is that this is always done using the 'force' attribute,
which gives a hint that the affinity setting can be outside of the online
CPU mask and the callsite set this flag with the knowledge that the
underlying interrupt controller knows to handle it.
This restores the expected behaviour on Marek's system.
Fixes: 33de0aa4ba ("genirq: Always limit the affinity to online CPUs")
Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Link: https://lore.kernel.org/r/4b7fc13c-887b-a664-26e8-45aed13f048a@samsung.com
Link: https://lore.kernel.org/r/20220414140011.541725-1-maz@kernel.org
When booting with maxcpus=<small number> (or even loading a driver
while most CPUs are offline), it is pretty easy to observe managed
affinities containing a mix of online and offline CPUs being passed
to the irqchip driver.
This means that the irqchip cannot trust the affinity passed down
from the core code, which is a bit annoying and requires (at least
in theory) all drivers to implement some sort of affinity narrowing.
In order to address this, always limit the cpumask to the set of
online CPUs.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20220405185040.206297-3-maz@kernel.org
Refer to housekeeping APIs using single feature types instead of flags.
This prevents from passing multiple isolation features at once to
housekeeping interfaces, which soon won't be possible anymore as each
isolation features will have their own cpumask.
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Juri Lelli <juri.lelli@redhat.com>
Reviewed-by: Phil Auld <pauld@redhat.com>
Link: https://lore.kernel.org/r/20220207155910.527133-5-frederic@kernel.org
The discussion about removing the side effect of irq_set_affinity_hint() of
actually applying the cpumask (if not NULL) as affinity to the interrupt,
unearthed a few unpleasantries:
1) The modular perf drivers rely on the current behaviour for the very
wrong reasons.
2) While none of the other drivers prevents user space from changing
the affinity, a cursorily inspection shows that there are at least
expectations in some drivers.
#1 needs to be cleaned up anyway, so that's not a problem
#2 might result in subtle regressions especially when irqbalanced (which
nowadays ignores the affinity hint) is disabled.
Provide new interfaces:
irq_update_affinity_hint() - Only sets the affinity hint pointer
irq_set_affinity_and_hint() - Set the pointer and apply the affinity to
the interrupt
Make irq_set_affinity_hint() a wrapper around irq_apply_affinity_hint() and
document it to be phased out.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Nitesh Narayan Lal <nitesh@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20210501021832.743094-1-jesse.brandeburg@intel.com
Link: https://lore.kernel.org/r/20210903152430.244937-2-nitesh@redhat.com
With enabled threaded interrupts the nouveau driver reported the
following:
| Chain exists of:
| &mm->mmap_lock#2 --> &device->mutex --> &cpuset_rwsem
|
| Possible unsafe locking scenario:
|
| CPU0 CPU1
| ---- ----
| lock(&cpuset_rwsem);
| lock(&device->mutex);
| lock(&cpuset_rwsem);
| lock(&mm->mmap_lock#2);
The device->mutex is nvkm_device::mutex.
Unblocking the lockchain at `cpuset_rwsem' is probably the easiest
thing to do. Move the priority assignment to the start of the newly
created thread.
Fixes: 710da3c8ea ("sched/core: Prevent race condition between cpuset and __sched_setscheduler()")
Reported-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
[bigeasy: Patch description]
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/a23a826af7c108ea5651e73b8fbae5e653f16e86.camel@gmx.de
On RT kernels, the use of migrate_disable()/migrate_enable() is sufficient
to guarantee a task isn't moved to another CPU. Update the
irq_set_irqchip_state() documentation to reflect this.
Signed-off-by: Josh Cartwright <joshc@ni.com>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20210917103055.92150-1-bigeasy@linutronix.de
Fixes: 61377ec144 ("genirq: Clarify documentation for request_threaded_irq()")
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
With CONFIG_IRQ_FORCED_THREADING=y, testing the boolean force_irqthreads
could incur a cache line miss in invoke_softirq() and other places.
Replace the test with a static key to avoid the potential cache miss.
[ tglx: Dropped the IDE part, removed the export and updated blk-mq ]
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Tanner Love <tannerlove@google.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20210602180338.3324213-1-tannerlove.kernel@gmail.com
Perf modules abuse irq_set_affinity_hint() to set the affinity of system
PMU interrupts just because irq_set_affinity() was not exported.
The fact that irq_set_affinity_hint() actually sets the affinity is a
non-documented side effect and the name is clearly saying it's a hint.
To clean this up, export the real affinity setter.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Link: https://lore.kernel.org/r/20210518093117.968251441@linutronix.de
The whole call to note_interrupt() can be avoided or return early when
interrupts would be marked accordingly. For IPI handlers which always
return HANDLED the whole procedure is pretty pointless to begin with.
Add a IRQF_NO_DEBUG flag and mark the interrupt accordingly if supplied
when the interrupt is requested.
When noirqdebug is set on the kernel commandline, then the interrupt is
marked unconditionally so that there is only one condition in the hotpath
to evaluate.
[ clg: Add changelog ]
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/7a8ad02f-63a8-c1aa-fdd1-39d973593d02@kaod.org
Core changes:
- Provide IRQF_NO_AUTOEN as a flag for request*_irq() so drivers can be
cleaned up which either use a seperate mechanism to prevent auto-enable
at request time or have a racy mechanism which disables the interrupt
right after request.
- Get rid of the last usage of irq_create_identity_mapping() and remove
the interface.
- An overhaul of tasklet_disable(). Most usage sites of tasklet_disable()
are in task context and usually in cleanup, teardown code pathes.
tasklet_disable() spinwaits for a tasklet which is currently executed.
That's not only a problem for PREEMPT_RT where this can lead to a live
lock when the disabling task preempts the softirq thread. It's also
problematic in context of virtualization when the vCPU which runs the
tasklet is scheduled out and the disabling code has to spin wait until
it's scheduled back in. Though there are a few code pathes which invoke
tasklet_disable() from non-sleepable context. For these a new disable
variant which still spinwaits is provided which allows to switch
tasklet_disable() to a sleep wait mechanism. For the atomic use cases
this does not solve the live lock issue on PREEMPT_RT. That is mitigated
by blocking on the RT specific softirq lock.
- The PREEMPT_RT specific implementation of softirq processing and
local_bh_disable/enable().
On RT enabled kernels soft interrupt processing happens always in task
context and all interrupt handlers, which are not explicitly marked to
be invoked in hard interrupt context are forced into task context as
well. This allows to protect against softirq processing with a per
CPU lock, which in turn allows to make BH disabled regions preemptible.
Most of the softirq handling code is still shared. The RT/non-RT
specific differences are addressed with a set of inline functions which
provide the context specific functionality. The local_bh_disable() /
local_bh_enable() mechanism are obviously seperate.
- The usual set of small improvements and cleanups
Driver changes:
- New drivers for Nuvoton WPCM450 and DT 79rc3243x interrupt controllers
- Extended functionality for MStar, STM32 and SC7280 irq chips
- Enhanced robustness for ARM GICv3/4.1 drivers
- The usual set of cleanups and improvements all over the place
-----BEGIN PGP SIGNATURE-----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=id4G
-----END PGP SIGNATURE-----
Merge tag 'irq-core-2021-04-26' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull irq updates from Thomas Gleixner:
"The usual updates from the irq departement:
Core changes:
- Provide IRQF_NO_AUTOEN as a flag for request*_irq() so drivers can
be cleaned up which either use a seperate mechanism to prevent
auto-enable at request time or have a racy mechanism which disables
the interrupt right after request.
- Get rid of the last usage of irq_create_identity_mapping() and
remove the interface.
- An overhaul of tasklet_disable().
Most usage sites of tasklet_disable() are in task context and
usually in cleanup, teardown code pathes. tasklet_disable()
spinwaits for a tasklet which is currently executed. That's not
only a problem for PREEMPT_RT where this can lead to a live lock
when the disabling task preempts the softirq thread. It's also
problematic in context of virtualization when the vCPU which runs
the tasklet is scheduled out and the disabling code has to spin
wait until it's scheduled back in.
There are a few code pathes which invoke tasklet_disable() from
non-sleepable context. For these a new disable variant which still
spinwaits is provided which allows to switch tasklet_disable() to a
sleep wait mechanism. For the atomic use cases this does not solve
the live lock issue on PREEMPT_RT. That is mitigated by blocking on
the RT specific softirq lock.
- The PREEMPT_RT specific implementation of softirq processing and
local_bh_disable/enable().
On RT enabled kernels soft interrupt processing happens always in
task context and all interrupt handlers, which are not explicitly
marked to be invoked in hard interrupt context are forced into task
context as well. This allows to protect against softirq processing
with a per CPU lock, which in turn allows to make BH disabled
regions preemptible.
Most of the softirq handling code is still shared. The RT/non-RT
specific differences are addressed with a set of inline functions
which provide the context specific functionality. The
local_bh_disable() / local_bh_enable() mechanism are obviously
seperate.
- The usual set of small improvements and cleanups
Driver changes:
- New drivers for Nuvoton WPCM450 and DT 79rc3243x interrupt
controllers
- Extended functionality for MStar, STM32 and SC7280 irq chips
- Enhanced robustness for ARM GICv3/4.1 drivers
- The usual set of cleanups and improvements all over the place"
* tag 'irq-core-2021-04-26' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (53 commits)
irqchip/xilinx: Expose Kconfig option for Zynq/ZynqMP
irqchip/gic-v3: Do not enable irqs when handling spurious interrups
dt-bindings: interrupt-controller: Add IDT 79RC3243x Interrupt Controller
irqchip: Add support for IDT 79rc3243x interrupt controller
irqdomain: Drop references to recusive irqdomain setup
irqdomain: Get rid of irq_create_strict_mappings()
irqchip/jcore-aic: Kill use of irq_create_strict_mappings()
ARM: PXA: Kill use of irq_create_strict_mappings()
irqchip/gic-v4.1: Disable vSGI upon (GIC CPUIF < v4.1) detection
irqchip/tb10x: Use 'fallthrough' to eliminate a warning
genirq: Reduce irqdebug cacheline bouncing
kernel: Initialize cpumask before parsing
irqchip/wpcm450: Drop COMPILE_TEST
irqchip/irq-mst: Support polarity configuration
irqchip: Add driver for WPCM450 interrupt controller
dt-bindings: interrupt-controller: Add nuvoton, wpcm450-aic
dt-bindings: qcom,pdc: Add compatible for sc7280
irqchip/stm32: Add usart instances exti direct event support
irqchip/gic-v3: Fix OF_BAD_ADDR error handling
irqchip/sifive-plic: Mark two global variables __ro_after_init
...
With interrupt force threading all device interrupt handlers are invoked
from kernel threads. Contrary to hard interrupt context the invocation only
disables bottom halfs, but not interrupts. This was an oversight back then
because any code like this will have an issue:
thread(irq_A)
irq_handler(A)
spin_lock(&foo->lock);
interrupt(irq_B)
irq_handler(B)
spin_lock(&foo->lock);
This has been triggered with networking (NAPI vs. hrtimers) and console
drivers where printk() happens from an interrupt which interrupted the
force threaded handler.
Now people noticed and started to change the spin_lock() in the handler to
spin_lock_irqsave() which affects performance or add IRQF_NOTHREAD to the
interrupt request which in turn breaks RT.
Fix the root cause and not the symptom and disable interrupts before
invoking the force threaded handler which preserves the regular semantics
and the usefulness of the interrupt force threading as a general debugging
tool.
For not RT this is not changing much, except that during the execution of
the threaded handler interrupts are delayed until the handler
returns. Vs. scheduling and softirq processing there is no difference.
For RT kernels there is no issue.
Fixes: 8d32a307e4 ("genirq: Provide forced interrupt threading")
Reported-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Johan Hovold <johan@kernel.org>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://lore.kernel.org/r/20210317143859.513307808@linutronix.de
Many drivers don't want interrupts enabled automatically via request_irq().
So they are handling this issue by either way of the below two:
(1)
irq_set_status_flags(irq, IRQ_NOAUTOEN);
request_irq(dev, irq...);
(2)
request_irq(dev, irq...);
disable_irq(irq);
The code in the second way is silly and unsafe. In the small time gap
between request_irq() and disable_irq(), interrupts can still come.
The code in the first way is safe though it's subobtimal.
Add a new IRQF_NO_AUTOEN flag which can be handed in by drivers to
request_irq() and request_nmi(). It prevents the automatic enabling of the
requested interrupt/nmi in the same safe way as #1 above. With that the
various usage sites of #1 and #2 above can be simplified and corrected.
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: dmitry.torokhov@gmail.com
Link: https://lore.kernel.org/r/20210302224916.13980-2-song.bao.hua@hisilicon.com
One of the users can be built modular:
ERROR: modpost: "irq_check_status_bit" [drivers/perf/arm_spe_pmu.ko] undefined!
Fixes: fdd0296304 ("genirq: Move status flag checks to core")
Reported-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20201227192049.GA195845@roeck-us.net
This function uses irq_to_desc() and is going to be used by modules to
replace the open coded irq_to_desc() (ab)usage. The final goal is to remove
the export of irq_to_desc() so driver cannot fiddle with it anymore.
Move it into the core code and fixup the usage sites to include the proper
header.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20201210194042.548936472@linutronix.de
Add a function to allow the affinity of an interrupt be switched to
managed, such that interrupts allocated for platform devices may be
managed.
This new interface has certain limitations, and attempts to use it in the
following circumstances will fail:
- For when the kernel is configured for generic IRQ reservation mode (in
config GENERIC_IRQ_RESERVATION_MODE). The reason being that it could
conflict with managed vs. non-managed interrupt accounting.
- The interrupt is already started, which should not be the case during
init
- The interrupt is already configured as managed, which means double init
Suggested-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/1606905417-183214-2-git-send-email-john.garry@huawei.com
A previous commit changed the notification mode from true/false to an
int, allowing notify-no, notify-yes, or signal-notify. This was
backwards compatible in the sense that any existing true/false user
would translate to either 0 (on notification sent) or 1, the latter
which mapped to TWA_RESUME. TWA_SIGNAL was assigned a value of 2.
Clean this up properly, and define a proper enum for the notification
mode. Now we have:
- TWA_NONE. This is 0, same as before the original change, meaning no
notification requested.
- TWA_RESUME. This is 1, same as before the original change, meaning
that we use TIF_NOTIFY_RESUME.
- TWA_SIGNAL. This uses TIF_SIGPENDING/JOBCTL_TASK_WORK for the
notification.
Clean up all the callers, switching their 0/1/false/true to using the
appropriate TWA_* mode for notifications.
Fixes: e91b481623 ("task_work: teach task_work_add() to do signal_wake_up()")
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In irq_set_irqchip_state(), the irq descriptor is not unlocked after an
error is encountered. While that should never happen in practice, a buggy
driver may trigger it. This would result in a lockup, so fix it.
Fixes: 1d0326f352 ("genirq: Check irq_data_get_irq_chip() return value before use")
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20200811180012.80269-1-linux@roeck-us.net
static priority level knowledge from non-scheduler code.
The three APIs for non-scheduler code to set SCHED_FIFO are:
- sched_set_fifo()
- sched_set_fifo_low()
- sched_set_normal()
These are two FIFO priority levels: default (high), and a 'low' priority level,
plus sched_set_normal() to set the policy back to non-SCHED_FIFO.
Since the changes affect a lot of non-scheduler code, we kept this in a separate
tree.
When merging to the latest upstream tree there's a conflict in drivers/spi/spi.c,
which can be resolved via:
sched_set_fifo(ctlr->kworker_task);
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-----BEGIN PGP SIGNATURE-----
iQJFBAABCgAvFiEEBpT5eoXrXCwVQwEKEnMQ0APhK1gFAl8pPQIRHG1pbmdvQGtl
cm5lbC5vcmcACgkQEnMQ0APhK1j0Jw/+LlSyX6gD2ATy3cizGL7DFPZogD5MVKTb
IXbhXH/ACpuPQlBe1+haRLbJj6XfXqbOlAleVKt7eh+jZ1jYjC972RCSTO4566mJ
0v8Iy9kkEeb2TDbYx1H3bnk78lf85t0CB+sCzyKUYFuTrXU04eRj7MtN3vAQyRQU
xJg83x/sT5DGdDTP50sL7lpbwk3INWkD0aDCJEaO/a9yHElMsTZiZBKoXxN/s30o
FsfzW56jqtng771H2bo8ERN7+abwJg10crQU5mIaLhacNMETuz0NZ/f8fY/fydCL
Ju8HAdNKNXyphWkAOmixQuyYtWKe2/GfbHg8hld0jmpwxkOSTgZjY+pFcv7/w306
g2l1TPOt8e1n5jbfnY3eig+9Kr8y0qHkXPfLfgRqKwMMaOqTTYixEzj+NdxEIRX9
Kr7oFAv6VEFfXGSpb5L1qyjIGVgQ5/JE/p3OC3GHEsw5VKiy5yjhNLoSmSGzdS61
1YurVvypSEUAn3DqTXgeGX76f0HH365fIKqmbFrUWxliF+YyflMhtrj2JFtejGzH
Md3RgAzxusE9S6k3gw1ev4byh167bPBbY8jz0w3Gd7IBRKy9vo92h6ZRYIl6xeoC
BU2To1IhCAydIr6hNsIiCSDTgiLbsYQzPuVVovUxNh+l1ZvKV2X+csEHhs8oW4pr
4BRU7dKL2NE=
=/7JH
-----END PGP SIGNATURE-----
Merge tag 'sched-fifo-2020-08-04' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull sched/fifo updates from Ingo Molnar:
"This adds the sched_set_fifo*() encapsulation APIs to remove static
priority level knowledge from non-scheduler code.
The three APIs for non-scheduler code to set SCHED_FIFO are:
- sched_set_fifo()
- sched_set_fifo_low()
- sched_set_normal()
These are two FIFO priority levels: default (high), and a 'low'
priority level, plus sched_set_normal() to set the policy back to
non-SCHED_FIFO.
Since the changes affect a lot of non-scheduler code, we kept this in
a separate tree"
* tag 'sched-fifo-2020-08-04' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (24 commits)
sched,tracing: Convert to sched_set_fifo()
sched: Remove sched_set_*() return value
sched: Remove sched_setscheduler*() EXPORTs
sched,psi: Convert to sched_set_fifo_low()
sched,rcutorture: Convert to sched_set_fifo_low()
sched,rcuperf: Convert to sched_set_fifo_low()
sched,locktorture: Convert to sched_set_fifo()
sched,irq: Convert to sched_set_fifo()
sched,watchdog: Convert to sched_set_fifo()
sched,serial: Convert to sched_set_fifo()
sched,powerclamp: Convert to sched_set_fifo()
sched,ion: Convert to sched_set_normal()
sched,powercap: Convert to sched_set_fifo*()
sched,spi: Convert to sched_set_fifo*()
sched,mmc: Convert to sched_set_fifo*()
sched,ivtv: Convert to sched_set_fifo*()
sched,drm/scheduler: Convert to sched_set_fifo*()
sched,msm: Convert to sched_set_fifo*()
sched,psci: Convert to sched_set_fifo*()
sched,drbd: Convert to sched_set_fifo*()
...
John reported that on a RK3288 system the perf per CPU interrupts are all
affine to CPU0 and provided the analysis:
"It looks like what happens is that because the interrupts are not per-CPU
in the hardware, armpmu_request_irq() calls irq_force_affinity() while
the interrupt is deactivated and then request_irq() with IRQF_PERCPU |
IRQF_NOBALANCING.
Now when irq_startup() runs with IRQ_STARTUP_NORMAL, it calls
irq_setup_affinity() which returns early because IRQF_PERCPU and
IRQF_NOBALANCING are set, leaving the interrupt on its original CPU."
This was broken by the recent commit which blocked interrupt affinity
setting in hardware before activation of the interrupt. While this works in
general, it does not work for this particular case. As contrary to the
initial analysis not all interrupt chip drivers implement an activate
callback, the safe cure is to make the deferred interrupt affinity setting
at activation time opt-in.
Implement the necessary core logic and make the two irqchip implementations
for which this is required opt-in. In hindsight this would have been the
right thing to do, but ...
Fixes: baedb87d1b ("genirq/affinity: Handle affinity setting on inactive interrupts correctly")
Reported-by: John Keeping <john@metanate.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Marc Zyngier <maz@kernel.org>
Acked-by: Marc Zyngier <maz@kernel.org>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/87blk4tzgm.fsf@nanos.tec.linutronix.de
Setting interrupt affinity on inactive interrupts is inconsistent when
hierarchical irq domains are enabled. The core code should just store the
affinity and not call into the irq chip driver for inactive interrupts
because the chip drivers may not be in a state to handle such requests.
X86 has a hacky workaround for that but all other irq chips have not which
causes problems e.g. on GIC V3 ITS.
Instead of adding more ugly hacks all over the place, solve the problem in
the core code. If the affinity is set on an inactive interrupt then:
- Store it in the irq descriptors affinity mask
- Update the effective affinity to reflect that so user space has
a consistent view
- Don't call into the irq chip driver
This is the core equivalent of the X86 workaround and works correctly
because the affinity setting is established in the irq chip when the
interrupt is activated later on.
Note, that this is only effective when hierarchical irq domains are enabled
by the architecture. Doing it unconditionally would break legacy irq chip
implementations.
For hierarchial irq domains this works correctly as none of the drivers can
have a dependency on affinity setting in inactive state by design.
Remove the X86 workaround as it is not longer required.
Fixes: 02edee152d ("x86/apic/vector: Ignore set_affinity call for inactive interrupts")
Reported-by: Ali Saidi <alisaidi@amazon.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Ali Saidi <alisaidi@amazon.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20200529015501.15771-1-alisaidi@amazon.com
Link: https://lkml.kernel.org/r/877dv2rv25.fsf@nanos.tec.linutronix.de
Because SCHED_FIFO is a broken scheduler model (see previous patches)
take away the priority field, the kernel can't possibly make an
informed decision.
Effectively no change.
Cc: tglx@linutronix.de
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
irq_data_get_irq_chip() can return NULL, however it is expected that this
never happens. If a buggy driver leads to NULL being returned from
irq_data_get_irq_chip(), warn about it instead of crashing the machine.
Signed-off-by: Marek Vasut <marex@denx.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
To: linux-arm-kernel@lists.infradead.org
The handling of notify->work did not properly maintain notify->kref in two
cases:
1) where the work was already scheduled, another irq_set_affinity_locked()
would get the ref and (no-op-ly) schedule the work. Thus when
irq_affinity_notify() ran, it would drop the original ref but not the
additional one.
2) when cancelling the (old) work in irq_set_affinity_notifier(), if there
was outstanding work a ref had been got for it but was never put.
Fix both by checking the return values of the work handling functions
(schedule_work() for (1) and cancel_work_sync() for (2)) and put the
extra ref if the return value indicates preexisting work.
Fixes: cd7eab44e9 ("genirq: Add IRQ affinity notifiers")
Fixes: 59c39840f5 ("genirq: Prevent use-after-free and work list corruption")
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Ben Hutchings <ben@decadent.org.uk>
Link: https://lkml.kernel.org/r/24f5983f-2ab5-e83a-44ee-a45b5f9300f5@solarflare.com
Qian Cai reported that the WARN_ON() in the x86/msi affinity setting code,
which catches cases where the affinity setting is not done on the CPU which
is the current target of the interrupt, triggers during CPU hotplug stress
testing.
It turns out that the warning which was added with the commit addressing
the MSI affinity race unearthed yet another long standing bug.
If user space writes a bogus affinity mask, i.e. it contains no online CPUs,
then it calls irq_select_affinity_usr(). This was introduced for ALPHA in
eee45269b0 ("[PATCH] Alpha: convert to generic irq framework (generic part)")
and subsequently made available for all architectures in
1840475676 ("genirq: Expose default irq affinity mask (take 3)")
which introduced the circumvention of the affinity setting restrictions for
interrupt which cannot be moved in process context.
The whole exercise is bogus in various aspects:
1) If the interrupt is already started up then there is absolutely
no point to honour a bogus interrupt affinity setting from user
space. The interrupt is already assigned to an online CPU and it
does not make any sense to reassign it to some other randomly
chosen online CPU.
2) If the interupt is not yet started up then there is no point
either. A subsequent startup of the interrupt will invoke
irq_setup_affinity() anyway which will chose a valid target CPU.
So the only correct solution is to just return -EINVAL in case user space
wrote an affinity mask which does not contain any online CPUs, except for
ALPHA which has it's own magic sauce for this.
Fixes: 1840475676 ("genirq: Expose default irq affinity mask (take 3)")
Reported-by: Qian Cai <cai@lca.pw>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Qian Cai <cai@lca.pw>
Link: https://lkml.kernel.org/r/878sl8xdbm.fsf@nanos.tec.linutronix.de
There's some confusion around if an irq that's disabled with disable_irq()
can still wake the system from sleep states such as "suspend to RAM".
Clarify this in the kernel documentation for irq_set_irq_wake() so that
it's clear that an irq can be disabled and still wake the system if it has
been marked for wakeup.
Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Link: https://lkml.kernel.org/r/20200206191521.94559-1-swboyd@chromium.org
The affinity of managed interrupts is completely handled in the kernel and
cannot be changed via the /proc/irq/* interfaces from user space. As the
kernel tries to spread out interrupts evenly accross CPUs on x86 to prevent
vector exhaustion, it can happen that a managed interrupt whose affinity
mask contains both isolated and housekeeping CPUs is routed to an isolated
CPU. As a consequence IO submitted on a housekeeping CPU causes interrupts
on the isolated CPU.
Add a new sub-parameter 'managed_irq' for 'isolcpus' and the corresponding
logic in the interrupt affinity selection code.
The subparameter indicates to the interrupt affinity selection logic that
it should try to avoid the above scenario.
This isolation is best effort and only effective if the automatically
assigned interrupt mask of a device queue contains isolated and
housekeeping CPUs. If housekeeping CPUs are online then such interrupts are
directed to the housekeeping CPU so that IO submitted on the housekeeping
CPU cannot disturb the isolated CPU.
If a queue's affinity mask contains only isolated CPUs then this parameter
has no effect on the interrupt routing decision, though interrupts are only
happening when tasks running on those isolated CPUs submit IO. IO submitted
on housekeeping CPUs has no influence on those queues.
If the affinity mask contains both housekeeping and isolated CPUs, but none
of the contained housekeeping CPUs is online, then the interrupt is also
routed to an isolated CPU. Interrupts are only delivered when one of the
isolated CPUs in the affinity mask submits IO. If one of the contained
housekeeping CPUs comes online, the CPU hotplug logic migrates the
interrupt automatically back to the upcoming housekeeping CPU. Depending on
the type of interrupt controller, this can require that at least one
interrupt is delivered to the isolated CPU in order to complete the
migration.
[ tglx: Removed unused parameter, added and edited comments/documentation
and rephrased the changelog so it contains more details. ]
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20200120091625.17912-1-ming.lei@redhat.com
Requesting a threaded IRQ with handler=NULL and !ONESHOT fails, but the
error message does not include the IRQ line name, which makes it harder to
find the offending driver.
Print the IRQ line name to clarify where the error comes from. Use the same
format as the other pr_err() above in the same function.
Signed-off-by: Luca Ceresoli <luca@lucaceresoli.net>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20191105140854.27893-1-luca@lucaceresoli.net
