Only calculate the affinity for the main I/O vectors, and skip the
pre or post vectors specified by struct irq_affinity.
Also remove the irq_affinity cpumask argument that has never been used.
If we ever need it in the future we can pass it through struct
irq_affinity.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Jens Axboe <axboe@kernel.dk>
Cc: linux-block@vger.kernel.org
Cc: linux-pci@vger.kernel.org
Link: http://lkml.kernel.org/r/1478654107-7384-4-git-send-email-hch@lst.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Only calculate the affinity for the main I/O vectors, and skip the pre or
post vectors specified by struct irq_affinity.
Also remove the irq_affinity cpumask argument that has never been used. If
we ever need it in the future we can pass it through struct irq_affinity.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Acked-by: Jens Axboe <axboe@kernel.dk>
Cc: linux-block@vger.kernel.org
Cc: linux-pci@vger.kernel.org
Link: http://lkml.kernel.org/r/1478654107-7384-3-git-send-email-hch@lst.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Some drivers (various network and RDMA adapter for example) have a MSI-X
vector layout where most of the vectors are used for I/O queues and should
have CPU affinity assigned to them, but some (usually 1 but sometimes more)
at the beginning or end are used for low-performance admin or configuration
work and should not have any explicit affinity assigned to them.
Add a new irq_affinity structure, which will be passed through a variant of
pci_irq_alloc_vectors that allows to specify these requirements (and is
extensible to any future quirks in that area) so that the core IRQ affinity
algorithm can take this quirks into account.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Acked-by: Jens Axboe <axboe@kernel.dk>
Cc: linux-block@vger.kernel.org
Cc: linux-pci@vger.kernel.org
Link: http://lkml.kernel.org/r/1478654107-7384-2-git-send-email-hch@lst.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The current irq spreading infrastructure is just looking at a cpumask and
tries to spread the interrupts over the mask. Thats suboptimal as it does
not take numa nodes into account.
Change the logic so the interrupts are spread across numa nodes and inside
the nodes. If there are more cpus than vectors per node, then we set the
affinity to several cpus. If HT siblings are available we take that into
account and try to set all siblings to a single vector.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: axboe@fb.com
Cc: keith.busch@intel.com
Cc: agordeev@redhat.com
Cc: linux-block@vger.kernel.org
Link: http://lkml.kernel.org/r/1473862739-15032-3-git-send-email-hch@lst.de
KASAN needs to know whether the allocation happens in an IRQ handler.
This lets us strip everything below the IRQ entry point to reduce the
number of unique stack traces needed to be stored.
Move the definition of __irq_entry to <linux/interrupt.h> so that the
users don't need to pull in <linux/ftrace.h>. Also introduce the
__softirq_entry macro which is similar to __irq_entry, but puts the
corresponding functions to the .softirqentry.text section.
Signed-off-by: Alexander Potapenko <glider@google.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Andrey Konovalov <adech.fo@gmail.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
Cc: Konstantin Serebryany <kcc@google.com>
Cc: Dmitry Chernenkov <dmitryc@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Per the x86-specific footnote to PCI spec r3.0, sec 6.2.4, the value 255 in
the Interrupt Line register means "unknown" or "no connection."
Previously, when we couldn't derive an IRQ from the _PRT, we fell back to
using the value from Interrupt Line as an IRQ. It's questionable whether
we should do that at all, but the spec clearly suggests we shouldn't do it
for the value 255 on x86.
Calling request_irq() with IRQ 255 may succeed, but the driver won't
receive any interrupts. Or, if IRQ 255 is shared with another device, it
may succeed, and the driver's ISR will be called at random times when the
*other* device interrupts. Or it may fail if another device is using IRQ
255 with incompatible flags. What we *want* is for request_irq() to fail
predictably so the driver can fall back to polling.
On x86, assume 255 in the Interrupt Line means the INTx line is not
connected. In that case, set dev->irq to IRQ_NOTCONNECTED so request_irq()
will fail gracefully with -ENOTCONN.
We found this problem on a system where Secure Boot firmware assigned
Interrupt Line 255 to an i801_smbus device and another device was already
using MSI-X IRQ 255. This was in v3.10, where i801_probe() fails if
request_irq() fails:
i801_smbus 0000:00:1f.3: enabling device (0140 -> 0143)
i801_smbus 0000:00:1f.3: can't derive routing for PCI INT C
i801_smbus 0000:00:1f.3: PCI INT C: no GSI
genirq: Flags mismatch irq 255. 00000080 (i801_smbus) vs. 00000000 (megasa)
CPU: 0 PID: 2487 Comm: kworker/0:1 Not tainted 3.10.0-229.el7.x86_64 #1
Hardware name: FUJITSU PRIMEQUEST 2800E2/D3736, BIOS PRIMEQUEST 2000 Serie5
Call Trace:
dump_stack+0x19/0x1b
__setup_irq+0x54a/0x570
request_threaded_irq+0xcc/0x170
i801_probe+0x32f/0x508 [i2c_i801]
local_pci_probe+0x45/0xa0
i801_smbus 0000:00:1f.3: Failed to allocate irq 255: -16
i801_smbus: probe of 0000:00:1f.3 failed with error -16
After aeb8a3d16a ("i2c: i801: Check if interrupts are disabled"),
i801_probe() will fall back to polling if request_irq() fails. But we
still need this patch because request_irq() may succeed or fail depending
on other devices in the system. If request_irq() fails, i801_smbus will
work by falling back to polling, but if it succeeds, i801_smbus won't work
because it expects interrupts that it may not receive.
Signed-off-by: Chen Fan <chen.fan.fnst@cn.fujitsu.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
- Remove usage of ib_query_device and instead store attributes in
ib_device struct
- Move iopoll out of block and into lib, rename to irqpoll, and use
in several places in the rdma stack as our new completion queue
polling library mechanism. Update the other block drivers that
already used iopoll to use the new mechanism too.
- Replace the per-entry GID table locks with a single GID table lock
- IPoIB multicast cleanup
- Cleanups to the IB MR facility
- Add support for 64bit extended IB counters
- Fix for netlink oops while parsing RDMA nl messages
- RoCEv2 support for the core IB code
- mlx4 RoCEv2 support
- mlx5 RoCEv2 support
- Cross Channel support for mlx5
- Timestamp support for mlx5
- Atomic support for mlx5
- Raw QP support for mlx5
- MAINTAINERS update for mlx4/mlx5
- Misc ocrdma, qib, nes, usNIC, cxgb3, cxgb4, mlx4, mlx5 updates
- Add support for remote invalidate to the iSER driver (pushed through the
RDMA tree due to dependencies, acknowledged by nab)
- Update to NFSoRDMA (pushed through the RDMA tree due to dependencies,
acknowledged by Bruce)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=5oO7
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma
Pull rdma updates from Doug Ledford:
"Initial roundup of 4.5 merge window patches
- Remove usage of ib_query_device and instead store attributes in
ib_device struct
- Move iopoll out of block and into lib, rename to irqpoll, and use
in several places in the rdma stack as our new completion queue
polling library mechanism. Update the other block drivers that
already used iopoll to use the new mechanism too.
- Replace the per-entry GID table locks with a single GID table lock
- IPoIB multicast cleanup
- Cleanups to the IB MR facility
- Add support for 64bit extended IB counters
- Fix for netlink oops while parsing RDMA nl messages
- RoCEv2 support for the core IB code
- mlx4 RoCEv2 support
- mlx5 RoCEv2 support
- Cross Channel support for mlx5
- Timestamp support for mlx5
- Atomic support for mlx5
- Raw QP support for mlx5
- MAINTAINERS update for mlx4/mlx5
- Misc ocrdma, qib, nes, usNIC, cxgb3, cxgb4, mlx4, mlx5 updates
- Add support for remote invalidate to the iSER driver (pushed
through the RDMA tree due to dependencies, acknowledged by nab)
- Update to NFSoRDMA (pushed through the RDMA tree due to
dependencies, acknowledged by Bruce)"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma: (169 commits)
IB/mlx5: Unify CQ create flags check
IB/mlx5: Expose Raw Packet QP to user space consumers
{IB, net}/mlx5: Move the modify QP operation table to mlx5_ib
IB/mlx5: Support setting Ethernet priority for Raw Packet QPs
IB/mlx5: Add Raw Packet QP query functionality
IB/mlx5: Add create and destroy functionality for Raw Packet QP
IB/mlx5: Refactor mlx5_ib_qp to accommodate other QP types
IB/mlx5: Allocate a Transport Domain for each ucontext
net/mlx5_core: Warn on unsupported events of QP/RQ/SQ
net/mlx5_core: Add RQ and SQ event handling
net/mlx5_core: Export transport objects
IB/mlx5: Expose CQE version to user-space
IB/mlx5: Add CQE version 1 support to user QPs and SRQs
IB/mlx5: Fix data validation in mlx5_ib_alloc_ucontext
IB/sa: Fix netlink local service GFP crash
IB/srpt: Remove redundant wc array
IB/qib: Improve ipoib UD performance
IB/mlx4: Advertise RoCE v2 support
IB/mlx4: Create and use another QP1 for RoCEv2
IB/mlx4: Enable send of RoCE QP1 packets with IP/UDP headers
...
The new name is irq_poll as iopoll is already taken. Better suggestions
welcome.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
Certain interrupt controller drivers have a register set that does not
make it easy to save/restore the mask of enabled/disabled interrupts
at suspend/resume time. At resume time, such drivers rely on the core
kernel irq subsystem to tell whether such or such interrupt is enabled
or not, in order to restore the proper state in the interrupt
controller register.
While the irqd_irq_disabled() provides the relevant information for
global interrupts, there is no similar function to query the
enabled/disabled state of a per-CPU interrupt.
Therefore, this commit complements the percpu_irq API with an
irq_percpu_is_enabled() function.
[ tglx: Simplified the implementation and added kerneldoc ]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: Tawfik Bayouk <tawfik@marvell.com>
Cc: Nadav Haklai <nadavh@marvell.com>
Cc: Lior Amsalem <alior@marvell.com>
Cc: Andrew Lunn <andrew@lunn.ch>
Cc: Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com>
Cc: Gregory Clement <gregory.clement@free-electrons.com>
Cc: Jason Cooper <jason@lakedaemon.net>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Link: http://lkml.kernel.org/r/1445347435-2333-2-git-send-email-thomas.petazzoni@free-electrons.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Force threading of interrupts does not really deal with interrupts
which are requested with a primary and a threaded handler. The current
policy is to leave them alone and let the primary handler run in
interrupt context, but we set the ONESHOT flag for those interrupts as
well.
Kohji Okuno debugged a problem with the SDHCI driver where the
interrupt thread waits for a hardware interrupt to trigger, which can't
work well because the hardware interrupt is masked due to the ONESHOT
flag being set. He proposed to set the ONESHOT flag only if the
interrupt does not provide a thread handler.
Though that does not work either because these interrupts can be
shared. So the other interrupt would rightfully get the ONESHOT flag
set and therefor the same situation would happen again.
To deal with this proper, we need to force thread the primary handler
of such interrupts as well. That means that the primary interrupt
handler is treated as any other primary interrupt handler which is not
marked IRQF_NO_THREAD. The threaded handler becomes a separate thread
so the SDHCI flow logic can be handled gracefully.
The same issue was reported against 4.1-rt.
Reported-and-tested-by: Kohji Okuno <okuno.kohji@jp.panasonic.com>
Reported-By: Michal Smucr <msmucr@gmail.com>
Reported-and-tested-by: Nathan Sullivan <nathan.sullivan@ni.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: http://lkml.kernel.org/r/alpine.DEB.2.11.1509211058080.5606@nanos
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
No user was ever interested whether the timer was active or not when
it was started. All abusers of the return value are gone, so get rid
of it.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: Preeti U Murthy <preeti@linux.vnet.ibm.com>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Link: http://lkml.kernel.org/r/20150414203503.483556394@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
hrtimer softirq is a leftover from the initial implementation and
serves only the purpose to handle the enqueueing of already expired
timers in the high resolution timer mode. We discussed whether we
change the return value and force all start sites to handle that the
timer is already expired, but that would be a Herculean task and I'm
not sure whether its a good idea to enforce that handling on
everyone.
A simpler solution is to enforce a timer interrupt instead of raising
and scheduling a softirq. Just use the existing infrastructure to do
so and remove all the softirq leftovers.
The HRTIMER softirq enum is now unused, but kept around because trace
parsers rely on the existing numbering.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: Preeti U Murthy <preeti@linux.vnet.ibm.com>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Link: http://lkml.kernel.org/r/20150414203501.840834708@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
There is a number of cases where a kernel subsystem may want to
introspect the state of an interrupt at the irqchip level:
- When a peripheral is shared between virtual machines,
its interrupt state becomes part of the guest's state,
and must be switched accordingly. KVM on arm/arm64 requires
this for its guest-visible timer
- Some GPIO controllers seem to require peeking into the
interrupt controller they are connected to to report
their internal state
This seem to be a pattern that is common enough for the core code
to try and support this without too many horrible hacks. Introduce
a pair of accessors (irq_get_irqchip_state/irq_set_irqchip_state)
to retrieve the bits that can be of interest to another subsystem:
pending, active, and masked.
- irq_get_irqchip_state returns the state of the interrupt according
to a parameter set to IRQCHIP_STATE_PENDING, IRQCHIP_STATE_ACTIVE,
IRQCHIP_STATE_MASKED or IRQCHIP_STATE_LINE_LEVEL.
- irq_set_irqchip_state similarly sets the state of the interrupt.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Bjorn Andersson <bjorn.andersson@sonymobile.com>
Tested-by: Bjorn Andersson <bjorn.andersson@sonymobile.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: Abhijeet Dharmapurikar <adharmap@codeaurora.org>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: Phong Vo <pvo@apm.com>
Cc: Linus Walleij <linus.walleij@linaro.org>
Cc: Tin Huynh <tnhuynh@apm.com>
Cc: Y Vo <yvo@apm.com>
Cc: Toan Le <toanle@apm.com>
Cc: Bjorn Andersson <bjorn@kryo.se>
Cc: Jason Cooper <jason@lakedaemon.net>
Cc: Arnd Bergmann <arnd@arndb.de>
Link: http://lkml.kernel.org/r/1426676484-21812-2-git-send-email-marc.zyngier@arm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The IRQF_DISABLED flag is a NOOP and has been scheduled for removal
since Linux v2.6.36 by commit 6932bf37be ("genirq: Remove
IRQF_DISABLED from core code").
According to commit e58aa3d2d0 ("genirq: Run irq handlers with
interrupts disabled"), running IRQ handlers with interrupts
enabled can cause stack overflows when the interrupt line of the
issuing device is still active.
This patch ends the grace period for IRQF_DISABLED (i.e.,
SA_INTERRUPT in older versions of Linux) and removes the
definition and all remaining usages of this flag.
There's still a few non-functional references left in the kernel
source:
- The bigger hunk in Documentation/scsi/ncr53c8xx.txt is removed entirely
as IRQF_DISABLED is gone now; the usage in older kernel versions
(including the old SA_INTERRUPT flag) should be discouraged. The
trouble of using IRQF_SHARED is a general problem and not specific to
any driver.
- I left the reference in Documentation/PCI/MSI-HOWTO.txt untouched since
it has already been removed in linux-next.
- All remaining references are changelogs that I suggest to keep.
Signed-off-by: Valentin Rothberg <valentinrothberg@gmail.com>
Cc: Afzal Mohammed <afzal@ti.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Brian Norris <computersforpeace@gmail.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Ewan Milne <emilne@redhat.com>
Cc: Eyal Perry <eyalpe@mellanox.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Hongliang Tao <taohl@lemote.com>
Cc: Huacai Chen <chenhc@lemote.com>
Cc: Jiri Kosina <jkosina@suse.cz>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Keerthy <j-keerthy@ti.com>
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Nishanth Menon <nm@ti.com>
Cc: Paul Bolle <pebolle@tiscali.nl>
Cc: Peter Ujfalusi <peter.ujfalusi@ti.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Quentin Lambert <lambert.quentin@gmail.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Santosh Shilimkar <santosh.shilimkar@ti.com>
Cc: Sricharan R <r.sricharan@ti.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tony Lindgren <tony@atomide.com>
Cc: Zhou Wang <wangzhou1@hisilicon.com>
Cc: iss_storagedev@hp.com
Cc: linux-mips@linux-mips.org
Cc: linux-mtd@lists.infradead.org
Link: http://lkml.kernel.org/r/1425565425-12604-1-git-send-email-valentinrothberg@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
It currently is required that all users of NO_SUSPEND interrupt
lines pass the IRQF_NO_SUSPEND flag when requesting the IRQ or the
WARN_ON_ONCE() in irq_pm_install_action() will trigger. That is
done to warn about situations in which unprepared interrupt handlers
may be run unnecessarily for suspended devices and may attempt to
access those devices by mistake. However, it may cause drivers
that have no technical reasons for using IRQF_NO_SUSPEND to set
that flag just because they happen to share the interrupt line
with something like a timer.
Moreover, the generic handling of wakeup interrupts introduced by
commit 9ce7a25849 (genirq: Simplify wakeup mechanism) only works
for IRQs without any NO_SUSPEND users, so the drivers of wakeup
devices needing to use shared NO_SUSPEND interrupt lines for
signaling system wakeup generally have to detect wakeup in their
interrupt handlers. Thus if they happen to share an interrupt line
with a NO_SUSPEND user, they also need to request that their
interrupt handlers be run after suspend_device_irqs().
In both cases the reason for using IRQF_NO_SUSPEND is not because
the driver in question has a genuine need to run its interrupt
handler after suspend_device_irqs(), but because it happens to
share the line with some other NO_SUSPEND user. Otherwise, the
driver would do without IRQF_NO_SUSPEND just fine.
To make it possible to specify that condition explicitly, introduce
a new IRQ action handler flag for shared IRQs, IRQF_COND_SUSPEND,
that, when set, will indicate to the IRQ core that the interrupt
user is generally fine with suspending the IRQ, but it also can
tolerate handler invocations after suspend_device_irqs() and, in
particular, it is capable of detecting system wakeup and triggering
it as appropriate from its interrupt handler.
That will allow us to work around a problem with a shared timer
interrupt line on at91 platforms.
Link: http://marc.info/?l=linux-kernel&m=142252777602084&w=2
Link: http://marc.info/?t=142252775300011&r=1&w=2
Link: https://lkml.org/lkml/2014/12/15/552
Reported-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Mark Rutland <mark.rutland@arm.com>
The IRQF_NO_SUSPEND flag is intended to be used for interrupts required
to be enabled during the suspend-resume cycle. This mostly consists of
IPIs and timer interrupts, potentially including chained irqchip
interrupts if these are necessary to handle timers or IPIs. If an
interrupt does not fall into one of the aforementioned categories,
requesting it with IRQF_NO_SUSPEND is likely incorrect.
Using IRQF_NO_SUSPEND does not guarantee that the interrupt can wake the
system from a suspended state. For an interrupt to be able to trigger a
wakeup, it may be necessary to program various components of the system.
In these cases it is necessary to use {enable,disabled}_irq_wake.
Unfortunately, several drivers assume that IRQF_NO_SUSPEND ensures that
an IRQ can wake up the system, and the documentation can be read
ambiguously w.r.t. this property.
This patch updates the documentation regarding IRQF_NO_SUSPEND to make
this caveat explicit, hopefully making future misuse rarer. Cleanup of
existing misuse will occur as part of later patch series.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
For things like netpoll there is a need to disable an interrupt from
atomic context. Currently netpoll uses disable_irq() which will
sleep-wait on threaded handlers and thus forced_irqthreads breaks
things.
Provide disable_hardirq(), which uses synchronize_hardirq() to only wait
for active hardirq handlers; also change synchronize_hardirq() to
return the status of threaded handlers.
This will allow one to try-disable an interrupt from atomic context, or
in case of request_threaded_irq() to only wait for the hardirq part.
Suggested-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: David Miller <davem@davemloft.net>
Cc: Eyal Perry <eyalpe@mellanox.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Quentin Lambert <lambert.quentin@gmail.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Russell King <linux@arm.linux.org.uk>
Link: http://lkml.kernel.org/r/20150205130623.GH5029@twins.programming.kicks-ass.net
[ Fixed typos and such. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Quentin Lambert <lambert.quentin@gmail.com>
Signed-off-by: Valentin Rothberg <valentinrothberg@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently we suspend wakeup interrupts by lazy disabling them and
check later whether the interrupt has fired, but that's not sufficient
for suspend to idle as there is no way to check that once we
transitioned into the CPU idle state.
So we change the mechanism in the following way:
1) Leave the wakeup interrupts enabled across suspend
2) Add a check to irq_may_run() which is called at the beginning of
each flow handler whether the interrupt is an armed wakeup source.
This check is basically free as it just extends the existing check
for IRQD_IRQ_INPROGRESS. So no new conditional in the hot path.
If the IRQD_WAKEUP_ARMED flag is set, then the interrupt is
disabled, marked as pending/suspended and the pm core is notified
about the wakeup event.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
[ rjw: syscore.c and put irq_pm_check_wakeup() into pm.c ]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Pull core irq updates from Thomas Gleixner:
"The irq department delivers:
- Another tree wide update to get rid of the horrible create_irq
interface along with its even more horrible variants. That also
gets rid of the last leftovers of the initial sparse irq hackery.
arch/driver specific changes have been either acked or ignored.
- A fix for the spurious interrupt detection logic with threaded
interrupts.
- A new ARM SoC interrupt controller
- The usual pile of fixes and improvements all over the place"
* 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (40 commits)
Documentation: brcmstb-l2: Add Broadcom STB Level-2 interrupt controller binding
irqchip: brcmstb-l2: Add Broadcom Set Top Box Level-2 interrupt controller
genirq: Improve documentation to match current implementation
ARM: iop13xx: fix msi support with sparse IRQ
genirq: Provide !SMP stub for irq_set_affinity_notifier()
irqchip: armada-370-xp: Move the devicetree binding documentation
irqchip: gic: Use mask field in GICC_IAR
genirq: Remove dynamic_irq mess
ia64: Use irq_init_desc
genirq: Replace dynamic_irq_init/cleanup
genirq: Remove irq_reserve_irq[s]
genirq: Replace reserve_irqs in core code
s390: Avoid call to irq_reserve_irqs()
s390: Remove pointless arch_show_interrupts()
s390: pci: Check return value of alloc_irq_desc() proper
sh: intc: Remove pointless irq_reserve_irqs() invocation
x86, irq: Remove pointless irq_reserve_irqs() call
genirq: Make create/destroy_irq() ia64 private
tile: Use SPARSE_IRQ
tile: pci: Use irq_alloc/free_hwirq()
...
Instead of requiring each consumer of the IRQ affinity notifier to have
themselves be explicitly dependent on CONFIG_SMP, make the definition of
struct irq_affinity_notify to exist independently of that config option
and introduce a stub for irq_set_affinity_notifier() under non SMP
configuration.
Fixes: 2eacc23 ("net/mlx4_core: Enforce irq affinity changes
immediatly")
Signed-off-by: Eyal Perry <eyalpe@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Cc: Ben Hutchings <bhutchings@solarflare.com>
Cc: Yevgeny Petrilin <yevgenyp@mellanox.com>
Cc: Or Gerlitz <ogerlitz@mellanox.com>
Cc: David S. Miller <davem@davemloft.net>
Link: http://lkml.kernel.org/r/1400597820-30685-1-git-send-email-amirv@mellanox.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Patch 01f8fa4f01 "genirq: Allow forcing cpu affinity of interrupts" added
an irq_force_affinity() function, and 30ccf03b4a "clocksource: Exynos_mct:
Use irq_force_affinity() in cpu bringup" subsequently uses it. However, the
driver can be used with CONFIG_SMP disabled, but the function declaration
is only available for CONFIG_SMP, leading to this build error:
drivers/clocksource/exynos_mct.c:431:3: error: implicit declaration of function 'irq_force_affinity' [-Werror=implicit-function-declaration]
irq_force_affinity(mct_irqs[MCT_L0_IRQ + cpu], cpumask_of(cpu));
This patch introduces a dummy helper function for the non-SMP case
that always returns success, to get rid of the build error.
Since the patches causing the problem are marked for stable backports,
this one should be as well.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Acked-by: Kukjin Kim <kgene.kim@samsung.com>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/5619084.0zmrrIUZLV@wuerfel
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Fix new kernel-doc warnings in <linux/interrupt.h>:
Warning(include/linux/interrupt.h:219): No description found for parameter 'cpumask'
Warning(include/linux/interrupt.h:219): Excess function parameter 'mask' description in 'irq_set_affinity'
Warning(include/linux/interrupt.h:236): No description found for parameter 'cpumask'
Warning(include/linux/interrupt.h:236): Excess function parameter 'mask' description in 'irq_force_affinity'
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Link: http://lkml.kernel.org/r/535DD2FD.7030804@infradead.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The current implementation of irq_set_affinity() refuses rightfully to
route an interrupt to an offline cpu.
But there is a special case, where this is actually desired. Some of
the ARM SoCs have per cpu timers which require setting the affinity
during cpu startup where the cpu is not yet in the online mask.
If we can't do that, then the local timer interrupt for the about to
become online cpu is routed to some random online cpu.
The developers of the affected machines tried to work around that
issue, but that results in a massive mess in that timer code.
We have a yet unused argument in the set_affinity callbacks of the irq
chips, which I added back then for a similar reason. It was never
required so it got not used. But I'm happy that I never removed it.
That allows us to implement a sane handling of the above scenario. So
the affected SoC drivers can add the required force handling to their
interrupt chip, switch the timer code to irq_force_affinity() and
things just work.
This does not affect any existing user of irq_set_affinity().
Tagged for stable to allow a simple fix of the affected SoC clock
event drivers.
Reported-and-tested-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Cc: Tomasz Figa <t.figa@samsung.com>,
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>,
Cc: Kukjin Kim <kgene.kim@samsung.com>
Cc: linux-arm-kernel@lists.infradead.org,
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/20140416143315.717251504@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
In course of the sdhci/sdio discussion with Russell about killing the
sdio kthread hackery we discovered the need to be able to wake an
interrupt thread from software.
The rationale for this is, that sdio hardware can lack proper
interrupt support for certain features. So the driver needs to poll
the status registers, but at the same time it needs to be woken up by
an hardware interrupt.
To be able to get rid of the home brewn kthread construct of sdio we
need a way to wake an irq thread independent of an actual hardware
interrupt.
Provide an irq_wake_thread() function which wakes up the thread which
is associated to a given dev_id. This allows sdio to invoke the irq
thread from the hardware irq handler via the IRQ_WAKE_THREAD return
value and provides a possibility to wake it via a timer for the
polling scenarios. That allows to simplify the sdio logic
significantly.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Chris Ball <chris@printf.net>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20140215003823.772565780@linutronix.de
This commit was incomplete in that code to remove items from the per-cpu
lists was missing and never acquired a user in the 5 years it has been in
the tree. We're going to implement what it seems to try to archive in a
simpler way, and this code is in the way of doing so.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Cc: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
All arch overriden implementations of do_softirq() share the following
common code: disable irqs (to avoid races with the pending check),
check if there are softirqs pending, then execute __do_softirq() on
a specific stack.
Consolidate the common parts such that archs only worry about the
stack switch.
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@au1.ibm.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul Mackerras <paulus@au1.ibm.com>
Cc: James Hogan <james.hogan@imgtec.com>
Cc: James E.J. Bottomley <jejb@parisc-linux.org>
Cc: Helge Deller <deller@gmx.de>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Andrew Morton <akpm@linux-foundation.org>
After the last architecture switched to generic hard irqs the config
options HAVE_GENERIC_HARDIRQS & GENERIC_HARDIRQS and the related code
for !CONFIG_GENERIC_HARDIRQS can be removed.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
In some cases, free_irq_cpu_rmap() is called while holding a lock (eg
rtnl). This can lead to deadlocks, because it invokes
flush_scheduled_work() which ends up waiting for whole system workqueue
to flush, but some pending works might try to acquire the lock we are
already holding.
This commit uses reference-counting to replace
irq_run_affinity_notifiers(). It also removes
irq_run_affinity_notifiers() altogether.
[akpm@linux-foundation.org: eliminate free_cpu_rmap, rename cpu_rmap_reclaim() to cpu_rmap_release(), propagate kref_put() retval from cpu_rmap_put()]
Signed-off-by: David Decotigny <decot@googlers.com>
Reviewed-by: Ben Hutchings <bhutchings@solarflare.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Or Gerlitz <ogerlitz@mellanox.com>
Acked-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The can_stop_idle_tick() function complains if a softirq vector is
raised too late in the idle-entry process, presumably in order to
prevent dangling softirq invocations from being delayed across the
full idle period, which might be indefinitely long -- and if softirq
was asserted any later than the call to this function, such a delay
might well happen.
However, RCU needs to be able to use softirq to stop idle entry in
order to be able to drain RCU callbacks from the current CPU, which in
turn enables faster entry into dyntick-idle mode, which in turn reduces
power consumption. Because RCU takes this action at a well-defined
point in the idle-entry path, it is safe for RCU to take this approach.
This commit therefore silences the error message that is sometimes
produced when the going-idle CPU suddenly finds that it has an RCU_SOFTIRQ
to process. The error message will continue to be issued for other
softirq vectors.
Reported-by: Sedat Dilek <sedat.dilek@gmail.com>
Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Sedat Dilek <sedat.dilek@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
exit_irq_thread() and task->irq_thread are needed to handle the unexpected
(and unlikely) exit of irq-thread.
We can use task_work instead and make this all private to
kernel/irq/manage.c, cleanup plus micro-optimization.
1. rename exit_irq_thread() to irq_thread_dtor(), make it
static, and move it up before irq_thread().
2. change irq_thread() to do task_work_add(irq_thread_dtor)
at the start and task_work_cancel() before return.
tracehook_notify_resume() can never play with kthreads,
only do_exit()->exit_task_work() can call the callback
and this is what we want.
3. remove task_struct->irq_thread and the special hook
in do_exit().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Cc: David Howells <dhowells@redhat.com>
Cc: Richard Kuo <rkuo@codeaurora.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Alexander Gordeev <agordeev@redhat.com>
Cc: Chris Zankel <chris@zankel.net>
Cc: David Smith <dsmith@redhat.com>
Cc: "Frank Ch. Eigler" <fche@redhat.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Larry Woodman <lwoodman@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
There is no need to have flags as unsigned long. Make it unsigned int
and put it together with irq so it does not waste space on 64 bit.
Make irq unsigned int while at it.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Remove all #inclusions of asm/system.h preparatory to splitting and killing
it. Performed with the following command:
perl -p -i -e 's!^#\s*include\s*<asm/system[.]h>.*\n!!' `grep -Irl '^#\s*include\s*<asm/system[.]h>' *`
Signed-off-by: David Howells <dhowells@redhat.com>
The __raise_softirq_irqoff() contains a tracepoint. As tracepoints in headers
can cause issues, and not to mention, bloats the kernel when they are
in a static inline, it is best to move the function that contains the
tracepoint out of the header and into softirq.c.
Link: http://lkml.kernel.org/r/20120118120711.GB14863@elte.hu
Suggested-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
This adds a mechanism to resume selected IRQs during syscore_resume
instead of dpm_resume_noirq.
Under Xen we need to resume IRQs associated with IPIs early enough
that the resched IPI is unmasked and we can therefore schedule
ourselves out of the stop_machine where the suspend/resume takes
place.
This issue was introduced by 676dc3cf5b "xen: Use IRQF_FORCE_RESUME".
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Jeremy Fitzhardinge <Jeremy.Fitzhardinge@citrix.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Link: http://lkml.kernel.org/r/1318713254.11016.52.camel@dagon.hellion.org.uk
Cc: stable@kernel.org (at least to 2.6.32.y)
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
As request_percpu_irq() doesn't allow for a percpu interrupt to have
its type configured (it is generally impossible to configure it on all
CPUs at once), add a 'type' argument to enable_percpu_irq().
This allows some low-level, board specific init code to be switched to
a generic API.
[ tglx: Added WARN_ON argument ]
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Cc: Abhijeet Dharmapurikar <adharmap@codeaurora.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The ARM GIC interrupt controller offers per CPU interrupts (PPIs),
which are usually used to connect local timers to each core. Each CPU
has its own private interface to the GIC, and only sees the PPIs that
are directly connect to it.
While these timers are separate devices and have a separate interrupt
line to a core, they all use the same IRQ number.
For these devices, request_irq() is not the right API as it assumes
that an IRQ number is visible by a number of CPUs (through the
affinity setting), but makes it very awkward to express that an IRQ
number can be handled by all CPUs, and yet be a different interrupt
line on each CPU, requiring a different dev_id cookie to be passed
back to the handler.
The *_percpu_irq() functions is designed to overcome these
limitations, by providing a per-cpu dev_id vector:
int request_percpu_irq(unsigned int irq, irq_handler_t handler,
const char *devname, void __percpu *percpu_dev_id);
void free_percpu_irq(unsigned int, void __percpu *);
int setup_percpu_irq(unsigned int irq, struct irqaction *new);
void remove_percpu_irq(unsigned int irq, struct irqaction *act);
void enable_percpu_irq(unsigned int irq);
void disable_percpu_irq(unsigned int irq);
The API has a number of limitations:
- no interrupt sharing
- no threading
- common handler across all the CPUs
Once the interrupt is requested using setup_percpu_irq() or
request_percpu_irq(), it must be enabled by each core that wishes its
local interrupt to be delivered.
Based on an initial patch by Thomas Gleixner.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Cc: linux-arm-kernel@lists.infradead.org
Link: http://lkml.kernel.org/r/1316793788-14500-2-git-send-email-marc.zyngier@arm.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This allows us to move duplicated code in <asm/atomic.h>
(atomic_inc_not_zero() for now) to <linux/atomic.h>
Signed-off-by: Arun Sharma <asharma@fb.com>
Reviewed-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit a26ac2455ffcf3(rcu: move TREE_RCU from softirq to kthread)
introduced performance regression. In an AIM7 test, this commit degraded
performance by about 40%.
The commit runs rcu callbacks in a kthread instead of softirq. We observed
high rate of context switch which is caused by this. Out test system has
64 CPUs and HZ is 1000, so we saw more than 64k context switch per second
which is caused by RCU's per-CPU kthread. A trace showed that most of
the time the RCU per-CPU kthread doesn't actually handle any callbacks,
but instead just does a very small amount of work handling grace periods.
This means that RCU's per-CPU kthreads are making the scheduler do quite
a bit of work in order to allow a very small amount of RCU-related
processing to be done.
Alex Shi's analysis determined that this slowdown is due to lock
contention within the scheduler. Unfortunately, as Peter Zijlstra points
out, the scheduler's real-time semantics require global action, which
means that this contention is inherent in real-time scheduling. (Yes,
perhaps someone will come up with a workaround -- otherwise, -rt is not
going to do well on large SMP systems -- but this patch will work around
this issue in the meantime. And "the meantime" might well be forever.)
This patch therefore re-introduces softirq processing to RCU, but only
for core RCU work. RCU callbacks are still executed in kthread context,
so that only a small amount of RCU work runs in softirq context in the
common case. This should minimize ksoftirqd execution, allowing us to
skip boosting of ksoftirqd for CONFIG_RCU_BOOST=y kernels.
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Tested-by: "Alex,Shi" <alex.shi@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
If RCU priority boosting is to be meaningful, callback invocation must
be boosted in addition to preempted RCU readers. Otherwise, in presence
of CPU real-time threads, the grace period ends, but the callbacks don't
get invoked. If the callbacks don't get invoked, the associated memory
doesn't get freed, so the system is still subject to OOM.
But it is not reasonable to priority-boost RCU_SOFTIRQ, so this commit
moves the callback invocations to a kthread, which can be boosted easily.
Also add comments and properly synchronized all accesses to
rcu_cpu_kthread_task, as suggested by Lai Jiangshan.
Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Christoph Hellwig