Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=64791
When a cpu is downed on a system, the irqs on the cpu are assigned to
other cpus. It is possible, however, that when a cpu is downed there
aren't enough free vectors on the remaining cpus to account for the
vectors from the cpu that is being downed.
This results in an interesting "overflow" condition where irqs are
"assigned" to a CPU but are not handled.
For example, when downing cpus on a 1-64 logical processor system:
<snip>
[ 232.021745] smpboot: CPU 61 is now offline
[ 238.480275] smpboot: CPU 62 is now offline
[ 245.991080] ------------[ cut here ]------------
[ 245.996270] WARNING: CPU: 0 PID: 0 at net/sched/sch_generic.c:264 dev_watchdog+0x246/0x250()
[ 246.005688] NETDEV WATCHDOG: p786p1 (ixgbe): transmit queue 0 timed out
[ 246.013070] Modules linked in: lockd sunrpc iTCO_wdt iTCO_vendor_support sb_edac ixgbe microcode e1000e pcspkr joydev edac_core lpc_ich ioatdma ptp mdio mfd_core i2c_i801 dca pps_core i2c_core wmi acpi_cpufreq isci libsas scsi_transport_sas
[ 246.037633] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 3.12.0+ #14
[ 246.044451] Hardware name: Intel Corporation S4600LH ........../SVRBD-ROW_T, BIOS SE5C600.86B.01.08.0003.022620131521 02/26/2013
[ 246.057371] 0000000000000009 ffff88081fa03d40 ffffffff8164fbf6 ffff88081fa0ee48
[ 246.065728] ffff88081fa03d90 ffff88081fa03d80 ffffffff81054ecc ffff88081fa13040
[ 246.074073] 0000000000000000 ffff88200cce0000 0000000000000040 0000000000000000
[ 246.082430] Call Trace:
[ 246.085174] <IRQ> [<ffffffff8164fbf6>] dump_stack+0x46/0x58
[ 246.091633] [<ffffffff81054ecc>] warn_slowpath_common+0x8c/0xc0
[ 246.098352] [<ffffffff81054fb6>] warn_slowpath_fmt+0x46/0x50
[ 246.104786] [<ffffffff815710d6>] dev_watchdog+0x246/0x250
[ 246.110923] [<ffffffff81570e90>] ? dev_deactivate_queue.constprop.31+0x80/0x80
[ 246.119097] [<ffffffff8106092a>] call_timer_fn+0x3a/0x110
[ 246.125224] [<ffffffff8106280f>] ? update_process_times+0x6f/0x80
[ 246.132137] [<ffffffff81570e90>] ? dev_deactivate_queue.constprop.31+0x80/0x80
[ 246.140308] [<ffffffff81061db0>] run_timer_softirq+0x1f0/0x2a0
[ 246.146933] [<ffffffff81059a80>] __do_softirq+0xe0/0x220
[ 246.152976] [<ffffffff8165fedc>] call_softirq+0x1c/0x30
[ 246.158920] [<ffffffff810045f5>] do_softirq+0x55/0x90
[ 246.164670] [<ffffffff81059d35>] irq_exit+0xa5/0xb0
[ 246.170227] [<ffffffff8166062a>] smp_apic_timer_interrupt+0x4a/0x60
[ 246.177324] [<ffffffff8165f40a>] apic_timer_interrupt+0x6a/0x70
[ 246.184041] <EOI> [<ffffffff81505a1b>] ? cpuidle_enter_state+0x5b/0xe0
[ 246.191559] [<ffffffff81505a17>] ? cpuidle_enter_state+0x57/0xe0
[ 246.198374] [<ffffffff81505b5d>] cpuidle_idle_call+0xbd/0x200
[ 246.204900] [<ffffffff8100b7ae>] arch_cpu_idle+0xe/0x30
[ 246.210846] [<ffffffff810a47b0>] cpu_startup_entry+0xd0/0x250
[ 246.217371] [<ffffffff81646b47>] rest_init+0x77/0x80
[ 246.223028] [<ffffffff81d09e8e>] start_kernel+0x3ee/0x3fb
[ 246.229165] [<ffffffff81d0989f>] ? repair_env_string+0x5e/0x5e
[ 246.235787] [<ffffffff81d095a5>] x86_64_start_reservations+0x2a/0x2c
[ 246.242990] [<ffffffff81d0969f>] x86_64_start_kernel+0xf8/0xfc
[ 246.249610] ---[ end trace fb74fdef54d79039 ]---
[ 246.254807] ixgbe 0000:c2:00.0 p786p1: initiating reset due to tx timeout
[ 246.262489] ixgbe 0000:c2:00.0 p786p1: Reset adapter
Last login: Mon Nov 11 08:35:14 from 10.18.17.119
[root@(none) ~]# [ 246.792676] ixgbe 0000:c2:00.0 p786p1: detected SFP+: 5
[ 249.231598] ixgbe 0000:c2:00.0 p786p1: NIC Link is Up 10 Gbps, Flow Control: RX/TX
[ 246.792676] ixgbe 0000:c2:00.0 p786p1: detected SFP+: 5
[ 249.231598] ixgbe 0000:c2:00.0 p786p1: NIC Link is Up 10 Gbps, Flow Control: RX/TX
(last lines keep repeating. ixgbe driver is dead until module reload.)
If the downed cpu has more vectors than are free on the remaining cpus on the
system, it is possible that some vectors are "orphaned" even though they are
assigned to a cpu. In this case, since the ixgbe driver had a watchdog, the
watchdog fired and notified that something was wrong.
This patch adds a function, check_vectors(), to compare the number of vectors
on the CPU going down and compares it to the number of vectors available on
the system. If there aren't enough vectors for the CPU to go down, an
error is returned and propogated back to userspace.
v2: Do not need to look at percpu irqs
v3: Need to check affinity to prevent counting of MSIs in IOAPIC Lowest
Priority Mode
v4: Additional changes suggested by Gong Chen.
v5/v6/v7/v8: Updated comment text
Signed-off-by: Prarit Bhargava <prarit@redhat.com>
Link: http://lkml.kernel.org/r/1389613861-3853-1-git-send-email-prarit@redhat.com
Reviewed-by: Gong Chen <gong.chen@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Michel Lespinasse <walken@google.com>
Cc: Seiji Aguchi <seiji.aguchi@hds.com>
Cc: Yang Zhang <yang.z.zhang@Intel.com>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Janet Morgan <janet.morgan@intel.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Ruiv Wang <ruiv.wang@gmail.com>
Cc: Gong Chen <gong.chen@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Cc: <stable@vger.kernel.org>
These handlers are all referenced from assembler stubs, so need
to be visible.
The handlers without arguments become asmlinkage, the others __visible
to not force regparms(0) on x86-32.
I put it all into a single patch, please let me know if you want
it it split up.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Link: http://lkml.kernel.org/r/1375740170-7446-4-git-send-email-andi@firstfloor.org
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
The following change fixes the x86 implementation of
trigger_all_cpu_backtrace(), which was previously (accidentally,
as far as I can tell) disabled to always return false as on
architectures that do not implement this function.
trigger_all_cpu_backtrace(), as defined in include/linux/nmi.h,
should call arch_trigger_all_cpu_backtrace() if available, or
return false if the underlying arch doesn't implement this
function.
x86 did provide a suitable arch_trigger_all_cpu_backtrace()
implementation, but it wasn't actually being used because it was
declared in asm/nmi.h, which linux/nmi.h doesn't include. Also,
linux/nmi.h couldn't easily be fixed by including asm/nmi.h,
because that file is not available on all architectures.
I am proposing to fix this by moving the x86 definition of
arch_trigger_all_cpu_backtrace() to asm/irq.h.
Tested via: echo l > /proc/sysrq-trigger
Before the change, this uses a fallback implementation which
shows backtraces on active CPUs (using
smp_call_function_interrupt() )
After the change, this shows NMI backtraces on all CPUs
Signed-off-by: Michel Lespinasse <walken@google.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/1370518875-1346-1-git-send-email-walken@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This patch adds minimal support for device tree on x86. The device
tree blob is passed to the kernel via setup_data which requires at
least boot protocol 2.09.
Memory size, restricted memory regions, boot arguments are gathered
the traditional way so things like cmd_line are just here to let the
code compile.
The current plan is use the device tree as an extension and to gather
information which can not be enumerated and would have to be hardcoded
otherwise. This includes things like
- which devices are on this I2C/SPI bus?
- how are the interrupts wired to IO APIC?
- where could my hpet be?
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Dirk Brandewie <dirk.brandewie@gmail.com>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Cc: sodaville@linutronix.de
Cc: devicetree-discuss@lists.ozlabs.org
LKML-Reference: <1298405266-1624-3-git-send-email-bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* 'x86-olpc-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86, olpc: Speed up device tree creation during boot
x86, olpc: Add OLPC device-tree support
x86, of: Define irq functions to allow drivers/of/* to build on x86
The x86 arch has shifted its use of the nmi_watchdog from a
local implementation to the global one provide by
kernel/watchdog.c. This shift has caused a whole bunch of
compile problems under different config options. I attempt to
simplify things with the patch below.
In order to simplify things, I had to come to terms with the
meaning of two terms ARCH_HAS_NMI_WATCHDOG and
CONFIG_HARDLOCKUP_DETECTOR. Basically they mean the same thing,
the former on a local level and the latter on a global level.
With the old x86 nmi watchdog gone, there is no need to rely on
defining the ARCH_HAS_NMI_WATCHDOG variable because it doesn't
make sense any more. x86 will now use the global
implementation.
The changes below do a few things. First it changes the few
places that relied on ARCH_HAS_NMI_WATCHDOG to use
CONFIG_X86_LOCAL_APIC (the former was an alias for the latter
anyway, so nothing unusual here). Those pieces of code were
relying more on local apic functionality the nmi watchdog
functionality, so the change should make sense.
Second, I removed the x86 implementation of
touch_nmi_watchdog(). It isn't need now, instead x86 will rely
on kernel/watchdog.c's implementation.
Third, I removed the #define ARCH_HAS_NMI_WATCHDOG itself from
x86. And tweaked the include/linux/nmi.h file to tell users to
look for an externally defined touch_nmi_watchdog in the case of
ARCH_HAS_NMI_WATCHDOG _or_ CONFIG_HARDLOCKUP_DETECTOR. This
changes removes some of the ugliness in that file.
Finally, I added a Kconfig dependency for
CONFIG_HARDLOCKUP_DETECTOR that said you can't have
ARCH_HAS_NMI_WATCHDOG _and_ CONFIG_HARDLOCKUP_DETECTOR. You can
only have one nmi_watchdog.
Tested with
ARCH=i386: allnoconfig, defconfig, allyesconfig, (various broken
configs) ARCH=x86_64: allnoconfig, defconfig, allyesconfig,
(various broken configs)
Hopefully, after this patch I won't get any more compile broken
emails. :-)
v3:
changed a couple of 'linux/nmi.h' -> 'asm/nmi.h' to pick-up correct function
prototypes when CONFIG_HARDLOCKUP_DETECTOR is not set.
Signed-off-by: Don Zickus <dzickus@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: fweisbec@gmail.com
LKML-Reference: <1293044403-14117-1-git-send-email-dzickus@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
- Define a stub irq_create_of_mapping for x86 as a stop-gap solution until
drivers/of/irq is further along.
- Define irq_dispose_mapping for x86 to appease of_i2c.c
These are needed to allow stuff in drivers/of/ to build on x86. This stuff
will eventually get replaced; quoting Grant,
"The long term plan is to have the drivers/of/ code handling the mapping
intelligently like powerpc currently does." But for now, just provide
these functions.
Signed-off-by: Andres Salomon <dilinger@queued.net>
LKML-Reference: <20101111214526.5de7121b@queued.net>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
The percpu allocator cannot handle alignments larger than one
page. Allocate the irq stacks seperately, and only keep the
pointers as percpu data.
Signed-off-by: Brian Gerst <brgerst@gmail.com>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: tj@kernel.org
LKML-Reference: <1288158182-1753-1-git-send-email-brgerst@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
IRQ stacks provide much better safety against unexpected stack use from
interrupts, at the minimal downside of slightly higher memory usage.
Enable irq stacks also for the default 8k stack on 32-bit kernels to
minimize the problem of stack overflows through interrupt activity.
This is what the 64-bit kernel and various other architectures already do.
Signed-off-by: Christoph Hellwig <hch@lst.de>
LKML-Reference: <20100628121554.GA6605@lst.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
nr_legacy_irqs and its ilk have moved to legacy_pic.
-v2: there is one in ioapic_.c
Singed-off-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <4B84AAC4.2020204@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Currently IRQ0..IRQ15 are assigned to IRQ0_VECTOR..IRQ15_VECTOR's on
all the cpu's.
If these IRQ's are handled by legacy pic controller, then the kernel
handles them only on cpu 0. So there is no need to block this vector
space on all cpu's.
Similarly if these IRQ's are handled by IO-APIC, then the IRQ affinity
will determine on which cpu's we need allocate the vector resource for
that particular IRQ. This can be done dynamically and here also there
is no need to block 16 vectors for IRQ0..IRQ15 on all cpu's.
Fix this by initially assigning IRQ0..IRQ15 to IRQ0_VECTOR..IRQ15_VECTOR's only
on cpu 0. If the legacy controllers like pic handles these irq's, then
this configuration will be fixed. If more modern controllers like IO-APIC
handle these IRQ's, then we start with this configuration and as IRQ's
migrate, vectors (/and cpu's) associated with these IRQ's change dynamically.
This will freeup the block of 16 vectors on other cpu's which don't handle
IRQ0..IRQ15, which can now be used for other IRQ's that the particular cpu
handle.
[ hpa: this also an architectural cleanup for future legacy-PIC-free
configurations. ]
[ hpa: fixed typo NR_LEGACY_IRQS -> NR_IRQS_LEGACY ]
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1263932453.2814.52.camel@sbs-t61.sc.intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
When a cpu goes offline, fixup_irqs() try to move irq's
currently destined to the offline cpu to a new cpu. But this
attempt will fail if the irq is recently moved to this cpu and
the irq still hasn't arrived at this cpu (for non intr-remapping
platforms this is when we free the vector allocation at the
previous destination) that is about to go offline.
This will endup with the interrupt subsystem still pointing the
irq to the offline cpu, causing that irq to not work any more.
Fix this by forcing the irq to complete its move (its been a
long time we moved the irq to this cpu which we are offlining
now) and then move this irq to a new cpu before this cpu goes
offline.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Acked-by: Gary Hade <garyhade@us.ibm.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
LKML-Reference: <20091026230001.848830905@sbs-t61.sc.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
irq_init is overridden by x86_quirks and by paravirts. Unify the whole
mess and make it an unconditional x86_init_ops function which defaults
to the standard function and can be overridden by the early platform
code.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Replace the quirk machinery by a x86_init_ops function which
defaults to the standard implementation. This is also a preparatory
patch for Moorestown support which needs to replace the default
init_ISA_irqs as well.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This patch allocates a system interrupt vector for various platform
specific uses.
Signed-off-by: Dimitri Sivanich <sivanich@sgi.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: john stultz <johnstul@us.ibm.com>
LKML-Reference: <20090304185605.GA24419@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
With the differences in interrupt handling hoisted into handle_irq(),
do_IRQ is more or less identical between 32 and 64 bit, so unify it.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Xen uses a different interrupt path, so introduce handle_irq() to
allow interrupts to be inserted into the normal interrupt path. This
is handled slightly differently on 32 and 64-bit.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* 'cpus4096-for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (66 commits)
x86: export vector_used_by_percpu_irq
x86: use logical apicid in x2apic_cluster's x2apic_cpu_mask_to_apicid_and()
sched: nominate preferred wakeup cpu, fix
x86: fix lguest used_vectors breakage, -v2
x86: fix warning in arch/x86/kernel/io_apic.c
sched: fix warning in kernel/sched.c
sched: move test_sd_parent() to an SMP section of sched.h
sched: add SD_BALANCE_NEWIDLE at MC and CPU level for sched_mc>0
sched: activate active load balancing in new idle cpus
sched: bias task wakeups to preferred semi-idle packages
sched: nominate preferred wakeup cpu
sched: favour lower logical cpu number for sched_mc balance
sched: framework for sched_mc/smt_power_savings=N
sched: convert BALANCE_FOR_xx_POWER to inline functions
x86: use possible_cpus=NUM to extend the possible cpus allowed
x86: fix cpu_mask_to_apicid_and to include cpu_online_mask
x86: update io_apic.c to the new cpumask code
x86: Introduce topology_core_cpumask()/topology_thread_cpumask()
x86: xen: use smp_call_function_many()
x86: use work_on_cpu in x86/kernel/cpu/mcheck/mce_amd_64.c
...
Fixed up trivial conflict in kernel/time/tick-sched.c manually
Impact: fix lguest, clean up
32-bit lguest used used_vectors to record vectors, but that model of
allocating vectors changed and got broken, after we changed vector
allocation to a per_cpu array.
Try enable that for 64bit, and the array is used for all vectors that
are not managed by vector_irq per_cpu array.
Also kill system_vectors[], that is now a duplication of the
used_vectors bitmap.
[ merged in cpus4096 due to io_apic.c cpumask changes. ]
[ -v2, fix build failure ]
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup, remove on-stack cpumask.
The "map" arg is always cpu_online_mask. Importantly, set_affinity
always ands the argument with cpu_online_mask anyway, so we don't need
to do it in fixup_irqs(), avoiding a temporary.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Impact: cleanup
CONFIG_IRQBALANCE was removed in commit 8b8e8c1bf; this ifdef was still
around.
Signed-off-by: Dan McGee <dpmcgee@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Change header guards named "ASM_X86__*" to "_ASM_X86_*" since:
a. the double underscore is ugly and pointless.
b. no leading underscore violates namespace constraints.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
