When a CPU is about to be offlined we call fixup_irqs() that resets IRQ
affinities related to the CPU in question. The same thing is also done when
the system is suspended to S-states like S3 (mem).
For each IRQ we try to complete any on-going move regardless whether the
IRQ is actually part of x86_vector_domain. For each IRQ descriptor we fetch
its chip_data, assume it is of type struct apic_chip_data and manipulate it
by clearing old_domain mask etc. For irq_chips that are not part of the
x86_vector_domain, like those created by various GPIO drivers, will find
their chip_data being changed unexpectly.
Below is an example where GPIO chip owned by pinctrl-sunrisepoint.c gets
corrupted after resume:
# cat /sys/kernel/debug/gpio
gpiochip0: GPIOs 360-511, parent: platform/INT344B:00, INT344B:00:
gpio-511 ( |sysfs ) in hi
# rtcwake -s10 -mmem
<10 seconds passes>
# cat /sys/kernel/debug/gpio
gpiochip0: GPIOs 360-511, parent: platform/INT344B:00, INT344B:00:
gpio-511 ( |sysfs ) in ?
Note '?' in the output. It means the struct gpio_chip ->get function is
NULL whereas before suspend it was there.
Fix this by first checking that the IRQ belongs to x86_vector_domain before
we try to use the chip_data as struct apic_chip_data.
Reported-and-tested-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Cc: stable@vger.kernel.org # 4.4+
Link: http://lkml.kernel.org/r/20161003101708.34795-1-mika.westerberg@linux.intel.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
If x86_vector_alloc_irq() fails x86_vector_free_irqs() is invoked to cleanup
the already allocated vectors. This subsequently calls clear_vector_irq().
The failed irq has no vector assigned, which triggers the BUG_ON(!vector) in
clear_vector_irq().
We cannot suppress the call to x86_vector_free_irqs() for the failed
interrupt, because the other data related to this irq must be cleaned up as
well. So calling clear_vector_irq() with vector == 0 is legitimate.
Remove the BUG_ON and return if vector is zero,
[ tglx: Massaged changelog ]
Fixes: b5dc8e6c21 "x86/irq: Use hierarchical irqdomain to manage CPU interrupt vectors"
Signed-off-by: Keith Busch <keith.busch@intel.com>
Cc: stable@vger.kernel.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Harry reported, that he's able to trigger a system freeze with cpu hot
unplug. The freeze turned out to be a live lock caused by recent changes in
irq_force_complete_move().
When fixup_irqs() and from there irq_force_complete_move() is called on the
dying cpu, then all other cpus are in stop machine an wait for the dying cpu
to complete the teardown. If there is a move of an interrupt pending then
irq_force_complete_move() sends the cleanup IPI to the cpus in the old_domain
mask and waits for them to clear the mask. That's obviously impossible as
those cpus are firmly stuck in stop machine with interrupts disabled.
I should have known that, but I completely overlooked it being concentrated on
the locking issues around the vectors. And the existance of the call to
__irq_complete_move() in the code, which actually sends the cleanup IPI made
it reasonable to wait for that cleanup to complete. That call was bogus even
before the recent changes as it was just a pointless distraction.
We have to look at two cases:
1) The move_in_progress flag of the interrupt is set
This means the ioapic has been updated with the new vector, but it has not
fired yet. In theory there is a race:
set_ioapic(new_vector) <-- Interrupt is raised before update is effective,
i.e. it's raised on the old vector.
So if the target cpu cannot handle that interrupt before the old vector is
cleaned up, we get a spurious interrupt and in the worst case the ioapic
irq line becomes stale, but my experiments so far have only resulted in
spurious interrupts.
But in case of cpu hotplug this should be a non issue because if the
affinity update happens right before all cpus rendevouz in stop machine,
there is no way that the interrupt can be blocked on the target cpu because
all cpus loops first with interrupts enabled in stop machine, so the old
vector is not yet cleaned up when the interrupt fires.
So the only way to run into this issue is if the delivery of the interrupt
on the apic/system bus would be delayed beyond the point where the target
cpu disables interrupts in stop machine. I doubt that it can happen, but at
least there is a theroretical chance. Virtualization might be able to
expose this, but AFAICT the IOAPIC emulation is not as stupid as the real
hardware.
I've spent quite some time over the weekend to enforce that situation,
though I was not able to trigger the delayed case.
2) The move_in_progress flag is not set and the old_domain cpu mask is not
empty.
That means, that an interrupt was delivered after the change and the
cleanup IPI has been sent to the cpus in old_domain, but not all CPUs have
responded to it yet.
In both cases we can assume that the next interrupt will arrive on the new
vector, so we can cleanup the old vectors on the cpus in the old_domain cpu
mask.
Fixes: 98229aa36c "x86/irq: Plug vector cleanup race"
Reported-by: Harry Junior <harryjr@outlook.fr>
Tested-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Joe Lawrence <joe.lawrence@stratus.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Ben Hutchings <ben@decadent.org.uk>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/alpine.DEB.2.11.1603140931430.3657@nanos
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
We still can end up with a stale vector due to the following:
CPU0 CPU1 CPU2
lock_vector()
data->move_in_progress=0
sendIPI()
unlock_vector()
set_affinity()
assign_irq_vector()
lock_vector() handle_IPI
move_in_progress = 1 lock_vector()
unlock_vector()
move_in_progress == 1
So we need to serialize the vector assignment against a pending cleanup. The
solution is rather simple now. We not only check for the move_in_progress flag
in assign_irq_vector(), we also check whether there is still a cleanup pending
in the old_domain cpumask. If so, we return -EBUSY to the caller and let him
deal with it. Though we have to be careful in the cpu unplug case. If the
cleanout has not yet completed then the following setaffinity() call would
return -EBUSY. Add code which prevents this.
Full context is here: http://lkml.kernel.org/r/5653B688.4050809@stratus.com
Reported-and-tested-by: Joe Lawrence <joe.lawrence@stratus.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Borislav Petkov <bp@alien8.de>
Cc: Jiang Liu <jiang.liu@linux.intel.com>
Cc: Jeremiah Mahler <jmmahler@gmail.com>
Cc: andy.shevchenko@gmail.com
Cc: Guenter Roeck <linux@roeck-us.net>
Cc: stable@vger.kernel.org #4.3+
Link: http://lkml.kernel.org/r/20151231160107.207265407@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
send_cleanup_vector() fiddles with the old_domain mask unprotected because it
relies on the protection by the move_in_progress flag. But this is fatal, as
the flag is reset after the IPI has been sent. So a cpu which receives the IPI
can still see the flag set and therefor ignores the cleanup request. If no
other cleanup request happens then the vector stays stale on that cpu and in
case of an irq removal the vector still persists. That can lead to use after
free when the next cleanup IPI happens.
Protect the code with vector_lock and clear move_in_progress before sending
the IPI.
This does not plug the race which Joe reported because:
CPU0 CPU1 CPU2
lock_vector()
data->move_in_progress=0
sendIPI()
unlock_vector()
set_affinity()
assign_irq_vector()
lock_vector() handle_IPI
move_in_progress = 1 lock_vector()
unlock_vector()
move_in_progress == 1
The full fix comes with a later patch.
Reported-and-tested-by: Joe Lawrence <joe.lawrence@stratus.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Borislav Petkov <bp@alien8.de>
Cc: Jiang Liu <jiang.liu@linux.intel.com>
Cc: Jeremiah Mahler <jmmahler@gmail.com>
Cc: andy.shevchenko@gmail.com
Cc: Guenter Roeck <linux@roeck-us.net>
Cc: stable@vger.kernel.org #4.3+
Link: http://lkml.kernel.org/r/20151231160106.892412198@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
__assign_irq_vector() uses the vector_cpumask which is assigned by
apic->vector_allocation_domain() without doing basic sanity checks. That can
result in a situation where the final assignement of a newly found vector
fails in apic->cpu_mask_to_apicid_and(). So we have to do rollbacks for no
reason.
apic->cpu_mask_to_apicid_and() only fails if
vector_cpumask & requested_cpumask & cpu_online_mask
is empty.
Check for this condition right away and if the result is empty try immediately
the next possible cpu in the requested mask. So in case of a failure the old
setting is unchanged and we can remove the rollback code.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Borislav Petkov <bp@alien8.de>
Tested-by: Joe Lawrence <joe.lawrence@stratus.com>
Cc: Jiang Liu <jiang.liu@linux.intel.com>
Cc: Jeremiah Mahler <jmmahler@gmail.com>
Cc: andy.shevchenko@gmail.com
Cc: Guenter Roeck <linux@roeck-us.net>
Cc: stable@vger.kernel.org #4.3+
Link: http://lkml.kernel.org/r/20151231160106.561877324@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
There's a race condition between
x86_vector_free_irqs()
{
free_apic_chip_data(irq_data->chip_data);
xxxxx //irq_data->chip_data has been freed, but the pointer
//hasn't been reset yet
irq_domain_reset_irq_data(irq_data);
}
and
smp_irq_move_cleanup_interrupt()
{
raw_spin_lock(&vector_lock);
data = apic_chip_data(irq_desc_get_irq_data(desc));
access data->xxxx // may access freed memory
raw_spin_unlock(&desc->lock);
}
which may cause smp_irq_move_cleanup_interrupt() to access freed memory.
Call irq_domain_reset_irq_data(), which clears the pointer with vector lock
held.
[ tglx: Free memory outside of lock held region. ]
Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com>
Tested-by: Borislav Petkov <bp@alien8.de>
Tested-by: Joe Lawrence <joe.lawrence@stratus.com>
Cc: Jeremiah Mahler <jmmahler@gmail.com>
Cc: andy.shevchenko@gmail.com
Cc: Guenter Roeck <linux@roeck-us.net>
Cc: stable@vger.kernel.org #4.3+
Link: http://lkml.kernel.org/r/1450880014-11741-3-git-send-email-jiang.liu@linux.intel.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Commit d32932d02e ("x86/irq: Convert IOAPIC to use hierarchical irqdomain
interfaces") brought a regression for Hyper-V Gen2 instances. These
instances don't have i8259 legacy PIC but they use legacy IRQs for serial
port, rtc, and acpi. With this commit included we end up with these IRQs
not initialized. Earlier, there was a special workaround for legacy IRQs
in mp_map_pin_to_irq() doing mp_irqdomain_map() without looking at
nr_legacy_irqs() and now we fail in __irq_domain_alloc_irqs() when
irq_domain_alloc_descs() returns -EEXIST.
The essence of the issue seems to be that early_irq_init() calls
arch_probe_nr_irqs() to figure out the number of legacy IRQs before
we probe for i8259 and gets 16. Later when init_8259A() is called we switch
to NULL legacy PIC and nr_legacy_irqs() starts to return 0 but we already
have 16 descs allocated.
Solve the issue by separating i8259 probe from init and calling it in
arch_probe_nr_irqs() before we actually use nr_legacy_irqs() information.
Fixes: d32932d02e ("x86/irq: Convert IOAPIC to use hierarchical irqdomain interfaces")
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Cc: Jiang Liu <jiang.liu@linux.intel.com>
Cc: K. Y. Srinivasan <kys@microsoft.com>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/1446543614-3621-1-git-send-email-vkuznets@redhat.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Pull x86 apic updates from Thomas Gleixner:
"This udpate contains:
- rework the irq vector array to store a pointer to the irq
descriptor instead of the irq number to avoid a lookup of the irq
descriptor in the irq entry path
- lguest interrupt handling cleanups
- conversion of the local apic timer to the new clockevent callbacks
- preparatory changes for the irq argument removal of interrupt flow
handlers"
* 'x86-apic-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/irq: Do not dereference irq descriptor before checking it
tools/lguest: Clean up include dir
tools/lguest: Fix redefinition of struct virtio_pci_cfg_cap
x86/irq: Store irq descriptor in vector array
genirq: Provide irq_desc_has_action
x86/irq: Get rid of an indentation level
x86/irq: Rename VECTOR_UNDEFINED to VECTOR_UNUSED
x86/irq: Replace numeric constant
x86/irq: Protect smp_cleanup_move
x86/lguest: Do not setup unused irq vectors
x86/lguest: Clean up lguest_setup_irq
x86/apic: Drop local_irq_save/restore in timer callbacks
x86/apic: Migrate apic timer to new set_state interface
x86/irq: Use access helper irq_data_get_affinity_mask()
x86/irq: Use accessor irq_data_get_irq_handler_data()
x86/irq: Use accessor irq_data_get_node()
Alex Deucher, Mark Rustad and Alexander Holler reported a regression
with the latest v4.2-rc4 kernel, which breaks some SATA controllers.
With multi-MSI capable SATA controllers, only the first port works,
all other ports time out when executing SATA commands.
This happens because the first argument to assign_irq_vector_policy()
is always the base linux irq number of the multi MSI interrupt block,
so all subsequent vector assignments operate on the base linux irq
number, so all MSI irqs are handled as the first irq number. Therefor
the other MSI irqs of a device are never set up correctly and never
fire.
Add the loop iterator to the base irq number so all vectors are
assigned correctly.
Fixes: b5dc8e6c21 "x86/irq: Use hierarchical irqdomain to manage CPU interrupt vectors"
Reported-and-tested-by: Alex Deucher <alexdeucher@gmail.com>
Reported-and-tested-by: Mark Rustad <mrustad@gmail.com>
Reported-and-tested-by: Alexander Holler <holler@ahsoftware.de>
Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com>
Cc: Tony Luck <tony.luck@intel.com>
Link: http://lkml.kernel.org/r/1439911228-9880-1-git-send-email-jiang.liu@linux.intel.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Use accessor irq_data_get_node() to hide struct irq_data
implementation detail, so we can move node to irq_data_common later.
Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com>
Cc: Borislav Petkov <bp@alien8.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Jin debugged a nasty cpu hotplug race which results in leaking a irq
vector on the newly hotplugged cpu.
cpu N cpu M
native_cpu_up device_shutdown
do_boot_cpu free_msi_irqs
start_secondary arch_teardown_msi_irqs
smp_callin default_teardown_msi_irqs
setup_vector_irq arch_teardown_msi_irq
__setup_vector_irq native_teardown_msi_irq
lock(vector_lock) destroy_irq
install vectors
unlock(vector_lock)
lock(vector_lock)
---> __clear_irq_vector
unlock(vector_lock)
lock(vector_lock)
set_cpu_online
unlock(vector_lock)
This leaves the irq vector(s) which are torn down on CPU M stale in
the vector array of CPU N, because CPU M does not see CPU N online
yet. There is a similar issue with concurrent newly setup interrupts.
The alloc/free protection of irq descriptors does not prevent the
above race, because it merily prevents interrupt descriptors from
going away or changing concurrently.
Prevent this by moving the call to setup_vector_irq() into the
vector_lock held region which protects set_cpu_online():
cpu N cpu M
native_cpu_up device_shutdown
do_boot_cpu free_msi_irqs
start_secondary arch_teardown_msi_irqs
smp_callin default_teardown_msi_irqs
lock(vector_lock) arch_teardown_msi_irq
setup_vector_irq()
__setup_vector_irq native_teardown_msi_irq
install vectors destroy_irq
set_cpu_online
unlock(vector_lock)
lock(vector_lock)
__clear_irq_vector
unlock(vector_lock)
So cpu M either sees the cpu N online before clearing the vector or
cpu N installs the vectors after cpu M has cleared it.
Reported-by: xiao jin <jin.xiao@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Joerg Roedel <jroedel@suse.de>
Cc: Borislav Petkov <bp@suse.de>
Cc: Yanmin Zhang <yanmin_zhang@linux.intel.com>
Link: http://lkml.kernel.org/r/20150705171102.141898931@linutronix.de
Nothing changes those ops. Make the initializers readable while at it.
Reported-by: Krzysztof Kozlowski <k.kozlowski.k@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
