In multi-cluster systems, the PMUs can be different across clusters, and
so our logical PMU may not be able to schedule events on all CPUs.

This patch adds a cpumask to encode which CPUs a PMU driver supports
controlling events for, and limits the driver to scheduling events on
those CPUs, and enabling and disabling the physical PMUs on those CPUs.
The cpumask is built based on the interrupt-affinity property, and in
the absence of such a property a homogenous system is assumed.
Acked-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

To support multiple PMUs we'll need to pass the arm_pmu instance around.
Update of_pmu_irq_cfg to take an arm_pmu, and acquire the platform
device from this.
Acked-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Historically, the PMU devicetree bindings have expected SPIs to be
listed in order of *logical* CPU number. This is problematic for
bootloaders, especially when the boot CPU (logical ID 0) isn't listed
first in the devicetree.

This patch adds a new optional property, interrupt-affinity, to the
PMU node which allows the interrupt affinity to be described using
a list of phandled to CPU nodes, with each entry in the list
corresponding to the SPI at the same index in the interrupts property.

Cc: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Scorpion supports a set of local performance monitor event
selection registers (LPM) sitting behind a cp15 based interface
that extend the architected PMU events to include Scorpion CPU
and Venum VFP specific events. To use these events the user is
expected to program the lpm register with the event code shifted
into the group they care about and then point the PMNx event at
that region+group combo by writing a LPMn_GROUPx event. Add
support for this hardware.

Note: the raw event number is a pure software construct that
allows us to map the multi-dimensional number space of regions,
groups, and event codes into a flat event number space suitable
for use by the perf framework.

This is based on code originally written by Sheetal Sahasrabudhe,
Ashwin Chaugule, and Neil Leeder [1].

[1] https://www.codeaurora.org/cgit/quic/la/kernel/msm/tree/arch/arm/kernel/perf_event_msm.c?h=msm-3.4

Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Neil Leeder <nleeder@codeaurora.org>
Cc: Ashwin Chaugule <ashwinc@codeaurora.org>
Cc: Sheetal Sahasrabudhe <sheetals@codeaurora.org>
Cc: <devicetree@vger.kernel.org>
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

of_device_ids (i.e. compatible strings and the respective data) are not
supposed to change at runtime. All functions working with of_device_ids
provided by <linux/of.h> work with const of_device_ids. So mark the
non-const structs in arch/arm as const, too.

While at it also add some __initconst annotations.
Acked-by: NJason Cooper <jason@lakedameon.net>
Signed-off-by: NUwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: NArnd Bergmann <arnd@arndb.de>

Handling multiple PMUs using a single hotplug notifier requires a list
of PMUs to be maintained, with synchronisation in the probe, remove, and
notify paths. This is error-prone and makes the code much harder to
maintain.

Instead of using a single notifier, we can dynamically allocate a
notifier block per-PMU. The end result is the same, but the list of PMUs
is implicit in the hotplug notifier list rather than within a perf-local
data structure, which makes the code far easier to handle.

Signed-off-by: Mark Rutland <mark.rutland at arm.com>
Reviewed-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

To support multiple PMUs, each PMU will need its own accounting data.
As we don't know how (in general) many PMUs we'll have to support at
compile-time, we must allocate the data at runtime dynamically
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Reviewed-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Currently the percpu_pmu pointers used as percpu_irq dev_id values are
defined separately from the other per-cpu accounting data, which make
dynamically allocating the data (as will be required for systems with
heterogeneous CPUs) difficult.

This patch moves the percpu_pmu pointers into pmu_hw_events (which is
itself allocated per cpu), which will allow for easier dynamic
allocation. Both percpu and regular irqs are requested using percpu_pmu
pointers as tokens, freeing us from having to know whether an irq is
percpu within the handler, and thus avoiding a radix tree lookup on the
handler path.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Reviewed-by: NWill Deacon <will.deacon@arm.com>
Reviewed-by: NStephen Boyd <sboyd@codeaurora.org>
Tested-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Now that the arm pmu code is limited to CPU PMUs the get_hw_events()
function is superfluous, as we'll always have a set of per-cpu
pmu_hw_events structures.

This patch removes the get_hw_events() function, replacing it with
a percpu hw_events pointer. Uses of get_hw_events are updated to use
this_cpu_ptr.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Reviewed-by: NWill Deacon <will.deacon@arm.com>
Reviewed-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Commit 3fc2c830 (ARM: perf: remove event limit from pmu_hw_events) got
rid of the upper limit on the number of events an arm_pmu could handle,
but introduced additional complexity and places a burden on each PMU
driver to allocate accounting data somehow. So far this has not
generally been useful as the only users of arm_pmu are the CPU backend
and the CCI driver.

Now that the CCI driver plugs into the perf subsystem directly, we can
remove some of the complexities that get in the way of supporting
heterogeneous CPU PMUs.

This patch restores the original limits on pmu_hw_events fields such
that the pmu_hw_events data can be allocated as a contiguous block. This
will simplify dynamic pmu_hw_events allocation in later patches.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Reviewed-by: NWill Deacon <will.deacon@arm.com>
Reviewed-by: NStephen Boyd <sboyd@codeaurora.org>
Tested-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

For systems with heterogeneous CPUs (e.g. big.LITTLE systems) the PMUs
can be different in each cluster, and not all events can be migrated
between clusters. To allow userspace to deal with this, it must be
possible to address each PMU independently.

This patch changes PMUs to be registered with dynamic (IDR) types,
allowing them to be targeted individually. Each PMU's type can be found
in ${SYSFS_ROOT}/bus/event_source/devices/${PMU_NAME}/type.

From userspace, raw events can be targeted at a specific PMU:
$ perf stat -e ${PMU_NAME}/config=V,config1=V1,.../

Doing this does not break existing tools which use existing perf types:
when perf core can't find a PMU of matching type (in perf_init_event)
it'll iterate over the set of all PMUs. If a compatible PMU exists,
it'll be found eventually. If more than one compatible PMU exists, the
event will be handled by whichever PMU happens to be earlier in the pmus
list (which currently will be the last compatible PMU registered).
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Reviewed-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The current PMU probing logic consists of a single switch statement,
which means that the core arm_pmu core in perf_event_cpu.c needs to know
about every CPU PMU variant supported by a driver using the arm_pmu
framework. This makes it rather difficult to decouple the drivers from
the (otherwise generic) probing code.

The patch refactors that switch statement to a table-driven lookup,
separating the logic and knowledge (in the form of the table). Later
patches will split the table across the relevant PMU drivers, which can
pass their tables to the generic probing function.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Reviewed-by: NWill Deacon <will.deacon@arm.com>
Reviewed-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Most of the pr_info format strings in perf_event_cpu.c are missing
newlines. Currently we get away with this as the format strings for
subsequent calls to printk (including all pr_* calls) begin with a log
prefix, and the printk core adds the omitted newline for this case.
While generates the output we expect, we probably should not rely on the
format of successive printk calls in order to get legible output.

This patch adds the missing newlines to pr_info format strings in
perf_event_cpu.c, making them consistent with the format strings for
other pr_info, warn, and pr_err calls, and preventing potentially
illegible output if the next printk/pr_* format string doesn't begin
with a log prefix.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Reviewed-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

There are a few remaining uses of printk in the ARM perf code, so move
them over to the pr_* variants instead.
Reported-by: NRussell King <linux@arm.linux.org.uk>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Use the more common pr_warn.

Other miscellanea:

o Coalesce formats
o Realign arguments
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Rob Clark reports a sleeping while atomic bug when using perf.

BUG: sleeping function called from invalid context at ../kernel/locking/mutex.c:583
in_atomic(): 1, irqs_disabled(): 128, pid: 0, name: swapper/0
------------[ cut here ]------------
WARNING: CPU: 2 PID: 4828 at ../kernel/locking/mutex.c:479 mutex_lock_nested+0x3a0/0x3e8()
DEBUG_LOCKS_WARN_ON(in_interrupt())
Modules linked in:
CPU: 2 PID: 4828 Comm: Xorg.bin Tainted: G        W      3.17.0-rc3-00234-gd535c45-dirty #819
[<c0216690>] (unwind_backtrace) from [<c0212174>] (show_stack+0x10/0x14)
[<c0212174>] (show_stack) from [<c0867cc0>] (dump_stack+0x98/0xb8)
[<c0867cc0>] (dump_stack) from [<c02492a4>] (warn_slowpath_common+0x70/0x8c)
[<c02492a4>] (warn_slowpath_common) from [<c02492f0>] (warn_slowpath_fmt+0x30/0x40)
[<c02492f0>] (warn_slowpath_fmt) from [<c086a3f8>] (mutex_lock_nested+0x3a0/0x3e8)
[<c086a3f8>] (mutex_lock_nested) from [<c0294d08>] (irq_find_host+0x20/0x9c)
[<c0294d08>] (irq_find_host) from [<c0769d50>] (of_irq_get+0x28/0x48)
[<c0769d50>] (of_irq_get) from [<c057d104>] (platform_get_irq+0x1c/0x8c)
[<c057d104>] (platform_get_irq) from [<c021a06c>] (cpu_pmu_enable_percpu_irq+0x14/0x38)
[<c021a06c>] (cpu_pmu_enable_percpu_irq) from [<c02b1634>] (flush_smp_call_function_queue+0x88/0x178)
[<c02b1634>] (flush_smp_call_function_queue) from [<c0214dc0>] (handle_IPI+0x88/0x160)
[<c0214dc0>] (handle_IPI) from [<c0208930>] (gic_handle_irq+0x64/0x68)
[<c0208930>] (gic_handle_irq) from [<c0212d04>] (__irq_svc+0x44/0x5c)
Exception stack(0xe63ddea0 to 0xe63ddee8)
dea0: 00000001 00000001 00000000 c2f3b200 c16db380 c032d4a0 e63ddf40 60010013
dec0: 00000000 001fbfd4 00000100 00000000 00000001 e63ddee8 c0284770 c02a2e30
dee0: 20010013 ffffffff
[<c0212d04>] (__irq_svc) from [<c02a2e30>] (ktime_get_ts64+0x1c8/0x200)
[<c02a2e30>] (ktime_get_ts64) from [<c032d4a0>] (poll_select_set_timeout+0x60/0xa8)
[<c032d4a0>] (poll_select_set_timeout) from [<c032df64>] (SyS_select+0xa8/0x118)
[<c032df64>] (SyS_select) from [<c020e8e0>] (ret_fast_syscall+0x0/0x48)
---[ end trace 0bb583b46342da6f ]---
INFO: lockdep is turned off.

We don't really need to get the platform irq again when we're
enabling or disabling the per-cpu irq. Furthermore, we don't
really need to set and clear bits in the active_irqs bitmask
because that's only used in the non-percpu irq case to figure out
when the last CPU PMU has been disabled. Just pass the irq
directly to the enable/disable functions to clean all this up.
This should be slightly more efficient and also fix the
scheduling while atomic bug.

Fixes: bbd64559 "ARM: perf: support percpu irqs for the CPU PMU"
Reported-by: NRob Clark <robdclark@gmail.com>
Acked-by: NWill Deacon <will.deacon@arm.com>
Cc: stable@vger.kernel.org
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Ensure that platform maintainers check the CPU part number in the right
manner: the CPU part number is meaningless without also checking the
CPU implement(e|o)r (choose your preferred spelling!)  Provide an
interface which returns both the implementer and part number together,
and update the definitions to include the implementer.

Mark the old function as being deprecated... indeed, using the old
function with the definitions will now always evaluate as false, so
people must update their un-merged code to the new function.  While
this could be avoided by adding new definitions, we'd also have to
create new names for them which would be awkward.
Acked-by: NNicolas Pitre <nico@linaro.org>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

When described in DT, PMUs are given specific compatible strings
(e.g. "arm,cortex-a15-pmu") which makes it very easy to reorganise the
way individual PMUs are handled (i.e. we can easily split them into
separate drivers). The same is not true of PMUs described in board
files, which are all use the platform_device_id "arm-pmu" and must all
be handled by the same driver.

To enable splitting the ARMv6, ARMv7, and XScale PMU drivers we need
board files to identify which variant they provide. As a first step,
this patch adds new platform_device_id values: "armv6-pmu", "armv7-pmu,
and "xscale-pmu".

Once board files are moved over and all existing uses of "arm-pmu" are
gone, we can split the existing driver apart.
Acked-by: NWill Deacon <will.deacon@arm.com>
Tested-by: NChristopher Covington <cov@codeaurora.org>
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The perf userspace tools can't handle dashes or spaces in PMU names,
which conflicts with the current naming scheme in the arm perf backend.
This prevents these PMUs from being accessed by name from the perf
tools. Additionally the ARMv6 pmus are named "v6", which does not fully
distinguish them in the sys/bus/event_source namespace.

This patch renames the PMUs consistently to a lower case form with
underscores, e.g. "armv6_1176", "armv7_cortex_a9". This is both readily
accepted by today's perf tool, and far easier to type than the
(apparently unused) convention in use previously. The OProfile name
conversion code is updated to handle this.

Due to a copy-paste error involving two "xscale1" entries, "xscale2" has
never been matched by the name OProfile name mapping. While we're
updating names, this is corrected.
Acked-by: NWill Deacon <will.deacon@arm.com>
Tested-by: NChristopher Covington <cov@codeaurora.org>
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
[sachin: fixed missing semicolons in armv6 backend]
Signed-off-by: NSachin Kamat <sachin.kamat@samsung.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Make the ARM perf code use the new common PMU interrupt disabled code.

This allows perf to work on ARM machines without a working PMU
interrupt (for example, raspberry pi).
Acked-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NVince Weaver <vincent.weaver@maine.edu>
[peterz: applied changes suggested by Will]
Signed-off-by: NPeter Zijlstra <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Grant Likely <grant.likely@linaro.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Rob Herring <robh+dt@kernel.org>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Will Deacon <will.deacon@arm.com>
Cc: devicetree@vger.kernel.org
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-kernel@vger.kernel.org
Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1405161712190.11099@vincent-weaver-1.umelst.maine.edu
[ Small readability tweaks to the code. ]
Signed-off-by: NIngo Molnar <mingo@kernel.org>
Signed-off-by: NIngo Molnar <mingo@kernel.org>

The Cortex-A17 PMU is identical to that of the A12, so wire up a new
compatible string to the existing event structures.
Signed-off-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Cortex-A12 implements Performance Monitors compliant with the PMUv2
architecture.

This patch adds support for the Cortex-A12 PMU to the ARM perf backend.
Signed-off-by: NAlbin Tonnerre <albin.tonnerre@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The ARM perf backend can discover the type of PMU it needs to drive
either from DT or by probing a CPU it is running on. For
Cortex-A{5,7,15} there are no platforms in mainline not using dt, and
this probing won't work well for big.LITTLE systems with heterogeneous
PMUs.

This patch drops the probing for those CPUs, relying on information from
dt instead. Future platforms should describe their PMU(s) with dt.
Suggested-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Add basic support for the Krait CPU PMU. This allows us to use
the architected functionality of the PMU.

This is based on code originally written by Ashwin Chaugule and
Neil Leeder [1].

[1] https://www.codeaurora.org/cgit/quic/la/kernel/msm/tree/arch/arm/kernel/perf_event_msm_krait.c?h=msm-3.4

Cc: Neil Leeder <nleeder@codeaurora.org>
Cc: Ashwin Chaugule <ashwinc@codeaurora.org>
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

We want to inspect the of_node that the pdev is pointing to in
the Krait CPU specific PMU initialization function. Assign it
earlier so that we don't crash with a NULL pointer dereference.
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Some CPU PMUs are wired up with one PPI for all the CPUs instead
of with a different SPI for each CPU. Add support for these
devices.
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

arch/arm/kernel/perf_event_cpu.c:274:25: warning: incorrect type in assignment (different modifiers)
arch/arm/kernel/perf_event_cpu.c:274:25: expected int ( *init_fn )( ... )
arch/arm/kernel/perf_event_cpu.c:274:25: got void const *const data
Acked-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

This is the ARM part of Christoph's patchset cleaning up the various
uses of __get_cpu_var across the tree.

The idea is to convert __get_cpu_var into either an explicit address
calculation using this_cpu_ptr() or into a use of this_cpu operations
that use the offset. Thereby address calculations are avoided and fewer
registers are used when code is generated.

[will: fixed debug ref counting checks and pcpu array accesses]
Acked-by: NCatalin Marinas <catalin.marinas@arm.com>
Signed-off-by: NChristoph Lameter <cl@linux.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

PMU interrupts must not be threaded.
Acked-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NSebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

The __cpuinit type of throwaway sections might have made sense
some time ago when RAM was more constrained, but now the savings
do not offset the cost and complications.  For example, the fix in
commit 5e427ec2 ("x86: Fix bit corruption at CPU resume time")
is a good example of the nasty type of bugs that can be created
with improper use of the various __init prefixes.

After a discussion on LKML[1] it was decided that cpuinit should go
the way of devinit and be phased out.  Once all the users are gone,
we can then finally remove the macros themselves from linux/init.h.

Note that some harmless section mismatch warnings may result, since
notify_cpu_starting() and cpu_up() are arch independent (kernel/cpu.c)
and are flagged as __cpuinit  -- so if we remove the __cpuinit from
the arch specific callers, we will also get section mismatch warnings.
As an intermediate step, we intend to turn the linux/init.h cpuinit
related content into no-ops as early as possible, since that will get
rid of these warnings.  In any case, they are temporary and harmless.

This removes all the ARM uses of the __cpuinit macros from C code,
and all __CPUINIT from assembly code.  It also had two ".previous"
section statements that were paired off against __CPUINIT
(aka .section ".cpuinit.text") that also get removed here.

[1] https://lkml.org/lkml/2013/5/20/589

Cc: Russell King <linux@arm.linux.org.uk>
Cc: Will Deacon <will.deacon@arm.com>
Cc: linux-arm-kernel@lists.infradead.org
Signed-off-by: NPaul Gortmaker <paul.gortmaker@windriver.com>

Currently perf_pmu_register may fail for several reasons (e.g. being
unable to allocate memory for the struct device it associates with each
PMU), and while any error is propagated by armpmu_register, it is
ignored by cpu_pmu_device_probe and not propagated to the caller.  This
also results in a leak of a struct arm_pmu.

This patch adds cleanup if armpmu_register fails, and updates the info
messages to better differentiate this type of failure from a failure to
probe the PMU type from the hardware or dt.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

cpu_pmu has already been dereferenced before we consider invoking the
->reset function, so remove the redundant NULL check.
Reported-by: NCong Ding <dinggnu@gmail.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Instead of decoding implementor numbers, part numbers and Xscale
architecture masks inline in the pmu probing function, use defines
and accessor functions from cputype.h, which can also be shared by
other subsystems, such as KVM.
Signed-off-by: NChristoffer Dall <c.dall@virtualopensystems.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

CONFIG_HOTPLUG is going away as an option.  As a result, the __dev*
markings need to be removed.

This change removes the use of __devinit, __devexit_p, __devinitdata,
and __devexit from these drivers.

Based on patches originally written by Bill Pemberton, but redone by me
in order to handle some of the coding style issues better, by hand.

Cc: Bill Pemberton <wfp5p@virginia.edu>
Cc: Russell King <linux@arm.linux.org.uk>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Perf has three ways to name a PMU: either by passing an explicit char *,
reading arm_pmu->name or accessing arm_pmu->pmu.name.

Just use arm_pmu->name consistently in the ARM backend.
Signed-off-by: NWill Deacon <will.deacon@arm.com>

When attempting to reset the PMU state for either a NULL PMU or a PMU
implementation without a reset function, return NOTIFY_DONE from the CPU
notifier as we don't care about the hotplug event.
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The current practice of registering the cpu hotplug notifier at PMU
registration time won't be safe with multiple PMUs, as we'll repeatedly
attempt to register the notifier. This has the unfortunate effect of
silently corrupting the notifier list, leading to boot stalling.

Instead, register the notifier at init time. Its sanity checks will
prevent anything bad from happening if the notifier is called before we
have any PMUs registered.
Signed-off-by: NMark Rutland <mark.rutland@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

The arm_pmu functions have wildly varied parameters which can often be
derived from struct perf_event.

This patch changes the arm_pmu function prototypes so that struct
perf_event pointers are passed in preference to fields that can be
derived from the event.
Signed-off-by: NSudeep KarkadaNagesha <Sudeep.KarkadaNagesha@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

Supporting multiple, heterogeneous CPU PMUs requires us to allocate the
arm_pmu structures dynamically as the devices are probed.

This patch removes the static structure definitions for each CPU PMU
type and instead passes pointers to the PMU-specific init functions.
Signed-off-by: NSudeep KarkadaNagesha <Sudeep.KarkadaNagesha@arm.com>
Signed-off-by: NWill Deacon <will.deacon@arm.com>

This patch moves the CPU-specific IRQ registration and parsing code into
the CPU PMU backend. This is required because a PMU may have more than
one interrupt, which in turn can be either PPI (per-cpu) or SPI
(requiring strict affinity setting at the interrupt distributor).
Signed-off-by: NSudeep KarkadaNagesha <Sudeep.KarkadaNagesha@arm.com>
[will: cosmetic edits and reworked interrupt dispatching]
Signed-off-by: NWill Deacon <will.deacon@arm.com>