Move the perf-events backend from arch/arm/oprofile into
drivers/oprofile so that the code can be shared between architectures.
This allows each architecture to maintain only a single copy of the PMU
accessor functions instead of one for both perf and OProfile. It also
becomes possible for other architectures to delete much of their
OProfile code in favour of the common code now available in
drivers/oprofile/oprofile_perf.c.
Signed-off-by: Matt Fleming <matt@console-pimps.org>
Tested-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
In preparation for moving the majority of this oprofile code into an
architecture-neutral place separate the architecture-independent code
into oprofile_perf_init() and oprofile_perf_exit().
Signed-off-by: Matt Fleming <matt@console-pimps.org>
Tested-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
In preparation for moving the generic functions out of this file, give
the functions more general names (e.g. remove "arm" from the names).
Signed-off-by: Matt Fleming <matt@console-pimps.org>
Tested-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
Make op_name_from_perf_id() global so that we have a way for each
architecture to construct an oprofile name for op->cpu_type. We need to
remove the argument from the function prototype so that we can hide all
implementation details inside the function.
Signed-off-by: Matt Fleming <matt@console-pimps.org>
Signed-off-by: Robert Richter <robert.richter@amd.com>
The number of counters for the registered pmu is needed in a few places
so provide a helper function that returns this number.
Signed-off-by: Matt Fleming <matt@console-pimps.org>
Tested-by: Will Deacon <will.deacon@arm.com>
Acked-by: Paul Mundt <lethal@linux-sh.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Robert Richter <robert.richter@amd.com>
The pointers must be NULL'ed to avoid double-freeing the pointers in
rare cases during reinitialization.
Signed-off-by: Robert Richter <robert.richter@amd.com>
Now that oprofile_arch_exit is only called when the OProfile module
is unloaded, it can assume that init completed successfully and not
have to worry about double frees or releasing NULL perf events.
This patch ensures that oprofile_arch_init fails gracefully on ARM
and simplifies the exit code based on the above.
Cc: Robert Richter <robert.richter@amd.com>
Cc: Matt Fleming <matt@console-pimps.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
This is a reworking of an original patch posted by Aaro Koskinen:
oprofile does not work with PM, because sysdev_suspend() is done with
interrupts disabled and oprofile needs a mutex. Implementing oprofile
as a platform device solves this problem.
Cc: Aaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Aaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Enable hardware perf-events if CPU_HAS_PMU and select
HAVE_OPROFILE if HAVE_PERF_EVENTS. If no hardware support
is present, OProfile will fall back to timer mode.
This patch also removes the old OProfile drivers in favour
of the code implemented by perf.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
There are currently two hardware performance monitoring subsystems in
the kernel for ARM: OProfile and perf-events. This creates the
following problems:
1.) Duplicate PMU accessor code. Inevitable code drift may lead to
bugs in one framework that are fixed in the other.
2.) Locking issues. OProfile doesn't reprogram hardware counters
between profiling runs if the events to be monitored have not been
changed. This means that other profiling frameworks cannot use the
counters if OProfile is in use.
3.) Due to differences in the two frameworks, it may not be possible to
compare the results obtained by OProfile with those obtained by perf.
This patch removes the OProfile PMU driver code and replaces it with
calls to perf, therefore solving the issues mentioned above.
The only userspace-visible change is the lack of SCU counter support
for 11MPCore. This is currently unsupported by OProfile userspace tools anyway and therefore shouldn't cause any problems.
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Robert Richter <robert.richter@amd.com>
Cc: Jamie Iles <jamie.iles@picochip.com>
Cc: Jean Pihet <jpihet@mvista.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Make sure that we have access to the performance counters and
that they aren't being used by perf events or anything else.
Cc: Will Deacon <will.deacon@arm.com>
Cc: Jean Pihet <jpihet@mvista.com>
Signed-off-by: Jamie Iles <jamie.iles@picochip.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
On Oprofile ARMv7 the PMNC_D bit was set to lower the PMU IRQs
and so to decrease the risk of errata #628216 from appearing.
The effect of setting the PMNC_D bit is that the CCNT counter
is divided by 64, making the program counter events count
inaccurate.
The new OMAP3 r4 cores should have that errata fixed.
The PMNC_D bit should not be set, this patch fixes it.
Signed-off-by: Jean Pihet <jpihet@mvista.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
The IRQ_* macros need to be made visible via the mach/irqs.h file but
without the additional macros defined in the board-*.h files.
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
* 'devel' of master.kernel.org:/home/rmk/linux-2.6-arm: (422 commits)
[ARM] 5435/1: fix compile warning in sanity_check_meminfo()
[ARM] 5434/1: ARM: OMAP: Fix mailbox compile for 24xx
[ARM] pxa: fix the bad assumption that PCMCIA sockets always start with 0
[ARM] pxa: fix Colibri PXA300 and PXA320 LCD backlight pins
imxfb: Fix TFT mode
i.MX21/27: remove ifdef CONFIG_FB_IMX
imxfb: add clock support
mxc: add arch_reset() function
clkdev: add possibility to get a clock based on the device name
i.MX1: remove fb support from mach-imx
[ARM] pxa: build arch/arm/plat-pxa/mfp.c only when PXA3xx or ARCH_MMP defined
Gemini: Add support for Teltonika RUT100
Gemini: gpiolib based GPIO support v2
MAINTAINERS: add myself as Gemini architecture maintainer
ARM: Add Gemini architecture v3
[ARM] OMAP: Fix compile for omap2_init_common_hw()
MAINTAINERS: Add myself as Faraday ARM core variant maintainer
ARM: Add support for FA526 v2
[ARM] acorn,ebsa110,footbridge,integrator,sa1100: Convert asm/io.h to linux/io.h
[ARM] collie: fix two minor formatting nits
...
This patch changes the walk_stacktrace and its callers for easier
integration of stack unwinding. The arch/arm/kernel/stacktrace.h file is
also moved to arch/arm/include/asm/stacktrace.h.
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Impact: cleanup, update to new cpumask API
Irq_desc.affinity and irq_desc.pending_mask are now cpumask_var_t's
so access to them should be using the new cpumask API.
Signed-off-by: Mike Travis <travis@sgi.com>
Impact: change existing irq_chip API
Not much point with gentle transition here: the struct irq_chip's
setaffinity method signature needs to change.
Fortunately, not widely used code, but hits a few architectures.
Note: In irq_select_affinity() I save a temporary in by mangling
irq_desc[irq].affinity directly. Ingo, does this break anything?
(Folded in fix from KOSAKI Motohiro)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Reviewed-by: Grant Grundler <grundler@parisc-linux.org>
Acked-by: Ingo Molnar <mingo@redhat.com>
Cc: ralf@linux-mips.org
Cc: grundler@parisc-linux.org
Cc: jeremy@xensource.com
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Add Oprofile kernel support for ARMv7.
Tested on OMAP3430 and OMAP3530 chipsets (Cortex-A8).
Signed-off-by: Jean Pihet <jpihet@mvista.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Add asm/cputype.h, moving functions and definitions from asm/system.h
there. Convert all users of 'processor_id' to the more efficient
read_cpuid_id() function.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Remove includes of asm/hardware.h in addition to asm/arch/hardware.h.
Then, since asm/hardware.h only exists to include asm/arch/hardware.h,
update everything to directly include asm/arch/hardware.h and remove
asm/hardware.h.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
It's never used and the comments refer to nonatomic and retry
interchangably. So get rid of it.
Acked-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
This patch fixes the Oprofile for ARM11MPCore compilation introduced by
changes to the RealView code. Only RealView/EB is supported.
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
All kobjects require a dynamically allocated name now. We no longer
need to keep track if the name is statically assigned, we can just
unconditionally free() all kobject names on cleanup.
Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Quoting Randy:
"It seems sad that this patch sources Kconfig.marker, a 7-line file,
20-something times. Yes, you (we) don't want to put those 7 lines into
20-something different files, so sourcing is the right thing.
However, what you did for avr32 seems more on the right track to me: make
_one_ Instrumentation support menu that includes PROFILING, OPROFILE, KPROBES,
and MARKERS and then use (source) that in all of the arches."
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
BUG: using smp_processor_id() in preemptible [00000001] code: opcontrol/427
Resolve this bug by ensuring that we're not using smp_processor_id() in
a preemptable context (by disabling preemption.)
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Fix lockdep warnings, caused by 'set_affinity' being called without
the correct locks taken and local interrupts disabled:
=================================
[ INFO: inconsistent lock state ]
2.6.22-rc2 #1
---------------------------------
inconsistent {in-hardirq-W} -> {hardirq-on-W} usage.
swapper/1 [HC0[0]:SC0[0]:HE1:SE1] takes:
(irq_controller_lock){++..}, at: [<c002be50>] gic_set_cpu+0x60/0xa0
{in-hardirq-W} state was registered at:
[<c005d9a8>] lock_acquire+0x58/0x6c
[<c0233068>] _spin_lock+0x40/0x50
[<c002c020>] gic_mask_irq+0x2c/0x6c
[<c0069c64>] handle_level_irq+0x11c/0x14c
[<c0020060>] asm_do_IRQ+0x60/0x84
[<c0020d2c>] __irq_svc+0x4c/0xc0
[<c000ed84>] __alloc_bootmem_nopanic+0x74/0x88
[<c000edb0>] __alloc_bootmem+0x18/0x3c
[<c000fa00>] alloc_large_system_hash+0x16c/0x200
[<c00108dc>] inode_init_early+0x5c/0xa4
[<c00106dc>] vfs_caches_init_early+0x24/0xa0
[<c0008e54>] start_kernel+0x220/0x2fc
[<00008078>] 0x8078
irq event stamp: 88438
hardirqs last enabled at (88438): [<c0020dc0>] preempt_return+0x20/0x2c
hardirqs last disabled at (88436): [<c00417bc>] __do_softirq+0xb0/0x138
softirqs last enabled at (88437): [<c0041810>] __do_softirq+0x104/0x138
softirqs last disabled at (88428): [<c0041d9c>] irq_exit+0x68/0x7c
other info that might help us debug this:
no locks held by swapper/1.
stack backtrace:
[<c0025ecc>] (dump_stack+0x0/0x14) from [<c005b1e4>] (print_usage_bug+0x138/0x168)
[<c005b0ac>] (print_usage_bug+0x0/0x168) from [<c005be80>] (mark_lock+0x484/0x6a0)
[<c005b9fc>] (mark_lock+0x0/0x6a0) from [<c005cc48>] (__lock_acquire+0x3c0/0x10c8)
[<c005c888>] (__lock_acquire+0x0/0x10c8) from [<c005d9a8>] (lock_acquire+0x58/0x6c)
[<c005d950>] (lock_acquire+0x0/0x6c) from [<c0233068>] (_spin_lock+0x40/0x50)
[<c0233028>] (_spin_lock+0x0/0x50) from [<c002be50>] (gic_set_cpu+0x60/0xa0)
[<c002bdf0>] (gic_set_cpu+0x0/0xa0) from [<c01b04cc>] (em_route_irq+0x38/0x40)
[<c01b0494>] (em_route_irq+0x0/0x40) from [<c01b04ec>] (em_setup+0x18/0xa4)
[<c01b04d4>] (em_setup+0x0/0xa4) from [<c001570c>] (oprofile_arch_init+0x24/0xe8)
[<c00156e8>] (oprofile_arch_init+0x0/0xe8) from [<c0015640>] (oprofile_init+0x1c/0x64)
[<c0015624>] (oprofile_init+0x0/0x64) from [<c0008a20>] (kernel_init+0x154/0x368)
[<c00088cc>] (kernel_init+0x0/0x368) from [<c003ef34>] (do_exit+0x0/0x904)
oprofile: using arm/mpcore
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Add support for stacktrace. Use the new stacktrace code with
oprofile instead of it's version; there's no point having
multiple versions of stacktracing in the kernel.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Always allow backtrace when using oprofile on ARM, even if a PMU
isn't present.
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Add the glue for ARM11 SMP oprofile support, which also supports the
performance monitor in the coherency unit.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Add oprofile glue for ARM11 (ARMv6) oprofile support. This
connects the ARM11 core profiling support to the oprofile code
for uniprocessor configurations.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Add basic support for the ARM11 profiling hardware. This is shared
between the ARM11 UP and ARM11 SMP oprofile support code.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
OP_MAX_COUNTER never referenced, and is a reminant of an earlier
oprofile implementation. Remove it.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Some ARM platforms were still broken as a result of the IRQ register
passing changes, mostly due to a missing linux/irq.h include.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Untested, but this should fix up the bulk of the totally mechanical
issues, and should make the actual detail fixing easier.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Remove the last trace of iop31x support from the kernel.
Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Since the iop32x code isn't iop321-specific, and the iop33x code isn't
iop331-specfic, do a s/iop321/iop32x/ and s/iop331/iop33x/, and tidy up
the code to conform to the coding style guidelines somewhat better.
Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Rename CONFIG_ARCH_IOP321 to CONFIG_ARCH_IOP32X and
CONFIG_ARCH_IOP331 to CONFIG_ARCH_IOP33X.
Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
The irgflags consolidation did conflict with the ARM to generic IRQ
conversion and was not applied for ARM. Fix it up.
Use the new IRQF_ constants and remove the SA_INTERRUPT define
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
We need this to be zero initialised. Since this is an array, use kcalloc
rather than kzalloc or kmalloc.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Russ Dill
The oprofile init code was broken in commit c6b9da. The new logic will
always return -ENODEV. This fixes oprofile_arch_init to return 0 on
success, and return the return value of spec->init() if applicable.
Signed-off-by: Russ Dill <Russ.Dill@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
check_user_page_readable is a problematic variant of follow_page. It's used
only by oprofile's i386 and arm backtrace code, at interrupt time, to
establish whether a userspace stackframe is currently readable.
This is problematic, because we want to push the page_table_lock down inside
follow_page, and later split it; whereas oprofile is doing a spin_trylock on
it (in the i386 case, forgotten in the arm case), and needs that to pin
perhaps two pages spanned by the stackframe (which might be covered by
different locks when we split).
I think oprofile is going about this in the wrong way: it doesn't need to know
the area is readable (neither i386 nor arm uses read protection of user
pages), it doesn't need to pin the memory, it should simply
__copy_from_user_inatomic, and see if that succeeds or not. Sorry, but I've
not got around to devising the sparse __user annotations for this.
Then we can eliminate check_user_page_readable, and return to a single
follow_page without the __follow_page variants.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
There is nothing special about having the init code separate from
the common code, so combine the two.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
The common oprofile code assumes the name "PMU" (from Intel's
performance management unit). This is misleading when we
start adding oprofile support for other machine types which
don't use the same terminology. Call it op_arm_* instead of
pmu_*.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
The oprofile suspend/resume was missing locking. If we failed
to start oprofile on resume, we still reported that it was
enabled. Instead, disable oprofile on error.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
