Currently, CONFIG_RCU_TORTURE_TEST=y is hardcoded into the
kvm-test-1-rcu.sh script and CONFIG_PRINTK_TIME=y is mentioned in each
and every configs file.  This commit creates a CFcommon file for these
two Kconfig parameters, and modifies kvm-test-1-rcu.sh to copy this new
file into the .config file during the build.  This change will allow
these scripts to operate on torture types other than just rcutorture.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

This commit adds a pair of files in the configs directory to allow
test-the-test runs of rcutorture via a "--configs BUSTED" argument to
the kvm.sh script.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

This commit creates a plug-in to allow kvm-recheck.sh to process
non-rcutorture console output.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

This commit prevents the results directory from being created for
dryruns.  However, a script generated from a dryrun will create
the results directory should it be run.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org

Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

This commit adds a "(!)" flag after the number of CPUs required by a
given test if that test requires more than the available number of CPUs.
Note that these flags appear only when the number of CPUs is specified
using the --cpus argument.  In the absence of a --cpus argument, no
tests are flagged.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

Running the standard set of rcutorture tests on 24 CPUs results in
the following sub-optimal schedule:

	----start batch----
	 TREE07 16
	----start batch----
	 TREE08 16
	 SRCU-P 8
	----start batch----
	 TREE01 8
	 TREE02 8
	 TREE03 8
	----start batch----
	 TREE04 8
	 TREE05 8
	 TREE06 8
	----start batch----
	 SRCU-N 4
	 TINY01 1
	 TINY02 1
	 TREE09 1

If one of the eight-CPU runs were to be moved into the first batch,
the test suite would complete in four batches rather than five.

This commit therefore uses a greedy algorithm to re-order the test
entries so that the sequential batching will produce an optimal schedule
in this case:

	----start batch----
	 TREE07 16
	 SRCU-P 8
	----start batch----
	 TREE08 16
	 TREE01 8
	----start batch----
	 TREE02 8
	 TREE03 8
	 TREE04 8
	----start batch----
	 TREE05 8
	 TREE06 8
	 SRCU-N 4
	 TINY01 1
	 TINY02 1
	 TREE09 1

Please note that this is still not an optimal bin-packing algorithm,
however, it does produce optimal solutions for most common scenarios.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

This commit fixes handling numbering of multiple runs of the same test
so as to disambiguate output.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

Actual rcutorture tests take considerable time and machine resources,
so it is inconvenient to actually do an rcutorture run when optimizing
the bin-packing algorithm.  This commit therefore adds a --dryrun
argument, which defaults to doing a run, but for which "sched"
says to simply print the run schedule and "script" dumps the script
without running it.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

The message complains about a build directory when it should instead
be complaining about the results directory, so this commit fixes it.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

The rcutorture tests run by default range from using one CPU to using
sixteen of them.  Therefore, rcutorture testing could be sped up
significantly simply by running the kernels in parallel.  Building
them in parallel is not all that helpful: "make -j" is usually a
better bet.  So this commit takes a new "--cpus" argument that
specifies how many CPUs rcutorture is permitted to use for its
parallel runs.  The default of zero does sequential runs as before.

The bin-packing is minimal, and will be grossly suboptimal for
some configurations.  However, powers of two work reasonably well.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

Both SRCU-P and SRCU-N specify eight CPUs, which results in four
iterations for a parallel run on 32 CPUs.  This commit reduces SRCU-N
to four CPUs (but leaving SRCU-P at eight) to speed up parallel runs,
while maintaining essentially the same test coverage.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

Currently, most qemu flags are calculated in kvm-test-1-rcu.sh,
except that -nographics is set up by kvm.sh.  This commit promotes
one-stop shopping by consolidating the determination of qemu flags into
kvm-test-1-rcu.sh.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

Parallel rcutorture runs is valuable on large systems, but it is not a
good idea to do (say) five builds in parallel if each build believes it
has the whole system at its disposal, especially if the system is shared.
It is also bad to restrict the build to (say) a single CPU just because
the corresponding rcutorture run uses only a single CPU.  This commit
therefore adds a kvm-test-1-rcu.sh ability to pause after the build
completes, which will allow kvm.sh to do a number of builds serially
(with each build thus having the full system at its disposal), then
allow the rcutorture runs to proceed in parallel.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

Currently, most boot flags are calculated in kvm-test-1-rcu.sh, except
that rcutorture.test_no_idle_hz and rcutorture.verbose are set up by
kvm.sh.  This commit promotes one-stop shopping by consolidating the
determination of boot flags into kvm-test-1-rcu.sh.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

Although the script name and arguments are logged in the results directory,
it is more convenient to see it in the output.  This commit therefore
adds the output of this information.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

Sometime problems can manifest themselves as unusually slow grace periods.
This commit therefore prints the number of rcutorture updates during the
test and the number per second.  These statistics are harvested from the
config.out and qemu-cmd files, and are silently omitted if these files
are not available, as would be the case if there was a build failure or
a boot-time hang.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

The new option allows just run turbostat and get dump of counter values. It's
useful when we have something more than one program to test.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

The -s is not used, let's remove it, and update quick help accordingly.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

perf buildid-cache does not make another copy of kcore if the buildid
and modules match an existing copy.

That does not take into account the possibility that the kernel has been
relocated.

Extend the check to check if the reference relocation symbol matches
too, otherwise do make a copy.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Tested-by: NJiri Olsa <jolsa@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1391004884-10334-10-git-send-email-adrian.hunter@intel.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

If the kernel is relocated at boot time, kallsyms will not match data
recorded previously.

That does not matter for modules because they are corrected anyway.  It
also does not matter if vmlinux is being used for symbols. But if perf
tools has only kallsyms then the symbols will not match.

Fix by applying the delta gained by comparing the old and current
addresses of the relocation reference symbol.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Tested-by: NJiri Olsa <jolsa@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1391004884-10334-9-git-send-email-adrian.hunter@intel.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Now that ref_reloc_sym is set up by machine__create_kernel_maps(), the
"vmlinux symtab matches kallsyms" test does have to do it.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Tested-by: NJiri Olsa <jolsa@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1391004884-10334-8-git-send-email-adrian.hunter@intel.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Use of kcore is predicated upon it matching the recorded data.  If the
kernel has been relocated at boot time (i.e. since the data was
recorded) then do not use kcore.

Note that it is possible to make a copy of kcore at the time the data is
recorded using 'perf buildid-cache'.  Then the perf tools will use the
copy because it does match the data.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Tested-by: NJiri Olsa <jolsa@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1391004884-10334-7-git-send-email-adrian.hunter@intel.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Now that ref_reloc_sym is set up when the kernel map is created,
'perf record' does not need to pass the symbol names to
perf_event__synthesize_kernel_mmap() which can read the values needed
from ref_reloc_sym directly.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Tested-by: NJiri Olsa <jolsa@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1391004884-10334-6-git-send-email-adrian.hunter@intel.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

The ref_reloc_sym is always needed for the kernel map in order to check
for relocation.  Consequently set it up when the kernel map is created.
Otherwise it was only being set up by 'perf record'.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Tested-by: NJiri Olsa <jolsa@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1391004884-10334-5-git-send-email-adrian.hunter@intel.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Separate out the logic used to make the kallsyms full path name for a
machine.  It will be reused in a subsequent patch.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Tested-by: NJiri Olsa <jolsa@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1391004884-10334-4-git-send-email-adrian.hunter@intel.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Separate out the logic used to find the start address of the reference
symbol used to track kernel relocation.  kallsyms__get_function_start()
is used in subsequent patches.
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Tested-by: NJiri Olsa <jolsa@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1391004884-10334-3-git-send-email-adrian.hunter@intel.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Kernel maps map memory addresses to file offsets.

For symbol annotation, objdump needs the object VMA addresses.  For an
unrelocated kernel, that is the same as the memory address.

The addresses passed to objdump for symbol annotation did not take into
account kernel relocation.

This patch fixes that.
Reported-by: NLinus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: NAdrian Hunter <adrian.hunter@intel.com>
Tested-by: NJiri Olsa <jolsa@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1391004884-10334-2-git-send-email-adrian.hunter@intel.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Commit 71ae8aac ("lib: introduce arch optimized hash library") added an
include to <linux/hash.h> for setting up an architecture specific fast
hash.

Since perf includes directly the non-uapi kernel header, it cannot find
<asm/hash.h> on non-x86 and thus prevents perf to be compiled on every
architecture other than x86.

The problem is the inclusion of <asm/hash.h> in hash.h that results in
the following error originating from util/evlist.c:

  fatal error: asm/hash.h: No such file or directory

This commit simply adds an empty <asm/hash.h> stub/file to fix the
compile issue on non-x86 architectures.

As perf does not use any of these new functions, it fixes the
compilation and therefore seems to be the most appropriate solution to
go with.
Signed-off-by: NFrancesco Fusco <ffusco@redhat.com>
Link: http://lkml.kernel.org/r/2cf8143aad65a6aa6fe30325ef8a65847141afa2.1390829373.git.ffusco@redhat.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Someone got the load and store barriers mixed up for AAAAARGH64.  Turn
them the right side up.
Reported-by: NWill Deacon <will.deacon@arm.com>
Signed-off-by: NPeter Zijlstra <peterz@infradead.org>
Fixes: a94d342b ("tools/perf: Add required memory barriers")
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Will Deacon <will.deacon@arm.com>
Link: http://lkml.kernel.org/r/20140124154002.GF31570@twins.programming.kicks-ass.netSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Some kernels contain C++ code, and thus their symbols need to be
demangled.  This allows 'perf kvm top' to generate readable output.
Signed-off-by: NAvi Kivity <avi@cloudius-systems.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/26f71bf5bf7ee1408e3f1a803556d5df18223ef1.1390420726.git.avi@cloudius-systems.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

set_perf_event_pending() was removed in e360adbe ("irq_work: Add
generic hardirq context callbacks").
Signed-off-by: NBaruch Siach <baruch@tkos.co.il>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Link: http://lkml.kernel.org/r/4c54761865d40210be0628cb84701afc5d57b5d8.1390686193.git.baruch@tkos.co.ilSigned-off-by: NIngo Molnar <mingo@kernel.org>

To help avoid an architecture failing to correctly check kernel/user
boundaries when handling copy_to_user, copy_from_user, put_user, or
get_user, perform some simple tests and fail to load if any of them
behave unexpectedly.

Specifically, this is to make sure there is a way to notice if things
like what was fixed in commit 8404663f ("ARM: 7527/1: uaccess:
explicitly check __user pointer when !CPU_USE_DOMAINS") ever regresses
again, for any architecture.

Additionally, adds new "user" selftest target, which loads this module.
Signed-off-by: NKees Cook <keescook@chromium.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Joe Perches <joe@perches.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

In map_groups__find_symbol() map->map_ip is used without ensuring the
map is loaded. Then the address passed to map->map_ip isn't mapped at
the first time.

E.g. below code always fails to get a symbol at the first call;

	addr = /* Somewhere in the kernel text */
	symbol_conf.try_vmlinux_path = true;
	symbol__init();
	host_machine = machine__new_host();
	sym = machine__find_kernel_function(host_machine,
					 addr, NULL, NULL);
	/* Note that machine__find_kernel_function calls
	   map_groups__find_symbol */

This ensures it by calling map__load before using it in
map_groups__find_symbol().
Signed-off-by: NMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: "David A. Long" <dave.long@linaro.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: "Steven Rostedt (Red Hat)" <rostedt@goodmis.org>
Cc: yrl.pp-manager.tt@hitachi.com
Link: http://lkml.kernel.org/r/20140123022950.7206.17357.stgit@kbuild-fedora.yrl.intra.hitachi.co.jpSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

The plugindir_SQ definition contains $(prefix) which is not needed as
the $(libdir) definition already contains prefix in it.  This leads to
the path including an extra prefix in it, e.g. /usr/usr/lib64.

The -DPLUGIN_DIR defintion includes DESTDIR.  This is incorrect, as it
sets the plugin search path to include the value of DESTDIR.  DESTDIR is
a mechanism to install in a non-standard location such as a chroot or an
RPM build root.  In the RPM case, this leads to the search path being
incorrect after the resulting RPM is installed (or in some cases an RPM
build failure).

Remove both of these unnecessary inclusions.
Signed-off-by: NJosh Boyer <jwboyer@fedoraproject.org>
Acked-by: NJiri Olsa <jolsa@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/20140122150147.GK16455@hansolo.jdub.homelinux.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

Gaurav reported that perf cannot profile JIT program if it executes the
code on heap.  This was because current map__new() only handle JIT on
anon mappings - extends it to handle no_dso (heap, stack) case too.

This patch assumes JIT profiling only provides dynamic function symbols
so check the mapping type to distinguish the case.  It'd provide no
symbols for data mapping - if we need to support symbols on data
mappings later it should be changed.
Reported-by: NGaurav Jain <gjain@fb.com>
Signed-off-by: NNamhyung Kim <namhyung@kernel.org>
Tested-by: NGaurav Jain <gjain@fb.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Gaurav Jain <gjain@fb.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1389836971-3549-1-git-send-email-namhyung@kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

This patch fixes a memory corruption problem with the xyarray when the
evsel fds get closed at the end of the run_perf_stat() call.

It could be triggered with:

 # perf stat -a -e power/energy-cores/ ls

When cpumask are used by events (.e.g, RAPL or uncores) then the evsel
fds are allocated based on the actual number of CPUs monitored. That
number can be smaller than the total number of CPUs on the system.

The problem arises at the end by perf stat closes the fds twice. When
fds are closed, their entry in the xyarray are set to -1.

The first close() on the evsel is made from __run_perf_stat() and it
uses the actual number of CPUS for the event which is how the xyarray
was allocated for.

The second is from perf_evlist_close() but that one is on the total
number of CPUs in the system, so it assume the xyarray was allocated to
cover it. However it was not, and some writes corrupt memory.

The fix is in perf_evlist_close() is to first try with the evsel->cpus
if present, if not use the evlist->cpus. That fixes the problem.
Signed-off-by: NStephane Eranian <eranian@google.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1389972846-6566-3-git-send-email-eranian@google.comSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

No need to set evsel->fd to NULL after calling perf_evsel__free_fd(), as
this method already does that.

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/n/tip-wu6kul8fpapr8iyqm685ewtf@git.kernel.orgSigned-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>