linux-stable.git/tools/perf/util/include/linux, branch linux-3.14.y

perf tools: Fix strict alias issue for find_first_bit

2014-02-28T13:39:40+00:00

When compiling perf tool code with gcc 4.4.7 I'm getting
following error:

    CC       util/session.o
  cc1: warnings being treated as errors
  util/session.c: In function ‘perf_session_deliver_event’:
  tools/perf/util/include/linux/bitops.h:109: error: dereferencing pointer ‘p’ does break strict-aliasing rules
  tools/perf/util/include/linux/bitops.h:101: error: dereferencing pointer ‘p’ does break strict-aliasing rules
  util/session.c:697: note: initialized from here
  tools/perf/util/include/linux/bitops.h:101: note: initialized from here
  make[1]: *** [util/session.o] Error 1
  make: *** [util/session.o] Error 2

The aliased types here are u64 and unsigned long pointers, which is safe
for the find_first_bit processing.

This error shows up for me only for gcc 4.4 on 32bit x86, even for
-Wstrict-aliasing=3, while newer gcc are quiet and scream here for
-Wstrict-aliasing={2,1}. Looks like newer gcc changed the rules for
strict alias warnings.

The gcc documentation offers workaround for valid aliasing by using
__may_alias__ attribute:

  http://gcc.gnu.org/onlinedocs/gcc-4.4.0/gcc/Type-Attributes.html

Using this workaround for the find_first_bit function.

Signed-off-by: Jiri Olsa 
Cc: Corey Ashford 
Cc: David Ahern 
Cc: Frederic Weisbecker 
Cc: Ingo Molnar 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/1393434867-20271-1-git-send-email-jolsa@redhat.com
Signed-off-by: Arnaldo Carvalho de Melo

tools include: Move perf's linux/compiler.h to a generic place

2014-01-13T13:06:25+00:00

So that it can be shared with others like libtraceevent.

Signed-off-by: Namhyung Kim 
Acked-by: Borislav Petkov 
Cc: Frederic Weisbecker 
Cc: Ingo Molnar 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Steven Rostedt 
Link: http://lkml.kernel.org/r/1389276059-8829-2-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf fs: Add procfs support

2013-11-05T18:15:07+00:00

Adding procfs support into fs class.

The interface function:
  const char *procfs__mountpoint(void);

provides existing mountpoint path for procfs.

Signed-off-by: Jiri Olsa 
Cc: Adrian Hunter 
Cc: Corey Ashford 
Cc: David Ahern 
Cc: Frederic Weisbecker 
Cc: Ingo Molnar 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/1383660887-1734-3-git-send-email-jolsa@redhat.com
[ Fixup namespace ]
Signed-off-by: Arnaldo Carvalho de Melo

tools/perf: Turn strlcpy() into a __weak function

2013-10-09T06:48:49+00:00

The strlcpy() feature check slows every build unnecessarily - so make it
a __weak function so it does not have to be auto-detected.

If the libc (or any other library) has an strlcpy() implementation it will
be used - otherwise our fallback is active.

Cc: Arnaldo Carvalho de Melo 
Cc: Peter Zijlstra 
Cc: Namhyung Kim 
Cc: David Ahern 
Cc: Jiri Olsa 
Link: http://lkml.kernel.org/n/tip-zjbrcupapu08ePsyYhhhxiwk@git.kernel.org
Signed-off-by: Ingo Molnar

tools/perf: Clean up util/include/linux/compiler.h

2013-10-09T06:48:48+00:00

Use the standard CPP style we use in the kernel:

 #ifndef foo
 # define foo bar
 #endif

Cc: Arnaldo Carvalho de Melo 
Cc: Peter Zijlstra 
Cc: Namhyung Kim 
Cc: David Ahern 
Cc: Jiri Olsa 
Link: http://lkml.kernel.org/n/tip-iqyVrrHqpn0eiwenvgwrh8lf@git.kernel.org
Signed-off-by: Ingo Molnar

perf tools: Default to cpu// for events v5

2013-07-12T16:53:52+00:00

When an event fails to parse and it's not in a new style format,
try to parse it again as a cpu event.

This allows to use sysfs exported events directly without //, so you can use

perf record -e mem-loads ...

instead of

perf record -e cpu/mem-loads/

Signed-off-by: Andi Kleen 
Link: http://lkml.kernel.org/r/1366480949-32292-1-git-send-email-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Fix PMU format parsing test failure

2013-01-24T19:40:47+00:00

On POWER, the 'perf format parsing' test always fails.

Looks like it is because memset() is being passed number of longs rather
than number of bytes. It is interesting that the test always passes on
my x86 box.

With this patch, the test passes on POWER and continues to pass on x86.

Signed-off-by: Sukadev Bhattiprolu 
Acked-by: Jiri Olsa 
Cc: Anton Blanchard 
Cc: Jiri Olsa 
Cc: Paul Mackerras 
Cc: linuxppc-dev@ozlabs.org
Link: http://lkml.kernel.org/r/20130117172814.GA18882@us.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo

perf: Fix UAPI fallout

2012-10-14T19:22:52+00:00

The UAPI commits forgot to test tooling builds such as tools/perf/,
and this fixes the fallout.

Manual conversion.

Signed-off-by: Ingo Molnar 
Signed-off-by: Linus Torvalds

perf: Handle new rbtree implementation

2012-10-13T01:12:09+00:00

Perf build fails with the new rbtree implementation:

  ../../lib/rbtree.c:24:36: fatal error: linux/rbtree_augmented.h: No such file or directory compilation terminated.

Fix by updating the Makefile and adding a btree_augmented.h
wrapper.

Reported-and-tested-by: Borislav Petkov 
Signed-off-by: Markus Trippelsdorf 
Cc: Arnaldo Carvalho de Melo 
Cc: Borislav Petkov 
Link: http://lkml.kernel.org/r/20121009180156.GA245@x4
Signed-off-by: Ingo Molnar 
Signed-off-by: Linus Torvalds

perf: fix duplicate header inclusion

2012-10-10T23:50:17+00:00

#include  somehow got duplicated on its way to linus's tree
(probably as a conflict resolution as things got sent through multiple
trees)

Signed-off-by: Michel Lespinasse 
Cc: Adrian Hunter 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds