linux-toradex.git/tools/perf/util, branch v4.9.16

perf callchain: Reference count maps

2017-03-12T05:41:47+00:00

commit aa33b9b9a2ebb00d33c83a5312d4fbf2d5aeba36 upstream.

If dso__load_kcore frees all of the existing maps, but one has already
been attached to a callchain cursor node, then we can get a SIGSEGV in
any function that happens to try to use this invalid cursor.  Use the
existing map refcount mechanism to forestall cleanup of a map until the
cursor iterates past the node.

Signed-off-by: Krister Johansen 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Frederic Weisbecker 
Cc: Masami Hiramatsu 
Cc: Namhyung Kim 
Fixes: 84c2cafa2889 ("perf tools: Reference count struct map")
Link: http://lkml.kernel.org/r/20170106062331.GB2707@templeofstupid.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Greg Kroah-Hartman

perf diff: Fix -o/--order option behavior (again)

2017-02-14T23:25:42+00:00

commit a1c9f97f0b64e6337d9cfcc08c134450934fdd90 upstream.

Commit 21e6d8428664 ("perf diff: Use perf_hpp__register_sort_field
interface") changed list_add() to perf_hpp__register_sort_field().

This resulted in a behavior change since the field was added to the tail
instead of the head.  So the -o option is mostly ignored due to its
order in the list.

This patch fixes it by adding perf_hpp__prepend_sort_field().

Signed-off-by: Namhyung Kim 
Acked-by: Jiri Olsa 
Cc: Peter Zijlstra 
Fixes: 21e6d8428664 ("perf diff: Use perf_hpp__register_sort_field interface")
Link: http://lkml.kernel.org/r/20170118051457.30946-2-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Greg Kroah-Hartman

perf jit: Enable jitdump support without dwarf

2017-01-26T07:24:36+00:00

commit 621cb4e7837e39d25a5af5a785ad282cdd2b4ce8 upstream.

This patch modifies the build dependencies on the jitdump support in
perf. As it stands jitdump was wrongfully made dependent 100% on using
DWARF. However, the dwarf dependency, only exist if generating the
source line table in genelf_debug.c. The rest of the support does not
need DWARF.

This patch removes the dependency on DWARF for the entire jitdump
support. It keeps it only for the genelf_debug.c support.

Signed-off-by: Maciej Debski 
Reviewed-by: Stephane Eranian 
Cc: Anton Blanchard 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/1476356383-30100-3-git-send-email-eranian@google.com
Fixes: e12b202f8fb9 ("perf jitdump: Build only on supported archs")
[ Make it build only if NO_LIBELF isn't defined, as jitdump.o will only be built in that case ]
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Greg Kroah-Hartman

perf scripting: Avoid leaking the scripting_context variable

2017-01-26T07:24:36+00:00

commit cf346d5bd4b9d61656df2f72565c9b354ef3ca0d upstream.

Both register_perl_scripting() and register_python_scripting() allocate
this variable, fix it by checking if it already was.

Cc: Adrian Hunter 
Cc: David Ahern 
Cc: Frederic Weisbecker 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Tom Zanussi 
Cc: Wang Nan 
Fixes: 7e4b21b84c43 ("perf/scripts: Add Python scripting engine")
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Greg Kroah-Hartman

perf callchain: Fixup help/config for no-unwinding

2017-01-26T07:24:36+00:00

commit c56cb33b56c13493eeb95612f80e4dd6e35cd109 upstream.

Since 841e3558b2d ("perf callchain: Recording 'dwarf' callchains do not
need DWARF unwinding support"), --call-graph dwarf is allowed in 'perf
record' even without unwind support.  A couple of other places don't
reflect this yet though: the help text should list dwarf as a valid
record mode and the dump_size config should be respected too.

Signed-off-by: Rabin Vincent 
Cc: He Kuang 
Fixes: 841e3558b2de ("perf callchain: Recording 'dwarf' callchains do not need DWARF unwinding support")
Link: http://lkml.kernel.org/r/1470837148-7642-1-git-send-email-rabin.vincent@axis.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Greg Kroah-Hartman

perf diff: Do not overwrite valid build id

2017-01-26T07:24:36+00:00

commit ed6c166cc7dc329736cace3affd2df984fb22ec8 upstream.

Fixes a perf diff regression issue which was introduced by commit
5baecbcd9c9a ("perf symbols: we can now read separate debug-info files
based on a build ID")

The binary name could be same when perf diff different binaries. Build
id is used to distinguish between them.
However, the previous patch assumes the same binary name has same build
id. So it overwrites the build id according to the binary name,
regardless of whether the build id is set or not.

Check the has_build_id in dso__load. If the build id is already set, use
it.

Before the fix:

  $ perf diff 1.perf.data 2.perf.data
  # Event 'cycles'
  #
  # Baseline    Delta  Shared Object     Symbol
  # ........  .......  ................  .............................
  #
    99.83%  -99.80%  tchain_edit       [.] f2
     0.12%  +99.81%  tchain_edit       [.] f3
     0.02%   -0.01%  [ixgbe]           [k] ixgbe_read_reg

  After the fix:
  $ perf diff 1.perf.data 2.perf.data
  # Event 'cycles'
  #
  # Baseline    Delta  Shared Object     Symbol
  # ........  .......  ................  .............................
  #
    99.83%   +0.10%  tchain_edit       [.] f3
     0.12%   -0.08%  tchain_edit       [.] f2

Signed-off-by: Kan Liang 
Cc: Andi Kleen 
CC: Dima Kogan 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Fixes: 5baecbcd9c9a ("perf symbols: we can now read separate debug-info files based on a build ID")
Link: http://lkml.kernel.org/r/1481642984-13593-1-git-send-email-kan.liang@intel.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Greg Kroah-Hartman

perf annotate: Don't throw error for zero length symbols

2017-01-09T07:32:16+00:00

commit edee44be59190bf22d5c6e521f3852b7ff16862f upstream.

'perf report --tui' exits with error when it finds a sample of zero
length symbol (i.e. addr == sym->start == sym->end). Actually these are
valid samples. Don't exit TUI and show report with such symbols.

Reported-and-Tested-by: Anton Blanchard 
Link: https://lkml.org/lkml/2016/10/8/189
Signed-off-by: Ravi Bangoria 
Cc: Alexander Shishkin 
Cc: Benjamin Herrenschmidt 
Cc: Chris Riyder 
Cc: linuxppc-dev@lists.ozlabs.org
Cc: Masami Hiramatsu 
Cc: Michael Ellerman 
Cc: Nicholas Piggin 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/1479804050-5028-1-git-send-email-ravi.bangoria@linux.vnet.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Greg Kroah-Hartman

perf hists: Fix column length on --hierarchy

2016-11-09T14:55:29+00:00

Markus reported that there's a weird behavior on perf top --hierarchy
regarding the column length.

Looking at the code, I found a dubious code which affects the symptoms.
When --hierarchy option is used, the last column length might be
inaccurate since it skips to update the length on leaf entries.

I cannot remember why it did and looks like a leftover from previous
version during the development.

Anyway, updating the column length often is not harmful.  So let's move
the code out.

Reported-and-Tested-by: Markus Trippelsdorf 
Signed-off-by: Namhyung Kim 
Cc: Jiri Olsa 
Cc: Peter Zijlstra 
Fixes: 1a3906a7e6b9 ("perf hists: Resort hist entries with hierarchy")
Link: http://lkml.kernel.org/r/20161108130833.9263-5-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf jevents: Handle events including .c and .o

2016-10-17T14:24:18+00:00

This patch helps with Sukadev's vendor event tree where such events can happen.

>From Andi Kleen:
 Any event including a .c/.o/.bpf currently triggers BPF compilation or loading
 and then an error. This can happen for some Intel vendor events, which cannot
 be used.

This patch fixes this problem by forbidding BPF file patch containing '{', '}'
and ',', make sure flex consumes the leading '{', instead of matching it using
a BPF file path.

Tested result:

  $ perf stat -e '{unc_p_clockticks,unc_p_power_state_occupancy.cores_c0}' -a -I 1000
  invalid or unsupported event: '{unc_p_clockticks,unc_p_power_state_occupancy.cores_c0}'
  Run 'perf list' for a list of valid events
  (as expected, interperted as event)

  $ perf stat -e 'aaa.c' -a -I 1000
  ERROR: problems with path aaa.c: No such file or directory
  (as expected, interpreted as BPF source)

  $ perf stat -e 'aaa.ccc' -a -I 1000
  invalid or unsupported event: 'aaa.ccc'
  (as expected, interpreted as event)

  $ perf stat -e '{aaa.c}' -a -I 1000
  ERROR: problems with path aaa.c: No such file or directory
  event syntax error: '{aaa.c}'
  
  (as expected, interpreted as BPF source)

  $ perf stat -e '{cycles,aaa.c}' -a -I 1000
  ERROR: problems with path aaa.c: No such file or directory
  event syntax error: '{cycles,aaa.c}'
  (as expected, interpreted as BPF source)

Signed-off-by: Wang Nan 
Tested-by: Arnaldo Carvalho de Melo 
Reported-by: Andi Kleen 
Cc: Jiri Olsa 
Cc: Sukadev Bhattiprolu 
Cc: Zefan Li 
Cc: pi3orama@163.com
Link: http://lkml.kernel.org/r/1475900185-37967-1-git-send-email-wangnan0@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo

perf header: Set nr_numa_nodes only when we parsed all the data

2016-10-13T14:12:29+00:00

Sukadev reported segfault on releasing perf env's numa data.  It's due
to nr_numa_nodes being set no matter if the numa data gets parsed
properly. The perf_env__exit crash the on releasing non existed data.

Setting nr_numa_nodes only when data are parsed out properly.

Signed-off-by: Jiri Olsa 
Reported-by: Sukadev Bhattiprolu 
Cc: Adrian Hunter 
Cc: David Ahern 
Cc: Namhyung Kim 
Cc: Wang Nan 
Link: http://lkml.kernel.org/n/tip-dt9c0zgkt4hybn2cr4xiawta@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo