<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux.git/tools/perf, branch v4.8-rc3</title>
<subtitle>Linux kernel source tree</subtitle>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/'/>
<entry>
<title>perf intel-pt: Fix occasional decoding errors when tracing system-wide</title>
<updated>2016-08-15T21:11:18+00:00</updated>
<author>
<name>Adrian Hunter</name>
<email>adrian.hunter@intel.com</email>
</author>
<published>2016-08-15T07:23:04+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=3d918fb13abdbeca7947578f5d7e426eafad7f5e'/>
<id>3d918fb13abdbeca7947578f5d7e426eafad7f5e</id>
<content type='text'>
In order to successfully decode Intel PT traces, context switch events
are needed from the moment the trace starts. Currently that is ensured
by using the 'immediate' flag which enables the switch event when it is
opened.

However, since commit 86c2786994bd ("perf intel-pt: Add support for
PERF_RECORD_SWITCH") that might not always happen. When tracing
system-wide the context switch event is added to the tracking event
which was not set as 'immediate'. Change that so it is.

Signed-off-by: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Jiri Olsa &lt;jolsa@redhat.com&gt;
Cc: stable@vger.kernel.org # v4.4+
Fixes: 86c2786994bd ("perf intel-pt: Add support for PERF_RECORD_SWITCH")
Link: http://lkml.kernel.org/r/1471245784-22580-1-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
In order to successfully decode Intel PT traces, context switch events
are needed from the moment the trace starts. Currently that is ensured
by using the 'immediate' flag which enables the switch event when it is
opened.

However, since commit 86c2786994bd ("perf intel-pt: Add support for
PERF_RECORD_SWITCH") that might not always happen. When tracing
system-wide the context switch event is added to the tracking event
which was not set as 'immediate'. Change that so it is.

Signed-off-by: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Jiri Olsa &lt;jolsa@redhat.com&gt;
Cc: stable@vger.kernel.org # v4.4+
Fixes: 86c2786994bd ("perf intel-pt: Add support for PERF_RECORD_SWITCH")
Link: http://lkml.kernel.org/r/1471245784-22580-1-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf probe: Release resources on error when handling exit paths</title>
<updated>2016-08-15T21:10:59+00:00</updated>
<author>
<name>Arnaldo Carvalho de Melo</name>
<email>acme@redhat.com</email>
</author>
<published>2016-08-15T20:06:47+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=60ebc159817fef86171616510b1228476d979556'/>
<id>60ebc159817fef86171616510b1228476d979556</id>
<content type='text'>
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Colin King &lt;colin.king@canonical.com&gt;
Cc: David Ahern &lt;dsahern@gmail.com&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Masami Hiramatsu &lt;mhiramat@kernel.org&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Wang Nan &lt;wangnan0@huawei.com&gt;
Link: http://lkml.kernel.org/n/tip-zh2j4iqimralugke5qq7dn6d@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Colin King &lt;colin.king@canonical.com&gt;
Cc: David Ahern &lt;dsahern@gmail.com&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Masami Hiramatsu &lt;mhiramat@kernel.org&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Wang Nan &lt;wangnan0@huawei.com&gt;
Link: http://lkml.kernel.org/n/tip-zh2j4iqimralugke5qq7dn6d@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf probe: Check for dup and fdopen failures</title>
<updated>2016-08-15T20:06:19+00:00</updated>
<author>
<name>Colin Ian King</name>
<email>colin.king@canonical.com</email>
</author>
<published>2016-08-12T21:44:56+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=0325862dc364d8af524bf2db53ef4360ed55b989'/>
<id>0325862dc364d8af524bf2db53ef4360ed55b989</id>
<content type='text'>
dup and fdopen can potentially fail, so add some extra
error handling checks rather than assuming they always work.

Signed-off-by: Colin King &lt;colin.king@canonical.com&gt;
Acked-by: Masami Hiramatsu &lt;mhiramat@kernel.org&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Wang Nan &lt;wangnan0@huawei.com&gt;
Link: http://lkml.kernel.org/r/1471038296-12956-1-git-send-email-colin.king@canonical.com
[ Free resources when those functions (now being verified) fail ]
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
dup and fdopen can potentially fail, so add some extra
error handling checks rather than assuming they always work.

Signed-off-by: Colin King &lt;colin.king@canonical.com&gt;
Acked-by: Masami Hiramatsu &lt;mhiramat@kernel.org&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Wang Nan &lt;wangnan0@huawei.com&gt;
Link: http://lkml.kernel.org/r/1471038296-12956-1-git-send-email-colin.king@canonical.com
[ Free resources when those functions (now being verified) fail ]
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf symbols: Fix annotation of objects with debuginfo files</title>
<updated>2016-08-15T19:49:57+00:00</updated>
<author>
<name>Anton Blanchard</name>
<email>anton@samba.org</email>
</author>
<published>2016-08-13T01:55:33+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=50de1a0c54cdbc69a6dbcbc323f53daf95a4050e'/>
<id>50de1a0c54cdbc69a6dbcbc323f53daf95a4050e</id>
<content type='text'>
Commit 73cdf0c6ea9c ("perf symbols: Record text offset in dso
to calculate objdump address") started storing the offset of
the text section for all DSOs:

       if (elf_section_by_name(elf, &amp;ehdr, &amp;tshdr, ".text", NULL))
               dso-&gt;text_offset = tshdr.sh_addr - tshdr.sh_offset;

Unfortunately this breaks debuginfo files, because we need to calculate
the offset of the text section in the associated executable file. As a
result perf annotate returns junk for all debuginfo files.

Fix this by using runtime_ss-&gt;elf which should point at the executable
when parsing a debuginfo file.

Signed-off-by: Anton Blanchard &lt;anton@samba.org&gt;
Reviewed-by: Naveen N. Rao &lt;naveen.n.rao@linux.vnet.ibm.com&gt;
Tested-by: Wang Nan &lt;wangnan0@huawei.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Ravi Bangoria &lt;ravi.bangoria@linux.vnet.ibm.com&gt;
Cc: stable@vger.kernel.org # v4.6+
Fixes: 73cdf0c6ea9c ("perf symbols: Record text offset in dso to calculate objdump address")
Link: http://lkml.kernel.org/r/20160813115533.6de17912@kryten
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Commit 73cdf0c6ea9c ("perf symbols: Record text offset in dso
to calculate objdump address") started storing the offset of
the text section for all DSOs:

       if (elf_section_by_name(elf, &amp;ehdr, &amp;tshdr, ".text", NULL))
               dso-&gt;text_offset = tshdr.sh_addr - tshdr.sh_offset;

Unfortunately this breaks debuginfo files, because we need to calculate
the offset of the text section in the associated executable file. As a
result perf annotate returns junk for all debuginfo files.

Fix this by using runtime_ss-&gt;elf which should point at the executable
when parsing a debuginfo file.

Signed-off-by: Anton Blanchard &lt;anton@samba.org&gt;
Reviewed-by: Naveen N. Rao &lt;naveen.n.rao@linux.vnet.ibm.com&gt;
Tested-by: Wang Nan &lt;wangnan0@huawei.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Ravi Bangoria &lt;ravi.bangoria@linux.vnet.ibm.com&gt;
Cc: stable@vger.kernel.org # v4.6+
Fixes: 73cdf0c6ea9c ("perf symbols: Record text offset in dso to calculate objdump address")
Link: http://lkml.kernel.org/r/20160813115533.6de17912@kryten
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf script: Don't disable use_callchain if input is pipe</title>
<updated>2016-08-15T16:23:49+00:00</updated>
<author>
<name>He Kuang</name>
<email>hekuang@huawei.com</email>
</author>
<published>2016-08-04T11:25:43+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=71ac899b5ed7edfd8fa2a4e075194380b1de2d7f'/>
<id>71ac899b5ed7edfd8fa2a4e075194380b1de2d7f</id>
<content type='text'>
Because perf data from pipe do not have a header with evsel attr, we
should not check that and disable symbol_conf.use_callchain. Otherwise,
perf script won't show callchains even if the data stream contains
callchain.

Before:
  $ perf record -g -o - uname |perf script
  Linux
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.000 MB - ]
  uname  1828 182630.186578:  250000 cpu-clock:  ..b9499 setup_arg_pages
  uname  1828 182630.186850:  250000 cpu-clock:  ..83b20 ___might_sleep
  uname  1828 182630.187153:  250000 cpu-clock:  ..4b6be file_map_prot_ch
  ...

After:
  $ perf record -g -o - uname |perf script
  Linux
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.000 MB - ]
  uname  1833 182675.927099:     250000 cpu-clock:
                  ba5520 _raw_spin_lock+0xfe200040 ([kernel.kallsyms])
                  389dd4 expand_downwards+0xfe200154 ([kernel.kallsyms])
                  389f34 expand_stack+0xfe200024 ([kernel.kallsyms])
                  3b957e setup_arg_pages+0xfe20019e ([kernel.kallsyms])
                  40c80f load_elf_binary+0xfe20042f ([kernel.kallsyms])
                  ...

Signed-off-by: He Kuang &lt;hekuang@huawei.com&gt;
Tested-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Wang Nan &lt;wangnan0@huawei.com&gt;
Link: http://lkml.kernel.org/r/1470309943-153909-2-git-send-email-hekuang@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Because perf data from pipe do not have a header with evsel attr, we
should not check that and disable symbol_conf.use_callchain. Otherwise,
perf script won't show callchains even if the data stream contains
callchain.

Before:
  $ perf record -g -o - uname |perf script
  Linux
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.000 MB - ]
  uname  1828 182630.186578:  250000 cpu-clock:  ..b9499 setup_arg_pages
  uname  1828 182630.186850:  250000 cpu-clock:  ..83b20 ___might_sleep
  uname  1828 182630.187153:  250000 cpu-clock:  ..4b6be file_map_prot_ch
  ...

After:
  $ perf record -g -o - uname |perf script
  Linux
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.000 MB - ]
  uname  1833 182675.927099:     250000 cpu-clock:
                  ba5520 _raw_spin_lock+0xfe200040 ([kernel.kallsyms])
                  389dd4 expand_downwards+0xfe200154 ([kernel.kallsyms])
                  389f34 expand_stack+0xfe200024 ([kernel.kallsyms])
                  3b957e setup_arg_pages+0xfe20019e ([kernel.kallsyms])
                  40c80f load_elf_binary+0xfe20042f ([kernel.kallsyms])
                  ...

Signed-off-by: He Kuang &lt;hekuang@huawei.com&gt;
Tested-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Wang Nan &lt;wangnan0@huawei.com&gt;
Link: http://lkml.kernel.org/r/1470309943-153909-2-git-send-email-hekuang@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf script: Show proper message when failed list scripts</title>
<updated>2016-08-15T16:17:10+00:00</updated>
<author>
<name>He Kuang</name>
<email>hekuang@huawei.com</email>
</author>
<published>2016-08-04T11:25:42+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=88ded4d8d94a550624e1827478e13fecf97a7b0a'/>
<id>88ded4d8d94a550624e1827478e13fecf97a7b0a</id>
<content type='text'>
Perf shows the usage message when perf scripts folder failed to open,
which misleads users to let them think the command is being mistyped.

This patch shows a proper message and guides users to check the
PERF_EXEC_PATH environment variable in that case.

Before:

  $ perf script --list

  Usage: perf script [&lt;options&gt;]
   or: perf script [&lt;options&gt;] record &lt;script&gt; [&lt;record-options&gt;] &lt;command&gt;
   or: perf script [&lt;options&gt;] report &lt;script&gt; [script-args]
   or: perf script [&lt;options&gt;] &lt;script&gt; [&lt;record-options&gt;] &lt;command&gt;
   or: perf script [&lt;options&gt;] &lt;top-script&gt; [script-args]

      -l, --list            list available scripts

After:

  $ perf script --list
  open(/home/user/perf-core/scripts) failed.
  Check for "PERF_EXEC_PATH" env to set scripts dir.

Signed-off-by: He Kuang &lt;hekuang@huawei.com&gt;
Tested-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Wang Nan &lt;wangnan0@huawei.com&gt;
Link: http://lkml.kernel.org/r/1470309943-153909-1-git-send-email-hekuang@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Perf shows the usage message when perf scripts folder failed to open,
which misleads users to let them think the command is being mistyped.

This patch shows a proper message and guides users to check the
PERF_EXEC_PATH environment variable in that case.

Before:

  $ perf script --list

  Usage: perf script [&lt;options&gt;]
   or: perf script [&lt;options&gt;] record &lt;script&gt; [&lt;record-options&gt;] &lt;command&gt;
   or: perf script [&lt;options&gt;] report &lt;script&gt; [script-args]
   or: perf script [&lt;options&gt;] &lt;script&gt; [&lt;record-options&gt;] &lt;command&gt;
   or: perf script [&lt;options&gt;] &lt;top-script&gt; [script-args]

      -l, --list            list available scripts

After:

  $ perf script --list
  open(/home/user/perf-core/scripts) failed.
  Check for "PERF_EXEC_PATH" env to set scripts dir.

Signed-off-by: He Kuang &lt;hekuang@huawei.com&gt;
Tested-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Wang Nan &lt;wangnan0@huawei.com&gt;
Link: http://lkml.kernel.org/r/1470309943-153909-1-git-send-email-hekuang@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf jitdump: Add the right header to get the major()/minor() definitions</title>
<updated>2016-08-15T16:10:28+00:00</updated>
<author>
<name>Arnaldo Carvalho de Melo</name>
<email>acme@redhat.com</email>
</author>
<published>2016-08-13T04:12:10+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=49a7f01064a94c4fd6afb6b4f7e3ccc37d0edf99'/>
<id>49a7f01064a94c4fd6afb6b4f7e3ccc37d0edf99</id>
<content type='text'>
Noticed on Fedora Rawhide:

  $ gcc --version
  gcc (GCC) 6.1.1 20160721 (Red Hat 6.1.1-4)
  $ rpm -q glibc
  glibc-2.24.90-1.fc26.x86_64
  $

    CC       /tmp/build/perf/util/jitdump.o
  util/jitdump.c: In function 'jit_repipe_code_load':
  util/jitdump.c:428:2: error: '__major_from_sys_types' is deprecated:
    In the GNU C Library, `major' is defined by &lt;sys/sysmacros.h&gt;.
    For historical compatibility, it is currently defined by
    &lt;sys/types.h&gt; as well, but we plan to remove this soon.
    To use `major', include &lt;sys/sysmacros.h&gt; directly.
    If you did not intend to use a system-defined macro `major',
    you should #undef it after including &lt;sys/types.h&gt;.
    [-Werror=deprecated-declarations]
    event-&gt;mmap2.maj   = major(st.st_dev);
    ^~~~~
  In file included from /usr/include/features.h:397:0,
                   from /usr/include/sys/types.h:25,
                   from util/jitdump.c:1:
  /usr/include/sys/sysmacros.h:87:1: note: declared here
   __SYSMACROS_DEFINE_MAJOR (__SYSMACROS_FST_IMPL_TEMPL)

Fix it following that recomendation.

Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: David Ahern &lt;dsahern@gmail.com&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Stephane Eranian &lt;eranian@google.com&gt;
Cc: Wang Nan &lt;wangnan0@huawei.com&gt;
Link: http://lkml.kernel.org/n/tip-3majvd0adhfr25rvx4v5e9te@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Noticed on Fedora Rawhide:

  $ gcc --version
  gcc (GCC) 6.1.1 20160721 (Red Hat 6.1.1-4)
  $ rpm -q glibc
  glibc-2.24.90-1.fc26.x86_64
  $

    CC       /tmp/build/perf/util/jitdump.o
  util/jitdump.c: In function 'jit_repipe_code_load':
  util/jitdump.c:428:2: error: '__major_from_sys_types' is deprecated:
    In the GNU C Library, `major' is defined by &lt;sys/sysmacros.h&gt;.
    For historical compatibility, it is currently defined by
    &lt;sys/types.h&gt; as well, but we plan to remove this soon.
    To use `major', include &lt;sys/sysmacros.h&gt; directly.
    If you did not intend to use a system-defined macro `major',
    you should #undef it after including &lt;sys/types.h&gt;.
    [-Werror=deprecated-declarations]
    event-&gt;mmap2.maj   = major(st.st_dev);
    ^~~~~
  In file included from /usr/include/features.h:397:0,
                   from /usr/include/sys/types.h:25,
                   from util/jitdump.c:1:
  /usr/include/sys/sysmacros.h:87:1: note: declared here
   __SYSMACROS_DEFINE_MAJOR (__SYSMACROS_FST_IMPL_TEMPL)

Fix it following that recomendation.

Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: David Ahern &lt;dsahern@gmail.com&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Stephane Eranian &lt;eranian@google.com&gt;
Cc: Wang Nan &lt;wangnan0@huawei.com&gt;
Link: http://lkml.kernel.org/n/tip-3majvd0adhfr25rvx4v5e9te@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf ppc64le: Fix build failure when libelf is not present</title>
<updated>2016-08-12T17:39:48+00:00</updated>
<author>
<name>Ravi Bangoria</name>
<email>ravi.bangoria@linux.vnet.ibm.com</email>
</author>
<published>2016-08-11T17:43:59+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=f046f3df665361d2f89e660f8c79ba164069f02a'/>
<id>f046f3df665361d2f89e660f8c79ba164069f02a</id>
<content type='text'>
arch__post_process_probe_trace_events() calls get_target_map() to
prepare symbol table. get_target_map() is defined inside
util/probe-event.c.

probe-event.c will only get included in perf binary if CONFIG_LIBELF is
set.  Hence arch__post_process_probe_trace_events() needs to be defined
inside #ifdef HAVE_LIBELF_SUPPORT to solve compilation error.

Reported-and-Tested-by: Anton Blanchard &lt;anton@samba.org&gt;
Tested-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Balbir Singh &lt;bsingharora@gmail.com&gt;
Cc: Naveen N. Rao &lt;naveen.n.rao@linux.vnet.ibm.com&gt;
Cc: Ananth N Mavinakayanahalli &lt;ananth@in.ibm.com&gt;
Cc: Masami Hiramatsu &lt;mhiramat@kernel.org&gt;
Cc: Wang Nan &lt;wangnan0@huawei.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Link: http://lkml.kernel.org/r/57ABFF88.8030905@linux.vnet.ibm.com
[ Thunderbird MUA mangled it, fix that ]
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
arch__post_process_probe_trace_events() calls get_target_map() to
prepare symbol table. get_target_map() is defined inside
util/probe-event.c.

probe-event.c will only get included in perf binary if CONFIG_LIBELF is
set.  Hence arch__post_process_probe_trace_events() needs to be defined
inside #ifdef HAVE_LIBELF_SUPPORT to solve compilation error.

Reported-and-Tested-by: Anton Blanchard &lt;anton@samba.org&gt;
Tested-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Balbir Singh &lt;bsingharora@gmail.com&gt;
Cc: Naveen N. Rao &lt;naveen.n.rao@linux.vnet.ibm.com&gt;
Cc: Ananth N Mavinakayanahalli &lt;ananth@in.ibm.com&gt;
Cc: Masami Hiramatsu &lt;mhiramat@kernel.org&gt;
Cc: Wang Nan &lt;wangnan0@huawei.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Link: http://lkml.kernel.org/r/57ABFF88.8030905@linux.vnet.ibm.com
[ Thunderbird MUA mangled it, fix that ]
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf tools mem: Fix -t store option for record command</title>
<updated>2016-08-12T17:39:48+00:00</updated>
<author>
<name>Jiri Olsa</name>
<email>jolsa@kernel.org</email>
</author>
<published>2016-08-11T08:50:57+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=33da54fa86e29b87fe1e83bd0f15b4ef2be53ecb'/>
<id>33da54fa86e29b87fe1e83bd0f15b4ef2be53ecb</id>
<content type='text'>
Michael reported 'perf mem -t store record' being broken.  The reason is
latest rework of this area:

  commit acbe613e0c03 ("perf tools: Add monitored events array")

We don't mark perf_mem_events store record when -t store option is
specified.

Committer notes:

Before:

  # perf mem -t store record usleep 1
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.020 MB perf.data (7 samples) ]
  # perf evlist
  cycles:ppp
  #

After:

  # perf mem -t store record usleep 1
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.020 MB perf.data (7 samples) ]
  # perf evlist
  cpu/mem-stores/P
  #

Reported-by: Michael Petlan &lt;mpetlan@redhat.com&gt;
Signed-off-by: Jiri Olsa &lt;jolsa@redhat.com&gt;
Tested-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Cc: David Ahern &lt;dsahern@gmail.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;a.p.zijlstra@chello.nl&gt;
Fixes: acbe613e0c03 ("perf tools: Add monitored events array")
Link: http://lkml.kernel.org/r/1470905457-18311-1-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Michael reported 'perf mem -t store record' being broken.  The reason is
latest rework of this area:

  commit acbe613e0c03 ("perf tools: Add monitored events array")

We don't mark perf_mem_events store record when -t store option is
specified.

Committer notes:

Before:

  # perf mem -t store record usleep 1
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.020 MB perf.data (7 samples) ]
  # perf evlist
  cycles:ppp
  #

After:

  # perf mem -t store record usleep 1
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.020 MB perf.data (7 samples) ]
  # perf evlist
  cpu/mem-stores/P
  #

Reported-by: Michael Petlan &lt;mpetlan@redhat.com&gt;
Signed-off-by: Jiri Olsa &lt;jolsa@redhat.com&gt;
Tested-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Cc: David Ahern &lt;dsahern@gmail.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;a.p.zijlstra@chello.nl&gt;
Fixes: acbe613e0c03 ("perf tools: Add monitored events array")
Link: http://lkml.kernel.org/r/1470905457-18311-1-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf intel-pt: Fix ip compression</title>
<updated>2016-08-12T17:39:48+00:00</updated>
<author>
<name>Adrian Hunter</name>
<email>adrian.hunter@intel.com</email>
</author>
<published>2016-07-20T09:00:06+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=e1717e0485af4f47fc4da1e979ac817f9ad61b0f'/>
<id>e1717e0485af4f47fc4da1e979ac817f9ad61b0f</id>
<content type='text'>
The June 2015 Intel SDM introduced IP Compression types 4 and 6. Refer
to section 36.4.2.2 Target IP (TIP) Packet - IP Compression.

Existing Intel PT packet decoder did not support type 4, and got type 6
wrong.  Because type 3 and type 4 have the same number of bytes, the
packet 'count' has been changed from being the number of ip bytes to
being the type code.  That allows the Intel PT decoder to correctly
decide whether to sign-extend or use the last ip.  However that also
meant the code had to be adjusted in a number of places.

Currently hardware is not using the new compression types, so this fix
has no effect on existing hardware.

Signed-off-by: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Jiri Olsa &lt;jolsa@redhat.com&gt;
Link: http://lkml.kernel.org/r/1469005206-3049-1-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The June 2015 Intel SDM introduced IP Compression types 4 and 6. Refer
to section 36.4.2.2 Target IP (TIP) Packet - IP Compression.

Existing Intel PT packet decoder did not support type 4, and got type 6
wrong.  Because type 3 and type 4 have the same number of bytes, the
packet 'count' has been changed from being the number of ip bytes to
being the type code.  That allows the Intel PT decoder to correctly
decide whether to sign-extend or use the last ip.  However that also
meant the code had to be adjusted in a number of places.

Currently hardware is not using the new compression types, so this fix
has no effect on existing hardware.

Signed-off-by: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Jiri Olsa &lt;jolsa@redhat.com&gt;
Link: http://lkml.kernel.org/r/1469005206-3049-1-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
