<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-stable.git/tools/perf/util/map.c, branch linux-5.0.y</title>
<subtitle>Linux kernel stable tree</subtitle>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/'/>
<entry>
<title>perf tools: Fix map reference counting</title>
<updated>2019-05-16T17:40:21+00:00</updated>
<author>
<name>Jiri Olsa</name>
<email>jolsa@kernel.org</email>
</author>
<published>2019-04-16T16:01:24+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=e9697ba264c288144aced5f9cc05d1d9f51d026d'/>
<id>e9697ba264c288144aced5f9cc05d1d9f51d026d</id>
<content type='text'>
[ Upstream commit b9abbdfa88024d52c8084d8f46ea4f161606c692 ]

By calling maps__insert() we assume to get 2 references on the map,
which we relese within maps__remove call.

However if there's already same map name, we currently don't bump the
reference and can crash, like:

  Program received signal SIGABRT, Aborted.
  0x00007ffff75e60f5 in raise () from /lib64/libc.so.6

  (gdb) bt
  #0  0x00007ffff75e60f5 in raise () from /lib64/libc.so.6
  #1  0x00007ffff75d0895 in abort () from /lib64/libc.so.6
  #2  0x00007ffff75d0769 in __assert_fail_base.cold () from /lib64/libc.so.6
  #3  0x00007ffff75de596 in __assert_fail () from /lib64/libc.so.6
  #4  0x00000000004fc006 in refcount_sub_and_test (i=1, r=0x1224e88) at tools/include/linux/refcount.h:131
  #5  refcount_dec_and_test (r=0x1224e88) at tools/include/linux/refcount.h:148
  #6  map__put (map=0x1224df0) at util/map.c:299
  #7  0x00000000004fdb95 in __maps__remove (map=0x1224df0, maps=0xb17d80) at util/map.c:953
  #8  maps__remove (maps=0xb17d80, map=0x1224df0) at util/map.c:959
  #9  0x00000000004f7d8a in map_groups__remove (map=&lt;optimized out&gt;, mg=&lt;optimized out&gt;) at util/map_groups.h:65
  #10 machine__process_ksymbol_unregister (sample=&lt;optimized out&gt;, event=0x7ffff7279670, machine=&lt;optimized out&gt;) at util/machine.c:728
  #11 machine__process_ksymbol (machine=&lt;optimized out&gt;, event=0x7ffff7279670, sample=&lt;optimized out&gt;) at util/machine.c:741
  #12 0x00000000004fffbb in perf_session__deliver_event (session=0xb11390, event=0x7ffff7279670, tool=0x7fffffffc7b0, file_offset=13936) at util/session.c:1362
  #13 0x00000000005039bb in do_flush (show_progress=false, oe=0xb17e80) at util/ordered-events.c:243
  #14 __ordered_events__flush (oe=0xb17e80, how=OE_FLUSH__ROUND, timestamp=&lt;optimized out&gt;) at util/ordered-events.c:322
  #15 0x00000000005005e4 in perf_session__process_user_event (session=session@entry=0xb11390, event=event@entry=0x7ffff72a4af8,
  ...

Add the map to the list and getting the reference event if we find the
map with same name.

Signed-off-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Alexei Starovoitov &lt;ast@kernel.org&gt;
Cc: Andi Kleen &lt;ak@linux.intel.com&gt;
Cc: Daniel Borkmann &lt;daniel@iogearbox.net&gt;
Cc: Eric Saint-Etienne &lt;eric.saint.etienne@oracle.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Song Liu &lt;songliubraving@fb.com&gt;
Fixes: 1e6285699b30 ("perf symbols: Fix slowness due to -ffunction-section")
Link: http://lkml.kernel.org/r/20190416160127.30203-10-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Signed-off-by: Sasha Levin &lt;sashal@kernel.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
[ Upstream commit b9abbdfa88024d52c8084d8f46ea4f161606c692 ]

By calling maps__insert() we assume to get 2 references on the map,
which we relese within maps__remove call.

However if there's already same map name, we currently don't bump the
reference and can crash, like:

  Program received signal SIGABRT, Aborted.
  0x00007ffff75e60f5 in raise () from /lib64/libc.so.6

  (gdb) bt
  #0  0x00007ffff75e60f5 in raise () from /lib64/libc.so.6
  #1  0x00007ffff75d0895 in abort () from /lib64/libc.so.6
  #2  0x00007ffff75d0769 in __assert_fail_base.cold () from /lib64/libc.so.6
  #3  0x00007ffff75de596 in __assert_fail () from /lib64/libc.so.6
  #4  0x00000000004fc006 in refcount_sub_and_test (i=1, r=0x1224e88) at tools/include/linux/refcount.h:131
  #5  refcount_dec_and_test (r=0x1224e88) at tools/include/linux/refcount.h:148
  #6  map__put (map=0x1224df0) at util/map.c:299
  #7  0x00000000004fdb95 in __maps__remove (map=0x1224df0, maps=0xb17d80) at util/map.c:953
  #8  maps__remove (maps=0xb17d80, map=0x1224df0) at util/map.c:959
  #9  0x00000000004f7d8a in map_groups__remove (map=&lt;optimized out&gt;, mg=&lt;optimized out&gt;) at util/map_groups.h:65
  #10 machine__process_ksymbol_unregister (sample=&lt;optimized out&gt;, event=0x7ffff7279670, machine=&lt;optimized out&gt;) at util/machine.c:728
  #11 machine__process_ksymbol (machine=&lt;optimized out&gt;, event=0x7ffff7279670, sample=&lt;optimized out&gt;) at util/machine.c:741
  #12 0x00000000004fffbb in perf_session__deliver_event (session=0xb11390, event=0x7ffff7279670, tool=0x7fffffffc7b0, file_offset=13936) at util/session.c:1362
  #13 0x00000000005039bb in do_flush (show_progress=false, oe=0xb17e80) at util/ordered-events.c:243
  #14 __ordered_events__flush (oe=0xb17e80, how=OE_FLUSH__ROUND, timestamp=&lt;optimized out&gt;) at util/ordered-events.c:322
  #15 0x00000000005005e4 in perf_session__process_user_event (session=session@entry=0xb11390, event=event@entry=0x7ffff72a4af8,
  ...

Add the map to the list and getting the reference event if we find the
map with same name.

Signed-off-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Alexei Starovoitov &lt;ast@kernel.org&gt;
Cc: Andi Kleen &lt;ak@linux.intel.com&gt;
Cc: Daniel Borkmann &lt;daniel@iogearbox.net&gt;
Cc: Eric Saint-Etienne &lt;eric.saint.etienne@oracle.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Song Liu &lt;songliubraving@fb.com&gt;
Fixes: 1e6285699b30 ("perf symbols: Fix slowness due to -ffunction-section")
Link: http://lkml.kernel.org/r/20190416160127.30203-10-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Signed-off-by: Sasha Levin &lt;sashal@kernel.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf maps: Purge all maps from the 'names' tree</title>
<updated>2019-04-20T07:16:54+00:00</updated>
<author>
<name>Changbin Du</name>
<email>changbin.du@gmail.com</email>
</author>
<published>2019-03-16T08:05:51+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=940df86f505d30aafc0620b2fdc9e52282c79dcc'/>
<id>940df86f505d30aafc0620b2fdc9e52282c79dcc</id>
<content type='text'>
[ Upstream commit da3a53a7390a89391bd63bead0c2e9af4c5ef3d6 ]

Add function __maps__purge_names() to purge all maps from the names
tree.  We need to cleanup the names tree in maps__exit().

Detected with gcc's ASan.

Signed-off-by: Changbin Du &lt;changbin.du@gmail.com&gt;
Reviewed-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Alexei Starovoitov &lt;ast@kernel.org&gt;
Cc: Daniel Borkmann &lt;daniel@iogearbox.net&gt;
Cc: Eric Saint-Etienne &lt;eric.saint.etienne@oracle.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Steven Rostedt (VMware) &lt;rostedt@goodmis.org&gt;
Fixes: 1e6285699b30 ("perf symbols: Fix slowness due to -ffunction-section")
Link: http://lkml.kernel.org/r/20190316080556.3075-12-changbin.du@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Signed-off-by: Sasha Levin &lt;sashal@kernel.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
[ Upstream commit da3a53a7390a89391bd63bead0c2e9af4c5ef3d6 ]

Add function __maps__purge_names() to purge all maps from the names
tree.  We need to cleanup the names tree in maps__exit().

Detected with gcc's ASan.

Signed-off-by: Changbin Du &lt;changbin.du@gmail.com&gt;
Reviewed-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Alexei Starovoitov &lt;ast@kernel.org&gt;
Cc: Daniel Borkmann &lt;daniel@iogearbox.net&gt;
Cc: Eric Saint-Etienne &lt;eric.saint.etienne@oracle.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Steven Rostedt (VMware) &lt;rostedt@goodmis.org&gt;
Fixes: 1e6285699b30 ("perf symbols: Fix slowness due to -ffunction-section")
Link: http://lkml.kernel.org/r/20190316080556.3075-12-changbin.du@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Signed-off-by: Sasha Levin &lt;sashal@kernel.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf map: Remove map from 'names' tree in __maps__remove()</title>
<updated>2019-04-20T07:16:54+00:00</updated>
<author>
<name>Changbin Du</name>
<email>changbin.du@gmail.com</email>
</author>
<published>2019-03-16T08:05:50+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=60b7f41c4aea4d09bc36c957014e4bc7aa056597'/>
<id>60b7f41c4aea4d09bc36c957014e4bc7aa056597</id>
<content type='text'>
[ Upstream commit b49265e04410b97b31a5ee66ef6782c1b2d6cd2c ]

There are two trees for each map inserted by maps__insert(), so remove
it from the 'names' tree in __maps__remove().

Detected with gcc's ASan.

Signed-off-by: Changbin Du &lt;changbin.du@gmail.com&gt;
Reviewed-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Alexei Starovoitov &lt;ast@kernel.org&gt;
Cc: Daniel Borkmann &lt;daniel@iogearbox.net&gt;
Cc: Eric Saint-Etienne &lt;eric.saint.etienne@oracle.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Steven Rostedt (VMware) &lt;rostedt@goodmis.org&gt;
Fixes: 1e6285699b30 ("perf symbols: Fix slowness due to -ffunction-section")
Link: http://lkml.kernel.org/r/20190316080556.3075-11-changbin.du@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Signed-off-by: Sasha Levin &lt;sashal@kernel.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
[ Upstream commit b49265e04410b97b31a5ee66ef6782c1b2d6cd2c ]

There are two trees for each map inserted by maps__insert(), so remove
it from the 'names' tree in __maps__remove().

Detected with gcc's ASan.

Signed-off-by: Changbin Du &lt;changbin.du@gmail.com&gt;
Reviewed-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Alexei Starovoitov &lt;ast@kernel.org&gt;
Cc: Daniel Borkmann &lt;daniel@iogearbox.net&gt;
Cc: Eric Saint-Etienne &lt;eric.saint.etienne@oracle.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Steven Rostedt (VMware) &lt;rostedt@goodmis.org&gt;
Fixes: 1e6285699b30 ("perf symbols: Fix slowness due to -ffunction-section")
Link: http://lkml.kernel.org/r/20190316080556.3075-11-changbin.du@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Signed-off-by: Sasha Levin &lt;sashal@kernel.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf tools: Support 'srccode' output</title>
<updated>2018-12-17T17:57:07+00:00</updated>
<author>
<name>Andi Kleen</name>
<email>ak@linux.intel.com</email>
</author>
<published>2018-12-04T00:18:48+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=dd2e18e9ac20e3ffc27cabf318e83b43ed5ddb92'/>
<id>dd2e18e9ac20e3ffc27cabf318e83b43ed5ddb92</id>
<content type='text'>
When looking at PT or brstackinsn traces with 'perf script' it can be
very useful to see the source code. This adds a simple facility to print
them with 'perf script', if the information is available through dwarf

  % perf record ...
  % perf script -F insn,ip,sym,srccode
  ...

            4004c6 main
  5               for (i = 0; i &lt; 10000000; i++)
             4004cd main
  5               for (i = 0; i &lt; 10000000; i++)
             4004c6 main
  5               for (i = 0; i &lt; 10000000; i++)
             4004cd main
  5               for (i = 0; i &lt; 10000000; i++)
             4004cd main
  5               for (i = 0; i &lt; 10000000; i++)
             4004cd main
  5               for (i = 0; i &lt; 10000000; i++)
             4004cd main
  5               for (i = 0; i &lt; 10000000; i++)
             4004cd main
  5               for (i = 0; i &lt; 10000000; i++)
             4004b3 main
  6                       v++;

  % perf record -b ...
  % perf script -F insn,ip,sym,srccode,brstackinsn

  ...
         main+22:
          0000000000400543        insn: e8 ca ff ff ff            # PRED
  |18                     f1();
          f1:
          0000000000400512        insn: 55
  |10       {
          0000000000400513        insn: 48 89 e5
          0000000000400516        insn: b8 00 00 00 00
  |11             f2();
          000000000040051b        insn: e8 d6 ff ff ff            # PRED
          f2:
          00000000004004f6        insn: 55
  |5        {
          00000000004004f7        insn: 48 89 e5
          00000000004004fa        insn: 8b 05 2c 0b 20 00
  |6              c = a / b;
          0000000000400500        insn: 8b 0d 2a 0b 20 00
          0000000000400506        insn: 99
          0000000000400507        insn: f7 f9
          0000000000400509        insn: 89 05 29 0b 20 00
          000000000040050f        insn: 90
  |7        }
          0000000000400510        insn: 5d
          0000000000400511        insn: c3                        # PRED
          f1+14:
          0000000000400520        insn: b8 00 00 00 00
  |12             f2();
          0000000000400525        insn: e8 cc ff ff ff            # PRED
          f2:
          00000000004004f6        insn: 55
  |5        {
          00000000004004f7        insn: 48 89 e5
          00000000004004fa        insn: 8b 05 2c 0b 20 00
  |6              c = a / b;

Not supported for callchains currently, would need some layout changes
there.

Committer notes:

Fixed the build on Alpine Linux (3.4 .. 3.8) by addressing this
warning:

  In file included from util/srccode.c:19:0:
  /usr/include/sys/fcntl.h:1:2: error: #warning redirecting incorrect #include &lt;sys/fcntl.h&gt; to &lt;fcntl.h&gt; [-Werror=cpp]
   #warning redirecting incorrect #include &lt;sys/fcntl.h&gt; to &lt;fcntl.h&gt;
    ^~~~~~~
  cc1: all warnings being treated as errors

Signed-off-by: Andi Kleen &lt;ak@linux.intel.com&gt;
Tested-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Link: http://lkml.kernel.org/r/20181204001848.24769-1-andi@firstfloor.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>
When looking at PT or brstackinsn traces with 'perf script' it can be
very useful to see the source code. This adds a simple facility to print
them with 'perf script', if the information is available through dwarf

  % perf record ...
  % perf script -F insn,ip,sym,srccode
  ...

            4004c6 main
  5               for (i = 0; i &lt; 10000000; i++)
             4004cd main
  5               for (i = 0; i &lt; 10000000; i++)
             4004c6 main
  5               for (i = 0; i &lt; 10000000; i++)
             4004cd main
  5               for (i = 0; i &lt; 10000000; i++)
             4004cd main
  5               for (i = 0; i &lt; 10000000; i++)
             4004cd main
  5               for (i = 0; i &lt; 10000000; i++)
             4004cd main
  5               for (i = 0; i &lt; 10000000; i++)
             4004cd main
  5               for (i = 0; i &lt; 10000000; i++)
             4004b3 main
  6                       v++;

  % perf record -b ...
  % perf script -F insn,ip,sym,srccode,brstackinsn

  ...
         main+22:
          0000000000400543        insn: e8 ca ff ff ff            # PRED
  |18                     f1();
          f1:
          0000000000400512        insn: 55
  |10       {
          0000000000400513        insn: 48 89 e5
          0000000000400516        insn: b8 00 00 00 00
  |11             f2();
          000000000040051b        insn: e8 d6 ff ff ff            # PRED
          f2:
          00000000004004f6        insn: 55
  |5        {
          00000000004004f7        insn: 48 89 e5
          00000000004004fa        insn: 8b 05 2c 0b 20 00
  |6              c = a / b;
          0000000000400500        insn: 8b 0d 2a 0b 20 00
          0000000000400506        insn: 99
          0000000000400507        insn: f7 f9
          0000000000400509        insn: 89 05 29 0b 20 00
          000000000040050f        insn: 90
  |7        }
          0000000000400510        insn: 5d
          0000000000400511        insn: c3                        # PRED
          f1+14:
          0000000000400520        insn: b8 00 00 00 00
  |12             f2();
          0000000000400525        insn: e8 cc ff ff ff            # PRED
          f2:
          00000000004004f6        insn: 55
  |5        {
          00000000004004f7        insn: 48 89 e5
          00000000004004fa        insn: 8b 05 2c 0b 20 00
  |6              c = a / b;

Not supported for callchains currently, would need some layout changes
there.

Committer notes:

Fixed the build on Alpine Linux (3.4 .. 3.8) by addressing this
warning:

  In file included from util/srccode.c:19:0:
  /usr/include/sys/fcntl.h:1:2: error: #warning redirecting incorrect #include &lt;sys/fcntl.h&gt; to &lt;fcntl.h&gt; [-Werror=cpp]
   #warning redirecting incorrect #include &lt;sys/fcntl.h&gt; to &lt;fcntl.h&gt;
    ^~~~~~~
  cc1: all warnings being treated as errors

Signed-off-by: Andi Kleen &lt;ak@linux.intel.com&gt;
Tested-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Link: http://lkml.kernel.org/r/20181204001848.24769-1-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf map: Remove extra indirection from map__find()</title>
<updated>2018-12-17T17:53:57+00:00</updated>
<author>
<name>Eric Saint-Etienne</name>
<email>eric.saint.etienne@oracle.com</email>
</author>
<published>2018-11-23T10:42:39+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=b18e088825883bcb8dc4c4a641494049cf8ccec3'/>
<id>b18e088825883bcb8dc4c4a641494049cf8ccec3</id>
<content type='text'>
A double pointer is used in map__find() where a single pointer is enough
because the function doesn't affect the rbtree and the rbtree is locked.

Signed-off-by: Eric Saint-Etienne &lt;eric.saint.etienne@oracle.com&gt;
Acked-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Eric Saint-Etienne &lt;eric.saintetienne@gmail.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lkml.kernel.org/r/1542969759-24346-1-git-send-email-eric.saint.etienne@oracle.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>
A double pointer is used in map__find() where a single pointer is enough
because the function doesn't affect the rbtree and the rbtree is locked.

Signed-off-by: Eric Saint-Etienne &lt;eric.saint.etienne@oracle.com&gt;
Acked-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Eric Saint-Etienne &lt;eric.saintetienne@gmail.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lkml.kernel.org/r/1542969759-24346-1-git-send-email-eric.saint.etienne@oracle.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf symbols: Fix slowness due to -ffunction-section</title>
<updated>2018-11-22T01:39:59+00:00</updated>
<author>
<name>Eric Saint-Etienne</name>
<email>eric.saint.etienne@oracle.com</email>
</author>
<published>2018-11-21T17:51:19+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=1e6285699b3034e6f4d1f091edd46d717580bf7c'/>
<id>1e6285699b3034e6f4d1f091edd46d717580bf7c</id>
<content type='text'>
Perf can take minutes to parse an image when -ffunction-section is used.
This is especially true with the kernel image when it is compiled this
way, which is the arm64 default since the patcheset "Enable deadcode
elimination at link time".

Perf organize maps using a rbtree. Whenever perf finds a new symbols, it
first searches this rbtree for the map it belongs to, by strcmp()'aring
section names.  When it finds the map with the right name, it uses it to
add the symbol. With a usual image there aren't so many maps but when
using -ffunction-section there's basically one map per function.  With
the kernel image that's north of 40,000 maps. For most symbols perf has
to parses the entire rbtree to eventually create a new map and add it.
Consequently perf spends most of the time browsing a rbtree that keeps
getting larger.

This performance fix introduces a secondary rbtree that indexes maps
based on the section name.

Signed-off-by: Eric Saint-Etienne &lt;eric.saint.etienne@oracle.com&gt;
Reviewed-by: Dave Kleikamp &lt;dave.kleikamp@oracle.com&gt;
Reviewed-by: David Aldridge &lt;david.aldridge@oracle.com&gt;
Reviewed-by: Rob Gardner &lt;rob.gardner@oracle.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Jiri Olsa &lt;jolsa@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lkml.kernel.org/r/1542822679-25591-1-git-send-email-eric.saint.etienne@oracle.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 can take minutes to parse an image when -ffunction-section is used.
This is especially true with the kernel image when it is compiled this
way, which is the arm64 default since the patcheset "Enable deadcode
elimination at link time".

Perf organize maps using a rbtree. Whenever perf finds a new symbols, it
first searches this rbtree for the map it belongs to, by strcmp()'aring
section names.  When it finds the map with the right name, it uses it to
add the symbol. With a usual image there aren't so many maps but when
using -ffunction-section there's basically one map per function.  With
the kernel image that's north of 40,000 maps. For most symbols perf has
to parses the entire rbtree to eventually create a new map and add it.
Consequently perf spends most of the time browsing a rbtree that keeps
getting larger.

This performance fix introduces a secondary rbtree that indexes maps
based on the section name.

Signed-off-by: Eric Saint-Etienne &lt;eric.saint.etienne@oracle.com&gt;
Reviewed-by: Dave Kleikamp &lt;dave.kleikamp@oracle.com&gt;
Reviewed-by: David Aldridge &lt;david.aldridge@oracle.com&gt;
Reviewed-by: Rob Gardner &lt;rob.gardner@oracle.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Jiri Olsa &lt;jolsa@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lkml.kernel.org/r/1542822679-25591-1-git-send-email-eric.saint.etienne@oracle.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>Merge tag 'v4.19-rc5' into perf/core, to pick up fixes</title>
<updated>2018-09-25T09:19:44+00:00</updated>
<author>
<name>Ingo Molnar</name>
<email>mingo@kernel.org</email>
</author>
<published>2018-09-25T09:19:44+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=fb437bc8fe36c964d9bf2f4b568e6c77c235c5e3'/>
<id>fb437bc8fe36c964d9bf2f4b568e6c77c235c5e3</id>
<content type='text'>
Signed-off-by: Ingo Molnar &lt;mingo@kernel.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Signed-off-by: Ingo Molnar &lt;mingo@kernel.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf tools: Fix maps__find_symbol_by_name()</title>
<updated>2018-09-11T17:12:51+00:00</updated>
<author>
<name>Adrian Hunter</name>
<email>adrian.hunter@intel.com</email>
</author>
<published>2018-09-07T08:51:16+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=03db8b583d1c3c84963e08e2abf6c79081da5c31'/>
<id>03db8b583d1c3c84963e08e2abf6c79081da5c31</id>
<content type='text'>
Commit 1c5aae7710bb ("perf machine: Create maps for x86 PTI entry
trampolines") revealed a problem with maps__find_symbol_by_name() that
resulted in probes not being found e.g.

	$ sudo perf probe xsk_mmap
	xsk_mmap is out of .text, skip it.
	Probe point 'xsk_mmap' not found.
	   Error: Failed to add events.

maps__find_symbol_by_name() can optionally return the map of the found
symbol. It can get the map wrong because, in fact, the symbol is found
on the map's dso, not allowing for the possibility that the dso has more
than one map. Fix by always checking the map contains the symbol.

Reported-by: Björn Töpel &lt;bjorn.topel@intel.com&gt;
Signed-off-by: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Tested-by: Björn Töpel &lt;bjorn.topel@intel.com&gt;
Cc: Jiri Olsa &lt;jolsa@redhat.com&gt;
Cc: stable@vger.kernel.org
Fixes: 1c5aae7710bb ("perf machine: Create maps for x86 PTI entry trampolines")
Link: http://lkml.kernel.org/r/20180907085116.25782-1-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>
Commit 1c5aae7710bb ("perf machine: Create maps for x86 PTI entry
trampolines") revealed a problem with maps__find_symbol_by_name() that
resulted in probes not being found e.g.

	$ sudo perf probe xsk_mmap
	xsk_mmap is out of .text, skip it.
	Probe point 'xsk_mmap' not found.
	   Error: Failed to add events.

maps__find_symbol_by_name() can optionally return the map of the found
symbol. It can get the map wrong because, in fact, the symbol is found
on the map's dso, not allowing for the possibility that the dso has more
than one map. Fix by always checking the map contains the symbol.

Reported-by: Björn Töpel &lt;bjorn.topel@intel.com&gt;
Signed-off-by: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Tested-by: Björn Töpel &lt;bjorn.topel@intel.com&gt;
Cc: Jiri Olsa &lt;jolsa@redhat.com&gt;
Cc: stable@vger.kernel.org
Fixes: 1c5aae7710bb ("perf machine: Create maps for x86 PTI entry trampolines")
Link: http://lkml.kernel.org/r/20180907085116.25782-1-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf map: Turn some pr_warning() to pr_debug()</title>
<updated>2018-09-04T19:51:12+00:00</updated>
<author>
<name>Arnaldo Carvalho de Melo</name>
<email>acme@redhat.com</email>
</author>
<published>2018-09-04T13:43:07+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=d8e75a110df7e3318990c9fb207ae0aa7812895a'/>
<id>d8e75a110df7e3318990c9fb207ae0aa7812895a</id>
<content type='text'>
Annoying when using it with --stdio/--stdio2, so just turn them debug,
we can get those using -v.

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: Wang Nan &lt;wangnan0@huawei.com&gt;
Link: https://lkml.kernel.org/n/tip-t3684lkugnf1w4lwcmpj9ivm@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>
Annoying when using it with --stdio/--stdio2, so just turn them debug,
we can get those using -v.

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: Wang Nan &lt;wangnan0@huawei.com&gt;
Link: https://lkml.kernel.org/n/tip-t3684lkugnf1w4lwcmpj9ivm@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf map: Optimize maps__fixup_overlappings()</title>
<updated>2018-08-08T18:56:00+00:00</updated>
<author>
<name>Konstantin Khlebnikov</name>
<email>khlebnikov@yandex-team.ru</email>
</author>
<published>2018-08-07T14:24:54+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=6a9405b56c274024564f9014bba97b92c91b34d6'/>
<id>6a9405b56c274024564f9014bba97b92c91b34d6</id>
<content type='text'>
This function splits and removes overlapping areas.

Maps in tree are ordered by start address thus we could find first
overlap and stop if next map does not overlap.

Signed-off-by: Konstantin Khlebnikov &lt;khlebnikov@yandex-team.ru&gt;
Acked-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lkml.kernel.org/r/153365189407.435244.7234821822450484712.stgit@buzz
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>
This function splits and removes overlapping areas.

Maps in tree are ordered by start address thus we could find first
overlap and stop if next map does not overlap.

Signed-off-by: Konstantin Khlebnikov &lt;khlebnikov@yandex-team.ru&gt;
Acked-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lkml.kernel.org/r/153365189407.435244.7234821822450484712.stgit@buzz
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
