<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-stable.git/tools/lib/perf/include/internal, branch linux-5.15.y</title>
<subtitle>Linux kernel stable tree</subtitle>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/'/>
<entry>
<title>libperf evlist: Avoid out-of-bounds access</title>
<updated>2024-04-13T11:01:45+00:00</updated>
<author>
<name>Ian Rogers</name>
<email>irogers@google.com</email>
</author>
<published>2024-02-29T07:07:57+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=be113e082b65433135b9e85e442fb8c5f0c87534'/>
<id>be113e082b65433135b9e85e442fb8c5f0c87534</id>
<content type='text'>
[ Upstream commit 1947b92464c3268381604bbe2ac977a3fd78192f ]

Parallel testing appears to show a race between allocating and setting
evsel ids. As there is a bounds check on the xyarray it yields a segv
like:

```
AddressSanitizer:DEADLYSIGNAL

=================================================================

==484408==ERROR: AddressSanitizer: SEGV on unknown address 0x000000000010

==484408==The signal is caused by a WRITE memory access.

==484408==Hint: address points to the zero page.

    #0 0x55cef5d4eff4 in perf_evlist__id_hash tools/lib/perf/evlist.c:256
    #1 0x55cef5d4f132 in perf_evlist__id_add tools/lib/perf/evlist.c:274
    #2 0x55cef5d4f545 in perf_evlist__id_add_fd tools/lib/perf/evlist.c:315
    #3 0x55cef5a1923f in store_evsel_ids util/evsel.c:3130
    #4 0x55cef5a19400 in evsel__store_ids util/evsel.c:3147
    #5 0x55cef5888204 in __run_perf_stat tools/perf/builtin-stat.c:832
    #6 0x55cef5888c06 in run_perf_stat tools/perf/builtin-stat.c:960
    #7 0x55cef58932db in cmd_stat tools/perf/builtin-stat.c:2878
...
```

Avoid this crash by early exiting the perf_evlist__id_add_fd and
perf_evlist__id_add is the access is out-of-bounds.

Signed-off-by: Ian Rogers &lt;irogers@google.com&gt;
Cc: Yang Jihong &lt;yangjihong1@huawei.com&gt;
Signed-off-by: Namhyung Kim &lt;namhyung@kernel.org&gt;
Link: https://lore.kernel.org/r/20240229070757.796244-1-irogers@google.com
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 1947b92464c3268381604bbe2ac977a3fd78192f ]

Parallel testing appears to show a race between allocating and setting
evsel ids. As there is a bounds check on the xyarray it yields a segv
like:

```
AddressSanitizer:DEADLYSIGNAL

=================================================================

==484408==ERROR: AddressSanitizer: SEGV on unknown address 0x000000000010

==484408==The signal is caused by a WRITE memory access.

==484408==Hint: address points to the zero page.

    #0 0x55cef5d4eff4 in perf_evlist__id_hash tools/lib/perf/evlist.c:256
    #1 0x55cef5d4f132 in perf_evlist__id_add tools/lib/perf/evlist.c:274
    #2 0x55cef5d4f545 in perf_evlist__id_add_fd tools/lib/perf/evlist.c:315
    #3 0x55cef5a1923f in store_evsel_ids util/evsel.c:3130
    #4 0x55cef5a19400 in evsel__store_ids util/evsel.c:3147
    #5 0x55cef5888204 in __run_perf_stat tools/perf/builtin-stat.c:832
    #6 0x55cef5888c06 in run_perf_stat tools/perf/builtin-stat.c:960
    #7 0x55cef58932db in cmd_stat tools/perf/builtin-stat.c:2878
...
```

Avoid this crash by early exiting the perf_evlist__id_add_fd and
perf_evlist__id_add is the access is out-of-bounds.

Signed-off-by: Ian Rogers &lt;irogers@google.com&gt;
Cc: Yang Jihong &lt;yangjihong1@huawei.com&gt;
Signed-off-by: Namhyung Kim &lt;namhyung@kernel.org&gt;
Link: https://lore.kernel.org/r/20240229070757.796244-1-irogers@google.com
Signed-off-by: Sasha Levin &lt;sashal@kernel.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>libperf tests: Fix verbose printing</title>
<updated>2021-08-24T18:05:35+00:00</updated>
<author>
<name>Shunsuke Nakamura</name>
<email>nakamura.shun@fujitsu.com</email>
</author>
<published>2021-08-20T09:39:07+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=37c3193fa4d7ddf93947264fa38d71120ff20c2a'/>
<id>37c3193fa4d7ddf93947264fa38d71120ff20c2a</id>
<content type='text'>
libperf's verbose printing checks the -v option every time the macro _T_ START
is called.

Since there are currently four libperf tests registered, the macro _T_ START is
called four times, but verbose printing after the second time is not output.

Resets the index of the element processed by getopt() and fix verbose printing
so that it prints in all tests.

Signed-off-by: Shunsuke Nakamura &lt;nakamura.shun@fujitsu.com&gt;
Acked-by: Rob Herring &lt;robh@kernel.org&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Jiri Olsa &lt;jolsa@redhat.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lore.kernel.org/lkml/20210820093908.734503-3-nakamura.shun@fujitsu.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>
libperf's verbose printing checks the -v option every time the macro _T_ START
is called.

Since there are currently four libperf tests registered, the macro _T_ START is
called four times, but verbose printing after the second time is not output.

Resets the index of the element processed by getopt() and fix verbose printing
so that it prints in all tests.

Signed-off-by: Shunsuke Nakamura &lt;nakamura.shun@fujitsu.com&gt;
Acked-by: Rob Herring &lt;robh@kernel.org&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Jiri Olsa &lt;jolsa@redhat.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lore.kernel.org/lkml/20210820093908.734503-3-nakamura.shun@fujitsu.com
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>libperf: Adopt evlist__set_leader() from tools/perf as perf_evlist__set_leader()</title>
<updated>2021-07-09T17:04:32+00:00</updated>
<author>
<name>Jiri Olsa</name>
<email>jolsa@redhat.com</email>
</author>
<published>2021-07-06T15:17:02+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=2e6263ab54322bac27589d1102f7a955d3f7ff80'/>
<id>2e6263ab54322bac27589d1102f7a955d3f7ff80</id>
<content type='text'>
Move the implementation of evlist__set_leader() to a new libperf
perf_evlist__set_leader() function with the same functionality make it a
libperf exported API.

Signed-off-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Requested-by: Shunsuke Nakamura &lt;nakamura.shun@fujitsu.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Michael Petlan &lt;mpetlan@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lore.kernel.org/lkml/20210706151704.73662-6-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>
Move the implementation of evlist__set_leader() to a new libperf
perf_evlist__set_leader() function with the same functionality make it a
libperf exported API.

Signed-off-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Requested-by: Shunsuke Nakamura &lt;nakamura.shun@fujitsu.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Michael Petlan &lt;mpetlan@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lore.kernel.org/lkml/20210706151704.73662-6-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>libperf: Move 'nr_groups' from tools/perf to evlist::nr_groups</title>
<updated>2021-07-09T17:04:32+00:00</updated>
<author>
<name>Jiri Olsa</name>
<email>jolsa@redhat.com</email>
</author>
<published>2021-07-06T15:17:01+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=3a683120d88586056c0f6728b11da895bbd773a3'/>
<id>3a683120d88586056c0f6728b11da895bbd773a3</id>
<content type='text'>
Move evsel::nr_groups to perf_evsel::nr_groups, so we can move the group
interface to libperf.

Signed-off-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Requested-by: Shunsuke Nakamura &lt;nakamura.shun@fujitsu.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Michael Petlan &lt;mpetlan@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lore.kernel.org/lkml/20210706151704.73662-5-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>
Move evsel::nr_groups to perf_evsel::nr_groups, so we can move the group
interface to libperf.

Signed-off-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Requested-by: Shunsuke Nakamura &lt;nakamura.shun@fujitsu.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Michael Petlan &lt;mpetlan@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lore.kernel.org/lkml/20210706151704.73662-5-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>libperf: Move 'leader' from tools/perf to perf_evsel::leader</title>
<updated>2021-07-09T17:04:31+00:00</updated>
<author>
<name>Jiri Olsa</name>
<email>jolsa@redhat.com</email>
</author>
<published>2021-07-06T15:17:00+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=fba7c86601e2e42d7057db47bf6d45865a208b8c'/>
<id>fba7c86601e2e42d7057db47bf6d45865a208b8c</id>
<content type='text'>
Move evsel::leader to perf_evsel::leader, so we can move the group
interface to libperf.

Also add several evsel helpers to ease up the transition:

  struct evsel *evsel__leader(struct evsel *evsel);
  - get leader evsel

  bool evsel__has_leader(struct evsel *evsel, struct evsel *leader);
  - true if evsel has leader as leader

  bool evsel__is_leader(struct evsel *evsel);
  - true if evsel is itw own leader

  void evsel__set_leader(struct evsel *evsel, struct evsel *leader);
  - set leader for evsel

Committer notes:

Fix this when building with 'make BUILD_BPF_SKEL=1'

  tools/perf/util/bpf_counter.c

  -       if (evsel-&gt;leader-&gt;core.nr_members &gt; 1) {
  +       if (evsel-&gt;core.leader-&gt;nr_members &gt; 1) {

Signed-off-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Requested-by: Shunsuke Nakamura &lt;nakamura.shun@fujitsu.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Michael Petlan &lt;mpetlan@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lore.kernel.org/lkml/20210706151704.73662-4-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>
Move evsel::leader to perf_evsel::leader, so we can move the group
interface to libperf.

Also add several evsel helpers to ease up the transition:

  struct evsel *evsel__leader(struct evsel *evsel);
  - get leader evsel

  bool evsel__has_leader(struct evsel *evsel, struct evsel *leader);
  - true if evsel has leader as leader

  bool evsel__is_leader(struct evsel *evsel);
  - true if evsel is itw own leader

  void evsel__set_leader(struct evsel *evsel, struct evsel *leader);
  - set leader for evsel

Committer notes:

Fix this when building with 'make BUILD_BPF_SKEL=1'

  tools/perf/util/bpf_counter.c

  -       if (evsel-&gt;leader-&gt;core.nr_members &gt; 1) {
  +       if (evsel-&gt;core.leader-&gt;nr_members &gt; 1) {

Signed-off-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Requested-by: Shunsuke Nakamura &lt;nakamura.shun@fujitsu.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Michael Petlan &lt;mpetlan@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lore.kernel.org/lkml/20210706151704.73662-4-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>libperf: Move 'idx' from tools/perf to perf_evsel::idx</title>
<updated>2021-07-09T17:04:28+00:00</updated>
<author>
<name>Jiri Olsa</name>
<email>jolsa@redhat.com</email>
</author>
<published>2021-07-06T15:16:59+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=38fe0e0156c037c060f81fe4e36549fae760322d'/>
<id>38fe0e0156c037c060f81fe4e36549fae760322d</id>
<content type='text'>
Move evsel::idx to perf_evsel::idx, so we can move the group interface
to libperf.

Committer notes:

Fixup evsel-&gt;idx usage in tools/perf/util/bpf_counter_cgroup.c, that
appeared in my tree in my local tree.

Also fixed up these:

$ find tools/perf/ -name "*.[ch]" | xargs grep 'evsel-&gt;idx'
tools/perf/ui/gtk/annotate.c:                      evsel-&gt;idx + i);
tools/perf/ui/gtk/annotate.c:                   evsel-&gt;idx);
$

That running 'make -C tools/perf build-test' caught.

Signed-off-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Requested-by: Shunsuke Nakamura &lt;nakamura.shun@fujitsu.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Michael Petlan &lt;mpetlan@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lore.kernel.org/lkml/20210706151704.73662-3-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>
Move evsel::idx to perf_evsel::idx, so we can move the group interface
to libperf.

Committer notes:

Fixup evsel-&gt;idx usage in tools/perf/util/bpf_counter_cgroup.c, that
appeared in my tree in my local tree.

Also fixed up these:

$ find tools/perf/ -name "*.[ch]" | xargs grep 'evsel-&gt;idx'
tools/perf/ui/gtk/annotate.c:                      evsel-&gt;idx + i);
tools/perf/ui/gtk/annotate.c:                   evsel-&gt;idx);
$

That running 'make -C tools/perf build-test' caught.

Signed-off-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Requested-by: Shunsuke Nakamura &lt;nakamura.shun@fujitsu.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Michael Petlan &lt;mpetlan@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lore.kernel.org/lkml/20210706151704.73662-3-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>libperf: Change tests to single static and shared binaries</title>
<updated>2021-07-07T14:41:58+00:00</updated>
<author>
<name>Jiri Olsa</name>
<email>jolsa@redhat.com</email>
</author>
<published>2021-07-06T15:16:58+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=3d970601dacaefc025a2d690dfd63f56b1bdde0b'/>
<id>3d970601dacaefc025a2d690dfd63f56b1bdde0b</id>
<content type='text'>
Make tests to be two binaries 'tests_static' and 'tests_shared', so the
maintenance is easier.

Adding tests under libperf build system, so we define all the flags just
once.

Adding make-tests tule to just compile tests without running them.

Signed-off-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Michael Petlan &lt;mpetlan@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Shunsuke Nakamura &lt;nakamura.shun@fujitsu.com&gt;
Link: http://lore.kernel.org/lkml/20210706151704.73662-2-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>
Make tests to be two binaries 'tests_static' and 'tests_shared', so the
maintenance is easier.

Adding tests under libperf build system, so we define all the flags just
once.

Adding make-tests tule to just compile tests without running them.

Signed-off-by: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Michael Petlan &lt;mpetlan@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Shunsuke Nakamura &lt;nakamura.shun@fujitsu.com&gt;
Link: http://lore.kernel.org/lkml/20210706151704.73662-2-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>libperf xyarray: Add bounds checks to xyarray__entry()</title>
<updated>2021-04-20T11:11:33+00:00</updated>
<author>
<name>Rob Herring</name>
<email>robh@kernel.org</email>
</author>
<published>2021-04-14T19:57:58+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=818869489ba3c4a4ed1360e22b2f66be488ea9f5'/>
<id>818869489ba3c4a4ed1360e22b2f66be488ea9f5</id>
<content type='text'>
xyarray__entry() is missing any bounds checking yet often the x and y
parameters come from external callers. Add bounds checks and an
unchecked __xyarray__entry().

Committer notes:

Make the 'x' and 'y' arguments to the new xyarray__entry() that does
bounds check to be of type 'size_t', so that we cover also the case
where 'x' and 'y' could be negative, which is needed anyway as having
them as 'int' breaks the build with:

  /home/acme/git/perf/tools/lib/perf/include/internal/xyarray.h: In function ‘xyarray__entry’:
  /home/acme/git/perf/tools/lib/perf/include/internal/xyarray.h:28:8: error: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Werror=sign-compare]
     28 |  if (x &gt;= xy-&gt;max_x || y &gt;= xy-&gt;max_y)
        |        ^~
  /home/acme/git/perf/tools/lib/perf/include/internal/xyarray.h:28:26: error: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Werror=sign-compare]
     28 |  if (x &gt;= xy-&gt;max_x || y &gt;= xy-&gt;max_y)
        |                          ^~
  cc1: all warnings being treated as errors

Signed-off-by: Rob Herring &lt;robh@kernel.org&gt;
Suggested-by: Arnaldo Carvalho de Melo &lt;acme@kernel.org&gt;
Suggested-by: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Jiri Olsa &lt;jolsa@redhat.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lore.kernel.org/lkml/20210414195758.4078803-1-robh@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>
xyarray__entry() is missing any bounds checking yet often the x and y
parameters come from external callers. Add bounds checks and an
unchecked __xyarray__entry().

Committer notes:

Make the 'x' and 'y' arguments to the new xyarray__entry() that does
bounds check to be of type 'size_t', so that we cover also the case
where 'x' and 'y' could be negative, which is needed anyway as having
them as 'int' breaks the build with:

  /home/acme/git/perf/tools/lib/perf/include/internal/xyarray.h: In function ‘xyarray__entry’:
  /home/acme/git/perf/tools/lib/perf/include/internal/xyarray.h:28:8: error: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Werror=sign-compare]
     28 |  if (x &gt;= xy-&gt;max_x || y &gt;= xy-&gt;max_y)
        |        ^~
  /home/acme/git/perf/tools/lib/perf/include/internal/xyarray.h:28:26: error: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Werror=sign-compare]
     28 |  if (x &gt;= xy-&gt;max_x || y &gt;= xy-&gt;max_y)
        |                          ^~
  cc1: all warnings being treated as errors

Signed-off-by: Rob Herring &lt;robh@kernel.org&gt;
Suggested-by: Arnaldo Carvalho de Melo &lt;acme@kernel.org&gt;
Suggested-by: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Jiri Olsa &lt;jolsa@redhat.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Link: http://lore.kernel.org/lkml/20210414195758.4078803-1-robh@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>libperf: Add support for user space counter access</title>
<updated>2021-04-20T11:10:45+00:00</updated>
<author>
<name>Rob Herring</name>
<email>robh@kernel.org</email>
</author>
<published>2021-04-14T16:07:39+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=47d01e7b9999b9591077a59a1ecec11c6ce570de'/>
<id>47d01e7b9999b9591077a59a1ecec11c6ce570de</id>
<content type='text'>
x86 and arm64 can both support direct access of event counters in
userspace. The access sequence is less than trivial and currently exists
in perf test code (tools/perf/arch/x86/tests/rdpmc.c) with copies in
projects such as PAPI and libpfm4.

In order to support userspace access, an event must be mmapped first
with perf_evsel__mmap(). Then subsequent calls to perf_evsel__read()
will use the fast path (assuming the arch supports it).

Committer notes:

Added a '__maybe_unused' attribute to the read_perf_counter() argument
to fix the build on arches other than x86_64 and arm.

Committer testing:

  Building and running the libperf tests in verbose mode (V=1) now shows
  those "loop = N, count = N" extra lines, testing user space counter
  access.

  # make V=1 -C tools/lib/perf tests
  make: Entering directory '/home/acme/git/perf/tools/lib/perf'
  make -f /home/acme/git/perf/tools/build/Makefile.build dir=. obj=libperf
  make -C /home/acme/git/perf/tools/lib/api/ O= libapi.a
  make -f /home/acme/git/perf/tools/build/Makefile.build dir=./fd obj=libapi
  make -f /home/acme/git/perf/tools/build/Makefile.build dir=./fs obj=libapi
  make -C tests
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -o test-cpumap-a test-cpumap.c ../libperf.a /home/acme/git/perf/tools/lib/api/libapi.a
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -o test-threadmap-a test-threadmap.c ../libperf.a /home/acme/git/perf/tools/lib/api/libapi.a
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -o test-evlist-a test-evlist.c ../libperf.a /home/acme/git/perf/tools/lib/api/libapi.a
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -o test-evsel-a test-evsel.c ../libperf.a /home/acme/git/perf/tools/lib/api/libapi.a
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -L.. -o test-cpumap-so test-cpumap.c /home/acme/git/perf/tools/lib/api/libapi.a -lperf
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -L.. -o test-threadmap-so test-threadmap.c /home/acme/git/perf/tools/lib/api/libapi.a -lperf
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -L.. -o test-evlist-so test-evlist.c /home/acme/git/perf/tools/lib/api/libapi.a -lperf
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -L.. -o test-evsel-so test-evsel.c /home/acme/git/perf/tools/lib/api/libapi.a -lperf
  make -C tests run
  running static:
  - running test-cpumap.c...OK
  - running test-threadmap.c...OK
  - running test-evlist.c...OK
  - running test-evsel.c...
  	loop = 65536, count = 333926
  	loop = 131072, count = 655781
  	loop = 262144, count = 1311141
  	loop = 524288, count = 2630126
  	loop = 1048576, count = 5256955
  	loop = 65536, count = 524594
  	loop = 131072, count = 1058916
  	loop = 262144, count = 2097458
  	loop = 524288, count = 4205429
  	loop = 1048576, count = 8406606
  OK
  running dynamic:
  - running test-cpumap.c...OK
  - running test-threadmap.c...OK
  - running test-evlist.c...OK
  - running test-evsel.c...
  	loop = 65536, count = 328102
  	loop = 131072, count = 655782
  	loop = 262144, count = 1317494
  	loop = 524288, count = 2627851
  	loop = 1048576, count = 5255187
  	loop = 65536, count = 524601
  	loop = 131072, count = 1048923
  	loop = 262144, count = 2107917
  	loop = 524288, count = 4194606
  	loop = 1048576, count = 8409322
  OK
  make: Leaving directory '/home/acme/git/perf/tools/lib/perf'
  #

Signed-off-by: Rob Herring &lt;robh@kernel.org&gt;
Acked-by: Jiri Olsa &lt;jolsa@redhat.com&gt;
Acked-by: Namhyung Kim &lt;namhyung@kernel.org&gt;
Tested-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Cc: Catalin Marinas &lt;catalin.marinas@arm.com&gt;
Cc: Itaru Kitayama &lt;itaru.kitayama@gmail.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Will Deacon &lt;will@kernel.org&gt;
Link: http://lore.kernel.org/lkml/20210414155412.3697605-4-robh@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>
x86 and arm64 can both support direct access of event counters in
userspace. The access sequence is less than trivial and currently exists
in perf test code (tools/perf/arch/x86/tests/rdpmc.c) with copies in
projects such as PAPI and libpfm4.

In order to support userspace access, an event must be mmapped first
with perf_evsel__mmap(). Then subsequent calls to perf_evsel__read()
will use the fast path (assuming the arch supports it).

Committer notes:

Added a '__maybe_unused' attribute to the read_perf_counter() argument
to fix the build on arches other than x86_64 and arm.

Committer testing:

  Building and running the libperf tests in verbose mode (V=1) now shows
  those "loop = N, count = N" extra lines, testing user space counter
  access.

  # make V=1 -C tools/lib/perf tests
  make: Entering directory '/home/acme/git/perf/tools/lib/perf'
  make -f /home/acme/git/perf/tools/build/Makefile.build dir=. obj=libperf
  make -C /home/acme/git/perf/tools/lib/api/ O= libapi.a
  make -f /home/acme/git/perf/tools/build/Makefile.build dir=./fd obj=libapi
  make -f /home/acme/git/perf/tools/build/Makefile.build dir=./fs obj=libapi
  make -C tests
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -o test-cpumap-a test-cpumap.c ../libperf.a /home/acme/git/perf/tools/lib/api/libapi.a
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -o test-threadmap-a test-threadmap.c ../libperf.a /home/acme/git/perf/tools/lib/api/libapi.a
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -o test-evlist-a test-evlist.c ../libperf.a /home/acme/git/perf/tools/lib/api/libapi.a
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -o test-evsel-a test-evsel.c ../libperf.a /home/acme/git/perf/tools/lib/api/libapi.a
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -L.. -o test-cpumap-so test-cpumap.c /home/acme/git/perf/tools/lib/api/libapi.a -lperf
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -L.. -o test-threadmap-so test-threadmap.c /home/acme/git/perf/tools/lib/api/libapi.a -lperf
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -L.. -o test-evlist-so test-evlist.c /home/acme/git/perf/tools/lib/api/libapi.a -lperf
  gcc -I/home/acme/git/perf/tools/lib/perf/include -I/home/acme/git/perf/tools/include -I/home/acme/git/perf/tools/lib -g -Wall -L.. -o test-evsel-so test-evsel.c /home/acme/git/perf/tools/lib/api/libapi.a -lperf
  make -C tests run
  running static:
  - running test-cpumap.c...OK
  - running test-threadmap.c...OK
  - running test-evlist.c...OK
  - running test-evsel.c...
  	loop = 65536, count = 333926
  	loop = 131072, count = 655781
  	loop = 262144, count = 1311141
  	loop = 524288, count = 2630126
  	loop = 1048576, count = 5256955
  	loop = 65536, count = 524594
  	loop = 131072, count = 1058916
  	loop = 262144, count = 2097458
  	loop = 524288, count = 4205429
  	loop = 1048576, count = 8406606
  OK
  running dynamic:
  - running test-cpumap.c...OK
  - running test-threadmap.c...OK
  - running test-evlist.c...OK
  - running test-evsel.c...
  	loop = 65536, count = 328102
  	loop = 131072, count = 655782
  	loop = 262144, count = 1317494
  	loop = 524288, count = 2627851
  	loop = 1048576, count = 5255187
  	loop = 65536, count = 524601
  	loop = 131072, count = 1048923
  	loop = 262144, count = 2107917
  	loop = 524288, count = 4194606
  	loop = 1048576, count = 8409322
  OK
  make: Leaving directory '/home/acme/git/perf/tools/lib/perf'
  #

Signed-off-by: Rob Herring &lt;robh@kernel.org&gt;
Acked-by: Jiri Olsa &lt;jolsa@redhat.com&gt;
Acked-by: Namhyung Kim &lt;namhyung@kernel.org&gt;
Tested-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
Cc: Catalin Marinas &lt;catalin.marinas@arm.com&gt;
Cc: Itaru Kitayama &lt;itaru.kitayama@gmail.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Will Deacon &lt;will@kernel.org&gt;
Link: http://lore.kernel.org/lkml/20210414155412.3697605-4-robh@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>libperf tests: Add support for verbose printing</title>
<updated>2021-04-15T19:40:15+00:00</updated>
<author>
<name>Rob Herring</name>
<email>robh@kernel.org</email>
</author>
<published>2021-04-14T16:07:38+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux-stable.git/commit/?id=d3003d9e686890a1e9f0cc7c08aa02ef2953b1f0'/>
<id>d3003d9e686890a1e9f0cc7c08aa02ef2953b1f0</id>
<content type='text'>
Add __T_VERBOSE() so tests can add verbose output. The verbose output is
enabled with the '-v' command line option. Running 'make tests V=1' will
enable the '-v' option when running the tests.

It'll be used in the next patch, for a user space counter access test.

Signed-off-by: Rob Herring &lt;robh@kernel.org&gt;
Acked-by: Jiri Olsa &lt;jolsa@redhat.com&gt;
Acked-by: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Catalin Marinas &lt;catalin.marinas@arm.com&gt;
Cc: Itaru Kitayama &lt;itaru.kitayama@gmail.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Will Deacon &lt;will@kernel.org&gt;
Link: http://lore.kernel.org/lkml/20210414155412.3697605-3-robh@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>
Add __T_VERBOSE() so tests can add verbose output. The verbose output is
enabled with the '-v' command line option. Running 'make tests V=1' will
enable the '-v' option when running the tests.

It'll be used in the next patch, for a user space counter access test.

Signed-off-by: Rob Herring &lt;robh@kernel.org&gt;
Acked-by: Jiri Olsa &lt;jolsa@redhat.com&gt;
Acked-by: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Catalin Marinas &lt;catalin.marinas@arm.com&gt;
Cc: Itaru Kitayama &lt;itaru.kitayama@gmail.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Will Deacon &lt;will@kernel.org&gt;
Link: http://lore.kernel.org/lkml/20210414155412.3697605-3-robh@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
