[PATCH 1/9] perf: test 15 fix python error on empty result

[PATCH 1/9] perf: test 15 fix python error on empty result

[Thomas Richter]

Commit d78ada4a767 ("perf tests attr: Do not store failed events")
does not create an event file in the /tmp directory when the
perf_open_event() system call failed.
This can lead to a situation where not /tmp/event-xx-yy-zz result
file exists at all (for example on a s390x virtual machine environment)
where no CPUMF hardware is available.

The following command then fails with a python call back chain
instead of printing failure:
[root@s8360046 perf]# /usr/bin/python2 ./tests/attr.py -d ./tests/attr/ \
    -p ./perf -v -ttest-stat-basic
running './tests/attr//test-stat-basic'
Traceback (most recent call last):
  File "./tests/attr.py", line 379, in <module>
    main()
  File "./tests/attr.py", line 370, in main
    run_tests(options)
  File "./tests/attr.py", line 311, in run_tests
    Test(f, options).run()
  File "./tests/attr.py", line 300, in run
    self.compare(self.expect, self.result)
  File "./tests/attr.py", line 248, in compare
    exp_event.diff(res_event)
UnboundLocalError: local variable 'res_event' referenced before assignment
[root@s8360046 perf]#

This patch catches this pitfall and prints an error message
instead:
[root@s8360047 perf]# /usr/bin/python2 ./tests/attr.py -d ./tests/attr/ \
     -p ./perf  -vvv -ttest-stat-basic
running './tests/attr//test-stat-basic'
  loading expected events
    Event event:base-stat
      fd = 1
      group_fd = -1
      flags = 0|8
      [....]
      sample_regs_user = 0
      sample_stack_user = 0
  'PERF_TEST_ATTR=/tmp/tmpJbMQMP ./perf stat -o /tmp/tmpJbMQMP/perf.data -e cycles kill >/dev/null 2>&1' ret '1', expected '1'
  loading result events
  compare
    matching [event:base-stat]
    match: [event:base-stat] matches []
    res_event is empty
FAILED './tests/attr//test-stat-basic' - match failure
[root@s8360047 perf]#

Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
---
 tools/perf/tests/attr.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/tools/perf/tests/attr.py b/tools/perf/tests/attr.py
index 6bb50e8..a13cd78 100644
--- a/tools/perf/tests/attr.py
+++ b/tools/perf/tests/attr.py
@@ -237,6 +237,7 @@ class Test(object):
         # events in result. Fail if there's not any.
         for exp_name, exp_event in expect.items():
             exp_list = []
+            res_event = {}
             log.debug("    matching [%s]" % exp_name)
             for res_name, res_event in result.items():
                 log.debug("      to [%s]" % res_name)
@@ -253,7 +254,10 @@ class Test(object):
                 if exp_event.optional():
                     log.debug("    %s does not match, but is optional" % exp_name)
                 else:
-                    exp_event.diff(res_event)
+                    if not res_event:
+                        log.debug("    res_event is empty");
+                    else:
+                        exp_event.diff(res_event)
                     raise Fail(self, 'match failure');
 
             match[exp_name] = exp_list
-- 
2.9.3

[PATCH 2/9] perf test: perf test case 15 result ignored

[Thomas Richter]

Command perf test -v 15 (Setup struct perf_event_attr test)
always reports success even if the test case fails.
It works correctly if you also specify -F (for don't fork).

   root@s35lp76 perf]# ./perf test -v 15
   15: Setup struct perf_event_attr               :
   --- start ---
   running './tests/attr/test-record-no-delay'
   [ perf record: Woken up 1 times to write data ]
   [ perf record: Captured and wrote 0.002 MB /tmp/tmp4E1h7R/perf.data
     (1 samples) ]
   expected task=0, got 1
   expected precise_ip=0, got 3
   expected wakeup_events=1, got 0
   FAILED './tests/attr/test-record-no-delay' - match failure
   test child finished with 0
   ---- end ----
   Setup struct perf_event_attr: Ok

The reason for the wrong error reporting is the return value of the
system() library call. It is called in run_dir() file tests/attr.c
and returns the exit status, in above case 0xff00.
This value is given as parameter to the exit() function which
can only handle values 0-0xff.
The child process terminates with exit value of 0 and the parent
does not detect any error.

This patch corrects the error reporting and prints the
correct test result.

Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
---
 tools/perf/tests/attr.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/tests/attr.c b/tools/perf/tests/attr.c
index 84c0eb5..9d932c6 100644
--- a/tools/perf/tests/attr.c
+++ b/tools/perf/tests/attr.c
@@ -166,7 +166,7 @@ static int run_dir(const char *d, const char *perf)
 	snprintf(cmd, 3*PATH_MAX, PYTHON " %s/attr.py -d %s/attr/ -p %s %.*s",
 		 d, d, perf, vcnt, v);
 
-	return system(cmd);
+	return system(cmd) ? TEST_FAIL : TEST_OK;
 }
 
 int test__attr(int subtest __maybe_unused)
-- 
2.9.3

[PATCH 3/9] perf: test 15 fix test-stat-group tests

[Thomas Richter]

Both test-stat-group and test-stat-group1 fail.
The reason is an invalid check for the read_format
return code.  The submitted value is 15 and this
value should be returned by the perf_event_open() call
(which it is).
The comparison however is done against an invalid
value of 3.
Fix this by setting the expected read_format value.

Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
---
 tools/perf/tests/attr/test-stat-group  | 2 ++
 tools/perf/tests/attr/test-stat-group1 | 2 ++
 2 files changed, 4 insertions(+)

diff --git a/tools/perf/tests/attr/test-stat-group b/tools/perf/tests/attr/test-stat-group
index fdc1596..8fb3a50 100644
--- a/tools/perf/tests/attr/test-stat-group
+++ b/tools/perf/tests/attr/test-stat-group
@@ -6,10 +6,12 @@ ret     = 1
 [event-1:base-stat]
 fd=1
 group_fd=-1
+read_format=15
 
 [event-2:base-stat]
 fd=2
 group_fd=1
 config=1
 disabled=0
+read_format=15
 enable_on_exec=0
diff --git a/tools/perf/tests/attr/test-stat-group1 b/tools/perf/tests/attr/test-stat-group1
index 2a1f86e..086aa2f 100644
--- a/tools/perf/tests/attr/test-stat-group1
+++ b/tools/perf/tests/attr/test-stat-group1
@@ -6,6 +6,7 @@ ret     = 1
 [event-1:base-stat]
 fd=1
 group_fd=-1
+read_format=15
 
 [event-2:base-stat]
 fd=2
@@ -13,3 +14,4 @@ group_fd=1
 config=1
 disabled=0
 enable_on_exec=0
+read_format=15
-- 
2.9.3

[PATCH 4/9] perf: test 15 add s390x support for stat test cases

[Thomas Richter]

The perf test 15 test cases test-stat-detailed[123]
and test-stat-default fail for s390x.
The reason is missing support of the CPUMF facility
for some of the test.

The tests are now marked optional. On s390x some of the
individual tests still fail, but when marked as optional,
do not cause the complete test to fail.

Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
---
 tools/perf/tests/attr/test-stat-default    |  4 ++++
 tools/perf/tests/attr/test-stat-detailed-1 | 12 ++++++++++++
 tools/perf/tests/attr/test-stat-detailed-2 | 22 ++++++++++++++++++++++
 tools/perf/tests/attr/test-stat-detailed-3 | 22 ++++++++++++++++++++++
 4 files changed, 60 insertions(+)

diff --git a/tools/perf/tests/attr/test-stat-default b/tools/perf/tests/attr/test-stat-default
index e911dbd..0dc3ef0 100644
--- a/tools/perf/tests/attr/test-stat-default
+++ b/tools/perf/tests/attr/test-stat-default
@@ -54,13 +54,17 @@ type=0
 config=1
 
 # PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_INSTRUCTIONS
+# Optional on s390x
 [event9:base-stat]
 fd=9
 type=0
 config=4
+optional=1
 
 # PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_MISSES
+# Optional on s390x
 [event10:base-stat]
 fd=10
 type=0
 config=5
+optional=1
diff --git a/tools/perf/tests/attr/test-stat-detailed-1 b/tools/perf/tests/attr/test-stat-detailed-1
index b39270a..14ec910 100644
--- a/tools/perf/tests/attr/test-stat-detailed-1
+++ b/tools/perf/tests/attr/test-stat-detailed-1
@@ -55,49 +55,61 @@ type=0
 config=1
 
 # PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_INSTRUCTIONS
+# Optional on s390x
 [event9:base-stat]
 fd=9
 type=0
 config=4
+optional=1
 
 # PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_MISSES
+# Optional on s390x
 [event10:base-stat]
 fd=10
 type=0
 config=5
+optional=1
 
 # PERF_TYPE_HW_CACHE /
 #  PERF_COUNT_HW_CACHE_L1D                <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+# Optional on s390x
 [event11:base-stat]
 fd=11
 type=3
 config=0
+optional=1
 
 # PERF_TYPE_HW_CACHE /
 #  PERF_COUNT_HW_CACHE_L1D                <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+# Optional on s390x
 [event12:base-stat]
 fd=12
 type=3
 config=65536
+optional=1
 
 # PERF_TYPE_HW_CACHE /
 #  PERF_COUNT_HW_CACHE_LL                 <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+# Optional on s390x
 [event13:base-stat]
 fd=13
 type=3
 config=2
+optional=1
 
 # PERF_TYPE_HW_CACHE,
 #  PERF_COUNT_HW_CACHE_LL                 <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+# Optional on s390x
 [event14:base-stat]
 fd=14
 type=3
 config=65538
+optional=1
diff --git a/tools/perf/tests/attr/test-stat-detailed-2 b/tools/perf/tests/attr/test-stat-detailed-2
index 45f8e6e..763e2f0 100644
--- a/tools/perf/tests/attr/test-stat-detailed-2
+++ b/tools/perf/tests/attr/test-stat-detailed-2
@@ -55,52 +55,64 @@ type=0
 config=1
 
 # PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_INSTRUCTIONS
+# Optional on s390x
 [event9:base-stat]
 fd=9
 type=0
 config=4
+optional=1
 
 # PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_MISSES
+# Optional on s390x
 [event10:base-stat]
 fd=10
 type=0
 config=5
+optional=1
 
 # PERF_TYPE_HW_CACHE /
 #  PERF_COUNT_HW_CACHE_L1D                <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+# Optional on s390x
 [event11:base-stat]
 fd=11
 type=3
 config=0
+optional=1
 
 # PERF_TYPE_HW_CACHE /
 #  PERF_COUNT_HW_CACHE_L1D                <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+# Optional on s390x
 [event12:base-stat]
 fd=12
 type=3
 config=65536
+optional=1
 
 # PERF_TYPE_HW_CACHE /
 #  PERF_COUNT_HW_CACHE_LL                 <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+# Optional on s390x
 [event13:base-stat]
 fd=13
 type=3
 config=2
+optional=1
 
 # PERF_TYPE_HW_CACHE,
 #  PERF_COUNT_HW_CACHE_LL                 <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+# Optional on s390x
 [event14:base-stat]
 fd=14
 type=3
 config=65538
+optional=1
 
 # PERF_TYPE_HW_CACHE,
 #  PERF_COUNT_HW_CACHE_L1I                <<  0  |
@@ -116,43 +128,53 @@ optional=1
 #  PERF_COUNT_HW_CACHE_L1I                <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+# Optional on s390x
 [event16:base-stat]
 fd=16
 type=3
 config=65537
+optional=1
 
 # PERF_TYPE_HW_CACHE,
 #  PERF_COUNT_HW_CACHE_DTLB               <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+# Optional on s390x
 [event17:base-stat]
 fd=17
 type=3
 config=3
+optional=1
 
 # PERF_TYPE_HW_CACHE,
 #  PERF_COUNT_HW_CACHE_DTLB               <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+# Optional on s390x
 [event18:base-stat]
 fd=18
 type=3
 config=65539
+optional=1
 
 # PERF_TYPE_HW_CACHE,
 #  PERF_COUNT_HW_CACHE_ITLB               <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+# Optional on s390x
 [event19:base-stat]
 fd=19
 type=3
 config=4
+optional=1
 
 # PERF_TYPE_HW_CACHE,
 #  PERF_COUNT_HW_CACHE_ITLB               <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+# Optional on s390x
 [event20:base-stat]
 fd=20
 type=3
 config=65540
+optional=1
diff --git a/tools/perf/tests/attr/test-stat-detailed-3 b/tools/perf/tests/attr/test-stat-detailed-3
index 30ae0fb..5ed7ce6 100644
--- a/tools/perf/tests/attr/test-stat-detailed-3
+++ b/tools/perf/tests/attr/test-stat-detailed-3
@@ -55,52 +55,64 @@ type=0
 config=1
 
 # PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_INSTRUCTIONS
+# Optional on s390x
 [event9:base-stat]
 fd=9
 type=0
 config=4
+optional=1
 
 # PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_MISSES
+# Optional on s390x
 [event10:base-stat]
 fd=10
 type=0
 config=5
+optional=1
 
 # PERF_TYPE_HW_CACHE /
 #  PERF_COUNT_HW_CACHE_L1D                <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+# Optional on s390x
 [event11:base-stat]
 fd=11
 type=3
 config=0
+optional=1
 
 # PERF_TYPE_HW_CACHE /
 #  PERF_COUNT_HW_CACHE_L1D                <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+# Optional on s390x
 [event12:base-stat]
 fd=12
 type=3
 config=65536
+optional=1
 
 # PERF_TYPE_HW_CACHE /
 #  PERF_COUNT_HW_CACHE_LL                 <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+# Optional on s390x
 [event13:base-stat]
 fd=13
 type=3
 config=2
+optional=1
 
 # PERF_TYPE_HW_CACHE,
 #  PERF_COUNT_HW_CACHE_LL                 <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+# Optional on s390x
 [event14:base-stat]
 fd=14
 type=3
 config=65538
+optional=1
 
 # PERF_TYPE_HW_CACHE,
 #  PERF_COUNT_HW_CACHE_L1I                <<  0  |
@@ -116,46 +128,56 @@ optional=1
 #  PERF_COUNT_HW_CACHE_L1I                <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+# Optional on s390x
 [event16:base-stat]
 fd=16
 type=3
 config=65537
+optional=1
 
 # PERF_TYPE_HW_CACHE,
 #  PERF_COUNT_HW_CACHE_DTLB               <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+# Optional on s390x
 [event17:base-stat]
 fd=17
 type=3
 config=3
+optional=1
 
 # PERF_TYPE_HW_CACHE,
 #  PERF_COUNT_HW_CACHE_DTLB               <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+# Optional on s390x
 [event18:base-stat]
 fd=18
 type=3
 config=65539
+optional=1
 
 # PERF_TYPE_HW_CACHE,
 #  PERF_COUNT_HW_CACHE_ITLB               <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+# Optional on s390x
 [event19:base-stat]
 fd=19
 type=3
 config=4
+optional=1
 
 # PERF_TYPE_HW_CACHE,
 #  PERF_COUNT_HW_CACHE_ITLB               <<  0  |
 # (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
 # (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+# Optional on s390x
 [event20:base-stat]
 fd=20
 type=3
 config=65540
+optional=1
 
 # PERF_TYPE_HW_CACHE,
 #  PERF_COUNT_HW_CACHE_L1D                <<  0  |
-- 
2.9.3

[PATCH 5/9] perf test 15: disable test-record-group tests on s390x

[Thomas Richter]

On s390x the CPUMF supports sampling of only one hardware
event, namely cycles. There grouping of multiple hardware
events is not supported.
Disable these tests for s390x.

Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
---
 tools/perf/tests/attr/test-record-group          | 1 +
 tools/perf/tests/attr/test-record-group-sampling | 1 +
 tools/perf/tests/attr/test-record-group1         | 1 +
 3 files changed, 3 insertions(+)

diff --git a/tools/perf/tests/attr/test-record-group b/tools/perf/tests/attr/test-record-group
index 6e7961f..aa3885d 100644
--- a/tools/perf/tests/attr/test-record-group
+++ b/tools/perf/tests/attr/test-record-group
@@ -2,6 +2,7 @@
 command = record
 args    = --group -e cycles,instructions kill >/dev/null 2>&1
 ret     = 1
+arch	= !s390x
 
 [event-1:base-record]
 fd=1
diff --git a/tools/perf/tests/attr/test-record-group-sampling b/tools/perf/tests/attr/test-record-group-sampling
index ef59afd..332dc28 100644
--- a/tools/perf/tests/attr/test-record-group-sampling
+++ b/tools/perf/tests/attr/test-record-group-sampling
@@ -2,6 +2,7 @@
 command = record
 args    = -e '{cycles,cache-misses}:S' kill >/dev/null 2>&1
 ret     = 1
+arch	= !s390x
 
 [event-1:base-record]
 fd=1
diff --git a/tools/perf/tests/attr/test-record-group1 b/tools/perf/tests/attr/test-record-group1
index 87a222d..1f5359d 100644
--- a/tools/perf/tests/attr/test-record-group1
+++ b/tools/perf/tests/attr/test-record-group1
@@ -2,6 +2,7 @@
 command = record
 args    = -e '{cycles,instructions}' kill >/dev/null 2>&1
 ret     = 1
+arch	= !s390x
 
 [event-1:base-record]
 fd=1
-- 
2.9.3

[PATCH 6/9] perf test 15: disable test-record-filter tests on s390x

[Thomas Richter]

On s390x the CPUMF does not support sampling of branch
events.  Disable these tests for s390x.

Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
---
 tools/perf/tests/attr/test-record-branch-any             | 1 +
 tools/perf/tests/attr/test-record-branch-filter-any      | 1 +
 tools/perf/tests/attr/test-record-branch-filter-any_call | 1 +
 tools/perf/tests/attr/test-record-branch-filter-any_ret  | 1 +
 tools/perf/tests/attr/test-record-branch-filter-hv       | 1 +
 tools/perf/tests/attr/test-record-branch-filter-ind_call | 1 +
 tools/perf/tests/attr/test-record-branch-filter-k        | 1 +
 tools/perf/tests/attr/test-record-branch-filter-u        | 1 +
 8 files changed, 8 insertions(+)

diff --git a/tools/perf/tests/attr/test-record-branch-any b/tools/perf/tests/attr/test-record-branch-any
index 81f839e..348bd18 100644
--- a/tools/perf/tests/attr/test-record-branch-any
+++ b/tools/perf/tests/attr/test-record-branch-any
@@ -2,6 +2,7 @@
 command = record
 args    = -b kill >/dev/null 2>&1
 ret     = 1
+arch	= !s390x
 
 [event:base-record]
 sample_type=2311
diff --git a/tools/perf/tests/attr/test-record-branch-filter-any b/tools/perf/tests/attr/test-record-branch-filter-any
index 357421f..a99e743 100644
--- a/tools/perf/tests/attr/test-record-branch-filter-any
+++ b/tools/perf/tests/attr/test-record-branch-filter-any
@@ -2,6 +2,7 @@
 command = record
 args    = -j any kill >/dev/null 2>&1
 ret     = 1
+arch	= !s390x
 
 [event:base-record]
 sample_type=2311
diff --git a/tools/perf/tests/attr/test-record-branch-filter-any_call b/tools/perf/tests/attr/test-record-branch-filter-any_call
index dbc55f2..bb13ebe 100644
--- a/tools/perf/tests/attr/test-record-branch-filter-any_call
+++ b/tools/perf/tests/attr/test-record-branch-filter-any_call
@@ -2,6 +2,7 @@
 command = record
 args    = -j any_call kill >/dev/null 2>&1
 ret     = 1
+arch	= !s390x
 
 [event:base-record]
 sample_type=2311
diff --git a/tools/perf/tests/attr/test-record-branch-filter-any_ret b/tools/perf/tests/attr/test-record-branch-filter-any_ret
index a0824ff..9d21181 100644
--- a/tools/perf/tests/attr/test-record-branch-filter-any_ret
+++ b/tools/perf/tests/attr/test-record-branch-filter-any_ret
@@ -2,6 +2,7 @@
 command = record
 args    = -j any_ret kill >/dev/null 2>&1
 ret     = 1
+arch	= !s390x
 
 [event:base-record]
 sample_type=2311
diff --git a/tools/perf/tests/attr/test-record-branch-filter-hv b/tools/perf/tests/attr/test-record-branch-filter-hv
index f34d6f1..c1c412d 100644
--- a/tools/perf/tests/attr/test-record-branch-filter-hv
+++ b/tools/perf/tests/attr/test-record-branch-filter-hv
@@ -2,6 +2,7 @@
 command = record
 args    = -j hv kill >/dev/null 2>&1
 ret     = 1
+arch	= !s390x
 
 [event:base-record]
 sample_type=2311
diff --git a/tools/perf/tests/attr/test-record-branch-filter-ind_call b/tools/perf/tests/attr/test-record-branch-filter-ind_call
index b86a352..187fa31 100644
--- a/tools/perf/tests/attr/test-record-branch-filter-ind_call
+++ b/tools/perf/tests/attr/test-record-branch-filter-ind_call
@@ -2,6 +2,7 @@
 command = record
 args    = -j ind_call kill >/dev/null 2>&1
 ret     = 1
+arch	= !s390x
 
 [event:base-record]
 sample_type=2311
diff --git a/tools/perf/tests/attr/test-record-branch-filter-k b/tools/perf/tests/attr/test-record-branch-filter-k
index d3fbc5e..236a2b0 100644
--- a/tools/perf/tests/attr/test-record-branch-filter-k
+++ b/tools/perf/tests/attr/test-record-branch-filter-k
@@ -2,6 +2,7 @@
 command = record
 args    = -j k kill >/dev/null 2>&1
 ret     = 1
+arch	= !s390x
 
 [event:base-record]
 sample_type=2311
diff --git a/tools/perf/tests/attr/test-record-branch-filter-u b/tools/perf/tests/attr/test-record-branch-filter-u
index a318f0d..518c823 100644
--- a/tools/perf/tests/attr/test-record-branch-filter-u
+++ b/tools/perf/tests/attr/test-record-branch-filter-u
@@ -2,6 +2,7 @@
 command = record
 args    = -j u kill >/dev/null 2>&1
 ret     = 1
+arch	= !s390x
 
 [event:base-record]
 sample_type=2311
-- 
2.9.3

[PATCH 7/9] perf test 15: disable test-record-graph-dwarf test on s390x

[Thomas Richter]

There is currently no stack unwinding support for dwarf
in the perf s390x architecture dependend code.
Disable this test for s390x.

Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
---
 tools/perf/tests/attr/test-record-graph-dwarf | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tools/perf/tests/attr/test-record-graph-dwarf b/tools/perf/tests/attr/test-record-graph-dwarf
index da2fa73..c7a22d8 100644
--- a/tools/perf/tests/attr/test-record-graph-dwarf
+++ b/tools/perf/tests/attr/test-record-graph-dwarf
@@ -2,6 +2,7 @@
 command = record
 args    = --call-graph dwarf -- kill >/dev/null 2>&1
 ret     = 1
+arch	= !s390x
 
 [event:base-record]
 sample_type=45359
@@ -10,3 +11,4 @@ sample_stack_user=8192
 # TODO different for each arch, no support for that now
 sample_regs_user=*
 mmap_data=1
+task=1
-- 
2.9.3

[PATCH 8/9] perf: test 15 change period and frequency for s390x

[Thomas Richter]

The perf record command can specify sampling period and frequency.
For s390x both values are too slow and needs to be increased.
On s390x there is a minimum value and when the specified
command line value is too low, the perf_event_system call
fails.
I have successfully tested these new values on Intel.

Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
---
 tools/perf/tests/attr/test-record-count  | 4 ++--
 tools/perf/tests/attr/test-record-period | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/tools/perf/tests/attr/test-record-count b/tools/perf/tests/attr/test-record-count
index 34f6cc5..34788fe 100644
--- a/tools/perf/tests/attr/test-record-count
+++ b/tools/perf/tests/attr/test-record-count
@@ -1,9 +1,9 @@
 [config]
 command = record
-args    = -c 123 kill >/dev/null 2>&1
+args    = -c 100000 kill >/dev/null 2>&1
 ret     = 1
 
 [event:base-record]
-sample_period=123
+sample_period=100000
 sample_type=7
 freq=0
diff --git a/tools/perf/tests/attr/test-record-period b/tools/perf/tests/attr/test-record-period
index 69bc748..6b53250 100644
--- a/tools/perf/tests/attr/test-record-period
+++ b/tools/perf/tests/attr/test-record-period
@@ -1,8 +1,8 @@
 [config]
 command = record
-args    = -c 100 -P kill >/dev/null 2>&1
+args    = -c 100000 -P kill >/dev/null 2>&1
 ret     = 1
 
 [event:base-record]
-sample_period=100
+sample_period=100000
 freq=0
-- 
2.9.3

[PATCH 9/9] perf: test 15 fix expected task value.

[Thomas Richter]

Running perf test 15 (event attr test) the expected value
for task is 1. This can be seen with:

[root@oc2666213455 ~]# perf record -vv -b -- sleep 1
------------------------------------------------------------
perf_event_attr:
  size                             104
  { sample_period, sample_freq }   4000
  sample_type                      IP|TID|TIME|PERIOD|BRANCH_STACK
  disabled                         1
  inherit                          1
  mmap                             1
  comm                             1
  freq                             1
  enable_on_exec                   1
  task                             1
  sample_id_all                    1
  exclude_guest                    1
  mmap2                            1
  comm_exec                        1
------------------------------------------------------------
[...]

Change the expected task value from 0 to 1 in the base-record
file to match the correct results and report success for the
test-record-* test cases.

Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
---
 tools/perf/tests/attr/base-record | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/tests/attr/base-record b/tools/perf/tests/attr/base-record
index 31e0b1d..3794066 100644
--- a/tools/perf/tests/attr/base-record
+++ b/tools/perf/tests/attr/base-record
@@ -23,7 +23,7 @@ comm=1
 freq=1
 inherit_stat=0
 enable_on_exec=1
-task=0
+task=1
 watermark=0
 precise_ip=0|1|2|3
 mmap_data=0
-- 
2.9.3

Re: [PATCH 1/9] perf: test 15 fix python error on empty result

[Jiri Olsa]

On Wed, Aug 09, 2017 at 08:40:32AM +0200, Thomas Richter wrote:
> Commit d78ada4a767 ("perf tests attr: Do not store failed events")
> does not create an event file in the /tmp directory when the
> perf_open_event() system call failed.
> This can lead to a situation where not /tmp/event-xx-yy-zz result
> file exists at all (for example on a s390x virtual machine environment)
> where no CPUMF hardware is available.

hi,
what's this patchset based on? I got 2nd patch failing to apply:

Applying: perf: test 15 fix python error on empty result
Applying: perf test: perf test case 15 result ignored
error: patch failed: tools/perf/tests/attr.c:166
error: tools/perf/tests/attr.c: patch does not apply
Patch failed at 0002 perf test: perf test case 15 result ignored

any chance you have this on a git repo somewhere?

thanks,
jirka

Re: [PATCH 1/9] perf: test 15 fix python error on empty result

[Thomas-Mich Richter]

On 08/09/2017 12:39 PM, Jiri Olsa wrote:
> On Wed, Aug 09, 2017 at 08:40:32AM +0200, Thomas Richter wrote:
>> Commit d78ada4a767 ("perf tests attr: Do not store failed events")
>> does not create an event file in the /tmp directory when the
>> perf_open_event() system call failed.
>> This can lead to a situation where not /tmp/event-xx-yy-zz result
>> file exists at all (for example on a s390x virtual machine environment)
>> where no CPUMF hardware is available.
> 
> hi,
> what's this patchset based on? I got 2nd patch failing to apply:
> 
> Applying: perf: test 15 fix python error on empty result
> Applying: perf test: perf test case 15 result ignored
> error: patch failed: tools/perf/tests/attr.c:166
> error: tools/perf/tests/attr.c: patch does not apply
> Patch failed at 0002 perf test: perf test case 15 result ignored
> 
> any chance you have this on a git repo somewhere?
> 
> thanks,
> jirka


Hi,

I have cloned branch perf/core from

url = git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux

dated 27-July-2017 and last commit before my 9 patches is
6b7007a perf data: Add doc when no conversion support compiled

I have no git repository to clone from. But I can rebase the
patches on the latest level and can send them again.

Thanks.

-- 
Thomas Richter, Dept 3303, IBM LTC Boeblingen Germany
--
Vorsitzende des Aufsichtsrats: Martina Koederitz 
Geschäftsführung: Dirk Wittkopp
Sitz der Gesellschaft: Böblingen / Registergericht: Amtsgericht Stuttgart, HRB 243294

Re: [PATCH 9/9] perf: test 15 fix expected task value.

[Jiri Olsa]

On Wed, Aug 09, 2017 at 08:40:40AM +0200, Thomas Richter wrote:
> Running perf test 15 (event attr test) the expected value
> for task is 1. This can be seen with:

I just sent email to Arnaldo (you're in CC), fix for this was already posted

jirka

> 
> [root@oc2666213455 ~]# perf record -vv -b -- sleep 1
> ------------------------------------------------------------
> perf_event_attr:
>   size                             104
>   { sample_period, sample_freq }   4000
>   sample_type                      IP|TID|TIME|PERIOD|BRANCH_STACK
>   disabled                         1
>   inherit                          1
>   mmap                             1
>   comm                             1
>   freq                             1
>   enable_on_exec                   1
>   task                             1
>   sample_id_all                    1
>   exclude_guest                    1
>   mmap2                            1
>   comm_exec                        1
> ------------------------------------------------------------
> [...]
> 
> Change the expected task value from 0 to 1 in the base-record
> file to match the correct results and report success for the
> test-record-* test cases.
> 
> Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
> ---
>  tools/perf/tests/attr/base-record | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/tools/perf/tests/attr/base-record b/tools/perf/tests/attr/base-record
> index 31e0b1d..3794066 100644
> --- a/tools/perf/tests/attr/base-record
> +++ b/tools/perf/tests/attr/base-record
> @@ -23,7 +23,7 @@ comm=1
>  freq=1
>  inherit_stat=0
>  enable_on_exec=1
> -task=0
> +task=1
>  watermark=0
>  precise_ip=0|1|2|3
>  mmap_data=0
> -- 
> 2.9.3
> 

Re: [PATCH 3/9] perf: test 15 fix test-stat-group tests

[Jiri Olsa]

On Wed, Aug 09, 2017 at 08:40:34AM +0200, Thomas Richter wrote:
> Both test-stat-group and test-stat-group1 fail.
> The reason is an invalid check for the read_format
> return code.  The submitted value is 15 and this
> value should be returned by the perf_event_open() call
> (which it is).
> The comparison however is done against an invalid
> value of 3.
> Fix this by setting the expected read_format value.
> 
> Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
> ---
>  tools/perf/tests/attr/test-stat-group  | 2 ++
>  tools/perf/tests/attr/test-stat-group1 | 2 ++
>  2 files changed, 4 insertions(+)
> 
> diff --git a/tools/perf/tests/attr/test-stat-group b/tools/perf/tests/attr/test-stat-group
> index fdc1596..8fb3a50 100644
> --- a/tools/perf/tests/attr/test-stat-group
> +++ b/tools/perf/tests/attr/test-stat-group
> @@ -6,10 +6,12 @@ ret     = 1
>  [event-1:base-stat]
>  fd=1
>  group_fd=-1
> +read_format=15

I suppose this is on new kernel, which now support group read
it still passes on mine which doesn't supprot that ;-)

the fix needs to allow both values, so it works on older kernels, like:

read_format=3|15

thanks,
jirka

Re: [PATCH 4/9] perf: test 15 add s390x support for stat test cases

[Jiri Olsa]

On Wed, Aug 09, 2017 at 08:40:35AM +0200, Thomas Richter wrote:
> The perf test 15 test cases test-stat-detailed[123]
> and test-stat-default fail for s390x.
> The reason is missing support of the CPUMF facility
> for some of the test.
> 
> The tests are now marked optional. On s390x some of the
> individual tests still fail, but when marked as optional,
> do not cause the complete test to fail.

hum I wonder we should switch the optional keyword with that
arch support you introduced recently.. this way we 'disable'
those tests for other archs

jirka

Re: [PATCH 4/9] perf: test 15 add s390x support for stat test cases

[Thomas Richter]

On Wed, Aug 09, 2017 at 05:22:25PM +0200, Jiri Olsa wrote:
> On Wed, Aug 09, 2017 at 08:40:35AM +0200, Thomas Richter wrote:
> > The perf test 15 test cases test-stat-detailed[123]
> > and test-stat-default fail for s390x.
> > The reason is missing support of the CPUMF facility
> > for some of the test.
> > 
> > The tests are now marked optional. On s390x some of the
> > individual tests still fail, but when marked as optional,
> > do not cause the complete test to fail.
> 
> hum I wonder we should switch the optional keyword with that
> arch support you introduced recently.. this way we 'disable'
> those tests for other archs

This is correct.
Using the arch support requires several test case files.
So we would have file
  tools/perf/tests/attr/test-stat-default
with an arch = !s390x statement for all arch's that support it
and another fle
  tools/perf/tests/attr/test-stat-default-s390x
with an arch = s390x statement for 390x.

I am fine with this approach. If we all agree I can change the
patch and resend it.

Thanks Thomas