diff mbox series

[4/5] perf test: Add test case for record tracking

Message ID 20230704074217.240939-5-yangjihong1@huawei.com
State New
Headers
Series perf record: Tracking side-band events for all CPUs when tracing selected CPUs |

Commit Message

Yang Jihong July 4, 2023, 7:42 a.m. UTC 
  Add a new test case to record tracking side-band events for all CPUs when
tracing selected CPUs

Test result:

  # ./perf test list 2>&1 | grep 'perf record tracking tests'
   95: perf record tracking tests
  f# ./perf test 95
   95: perf record tracking tests                                      : Ok

Signed-off-by: Yang Jihong <yangjihong1@huawei.com>
---
 tools/perf/tests/shell/record_tracking.sh | 44 +++++++++++++++++++++++
 1 file changed, 44 insertions(+)
 create mode 100755 tools/perf/tests/shell/record_tracking.sh

Comments

Ian Rogers July 12, 2023, 8:48 p.m. UTC | #1 
On Tue, Jul 4, 2023 at 12:44 AM Yang Jihong <yangjihong1@huawei.com> wrote:
>
> Add a new test case to record tracking side-band events for all CPUs when
> tracing selected CPUs

We're using "tracking" and "sideband" as synonyms, I think it would be
clearer to just use "sideband" and not use "tracking". I see both
terms in tools/perf/Documentation/perf-intel-pt.txt

Would it be possible to get some consistency here?

Thanks,
Ian

> Test result:
>
>   # ./perf test list 2>&1 | grep 'perf record tracking tests'
>    95: perf record tracking tests
>   f# ./perf test 95
>    95: perf record tracking tests                                      : Ok
>
> Signed-off-by: Yang Jihong <yangjihong1@huawei.com>
> ---
>  tools/perf/tests/shell/record_tracking.sh | 44 +++++++++++++++++++++++
>  1 file changed, 44 insertions(+)
>  create mode 100755 tools/perf/tests/shell/record_tracking.sh
>
> diff --git a/tools/perf/tests/shell/record_tracking.sh b/tools/perf/tests/shell/record_tracking.sh
> new file mode 100755
> index 000000000000..fe05f4772999
> --- /dev/null
> +++ b/tools/perf/tests/shell/record_tracking.sh
> @@ -0,0 +1,44 @@
> +#!/bin/sh
> +# perf record tracking tests
> +# SPDX-License-Identifier: GPL-2.0
> +
> +set -e
> +
> +err=0
> +perfdata=$(mktemp /tmp/__perf_test.perf.data.XXXXX)
> +
> +can_cpu_wide()
> +{
> +    if ! perf record -o ${perfdata} -BN --no-bpf-event -e dummy:u -C $1 true 2>&1 >/dev/null
> +    then
> +       echo "record tracking test [Skipped cannot record cpu$1]"
> +       err=2
> +    fi
> +
> +    rm -f ${perfdata}
> +    return $err
> +}
> +
> +test_system_wide_tracking()
> +{
> +    # Need CPU 0 and CPU 1
> +    can_cpu_wide 0 || return 0
> +    can_cpu_wide 1 || return 0
> +
> +    # Record on CPU 0 a task running on CPU 1
> +    perf record -BN --no-bpf-event -o ${perfdata} -e dummy:u -C 0 -- taskset --cpu-list 1 true
> +
> +    # Should get MMAP events from CPU 1
> +    mmap_cnt=`perf script -i ${perfdata} --show-mmap-events -C 1 2>/dev/null | grep MMAP | wc -l`
> +
> +    rm -f ${perfdata}
> +
> +    if [ ${mmap_cnt} -gt 0 ] ; then
> +       return 0
> +    fi
> +
> +    echo "Failed to record MMAP events on CPU 1 when tracing CPU 0"
> +    return 1
> +}
> +
> +test_system_wide_tracking
> --
> 2.30.GIT
>
Yang Jihong July 13, 2023, 6:59 a.m. UTC | #2 
Hello,

On 2023/7/13 4:48, Ian Rogers wrote:
> On Tue, Jul 4, 2023 at 12:44 AM Yang Jihong <yangjihong1@huawei.com> wrote:
>>
>> Add a new test case to record tracking side-band events for all CPUs when
>> tracing selected CPUs
> 
> We're using "tracking" and "sideband" as synonyms, I think it would be
> clearer to just use "sideband" and not use "tracking". I see both
> terms in tools/perf/Documentation/perf-intel-pt.txt
> 
Okay, will use "sideband events" instead of "tracking events" in the 
next version.

Thanks,
Yang
diff mbox series

Patch

diff --git a/tools/perf/tests/shell/record_tracking.sh b/tools/perf/tests/shell/record_tracking.sh
new file mode 100755
index 000000000000..fe05f4772999
--- /dev/null
+++ b/tools/perf/tests/shell/record_tracking.sh
@@ -0,0 +1,44 @@ 
+#!/bin/sh
+# perf record tracking tests
+# SPDX-License-Identifier: GPL-2.0
+
+set -e
+
+err=0
+perfdata=$(mktemp /tmp/__perf_test.perf.data.XXXXX)
+
+can_cpu_wide()
+{
+    if ! perf record -o ${perfdata} -BN --no-bpf-event -e dummy:u -C $1 true 2>&1 >/dev/null
+    then
+	echo "record tracking test [Skipped cannot record cpu$1]"
+	err=2
+    fi
+
+    rm -f ${perfdata}
+    return $err
+}
+
+test_system_wide_tracking()
+{
+    # Need CPU 0 and CPU 1
+    can_cpu_wide 0 || return 0
+    can_cpu_wide 1 || return 0
+
+    # Record on CPU 0 a task running on CPU 1
+    perf record -BN --no-bpf-event -o ${perfdata} -e dummy:u -C 0 -- taskset --cpu-list 1 true
+
+    # Should get MMAP events from CPU 1
+    mmap_cnt=`perf script -i ${perfdata} --show-mmap-events -C 1 2>/dev/null | grep MMAP | wc -l`
+
+    rm -f ${perfdata}
+
+    if [ ${mmap_cnt} -gt 0 ] ; then
+	return 0
+    fi
+
+    echo "Failed to record MMAP events on CPU 1 when tracing CPU 0"
+    return 1
+}
+
+test_system_wide_tracking