Add a shell script to check that the call-graphs generated using frame
pointers (--call-graph fp) are complete and not missing leaf functions:
| $ perf test 88 -v
| 88: Check Arm64 callgraphs are complete in fp mode :
| --- start ---
| test child forked, pid 8734
| + Compiling test program (/tmp/test_program.Cz3yL)...
| + Recording (PID=8749)...
| + Stopping perf-record...
| test_program.Cz
| 728 leaf
| 753 parent
| 76c main
| test child finished with 0
| ---- end ----
| Check Arm SPE callgraphs are complete in fp mode: Ok
Fixes: b9f6fbb3b2c2 ("perf arm64: Inject missing frames when using 'perf record --call-graph=fp'")
Suggested-by: Jiri Olsa <[email protected]>
Signed-off-by: German Gomez <[email protected]>
---
Changes since v1: https://lore.kernel.org/all/[email protected]/
- Add explicit '-g' flag to GCC command
---
.../perf/tests/shell/test_arm_callgraph_fp.sh | 68 +++++++++++++++++++
1 file changed, 68 insertions(+)
create mode 100755 tools/perf/tests/shell/test_arm_callgraph_fp.sh
diff --git a/tools/perf/tests/shell/test_arm_callgraph_fp.sh b/tools/perf/tests/shell/test_arm_callgraph_fp.sh
new file mode 100755
index 000000000..ea1b4e6bb
--- /dev/null
+++ b/tools/perf/tests/shell/test_arm_callgraph_fp.sh
@@ -0,0 +1,68 @@
+#!/bin/sh
+# Check Arm64 callgraphs are complete in fp mode
+# SPDX-License-Identifier: GPL-2.0
+
+lscpu | grep -q "aarch64" || exit 2
+
+if ! [ -x "$(command -v cc)" ]; then
+ echo "failed: no compiler, install gcc"
+ exit 2
+fi
+
+PERF_DATA=$(mktemp /tmp/__perf_test.perf.data.XXXXX)
+TEST_PROGRAM_SOURCE=$(mktemp /tmp/test_program.XXXXX.c)
+TEST_PROGRAM=$(mktemp /tmp/test_program.XXXXX)
+
+cleanup_files()
+{
+ rm -f $PERF_DATA
+ rm -f $TEST_PROGRAM_SOURCE
+ rm -f $TEST_PROGRAM
+}
+
+trap cleanup_files exit term int
+
+cat << EOF > $TEST_PROGRAM_SOURCE
+int a = 0;
+void leaf(void) {
+ for (;;)
+ a += a;
+}
+void parent(void) {
+ leaf();
+}
+int main(void) {
+ parent();
+ return 0;
+}
+EOF
+
+echo " + Compiling test program ($TEST_PROGRAM)..."
+
+CFLAGS="-g -O0 -fno-inline -fno-omit-frame-pointer"
+cc $CFLAGS $TEST_PROGRAM_SOURCE -o $TEST_PROGRAM || exit 1
+
+# Add a 1 second delay to skip samples that are not in the leaf() function
+perf record -o $PERF_DATA --call-graph fp -e cycles//u -D 1000 -- $TEST_PROGRAM 2> /dev/null &
+PID=$!
+
+echo " + Recording (PID=$PID)..."
+sleep 2
+echo " + Stopping perf-record..."
+
+kill $PID
+wait $PID
+
+# expected perf-script output:
+#
+# program
+# 728 leaf
+# 753 parent
+# 76c main
+# ...
+
+perf script -i $PERF_DATA -F comm,ip,sym | head -n4
+perf script -i $PERF_DATA -F comm,ip,sym | head -n4 | \
+ awk '{ if ($2 != "") sym[i++] = $2 } END { if (sym[0] != "leaf" ||
+ sym[1] != "parent" ||
+ sym[2] != "main") exit 1 }'
--
2.25.1
On 16/03/2022 17:20, German Gomez wrote:
> Add a shell script to check that the call-graphs generated using frame
> pointers (--call-graph fp) are complete and not missing leaf functions:
>
> | $ perf test 88 -v
> | 88: Check Arm64 callgraphs are complete in fp mode :
> | --- start ---
> | test child forked, pid 8734
> | + Compiling test program (/tmp/test_program.Cz3yL)...
> | + Recording (PID=8749)...
> | + Stopping perf-record...
> | test_program.Cz
> | 728 leaf
> | 753 parent
> | 76c main
> | test child finished with 0
> | ---- end ----
> | Check Arm SPE callgraphs are complete in fp mode: Ok
>
Ran it on N1SDP and it passes, and it fails if b9f6fbb3b2c2 isn't applied.
Reviewed-by: James Clark <[email protected]>
> Fixes: b9f6fbb3b2c2 ("perf arm64: Inject missing frames when using 'perf record --call-graph=fp'")
> Suggested-by: Jiri Olsa <[email protected]>
> Signed-off-by: German Gomez <[email protected]>
> ---
> Changes since v1: https://lore.kernel.org/all/[email protected]/
> - Add explicit '-g' flag to GCC command
> ---
> .../perf/tests/shell/test_arm_callgraph_fp.sh | 68 +++++++++++++++++++
> 1 file changed, 68 insertions(+)
> create mode 100755 tools/perf/tests/shell/test_arm_callgraph_fp.sh
>
> diff --git a/tools/perf/tests/shell/test_arm_callgraph_fp.sh b/tools/perf/tests/shell/test_arm_callgraph_fp.sh
> new file mode 100755
> index 000000000..ea1b4e6bb
> --- /dev/null
> +++ b/tools/perf/tests/shell/test_arm_callgraph_fp.sh
> @@ -0,0 +1,68 @@
> +#!/bin/sh
> +# Check Arm64 callgraphs are complete in fp mode
> +# SPDX-License-Identifier: GPL-2.0
> +
> +lscpu | grep -q "aarch64" || exit 2
> +
> +if ! [ -x "$(command -v cc)" ]; then
> + echo "failed: no compiler, install gcc"
> + exit 2
> +fi
> +
> +PERF_DATA=$(mktemp /tmp/__perf_test.perf.data.XXXXX)
> +TEST_PROGRAM_SOURCE=$(mktemp /tmp/test_program.XXXXX.c)
> +TEST_PROGRAM=$(mktemp /tmp/test_program.XXXXX)
> +
> +cleanup_files()
> +{
> + rm -f $PERF_DATA
> + rm -f $TEST_PROGRAM_SOURCE
> + rm -f $TEST_PROGRAM
> +}
> +
> +trap cleanup_files exit term int
> +
> +cat << EOF > $TEST_PROGRAM_SOURCE
> +int a = 0;
> +void leaf(void) {
> + for (;;)
> + a += a;
> +}
> +void parent(void) {
> + leaf();
> +}
> +int main(void) {
> + parent();
> + return 0;
> +}
> +EOF
> +
> +echo " + Compiling test program ($TEST_PROGRAM)..."
> +
> +CFLAGS="-g -O0 -fno-inline -fno-omit-frame-pointer"
> +cc $CFLAGS $TEST_PROGRAM_SOURCE -o $TEST_PROGRAM || exit 1
> +
> +# Add a 1 second delay to skip samples that are not in the leaf() function
> +perf record -o $PERF_DATA --call-graph fp -e cycles//u -D 1000 -- $TEST_PROGRAM 2> /dev/null &
> +PID=$!
> +
> +echo " + Recording (PID=$PID)..."
> +sleep 2
> +echo " + Stopping perf-record..."
> +
> +kill $PID
> +wait $PID
> +
> +# expected perf-script output:
> +#
> +# program
> +# 728 leaf
> +# 753 parent
> +# 76c main
> +# ...
> +
> +perf script -i $PERF_DATA -F comm,ip,sym | head -n4
> +perf script -i $PERF_DATA -F comm,ip,sym | head -n4 | \
> + awk '{ if ($2 != "") sym[i++] = $2 } END { if (sym[0] != "leaf" ||
> + sym[1] != "parent" ||
> + sym[2] != "main") exit 1 }'
On 16/03/2022 17:30, James Clark wrote:
>
> On 16/03/2022 17:20, German Gomez wrote:
>> Add a shell script to check that the call-graphs generated using frame
>> pointers (--call-graph fp) are complete and not missing leaf functions:
>>
>> | $ perf test 88 -v
>> | 88: Check Arm64 callgraphs are complete in fp mode :
>> | --- start ---
>> | test child forked, pid 8734
>> | + Compiling test program (/tmp/test_program.Cz3yL)...
>> | + Recording (PID=8749)...
>> | + Stopping perf-record...
>> | test_program.Cz
>> | 728 leaf
>> | 753 parent
>> | 76c main
>> | test child finished with 0
>> | ---- end ----
>> | Check Arm SPE callgraphs are complete in fp mode: Ok
>>
> Ran it on N1SDP and it passes, and it fails if b9f6fbb3b2c2 isn't applied.
>
> Reviewed-by: James Clark <[email protected]>
Thanks!
On 16/03/2022 17:30, James Clark wrote:
>
> On 16/03/2022 17:20, German Gomez wrote:
>> Add a shell script to check that the call-graphs generated using frame
>> pointers (--call-graph fp) are complete and not missing leaf functions:
>>
>> | $ perf test 88 -v
>> | 88: Check Arm64 callgraphs are complete in fp mode :
>> | --- start ---
>> | test child forked, pid 8734
>> | + Compiling test program (/tmp/test_program.Cz3yL)...
>> | + Recording (PID=8749)...
>> | + Stopping perf-record...
>> | test_program.Cz
>> | 728 leaf
>> | 753 parent
>> | 76c main
>> | test child finished with 0
>> | ---- end ----
>> | Check Arm SPE callgraphs are complete in fp mode: Ok
>>
> Ran it on N1SDP and it passes, and it fails if b9f6fbb3b2c2 isn't applied.
I forgot to mention in the notes that it's supposed to work with both unwinders:
$ make # for libunwind (default)
$ make NO_LIBUNWIND=1 # for libdw
>
> Reviewed-by: James Clark <[email protected]>
>
>> Fixes: b9f6fbb3b2c2 ("perf arm64: Inject missing frames when using 'perf record --call-graph=fp'")
>> Suggested-by: Jiri Olsa <[email protected]>
>> Signed-off-by: German Gomez <[email protected]>
>> ---
>> Changes since v1: https://lore.kernel.org/all/[email protected]/
>> - Add explicit '-g' flag to GCC command
>> ---
>> .../perf/tests/shell/test_arm_callgraph_fp.sh | 68 +++++++++++++++++++
>> 1 file changed, 68 insertions(+)
>> create mode 100755 tools/perf/tests/shell/test_arm_callgraph_fp.sh
>>
>> diff --git a/tools/perf/tests/shell/test_arm_callgraph_fp.sh b/tools/perf/tests/shell/test_arm_callgraph_fp.sh
>> new file mode 100755
>> index 000000000..ea1b4e6bb
>> --- /dev/null
>> +++ b/tools/perf/tests/shell/test_arm_callgraph_fp.sh
>> @@ -0,0 +1,68 @@
>> +#!/bin/sh
>> +# Check Arm64 callgraphs are complete in fp mode
>> +# SPDX-License-Identifier: GPL-2.0
>> +
>> +lscpu | grep -q "aarch64" || exit 2
>> +
>> +if ! [ -x "$(command -v cc)" ]; then
>> + echo "failed: no compiler, install gcc"
>> + exit 2
>> +fi
>> +
>> +PERF_DATA=$(mktemp /tmp/__perf_test.perf.data.XXXXX)
>> +TEST_PROGRAM_SOURCE=$(mktemp /tmp/test_program.XXXXX.c)
>> +TEST_PROGRAM=$(mktemp /tmp/test_program.XXXXX)
>> +
>> +cleanup_files()
>> +{
>> + rm -f $PERF_DATA
>> + rm -f $TEST_PROGRAM_SOURCE
>> + rm -f $TEST_PROGRAM
>> +}
>> +
>> +trap cleanup_files exit term int
>> +
>> +cat << EOF > $TEST_PROGRAM_SOURCE
>> +int a = 0;
>> +void leaf(void) {
>> + for (;;)
>> + a += a;
>> +}
>> +void parent(void) {
>> + leaf();
>> +}
>> +int main(void) {
>> + parent();
>> + return 0;
>> +}
>> +EOF
>> +
>> +echo " + Compiling test program ($TEST_PROGRAM)..."
>> +
>> +CFLAGS="-g -O0 -fno-inline -fno-omit-frame-pointer"
>> +cc $CFLAGS $TEST_PROGRAM_SOURCE -o $TEST_PROGRAM || exit 1
>> +
>> +# Add a 1 second delay to skip samples that are not in the leaf() function
>> +perf record -o $PERF_DATA --call-graph fp -e cycles//u -D 1000 -- $TEST_PROGRAM 2> /dev/null &
>> +PID=$!
>> +
>> +echo " + Recording (PID=$PID)..."
>> +sleep 2
>> +echo " + Stopping perf-record..."
>> +
>> +kill $PID
>> +wait $PID
>> +
>> +# expected perf-script output:
>> +#
>> +# program
>> +# 728 leaf
>> +# 753 parent
>> +# 76c main
>> +# ...
>> +
>> +perf script -i $PERF_DATA -F comm,ip,sym | head -n4
>> +perf script -i $PERF_DATA -F comm,ip,sym | head -n4 | \
>> + awk '{ if ($2 != "") sym[i++] = $2 } END { if (sym[0] != "leaf" ||
>> + sym[1] != "parent" ||
>> + sym[2] != "main") exit 1 }'
Em Wed, Mar 16, 2022 at 05:48:50PM +0000, German Gomez escreveu:
>
> On 16/03/2022 17:30, James Clark wrote:
> >
> > On 16/03/2022 17:20, German Gomez wrote:
> >> Add a shell script to check that the call-graphs generated using frame
> >> pointers (--call-graph fp) are complete and not missing leaf functions:
> >>
> >> | $ perf test 88 -v
> >> | 88: Check Arm64 callgraphs are complete in fp mode :
> >> | --- start ---
> >> | test child forked, pid 8734
> >> | + Compiling test program (/tmp/test_program.Cz3yL)...
> >> | + Recording (PID=8749)...
> >> | + Stopping perf-record...
> >> | test_program.Cz
> >> | 728 leaf
> >> | 753 parent
> >> | 76c main
> >> | test child finished with 0
> >> | ---- end ----
> >> | Check Arm SPE callgraphs are complete in fp mode: Ok
> >>
> > Ran it on N1SDP and it passes, and it fails if b9f6fbb3b2c2 isn't applied.
>
> I forgot to mention in the notes that it's supposed to work with both unwinders:
>
> $ make # for libunwind (default)
> $ make NO_LIBUNWIND=1 # for libdw
I'll add it to the cset commit log, and also will fix this:
⬢[acme@toolbox perf]$ git am ./v2_20220316_german_gomez_perf_test_arm64_test_unwinding_using_fame_pointer_fp_mode.mbx
Applying: perf test arm64: Test unwinding using fame-pointer (fp) mode
.git/rebase-apply/patch:72: trailing whitespace.
# program
warning: 1 line adds whitespace errors.
tools/perf/tests/shell/test_arm_callgraph_fp.sh:58: trailing whitespace.
+# program
⬢[acme@toolbox perf]$
> >
> > Reviewed-by: James Clark <[email protected]>
> >
> >> Fixes: b9f6fbb3b2c2 ("perf arm64: Inject missing frames when using 'perf record --call-graph=fp'")
> >> Suggested-by: Jiri Olsa <[email protected]>
> >> Signed-off-by: German Gomez <[email protected]>
> >> ---
> >> Changes since v1: https://lore.kernel.org/all/[email protected]/
> >> - Add explicit '-g' flag to GCC command
> >> ---
> >> .../perf/tests/shell/test_arm_callgraph_fp.sh | 68 +++++++++++++++++++
> >> 1 file changed, 68 insertions(+)
> >> create mode 100755 tools/perf/tests/shell/test_arm_callgraph_fp.sh
> >>
> >> diff --git a/tools/perf/tests/shell/test_arm_callgraph_fp.sh b/tools/perf/tests/shell/test_arm_callgraph_fp.sh
> >> new file mode 100755
> >> index 000000000..ea1b4e6bb
> >> --- /dev/null
> >> +++ b/tools/perf/tests/shell/test_arm_callgraph_fp.sh
> >> @@ -0,0 +1,68 @@
> >> +#!/bin/sh
> >> +# Check Arm64 callgraphs are complete in fp mode
> >> +# SPDX-License-Identifier: GPL-2.0
> >> +
> >> +lscpu | grep -q "aarch64" || exit 2
> >> +
> >> +if ! [ -x "$(command -v cc)" ]; then
> >> + echo "failed: no compiler, install gcc"
> >> + exit 2
> >> +fi
> >> +
> >> +PERF_DATA=$(mktemp /tmp/__perf_test.perf.data.XXXXX)
> >> +TEST_PROGRAM_SOURCE=$(mktemp /tmp/test_program.XXXXX.c)
> >> +TEST_PROGRAM=$(mktemp /tmp/test_program.XXXXX)
> >> +
> >> +cleanup_files()
> >> +{
> >> + rm -f $PERF_DATA
> >> + rm -f $TEST_PROGRAM_SOURCE
> >> + rm -f $TEST_PROGRAM
> >> +}
> >> +
> >> +trap cleanup_files exit term int
> >> +
> >> +cat << EOF > $TEST_PROGRAM_SOURCE
> >> +int a = 0;
> >> +void leaf(void) {
> >> + for (;;)
> >> + a += a;
> >> +}
> >> +void parent(void) {
> >> + leaf();
> >> +}
> >> +int main(void) {
> >> + parent();
> >> + return 0;
> >> +}
> >> +EOF
> >> +
> >> +echo " + Compiling test program ($TEST_PROGRAM)..."
> >> +
> >> +CFLAGS="-g -O0 -fno-inline -fno-omit-frame-pointer"
> >> +cc $CFLAGS $TEST_PROGRAM_SOURCE -o $TEST_PROGRAM || exit 1
> >> +
> >> +# Add a 1 second delay to skip samples that are not in the leaf() function
> >> +perf record -o $PERF_DATA --call-graph fp -e cycles//u -D 1000 -- $TEST_PROGRAM 2> /dev/null &
> >> +PID=$!
> >> +
> >> +echo " + Recording (PID=$PID)..."
> >> +sleep 2
> >> +echo " + Stopping perf-record..."
> >> +
> >> +kill $PID
> >> +wait $PID
> >> +
> >> +# expected perf-script output:
> >> +#
> >> +# program
> >> +# 728 leaf
> >> +# 753 parent
> >> +# 76c main
> >> +# ...
> >> +
> >> +perf script -i $PERF_DATA -F comm,ip,sym | head -n4
> >> +perf script -i $PERF_DATA -F comm,ip,sym | head -n4 | \
> >> + awk '{ if ($2 != "") sym[i++] = $2 } END { if (sym[0] != "leaf" ||
> >> + sym[1] != "parent" ||
> >> + sym[2] != "main") exit 1 }'
--
- Arnaldo
On 21/03/2022 20:52, Arnaldo Carvalho de Melo wrote:
> Em Wed, Mar 16, 2022 at 05:48:50PM +0000, German Gomez escreveu:
>> On 16/03/2022 17:30, James Clark wrote:
>>> On 16/03/2022 17:20, German Gomez wrote:
>>>> Add a shell script to check that the call-graphs generated using frame
>>>> pointers (--call-graph fp) are complete and not missing leaf functions:
>>>>
>>>> | $ perf test 88 -v
>>>> | 88: Check Arm64 callgraphs are complete in fp mode :
>>>> | --- start ---
>>>> | test child forked, pid 8734
>>>> | + Compiling test program (/tmp/test_program.Cz3yL)...
>>>> | + Recording (PID=8749)...
>>>> | + Stopping perf-record...
>>>> | test_program.Cz
>>>> | 728 leaf
>>>> | 753 parent
>>>> | 76c main
>>>> | test child finished with 0
>>>> | ---- end ----
>>>> | Check Arm SPE callgraphs are complete in fp mode: Ok
>>>>
>>> Ran it on N1SDP and it passes, and it fails if b9f6fbb3b2c2 isn't applied.
>> I forgot to mention in the notes that it's supposed to work with both unwinders:
>>
>> $ make # for libunwind (default)
>> $ make NO_LIBUNWIND=1 # for libdw
> I'll add it to the cset commit log, and also will fix this:
>
> ⬢[acme@toolbox perf]$ git am ./v2_20220316_german_gomez_perf_test_arm64_test_unwinding_using_fame_pointer_fp_mode.mbx
> Applying: perf test arm64: Test unwinding using fame-pointer (fp) mode
> .git/rebase-apply/patch:72: trailing whitespace.
> # program
> warning: 1 line adds whitespace errors.
> tools/perf/tests/shell/test_arm_callgraph_fp.sh:58: trailing whitespace.
> +# program
> ⬢[acme@toolbox perf]$
>
Thanks Arnaldo, I forgot to run the checkpatch script.
>>> Reviewed-by: James Clark <[email protected]>
>>>
>>>> Fixes: b9f6fbb3b2c2 ("perf arm64: Inject missing frames when using 'perf record --call-graph=fp'")
>>>> Suggested-by: Jiri Olsa <[email protected]>
>>>> Signed-off-by: German Gomez <[email protected]>