tool/perf/test: Fix shellcheck coding/formatting issues of test shell scripts
cover tool/perf/test: Fix shellcheck coding/formatting issues of test shell scripts
Patchset covers a set of fixes for coding/formatting issues observed while running shellcheck tool on the perf test shell scripts. Shellcheck is a static analysis tool that can find semantic/syntax bugs in the shell scripts.
Patches 1-14 fixes the issues found with shellcheck. Patch 15, 16 and patch 17 address a fix in task_analyzer test.
This cleanup is a pre-requisite to include a build option for shellcheck discussed here: https://www.spinics.net/lists/linux-perf-users/msg25553.html
Also this is first set of patches. There will be one more set which will include build option for shellcheck as discussed in the mail thread.
Abhirup Deb (2):
tools/perf/tests: fix test_arm_spe.sh signal case issues perf/tests/shell: fix shellscript errors for lock_contention.sh
Aboorva Devarajan (1):
tools/perf/tests: Fix shellcheck issues in test_task_analyzer.sh file
Aditya Gupta (3):
perf tests task_analyzer: fix bad substitution ${$1} perf tests task_analyzer: print command on failure perf tests task_analyzer: skip tests if no libtraceevent support
Akanksha J N (1):
tools/perf/tests: Fix shellcheck warnings for
trace+probe_vfs_getname.sh
Anushree Mathur (1):
perf/tests/shell : Shellcheck fixes for perf test
"test_arm_coresight.sh"
Barnali Guha Thakurata (1):
tools/perf/tests/shell/stat_all_metrics: Fix shellcheck warning SC2076
in stat_all_metrics.sh
Disha Goel (1):
tools/perf/tests: fix shellcheck warning for stat+json_output
Geetika (1):
tools/perf/tests: Fix all POSIX sh warnings in perf shell test
test_brstack.sh
Korrapati Likhitha (1):
tools/perf/tests: Fix shellcheck warnings for stat+csv_output
Samir Mulani (1):
tools/perf/tests: fixed shellcheck warnings for perf shell scripts
Shirisha G (1):
tools/perf/tests: fix shellcheck warnings for daemon.sh
Sourabh Jain (1):
perf: get rid of unused import
Spoorthy S (2):
shellcheck : fixing signal names and adding double quotes for
expression in test_arm_callgraph_fp
tools/perf/tests: Fix all POSIX sh warnings in stat+shadow_stat.sh
.../scripts/python/arm-cs-trace-disasm.py | 1 - tools/perf/tests/shell/buildid.sh | 12 +- tools/perf/tests/shell/daemon.sh | 113 ++++++++++++------ tools/perf/tests/shell/lock_contention.sh | 70 +++++------ .../shell/record+probe_libc_inet_pton.sh | 6 +- .../shell/record+script_probe_vfs_getname.sh | 4 +- tools/perf/tests/shell/stat+csv_output.sh | 4 +- tools/perf/tests/shell/stat+json_output.sh | 2 +- tools/perf/tests/shell/stat+shadow_stat.sh | 4 +- tools/perf/tests/shell/stat_all_metrics.sh | 6 +- .../perf/tests/shell/test_arm_callgraph_fp.sh | 6 +- tools/perf/tests/shell/test_arm_coresight.sh | 6 +- tools/perf/tests/shell/test_arm_spe.sh | 2 +- tools/perf/tests/shell/test_brstack.sh | 12 +- tools/perf/tests/shell/test_task_analyzer.sh | 98 ++++++++------- .../tests/shell/trace+probe_vfs_getname.sh | 6 +- 16 files changed, 203 insertions(+), 149 deletions(-)
Signature · 1 line
2.39.1
PATCH 1/17 perf: get rid of unused import
From: Sourabh Jain <[email protected]>
Script doesn't use sys library, so remove it.
Report by pylint:
W0611: Unused import sys (unused-import)
--- tools/perf/scripts/python/arm-cs-trace-disasm.py | 1 - 1 file changed, 1 deletion(-)
tools/perf/scripts/python/arm-cs-trace-disasm.py +0 -1
@@ -9,7 +9,6 @@ from __future__ import print_function import os from os import path -import sys import re from subprocess import * from optparse import OptionParser, make_option
Signature · 1 line
2.39.1
Re: [PATCH 01/17] perf: get rid of unused import
Em Tue, Jun 13, 2023 at 10:11:29PM +0530, Athira Rajeev escreveu:
Quoted 3 lines
From: Sourabh Jain <[email protected]>
Script doesn't use sys library, so remove it.
Please Cc the persons working on that file, I added Leo to the CC list of this message.
Thanks, applied.
- Arnaldo
Quoted 25 lines
Report by pylint:
W0611: Unused import sys (unused-import)Signed-off-by Athira Rajeev <[email protected]>Signed-off-by Kajol Jain <[email protected]>Signed-off-by Sourabh Jain <[email protected]>--- tools/perf/scripts/python/arm-cs-trace-disasm.py | 1 - 1 file changed, 1 deletion(-)tools/perf/scripts/python/arm-cs-trace-disasm.py +0
-1@@ -9,7 +9,6 @@ from __future__ import print_function import os from os import path -import sys import re from subprocess import * from optparse import OptionParser, make_optionSignature · 1 line
2.39.1
Signature · 1 line
- Arnaldo
Re: [PATCH 01/17] perf: get rid of unused import
Quoted 7 lines · Arnaldo Carvalho de Melo
Em Tue, Jun 13, 2023 at 10:11:29PM +0530, Athira Rajeev escreveu:
Quoted 3 lines
From: Sourabh Jain <[email protected]>
Script doesn't use sys library, so remove it.
Please Cc the persons working on that file, I added Leo to the CC list of this message.
Thanks, Arnaldo & Athira. The change looks good to me.
Quoted 1 line
Thanks, applied.
Since have applied this patch, it's no need to give my review tag :)
Thanks,
Leo
Quoted 31 lines
- Arnaldo
Quoted 25 lines
Report by pylint:
W0611: Unused import sys (unused-import)Signed-off-by Athira Rajeev <[email protected]>Signed-off-by Kajol Jain <[email protected]>Signed-off-by Sourabh Jain <[email protected]>--- tools/perf/scripts/python/arm-cs-trace-disasm.py | 1 - 1 file changed, 1 deletion(-)tools/perf/scripts/python/arm-cs-trace-disasm.py +0
-1@@ -9,7 +9,6 @@ from __future__ import print_function import os from os import path -import sys import re from subprocess import * from optparse import OptionParser, make_optionSignature · 1 line
2.39.1Signature · 1 line
- Arnaldo
Arnaldo Carvalho de Melo
maintainerRe: [PATCH 01/17] perf: get rid of unused import
Em Wed, Jun 14, 2023 at 09:59:14AM +0800, Leo Yan escreveu:
Quoted 6 lines · Arnaldo Carvalho de Melo
Quoted 5 lines
Em Tue, Jun 13, 2023 at 10:11:29PM +0530, Athira Rajeev escreveu:
Quoted 3 lines
From: Sourabh Jain <[email protected]>
Script doesn't use sys library, so remove it.
Quoted 8 lines
Quoted 2 lines
Please Cc the persons working on that file, I added Leo to the CC list of this message.
Thanks, Arnaldo & Athira. The change looks good to me.
Quoted 1 line
Thanks, applied.
Since have applied this patch, it's no need to give my review tag :)
No, I usually can add a Reviewed-by tag even after having applied it to my local tree, as I still need to run tests before making it available via perf-tools-next, when I should not make any further changes.
- Arnaldo
Quoted 34 lines
Thanks,
LeoQuoted 31 lines
- Arnaldo
Quoted 25 lines
Report by pylint:
W0611: Unused import sys (unused-import)Signed-off-by Athira Rajeev <[email protected]>Signed-off-by Kajol Jain <[email protected]>Signed-off-by Sourabh Jain <[email protected]>--- tools/perf/scripts/python/arm-cs-trace-disasm.py | 1 - 1 file changed, 1 deletion(-)tools/perf/scripts/python/arm-cs-trace-disasm.py +0
-1@@ -9,7 +9,6 @@ from __future__ import print_function import os from os import path -import sys import re from subprocess import * from optparse import OptionParser, make_optionSignature · 1 line
2.39.1Signature · 1 line
- Arnaldo
Signature · 1 line
- Arnaldo
Leo Yan
Reviewed-byRe: [PATCH 01/17] perf: get rid of unused import
On Tue, Jun 13, 2023 at 11:34:11PM -0300, Arnaldo Carvalho de Melo wrote:
[...]
Quoted 5 lines
Quoted 1 line
Since have applied this patch, it's no need to give my review tag :)
No, I usually can add a Reviewed-by tag even after having applied it to my local tree, as I still need to run tests before making it available via perf-tools-next, when I should not make any further changes.
Understand now. Here is my review tag:
PATCH 2/17 tools/perf/tests: fix shellcheck warning for stat+json_output
In reply to Athira RajeevFrom: Disha Goel <[email protected]>
Running shellcheck on stat+json_output testcase, generates below warning:
[ $(id -u) != 0 ] && [ $(cat /proc/sys/kernel/perf_event_paranoid) -gt $1 ]
^------^ SC2046 (warning): Quote this to prevent word splitting.
^-- SC2046 (warning): Quote this to prevent word splitting.
Fixed the warning by adding quotes to avoid word splitting.
ShellCheck result with patch:
# shellcheck -S warning stat+json_output.sh #
perf test result after the change:
94: perf stat JSON output linter : Ok
--- tools/perf/tests/shell/stat+json_output.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/tests/shell/stat+json_output.sh +1 -1
@@ -40,7 +40,7 @@ trap trap_cleanup EXIT TERM INT # Return true if perf_event_paranoid is > $1 and not running as root. function ParanoidAndNotRoot() { - [ $(id -u) != 0 ] && [ $(cat /proc/sys/kernel/perf_event_paranoid) -gt $1 ] + [ "$(id -u)" != 0 ] && [ "$(cat /proc/sys/kernel/perf_event_paranoid)" -gt $1 ] } check_no_args()
Signature · 1 line
2.39.1
PATCH 3/17 shellcheck : fixing signal names and adding double quotes for expression in test_arm_callgraph_fp
In reply to Athira RajeevFrom: Spoorthy S <[email protected]>
Running shellcheck -S on test_arm_calligraph_fp throws warnings SC2086 and SC3049,
$shellcheck -S warning tests/shell/test_arm_callgraph_fp.sh
rm -f $PERF_DATA
: Double quote to prevent globbing and word splitting.
trap cleanup_files exit term int
: In POSIX sh, using lower/mixed case for signal names is undefined.
After fixing the warnings,
$shellcheck tests/shell/test_arm_callgraph_fp.sh
$ echo $? 0
To address the POSIX shell warnings added changes to convert Lowercase signal names to uppercase in the script and double quoted the command substitutions($fix to "$fix") to solve Globbing warnings.
--- tools/perf/tests/shell/test_arm_callgraph_fp.sh | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
tools/perf/tests/shell/test_arm_callgraph_fp.sh +3 -3
@@ -9,13 +9,13 @@ TEST_PROGRAM="perf test -w leafloop" cleanup_files() { - rm -f $PERF_DATA + rm -f "$PERF_DATA" } -trap cleanup_files exit term int +trap cleanup_files EXIT TERM INT # 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 --user-callchains -- $TEST_PROGRAM 2> /dev/null & +perf record -o "$PERF_DATA" --call-graph fp -e cycles//u -D 1000 --user-callchains -- "$TEST_PROGRAM" 2> /dev/null & PID=$! echo " + Recording (PID=$PID)..."
Signature · 1 line
2.39.1
PATCH 4/17 tools/perf/tests: fix shellcheck warnings for daemon.sh
In reply to Athira RajeevFrom: Shirisha G <[email protected]>
Running shellcheck -S on daemon.sh throws below warnings:
Result from shellcheck:
# shellcheck -S warning daemon.sh
local line_name=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $2 }'`
^-------^ SC2155: Declare and assign separately to avoid masking return values. trap "echo 'FAILED: Signal caught'; daemon_exit ${config}; exit 1" SIGINT SIGTERM
^-------^ SC2064: Use single quotes, otherwise this expands now rather than when signalled.
count=`ls ${base}/session-test/ | grep perf.data | wc -l`
^-- SC2010: Don't use ls | grep. Use a glob or a for loop with a condition to allow non-alphanumeric filenames.
if [ ${size} != "OK" -o ${type} != "OK" ]; then
^-- SC2166: Prefer [ p ] || [ q ] as [ p -o q ] is not well defined.Fixed above warnings by:
- declaring and assigning local variables separately
- To fix SC2010, instead of using "ls | grep", used glob to allow non-alphanumeric filenames
- Used single quotes to prevent expanding.
Result from shellcheck after patch changes:
$ shellcheck -S warning daemon.sh
$ echo $?
0--- tools/perf/tests/shell/daemon.sh | 113 ++++++++++++++++++++----------- 1 file changed, 75 insertions(+), 38 deletions(-)
tools/perf/tests/shell/daemon.sh +75 -38
@@ -11,11 +11,16 @@ check_line_first() local lock=$5 local up=$6 - local line_name=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $2 }'` - local line_base=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $3 }'` - local line_output=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $4 }'` - local line_lock=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $5 }'` - local line_up=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $6 }'` + local line_name + line_name=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $2 }'` + local line_base + line_base=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $3 }'` + local line_output + line_output=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $4 }'` + local line_lock + line_lock=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $5 }'` + local line_up + line_up=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $6 }'` if [ "${name}" != "${line_name}" ]; then echo "FAILED: wrong name"
@@ -54,13 +59,20 @@ check_line_other() local ack=$7 local up=$8 - local line_name=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $2 }'` - local line_run=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $3 }'` - local line_base=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $4 }'` - local line_output=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $5 }'` - local line_control=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $6 }'` - local line_ack=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $7 }'` - local line_up=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $8 }'` + local line_name + line_name=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $2 }'` + local line_run + line_run=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $3 }'` + local line_base + line_base=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $4 }'` + local line_output + line_output=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $5 }'` + local line_control + line_control=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $6 }'` + local line_ack + line_ack=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $7 }'` + local line_up + line_up=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $8 }'` if [ "${name}" != "${line_name}" ]; then echo "FAILED: wrong name"
@@ -102,8 +114,10 @@ daemon_exit() { local config=$1 - local line=`perf daemon --config ${config} -x: | head -1` - local pid=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $1 }'` + local line + line=`perf daemon --config ${config} -x: | head -1` + local pid + pid=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $1 }'` # Reset trap handler. trap - SIGINT SIGTERM
@@ -123,7 +137,7 @@ daemon_start() perf daemon start --config ${config} # Clean up daemon if interrupted. - trap "echo 'FAILED: Signal caught'; daemon_exit ${config}; exit 1" SIGINT SIGTERM + trap 'echo "FAILED: Signal caught"; daemon_exit "${config}"; exit 1' SIGINT SIGTERM # wait for the session to ping local state="FAIL"
@@ -144,8 +158,10 @@ test_list() { echo "test daemon list" - local config=$(mktemp /tmp/perf.daemon.config.XXX) - local base=$(mktemp -d /tmp/perf.daemon.base.XXX) + local config + config=$(mktemp /tmp/perf.daemon.config.XXX) + local base + base=$(mktemp -d /tmp/perf.daemon.base.XXX) cat <<EOF > ${config} [daemon]
@@ -165,19 +181,22 @@ EOF # check first line # pid:daemon:base:base/output:base/lock - local line=`perf daemon --config ${config} -x: | head -1` + local line + line=`perf daemon --config ${config} -x: | head -1` check_line_first ${line} daemon ${base} ${base}/output ${base}/lock "0" # check 1st session # pid:size:-e cpu-clock:base/size:base/size/output:base/size/control:base/size/ack:0 - local line=`perf daemon --config ${config} -x: | head -2 | tail -1` + local line + line=`perf daemon --config ${config} -x: | head -2 | tail -1` check_line_other "${line}" size "-e cpu-clock -m 1 sleep 10" ${base}/session-size \ ${base}/session-size/output ${base}/session-size/control \ ${base}/session-size/ack "0" # check 2nd session # pid:time:-e task-clock:base/time:base/time/output:base/time/control:base/time/ack:0 - local line=`perf daemon --config ${config} -x: | head -3 | tail -1` + local line + line=`perf daemon --config ${config} -x: | head -3 | tail -1` check_line_other "${line}" time "-e task-clock -m 1 sleep 10" ${base}/session-time \ ${base}/session-time/output ${base}/session-time/control \ ${base}/session-time/ack "0"
@@ -193,8 +212,10 @@ test_reconfig() { echo "test daemon reconfig" - local config=$(mktemp /tmp/perf.daemon.config.XXX) - local base=$(mktemp -d /tmp/perf.daemon.base.XXX) + local config + config=$(mktemp /tmp/perf.daemon.config.XXX) + local base + base=$(mktemp -d /tmp/perf.daemon.base.XXX) # prepare config cat <<EOF > ${config}
@@ -215,10 +236,12 @@ EOF # check 2nd session # pid:time:-e task-clock:base/time:base/time/output:base/time/control:base/time/ack:0 - local line=`perf daemon --config ${config} -x: | head -3 | tail -1` + local line + line=`perf daemon --config ${config} -x: | head -3 | tail -1` check_line_other "${line}" time "-e task-clock -m 1 sleep 10" ${base}/session-time \ ${base}/session-time/output ${base}/session-time/control ${base}/session-time/ack "0" - local pid=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $1 }'` + local pid + pid=`echo "${line}" | awk 'BEGIN { FS = ":" } ; { print $1 }'` # prepare new config local config_new=${config}.new
@@ -249,7 +272,8 @@ EOF # check reconfigured 2nd session # pid:time:-e task-clock:base/time:base/time/output:base/time/control:base/time/ack:0 - local line=`perf daemon --config ${config} -x: | head -3 | tail -1` + local line + line=`perf daemon --config ${config} -x: | head -3 | tail -1` check_line_other "${line}" time "-e cpu-clock -m 1 sleep 10" ${base}/session-time \ ${base}/session-time/output ${base}/session-time/control ${base}/session-time/ack "0"
@@ -276,7 +300,8 @@ EOF state=`perf daemon ping --config ${config} --session size | awk '{ print $1 }'` done - local one=`perf daemon --config ${config} -x: | wc -l` + local one + one=`perf daemon --config ${config} -x: | wc -l` if [ ${one} -ne "1" ]; then echo "FAILED: wrong list output"
@@ -312,8 +337,10 @@ test_stop() { echo "test daemon stop" - local config=$(mktemp /tmp/perf.daemon.config.XXX) - local base=$(mktemp -d /tmp/perf.daemon.base.XXX) + local config + config=$(mktemp /tmp/perf.daemon.config.XXX) + local base + base=$(mktemp -d /tmp/perf.daemon.base.XXX) # prepare config cat <<EOF > ${config}
@@ -332,8 +359,12 @@ EOF # start daemon daemon_start ${config} size - local pid_size=`perf daemon --config ${config} -x: | head -2 | tail -1 | awk 'BEGIN { FS = ":" } ; { print $1 }'` - local pid_time=`perf daemon --config ${config} -x: | head -3 | tail -1 | awk 'BEGIN { FS = ":" } ; { print $1 }'` + local pid_size + pid_size=`perf daemon --config ${config} -x: | head -2 | tail -1 | + awk 'BEGIN { FS = ":" } ; { print $1 }'` + local pid_time + pid_time=`perf daemon --config ${config} -x: | head -3 | tail -1 | + awk 'BEGIN { FS = ":" } ; { print $1 }'` # check that sessions are running if [ ! -d "/proc/${pid_size}" ]; then
@@ -364,8 +395,10 @@ test_signal() { echo "test daemon signal" - local config=$(mktemp /tmp/perf.daemon.config.XXX) - local base=$(mktemp -d /tmp/perf.daemon.base.XXX) + local config + config=$(mktemp /tmp/perf.daemon.config.XXX) + local base + base=$(mktemp -d /tmp/perf.daemon.base.XXX) # prepare config cat <<EOF > ${config}
@@ -389,7 +422,7 @@ EOF daemon_exit ${config} # count is 2 perf.data for signals and 1 for perf record finished - count=`ls ${base}/session-test/ | grep perf.data | wc -l` + count=`ls ${base}/session-test/*perf.data* | wc -l` if [ ${count} -ne 3 ]; then error=1 echo "FAILED: perf data no generated"
@@ -403,8 +436,10 @@ test_ping() { echo "test daemon ping" - local config=$(mktemp /tmp/perf.daemon.config.XXX) - local base=$(mktemp -d /tmp/perf.daemon.base.XXX) + local config + config=$(mktemp /tmp/perf.daemon.config.XXX) + local base + base=$(mktemp -d /tmp/perf.daemon.base.XXX) # prepare config cat <<EOF > ${config}
@@ -426,7 +461,7 @@ EOF size=`perf daemon ping --config ${config} --session size | awk '{ print $1 }'` type=`perf daemon ping --config ${config} --session time | awk '{ print $1 }'` - if [ ${size} != "OK" -o ${type} != "OK" ]; then + if [ ${size} != "OK" ] || [ ${type} != "OK" ]; then error=1 echo "FAILED: daemon ping failed" fi
@@ -442,8 +477,10 @@ test_lock() { echo "test daemon lock" - local config=$(mktemp /tmp/perf.daemon.config.XXX) - local base=$(mktemp -d /tmp/perf.daemon.base.XXX) + local config + config=$(mktemp /tmp/perf.daemon.config.XXX) + local base + base=$(mktemp -d /tmp/perf.daemon.base.XXX) # prepare config cat <<EOF > ${config}
Signature · 1 line
2.39.1
PATCH 5/17 tools/perf/tests: Fix shellcheck warnings for stat+csv_output
In reply to Athira RajeevFrom: Korrapati Likhitha <[email protected]>
Running the shellcheck on stat+csv_output resulted in the following warning.
Result with shellcheck without patch:
=====
$ shellcheck -S warning stat+csv_output.sh
In stat+csv_output.sh line 23:
[ $(uname -m) = "s390x" ] && exp='^[6-7]$'
^---------^ SC2046: Quote this to prevent word splitting.In stat+csv_output.sh line 51:
[ $(id -u) != 0 ] && [ $(cat /proc/sys/kernel/perf_event_paranoid) -gt $1 ]
^------^ SC2046: Quote this to prevent word splitting.
^-- SC2046: Quote this to prevent word splitting.
=====
Fixed the warning SC2046 by adding quotes to prevent word splitting.
Result with shellcheck with patch:
=====
$ shellcheck -S warning tests/shell/stat+csv_output.sh $ ./perf test "stat CSV output linter" 96: perf stat CSV output linter : Ok =====
--- tools/perf/tests/shell/stat+csv_output.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/tests/shell/stat+csv_output.sh +2 -2
@@ -35,7 +35,7 @@ function commachecker() ;; "--interval") exp=7 ;; "--per-thread") exp=7 ;; "--system-wide-no-aggr") exp=7 - [ $(uname -m) = "s390x" ] && exp='^[6-7]$' + [ "$(uname -m)" = "s390x" ] && exp='^[6-7]$' ;; "--per-core") exp=8 ;; "--per-socket") exp=8 ;; "--per-node") exp=8
@@ -65,7 +65,7 @@ function commachecker() # Return true if perf_event_paranoid is > $1 and not running as root. function ParanoidAndNotRoot() { - [ $(id -u) != 0 ] && [ $(cat /proc/sys/kernel/perf_event_paranoid) -gt $1 ] + [ "$(id -u)" != 0 ] && [ "$(cat /proc/sys/kernel/perf_event_paranoid)" -gt $1 ] } check_no_args()
Signature · 1 line
2.39.1
PATCH 6/17 tools/perf/tests: Fix shellcheck warnings for trace+probe_vfs_getname.sh
In reply to Athira RajeevFrom: Akanksha J N <[email protected]>
Fix the shellcheck warnings on powerpc and x86 for testcase trace+probe_vfs_getname.sh. Add quotes to prevent word splitting which are caused by unquoted command expansions.
Before fix:
$ shellcheck -S warning trace+probe_vfs_getname.sh
In trace+probe_vfs_getname.sh line 13: . $(dirname $0)/lib/probe.sh ^-----------^ SC2046 (warning): Quote this to prevent word splitting. In trace+probe_vfs_getname.sh line 18: . $(dirname $0)/lib/probe_vfs_getname.sh ^-----------^ SC2046 (warning): Quote this to prevent word splitting. In trace+probe_vfs_getname.sh line 21: evts=$(echo $(perf list syscalls:sys_enter_open* 2>/dev/null | grep -E 'open(at)? ' | sed -r 's/.*sys_enter_([a-z]+) +\[.*$/\1/') | sed 's/ /,/') ^-- SC2046 (warning): Quote this to prevent word splitting. 100: Check open filename arg using perf trace + vfs_getname : Ok
After the fix:
$ shellcheck -S warning trace+probe_vfs_getname.sh
100: Check open filename arg using perf trace + vfs_getname : Ok
--- tools/perf/tests/shell/trace+probe_vfs_getname.sh | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
tools/perf/tests/shell/trace+probe_vfs_getname.sh +3 -3
@@ -10,15 +10,15 @@ # SPDX-License-Identifier: GPL-2.0 # Arnaldo Carvalho de Melo <[email protected]>, 2017 -. $(dirname $0)/lib/probe.sh +. "$(dirname $0)"/lib/probe.sh skip_if_no_perf_probe || exit 2 skip_if_no_perf_trace || exit 2 -. $(dirname $0)/lib/probe_vfs_getname.sh +. "$(dirname $0)"/lib/probe_vfs_getname.sh trace_open_vfs_getname() { - evts=$(echo $(perf list syscalls:sys_enter_open* 2>/dev/null | grep -E 'open(at)? ' | sed -r 's/.*sys_enter_([a-z]+) +\[.*$/\1/') | sed 's/ /,/') + evts=$(echo "$(perf list syscalls:sys_enter_open* 2>/dev/null | grep -E 'open(at)? ' | sed -r 's/.*sys_enter_([a-z]+) +\[.*$/\1/')" | sed 's/ /,/') perf trace -e $evts touch $file 2>&1 | \ grep -E " +[0-9]+\.[0-9]+ +\( +[0-9]+\.[0-9]+ ms\): +touch\/[0-9]+ open(at)?\((dfd: +CWD, +)?filename: +${file}, +flags: CREAT\|NOCTTY\|NONBLOCK\|WRONLY, +mode: +IRUGO\|IWUGO\) += +[0-9]+$" }
Signature · 1 line
2.39.1
Arnaldo Carvalho de Melo
maintainerRe: [PATCH 06/17] tools/perf/tests: Fix shellcheck warnings for trace+probe_vfs_getname.sh
Em Tue, Jun 13, 2023 at 10:11:34PM +0530, Athira Rajeev escreveu:
Quoted 24 lines
From: Akanksha J N <[email protected]>
Fix the shellcheck warnings on powerpc and x86 for testcase
trace+probe_vfs_getname.sh. Add quotes to prevent word splitting which are caused by unquoted command expansions.Before fix:
$ shellcheck -S warning trace+probe_vfs_getname.shIn trace+probe_vfs_getname.sh line 13: . $(dirname $0)/lib/probe.sh ^-----------^ SC2046 (warning): Quote this to prevent word splitting. In trace+probe_vfs_getname.sh line 18: . $(dirname $0)/lib/probe_vfs_getname.sh ^-----------^ SC2046 (warning): Quote this to prevent word splitting. In trace+probe_vfs_getname.sh line 21: evts=$(echo $(perf list syscalls:sys_enter_open* 2>/dev/null | grep -E 'open(at)? ' | sed -r 's/.*sys_enter_([a-z]+) +\[.*$/\1/') | sed 's/ /,/') ^-- SC2046 (warning): Quote this to prevent word splitting. 100: Check open filename arg using perf trace + vfs_getname : Ok
Quoted 5 lines
After the fix:
$ shellcheck -S warning trace+probe_vfs_getname.sh100: Check open filename arg using perf trace +
vfs_getname: Ok
So, I tried this on x86_64, fedora and get:
[root@quaco ~]# perf test "trace + vfs"
115: Check open filename arg using perf trace + vfs_getname : FAILED!
[root@quaco ~]#
Then, looking at the change:
tools/perf/tests/shell/trace+probe_vfs_getname.sh line 21
Quoted 2 lines
tools/perf/tests/shell/trace+probe_vfs_getname.sh +1
-1- evts=$(echo $(perf list syscalls:sys_enter_open* 2>/dev/null | grep -E 'open(at)? ' | sed -r 's/.*sys_enter_([a-z]+) +\[.*$/\1/') | sed 's/ /,/') + evts=$(echo "$(perf list syscalls:sys_enter_open* 2>/dev/null | grep -E 'open(at)? ' | sed -r 's/.*sys_enter_([a-z]+) +\[.*$/\1/')" | sed 's/ /,/')
So, before:
[root@quaco ~]# evts=$(echo $(perf list syscalls:sys_enter_open* 2>/dev/null | grep -E 'open(at)? ' | sed -r 's/.*sys_enter_([a-z]+) +\[.*$/\1/') | sed 's/ /,/') [root@quaco ~]# echo $evts
open,openat
[root@quaco ~]#
Then after:
[root@quaco ~]# evts=$(echo "$(perf list syscalls:sys_enter_open* 2>/dev/null | grep -E 'open(at)? ' | sed -r 's/.*sys_enter_([a-z]+) +\[.*$/\1/')" | sed 's/ /,/') [root@quaco ~]# echo $evts
open openat
[root@quaco ~]#
Not equivalent, so I'm removing this patch, please check and resubmit, please.
- Arnaldo
Quoted 37 lines
Signed-off-by Athira Rajeev <[email protected]>Signed-off-by Kajol Jain <[email protected]>Signed-off-by Akanksha J N <[email protected]>Signed-off-by Abhishek Singh Tomar <[email protected]>Signed-off-by Saket <[email protected]>Signed-off-by Avnish Chouhan <[email protected]>--- tools/perf/tests/shell/trace+probe_vfs_getname.sh | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)tools/perf/tests/shell/trace+probe_vfs_getname.sh +3
-3@@ -10,15 +10,15 @@ # SPDX-License-Identifier: GPL-2.0 # Arnaldo Carvalho de Melo <[email protected]>, 2017 -. $(dirname $0)/lib/probe.sh +. "$(dirname $0)"/lib/probe.sh skip_if_no_perf_probe || exit 2 skip_if_no_perf_trace || exit 2 -. $(dirname $0)/lib/probe_vfs_getname.sh +. "$(dirname $0)"/lib/probe_vfs_getname.sh trace_open_vfs_getname() { - evts=$(echo $(perf list syscalls:sys_enter_open* 2>/dev/null | grep -E 'open(at)? ' | sed -r 's/.*sys_enter_([a-z]+) +\[.*$/\1/') | sed 's/ /,/') + evts=$(echo "$(perf list syscalls:sys_enter_open* 2>/dev/null | grep -E 'open(at)? ' | sed -r 's/.*sys_enter_([a-z]+) +\[.*$/\1/')" | sed 's/ /,/') perf trace -e $evts touch $file 2>&1 | \ grep -E " +[0-9]+\.[0-9]+ +\( +[0-9]+\.[0-9]+ ms\): +touch\/[0-9]+ open(at)?\((dfd: +CWD, +)?filename: +${file}, +flags: CREAT\|NOCTTY\|NONBLOCK\|WRONLY, +mode: +IRUGO\|IWUGO\) += +[0-9]+$" }Signature · 1 line
2.39.1
Signature · 1 line
- Arnaldo
PATCH 7/17 perf/tests/shell : Shellcheck fixes for perf test "test_arm_coresight.sh"
In reply to Athira RajeevFrom: Anushree Mathur <[email protected]>
Fixed the following shellcheck issues in test_arm_coresight.sh file:
In tools/perf/tests/shell/test_arm_coresight.sh line 31:
trap - exit term int
^--^ SC2039: In POSIX sh, using lower/mixed case for signal names is undefined.
^--^ SC2039: In POSIX sh, using lower/mixed case for signal names is undefined.
^-^ SC2039: In POSIX sh, using lower/mixed case for signal names is undefined.
In tools/perf/tests/shell/test_arm_coresight.sh line 35:
trap cleanup_files exit term int
^--^ SC2039: In POSIX sh, using lower/mixed case for signal names is undefined.
^--^ SC2039: In POSIX sh, using lower/mixed case for signal names is undefined.
^-^ SC2039: In POSIX sh, using lower/mixed case for signal names is undefined.
In tools/perf/tests/shell/test_arm_coresight.sh line 92:
if [ $? -eq 0 -a -e "$1/enable_sink" ]; then
^-- SC2166: Prefer [ p ] && [ q ] as [ p -a q ] is not well defined.Fixed above warnings by:
1)Capitalize signals(INT, TERM, EXIT) to avoid mixed/lower case naming of signals.
2)Expression [p -a q] was not defined,changed it to [p] && [q] to avoid the ambiguity as this is older format using -a or -o ,now we use [p] && [q] in place of [p -a q] and [p] || [q] in place of [p -o q].
Result after fixing the issues:
shell$ shellcheck -S warning test_arm_coresight.sh shell$
--- tools/perf/tests/shell/test_arm_coresight.sh | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
tools/perf/tests/shell/test_arm_coresight.sh +3 -3
@@ -28,11 +28,11 @@ cleanup_files() rm -f ${perfdata} rm -f ${file} rm -f "${perfdata}.old" - trap - exit term int + trap - EXIT TERM INT exit $glb_err } -trap cleanup_files exit term int +trap cleanup_files EXIT TERM INT record_touch_file() { echo "Recording trace (only user mode) with path: CPU$2 => $1"
@@ -89,7 +89,7 @@ is_device_sink() { # cannot support perf PMU. echo "$1" | grep -E -q -v "tpiu" - if [ $? -eq 0 -a -e "$1/enable_sink" ]; then + if [ $? -eq 0 ] && [ -e "$1/enable_sink" ]; then pmu_dev="/sys/bus/event_source/devices/cs_etm/sinks/$2"
Signature · 1 line
2.39.1
PATCH 8/17 tools/perf/tests/shell/stat_all_metrics: Fix shellcheck warning SC2076 in stat_all_metrics.sh
In reply to Athira RajeevFrom: Barnali Guha Thakurata <[email protected]>
Fixed shellcheck warning SC2076 in stat_all_metrics.sh.
Before the patch:
shell$ shellcheck stat_all_metrics.sh
In stat_all_metrics.sh line 9:
if [[ "$result" =~ "${m:0:50}" ]] || [[ "$result" =~ "<not supported>" ]]
^---------^ SC2076: Don't quote right-hand
side of =~, it'll match literally rather than as a regex.
In stat_all_metrics.sh line 15:
if [[ "$result" =~ "${m:0:50}" ]]
^---------^ SC2076: Don't quote right-hand
side of =~, it'll match literally rather than as a regex.
In stat_all_metrics.sh line 22:
if [[ "$result" =~ "${m:0:50}" ]]
^---------^ SC2076: Don't quote right-hand
side of =~, it'll match literally rather than as a regex.
After the patch:
shell$ shellcheck stat_all_metrics.sh
shell$
--- tools/perf/tests/shell/stat_all_metrics.sh | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
tools/perf/tests/shell/stat_all_metrics.sh +3 -3
@@ -6,20 +6,20 @@ err=0 for m in $(perf list --raw-dump metrics); do echo "Testing $m" result=$(perf stat -M "$m" true 2>&1) - if [[ "$result" =~ "${m:0:50}" ]] || [[ "$result" =~ "<not supported>" ]] + if [[ "$result" =~ ${m:0:50} ]] || [[ "$result" =~ "<not supported>" ]] then continue fi # Failed so try system wide. result=$(perf stat -M "$m" -a sleep 0.01 2>&1) - if [[ "$result" =~ "${m:0:50}" ]] + if [[ "$result" =~ ${m:0:50} ]] then continue fi # Failed again, possibly the workload was too small so retry with something # longer. result=$(perf stat -M "$m" perf bench internals synthesize 2>&1) - if [[ "$result" =~ "${m:0:50}" ]] + if [[ "$result" =~ ${m:0:50} ]] then continue fi
Signature · 1 line
2.39.1
PATCH 9/17 tools/perf/tests: Fix shellcheck issues in test_task_analyzer.sh file
In reply to Athira RajeevFrom: Aboorva Devarajan <[email protected]>
Fixed the following shellcheck issues in test_task_analyzer.sh file:
SC2086: Double quote to prevent globbing and word splitting warnings in shell-check.
Fixes the following shellcheck issues,
SC2086: Double quote to prevent globbing and word splitting warnings in shell-check.
Before Patch:
$ shellcheck ./test_task_analyzer.sh | grep "SC2086" | ... In ./test_task_analyzer.sh line 13: SC2086: Double quote to prevent globbing and word splitting. In ./test_task_analyzer.sh line 24: SC2086: Double quote to prevent globbing and word splitting. In ./test_task_analyzer.sh line 39: SC2086: Double quote to prevent globbing and word splitting.
After Patch:
$ shellcheck ./test_task_analyzer.sh | grep -i "SC2086" None
perf test result after patch:
PASS: "test_basic"
PASS: "test_ns_rename"
PASS: "test_ms_filtertasks_highlight"
PASS: "test_extended_times_timelimit_limittasks"
PASS: "test_summary"
PASS: "test_summaryextended"
PASS: "test_summaryonly"
PASS: "test_extended_times_summary_ns"
PASS: "test_extended_times_summary_ns"
PASS: "test_csv"
PASS: "test_csvsummary"
PASS: "test_csv_extended_times"
PASS: "test_csvsummary_extended"
--- tools/perf/tests/shell/test_task_analyzer.sh | 54 ++++++++++---------- 1 file changed, 27 insertions(+), 27 deletions(-)
tools/perf/tests/shell/test_task_analyzer.sh +27 -27
@@ -10,7 +10,7 @@ cleanup() { rm -f perf.data.old rm -f csv rm -f csvsummary - rm -rf $tmpdir + rm -rf "$tmpdir" trap - exit term int }
@@ -21,7 +21,7 @@ trap_cleanup() { trap trap_cleanup exit term int report() { - if [ $1 = 0 ]; then + if [ "$1" = 0 ]; then echo "PASS: \"$2\"" else echo "FAIL: \"$2\" Error message: \"$3\""
@@ -36,11 +36,11 @@ check_exec_0() { } find_str_or_fail() { - grep -q "$1" $2 - if [ $? != 0 ]; then - report 1 $3 "Failed to find required string:'${1}'." + grep -q "$1" "$2" + if [ "$?" != 0 ]; then + report 1 "$3" "Failed to find required string:'${1}'." else - report 0 $3 + report 0 "$3" fi }
@@ -52,86 +52,86 @@ prepare_perf_data() { # check standard inkvokation with no arguments test_basic() { out="$tmpdir/perf.out" - perf script report task-analyzer > $out + perf script report task-analyzer > "$out" check_exec_0 "perf" - find_str_or_fail "Comm" $out ${FUNCNAME[0]} + find_str_or_fail "Comm" "$out" "${FUNCNAME[0]}" } test_ns_rename(){ out="$tmpdir/perf.out" - perf script report task-analyzer --ns --rename-comms-by-tids 0:random > $out + perf script report task-analyzer --ns --rename-comms-by-tids 0:random > "$out" check_exec_0 "perf" - find_str_or_fail "Comm" $out ${FUNCNAME[0]} + find_str_or_fail "Comm" "$out" "${FUNCNAME[0]}" } test_ms_filtertasks_highlight(){ out="$tmpdir/perf.out" perf script report task-analyzer --ms --filter-tasks perf --highlight-tasks perf \ - > $out + > "$out" check_exec_0 "perf" - find_str_or_fail "Comm" $out ${FUNCNAME[0]} + find_str_or_fail "Comm" "$out" "${FUNCNAME[0]}" } test_extended_times_timelimit_limittasks() { out="$tmpdir/perf.out" perf script report task-analyzer --extended-times --time-limit :99999 \ - --limit-to-tasks perf > $out + --limit-to-tasks perf > "$out" check_exec_0 "perf" - find_str_or_fail "Out-Out" $out ${FUNCNAME[0]} + find_str_or_fail "Out-Out" "$out" "${FUNCNAME[0]}" } test_summary() { out="$tmpdir/perf.out" - perf script report task-analyzer --summary > $out + perf script report task-analyzer --summary > "$out" check_exec_0 "perf" - find_str_or_fail "Summary" $out ${FUNCNAME[0]} + find_str_or_fail "Summary" "$out" "${FUNCNAME[0]}" } test_summaryextended() { out="$tmpdir/perf.out" - perf script report task-analyzer --summary-extended > $out + perf script report task-analyzer --summary-extended > "$out" check_exec_0 "perf" - find_str_or_fail "Inter Task Times" $out ${FUNCNAME[0]} + find_str_or_fail "Inter Task Times" "$out" "${FUNCNAME[0]}" } test_summaryonly() { out="$tmpdir/perf.out" - perf script report task-analyzer --summary-only > $out + perf script report task-analyzer --summary-only > "$out" check_exec_0 "perf" - find_str_or_fail "Summary" $out ${FUNCNAME[0]} + find_str_or_fail "Summary" "$out" "${FUNCNAME[0]}" } test_extended_times_summary_ns() { out="$tmpdir/perf.out" - perf script report task-analyzer --extended-times --summary --ns > $out + perf script report task-analyzer --extended-times --summary --ns > "$out" check_exec_0 "perf" - find_str_or_fail "Out-Out" $out ${FUNCNAME[0]} - find_str_or_fail "Summary" $out ${FUNCNAME[0]} + find_str_or_fail "Out-Out" "$out" "${FUNCNAME[0]}" + find_str_or_fail "Summary" "$out" "${FUNCNAME[0]}" } test_csv() { perf script report task-analyzer --csv csv > /dev/null check_exec_0 "perf" - find_str_or_fail "Comm;" csv ${FUNCNAME[0]} + find_str_or_fail "Comm;" csv "${FUNCNAME[0]}" } test_csv_extended_times() { perf script report task-analyzer --csv csv --extended-times > /dev/null check_exec_0 "perf" - find_str_or_fail "Out-Out;" csv ${FUNCNAME[0]} + find_str_or_fail "Out-Out;" csv "${FUNCNAME[0]}" } test_csvsummary() { perf script report task-analyzer --csv-summary csvsummary > /dev/null check_exec_0 "perf" - find_str_or_fail "Comm;" csvsummary ${FUNCNAME[0]} + find_str_or_fail "Comm;" csvsummary "${FUNCNAME[0]}" } test_csvsummary_extended() { perf script report task-analyzer --csv-summary csvsummary --summary-extended \ >/dev/null check_exec_0 "perf" - find_str_or_fail "Out-Out;" csvsummary ${FUNCNAME[0]} + find_str_or_fail "Out-Out;" csvsummary "${FUNCNAME[0]}" } prepare_perf_data
Signature · 1 line
2.39.1
PATCH 10/17 tools/perf/tests: fix test_arm_spe.sh signal case issues
In reply to Athira RajeevFrom: Abhirup Deb <[email protected]>
Running shellcheck -S on test_arm_spe.sh throws below warnings:
#shellcheck -S warning tests/shell/test_arm_spe.sh
In tests/shell/test_arm_spe.sh line 30:
trap cleanup_files exit term int
^--^ SC3049 (warning): In POSIX sh, using lower/mixed case for signal names is undefined.
^--^ SC3049 (warning): In POSIX sh, using lower/mixed case for signal names is undefined.
^-^ SC3049 (warning): In POSIX sh, using lower/mixed case for signal names is undefined.
Fixed this issue by using uppercase for "EXIT", "TERM" and "INIT" signals to avoid using lower/mixed case for signal names as input.
--- tools/perf/tests/shell/test_arm_spe.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/tests/shell/test_arm_spe.sh +1 -1
@@ -27,7 +27,7 @@ cleanup_files() exit $glb_err } -trap cleanup_files exit term int +trap cleanup_files EXIT TERM INT arm_spe_report() { if [ $2 = 0 ]; then
Signature · 1 line
2.39.1
PATCH 11/17 perf/tests/shell: fix shellscript errors for lock_contention.sh
In reply to Athira RajeevFrom: Abhirup Deb <[email protected]>
Use quotes around variables to prevent POSIX word expansion, use uppercase for signals(INT, TERM, EXIT) to avoid mixed/lower case naming of signals and replace "==" with "=" as "==" is not supported by POSIX shell.
--- tools/perf/tests/shell/lock_contention.sh | 70 +++++++++++------------ 1 file changed, 35 insertions(+), 35 deletions(-)
tools/perf/tests/shell/lock_contention.sh +35 -35
@@ -11,14 +11,14 @@ result=$(mktemp /tmp/__perf_test.result.XXXXX) cleanup() { rm -f ${perfdata} rm -f ${result} - trap - exit term int + trap - EXIT TERM INT } trap_cleanup() { cleanup exit ${err} } -trap trap_cleanup exit term int +trap trap_cleanup EXIT TERM INT check() { if [ `id -u` != 0 ]; then
@@ -40,8 +40,8 @@ test_record() perf lock record -o ${perfdata} -- perf bench sched messaging > /dev/null 2>&1 # the output goes to the stderr and we expect only 1 output (-E 1) perf lock contention -i ${perfdata} -E 1 -q 2> ${result} - if [ $(cat "${result}" | wc -l) != "1" ]; then - echo "[Fail] Recorded result count is not 1:" $(cat "${result}" | wc -l) + if [ "$(cat "${result}" | wc -l)" != "1" ]; then + echo "[Fail] Recorded result count is not 1:" "$(cat "${result}" | wc -l)" err=1 exit fi
@@ -58,8 +58,8 @@ test_bpf() # the perf lock contention output goes to the stderr perf lock con -a -b -E 1 -q -- perf bench sched messaging > /dev/null 2> ${result} - if [ $(cat "${result}" | wc -l) != "1" ]; then - echo "[Fail] BPF result count is not 1:" $(cat "${result}" | wc -l) + if [ "$(cat "${result}" | wc -l)" != "1" ]; then + echo "[Fail] BPF result count is not 1:" "$(cat "${result}" | wc -l)" err=1 exit fi
@@ -70,8 +70,8 @@ test_record_concurrent() echo "Testing perf lock record and perf lock contention at the same time" perf lock record -o- -- perf bench sched messaging 2> /dev/null | \ perf lock contention -i- -E 1 -q 2> ${result} - if [ $(cat "${result}" | wc -l) != "1" ]; then - echo "[Fail] Recorded result count is not 1:" $(cat "${result}" | wc -l) + if [ "$(cat "${result}" | wc -l)" != "1" ]; then + echo "[Fail] Recorded result count is not 1:" "$(cat "${result}" | wc -l)" err=1 exit fi
@@ -81,8 +81,8 @@ test_aggr_task() { echo "Testing perf lock contention --threads" perf lock contention -i ${perfdata} -t -E 1 -q 2> ${result} - if [ $(cat "${result}" | wc -l) != "1" ]; then - echo "[Fail] Recorded result count is not 1:" $(cat "${result}" | wc -l) + if [ "$(cat "${result}" | wc -l)" != "1" ]; then + echo "[Fail] Recorded result count is not 1:" "$(cat "${result}" | wc -l)" err=1 exit fi
@@ -93,8 +93,8 @@ test_aggr_task() # the perf lock contention output goes to the stderr perf lock con -a -b -t -E 1 -q -- perf bench sched messaging > /dev/null 2> ${result} - if [ $(cat "${result}" | wc -l) != "1" ]; then - echo "[Fail] BPF result count is not 1:" $(cat "${result}" | wc -l) + if [ "$(cat "${result}" | wc -l)" != "1" ]; then + echo "[Fail] BPF result count is not 1:" "$(cat "${result}" | wc -l)" err=1 exit fi
@@ -104,8 +104,8 @@ test_aggr_addr() { echo "Testing perf lock contention --lock-addr" perf lock contention -i ${perfdata} -l -E 1 -q 2> ${result} - if [ $(cat "${result}" | wc -l) != "1" ]; then - echo "[Fail] Recorded result count is not 1:" $(cat "${result}" | wc -l) + if [ "$(cat "${result}" | wc -l)" != "1" ]; then + echo "[Fail] Recorded result count is not 1:" "$(cat "${result}" | wc -l)" err=1 exit fi
@@ -116,8 +116,8 @@ test_aggr_addr() # the perf lock contention output goes to the stderr perf lock con -a -b -l -E 1 -q -- perf bench sched messaging > /dev/null 2> ${result} - if [ $(cat "${result}" | wc -l) != "1" ]; then - echo "[Fail] BPF result count is not 1:" $(cat "${result}" | wc -l) + if [ "$(cat "${result}" | wc -l)" != "1" ]; then + echo "[Fail] BPF result count is not 1:" "$(cat "${result}" | wc -l)" err=1 exit fi
@@ -127,8 +127,8 @@ test_type_filter() { echo "Testing perf lock contention --type-filter (w/ spinlock)" perf lock contention -i ${perfdata} -Y spinlock -q 2> ${result} - if [ $(grep -c -v spinlock "${result}") != "0" ]; then - echo "[Fail] Recorded result should not have non-spinlocks:" $(cat "${result}") + if [ "$(grep -c -v spinlock "${result}")" != "0" ]; then + echo "[Fail] Recorded result should not have non-spinlocks:" "$(cat "${result}")" err=1 exit fi
@@ -138,8 +138,8 @@ test_type_filter() fi perf lock con -a -b -Y spinlock -q -- perf bench sched messaging > /dev/null 2> ${result} - if [ $(grep -c -v spinlock "${result}") != "0" ]; then - echo "[Fail] BPF result should not have non-spinlocks:" $(cat "${result}") + if [ "$(grep -c -v spinlock "${result}")" != "0" ]; then + echo "[Fail] BPF result should not have non-spinlocks:" "$(cat "${result}")" err=1 exit fi
@@ -149,7 +149,7 @@ test_lock_filter() { echo "Testing perf lock contention --lock-filter (w/ tasklist_lock)" perf lock contention -i ${perfdata} -l -q 2> ${result} - if [ $(grep -c tasklist_lock "${result}") != "1" ]; then + if [ "$(grep -c tasklist_lock "${result}")" != "1" ]; then echo "[Skip] Could not find 'tasklist_lock'" return fi
@@ -159,8 +159,8 @@ test_lock_filter() # find out the type of tasklist_lock local type=$(head -1 "${result}" | awk '{ print $8 }' | sed -e 's/:.*//') - if [ $(grep -c -v "${type}" "${result}") != "0" ]; then - echo "[Fail] Recorded result should not have non-${type} locks:" $(cat "${result}") + if [ "$(grep -c -v "${type}" "${result}")" != "0" ]; then + echo "[Fail] Recorded result should not have non-${type} locks:" "$(cat "${result}")" err=1 exit fi
@@ -170,8 +170,8 @@ test_lock_filter() fi perf lock con -a -b -L tasklist_lock -q -- perf bench sched messaging > /dev/null 2> ${result} - if [ $(grep -c -v "${type}" "${result}") != "0" ]; then - echo "[Fail] BPF result should not have non-${type} locks:" $(cat "${result}") + if [ "$(grep -c -v "${type}" "${result}")" != "0" ]; then + echo "[Fail] BPF result should not have non-${type} locks:" "$(cat "${result}")" err=1 exit fi
@@ -181,14 +181,14 @@ test_stack_filter() { echo "Testing perf lock contention --callstack-filter (w/ unix_stream)" perf lock contention -i ${perfdata} -v -q 2> ${result} - if [ $(grep -c unix_stream "${result}") == "0" ]; then + if [ "$(grep -c unix_stream "${result}")" = "0" ]; then echo "[Skip] Could not find 'unix_stream'" return fi perf lock contention -i ${perfdata} -E 1 -S unix_stream -q 2> ${result} - if [ $(cat "${result}" | wc -l) != "1" ]; then - echo "[Fail] Recorded result should have a lock from unix_stream:" $(cat "${result}") + if [ "$(cat "${result}" | wc -l)" != "1" ]; then + echo "[Fail] Recorded result should have a lock from unix_stream:" "$(cat "${result}")" err=1 exit fi
@@ -198,8 +198,8 @@ test_stack_filter() fi perf lock con -a -b -S unix_stream -E 1 -q -- perf bench sched messaging > /dev/null 2> ${result} - if [ $(cat "${result}" | wc -l) != "1" ]; then - echo "[Fail] BPF result should have a lock from unix_stream:" $(cat "${result}") + if [ "$(cat "${result}" | wc -l)" != "1" ]; then + echo "[Fail] BPF result should have a lock from unix_stream:" "$(cat "${result}")" err=1 exit fi
@@ -209,14 +209,14 @@ test_aggr_task_stack_filter() { echo "Testing perf lock contention --callstack-filter with task aggregation" perf lock contention -i ${perfdata} -v -q 2> ${result} - if [ $(grep -c unix_stream "${result}") == "0" ]; then + if [ "$(grep -c unix_stream "${result}")" = "0" ]; then echo "[Skip] Could not find 'unix_stream'" return fi perf lock contention -i ${perfdata} -t -E 1 -S unix_stream -q 2> ${result} - if [ $(cat "${result}" | wc -l) != "1" ]; then - echo "[Fail] Recorded result should have a task from unix_stream:" $(cat "${result}") + if [ "$(cat "${result}" | wc -l)" != "1" ]; then + echo "[Fail] Recorded result should have a task from unix_stream:" "$(cat "${result}")" err=1 exit fi
@@ -226,8 +226,8 @@ test_aggr_task_stack_filter() fi perf lock con -a -b -t -S unix_stream -E 1 -q -- perf bench sched messaging > /dev/null 2> ${result} - if [ $(cat "${result}" | wc -l) != "1" ]; then - echo "[Fail] BPF result should have a task from unix_stream:" $(cat "${result}") + if [ "$(cat "${result}" | wc -l)" != "1" ]; then + echo "[Fail] BPF result should have a task from unix_stream:" "$(cat "${result}")" err=1 exit fi
Signature · 1 line
2.39.1
PATCH 12/17 tools/perf/tests: fixed shellcheck warnings for perf shell scripts
In reply to Athira RajeevFrom: Samir Mulani <[email protected]>
Fixed the shellcheck warnings in buildid.sh, record+probe_libc_inet_pton.sh and record+script_probe_vfs_getname.sh perf shell scripts:
1. Prefer [ p ] && [ q ] as [ p -a q ] is not well defined.
2. Prefer [ p ] || [ q ] as [ p -o q ] is not well defined.
3. Used * argument to avoid the argument mixes string and array
4. Resolved issue for variable refernce, where the variable is being used before it has been initialized.
5. Resolved word splitting issue (syntax error).
6. The "err" variable has been removed from buildid.sh since it is not used anywhere in the code.
--- tools/perf/tests/shell/buildid.sh | 12 ++++++------ .../perf/tests/shell/record+probe_libc_inet_pton.sh | 6 +++--- .../tests/shell/record+script_probe_vfs_getname.sh | 4 ++-- 3 files changed, 11 insertions(+), 11 deletions(-)
tools/perf/tests/shell/buildid.sh +6 -6
@@ -83,12 +83,12 @@ check() # in case of pe-file.exe file echo $1 | grep ".exe" if [ $? -eq 0 ]; then - if [ -x $1 -a ! -x $file ]; then + if [ -x $1 ] && [ ! -x $file ]; then echo "failed: file ${file} executable does not exist" exit 1 fi - if [ ! -x $file -a ! -e $file ]; then + if [ ! -x $file ] && [ ! -e $file ]; then echo "failed: file ${file} does not exist" exit 1 fi
@@ -136,10 +136,10 @@ test_record() log_err=$(mktemp /tmp/perf.log.err.XXX) perf="perf --buildid-dir ${build_id_dir}" - echo "running: perf record $@" - ${perf} record --buildid-all -o ${data} $@ 1>${log_out} 2>${log_err} + echo "running: perf record $*" + ${perf} record --buildid-all -o ${data} "$@" 1>${log_out} 2>${log_err} if [ $? -ne 0 ]; then - echo "failed: record $@" + echo "failed: record $*" echo "see log: ${log_err}" exit 1 fi
@@ -172,4 +172,4 @@ if [ ${run_pe} -eq 1 ]; then rm -r ${wineprefix} fi -exit ${err} +exit 0
tools/perf/tests/shell/record+probe_libc_inet_pton.sh +3 -3
@@ -10,8 +10,8 @@ # SPDX-License-Identifier: GPL-2.0 # Arnaldo Carvalho de Melo <[email protected]>, 2017 -. $(dirname $0)/lib/probe.sh -. $(dirname $0)/lib/probe_vfs_getname.sh +. "$(dirname "$0")/lib/probe.sh" +. "$(dirname "$0")/lib/probe_vfs_getname.sh" libc=$(grep -w libc /proc/self/maps | head -1 | sed -r 's/.*[[:space:]](\/.*)/\1/g') nm -Dg $libc 2>/dev/null | fgrep -q inet_pton || exit 254
@@ -23,7 +23,7 @@ add_libc_inet_pton_event() { event_name=$(perf probe -f -x $libc -a inet_pton 2>&1 | tail -n +2 | head -n -5 | \ grep -P -o "$event_pattern(?=[[:space:]]\(on inet_pton in $libc\))") - if [ $? -ne 0 -o -z "$event_name" ] ; then + if [ $? -ne 0 ] || [ -z "$event_name" ] ; then printf "FAIL: could not add event\n" return 1 fi
tools/perf/tests/shell/record+script_probe_vfs_getname.sh +2 -2
@@ -9,11 +9,11 @@ # SPDX-License-Identifier: GPL-2.0 # Arnaldo Carvalho de Melo <[email protected]>, 2017 -. $(dirname $0)/lib/probe.sh +. "$(dirname "$0")/lib/probe.sh" skip_if_no_perf_probe || exit 2 -. $(dirname $0)/lib/probe_vfs_getname.sh +. "$(dirname "$0")/lib/probe_vfs_getname.sh" record_open_file() { echo "Recording open file:"
Signature · 1 line
2.39.1
PATCH 13/17 tools/perf/tests: Fix all POSIX sh warnings in perf shell test test_brstack.sh
In reply to Athira RajeevFrom: Geetika <[email protected]>
Fix all the POSIX sh warnings in perf shell test test_brstack.sh Warnings fixed :
* In POSIX sh, using lower/mixed case for signal names is undefined. Correcting that in this script.
* In POSIX sh, 'local' is undefined.
local is supported in many shells, but it's not in POSIX. In POSIX sh, you can adopt some convention to avoid accidentally overwriting variables names, e.g. prefixing with the function name, that is what I have done here.
--- tools/perf/tests/shell/test_brstack.sh | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
tools/perf/tests/shell/test_brstack.sh +6 -6
@@ -18,7 +18,7 @@ cleanup() { rm -rf $TMPDIR } -trap cleanup exit term int +trap cleanup EXIT TERM INT test_user_branches() { echo "Testing user branch stack sampling"
@@ -47,17 +47,17 @@ test_user_branches() { # first argument <arg0> is the argument passed to "--branch-stack <arg0>,save_type,u" # second argument are the expected branch types for the given filter test_filter() { - local filter=$1 - local expect=$2 + test_filter_filter=$1 + test_filter_expect=$2 - echo "Testing branch stack filtering permutation ($filter,$expect)" + echo "Testing branch stack filtering permutation ($test_filter_filter,$test_filter_expect)" - perf record -o $TMPDIR/perf.data --branch-filter $filter,save_type,u -- ${TESTPROG} > /dev/null 2>&1 + perf record -o $TMPDIR/perf.data --branch-filter $test_filter_filter,save_type,u -- ${TESTPROG} > /dev/null 2>&1 perf script -i $TMPDIR/perf.data --fields brstack | xargs -n1 > $TMPDIR/perf.script # fail if we find any branch type that doesn't match any of the expected ones # also consider UNKNOWN branch types (-) - if grep -E -vm1 "^[^ ]*/($expect|-|( *))/.*$" $TMPDIR/perf.script; then + if grep -E -vm1 "^[^ ]*/($test_filter_expect|-|( *))/.*$" $TMPDIR/perf.script; then return 1 fi }
Signature · 1 line
2.39.1
PATCH 14/17 tools/perf/tests: Fix all POSIX sh warnings in stat+shadow_stat.sh
In reply to Athira RajeevFrom: Spoorthy S <[email protected]>
Running shellcheck -S on stat+shadow_stat.sh testcase, generates SC2046 and SC2034 warnings,
$ shellcheck -S warning tests/shell/stat+shadow_stat.sh res=`printf "%.2f" $(echo "scale=6; $num / $cyc" | bc -q)` : Quote this to prevent word splitting
To address the POSIX shell warnings used quotes in the printf expressions, to prevent word splitting.
--- tools/perf/tests/shell/stat+shadow_stat.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/tests/shell/stat+shadow_stat.sh +2 -2
@@ -33,7 +33,7 @@ test_global_aggr() fi # use printf for rounding and a leading zero - res=`printf "%.2f" $(echo "scale=6; $num / $cyc" | bc -q)` + res=`printf "%.2f" "$(echo "scale=6; $num / $cyc" | bc -q)"` if [ "$ipc" != "$res" ]; then echo "IPC is different: $res != $ipc ($num / $cyc)" exit 1
@@ -67,7 +67,7 @@ test_no_aggr() fi # use printf for rounding and a leading zero - res=`printf "%.2f" $(echo "scale=6; $num / $cyc" | bc -q)` + res=`printf "%.2f" "$(echo "scale=6; $num / $cyc" | bc -q)"` if [ "$ipc" != "$res" ]; then echo "IPC is different for $cpu: $res != $ipc ($num / $cyc)" exit 1
Signature · 1 line
2.39.1
PATCH 15/17 perf tests task_analyzer: fix bad substitution ${$1}
In reply to Athira RajeevFrom: Aditya Gupta <[email protected]>
${$1} gives bad substitution error on sh, bash, and zsh. This seems like a typo, and this patch modifies it to $1, since that is what it's usage looks like from wherever check_exec_0 is called.
This issue due to ${$1} caused all function calls to give error in find_str_or_fail line, and so no test runs completely. But 'perf test "perf script task-analyzer tests"' wrongly reports that tests passed with the status OK, which is wrong considering the tests didn't even run completely
--- tools/perf/tests/shell/test_task_analyzer.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/tests/shell/test_task_analyzer.sh +1 -1
@@ -31,7 +31,7 @@ report() { check_exec_0() { if [ $? != 0 ]; then - report 1 "invokation of ${$1} command failed" + report 1 "invocation of $1 command failed" fi }
Signature · 1 line
2.39.1
Arnaldo Carvalho de Melo
maintainerRe: [PATCH 15/17] perf tests task_analyzer: fix bad substitution ${$1}
Em Tue, Jun 13, 2023 at 10:11:43PM +0530, Athira Rajeev escreveu:
Quoted 5 lines
From: Aditya Gupta <[email protected]>
${$1} gives bad substitution error on sh, bash, and zsh. This seems like a typo, and this patch modifies it to $1, since that is what it's usage looks like from wherever
check_exec_0is called.
Nicely spotted!
Please add the people that last touched the problem to the cc list, specially when it fixes a bug.
Thanks for adding a Fixes tag, that helps the [email protected] guys to get this propagated to supported kernel releases.
I've added the test author to the CC list in this message.
thanks!
- Arnaldo
Quoted 30 lines
This issue due to ${$1} caused all function calls to give error in
find_str_or_failline, and so no test runs completely. But 'perf test "perf script task-analyzer tests"' wrongly reports that tests passed with the status OK, which is wrong considering the tests didn't even run completelyFixes e8478b84d6ba ("perf test: add new task-analyzer tests")Signed-off-by Athira Rajeev <[email protected]>Signed-off-by Kajol Jain <[email protected]>Signed-off-by Aditya Gupta <[email protected]>--- tools/perf/tests/shell/test_task_analyzer.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)tools/perf/tests/shell/test_task_analyzer.sh +1
-1@@ -31,7 +31,7 @@ report() { check_exec_0() { if [ $? != 0 ]; then - report 1 "invokation of ${$1} command failed" + report 1 "invocation of $1 command failed" fi }Signature · 1 line
2.39.1
Signature · 1 line
- Arnaldo
Namhyung Kim
maintainerRe: [PATCH 15/17] perf tests task_analyzer: fix bad substitution ${$1}
Hello,
Quoted 13 lines · Arnaldo Carvalho de Melo
Em Tue, Jun 13, 2023 at 10:11:43PM +0530, Athira Rajeev escreveu:
Quoted 5 lines
From: Aditya Gupta <[email protected]>
${$1} gives bad substitution error on sh, bash, and zsh. This seems like a typo, and this patch modifies it to $1, since that is what it's usage looks like from wherever
check_exec_0is called.Nicely spotted!
Please add the people that last touched the problem to the cc list, specially when it fixes a bug.
Quoted 20 lines
Thanks for adding a Fixes tag, that helps the [email protected] guys to get this propagated to supported kernel releases.
I've added the test author to the CC list in this message.
thanks!
- Arnaldo
Quoted 11 lines
This issue due to ${$1} caused all function calls to give error in
find_str_or_failline, and so no test runs completely. But 'perf test "perf script task-analyzer tests"' wrongly reports that tests passed with the status OK, which is wrong considering the tests didn't even run completelyFixes e8478b84d6ba ("perf test: add new task-analyzer tests")Signed-off-by Athira Rajeev <[email protected]>Signed-off-by Kajol Jain <[email protected]>Signed-off-by Aditya Gupta <[email protected]>---
I'm seeing a different error even after this fix.
Can you please take a look?
Thanks,
Namhyung
$ sudo ./perf test -v task 114: perf script task-analyzer tests : --- start --- test child forked, pid 1771042 Please specify a valid report script(see 'perf script -l' for listing) FAIL: "invocation of perf command failed" Error message: "" FAIL: "test_basic" Error message: "Failed to find required string:'Comm'." Please specify a valid report script(see 'perf script -l' for listing) FAIL: "invocation of perf command failed" Error message: "" FAIL: "test_ns_rename" Error message: "Failed to find required string:'Comm'." ... test child finished with -1 ---- end ---- perf script task-analyzer tests: FAILED!
Re: [PATCH 15/17] perf tests task_analyzer: fix bad substitution ${$1}
Hello Namhyung,
Quoted 23 lines · Namhyung Kim
Hello,
Quoted 12 lines · Arnaldo Carvalho de Melo
Em Tue, Jun 13, 2023 at 10:11:43PM +0530, Athira Rajeev escreveu:
Quoted 11 lines
This issue due to ${$1} caused all function calls to give error in
find_str_or_failline, and so no test runs completely. But 'perf test "perf script task-analyzer tests"' wrongly reports that tests passed with the status OK, which is wrong considering the tests didn't even run completelyFixes e8478b84d6ba ("perf test: add new task-analyzer tests")Signed-off-by Athira Rajeev <[email protected]>Signed-off-by Kajol Jain <[email protected]>Signed-off-by Aditya Gupta <[email protected]>---
I'm seeing a different error even after this fix.
Can you please take a look?Thanks,
Namhyung
Quoted 14 lines
$ sudo ./perf test -v task 114: perf script task-analyzer tests : --- start --- test child forked, pid 1771042 Please specify a valid report script(see 'perf script -l' for listing) FAIL: "invocation of perf command failed" Error message: "" FAIL: "test_basic" Error message: "Failed to find required string:'Comm'." Please specify a valid report script(see 'perf script -l' for listing) FAIL: "invocation of perf command failed" Error message: "" FAIL: "test_ns_rename" Error message: "Failed to find required string:'Comm'." ... test child finished with -1 ---- end ---- perf script task-analyzer tests: FAILED!
Can you please check if your environment has libtraceevent devel libraries (or did you compile with make NO_LIBTRACEEVENT=1) ? When libtraceevent support is not there, perf record fails and so perf.data doesn't contain the strings it's searching for and hence those errors
The error you mentioned has been mentioned and fixed in patch 17/17 of this series.
Thanks
- Aditya
Namhyung Kim
maintainerRe: [PATCH 15/17] perf tests task_analyzer: fix bad substitution ${$1}
Hello Aditya,
Quoted 41 lines · Aditya Gupta
Hello Namhyung,
Quoted 23 lines · Namhyung Kim
Hello,
Quoted 12 lines · Arnaldo Carvalho de Melo
Em Tue, Jun 13, 2023 at 10:11:43PM +0530, Athira Rajeev escreveu:
Quoted 11 lines
This issue due to ${$1} caused all function calls to give error in
find_str_or_failline, and so no test runs completely. But 'perf test "perf script task-analyzer tests"' wrongly reports that tests passed with the status OK, which is wrong considering the tests didn't even run completelyFixes e8478b84d6ba ("perf test: add new task-analyzer tests")Signed-off-by Athira Rajeev <[email protected]>Signed-off-by Kajol Jain <[email protected]>Signed-off-by Aditya Gupta <[email protected]>---
I'm seeing a different error even after this fix.
Can you please take a look?Thanks,
NamhyungQuoted 14 lines
$ sudo ./perf test -v task 114: perf script task-analyzer tests : --- start --- test child forked, pid 1771042 Please specify a valid report script(see 'perf script -l' for listing) FAIL: "invocation of perf command failed" Error message: "" FAIL: "test_basic" Error message: "Failed to find required string:'Comm'." Please specify a valid report script(see 'perf script -l' for listing) FAIL: "invocation of perf command failed" Error message: "" FAIL: "test_ns_rename" Error message: "Failed to find required string:'Comm'." ... test child finished with -1 ---- end ---- perf script task-analyzer tests: FAILED!
Quoted 7 lines
Can you please check if your environment has libtraceevent devel libraries (or did you compile with
make NO_LIBTRACEEVENT=1) ? When libtraceevent support is not there, perf record fails and so perf.data doesn't contain the strings it's searching for and hence those errorsThe error you mentioned has been mentioned and fixed in patch 17/17 of this series.
Thanks for your reply but It has libtraceevent. Also, shouldn't it skip if it's not?
Thanks,
Namhyung
Re: [PATCH 15/17] perf tests task_analyzer: fix bad substitution ${$1}
Hello Namhyung,
Quoted 27 lines · Namhyung Kim
Hello Aditya,
Quoted 24 lines · Aditya Gupta
Hello Namhyung,
Quoted 16 lines · Namhyung Kim
...
$ sudo ./perf test -v task 114: perf script task-analyzer tests : --- start --- test child forked, pid 1771042 Please specify a valid report script(see 'perf script -l' for listing) FAIL: "invocation of perf command failed" Error message: "" FAIL: "test_basic" Error message: "Failed to find required string:'Comm'." Please specify a valid report script(see 'perf script -l' for listing) FAIL: "invocation of perf command failed" Error message: "" FAIL: "test_ns_rename" Error message: "Failed to find required string:'Comm'." ... test child finished with -1 ---- end ---- perf script task-analyzer tests: FAILED!Can you please check if your environment has libtraceevent devel libraries (or did you compile with
make NO_LIBTRACEEVENT=1) ? When libtraceevent support is not there, perf record fails and so perf.data doesn't contain the strings it's searching for and hence those errors
Quoted 7 lines
Quoted 2 lines
The error you mentioned has been mentioned and fixed in patch 17/17 of this series.
Thanks for your reply but It has libtraceevent. Also, shouldn't it skip if it's not?
Thanks,
Namhyung
The skipping is handled in the 17th patch in this series, and considering that patch has also been applied, it will skip the tests if perf wasn't built with proper libtraceevent support.
Back to the error, Sorry but I tested again on my system and am unable to reproduce the issue you are seeing when built with libtraceevent support.
This is what I did:
0. git clone --depth=1 https://github.com/torvalds/linux
0. cd linux/tools/perf
0. git am patch_15/17.patch
Quoted 1 line
I applied ONLY this patch (15/17) of this series, to a fresh linux tree
1. dnf install libtraceevent-devel
2. make clean && make
3. sudo ./perf test -v "perf script task-analyzer tests" # Working fine, tests passed
4. dnf remove libtraceevent-devel
5. make clean && make # There will also be a warning during build: "libtraceevent is missing limiting functionality"
6. sudo ./perf test -v "perf script task-analyzer tests" # Fails with the error you posted, which was the case till now, it's skipped when the 17th patch is also applied and perf built without libtraceevent support
The error in the second case (without proper libtraceevent support) is expected, as it was the case till now, that is fixed by the 17th patch, try applying that also and build perf with make NO_LIBTRACEEVENT=1, it will skip then.
Can you guide me with the steps to reproduce the error ?
Sidenote: Just in case, please ensure you are running the perf as root here as perf record -e sched:sched_switch -a -- sleep 1 requires root, which has been used in prepare_perf_data
Thanks,
- Aditya
Namhyung Kim
maintainerRe: [PATCH 15/17] perf tests task_analyzer: fix bad substitution ${$1}
Hello Aditya,
Quoted 36 lines · Aditya Gupta
Hello Namhyung,
Quoted 32 lines · Namhyung Kim
Hello Aditya,
Quoted 19 lines · Aditya Gupta
Hello Namhyung,
Quoted 16 lines · Namhyung Kim
...
$ sudo ./perf test -v task 114: perf script task-analyzer tests : --- start --- test child forked, pid 1771042 Please specify a valid report script(see 'perf script -l' for listing) FAIL: "invocation of perf command failed" Error message: "" FAIL: "test_basic" Error message: "Failed to find required string:'Comm'." Please specify a valid report script(see 'perf script -l' for listing) FAIL: "invocation of perf command failed" Error message: "" FAIL: "test_ns_rename" Error message: "Failed to find required string:'Comm'." ... test child finished with -1 ---- end ---- perf script task-analyzer tests: FAILED!Quoted 7 lines
Can you please check if your environment has libtraceevent devel libraries (or did you compile with
make NO_LIBTRACEEVENT=1) ? When libtraceevent support is not there, perf record fails and so perf.data doesn't contain the strings it's searching for and hence those errorsThe error you mentioned has been mentioned and fixed in patch 17/17 of this series.
Thanks for your reply but It has libtraceevent. Also, shouldn't it skip if it's not?
Quoted 6 lines
Quoted 2 lines
Thanks,
NamhyungThe skipping is handled in the 17th patch in this series, and considering that patch has also been applied, it will skip the tests if perf wasn't built with proper libtraceevent support.
Right, I tested it with the whole series applied. It skipped to run when I disable libtraceevent build.
Quoted 15 lines
Back to the error, Sorry but I tested again on my system and am unable to reproduce the issue you are seeing when built with libtraceevent support.
This is what I did:
0. git clone --depth=1 https://github.com/torvalds/linux
0. cd linux/tools/perf
0. git am patch_15/17.patchQuoted 1 line
I applied ONLY this patch (15/17) of this series, to a fresh linux tree
1. dnf install libtraceevent-devel
2. make clean && make
3. sudo ./perf test -v "perf script task-analyzer tests" # Working fine, tests passed
Quoted 7 lines
4. dnf remove libtraceevent-devel
5. make clean && make # There will also be a warning during build: "libtraceevent is missing limiting functionality"
6. sudo ./perf test -v "perf script task-analyzer tests" # Fails with the error you posted, which was the case till now, it's skipped when the 17th patch is also applied and perf built without libtraceevent supportThe error in the second case (without proper libtraceevent support) is expected, as it was the case till now, that is fixed by the 17th patch, try applying that also and build perf with
make NO_LIBTRACEEVENT=1, it will skip then.Can you guide me with the steps to reproduce the error ?
You can try the perf-tools-next branch in the perf/perft-tools-next.git repo.
git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git
$ make clean all $ sudo ./perf test -v task
Quoted 2 lines
Sidenote: Just in case, please ensure you are running the perf as root here as
perf record -e sched:sched_switch -a -- sleep 1requires root, which has been used inprepare_perf_data
Sure, I ran the perf test as root.
Thanks,
Namhyung
Namhyung Kim
maintainerRe: [PATCH 15/17] perf tests task_analyzer: fix bad substitution ${$1}
Ok, I found two problems.
Quoted 21 lines · Namhyung Kim
Hello Aditya,
Quoted 17 lines · Aditya Gupta
Hello Namhyung,
Quoted 13 lines · Namhyung Kim
Hello Aditya,
Quoted 10 lines · Aditya Gupta
Hello Namhyung,
Quoted 7 lines · Namhyung Kim
...
$ sudo ./perf test -v task 114: perf script task-analyzer tests : --- start --- test child forked, pid 1771042 Please specify a valid report script(see 'perf script -l' for listing)
The first one is related to this message. It couldn't find the script (task-analyzer.py) because PERF_EXEC_PATH is not set. Running with --exec-path=$PWD was ok, but I got a segfault.
The other problem is in set_regs_in_dict(). It tries to capture register values in the sample and save them to a dictionary. The sample doesn't have registers so it should have no problem. But the 'bf' was not initialized properly when size is 0, and it led PyUnicode_FromString() returning NULL.
After the changes, it ran ok:
$ sudo ./perf test -v task 116: perf script task-analyzer tests : --- start --- test child forked, pid 204088 PASS: "test_basic" PASS: "test_ns_rename" PASS: "test_ms_filtertasks_highlight" PASS: "test_extended_times_timelimit_limittasks" PASS: "test_summary" PASS: "test_summaryextended" PASS: "test_summaryonly" PASS: "test_extended_times_summary_ns" PASS: "test_extended_times_summary_ns" PASS: "test_csv" PASS: "test_csvsummary" PASS: "test_csv_extended_times" PASS: "test_csvsummary_extended" test child finished with 0 ---- end ---- perf script task-analyzer tests: Ok
I'll send the fixes soon.
Thanks,
Namhyung
Quoted 66 lines
Quoted 19 lines
Quoted 19 lines
Quoted 9 lines
Quoted 9 lines
FAIL: "invocation of perf command failed" Error message: ""
FAIL: "test_basic" Error message: "Failed to find required string:'Comm'." Please specify a valid report script(see 'perf script -l' for listing)
FAIL: "invocation of perf command failed" Error message: ""
FAIL: "test_ns_rename" Error message: "Failed to find required string:'Comm'."
...
test child finished with -1
---- end ----
perf script task-analyzer tests: FAILED!Quoted 7 lines
Can you please check if your environment has libtraceevent devel libraries (or did you compile with
make NO_LIBTRACEEVENT=1) ? When libtraceevent support is not there, perf record fails and so perf.data doesn't contain the strings it's searching for and hence those errorsThe error you mentioned has been mentioned and fixed in patch 17/17 of this series.
Thanks for your reply but It has libtraceevent. Also, shouldn't it skip if it's not?
Quoted 6 lines
Quoted 2 lines
Thanks,
NamhyungThe skipping is handled in the 17th patch in this series, and considering that patch has also been applied, it will skip the tests if perf wasn't built with proper libtraceevent support.
Right, I tested it with the whole series applied. It skipped to run when I disable libtraceevent build.
Quoted 15 lines
Back to the error, Sorry but I tested again on my system and am unable to reproduce the issue you are seeing when built with libtraceevent support.
This is what I did:
0. git clone --depth=1 https://github.com/torvalds/linux
0. cd linux/tools/perf
0. git am patch_15/17.patchQuoted 1 line
I applied ONLY this patch (15/17) of this series, to a fresh linux tree
1. dnf install libtraceevent-devel
2. make clean && make
3. sudo ./perf test -v "perf script task-analyzer tests" # Working fine, tests passedQuoted 7 lines
4. dnf remove libtraceevent-devel
5. make clean && make # There will also be a warning during build: "libtraceevent is missing limiting functionality"
6. sudo ./perf test -v "perf script task-analyzer tests" # Fails with the error you posted, which was the case till now, it's skipped when the 17th patch is also applied and perf built without libtraceevent supportThe error in the second case (without proper libtraceevent support) is expected, as it was the case till now, that is fixed by the 17th patch, try applying that also and build perf with
make NO_LIBTRACEEVENT=1, it will skip then.Can you guide me with the steps to reproduce the error ?
You can try the perf-tools-next branch in the perf/perft-tools-next.git repo.
git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git$ make clean all $ sudo ./perf test -v taskQuoted 2 lines
Sidenote: Just in case, please ensure you are running the perf as root here as
perf record -e sched:sched_switch -a -- sleep 1requires root, which has been used inprepare_perf_dataSure, I ran the perf test as root.
Thanks,
Namhyung
Re: [PATCH 15/17] perf tests task_analyzer: fix bad substitution ${$1}
Hello, Namhyung,
Quoted 28 lines · Namhyung Kim
Ok, I found two problems.
...
The first one is related to this message. It couldn't find the script (
task-analyzer.py) becausePERF_EXEC_PATHis not set. Running with --exec-path=$PWD was ok, but I got a segfault.The other problem is in
set_regs_in_dict(). It tries to capture register values in the sample and save them to a dictionary. The sample doesn't have registers so it should have no problem. But the 'bf' was not initialized properly when size is 0, and it ledPyUnicode_FromString()returning NULL.After the changes, it ran ok:
$ sudo ./perf test -v task 116: perf script task-analyzer tests : --- start --- test child forked, pid 204088 PASS: "test_basic" PASS: "test_ns_rename" PASS: "test_ms_filtertasks_highlight" PASS: "test_extended_times_timelimit_limittasks" PASS: "test_summary" PASS: "test_summaryextended" PASS: "test_summaryonly" PASS: "test_extended_times_summary_ns"
Quoted 8 lines
PASS: "test_extended_times_summary_ns" PASS: "test_csv" PASS: "test_csvsummary" PASS: "test_csv_extended_times" PASS: "test_csvsummary_extended" test child finished with 0 ---- end ---- perf script task-analyzer tests: Ok
This is interesting. I did not encounter these earlier.
Quoted 2 lines
I'll send the fixes soon.
Thanks for fixing it.
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
You can try the perf-tools-next branch in the perf/perft-tools-next.git repo.
git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git$ make clean all $ sudo ./perf test -v task
I tested with the steps provided, i.e.. clone, /*apply patch 15 (added this step myself)*/, clean, make, ./perf test -v test. But still- test passed:
'''
~/temp_clones/perf-tools-next/tools/perf git:(master) sudo ./perf test -v test
...
--- start ---
test child forked, pid 75261
PASS: "test_basic"
PASS: "test_ns_rename"
PASS: "test_ms_filtertasks_highlight"
PASS: "test_extended_times_timelimit_limittasks"
PASS: "test_summary"
PASS: "test_summaryextended"
PASS: "test_summaryonly"
PASS: "test_extended_times_summary_ns"
PASS: "test_extended_times_summary_ns"
PASS: "test_csv"
PASS: "test_csvsummary"
PASS: "test_csv_extended_times"
PASS: "test_csvsummary_extended"
test child finished with 0
---- end ----
'''
Maybe my environment had that other things set in a way that, I did not face any issue (I don't recall doing any special thing with my environment though, and these patches have been tested multiple times by more people also). But thanks for your efforts to debug and fix the issue.
Thanks,
- AdityaG
Namhyung Kim
maintainerRe: [PATCH 15/17] perf tests task_analyzer: fix bad substitution ${$1}
Hello Aditya,
Quoted 40 lines · Aditya Gupta
Hello, Namhyung,
Quoted 28 lines · Namhyung Kim
Ok, I found two problems.
...
The first one is related to this message. It couldn't find the script (
task-analyzer.py) becausePERF_EXEC_PATHis not set. Running with --exec-path=$PWD was ok, but I got a segfault.The other problem is in
set_regs_in_dict(). It tries to capture register values in the sample and save them to a dictionary. The sample doesn't have registers so it should have no problem. But the 'bf' was not initialized properly when size is 0, and it ledPyUnicode_FromString()returning NULL.After the changes, it ran ok:
$ sudo ./perf test -v task 116: perf script task-analyzer tests : --- start --- test child forked, pid 204088 PASS: "test_basic" PASS: "test_ns_rename" PASS: "test_ms_filtertasks_highlight" PASS: "test_extended_times_timelimit_limittasks" PASS: "test_summary" PASS: "test_summaryextended" PASS: "test_summaryonly" PASS: "test_extended_times_summary_ns"Quoted 8 lines
PASS: "test_extended_times_summary_ns" PASS: "test_csv" PASS: "test_csvsummary" PASS: "test_csv_extended_times" PASS: "test_csvsummary_extended" test child finished with 0 ---- end ---- perf script task-analyzer tests: Ok
Quoted 5 lines
This is interesting. I did not encounter these earlier.
Quoted 2 lines
I'll send the fixes soon.
Thanks for fixing it.
No problem.
Quoted 37 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
You can try the perf-tools-next branch in the perf/perft-tools-next.git repo.
git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git$ make clean all $ sudo ./perf test -v taskI tested with the steps provided, i.e.. clone, /*apply patch 15 (added this step myself)*/, clean, make, ./perf test -v test. But still- test passed:
'''
~/temp_clones/perf-tools-next/tools/perf git:(master) sudo ./perf test -v test...
--- start ---
test child forked, pid 75261
PASS: "test_basic"
PASS: "test_ns_rename"
PASS: "test_ms_filtertasks_highlight"
PASS: "test_extended_times_timelimit_limittasks"
PASS: "test_summary"
PASS: "test_summaryextended"
PASS: "test_summaryonly"
PASS: "test_extended_times_summary_ns"
PASS: "test_extended_times_summary_ns"
PASS: "test_csv"
PASS: "test_csvsummary"
PASS: "test_csv_extended_times"
PASS: "test_csvsummary_extended"
test child finished with 0
---- end ----
'''
Quoted 5 lines
Maybe my environment had that other things set in a way that, I did not face any issue (I don't recall doing any special thing with my environment though, and these patches have been tested multiple times by more people also).
But thanks for your efforts to debug and fix the issue.
I don't know but you might either set PERF_EXEC_PATH env or install the scripts somewhere that can be found by default.
Thanks,
Namhyung
Re: [PATCH 15/17] perf tests task_analyzer: fix bad substitution ${$1}
In reply to Namhyung KimHello Namhyung,
Quoted 27 lines · Namhyung Kim
Ok, I found two problems.
...
The first one is related to this message. It couldn't find the script (task-analyzer.py) becausePERF_EXEC_PATHis not set. Running with --exec-path=$PWD was ok, but I got a segfault.The other problem is in
set_regs_in_dict(). It tries to capture register values in the sample and save them to a dictionary. The sample doesn't have registers so it should have no problem. But the 'bf' was not initialized properly when size is 0, and it ledPyUnicode_FromString()returning NULL.After the changes, it ran ok:
$ sudo ./perf test -v task 116: perf script task-analyzer tests : --- start --- test child forked, pid 204088 PASS: "test_basic" PASS: "test_ns_rename" PASS: "test_ms_filtertasks_highlight" PASS: "test_extended_times_timelimit_limittasks" PASS: "test_summary" PASS: "test_summaryextended" PASS: "test_summaryonly" PASS: "test_extended_times_summary_ns"
Quoted 9 lines
PASS: "test_extended_times_summary_ns" PASS: "test_csv" PASS: "test_csvsummary" PASS: "test_csv_extended_times" PASS: "test_csvsummary_extended" test child finished with 0 ---- end ---- perf script task-analyzer tests: Ok
This is interesting. I did not encounter these earlier.
Quoted 2 lines
I'll send the fixes soon.
Thanks for fixing it.
Quoted 7 lines
Quoted 7 lines
You can try the perf-tools-next branch in the perf/perft-tools-next.git repo.
git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git$ make clean all $ sudo ./perf test -v task
I tested with the steps provided, i.e.. clone, /*apply patch 15 (added this step myself)*/, clean, make, ./perf test -v test. But still- test passed:
'''
~/temp_clones/perf-tools-next/tools/perf git:(master) sudo ./perf test -v test
...
--- start ---
test child forked, pid 75261
PASS: "test_basic"
PASS: "test_ns_rename"
PASS: "test_ms_filtertasks_highlight"
PASS: "test_extended_times_timelimit_limittasks"
PASS: "test_summary"
PASS: "test_summaryextended"
PASS: "test_summaryonly"
PASS: "test_extended_times_summary_ns"
PASS: "test_extended_times_summary_ns"
PASS: "test_csv"
PASS: "test_csvsummary"
PASS: "test_csv_extended_times"
PASS: "test_csvsummary_extended"
test child finished with 0
---- end ----
'''
Maybe my environment had that other things set in such a way that, I did not face any issue (I don't recall doing so though, and these patches have been tested multiple times by more people also). But thanks for your efforts to debug and fix the issue.
Thanks,
- AdityaG
Re: [PATCH 15/17] perf tests task_analyzer: fix bad substitution ${$1}
In reply to Namhyung KimHello Namhyung,
Quoted 27 lines · Namhyung Kim
Ok, I found two problems.
...
The first one is related to this message. It couldn't find the script (task-analyzer.py) becausePERF_EXEC_PATHis not set. Running with --exec-path=$PWD was ok, but I got a segfault.The other problem is in
set_regs_in_dict(). It tries to capture register values in the sample and save them to a dictionary. The sample doesn't have registers so it should have no problem. But the 'bf' was not initialized properly when size is 0, and it ledPyUnicode_FromString()returning NULL.After the changes, it ran ok:
$ sudo ./perf test -v task 116: perf script task-analyzer tests : --- start --- test child forked, pid 204088 PASS: "test_basic" PASS: "test_ns_rename" PASS: "test_ms_filtertasks_highlight" PASS: "test_extended_times_timelimit_limittasks" PASS: "test_summary" PASS: "test_summaryextended" PASS: "test_summaryonly" PASS: "test_extended_times_summary_ns"
Quoted 9 lines
PASS: "test_extended_times_summary_ns" PASS: "test_csv" PASS: "test_csvsummary" PASS: "test_csv_extended_times" PASS: "test_csvsummary_extended" test child finished with 0 ---- end ---- perf script task-analyzer tests: Ok
This is interesting. I did not encounter these earlier.
Quoted 2 lines
I'll send the fixes soon.
Thanks for fixing it.
Quoted 7 lines
Quoted 7 lines
You can try the perf-tools-next branch in the perf/perft-tools-next.git repo.
git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git$ make clean all $ sudo ./perf test -v task
I tested with the steps provided, i.e.. clone, /*apply patch 15 (added this step myself)*/, clean, make, ./perf test -v test. But still- test passed:
'''
~/temp_clones/perf-tools-next/tools/perf git:(master) sudo ./perf test -v test
...
--- start ---
test child forked, pid 75261
PASS: "test_basic"
PASS: "test_ns_rename"
PASS: "test_ms_filtertasks_highlight"
PASS: "test_extended_times_timelimit_limittasks"
PASS: "test_summary"
PASS: "test_summaryextended"
PASS: "test_summaryonly"
PASS: "test_extended_times_summary_ns"
PASS: "test_extended_times_summary_ns"
PASS: "test_csv"
PASS: "test_csvsummary"
PASS: "test_csv_extended_times"
PASS: "test_csvsummary_extended"
test child finished with 0
---- end ----
'''
Maybe my environment had that other things set in such a way that, I did not face any issue (I don't recall doing so though, and these patches have been tested multiple times by more people also). But thanks, that you debugged this and fixed the issue also, I was unsure of the reason since I did not encounter it.
Thanks,
- AdityaG
PATCH 16/17 perf tests task_analyzer: print command on failure
In reply to Athira RajeevFrom: Aditya Gupta <[email protected]>
Instead of printing "perf command failed" everytime, print the exact command that run earlier
--- tools/perf/tests/shell/test_task_analyzer.sh | 24 ++++++++++---------- 1 file changed, 12 insertions(+), 12 deletions(-)
tools/perf/tests/shell/test_task_analyzer.sh +12 -12
@@ -53,14 +53,14 @@ prepare_perf_data() { test_basic() { out="$tmpdir/perf.out" perf script report task-analyzer > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer" find_str_or_fail "Comm" "$out" "${FUNCNAME[0]}" } test_ns_rename(){ out="$tmpdir/perf.out" perf script report task-analyzer --ns --rename-comms-by-tids 0:random > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --ns --rename-comms-by-tids 0:random" find_str_or_fail "Comm" "$out" "${FUNCNAME[0]}" }
@@ -68,7 +68,7 @@ test_ms_filtertasks_highlight(){ out="$tmpdir/perf.out" perf script report task-analyzer --ms --filter-tasks perf --highlight-tasks perf \ > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --ms --filter-tasks perf --highlight-tasks perf" find_str_or_fail "Comm" "$out" "${FUNCNAME[0]}" }
@@ -76,61 +76,61 @@ test_extended_times_timelimit_limittasks() { out="$tmpdir/perf.out" perf script report task-analyzer --extended-times --time-limit :99999 \ --limit-to-tasks perf > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --extended-times --time-limit :99999 --limit-to-tasks perf" find_str_or_fail "Out-Out" "$out" "${FUNCNAME[0]}" } test_summary() { out="$tmpdir/perf.out" perf script report task-analyzer --summary > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --summary" find_str_or_fail "Summary" "$out" "${FUNCNAME[0]}" } test_summaryextended() { out="$tmpdir/perf.out" perf script report task-analyzer --summary-extended > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --summary-extended" find_str_or_fail "Inter Task Times" "$out" "${FUNCNAME[0]}" } test_summaryonly() { out="$tmpdir/perf.out" perf script report task-analyzer --summary-only > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --summary-only" find_str_or_fail "Summary" "$out" "${FUNCNAME[0]}" } test_extended_times_summary_ns() { out="$tmpdir/perf.out" perf script report task-analyzer --extended-times --summary --ns > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --extended-times --summary --ns" find_str_or_fail "Out-Out" "$out" "${FUNCNAME[0]}" find_str_or_fail "Summary" "$out" "${FUNCNAME[0]}" } test_csv() { perf script report task-analyzer --csv csv > /dev/null - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --csv csv" find_str_or_fail "Comm;" csv "${FUNCNAME[0]}" } test_csv_extended_times() { perf script report task-analyzer --csv csv --extended-times > /dev/null - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --csv csv --extended-times" find_str_or_fail "Out-Out;" csv "${FUNCNAME[0]}" } test_csvsummary() { perf script report task-analyzer --csv-summary csvsummary > /dev/null - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --csv-summary csvsummary" find_str_or_fail "Comm;" csvsummary "${FUNCNAME[0]}" } test_csvsummary_extended() { perf script report task-analyzer --csv-summary csvsummary --summary-extended \ >/dev/null - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --csv-summary csvsummary --summary-extended" find_str_or_fail "Out-Out;" csvsummary "${FUNCNAME[0]}" }
Signature · 1 line
2.39.1
Re: [PATCH 16/17] perf tests task_analyzer: print command on failure
Em Tue, Jun 13, 2023 at 10:11:44PM +0530, Athira Rajeev escreveu:
Quoted 4 lines
From: Aditya Gupta <[email protected]>
Instead of printing "perf command failed" everytime, print the exact command that run earlier
Looks like a nice improvement, added the test authors to the CC list on this message,
Thanks, applied.
- Arnaldo
Quoted 111 lines
Signed-off-by Athira Rajeev <[email protected]>Signed-off-by Kajol Jain <[email protected]>Signed-off-by Aditya Gupta <[email protected]>--- tools/perf/tests/shell/test_task_analyzer.sh | 24 ++++++++++---------- 1 file changed, 12 insertions(+), 12 deletions(-)tools/perf/tests/shell/test_task_analyzer.sh +12
-12@@ -53,14 +53,14 @@ prepare_perf_data() { test_basic() { out="$tmpdir/perf.out" perf script report task-analyzer > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer" find_str_or_fail "Comm" "$out" "${FUNCNAME[0]}" } test_ns_rename(){ out="$tmpdir/perf.out" perf script report task-analyzer --ns --rename-comms-by-tids 0:random > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --ns --rename-comms-by-tids 0:random" find_str_or_fail "Comm" "$out" "${FUNCNAME[0]}" }@@ -68,7 +68,7 @@ test_ms_filtertasks_highlight(){ out="$tmpdir/perf.out" perf script report task-analyzer --ms --filter-tasks perf --highlight-tasks perf \ > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --ms --filter-tasks perf --highlight-tasks perf" find_str_or_fail "Comm" "$out" "${FUNCNAME[0]}" }@@ -76,61 +76,61 @@ test_extended_times_timelimit_limittasks() { out="$tmpdir/perf.out" perf script report task-analyzer --extended-times --time-limit :99999 \ --limit-to-tasks perf > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --extended-times --time-limit :99999 --limit-to-tasks perf" find_str_or_fail "Out-Out" "$out" "${FUNCNAME[0]}" } test_summary() { out="$tmpdir/perf.out" perf script report task-analyzer --summary > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --summary" find_str_or_fail "Summary" "$out" "${FUNCNAME[0]}" } test_summaryextended() { out="$tmpdir/perf.out" perf script report task-analyzer --summary-extended > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --summary-extended" find_str_or_fail "Inter Task Times" "$out" "${FUNCNAME[0]}" } test_summaryonly() { out="$tmpdir/perf.out" perf script report task-analyzer --summary-only > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --summary-only" find_str_or_fail "Summary" "$out" "${FUNCNAME[0]}" } test_extended_times_summary_ns() { out="$tmpdir/perf.out" perf script report task-analyzer --extended-times --summary --ns > "$out" - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --extended-times --summary --ns" find_str_or_fail "Out-Out" "$out" "${FUNCNAME[0]}" find_str_or_fail "Summary" "$out" "${FUNCNAME[0]}" } test_csv() { perf script report task-analyzer --csv csv > /dev/null - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --csv csv" find_str_or_fail "Comm;" csv "${FUNCNAME[0]}" } test_csv_extended_times() { perf script report task-analyzer --csv csv --extended-times > /dev/null - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --csv csv --extended-times" find_str_or_fail "Out-Out;" csv "${FUNCNAME[0]}" } test_csvsummary() { perf script report task-analyzer --csv-summary csvsummary > /dev/null - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --csv-summary csvsummary" find_str_or_fail "Comm;" csvsummary "${FUNCNAME[0]}" } test_csvsummary_extended() { perf script report task-analyzer --csv-summary csvsummary --summary-extended \ >/dev/null - check_exec_0 "perf" + check_exec_0 "perf script report task-analyzer --csv-summary csvsummary --summary-extended" find_str_or_fail "Out-Out;" csvsummary "${FUNCNAME[0]}" }Signature · 1 line
2.39.1
Signature · 1 line
- Arnaldo
Hagen Paul Pfeifer
Acked-byRe: [PATCH 16/17] perf tests task_analyzer: print command on failure
Quoted 7 lines · * Arnaldo Carvalho de Melo | 2023-06-13 17:07:52 [-0300]:
Quoted 4 lines
From: Aditya Gupta <[email protected]>
Instead of printing "perf command failed" everytime, print the exact command that run earlier
Looks like a nice improvement, added the test authors to the CC list on this message,
Thank you Aditya and really great catch - thank you very much! Shell scripting can be nasty. This is another example where you can see that with one stupid typo the whole system can be undermined and you don't get a single hint.
PATCH 17/17 perf tests task_analyzer: skip tests if no libtraceevent support
In reply to Athira RajeevFrom: Aditya Gupta <[email protected]>
Test "perf script task-analyzer tests" fails in environment with missing libtraceevent support, as perf record fails to create the perf.data file, which further tests depend on.
Instead, when perf is not compiled with libtraceevent support, skip those tests instead of failing them, by checking the output of perf record --dry-run to see if it prints the error "libtraceevent is necessary for tracepoint support"
For the following output, perf compiled with: make NO_LIBTRACEEVENT=1
Before the patch:
108: perf script task-analyzer tests :
test child forked, pid 24105
failed to open perf.data: No such file or directory (try 'perf record' first)
FAIL: "invokation of perf script report task-analyzer command failed" Error message: ""
FAIL: "test_basic" Error message: "Failed to find required string:'Comm'."
failed to open perf.data: No such file or directory (try 'perf record' first)
FAIL: "invokation of perf script report task-analyzer --ns --rename-comms-by-tids 0:random command failed" Error message: ""
FAIL: "test_ns_rename" Error message: "Failed to find required string:'Comm'."
failed to open perf.data: No such file or directory (try 'perf record' first)
<...>
perf script task-analyzer tests: FAILED!
With this patch, the script instead returns 2 signifying SKIP, and after the patch:
108: perf script task-analyzer tests :
test child forked, pid 26010
libtraceevent is necessary for tracepoint support
WARN: Skipping tests. No libtraceevent support
test child finished with -2
perf script task-analyzer tests: Skip
--- tools/perf/tests/shell/test_task_analyzer.sh | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+)
tools/perf/tests/shell/test_task_analyzer.sh +18 -0
@@ -44,9 +44,20 @@ find_str_or_fail() { fi } +# check if perf is compiled with libtraceevent support +skip_no_probe_record_support() { + perf record -e "sched:sched_switch" -a -- sleep 1 2>&1 | grep "libtraceevent is necessary for tracepoint support" && return 2 + return 0 +} + prepare_perf_data() { # 1s should be sufficient to catch at least some switches perf record -e sched:sched_switch -a -- sleep 1 > /dev/null 2>&1 + # check if perf data file got created in above step. + if [ ! -e "perf.data" ]; then + printf "FAIL: perf record failed to create \"perf.data\" \n" + return 1 + fi } # check standard inkvokation with no arguments
@@ -134,6 +145,13 @@ test_csvsummary_extended() { find_str_or_fail "Out-Out;" csvsummary "${FUNCNAME[0]}" } +skip_no_probe_record_support +err=$? +if [ $err -ne 0 ]; then + echo "WARN: Skipping tests. No libtraceevent support" + cleanup + exit $err +fi prepare_perf_data test_basic test_ns_rename
Signature · 1 line
2.39.1
Ian Rogers
reviewerRe: [PATCH 17/17] perf tests task_analyzer: skip tests if no libtraceevent support
Quoted 51 lines · Athira Rajeev
From: Aditya Gupta <[email protected]>
Test "perf script task-analyzer tests" fails in environment with missing libtraceevent support, as perf record fails to create the perf.data file, which further tests depend on.
Instead, when perf is not compiled with libtraceevent support, skip those tests instead of failing them, by checking the output of
perf record --dry-runto see if it prints the error "libtraceevent is necessary for tracepoint support"For the following output, perf compiled with:
make NO_LIBTRACEEVENT=1Before the patch:
108: perf script task-analyzer tests :
test child forked, pid 24105
failed to open perf.data: No such file or directory (try 'perf record' first)
FAIL: "invokation of perf script report task-analyzer command failed" Error message: ""
FAIL: "test_basic" Error message: "Failed to find required string:'Comm'."
failed to open perf.data: No such file or directory (try 'perf record' first)
FAIL: "invokation of perf script report task-analyzer --ns --rename-comms-by-tids 0:random command failed" Error message: ""
FAIL: "test_ns_rename" Error message: "Failed to find required string:'Comm'."
failed to open perf.data: No such file or directory (try 'perf record' first)
<...>
perf script task-analyzer tests: FAILED!With this patch, the script instead returns 2 signifying SKIP, and after the patch:
108: perf script task-analyzer tests :
test child forked, pid 26010
libtraceevent is necessary for tracepoint support
WARN: Skipping tests. No libtraceevent support
test child finished with -2
perf script task-analyzer tests: SkipFixes e8478b84d6ba ("perf test: add new task-analyzer tests")Signed-off-by Athira Rajeev <[email protected]>Signed-off-by Kajol Jain <[email protected]>Signed-off-by Aditya Gupta <[email protected]>--- tools/perf/tests/shell/test_task_analyzer.sh | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+)tools/perf/tests/shell/test_task_analyzer.sh +0
-0
tools/perf/tests/shell/test_task_analyzer.sh find_str_or_fail() { · line 49
Quoted 6 lines
tools/perf/tests/shell/test_task_analyzer.sh +3
-0@@ -44,9 +44,20 @@ find_str_or_fail() { fi } +# check if perf is compiled with libtraceevent support +skip_no_probe_record_support() { + perf record -e "sched:sched_switch" -a -- sleep 1 2>&1 | grep "libtraceevent is necessary for tracepoint support" && return 2
Fwiw, another way to detect build options used in other shell tests is:
perf version --build-options | grep HAVE_LIBTRACEEVENT | grep -q OFF && return 2
Thanks,
Ian
Quoted 31 lines
tools/perf/tests/shell/test_task_analyzer.sh +15
-1+ return 0 +} + prepare_perf_data() { # 1s should be sufficient to catch at least some switches perf record -e sched:sched_switch -a -- sleep 1 > /dev/null 2>&1 + # check if perf data file got created in above step. + if [ ! -e "perf.data" ]; then + printf "FAIL: perf record failed to create \"perf.data\" \n" + return 1 + fi } # check standard inkvokation with no arguments@@ -134,6 +145,13 @@ test_csvsummary_extended() { find_str_or_fail "Out-Out;" csvsummary "${FUNCNAME[0]}" } +skip_no_probe_record_support +err=$? +if [ $err -ne 0 ]; then + echo "WARN: Skipping tests. No libtraceevent support" + cleanup + exit $err +fi prepare_perf_data test_basic test_ns_rename --2.39.1
Re: [PATCH 17/17] perf tests task_analyzer: skip tests if no libtraceevent support
Hello Ian,
Quoted 23 lines · Ian Rogers
Quoted 14 lines · Athira Rajeev
From: Aditya Gupta <[email protected]>
Test "perf script task-analyzer tests" fails in environment with missing libtraceevent support, as perf record fails to create the perf.data file, which further tests depend on.
Instead, when perf is not compiled with libtraceevent support, skip those tests instead of failing them, by checking the output of
perf record --dry-runto see if it prints the error "libtraceevent is necessary for tracepoint support"...
tools/perf/tests/shell/test_task_analyzer.sh find_str_or_fail() { · line 49
Quoted 3 lines
tools/perf/tests/shell/test_task_analyzer.sh +3
-0+# check if perf is compiled with libtraceevent support +skip_no_probe_record_support() { + perf record -e "sched:sched_switch" -a -- sleep 1 2>&1 | grep "libtraceevent is necessary for tracepoint support" && return 2Fwiw, another way to detect build options used in other shell tests is:
perf version --build-options | grepHAVE_LIBTRACEEVENT| grep -q OFF && return 2
Quoted 3 lines
Thanks,
Ian
Thanks for the review. That seems more straightforward way to check for libtraceevent support. I will work on a patch for this, and test it.
Thanks,
Aditya G
Em Tue, Jun 13, 2023 at 10:11:28PM +0530, Athira Rajeev escreveu:
Quoted 3 lines
Patchset covers a set of fixes for coding/formatting issues observed while running shellcheck tool on the perf test shell scripts. Shellcheck is a static analysis tool that can find semantic/syntax bugs in the shell scripts.
Thanks, applied the series.
- Arnaldo
Quoted 77 lines
Patches 1-14 fixes the issues found with shellcheck. Patch 15, 16 and patch 17 address a fix in
task_analyzertest.This cleanup is a pre-requisite to include a build option for shellcheck discussed here: https://www.spinics.net/lists/linux-perf-users/msg25553.html
Also this is first set of patches. There will be one more set which will include build option for shellcheck as discussed in the mail thread.Abhirup Deb (2):
tools/perf/tests: fixtest_arm_spe.shsignal case issues perf/tests/shell: fix shellscript errors forlock_contention.shAboorva Devarajan (1):
tools/perf/tests: Fix shellcheck issues intest_task_analyzer.shfileAditya Gupta (3):
perf teststask_analyzer: fix bad substitution ${$1} perf teststask_analyzer: print command on failure perf teststask_analyzer: skip tests if no libtraceevent supportAkanksha J N (1):
tools/perf/tests: Fix shellcheck warnings for
trace+probe_vfs_getname.shAnushree Mathur (1):
perf/tests/shell : Shellcheck fixes for perf test
"test_arm_coresight.sh"Barnali Guha Thakurata (1):
tools/perf/tests/shell/stat_all_metrics: Fix shellcheck warning SC2076
instat_all_metrics.shDisha Goel (1):
tools/perf/tests: fix shellcheck warning for stat+json_outputGeetika (1):
tools/perf/tests: Fix all POSIX sh warnings in perf shell test
test_brstack.shKorrapati Likhitha (1):
tools/perf/tests: Fix shellcheck warnings for stat+csv_outputSamir Mulani (1):
tools/perf/tests: fixed shellcheck warnings for perf shell scriptsShirisha G (1):
tools/perf/tests: fix shellcheck warnings fordaemon.shSourabh Jain (1):
perf: get rid of unused importSpoorthy S (2):
shellcheck : fixing signal names and adding double quotes for
expression intest_arm_callgraph_fp
tools/perf/tests: Fix all POSIX sh warnings instat+shadow_stat.sh.../scripts/python/arm-cs-trace-disasm.py | 1 - tools/perf/tests/shell/buildid.sh | 12 +- tools/perf/tests/shell/daemon.sh | 113 ++++++++++++------ tools/perf/tests/shell/lock_contention.sh | 70 +++++------ .../shell/record+probe_libc_inet_pton.sh | 6 +- .../shell/record+script_probe_vfs_getname.sh | 4 +- tools/perf/tests/shell/stat+csv_output.sh | 4 +- tools/perf/tests/shell/stat+json_output.sh | 2 +- tools/perf/tests/shell/stat+shadow_stat.sh | 4 +- tools/perf/tests/shell/stat_all_metrics.sh | 6 +- .../perf/tests/shell/test_arm_callgraph_fp.sh | 6 +- tools/perf/tests/shell/test_arm_coresight.sh | 6 +- tools/perf/tests/shell/test_arm_spe.sh | 2 +- tools/perf/tests/shell/test_brstack.sh | 12 +- tools/perf/tests/shell/test_task_analyzer.sh | 98 ++++++++------- .../tests/shell/trace+probe_vfs_getname.sh | 6 +- 16 files changed, 203 insertions(+), 149 deletions(-)Signature · 1 line
2.39.1
Signature · 1 line
- Arnaldo
Quoted 10 lines · Arnaldo Carvalho de Melo
Em Tue, Jun 13, 2023 at 10:11:28PM +0530, Athira Rajeev escreveu:
Quoted 3 lines
Patchset covers a set of fixes for coding/formatting issues observed while running shellcheck tool on the perf test shell scripts. Shellcheck is a static analysis tool that can find semantic/syntax bugs in the shell scripts.
Thanks, applied the series.
Hi,
Thanks Arnaldo for picking up the patchset.
We will check and resubmit patch 6.
Thanks
Athira
Quoted 84 lines
- Arnaldo
Quoted 77 lines
Patches 1-14 fixes the issues found with shellcheck. Patch 15, 16 and patch 17 address a fix in
task_analyzertest.This cleanup is a pre-requisite to include a build option for shellcheck discussed here: https://www.spinics.net/lists/linux-perf-users/msg25553.html
Also this is first set of patches. There will be one more set which will include build option for shellcheck as discussed in the mail thread.Abhirup Deb (2):
tools/perf/tests: fixtest_arm_spe.shsignal case issues perf/tests/shell: fix shellscript errors forlock_contention.shAboorva Devarajan (1):
tools/perf/tests: Fix shellcheck issues intest_task_analyzer.shfileAditya Gupta (3):
perf teststask_analyzer: fix bad substitution ${$1} perf teststask_analyzer: print command on failure perf teststask_analyzer: skip tests if no libtraceevent supportAkanksha J N (1):
tools/perf/tests: Fix shellcheck warnings for
trace+probe_vfs_getname.shAnushree Mathur (1):
perf/tests/shell : Shellcheck fixes for perf test
"test_arm_coresight.sh"Barnali Guha Thakurata (1):
tools/perf/tests/shell/stat_all_metrics: Fix shellcheck warning SC2076
instat_all_metrics.shDisha Goel (1):
tools/perf/tests: fix shellcheck warning for stat+json_outputGeetika (1):
tools/perf/tests: Fix all POSIX sh warnings in perf shell test
test_brstack.shKorrapati Likhitha (1):
tools/perf/tests: Fix shellcheck warnings for stat+csv_outputSamir Mulani (1):
tools/perf/tests: fixed shellcheck warnings for perf shell scriptsShirisha G (1):
tools/perf/tests: fix shellcheck warnings fordaemon.shSourabh Jain (1):
perf: get rid of unused importSpoorthy S (2):
shellcheck : fixing signal names and adding double quotes for
expression intest_arm_callgraph_fp
tools/perf/tests: Fix all POSIX sh warnings instat+shadow_stat.sh.../scripts/python/arm-cs-trace-disasm.py | 1 - tools/perf/tests/shell/buildid.sh | 12 +- tools/perf/tests/shell/daemon.sh | 113 ++++++++++++------ tools/perf/tests/shell/lock_contention.sh | 70 +++++------ .../shell/record+probe_libc_inet_pton.sh | 6 +- .../shell/record+script_probe_vfs_getname.sh | 4 +- tools/perf/tests/shell/stat+csv_output.sh | 4 +- tools/perf/tests/shell/stat+json_output.sh | 2 +- tools/perf/tests/shell/stat+shadow_stat.sh | 4 +- tools/perf/tests/shell/stat_all_metrics.sh | 6 +- .../perf/tests/shell/test_arm_callgraph_fp.sh | 6 +- tools/perf/tests/shell/test_arm_coresight.sh | 6 +- tools/perf/tests/shell/test_arm_spe.sh | 2 +- tools/perf/tests/shell/test_brstack.sh | 12 +- tools/perf/tests/shell/test_task_analyzer.sh | 98 ++++++++------- .../tests/shell/trace+probe_vfs_getname.sh | 6 +- 16 files changed, 203 insertions(+), 149 deletions(-)Signature · 1 line
2.39.1Signature · 1 line
- Arnaldo