linux-kernel - [PATCH v2] selftests/run_kselftest.sh: exit with error if tests fail

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <20251014-b4-ksft-error-on-fail-v2-1-b3e2657237b8@google.com>
Date: Tue, 14 Oct 2025 14:35:41 +0000
From: Brendan Jackman <jackmanb@...gle.com>
To: Shuah Khan <shuah@...nel.org>
Cc: "Thomas Weißschuh" <thomas.weissschuh@...utronix.de>, linux-kselftest@...r.kernel.org, 
	linux-kernel@...r.kernel.org, Brendan Jackman <jackmanb@...gle.com>
Subject: [PATCH v2] selftests/run_kselftest.sh: exit with error if tests fail

Parsing KTAP is quite an inconvenience, but most of the time the thing
you really want to know is "did anything fail"?

Let's give the user the his information without them needing
to parse anything.

Because of the use of subshells and namespaces, this needs to be
communicated via a file. Just write arbitrary data into the file and
treat non-empty content as a signal that something failed.

In case any user depends on the current behaviour, such as running this
from a script with `set -e` and parsing the result for failures
afterwards, add a flag they can set to get the old behaviour, namely
--no-error-on-fail.

Signed-off-by: Brendan Jackman <jackmanb@...gle.com>
---
Changes in v2:
- Fixed bug in report_failure()
- Made error-on-fail the default
- Link to v1: https://lore.kernel.org/r/20251007-b4-ksft-error-on-fail-v1-1-71bf058f5662@google.com
---
 tools/testing/selftests/kselftest/runner.sh | 14 ++++++++++----
 tools/testing/selftests/run_kselftest.sh    | 14 ++++++++++++++
 2 files changed, 24 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/kselftest/runner.sh b/tools/testing/selftests/kselftest/runner.sh
index 2c3c58e65a419f5ee8d7dc51a37671237a07fa0b..3a62039fa6217f3453423ff011575d0a1eb8c275 100644
--- a/tools/testing/selftests/kselftest/runner.sh
+++ b/tools/testing/selftests/kselftest/runner.sh
@@ -44,6 +44,12 @@ tap_timeout()
 	fi
 }
 
+report_failure()
+{
+	echo "not ok $*"
+	echo "$*" >> "$kselftest_failures_file"
+}
+
 run_one()
 {
 	DIR="$1"
@@ -105,7 +111,7 @@ run_one()
 	echo "# $TEST_HDR_MSG"
 	if [ ! -e "$TEST" ]; then
 		echo "# Warning: file $TEST is missing!"
-		echo "not ok $test_num $TEST_HDR_MSG"
+		report_failure "$test_num $TEST_HDR_MSG"
 	else
 		if [ -x /usr/bin/stdbuf ]; then
 			stdbuf="/usr/bin/stdbuf --output=L "
@@ -123,7 +129,7 @@ run_one()
 				interpreter=$(head -n 1 "$TEST" | cut -c 3-)
 				cmd="$stdbuf $interpreter ./$BASENAME_TEST"
 			else
-				echo "not ok $test_num $TEST_HDR_MSG"
+				report_failure "$test_num $TEST_HDR_MSG"
 				return
 			fi
 		fi
@@ -137,9 +143,9 @@ run_one()
 			echo "ok $test_num $TEST_HDR_MSG # SKIP"
 		elif [ $rc -eq $timeout_rc ]; then \
 			echo "#"
-			echo "not ok $test_num $TEST_HDR_MSG # TIMEOUT $kselftest_timeout seconds"
+			report_failure "$test_num $TEST_HDR_MSG # TIMEOUT $kselftest_timeout seconds"
 		else
-			echo "not ok $test_num $TEST_HDR_MSG # exit=$rc"
+			report_failure "$test_num $TEST_HDR_MSG # exit=$rc"
 		fi)
 		cd - >/dev/null
 	fi
diff --git a/tools/testing/selftests/run_kselftest.sh b/tools/testing/selftests/run_kselftest.sh
index 0443beacf3621ae36cb12ffd57f696ddef3526b5..cc1b4190edacedadafd9b993a351e4cfbf17ccd5 100755
--- a/tools/testing/selftests/run_kselftest.sh
+++ b/tools/testing/selftests/run_kselftest.sh
@@ -33,6 +33,7 @@ Usage: $0 [OPTIONS]
   -c | --collection COLLECTION	Run all tests from COLLECTION
   -l | --list			List the available collection:test entries
   -d | --dry-run		Don't actually run any tests
+  -f | --no-error-on-fail	Don't exit with an error just because tests failed
   -n | --netns			Run each test in namespace
   -h | --help			Show this usage info
   -o | --override-timeout	Number of seconds after which we timeout
@@ -44,6 +45,7 @@ COLLECTIONS=""
 TESTS=""
 dryrun=""
 kselftest_override_timeout=""
+ERROR_ON_FAIL=true
 while true; do
 	case "$1" in
 		-s | --summary)
@@ -65,6 +67,9 @@ while true; do
 		-d | --dry-run)
 			dryrun="echo"
 			shift ;;
+		-f | --no-error-on-fail)
+			ERROR_ON_FAIL=false
+			shift ;;
 		-n | --netns)
 			RUN_IN_NETNS=1
 			shift ;;
@@ -105,9 +110,18 @@ if [ -n "$TESTS" ]; then
 	available="$(echo "$valid" | sed -e 's/ /\n/g')"
 fi
 
+kselftest_failures_file=$(mktemp --tmpdir kselftest-failures-XXXXXX)
+export kselftest_failures_file
+
 collections=$(echo "$available" | cut -d: -f1 | sort | uniq)
 for collection in $collections ; do
 	[ -w /dev/kmsg ] && echo "kselftest: Running tests in $collection" >> /dev/kmsg
 	tests=$(echo "$available" | grep "^$collection:" | cut -d: -f2)
 	($dryrun cd "$collection" && $dryrun run_many $tests)
 done
+
+failures="$(cat "$kselftest_failures_file")"
+rm "$kselftest_failures_file"
+if "$ERROR_ON_FAIL" && [ "$failures" ]; then
+	exit 1
+fi

---
base-commit: 8f5ae30d69d7543eee0d70083daf4de8fe15d585
change-id: 20251007-b4-ksft-error-on-fail-0c2cb3246041

Best regards,
-- 
Brendan Jackman <jackmanb@...gle.com>