* [Buildroot] [PATCH v2 1/2] package/watchdogd: fix use of uninitialized exit status
@ 2025-02-17 13:44 Fiona Klute via buildroot
2025-02-17 13:44 ` [Buildroot] [PATCH v2 2/2] package/watchdogd: fix sysv init script Fiona Klute via buildroot
2025-02-22 14:45 ` [Buildroot] [PATCH v2 1/2] package/watchdogd: fix use of uninitialized exit status Peter Korsgaard
0 siblings, 2 replies; 6+ messages in thread
From: Fiona Klute via buildroot @ 2025-02-17 13:44 UTC (permalink / raw)
To: buildroot; +Cc: Fiona Klute, Joachim Wiberg
From: "Fiona Klute (WIWA)" <fiona.klute@gmx.de>
The bug could lead to incorrect "critical error" reports if that
uninitialized memory happened to contain a value that interpreted as
an int was above the critical threshold. It affected primarily scripts
running approximately one second or longer, because access happens by
timer.
Signed-off-by: Fiona Klute (WIWA) <fiona.klute@gmx.de>
---
Changed v1 -> v2: Update patch to the version merged upstream, with a
separate field to track if the process has exited.
...Fix-use-of-uninitialized-exit-status.patch | 85 +++++++++++++++++++
1 file changed, 85 insertions(+)
create mode 100644 package/watchdogd/0001-Fix-use-of-uninitialized-exit-status.patch
diff --git a/package/watchdogd/0001-Fix-use-of-uninitialized-exit-status.patch b/package/watchdogd/0001-Fix-use-of-uninitialized-exit-status.patch
new file mode 100644
index 0000000000..69e7b36ef4
--- /dev/null
+++ b/package/watchdogd/0001-Fix-use-of-uninitialized-exit-status.patch
@@ -0,0 +1,85 @@
+From f6d56c374723923e276ccfd442fdd0cabf23d095 Mon Sep 17 00:00:00 2001
+From: Fiona Klute <fiona.klute@gmx.de>
+Date: Fri, 14 Feb 2025 19:26:24 +0100
+Subject: [PATCH] Fix use of uninitialized exit status
+
+Depending on the timing between the SIGCHLD callback run and the first
+generic_cb() timer callback run, script_exit_status() sometimes
+returned uninitialized memory as the exit status of a script. This
+could lead to incorrect "critical error" reports if that uninitialized
+memory happened to contain a value that interpreted as an int was
+above the critical threshold.
+
+Additionally, script_exit_status() unconditionally removed the process
+information from the queue, meaning that if a process had not
+completed by the time its status was first probed, its status could
+never be updated or successfully probed again, leading to timeout. Fix
+this by removing the process only if the status indicates it has
+terminated (successful or not).
+
+The problems affected primarily scripts running approximately one
+second or longer, because very short running scripts will very likely
+have their exit status collected by the SIGCHLD callback before the
+first timer callback run.
+
+Signed-off-by: Fiona Klute <fiona.klute@gmx.de>
+Upstream: https://github.com/troglobit/watchdogd/pull/52
+---
+ src/script.c | 13 ++++++++++---
+ 1 file changed, 10 insertions(+), 3 deletions(-)
+
+diff --git a/src/script.c b/src/script.c
+index 1220fb9..f6e0763 100644
+--- a/src/script.c
++++ b/src/script.c
+@@ -16,6 +16,7 @@
+ */
+
+ #include <errno.h>
++#include <stdbool.h> /* bool type (before C23) */
+ #include <stdlib.h> /* setenv() */
+ #include <sys/wait.h> /* waitpid() */
+ #include <unistd.h> /* execv(), _exit() */
+@@ -28,6 +29,7 @@
+ struct exec_info {
+ pid_t pid;
+ int exit_status;
++ bool exited;
+ void (*cb)(void *arg);
+ void *arg;
+ LIST_ENTRY(exec_info) entry;
+@@ -64,6 +66,8 @@ static void add(pid_t pid, void (*cb)(void *), void *arg)
+ }
+
+ info->pid = pid;
++ info->exit_status = 0;
++ info->exited = false;
+ info->cb = cb;
+ info->arg = arg;
+ LIST_INSERT_HEAD(&exec_info_head, info, entry);
+@@ -78,6 +82,7 @@ static int exec(pid_t pid, int status)
+ continue;
+
+ info->exit_status = status;
++ info->exited = true;
+ if (info->cb)
+ info->cb(info->arg);
+
+@@ -96,9 +101,11 @@ int script_exit_status(pid_t pid)
+ if (info->pid != pid)
+ continue;
+
+- status = info->exit_status;
+- LIST_REMOVE(info, entry);
+- free(info);
++ if (info->exited) {
++ status = info->exit_status;
++ LIST_REMOVE(info, entry);
++ free(info);
++ }
+ break;
+ }
+
+--
+2.47.2
+
--
2.47.2
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot
^ permalink raw reply related [flat|nested] 6+ messages in thread* [Buildroot] [PATCH v2 2/2] package/watchdogd: fix sysv init script
2025-02-17 13:44 [Buildroot] [PATCH v2 1/2] package/watchdogd: fix use of uninitialized exit status Fiona Klute via buildroot
@ 2025-02-17 13:44 ` Fiona Klute via buildroot
2025-02-22 14:47 ` Peter Korsgaard
2025-02-22 14:45 ` [Buildroot] [PATCH v2 1/2] package/watchdogd: fix use of uninitialized exit status Peter Korsgaard
1 sibling, 1 reply; 6+ messages in thread
From: Fiona Klute via buildroot @ 2025-02-17 13:44 UTC (permalink / raw)
To: buildroot; +Cc: Fiona Klute, Joachim Wiberg
From: "Fiona Klute (WIWA)" <fiona.klute@gmx.de>
PIDFILE was set incorrectly, watchdogd writes its PID file to
/var/run/watchdogd/pid (note the slash), which is not configurable
without patching.
Restructure the rest of the script to match current style while at it.
Signed-off-by: Fiona Klute (WIWA) <fiona.klute@gmx.de>
---
.checkpackageignore | 1 -
package/watchdogd/S01watchdogd | 102 +++++++++++++++++++++------------
2 files changed, 66 insertions(+), 37 deletions(-)
mode change 100755 => 100644 package/watchdogd/S01watchdogd
diff --git a/.checkpackageignore b/.checkpackageignore
index 6e46b44b09..3366122d04 100644
--- a/.checkpackageignore
+++ b/.checkpackageignore
@@ -1262,7 +1262,6 @@ package/waffle/0003-drop-C-dependency.patch lib_patch.Upstream
package/wampcc/0001-Add-RISC-V-endian-detection.patch lib_patch.Upstream
package/wampcc/0002-include-wampcc-platform.h-fix-build-with-musl-1.2.0.patch lib_patch.Upstream
package/wampcc/0003-Broken-build-on-Windows.patch lib_patch.Upstream
-package/watchdogd/S01watchdogd NotExecutable lib_sysv.Indent
package/wget/0001-lib-getrandom.c-fix-build-with-uclibc-1.0.35.patch lib_patch.Upstream
package/wilc-driver/0001-cfg80211.c-fix-missing-prandom_u32-with-Linux-6.1.0.patch lib_patch.Upstream
package/wilc-driver/0002-spi.c-fix-build-failure-on-remove-callback.patch lib_patch.Upstream
diff --git a/package/watchdogd/S01watchdogd b/package/watchdogd/S01watchdogd
old mode 100755
new mode 100644
index df57b8ad7a..8f811ac0fc
--- a/package/watchdogd/S01watchdogd
+++ b/package/watchdogd/S01watchdogd
@@ -1,47 +1,77 @@
#!/bin/sh
DAEMON="watchdogd"
-PIDFILE="/var/run/$DAEMON.pid"
+PIDFILE="/var/run/$DAEMON/pid"
# shellcheck source=/dev/null
[ -r "/etc/default/$DAEMON" ] && . "/etc/default/$DAEMON"
-cmd()
-{
- start-stop-daemon -q -p "$PIDFILE" -x "/usr/sbin/$DAEMON" "$@"
- status=$?
- if [ $status -eq 0 ]; then
- echo "OK"
- else
- echo "FAIL"
- fi
- return $status
+start() {
+ printf "Starting %s: " "$DAEMON"
+ # shellcheck disable=SC2086 # we need the word splitting
+ start-stop-daemon --start --pidfile "$PIDFILE" \
+ --exec "/usr/sbin/$DAEMON" -- $WATCHDOGD_ARGS
+ status=$?
+ if [ "$status" -eq 0 ]; then
+ echo "OK"
+ else
+ echo "FAIL"
+ fi
+ return "$status"
+}
+
+stop() {
+ printf "Stopping %s: " "$DAEMON"
+ start-stop-daemon --stop --pidfile "$PIDFILE" \
+ --exec "/usr/sbin/$DAEMON"
+ status=$?
+ if [ "$status" -eq 0 ]; then
+ echo "OK"
+ else
+ echo "FAIL"
+ fi
+ # watchdogd deletes its PID file on exit, wait for it to be gone
+ while [ -f "$PIDFILE" ]; do
+ sleep 0.1
+ done
+ return "$status"
+}
+
+restart() {
+ stop
+ start
+}
+
+reload() {
+ printf "Reloading %s config: " "$DAEMON"
+ start-stop-daemon --stop --signal HUP -q --pidfile "$PIDFILE" \
+ --exec "/usr/sbin/$DAEMON"
+ status=$?
+ if [ "$status" -eq 0 ]; then
+ echo "OK"
+ else
+ echo "FAIL"
+ fi
+ return "$status"
}
case "$1" in
- start)
- printf 'Starting %s: ' "$DAEMON"
- # shellcheck disable=SC2086 # we need the word splitting
- cmd -S -- $SMCROUTED_ARGS
- ;;
-
- stop)
- printf 'Stopping %s: ' "$DAEMON"
- cmd -K
- ;;
-
- restart)
- $0 stop
- sleep 1
- $0 start
- ;;
-
- reload)
- printf 'Reloading %s: ' "$DAEMON"
- cmd -K -s HUP
- ;;
-
- *)
- echo "Usage: $0 {start|stop|restart|reload}"
- exit 1
+ start)
+ start
+ ;;
+ stop)
+ stop
+ ;;
+ restart)
+ restart
+ ;;
+ reload)
+ reload
+ ;;
+ *)
+ echo "Usage: $0 {start|stop|restart|reload}"
+ exit 1
+ ;;
esac
+
+exit $?
--
2.47.2
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot
^ permalink raw reply related [flat|nested] 6+ messages in thread* Re: [Buildroot] [PATCH v2 2/2] package/watchdogd: fix sysv init script
2025-02-17 13:44 ` [Buildroot] [PATCH v2 2/2] package/watchdogd: fix sysv init script Fiona Klute via buildroot
@ 2025-02-22 14:47 ` Peter Korsgaard
2025-02-25 20:02 ` Peter Korsgaard
0 siblings, 1 reply; 6+ messages in thread
From: Peter Korsgaard @ 2025-02-22 14:47 UTC (permalink / raw)
To: Fiona Klute via buildroot; +Cc: Fiona Klute, Joachim Wiberg
>>>>> "Fiona" == Fiona Klute via buildroot <buildroot@buildroot.org> writes:
> From: "Fiona Klute (WIWA)" <fiona.klute@gmx.de>
> PIDFILE was set incorrectly, watchdogd writes its PID file to
> /var/run/watchdogd/pid (note the slash), which is not configurable
> without patching.
> Restructure the rest of the script to match current style while at it.
> Signed-off-by: Fiona Klute (WIWA) <fiona.klute@gmx.de>
Committed, thanks.
--
Bye, Peter Korsgaard
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Buildroot] [PATCH v2 2/2] package/watchdogd: fix sysv init script
2025-02-22 14:47 ` Peter Korsgaard
@ 2025-02-25 20:02 ` Peter Korsgaard
0 siblings, 0 replies; 6+ messages in thread
From: Peter Korsgaard @ 2025-02-25 20:02 UTC (permalink / raw)
To: Fiona Klute via buildroot; +Cc: Fiona Klute, Joachim Wiberg
>>>>> "Peter" == Peter Korsgaard <peter@korsgaard.com> writes:
>>>>> "Fiona" == Fiona Klute via buildroot <buildroot@buildroot.org> writes:
>> From: "Fiona Klute (WIWA)" <fiona.klute@gmx.de>
>> PIDFILE was set incorrectly, watchdogd writes its PID file to
>> /var/run/watchdogd/pid (note the slash), which is not configurable
>> without patching.
>> Restructure the rest of the script to match current style while at it.
>> Signed-off-by: Fiona Klute (WIWA) <fiona.klute@gmx.de>
> Committed, thanks.
Committed to 2024.02.x and 2024.11.x, thanks.
--
Bye, Peter Korsgaard
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Buildroot] [PATCH v2 1/2] package/watchdogd: fix use of uninitialized exit status
2025-02-17 13:44 [Buildroot] [PATCH v2 1/2] package/watchdogd: fix use of uninitialized exit status Fiona Klute via buildroot
2025-02-17 13:44 ` [Buildroot] [PATCH v2 2/2] package/watchdogd: fix sysv init script Fiona Klute via buildroot
@ 2025-02-22 14:45 ` Peter Korsgaard
2025-02-25 19:57 ` Peter Korsgaard
1 sibling, 1 reply; 6+ messages in thread
From: Peter Korsgaard @ 2025-02-22 14:45 UTC (permalink / raw)
To: Fiona Klute via buildroot; +Cc: Fiona Klute, Joachim Wiberg
>>>>> "Fiona" == Fiona Klute via buildroot <buildroot@buildroot.org> writes:
> From: "Fiona Klute (WIWA)" <fiona.klute@gmx.de>
> The bug could lead to incorrect "critical error" reports if that
> uninitialized memory happened to contain a value that interpreted as
> an int was above the critical threshold. It affected primarily scripts
> running approximately one second or longer, because access happens by
> timer.
> Signed-off-by: Fiona Klute (WIWA) <fiona.klute@gmx.de>
> ---
> Changed v1 -> v2: Update patch to the version merged upstream, with a
> separate field to track if the process has exited.
Committed, thanks.
--
Bye, Peter Korsgaard
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Buildroot] [PATCH v2 1/2] package/watchdogd: fix use of uninitialized exit status
2025-02-22 14:45 ` [Buildroot] [PATCH v2 1/2] package/watchdogd: fix use of uninitialized exit status Peter Korsgaard
@ 2025-02-25 19:57 ` Peter Korsgaard
0 siblings, 0 replies; 6+ messages in thread
From: Peter Korsgaard @ 2025-02-25 19:57 UTC (permalink / raw)
To: Fiona Klute via buildroot; +Cc: Fiona Klute, Joachim Wiberg
>>>>> "Peter" == Peter Korsgaard <peter@korsgaard.com> writes:
>>>>> "Fiona" == Fiona Klute via buildroot <buildroot@buildroot.org> writes:
>> From: "Fiona Klute (WIWA)" <fiona.klute@gmx.de>
>> The bug could lead to incorrect "critical error" reports if that
>> uninitialized memory happened to contain a value that interpreted as
>> an int was above the critical threshold. It affected primarily scripts
>> running approximately one second or longer, because access happens by
>> timer.
>> Signed-off-by: Fiona Klute (WIWA) <fiona.klute@gmx.de>
>> ---
>> Changed v1 -> v2: Update patch to the version merged upstream, with a
>> separate field to track if the process has exited.
Committed to 2024.02.x and 2024.11.x, thanks.
--
Bye, Peter Korsgaard
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2025-02-25 20:02 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-02-17 13:44 [Buildroot] [PATCH v2 1/2] package/watchdogd: fix use of uninitialized exit status Fiona Klute via buildroot
2025-02-17 13:44 ` [Buildroot] [PATCH v2 2/2] package/watchdogd: fix sysv init script Fiona Klute via buildroot
2025-02-22 14:47 ` Peter Korsgaard
2025-02-25 20:02 ` Peter Korsgaard
2025-02-22 14:45 ` [Buildroot] [PATCH v2 1/2] package/watchdogd: fix use of uninitialized exit status Peter Korsgaard
2025-02-25 19:57 ` Peter Korsgaard
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox