taskstats: fix TGID dead-thread stat retention

fill_stats_for_tgid() builds TGID stats from two sources: the cached aggregate in signal->stats and a scan of the live threads in the group.

However, fill_tgid_exit() only accumulates delay accounting into signal->stats. This means that once a thread exits, TGID queries lose the fields that fill_stats_for_tgid() adds for live threads.

This gap was introduced incrementally by two earlier changes that extended fill_stats_for_tgid() but did not make the corresponding update to fill_tgid_exit():

- commit 8c733420bdd5 ("taskstats: add e/u/stime for TGID command") added ac_etime, ac_utime, and ac_stime to the TGID query path.
- commit b663a79c1915 ("taskstats: add context-switch counters") added nvcsw and nivcsw to the TGID query path.

As a result, those fields were accounted for live threads in TGID queries, but were dropped from the cached TGID aggregate after thread exit. The final TGID exit notification emitted when group_dead is true also copies that cached aggregate, so it loses the same fields.

Factor the per-task TGID accumulation into tgid_stats_add_task() and use it in both fill_stats_for_tgid() and fill_tgid_exit(). This keeps the cached aggregate used for dead threads aligned with the live-thread accumulation used by TGID queries.

Fixes 8c733420bdd5 ("taskstats: add e/u/stime for TGID command")

Fixes b663a79c1915 ("taskstats: add context-switch counters")

Cc [email protected]

Signed-off-by Yiyang Chen <[email protected]>

kernel/taskstats.c +1 -0

index 0cd680ccc7e5..a80be5d9f52b 100644
@@ -210,13 +210,39 @@ static int fill_stats_for_pid(pid_t pid, struct taskstats *stats)
 	return 0;
 }
 
+static void tgid_stats_add_task(struct taskstats *stats,

kernel/taskstats.c +20 -24

+	delayacct_add_tsk(stats, tsk);
+
+	/* calculate task elapsed time in nsec */
+	delta = now_ns - tsk->start_time;
+	/* Convert to micro seconds */
+	do_div(delta, NSEC_PER_USEC);
+	stats->ac_etime += delta;
+
+	task_cputime(tsk, &utime, &stime);
+	stats->ac_utime += div_u64(utime, NSEC_PER_USEC);
+	stats->ac_stime += div_u64(stime, NSEC_PER_USEC);
+
+	stats->nvcsw += tsk->nvcsw;
+	stats->nivcsw += tsk->nivcsw;
+}
+
 static int fill_stats_for_tgid(pid_t tgid, struct taskstats *stats)
 {
 	struct task_struct *tsk, *first;
 	unsigned long flags;
 	int rc = -ESRCH;
-	u64 delta, utime, stime;
-	u64 start_time;
+	u64 now_ns;
 
 	/*
 	 * Add additional stats from live tasks except zombie thread group

@@ -233,30 +259,12 @@ static int fill_stats_for_tgid(pid_t tgid, struct taskstats *stats)
 	else
 		memset(stats, 0, sizeof(*stats));
 
-	start_time = ktime_get_ns();
+	now_ns = ktime_get_ns();
 	for_each_thread(first, tsk) {
 		if (tsk->exit_state)
 			continue;
-		/*
-		 * Accounting subsystem can call its functions here to
-		 * fill in relevant parts of struct taskstsats as follows
-		 *
-		 *	per-task-foo(stats, tsk);
-		 */
-		delayacct_add_tsk(stats, tsk);
-
-		/* calculate task elapsed time in nsec */
-		delta = start_time - tsk->start_time;
-		/* Convert to micro seconds */
-		do_div(delta, NSEC_PER_USEC);
-		stats->ac_etime += delta;
 
-		task_cputime(tsk, &utime, &stime);
-		stats->ac_utime += div_u64(utime, NSEC_PER_USEC);
-		stats->ac_stime += div_u64(stime, NSEC_PER_USEC);
-
-		stats->nvcsw += tsk->nvcsw;
-		stats->nivcsw += tsk->nivcsw;
+		tgid_stats_add_task(stats, tsk, now_ns);
 	}
 
 	unlock_task_sighand(first, &flags);

@@ -275,18 +283,14 @@ static int fill_stats_for_tgid(pid_t tgid, struct taskstats *stats)
 static void fill_tgid_exit(struct task_struct *tsk)
 {
 	unsigned long flags;
+	u64 now_ns;
 
 	spin_lock_irqsave(&tsk->sighand->siglock, flags);
 	if (!tsk->signal->stats)
 		goto ret;
 
-	/*
-	 * Each accounting subsystem calls its functions here to

Add a kselftest for the taskstats TGID aggregation fix.

The test creates a worker thread, snapshots TGID taskstats while the worker is still alive, lets the worker exit, and then verifies that the TGID CPU total does not regress after the thread has been reaped.

The pass/fail check intentionally keys off ac_utime + ac_stime only, which is the primary user-visible regression fixed by the taskstats change and is less sensitive to scheduling noise than context-switch counters.

Signed-off-by Yiyang Chen <[email protected]>

tools/testing/selftests/acct/.gitignore +2 -1

index 7e78aac19038..9e9c61c5bfd6 100644
@@ -1,3 +1,4 @@
 acct_syscall
+taskstats_fill_stats_tgid
 config
-process_log
\ No newline at end of file
+process_log

tools/testing/selftests/acct/Makefile +4 -1

index 7e025099cf65..083cab5ddb72 100644
@@ -1,5 +1,8 @@
 # SPDX-License-Identifier: GPL-2.0
 TEST_GEN_PROGS := acct_syscall
+TEST_GEN_PROGS += taskstats_fill_stats_tgid
+
 CFLAGS += -Wall
+LDLIBS += -lpthread
 
-include ../lib.mk
\ No newline at end of file
+include ../lib.mk

tools/testing/selftests/acct/taskstats_fill_stats_tgid.c new file +365 -0

new file mode 100644
index 000000000000..d6cab4ae26f2
@@ -0,0 +1,375 @@
+// SPDX-License-Identifier: GPL-2.0
+#define _GNU_SOURCE
+
+#include <errno.h>
+#include <linux/genetlink.h>
+#include <linux/netlink.h>
+#include <linux/taskstats.h>
+#include <pthread.h>
+#include <stdbool.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/socket.h>
+#include <sys/types.h>
+#include <time.h>
+#include <unistd.h>
+
+#include "kselftest.h"
+
+#ifndef NLA_ALIGN
+#define NLA_ALIGNTO 4
+#define NLA_ALIGN(len) (((len) + NLA_ALIGNTO - 1) & ~(NLA_ALIGNTO - 1))
+#define NLA_HDRLEN ((int)NLA_ALIGN(sizeof(struct nlattr)))
+#endif
+
+#define BUSY_NS (200ULL * 1000 * 1000)
+
+struct worker_ctx {
+	pthread_mutex_t lock;
+	pthread_cond_t cond;
+	bool ready;
+	bool release;
+};
+
+static unsigned long busy_sink;
+
+static void *taskstats_nla_data(const struct nlattr *na)
+{
+	return (void *)((char *)na + NLA_HDRLEN);
+}
+
+static bool taskstats_nla_ok(const struct nlattr *na, int remaining)
+{
+	return remaining >= (int)sizeof(*na) &&
+	       na->nla_len >= sizeof(*na) &&
+	       na->nla_len <= remaining;
+}
+
+static struct nlattr *taskstats_nla_next(const struct nlattr *na, int *remaining)
+{
+	int aligned_len = NLA_ALIGN(na->nla_len);
+
+	*remaining -= aligned_len;
+	return (struct nlattr *)((char *)na + aligned_len);
+}
+
+static uint64_t timespec_diff_ns(const struct timespec *start,
+				 const struct timespec *end)
+{
+	return (uint64_t)(end->tv_sec - start->tv_sec) * 1000000000ULL +
+	       (uint64_t)(end->tv_nsec - start->tv_nsec);
+}
+
+static void burn_cpu_for_ns(uint64_t runtime_ns)
+{
+	struct timespec start, now;
+	unsigned long acc = 0;
+
+	if (clock_gettime(CLOCK_MONOTONIC, &start)) {
+		perror("clock_gettime");
+		exit(EXIT_FAILURE);
+	}
+
+	do {
+		for (int i = 0; i < 100000; i++)
+			acc += i;
+		if (clock_gettime(CLOCK_MONOTONIC, &now)) {
+			perror("clock_gettime");
+			exit(EXIT_FAILURE);
+		}
+	} while (timespec_diff_ns(&start, &now) < runtime_ns);
+
+	busy_sink = acc;
+}
+
+static int netlink_open(void)
+{
+	struct sockaddr_nl addr = {
+		.nl_family = AF_NETLINK,
+		.nl_pid = getpid(),
+	};
+	int fd;
+
+	fd = socket(AF_NETLINK, SOCK_RAW, NETLINK_GENERIC);
+	if (fd < 0)
+		return -errno;
+
+	if (bind(fd, (struct sockaddr *)&addr, sizeof(addr)) < 0) {
+		int err = -errno;
+
+		close(fd);
+		return err;
+	}
+
+	return fd;
+}
+
+static int send_request(int fd, void *buf, size_t len)
+{
+	struct sockaddr_nl addr = {
+		.nl_family = AF_NETLINK,
+	};
+
+	if (sendto(fd, buf, len, 0, (struct sockaddr *)&addr, sizeof(addr)) < 0)
+		return -errno;
+
+	return 0;
+}
+
+static int get_family_id(int fd, const char *name)
+{
+	struct {
+		struct nlmsghdr nlh;
+		struct genlmsghdr genl;
+		char buf[256];
+	} req = { 0 };
+	char resp[8192];
+	struct nlmsghdr *nlh;
+	struct genlmsghdr *genl;
+	struct nlattr *na;
+	int len;
+	int rem;
+	int ret;
+
+	req.nlh.nlmsg_len = NLMSG_LENGTH(GENL_HDRLEN);
+	req.nlh.nlmsg_type = GENL_ID_CTRL;
+	req.nlh.nlmsg_flags = NLM_F_REQUEST;
+	req.nlh.nlmsg_seq = 1;
+	req.nlh.nlmsg_pid = getpid();
+
+	req.genl.cmd = CTRL_CMD_GETFAMILY;
+	req.genl.version = 1;
+
+	na = (struct nlattr *)((char *)&req + NLMSG_ALIGN(req.nlh.nlmsg_len));
+	na->nla_type = CTRL_ATTR_FAMILY_NAME;
+	na->nla_len = NLA_HDRLEN + strlen(name) + 1;
+	memcpy(taskstats_nla_data(na), name, strlen(name) + 1);
+	req.nlh.nlmsg_len = NLMSG_ALIGN(req.nlh.nlmsg_len) + NLA_ALIGN(na->nla_len);
+
+	ret = send_request(fd, &req, req.nlh.nlmsg_len);
+	if (ret)
+		return ret;
+
+	len = recv(fd, resp, sizeof(resp), 0);
+	if (len < 0)
+		return -errno;
+
+	for (nlh = (struct nlmsghdr *)resp; NLMSG_OK(nlh, len);
+	     nlh = NLMSG_NEXT(nlh, len)) {
+		if (nlh->nlmsg_type == NLMSG_ERROR) {
+			struct nlmsgerr *err = NLMSG_DATA(nlh);
+
+			return err->error ? err->error : -ENOENT;
+		}
+
+		genl = (struct genlmsghdr *)NLMSG_DATA(nlh);
+		rem = nlh->nlmsg_len - NLMSG_HDRLEN - GENL_HDRLEN;
+		na = (struct nlattr *)((char *)genl + GENL_HDRLEN);
+		while (taskstats_nla_ok(na, rem)) {
+			if (na->nla_type == CTRL_ATTR_FAMILY_ID)
+				return *(uint16_t *)taskstats_nla_data(na);
+			na = taskstats_nla_next(na, &rem);
+		}
+	}
+
+	return -ENOENT;
+}
+
+static int get_taskstats(int fd, int family_id, uint16_t attr_type, uint32_t id,
+			 struct taskstats *stats)
+{
+	struct {
+		struct nlmsghdr nlh;
+		struct genlmsghdr genl;
+		char buf[256];
+	} req = { 0 };
+	char resp[16384];
+	struct nlmsghdr *nlh;
+	struct genlmsghdr *genl;
+	struct nlattr *na;
+	struct nlattr *nested;
+	int len;
+	int rem;
+	int nrem;
+	int ret;
+
+	memset(stats, 0, sizeof(*stats));
+
+	req.nlh.nlmsg_len = NLMSG_LENGTH(GENL_HDRLEN);
+	req.nlh.nlmsg_type = family_id;
+	req.nlh.nlmsg_flags = NLM_F_REQUEST;
+	req.nlh.nlmsg_seq = 2;
+	req.nlh.nlmsg_pid = getpid();
+
+	req.genl.cmd = TASKSTATS_CMD_GET;
+	req.genl.version = 1;
+
+	na = (struct nlattr *)((char *)&req + NLMSG_ALIGN(req.nlh.nlmsg_len));
+	na->nla_type = attr_type;
+	na->nla_len = NLA_HDRLEN + sizeof(id);
+	memcpy(taskstats_nla_data(na), &id, sizeof(id));
+	req.nlh.nlmsg_len = NLMSG_ALIGN(req.nlh.nlmsg_len) + NLA_ALIGN(na->nla_len);
+
+	ret = send_request(fd, &req, req.nlh.nlmsg_len);
+	if (ret)
+		return ret;
+
+	len = recv(fd, resp, sizeof(resp), 0);
+	if (len < 0)
+		return -errno;
+
+	for (nlh = (struct nlmsghdr *)resp; NLMSG_OK(nlh, len);
+	     nlh = NLMSG_NEXT(nlh, len)) {
+		if (nlh->nlmsg_type == NLMSG_ERROR) {
+			struct nlmsgerr *err = NLMSG_DATA(nlh);
+
+			return err->error ? err->error : -ENOENT;
+		}
+
+		genl = (struct genlmsghdr *)NLMSG_DATA(nlh);
+		rem = nlh->nlmsg_len - NLMSG_HDRLEN - GENL_HDRLEN;
+		na = (struct nlattr *)((char *)genl + GENL_HDRLEN);
+		while (taskstats_nla_ok(na, rem)) {
+			if (na->nla_type == TASKSTATS_TYPE_AGGR_PID ||
+			    na->nla_type == TASKSTATS_TYPE_AGGR_TGID) {
+				nested = (struct nlattr *)taskstats_nla_data(na);
+				nrem = na->nla_len - NLA_HDRLEN;
+				while (taskstats_nla_ok(nested, nrem)) {
+					if (nested->nla_type == TASKSTATS_TYPE_STATS) {
+						memcpy(stats, taskstats_nla_data(nested),
+						       sizeof(*stats));
+						return 0;
+					}
+					nested = taskstats_nla_next(nested, &nrem);
+				}
+			}
+			na = taskstats_nla_next(na, &rem);
+		}
+	}
+
+	return -ENOENT;
+}
+
+static uint64_t cpu_total(const struct taskstats *stats)
+{
+	return (uint64_t)stats->ac_utime + (uint64_t)stats->ac_stime;
+}
+
+static void print_stats(const char *label, const struct taskstats *stats)
+{
+	ksft_print_msg("%s: cpu_total=%llu nvcsw=%llu nivcsw=%llu\n",
+		       label, (unsigned long long)cpu_total(stats),
+		       (unsigned long long)stats->nvcsw,
+		       (unsigned long long)stats->nivcsw);
+}
+
+static void *worker_thread(void *arg)
+{
+	struct worker_ctx *ctx = arg;
+
+	burn_cpu_for_ns(BUSY_NS);
+
+	pthread_mutex_lock(&ctx->lock);
+	ctx->ready = true;
+	pthread_cond_broadcast(&ctx->cond);
+	while (!ctx->release)
+		pthread_cond_wait(&ctx->cond, &ctx->lock);
+	pthread_mutex_unlock(&ctx->lock);
+
+	return NULL;
+}
+
+int main(void)
+{
+	struct worker_ctx ctx = {
+		.lock = PTHREAD_MUTEX_INITIALIZER,
+		.cond = PTHREAD_COND_INITIALIZER,
+	};
+	struct taskstats before, after;
+	pthread_t thread;
+	pid_t tgid = getpid();
+	int family_id;
+	int fd;
+	int ret;
+
+	ksft_print_header();
+	ksft_set_plan(1);
+
+	if (geteuid())
+		ksft_exit_skip("taskstats_fill_stats_tgid needs root\n");
+
+	fd = netlink_open();
+	if (fd < 0)
+		ksft_exit_skip("failed to open generic netlink socket: %s\n",
+			       strerror(-fd));
+
+	family_id = get_family_id(fd, TASKSTATS_GENL_NAME);
+	if (family_id < 0)
+		ksft_exit_skip("taskstats generic netlink family unavailable: %s\n",
+			       strerror(-family_id));
+
+	/* Create worker thread that burns 200ms of CPU */
+	if (pthread_create(&thread, NULL, worker_thread, &ctx) != 0)
+		ksft_exit_fail_msg("pthread_create failed: %s\n", strerror(errno));
+
+	/* Wait for worker to finish generating activity */
+	pthread_mutex_lock(&ctx.lock);
+	while (!ctx.ready)
+		pthread_cond_wait(&ctx.cond, &ctx.lock);
+	pthread_mutex_unlock(&ctx.lock);
+
+	/*
+	 * Snapshot A: TGID stats while worker is alive and sleeping.
+	 * Contains main thread + worker contributions.
+	 */
+	ret = get_taskstats(fd, family_id, TASKSTATS_CMD_ATTR_TGID, tgid, &before);
+	if (ret)
+		ksft_exit_fail_msg("TGID query before exit failed: %s\n",
+				   strerror(-ret));
+
+	/* Release worker so it can exit, then join (deterministic wait).
+	 *
+	 * Kernel exit path ordering guarantees:
+	 *   do_exit()
+	 *     taskstats_exit() -> fill_tgid_exit()  (accumulates worker into signal->stats)
+	 *     exit_notify()                         (releases the thread)
+	 *     do_task_dead() -> __schedule()        (wakes joiner)
+	 *
+	 * So pthread_join() returns only after fill_tgid_exit() has completed.
+	 */
+	pthread_mutex_lock(&ctx.lock);
+	ctx.release = true;
+	pthread_cond_broadcast(&ctx.cond);
+	pthread_mutex_unlock(&ctx.lock);
+
+	pthread_join(thread, NULL);
+
+	/*
+	 * Snapshot B: TGID stats after worker has exited.
+	 * fill_stats_for_tgid() does:
+	 *   memcpy(signal->stats)   <- includes fill_tgid_exit accumulation
+	 *   + scan live threads      <- only main thread now
+	 */
+	ret = get_taskstats(fd, family_id, TASKSTATS_CMD_ATTR_TGID, tgid, &after);
+	if (ret)
+		ksft_exit_fail_msg("TGID query after exit failed: %s\n",
+				   strerror(-ret));
+
+	print_stats("TGID before worker exit", &before);
+	print_stats("TGID after  worker exit", &after);
+
+	/*
+	 * The worker burned 200ms of CPU before the first snapshot.
+	 * If the kernel correctly retained its contribution via