From nobody Mon Jun 15 18:00:18 2026 Received: from mail-pg1-f172.google.com (mail-pg1-f172.google.com [209.85.215.172]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 97E813DD520 for ; Mon, 13 Apr 2026 15:46:00 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.215.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1776095161; cv=none; b=n08W/o85dO0dclDnXPKjzUp0b783y1i68XqeSFqS3JQSdmIEzjMJdhrXIQcvPWDEahpbFCGz27LZCpxxtEKhv15+BDjPZxc4q2R/Br3A//CEUBKswFu1p/zTSzfrnL8XnJccFTvd98rP+qTlW0TJ9Y506V4jKKqKsm9TYuvAwYc= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1776095161; c=relaxed/simple; bh=4CemZu+suVUtzFuZOxj3sutsvIitsmDpqggC93wWMTw=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=XtUY23wGuXipgNzMun8BjU9Fct4aCJVgDGqjKB1U90XffXYlaajgWGd8iyMD7XLEt8730WySNcG0PmEFr52F8SJ0guqg9PjZcxAYyJGCLXaNMonYKzMYIIk6tfb/rNAsRgjLWtY5KBFfi0YirZL1zjGnwn7Chg+GMt+f3yuqjDg= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=jOLZKnZ6; arc=none smtp.client-ip=209.85.215.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="jOLZKnZ6" Received: by mail-pg1-f172.google.com with SMTP id 41be03b00d2f7-c70e27e2b74so1693671a12.0 for ; Mon, 13 Apr 2026 08:46:00 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20251104; t=1776095160; x=1776699960; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=+JusnDR0Yh9TDaRsWoQKtZaA9CPLjEJexzqPRXw9RNc=; b=jOLZKnZ6lAysEHloeIMR0dvx5BrUykpwrigoAoKIcSwehxZPhir9Z8loaQHzDOKDH6 pBr5S+GJQJRbz1rGjio4HLogRjxdHCgBK42YdCG9nrbAc+5yLno0NATxSTOhVpvc57vZ hZv/lxhYbBF4P8LV2WdRj+625g1LsQqILQ5DvJ4Je4oaPTq7S9rZBM/OUpiWvW8HN+Vp s23HBtaZ9Jqo2UHtArjjaKBh2FugIXWOqwx3xN9kdw5vrRM00Cw1imYolNVL+lBW1/NF wvyzH5GwEppwMR5UGZiSgsW5M6cC0b+S/x+Zg+DVIO+GikBlaYVmHnM8GLb3gsHHJ4bS JHSw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1776095160; x=1776699960; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-gg:x-gm-message-state:from :to:cc:subject:date:message-id:reply-to; bh=+JusnDR0Yh9TDaRsWoQKtZaA9CPLjEJexzqPRXw9RNc=; b=O+ImWVhStyD+oGb+SMD8LPHQ+XHlWOCIlwbl86nyn6dRXBHDBEeJv9VDplRTid9RH5 zgibWirJaJNXXBDZAYFlnEqXIztkUlf1rULcpJ+T2phvBv12wHRv1uaFMdcTkxWpIaym MeEwIuIVYZEKP0xHtcowihalNcqVzYPzjciHG3goaXpUtAUOL0gIvMVBrUK7vtGyy9hX Fl83VX5asP72IlJXfHu8o1b/wiPALu/MvxXrEIOc1o95szFB2YZKq/KyApeb+DgSqxjQ 1Y5nB6o8KAO3DEI07wQwTEJJoSufihe/Ci5CPPdtvjX2J/TmfzKFLVkc1qzkIHH4wQgI 411w== X-Gm-Message-State: AOJu0YzWS+3HzxEKzlLalhgPpdbLn5703EXa3lCQ210zd3kBkW6ukeQz 5g71JJmbBiIHt6BeF0t/u1YG+xr5M18iUTLlKfGR1BQnf1PJJv/NgGnr X-Gm-Gg: AeBDieuhmGXDojMmYTMsDkOruOB4Y+qg7p4/6a7nnMP7fIKefPXhnBIkG09T5w/PJRs WzhDQN1ebScBt7Z532bFb69FU6MamNlZdbjkrRBRlKZt3vINMK5rQXIMjeV3sNtMT2fabfNs28k S8PjsIBFUUEo7RfV+pOxYlGXofj+7f7ey1hkez0x16oSbjwyrEcVYXKuGRWWalEIC1Ir+7hhfH/ f2yGdG7fzJ0faDeBtCb05fnggTFF8CbUnFvVwquCWJ4d6/QaJhML3f2gDL66oheE8NzsicZOd6B WTR99BpLE1KsqW2wZxivrWOu/cl8Qwws/6AgrOgi6OHU4/DcQY41+A92hQ4hYsYG3hMQJt0CQD6 CTpSkLZ4u4CReVAYWf3QnT57nnSwKNRiECGPSIw1rnCGyaHX769lJBsF/zKxeGgRVNsljNqJepo JutABnfmpTSiqt8N1BF//AnA== X-Received: by 2002:a05:6a20:7486:b0:398:b346:b13 with SMTP id adf61e73a8af0-39fe3d0dfafmr15810712637.16.1776095159705; Mon, 13 Apr 2026 08:45:59 -0700 (PDT) Received: from ubuntu24.lan ([2602:ffe4:1:2113:9dfd:1ff:3726:3839]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-c793488d824sm6318233a12.16.2026.04.13.08.45.57 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 13 Apr 2026 08:45:59 -0700 (PDT) From: Yiyang Chen To: Balbir Singh , Yang Yang , Wang Yaxin Cc: linux-kernel@vger.kernel.org, Oleg Nesterov , "Dr . Thomas Orgis" , Andrew Morton , stable@vger.kernel.org, Yiyang Chen Subject: [PATCH v3 1/2] taskstats: retain dead thread stats in TGID queries Date: Mon, 13 Apr 2026 23:45:44 +0800 Message-ID: X-Mailer: git-send-email 2.43.0 In-Reply-To: References: Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" 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: stable@vger.kernel.org Acked-by: Balbir Singh Signed-off-by: Yiyang Chen diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 0cd680ccc7e5..f572f27a5828 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -210,13 +210,39 @@ static int fill_stats_for_pid(pid_t pid, struct tasks= tats *stats) return 0; } =20 +static void tgid_stats_add_task(struct taskstats *stats, + struct task_struct *tsk, u64 now_ns) +{ + u64 delta, utime, stime; + + /* + * Each accounting subsystem calls its functions here to + * accumulate its per-task stats for tsk, into the per-tgid structure + * + * per-task-foo(stats, tsk); + */ + delayacct_add_tsk(stats, tsk); + + /* calculate task elapsed time in nsec */ + delta =3D now_ns - tsk->start_time; + /* Convert to micro seconds */ + do_div(delta, NSEC_PER_USEC); + stats->ac_etime +=3D delta; + + task_cputime(tsk, &utime, &stime); + stats->ac_utime +=3D div_u64(utime, NSEC_PER_USEC); + stats->ac_stime +=3D div_u64(stime, NSEC_PER_USEC); + + stats->nvcsw +=3D tsk->nvcsw; + stats->nivcsw +=3D tsk->nivcsw; +} + static int fill_stats_for_tgid(pid_t tgid, struct taskstats *stats) { struct task_struct *tsk, *first; unsigned long flags; int rc =3D -ESRCH; - u64 delta, utime, stime; - u64 start_time; + u64 now_ns; =20 /* * Add additional stats from live tasks except zombie thread group @@ -233,30 +259,12 @@ static int fill_stats_for_tgid(pid_t tgid, struct tas= kstats *stats) else memset(stats, 0, sizeof(*stats)); =20 - start_time =3D ktime_get_ns(); + now_ns =3D 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 =3D start_time - tsk->start_time; - /* Convert to micro seconds */ - do_div(delta, NSEC_PER_USEC); - stats->ac_etime +=3D delta; =20 - task_cputime(tsk, &utime, &stime); - stats->ac_utime +=3D div_u64(utime, NSEC_PER_USEC); - stats->ac_stime +=3D div_u64(stime, NSEC_PER_USEC); - - stats->nvcsw +=3D tsk->nvcsw; - stats->nivcsw +=3D tsk->nivcsw; + tgid_stats_add_task(stats, tsk, now_ns); } =20 unlock_task_sighand(first, &flags); @@ -275,18 +283,14 @@ static int fill_stats_for_tgid(pid_t tgid, struct tas= kstats *stats) static void fill_tgid_exit(struct task_struct *tsk) { unsigned long flags; + u64 now_ns; =20 spin_lock_irqsave(&tsk->sighand->siglock, flags); if (!tsk->signal->stats) goto ret; =20 - /* - * Each accounting subsystem calls its functions here to - * accumalate its per-task stats for tsk, into the per-tgid structure - * - * per-task-foo(tsk->signal->stats, tsk); - */ - delayacct_add_tsk(tsk->signal->stats, tsk); + now_ns =3D ktime_get_ns(); + tgid_stats_add_task(tsk->signal->stats, tsk, now_ns); ret: spin_unlock_irqrestore(&tsk->sighand->siglock, flags); return; --=20 2.43.0 From nobody Mon Jun 15 18:00:18 2026 Received: from mail-pg1-f176.google.com (mail-pg1-f176.google.com [209.85.215.176]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 087183D3314 for ; Mon, 13 Apr 2026 15:46:04 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.215.176 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1776095166; cv=none; b=OxGrttZdtRLpyOEnC7rg89b8ukATTU+Bv6B/1uNPdnHWbpuSdGzYlxEP4wApff2usnEhmWM70WrormXTrsruKe9O+ffsro73/e86lNiNtuoN+HOSnsM2txP7GPESFsUxCxQ/K2CBrtyv4eYmo32SZ4PXmlq77IpGl694bMGsN3Y= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1776095166; c=relaxed/simple; bh=RPMP5nN3XQcJ4Sp8oYrKvpa28GuzhchqiWFFEbEuP28=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=SVijfI4hD0ohQrC6M8bfdmmUg0ijBpeiY35O5RLFuYDtKcL/XsrcDKaznkNgspPqMVGA9uG4thlg8yh7xZU54MGEcfdJlCKbM4jq8r4SYlcvAWp1elitJ/W9pj2X0wLJWoX7ARGkmT7AkLf8G80Zbp6lU5v/OG9UnCun4aABr2E= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=ceWTSMTo; arc=none smtp.client-ip=209.85.215.176 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="ceWTSMTo" Received: by mail-pg1-f176.google.com with SMTP id 41be03b00d2f7-c79467f128cso149864a12.3 for ; Mon, 13 Apr 2026 08:46:04 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20251104; t=1776095164; x=1776699964; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=Gk406e4B6yaA1AYsopBQuAcW9y1VonUX3B3BnIn8G5g=; b=ceWTSMToBgebIGAeejWWcD3kTohblOkjh7e9mIfyke0gklHQfpe0jC9P0U04sLVxTQ ngz31dBiIVfZb+upipe+BFgYlQv397uukorXXh3PY6BPUFIOG5hXE+DCOu+bv4P9gKTz 1pb8dbFBifdqgdF06gPpunqFBviwykdZPfrUQSpa3ufJCOPNeecNP0v5C3HHF/+v7Qgw wStPJOXG/JHG6DKPvQVyU4I5f/h6dt6lnlxZYgk5RzRi9VyUwEN9ufF4I2kB961iWLY+ cC+eu4kA8AF1LXBWjsANFTPlx38UVUJB7nqaiC7Zl9OSbhq51CfCDEIreSsXsL7Sw8B1 PW+w== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1776095164; x=1776699964; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-gg:x-gm-message-state:from :to:cc:subject:date:message-id:reply-to; bh=Gk406e4B6yaA1AYsopBQuAcW9y1VonUX3B3BnIn8G5g=; b=fRdrUVGzk06y8YkUkbXl7vZO9LZdS/PgYHR+vfDgnmRIo4CPReK6WJj7Z2E7pFMhS0 8NR7nI5rmPSmuiRCwEtMHvPbMtUWw9g0t+PLTnF6Fy7TZrbuA3FGJg7ZP2dU8uLOHoC6 OCLZVCMEAq6Ynqf+G65G0ox5gWJhMb5PENsf9HydHQQm3VeuqStpUC/aJEXkgCZm8FTN NgTdOfUlqBV/0PyDY2KFAihWn+DeRX0a1b5UBuLHmaOcZIlp4EGlKDloFSGL0GIpttNJ SN9e5uyE+601M2RavgJdGjtFUsTUQxvXZ3Ecub33yCYK0P5xaO7vftJctnhiHZc9MInA jdmA== X-Gm-Message-State: AOJu0YwBy3Kl/pZ3ittWo3h5BTkeQ+Ni0DFN/d5Rd4iVMESqTXxt9qhQ F4LbBWapodp+cmGH3oyk3XVS6z6bxRFrc+lg8OyNycB1IvbeYGF8ye9gBaDjLH8ddfw= X-Gm-Gg: AeBDies6Q9IcnsYkFCmAiLKFUbjme1PLJ/QWpPRn5bnOFDChSgY1fwCLkVZZ4IR/Ivp +0wXmZqQLy6od4WA044icIustpFp0LL2u2vlaSHu3S9Wda8fmTp11xqu0UEt6I2PMfOYNoISPXN iEXVuYBZXVWDhj4s8pEk+I4ZlPKPLtWK0RK0FxTlmreRS2ENW+jp5Ulq6InjYJwhhlLqByKyDKp XF8yK7URjHpUqhpz/4tDyk+Bb/1ec/XzBcjAO/OBV/FPaqbLFXxoWeOON1//iFtzEE1ZSoeBTpy zziAXQno84Lp900FA6xFGx9gckwcw8RPMQd0aLa6hoyKPi+SZyuA2/vVc4gbKaoKdvMULqikCuV I3ShATILdVBBM83f9dLpbMO6ooZKcXsAP3Tk/jrWos2zlH468r2X/4IZfRNnjp7gXmUWQ2AVGAT xYF5QphaztYqBvallesKm8BQ4sSqNo2iYC X-Received: by 2002:a05:6a20:258a:b0:3a0:129a:974b with SMTP id adf61e73a8af0-3a0129a9aa3mr7223876637.48.1776095164257; Mon, 13 Apr 2026 08:46:04 -0700 (PDT) Received: from ubuntu24.lan ([2602:ffe4:1:2113:9dfd:1ff:3726:3839]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-c793488d824sm6318233a12.16.2026.04.13.08.46.01 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 13 Apr 2026 08:46:03 -0700 (PDT) From: Yiyang Chen To: Balbir Singh , Yang Yang , Wang Yaxin Cc: linux-kernel@vger.kernel.org, Oleg Nesterov , "Dr . Thomas Orgis" , Andrew Morton , stable@vger.kernel.org, Yiyang Chen Subject: [PATCH v3 2/2] selftests/acct: add taskstats TGID retention test Date: Mon, 13 Apr 2026 23:45:45 +0800 Message-ID: <0d55354911c54cd1b9f10a09f6fd378af85c8d43.1776094300.git.cyyzero16@gmail.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: References: Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" 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. Acked-by: Balbir Singh Signed-off-by: Yiyang Chen diff --git a/tools/testing/selftests/acct/.gitignore b/tools/testing/selfte= sts/acct/.gitignore index 7e78aac19038..9e9c61c5bfd6 100644 --- a/tools/testing/selftests/acct/.gitignore +++ b/tools/testing/selftests/acct/.gitignore @@ -1,3 +1,4 @@ acct_syscall +taskstats_fill_stats_tgid config -process_log \ No newline at end of file +process_log diff --git a/tools/testing/selftests/acct/Makefile b/tools/testing/selftest= s/acct/Makefile index 7e025099cf65..083cab5ddb72 100644 --- a/tools/testing/selftests/acct/Makefile +++ b/tools/testing/selftests/acct/Makefile @@ -1,5 +1,8 @@ # SPDX-License-Identifier: GPL-2.0 TEST_GEN_PROGS :=3D acct_syscall +TEST_GEN_PROGS +=3D taskstats_fill_stats_tgid + CFLAGS +=3D -Wall +LDLIBS +=3D -lpthread =20 -include ../lib.mk \ No newline at end of file +include ../lib.mk diff --git a/tools/testing/selftests/acct/taskstats_fill_stats_tgid.c b/too= ls/testing/selftests/acct/taskstats_fill_stats_tgid.c new file mode 100644 index 000000000000..d6cab4ae26f2 --- /dev/null +++ b/tools/testing/selftests/acct/taskstats_fill_stats_tgid.c @@ -0,0 +1,375 @@ +// SPDX-License-Identifier: GPL-2.0 +#define _GNU_SOURCE + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#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 >=3D (int)sizeof(*na) && + na->nla_len >=3D sizeof(*na) && + na->nla_len <=3D remaining; +} + +static struct nlattr *taskstats_nla_next(const struct nlattr *na, int *rem= aining) +{ + int aligned_len =3D NLA_ALIGN(na->nla_len); + + *remaining -=3D 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 =3D 0; + + if (clock_gettime(CLOCK_MONOTONIC, &start)) { + perror("clock_gettime"); + exit(EXIT_FAILURE); + } + + do { + for (int i =3D 0; i < 100000; i++) + acc +=3D i; + if (clock_gettime(CLOCK_MONOTONIC, &now)) { + perror("clock_gettime"); + exit(EXIT_FAILURE); + } + } while (timespec_diff_ns(&start, &now) < runtime_ns); + + busy_sink =3D acc; +} + +static int netlink_open(void) +{ + struct sockaddr_nl addr =3D { + .nl_family =3D AF_NETLINK, + .nl_pid =3D getpid(), + }; + int fd; + + fd =3D socket(AF_NETLINK, SOCK_RAW, NETLINK_GENERIC); + if (fd < 0) + return -errno; + + if (bind(fd, (struct sockaddr *)&addr, sizeof(addr)) < 0) { + int err =3D -errno; + + close(fd); + return err; + } + + return fd; +} + +static int send_request(int fd, void *buf, size_t len) +{ + struct sockaddr_nl addr =3D { + .nl_family =3D 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 =3D { 0 }; + char resp[8192]; + struct nlmsghdr *nlh; + struct genlmsghdr *genl; + struct nlattr *na; + int len; + int rem; + int ret; + + req.nlh.nlmsg_len =3D NLMSG_LENGTH(GENL_HDRLEN); + req.nlh.nlmsg_type =3D GENL_ID_CTRL; + req.nlh.nlmsg_flags =3D NLM_F_REQUEST; + req.nlh.nlmsg_seq =3D 1; + req.nlh.nlmsg_pid =3D getpid(); + + req.genl.cmd =3D CTRL_CMD_GETFAMILY; + req.genl.version =3D 1; + + na =3D (struct nlattr *)((char *)&req + NLMSG_ALIGN(req.nlh.nlmsg_len)); + na->nla_type =3D CTRL_ATTR_FAMILY_NAME; + na->nla_len =3D NLA_HDRLEN + strlen(name) + 1; + memcpy(taskstats_nla_data(na), name, strlen(name) + 1); + req.nlh.nlmsg_len =3D NLMSG_ALIGN(req.nlh.nlmsg_len) + NLA_ALIGN(na->nla_= len); + + ret =3D send_request(fd, &req, req.nlh.nlmsg_len); + if (ret) + return ret; + + len =3D recv(fd, resp, sizeof(resp), 0); + if (len < 0) + return -errno; + + for (nlh =3D (struct nlmsghdr *)resp; NLMSG_OK(nlh, len); + nlh =3D NLMSG_NEXT(nlh, len)) { + if (nlh->nlmsg_type =3D=3D NLMSG_ERROR) { + struct nlmsgerr *err =3D NLMSG_DATA(nlh); + + return err->error ? err->error : -ENOENT; + } + + genl =3D (struct genlmsghdr *)NLMSG_DATA(nlh); + rem =3D nlh->nlmsg_len - NLMSG_HDRLEN - GENL_HDRLEN; + na =3D (struct nlattr *)((char *)genl + GENL_HDRLEN); + while (taskstats_nla_ok(na, rem)) { + if (na->nla_type =3D=3D CTRL_ATTR_FAMILY_ID) + return *(uint16_t *)taskstats_nla_data(na); + na =3D 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 =3D { 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 =3D NLMSG_LENGTH(GENL_HDRLEN); + req.nlh.nlmsg_type =3D family_id; + req.nlh.nlmsg_flags =3D NLM_F_REQUEST; + req.nlh.nlmsg_seq =3D 2; + req.nlh.nlmsg_pid =3D getpid(); + + req.genl.cmd =3D TASKSTATS_CMD_GET; + req.genl.version =3D 1; + + na =3D (struct nlattr *)((char *)&req + NLMSG_ALIGN(req.nlh.nlmsg_len)); + na->nla_type =3D attr_type; + na->nla_len =3D NLA_HDRLEN + sizeof(id); + memcpy(taskstats_nla_data(na), &id, sizeof(id)); + req.nlh.nlmsg_len =3D NLMSG_ALIGN(req.nlh.nlmsg_len) + NLA_ALIGN(na->nla_= len); + + ret =3D send_request(fd, &req, req.nlh.nlmsg_len); + if (ret) + return ret; + + len =3D recv(fd, resp, sizeof(resp), 0); + if (len < 0) + return -errno; + + for (nlh =3D (struct nlmsghdr *)resp; NLMSG_OK(nlh, len); + nlh =3D NLMSG_NEXT(nlh, len)) { + if (nlh->nlmsg_type =3D=3D NLMSG_ERROR) { + struct nlmsgerr *err =3D NLMSG_DATA(nlh); + + return err->error ? err->error : -ENOENT; + } + + genl =3D (struct genlmsghdr *)NLMSG_DATA(nlh); + rem =3D nlh->nlmsg_len - NLMSG_HDRLEN - GENL_HDRLEN; + na =3D (struct nlattr *)((char *)genl + GENL_HDRLEN); + while (taskstats_nla_ok(na, rem)) { + if (na->nla_type =3D=3D TASKSTATS_TYPE_AGGR_PID || + na->nla_type =3D=3D TASKSTATS_TYPE_AGGR_TGID) { + nested =3D (struct nlattr *)taskstats_nla_data(na); + nrem =3D na->nla_len - NLA_HDRLEN; + while (taskstats_nla_ok(nested, nrem)) { + if (nested->nla_type =3D=3D TASKSTATS_TYPE_STATS) { + memcpy(stats, taskstats_nla_data(nested), + sizeof(*stats)); + return 0; + } + nested =3D taskstats_nla_next(nested, &nrem); + } + } + na =3D 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=3D%llu nvcsw=3D%llu nivcsw=3D%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 =3D arg; + + burn_cpu_for_ns(BUSY_NS); + + pthread_mutex_lock(&ctx->lock); + ctx->ready =3D 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 =3D { + .lock =3D PTHREAD_MUTEX_INITIALIZER, + .cond =3D PTHREAD_COND_INITIALIZER, + }; + struct taskstats before, after; + pthread_t thread; + pid_t tgid =3D 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 =3D netlink_open(); + if (fd < 0) + ksft_exit_skip("failed to open generic netlink socket: %s\n", + strerror(-fd)); + + family_id =3D 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) !=3D 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 =3D get_taskstats(fd, family_id, TASKSTATS_CMD_ATTR_TGID, tgid, &befo= re); + 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 sig= nal->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 =3D 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 =3D get_taskstats(fd, family_id, TASKSTATS_CMD_ATTR_TGID, tgid, &afte= r); + 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 + * fill_tgid_exit(), then the TGID CPU total after exit must be at + * least as large as the TGID CPU total before exit. + */ + ksft_test_result(cpu_total(&after) >=3D cpu_total(&before), + "TGID CPU stats should not regress after thread exit\n"); + + close(fd); + ksft_finished(); + return ksft_get_fail_cnt() ? KSFT_FAIL : KSFT_PASS; +} --=20 2.43.0