From nobody Sun Feb 8 21:27:11 2026 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 2CBAAC77B7D for ; Mon, 15 May 2023 18:32:30 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S243264AbjEOScK (ORCPT ); Mon, 15 May 2023 14:32:10 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:37912 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S239142AbjEOScI (ORCPT ); Mon, 15 May 2023 14:32:08 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7CC88100EB for ; Mon, 15 May 2023 11:31:26 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1684175485; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=DcXHSTN5+9crGs0d1ByPA1BOvgEKNfvLeXclyFifNvo=; b=f6kUakMbrOZ4e/vNxMAVYwn6orK1Qe02nd+0L8r59E0kjpYO9aYMiWXeSGoHDKKl7mntEE z0pIMaTgeeYSvVE/eW160SpvzhDIbrvN+w5DonERgoiTjKc01zmzQSpWSlGLTHnpMGJQG8 5zJ4Cu/m92EJXqn9MgRP4YtXGhKaKok= Received: from mail-oa1-f71.google.com (mail-oa1-f71.google.com [209.85.160.71]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-220-g29-QMw9MiWZSxqgIyFfRw-1; Mon, 15 May 2023 14:31:24 -0400 X-MC-Unique: g29-QMw9MiWZSxqgIyFfRw-1 Received: by mail-oa1-f71.google.com with SMTP id 586e51a60fabf-192b4dc7ef1so3433514fac.3 for ; Mon, 15 May 2023 11:31:24 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1684175483; x=1686767483; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=DcXHSTN5+9crGs0d1ByPA1BOvgEKNfvLeXclyFifNvo=; b=CkxQVlSTcd36AANsIijckDpOZwG9QtphiOrcrnkdJYIC2ZqExeV9T/X1LN+DW6qk/R 5op/O9Y/ioTsKkfFLi/6c7Yt2p0cz5jr6P6cPkZ5yhxxjR2av5dckqu0brxLtwMrw7l4 IyEF2+TMDPPb3XUdasX3boeD+lA2Vj+qsBkAD2N0Gi1/ML9ao8hSZpF/v1JO5Thwokw2 SQTZ4K3oRk82u9E/3oZDlQeL+xoD4ActnG+3MXJ4vl0iozswKzjr2hL5UyK4awZXR+Dx HVZ9iYyrNSfLfQFE17CrKySN4cwgl4waVQn09SfyM6DT8tuciW6wo6BnFYBz55XDx3XS dkaw== X-Gm-Message-State: AC+VfDzH1tofu8v3CbAJOOhVLBnInMTySAGIcafpVjxOvblJzmHOE2fC GBi7rvllRcMpb7dvf3XSH4Q0YZ6N/5Tli4FXQ6h/yRB3UiVs8uFqZDopB5Q575dmSTnjU8AhCJk KETjrjlY3bZPi2tP1Ky3exHaO X-Received: by 2002:a05:6871:40f:b0:187:c066:a395 with SMTP id d15-20020a056871040f00b00187c066a395mr16141526oag.0.1684175483575; Mon, 15 May 2023 11:31:23 -0700 (PDT) X-Google-Smtp-Source: ACHHUZ7FlCRobycL533zZedNWGUsksKyld4HpV5MvQyWK6XXNRovrdvGIesK3IN2aMnipjczwWzz7Q== X-Received: by 2002:a05:6871:40f:b0:187:c066:a395 with SMTP id d15-20020a056871040f00b00187c066a395mr16141506oag.0.1684175483343; Mon, 15 May 2023 11:31:23 -0700 (PDT) Received: from localhost.localdomain ([2804:1b3:a803:46cc:5b68:5c23:dd7a:8cb3]) by smtp.gmail.com with ESMTPSA id ea20-20020a056870071400b001964dc3dadesm6094075oab.45.2023.05.15.11.31.19 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 15 May 2023 11:31:22 -0700 (PDT) From: Leonardo Bras To: Thomas Gleixner , Sebastian Andrzej Siewior , Nadav Amit , Chen Zhongjin , Daniel Bristot de Oliveira , Marcelo Tosatti , Steven Rostedt , Masami Hiramatsu , Leonardo Bras , Peter Zijlstra , "Paul E. McKenney" , Juergen Gross , Valentin Schneider , Yury Norov Cc: linux-kernel@vger.kernel.org, linux-trace-kernel@vger.kernel.org Subject: [RFC PATCH v4 1/2] trace,smp: Add tracepoints around remotelly called functions Date: Mon, 15 May 2023 15:30:45 -0300 Message-Id: <20230515183045.654199-2-leobras@redhat.com> X-Mailer: git-send-email 2.40.1 In-Reply-To: <20230515183045.654199-1-leobras@redhat.com> References: <20230515183045.654199-1-leobras@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" The recently added ipi_send_{cpu,cpumask} tracepoints allow finding sources of IPIs targeting CPUs running latency-sensitive applications. For NOHZ_FULL CPUs, all IPIs are interference, and those tracepoints are sufficient to find them and work on getting rid of them. In some setups however, not *all* IPIs are to be suppressed, but long-running IPI callbacks can still be problematic. Add a pair of tracepoints to mark the start and end of processing a CSD IPI callback, similar to what exists for softirq, workqueue or timer callbacks. Signed-off-by: Leonardo Bras --- include/trace/events/smp.h | 45 ++++++++++++++++++++++++++++++++++++++ kernel/smp.c | 25 ++++++++++++++++----- 2 files changed, 64 insertions(+), 6 deletions(-) create mode 100644 include/trace/events/smp.h diff --git a/include/trace/events/smp.h b/include/trace/events/smp.h new file mode 100644 index 000000000000..547f536e7ecd --- /dev/null +++ b/include/trace/events/smp.h @@ -0,0 +1,45 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM smp + +#if !defined(_TRACE_SMP_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_SMP_H + +#include + +/* + * Tracepoints for a function which is called as an effect of smp_call_fun= ction.* + */ +DECLARE_EVENT_CLASS(csd_function, + + TP_PROTO(smp_call_func_t func, call_single_data_t *csd), + + TP_ARGS(func, csd), + + TP_STRUCT__entry( + __field(void *, func) + __field(void *, csd) + ), + + TP_fast_assign( + __entry->func =3D func; + __entry->csd =3D csd; + ), + + TP_printk("func=3D%ps, csd=3D%p", __entry->func, __entry->csd) +); + +DEFINE_EVENT(csd_function, csd_function_entry, + TP_PROTO(smp_call_func_t func, call_single_data_t *csd), + TP_ARGS(func, csd) +); + +DEFINE_EVENT(csd_function, csd_function_exit, + TP_PROTO(smp_call_func_t func, call_single_data_t *csd), + TP_ARGS(func, csd) +); + +#endif /* _TRACE_SMP_H */ + +/* This part must be outside protection */ +#include diff --git a/kernel/smp.c b/kernel/smp.c index 919387be6d4e..eecdd452619c 100644 --- a/kernel/smp.c +++ b/kernel/smp.c @@ -27,6 +27,9 @@ #include =20 #include +#define CREATE_TRACE_POINTS +#include +#undef CREATE_TRACE_POINTS =20 #include "smpboot.h" #include "sched/smp.h" @@ -121,6 +124,14 @@ send_call_function_ipi_mask(struct cpumask *mask) arch_send_call_function_ipi_mask(mask); } =20 +static __always_inline void +csd_do_func(smp_call_func_t func, void *info, struct __call_single_data *c= sd) +{ + trace_csd_function_entry(func, csd); + func(info); + trace_csd_function_exit(func, csd); +} + #ifdef CONFIG_CSD_LOCK_WAIT_DEBUG =20 static DEFINE_STATIC_KEY_MAYBE(CONFIG_CSD_LOCK_WAIT_DEBUG_DEFAULT, csdlock= _debug_enabled); @@ -375,7 +386,7 @@ static int generic_exec_single(int cpu, call_single_dat= a_t *csd) csd_lock_record(csd); csd_unlock(csd); local_irq_save(flags); - func(info); + csd_do_func(func, info, csd); csd_lock_record(NULL); local_irq_restore(flags); return 0; @@ -477,7 +488,7 @@ static void __flush_smp_call_function_queue(bool warn_c= pu_offline) } =20 csd_lock_record(csd); - func(info); + csd_do_func(func, info, csd); csd_unlock(csd); csd_lock_record(NULL); } else { @@ -508,7 +519,7 @@ static void __flush_smp_call_function_queue(bool warn_c= pu_offline) =20 csd_lock_record(csd); csd_unlock(csd); - func(info); + csd_do_func(func, info, csd); csd_lock_record(NULL); } else if (type =3D=3D CSD_TYPE_IRQ_WORK) { irq_work_single(csd); @@ -522,8 +533,10 @@ static void __flush_smp_call_function_queue(bool warn_= cpu_offline) /* * Third; only CSD_TYPE_TTWU is left, issue those. */ - if (entry) - sched_ttwu_pending(entry); + if (entry) { + csd =3D llist_entry(entry, typeof(*csd), node.llist); + csd_do_func(sched_ttwu_pending, entry, csd); + } } =20 =20 @@ -816,7 +829,7 @@ static void smp_call_function_many_cond(const struct cp= umask *mask, unsigned long flags; =20 local_irq_save(flags); - func(info); + csd_do_func(func, info, NULL); local_irq_restore(flags); } =20 --=20 2.40.1 From nobody Sun Feb 8 21:27:11 2026 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id E808EC7EE26 for ; Mon, 15 May 2023 18:32:30 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S243462AbjEOScP (ORCPT ); Mon, 15 May 2023 14:32:15 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:37994 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S242642AbjEOScI (ORCPT ); Mon, 15 May 2023 14:32:08 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 8858418AA9 for ; Mon, 15 May 2023 11:31:31 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1684175490; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=ufrYLf3babsT3oqZ0wZvk4paQN3lMBWwcV5dtS3ah1Q=; b=ddy/GnIq+49gvOfVRiqKPgC3Oq2+TQq7/dnL+EkfzU6heihqgbvbiazs2xCV7Vp/vxEIOi nMr2WLdEV2L+jxr2/VK72bvNmBH3aWdOXOt2ETP9fdZlU1Vfg/M4iBgLMSbFf9LbRfcX+f IdCKc8jAmWgMSR9lOMIOBiHR6T4ZKG4= Received: from mail-oa1-f71.google.com (mail-oa1-f71.google.com [209.85.160.71]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-232-Zx8KvWI3N9KGzAUh5Ize1Q-1; Mon, 15 May 2023 14:31:29 -0400 X-MC-Unique: Zx8KvWI3N9KGzAUh5Ize1Q-1 Received: by mail-oa1-f71.google.com with SMTP id 586e51a60fabf-1877e4dff8aso8561678fac.2 for ; Mon, 15 May 2023 11:31:29 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1684175488; x=1686767488; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=ufrYLf3babsT3oqZ0wZvk4paQN3lMBWwcV5dtS3ah1Q=; b=Fflj8xqTKm5mbANTBZlFF54K1KlGIQ3GbdNVfuccKJoeG296gQHpJH50tNbkwu9m9b 53injwDE5TOrhQC+s/mJV3uHCcBmHuvreqB/bdjOUznXChQxxNwK8tlPSbnO3Jb9Gdm1 Ur9UplTda9KOZR2NskuYRvVL1XSjitBjPgT7Df6hs2tqYyVfR9iKqbYN0JIjj9qvOkWq 9bEg700zETZBtFqqnU2fnXttwNvqo/bP8u6ZZpQnF5/rKx2krk4UIjnr0df/6FEhA8do wTEI+Wxj8Tqv40KiRLXwFTL7d1qaWGepeu2blqUjGQhOVvlfhhJ/oYzE5ySFHYJuQgDE lx4w== X-Gm-Message-State: AC+VfDwr0lrFb03w2jZAxmKuHLA7h2Be5IR3VJMV6zgyeoriDDjw6KkD xe0jnExn0h8kbKgK3tUsNY8qYYurEMTuEKsaeoPWZba+kjVP/+qkEE8gjwIHgKX2ZTgOznXDY1n lkVt5CTiNCX4KiL2+m1lzj8ln X-Received: by 2002:a05:6871:542:b0:17e:cb7:29a3 with SMTP id t2-20020a056871054200b0017e0cb729a3mr20218565oal.15.1684175488163; Mon, 15 May 2023 11:31:28 -0700 (PDT) X-Google-Smtp-Source: ACHHUZ5Y4iVbZSf8VFwKfzpNfODiFE0E+amUJssftuTgUm2b+oSC3QPDreTO3rUAMSlBmnWwIp/nJw== X-Received: by 2002:a05:6871:542:b0:17e:cb7:29a3 with SMTP id t2-20020a056871054200b0017e0cb729a3mr20218555oal.15.1684175487945; Mon, 15 May 2023 11:31:27 -0700 (PDT) Received: from localhost.localdomain ([2804:1b3:a803:46cc:5b68:5c23:dd7a:8cb3]) by smtp.gmail.com with ESMTPSA id ea20-20020a056870071400b001964dc3dadesm6094075oab.45.2023.05.15.11.31.23 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 15 May 2023 11:31:27 -0700 (PDT) From: Leonardo Bras To: Thomas Gleixner , Sebastian Andrzej Siewior , Nadav Amit , Chen Zhongjin , Daniel Bristot de Oliveira , Marcelo Tosatti , Steven Rostedt , Masami Hiramatsu , Leonardo Bras , Peter Zijlstra , "Paul E. McKenney" , Juergen Gross , Valentin Schneider , Yury Norov Cc: linux-kernel@vger.kernel.org, linux-trace-kernel@vger.kernel.org Subject: [RFC PATCH v4 2/2] trace,smp: Add tracepoints for scheduling remotelly called functions Date: Mon, 15 May 2023 15:30:46 -0300 Message-Id: <20230515183045.654199-3-leobras@redhat.com> X-Mailer: git-send-email 2.40.1 In-Reply-To: <20230515183045.654199-1-leobras@redhat.com> References: <20230515183045.654199-1-leobras@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Add a tracepoint for when a CSD is queued to a remote CPU's call_single_queue. This allows finding exactly which CPU queued a given CSD when looking at a csd_function_{entry,exit} event, and also enables us to accurately measure IPI delivery time with e.g. a synthetic event: $ echo 'hist:keys=3Dcpu,csd.hex:ts=3Dcommon_timestamp.usecs' >\ /sys/kernel/tracing/events/smp/csd_queue_cpu/trigger $ echo 'csd_latency unsigned int dst_cpu; unsigned long csd; u64 time' >\ /sys/kernel/tracing/synthetic_events $ echo \ 'hist:keys=3Dcommon_cpu,csd.hex:'\ 'time=3Dcommon_timestamp.usecs-$ts:'\ 'onmatch(smp.csd_queue_cpu).trace(csd_latency,common_cpu,csd,$time)' >\ /sys/kernel/tracing/events/smp/csd_function_entry/trigger $ trace-cmd record -e 'synthetic:csd_latency' hackbench $ trace-cmd report <...>-467 [001] 21.824263: csd_queue_cpu: cpu=3D0 callsite=3D= try_to_wake_up+0x2ea func=3Dsched_ttwu_pending csd=3D0xffff8880076148b8 <...>-467 [001] 21.824280: ipi_send_cpu: cpu=3D0 callsite=3D= try_to_wake_up+0x2ea callback=3Dgeneric_smp_call_function_single_interrupt+= 0x0 <...>-489 [000] 21.824299: csd_function_entry: func=3Dsched_ttwu_p= ending csd=3D0xffff8880076148b8 <...>-489 [000] 21.824320: csd_latency: dst_cpu=3D0, csd=3D= 18446612682193848504, time=3D36 Suggested-by: Valentin Schneider Signed-off-by: Leonardo Bras --- include/trace/events/smp.h | 27 +++++++++++++++++++++++++++ kernel/smp.c | 16 +++++----------- 2 files changed, 32 insertions(+), 11 deletions(-) diff --git a/include/trace/events/smp.h b/include/trace/events/smp.h index 547f536e7ecd..b05259284eb0 100644 --- a/include/trace/events/smp.h +++ b/include/trace/events/smp.h @@ -7,6 +7,33 @@ =20 #include =20 +TRACE_EVENT(csd_queue_cpu, + + TP_PROTO(const unsigned int cpu, + unsigned long callsite, + smp_call_func_t func, + call_single_data_t *csd), + + TP_ARGS(cpu, callsite, func, csd), + + TP_STRUCT__entry( + __field(unsigned int, cpu) + __field(void *, callsite) + __field(void *, func) + __field(void *, csd) + ), + + TP_fast_assign( + __entry->cpu =3D cpu; + __entry->callsite =3D (void *)callsite; + __entry->func =3D func; + __entry->csd =3D csd; + ), + + TP_printk("cpu=3D%u callsite=3D%pS func=3D%ps csd=3D%p", + __entry->cpu, __entry->callsite, __entry->func, __entry->csd) + ); + /* * Tracepoints for a function which is called as an effect of smp_call_fun= ction.* */ diff --git a/kernel/smp.c b/kernel/smp.c index eecdd452619c..e70497f1c3c6 100644 --- a/kernel/smp.c +++ b/kernel/smp.c @@ -340,7 +340,7 @@ void __smp_call_single_queue(int cpu, struct llist_node= *node) * even if we haven't sent the smp_call IPI yet (e.g. the stopper * executes migration_cpu_stop() on the remote CPU). */ - if (trace_ipi_send_cpu_enabled()) { + if (trace_csd_queue_cpu_enabled()) { call_single_data_t *csd; smp_call_func_t func; =20 @@ -348,7 +348,7 @@ void __smp_call_single_queue(int cpu, struct llist_node= *node) func =3D CSD_TYPE(csd) =3D=3D CSD_TYPE_TTWU ? sched_ttwu_pending : csd->func; =20 - trace_ipi_send_cpu(cpu, _RET_IP_, func); + trace_csd_queue_cpu(cpu, _RET_IP_, func, csd); } =20 /* @@ -741,7 +741,7 @@ static void smp_call_function_many_cond(const struct cp= umask *mask, int cpu, last_cpu, this_cpu =3D smp_processor_id(); struct call_function_data *cfd; bool wait =3D scf_flags & SCF_WAIT; - int nr_cpus =3D 0, nr_queued =3D 0; + int nr_cpus =3D 0; bool run_remote =3D false; bool run_local =3D false; =20 @@ -799,21 +799,15 @@ static void smp_call_function_many_cond(const struct = cpumask *mask, csd->node.src =3D smp_processor_id(); csd->node.dst =3D cpu; #endif + trace_csd_queue_cpu(cpu, _RET_IP_, func, csd); + if (llist_add(&csd->node.llist, &per_cpu(call_single_queue, cpu))) { __cpumask_set_cpu(cpu, cfd->cpumask_ipi); nr_cpus++; last_cpu =3D cpu; } - nr_queued++; } =20 - /* - * Trace each smp_function_call_*() as an IPI, actual IPIs - * will be traced with func=3D=3Dgeneric_smp_call_function_single_ipi(). - */ - if (nr_queued) - trace_ipi_send_cpumask(cfd->cpumask, _RET_IP_, func); - /* * Choose the most efficient way to send an IPI. Note that the * number of CPUs might be zero due to concurrent changes to the --=20 2.40.1