From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-pg1-f173.google.com (mail-pg1-f173.google.com [209.85.215.173]) (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 2091B322DB7 for ; Wed, 15 Oct 2025 11:49:58 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.215.173 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1760529000; cv=none; b=f5kXORVyJy9HanaIqW/jBA06HgiiZFJlM71LO9MaqBmvfUesbfJ2jGbW15SKK20j0RGuUfQGlLZftPOdZMG0gYXeKlFhm+bYrAlVODdluHz7ojzNDKgmLUlx6+HjumN8NawIWGLymQx8vidHG/vOAohBIO1rcF2bX6rHwChAsfc= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1760529000; c=relaxed/simple; bh=67bPLpk4XtiM2chNuj/NIzxAtVYUBkPeAY7FycKz6x8=; h=From:To:Cc:Subject:Date:Message-ID:MIME-Version; b=HeVWoPQfcVdzuZoZqUteo58fQOlJ6JgSvcvo++YHiiJdREUDh1F7FGIQ+2r8qdlF8ySEJvLlb/FkmhpItMuJnxDtUc+a+9QnIsx0bk4z0bm8DFVGXK5GtHxxo0qiRGcconuYpjGUDzjPflQAdbHbyWHe5QTzP7eiY63uikRsJzA= 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=hk/tktpQ; arc=none smtp.client-ip=209.85.215.173 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="hk/tktpQ" Received: by mail-pg1-f173.google.com with SMTP id 41be03b00d2f7-b555ab7fabaso5944868a12.0 for ; Wed, 15 Oct 2025 04:49:58 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1760528998; x=1761133798; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:message-id:date:subject:cc :to:from:from:to:cc:subject:date:message-id:reply-to; bh=o7DX/MsjusdVTQDdPxzEiOw94/h8+AXMbc90Y8BQxVw=; b=hk/tktpQXkqqHRT7an5sKJ65tB4T2xXnYJz6ktReFg7+0UR6ndpyr3Ld6K1JnI4YW8 dKJchGk1WPnPrIkypO2+GG+DYc3xEW/ALRkMgME9b3YOvmy+M+YvQizzCN3DQTNfNfEt 5fNOHtip/toozfFzzNj9TQum3yED16IFUZkMPZvHypHccBb/8vCNKs/C2Rz0cV5j8H/P GIiD+pi8/PuN0httHMaO87M6SPVGw0jnISAoR0+SMWPL7DACV5lmzWmrDh321SDNM1LO fwqtTNda8+643ySqhWjBWJrAyk3R0x63PGj1W6fqIYv9DPHZmFVq4K9YRGazVHRXR3pC D7YQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1760528998; x=1761133798; h=content-transfer-encoding:mime-version:message-id:date:subject:cc :to:from:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to; bh=o7DX/MsjusdVTQDdPxzEiOw94/h8+AXMbc90Y8BQxVw=; b=QslSDyjLPC2itDYXqh53IRehtuWk763+hcaX67lTIR5h3FDpxTDeq3t66NOOmK1yPd lngoFEZYUevbE6oj0rWGTasXApdMWvdr/qC8FNMD4OVkUpSl1T3xNWONlpZlp7eu/s8R 0B1ysXVmJn574OnepMRz454r8mkK23cf3yUKF48KFxIsaYB7GKaG6r1nlqHRazwntZKr H8aFsGaTSnsynyfktrOXHWi9flm3BIkTNzuKd9eJkA61fRJVGjCNW8JmQ/pP5NLCur23 oMZAdxBs5/7MhIQdaHqbGZeX845W3YtjqDb8MypQFWx1uQpx+9PrchYnx62I+vpCgb9D 5V/w== X-Forwarded-Encrypted: i=1; AJvYcCVCcDzaQK+mJKMtfPzcGTw5e5S8Us5nzozCU9T5uCpncVtRlrd8vHxwBqeAO6jIkrp8sB3VgfnTnIy89XeZMSOGsMM=@vger.kernel.org X-Gm-Message-State: AOJu0YzcNAKo3P+gsvzHudnyCTQAwA9gDyipYP2CtpCAqZCKtxhJk7Wg tQwBkKQiatxa4JiJuRyedz2Ao3DNyKnG3jf/i+iAXEbo8HyM294qMUNb/1kTfvvL X-Gm-Gg: ASbGncvy5JiCUn6FS7gkYvdQ88abMPbIFibOsPsNMQiaol+3h9XlMKtGxefxYnK78sr kh4tJK1uvWAp0JBXGiajFa/tTxpgUjxBunhtv+sqjcgwwAxcrqbZeR2JQu8iyOIFMgyzzNzqLSs mXkV7MRl15Us/EpUr1jU4KtJRyrbF/nTFyXHsXS8DKJ6XqmLurQocrt2VykCWuAmHv0aLHy+UyR X1pWGEF/KQNFjU6B+y72SszNDdyfPg57j0OPm3L2Vr6WPulkVNJj7ZAv3aoz7RNYSvtZhjHw7HU 1TM96XV43JNdA+h80OnGSJiUeVV6wyPFhaijcyCyv8kMky0PGvuUrK7TTqJwWJMoflvqGvmZ+d1 jUt/UWScYpP97dcowhBNQINWpeFznl2yk409Xb3q06+5KGFtF1+JjtO3jHqyOfzrY3teb9sN8rQ idSasP5L7y4Vr/+YA= X-Google-Smtp-Source: AGHT+IHA7jmacf+2jyZ8effG47uhvpwUPR/qPm49iok5/wykenkm7y4wr4qNXbSUS6lRbQH9DgwwKA== X-Received: by 2002:a17:90b:33d1:b0:339:d1f0:c740 with SMTP id 98e67ed59e1d1-33b51149673mr39484383a91.1.1760528998341; Wed, 15 Oct 2025 04:49:58 -0700 (PDT) Received: from VM-119-80-tencentos.localdomain ([14.22.11.162]) by smtp.gmail.com with ESMTPSA id 98e67ed59e1d1-33b52a4f8bfsm9632658a91.10.2025.10.15.04.49.55 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 15 Oct 2025 04:49:57 -0700 (PDT) From: Yongliang Gao To: rostedt@goodmis.org, mhiramat@kernel.org, mathieu.desnoyers@efficios.com Cc: linux-kernel@vger.kernel.org, linux-trace-kernel@vger.kernel.org, Yongliang Gao , Huang Cun Subject: [PATCH] trace/pid_list: optimize pid_list->lock contention Date: Wed, 15 Oct 2025 19:49:52 +0800 Message-ID: <20251015114952.4014352-1-leonylgao@gmail.com> X-Mailer: git-send-email 2.43.5 Precedence: bulk X-Mailing-List: linux-trace-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit From: Yongliang Gao When the system has many cores and task switching is frequent, setting set_ftrace_pid can cause frequent pid_list->lock contention and high system sys usage. For example, in a vmcore environment with 288 cores, We found 267 CPUs are in pid_list->lock contention. #4 [ffffa6226fb4bc70] native_queued_spin_lock_slowpath at ffffffff99cd4b7e #5 [ffffa6226fb4bc90] _raw_spin_lock_irqsave at ffffffff99cd3e36 #6 [ffffa6226fb4bca0] trace_pid_list_is_set at ffffffff99267554 #7 [ffffa6226fb4bcc0] trace_ignore_this_task at ffffffff9925c288 #8 [ffffa6226fb4bcd8] ftrace_filter_pid_sched_switch_probe at ffffffff99246efe #9 [ffffa6226fb4bcf0] __schedule at ffffffff99ccd161 Signed-off-by: Yongliang Gao Reviewed-by: Huang Cun --- kernel/trace/pid_list.c | 26 +++++++++++++------------- kernel/trace/pid_list.h | 2 +- 2 files changed, 14 insertions(+), 14 deletions(-) diff --git a/kernel/trace/pid_list.c b/kernel/trace/pid_list.c index 090bb5ea4a19..62082a4f60db 100644 --- a/kernel/trace/pid_list.c +++ b/kernel/trace/pid_list.c @@ -138,14 +138,14 @@ bool trace_pid_list_is_set(struct trace_pid_list *pid_list, unsigned int pid) if (pid_split(pid, &upper1, &upper2, &lower) < 0) return false; - raw_spin_lock_irqsave(&pid_list->lock, flags); + read_lock_irqsave(&pid_list->lock, flags); upper_chunk = pid_list->upper[upper1]; if (upper_chunk) { lower_chunk = upper_chunk->data[upper2]; if (lower_chunk) ret = test_bit(lower, lower_chunk->data); } - raw_spin_unlock_irqrestore(&pid_list->lock, flags); + read_unlock_irqrestore(&pid_list->lock, flags); return ret; } @@ -177,7 +177,7 @@ int trace_pid_list_set(struct trace_pid_list *pid_list, unsigned int pid) if (pid_split(pid, &upper1, &upper2, &lower) < 0) return -EINVAL; - raw_spin_lock_irqsave(&pid_list->lock, flags); + write_lock_irqsave(&pid_list->lock, flags); upper_chunk = pid_list->upper[upper1]; if (!upper_chunk) { upper_chunk = get_upper_chunk(pid_list); @@ -199,7 +199,7 @@ int trace_pid_list_set(struct trace_pid_list *pid_list, unsigned int pid) set_bit(lower, lower_chunk->data); ret = 0; out: - raw_spin_unlock_irqrestore(&pid_list->lock, flags); + write_unlock_irqrestore(&pid_list->lock, flags); return ret; } @@ -229,7 +229,7 @@ int trace_pid_list_clear(struct trace_pid_list *pid_list, unsigned int pid) if (pid_split(pid, &upper1, &upper2, &lower) < 0) return -EINVAL; - raw_spin_lock_irqsave(&pid_list->lock, flags); + write_lock_irqsave(&pid_list->lock, flags); upper_chunk = pid_list->upper[upper1]; if (!upper_chunk) goto out; @@ -250,7 +250,7 @@ int trace_pid_list_clear(struct trace_pid_list *pid_list, unsigned int pid) } } out: - raw_spin_unlock_irqrestore(&pid_list->lock, flags); + write_unlock_irqrestore(&pid_list->lock, flags); return 0; } @@ -282,7 +282,7 @@ int trace_pid_list_next(struct trace_pid_list *pid_list, unsigned int pid, if (pid_split(pid, &upper1, &upper2, &lower) < 0) return -EINVAL; - raw_spin_lock_irqsave(&pid_list->lock, flags); + read_lock_irqsave(&pid_list->lock, flags); for (; upper1 <= UPPER_MASK; upper1++, upper2 = 0) { upper_chunk = pid_list->upper[upper1]; @@ -302,7 +302,7 @@ int trace_pid_list_next(struct trace_pid_list *pid_list, unsigned int pid, } found: - raw_spin_unlock_irqrestore(&pid_list->lock, flags); + read_unlock_irqrestore(&pid_list->lock, flags); if (upper1 > UPPER_MASK) return -1; @@ -339,10 +339,10 @@ static void pid_list_refill_irq(struct irq_work *iwork) int lcnt = 0; again: - raw_spin_lock(&pid_list->lock); + write_lock(&pid_list->lock); upper_count = CHUNK_ALLOC - pid_list->free_upper_chunks; lower_count = CHUNK_ALLOC - pid_list->free_lower_chunks; - raw_spin_unlock(&pid_list->lock); + write_unlock(&pid_list->lock); if (upper_count <= 0 && lower_count <= 0) return; @@ -369,7 +369,7 @@ static void pid_list_refill_irq(struct irq_work *iwork) lcnt++; } - raw_spin_lock(&pid_list->lock); + write_lock(&pid_list->lock); if (upper) { *upper_next = pid_list->upper_list; pid_list->upper_list = upper; @@ -380,7 +380,7 @@ static void pid_list_refill_irq(struct irq_work *iwork) pid_list->lower_list = lower; pid_list->free_lower_chunks += lcnt; } - raw_spin_unlock(&pid_list->lock); + write_unlock(&pid_list->lock); /* * On success of allocating all the chunks, both counters @@ -418,7 +418,7 @@ struct trace_pid_list *trace_pid_list_alloc(void) init_irq_work(&pid_list->refill_irqwork, pid_list_refill_irq); - raw_spin_lock_init(&pid_list->lock); + rwlock_init(&pid_list->lock); for (i = 0; i < CHUNK_ALLOC; i++) { union upper_chunk *chunk; diff --git a/kernel/trace/pid_list.h b/kernel/trace/pid_list.h index 62e73f1ac85f..da200834f4ad 100644 --- a/kernel/trace/pid_list.h +++ b/kernel/trace/pid_list.h @@ -76,7 +76,7 @@ union upper_chunk { }; struct trace_pid_list { - raw_spinlock_t lock; + rwlock_t lock; struct irq_work refill_irqwork; union upper_chunk *upper[UPPER1_SIZE]; // 1 or 2K in size union upper_chunk *upper_list; -- 2.43.5