From nobody Wed Dec 24 01:50:06 2025 Received: from out30-97.freemail.mail.aliyun.com (out30-97.freemail.mail.aliyun.com [115.124.30.97]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 823CD5477C for ; Tue, 30 Jan 2024 07:47:57 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=115.124.30.97 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1706600881; cv=none; b=sGGgjHVUR6Ku8P2O9ivMY/gecSOIh2a5ML/Oqgw4mYOm/M/fck23wUcxmI0UWNJnu0zBiW4VIAnNlCZ7z3i6u6Z9ln4Rf3nLMzMruMj9KFXdQGPbuAHQYRXlZQ1UbHHxOOn96JChPqRjD46jNF550t25AJzv/SjEmM6uPPjTzDg= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1706600881; c=relaxed/simple; bh=P40AWZnIMfAe79apwlrCNlsMdNUznsPqMxrTgzPvmF4=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=erD3nyJvm+zwX5fVYNOSL5lSXMmtslILYkH+dBcYANKCGmiEoWwY+Ge4PiPZpxufDGAqQxFiAlhxhhTDuWO/3TBvBqdc4smS0q59BOwmmtRhYCiSYuEU1rH0OvNktxYWJAd+4dq3wu8a7JbaTj9TCJFf+ZIIF7xuJVFvsUo/gCQ= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.alibaba.com; spf=pass smtp.mailfrom=linux.alibaba.com; dkim=pass (1024-bit key) header.d=linux.alibaba.com header.i=@linux.alibaba.com header.b=LLEDZpsc; arc=none smtp.client-ip=115.124.30.97 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.alibaba.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.alibaba.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux.alibaba.com header.i=@linux.alibaba.com header.b="LLEDZpsc" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux.alibaba.com; s=default; t=1706600876; h=From:To:Subject:Date:Message-Id:MIME-Version; bh=jxEQopJGdzQZWNbL3DL8o7X/qrGyOjoHI79VdE+/YrA=; b=LLEDZpscNFtDixQfRtVzxSWhWcAB5LVSbxCk+krEF8INtjG1K/wWOcBxQNU52OxdAhReCfjG8iCoLRoSo9IZroTIy1HlyyMKdV4CY/LW7SLyCYvYCVYdE/ZtAFEHnebECQsTWHNzAzLZkTmgsbvQUp5r0nrm5vv4C6eEUfL85Cs= X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R421e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=ay29a033018045192;MF=yaoma@linux.alibaba.com;NM=1;PH=DS;RN=8;SR=0;TI=SMTPD_---0W.fTvc0_1706600873; Received: from localhost.localdomain(mailfrom:yaoma@linux.alibaba.com fp:SMTPD_---0W.fTvc0_1706600873) by smtp.aliyun-inc.com; Tue, 30 Jan 2024 15:47:55 +0800 From: Bitao Hu To: dianders@chromium.org, akpm@linux-foundation.org, pmladek@suse.com, lecopzer.chen@mediatek.com, kernelfans@gmail.com, liusong@linux.alibaba.com Cc: linux-kernel@vger.kernel.org, yaoma@linux.alibaba.com Subject: [PATCHv2 2/2] watchdog/softlockup: report the most frequent interrupts Date: Tue, 30 Jan 2024 15:47:44 +0800 Message-Id: <20240130074744.45759-3-yaoma@linux.alibaba.com> X-Mailer: git-send-email 2.37.1 (Apple Git-137.1) In-Reply-To: <20240130074744.45759-1-yaoma@linux.alibaba.com> References: <20240130074744.45759-1-yaoma@linux.alibaba.com> 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" When the watchdog determines that the current soft lockup is due to an interrupt storm based on CPU utilization, reporting the most frequent interrupts could be good enough for further troubleshooting. Below is an example of interrupt storm. The call tree does not provide useful information, but we can analyze which interrupt caused the soft lockup by comparing the counts of interrupts. [ 2987.488075] watchdog: BUG: soft lockup - CPU#9 stuck for 23s! [kworker/9= :1:214] [ 2987.488607] CPU#9 Utilization every 4s during lockup: [ 2987.488941] #1: 0% system, 0% softirq, 100% hardirq, 0= % idle [ 2987.489357] #2: 0% system, 0% softirq, 100% hardirq, 0= % idle [ 2987.489771] #3: 0% system, 0% softirq, 100% hardirq, 0= % idle [ 2987.490186] #4: 0% system, 0% softirq, 100% hardirq, 0= % idle [ 2987.490601] #5: 0% system, 0% softirq, 100% hardirq, 0= % idle [ 2987.491034] CPU#9 Detect HardIRQ Time exceeds 50%. Most frequent HardIRQ= s: [ 2987.491493] #1: 330985 irq#7(IPI) [ 2987.491743] #2: 5000 irq#10(arch_timer) [ 2987.492039] #3: 9 irq#91(nvme0q2) [ 2987.492318] #4: 3 irq#118(virtio1-output.12) ... [ 2987.492728] Call trace: [ 2987.492729] __do_softirq+0xa8/0x364 Signed-off-by: Bitao Hu --- kernel/watchdog.c | 150 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 150 insertions(+) diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 0efe9604c3c2..38fb18e17d71 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -25,6 +25,9 @@ #include #include #include +#include +#include +#include =20 #include #include @@ -431,11 +434,15 @@ void touch_softlockup_watchdog_sync(void) __this_cpu_write(watchdog_report_ts, SOFTLOCKUP_DELAY_REPORT); } =20 +static void set_potential_softlockup(unsigned long now, unsigned long touc= h_ts); + static int is_softlockup(unsigned long touch_ts, unsigned long period_ts, unsigned long now) { if ((watchdog_enabled & WATCHDOG_SOFTOCKUP_ENABLED) && watchdog_thresh) { + /* Softlockup may occur in the current period */ + set_potential_softlockup(now, period_ts); /* Warn about unreasonable delays. */ if (time_after(now, period_ts + get_softlockup_thresh())) return now - touch_ts; @@ -457,6 +464,8 @@ static enum cpu_usage_stat idx_to_stat[NUM_STATS_PER_GR= OUP] =3D { CPUTIME_SYSTEM, CPUTIME_SOFTIRQ, CPUTIME_IRQ, CPUTIME_IDLE }; =20 +static void print_hardirq_counts(void); + static void update_cpustat(void) { u8 i; @@ -504,10 +513,150 @@ static void print_cpustat(void) utilization[i][STATS_SYSTEM], utilization[i][STATS_SOFTIRQ], utilization[i][STATS_HARDIRQ], utilization[i][STATS_IDLE]); } + print_hardirq_counts(); +} + +#define HARDIRQ_PERCENT_THRESH 50 +#define NUM_HARDIRQ_REPORT 5 +static DECLARE_BITMAP(softlockup_hardirq_cpus, CONFIG_NR_CPUS); +static DEFINE_PER_CPU(u32 *, hardirq_counts); + +static void find_counts_top(u32 *irq_counts, int *irq, u32 perirq_counts, = int perirq_id, int range) +{ + unsigned int i, j; + + for (i =3D 0; i < range; i++) { + if (perirq_counts > irq_counts[i]) { + for (j =3D range - 1; j > i; j--) { + irq_counts[j] =3D irq_counts[j - 1]; + irq[j] =3D irq[j - 1]; + } + irq_counts[j] =3D perirq_counts; + irq[j] =3D perirq_id; + break; + } + } +} + +/* + * If the proportion of time spent handling irq exceeds HARDIRQ_PERCENT_TH= RESH% + * during sample_period, then it is necessary to record the counts of each= irq. + */ +static inline bool need_record_irq_counts(int type) +{ + int tail =3D this_cpu_read(cpustat_tail); + u8 utilization; + + if (--tail =3D=3D -1) + tail =3D 4; + utilization =3D this_cpu_read(cpustat_utilization[tail][type]); + return utilization > HARDIRQ_PERCENT_THRESH; } + +/* + * Mark softlockup as potentially caused by hardirq + */ +static void set_potential_softlockup_hardirq(void) +{ + u32 i; + u32 *counts =3D __this_cpu_read(hardirq_counts); + int cpu =3D smp_processor_id(); + struct irq_desc *desc; + + if (!need_record_irq_counts(STATS_HARDIRQ)) + return; + + if (!test_bit(cpu, softlockup_hardirq_cpus)) { + counts =3D kmalloc_array(nr_irqs, sizeof(u32), GFP_ATOMIC); + if (!counts) + return; + for_each_irq_desc(i, desc) { + if (!desc) + continue; + counts[i] =3D desc->kstat_irqs ? + *this_cpu_ptr(desc->kstat_irqs) : 0; + } + __this_cpu_write(hardirq_counts, counts); + set_bit(cpu, softlockup_hardirq_cpus); + } +} + +static void clear_potential_softlockup_hardirq(void) +{ + u32 *counts =3D __this_cpu_read(hardirq_counts); + int cpu =3D smp_processor_id(); + + if (test_bit(cpu, softlockup_hardirq_cpus)) { + kfree(counts); + counts =3D NULL; + __this_cpu_write(hardirq_counts, counts); + clear_bit(cpu, softlockup_hardirq_cpus); + } +} + +/* + * Mark that softlockup may occur + */ +static void set_potential_softlockup(unsigned long now, unsigned long peri= od_ts) +{ + if (time_after_eq(now, period_ts + get_softlockup_thresh() / 5)) + set_potential_softlockup_hardirq(); +} + +static void clear_potential_softlockup(void) +{ + clear_potential_softlockup_hardirq(); +} + +static void print_hardirq_counts(void) +{ + u32 i; + struct irq_desc *desc; + u32 counts_diff; + u32 *counts =3D __this_cpu_read(hardirq_counts); + int cpu =3D smp_processor_id(); + u32 hardirq_counts_top[NUM_HARDIRQ_REPORT] =3D {0, 0, 0, 0, 0}; + int hardirq_top[NUM_HARDIRQ_REPORT] =3D {-1, -1, -1, -1, -1}; + + if (test_bit(cpu, softlockup_hardirq_cpus)) { + /* Find the top NUM_HARDIRQ_REPORT most frequent interrupts */ + for_each_irq_desc(i, desc) { + if (!desc) + continue; + counts_diff =3D desc->kstat_irqs ? + *this_cpu_ptr(desc->kstat_irqs) - counts[i] : 0; + find_counts_top(hardirq_counts_top, hardirq_top, + counts_diff, i, NUM_HARDIRQ_REPORT); + } + /* + * We do not want the "watchdog: " prefix on every line, + * hence we use "printk" instead of "pr_crit". + */ + printk(KERN_CRIT "CPU#%d Detect HardIRQ Time exceeds %d%%. Most frequent= HardIRQs:\n", + smp_processor_id(), HARDIRQ_PERCENT_THRESH); + for (i =3D 0; i < NUM_HARDIRQ_REPORT; i++) { + if (hardirq_top[i] =3D=3D -1) + break; + desc =3D irq_to_desc(hardirq_top[i]); + if (desc && desc->action) + printk(KERN_CRIT "\t#%u: %-10u\tirq#%d(%s)\n", + i+1, hardirq_counts_top[i], + hardirq_top[i], desc->action->name); + else + printk(KERN_CRIT "\t#%u: %-10u\tirq#%d\n", + i+1, hardirq_counts_top[i], + hardirq_top[i]); + } + if (!need_record_irq_counts(STATS_HARDIRQ)) + clear_potential_softlockup_hardirq(); + } +} + #else static inline void update_cpustat(void) { } static inline void print_cpustat(void) { } +static inline void set_potential_softlockup(unsigned long now, unsigned lo= ng period_ts) { } +static inline void clear_potential_softlockup(void) { } #endif =20 /* watchdog detector functions */ @@ -525,6 +674,7 @@ static DEFINE_PER_CPU(struct cpu_stop_work, softlockup_= stop_work); static int softlockup_fn(void *data) { update_touch_ts(); + clear_potential_softlockup(); complete(this_cpu_ptr(&softlockup_completion)); =20 return 0; --=20 2.37.1 (Apple Git-137.1)