linux-kernel - Re: [PATCH v2] delayaccy/sched: add SOFTIRQ delay

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20250827111444659yr8tENNnk9xU2ZFnz94FO@zte.com.cn>
Date: Wed, 27 Aug 2025 11:14:44 +0800 (CST)
From: <wang.yaxin@....com.cn>
To: <tiozhang@...iglobal.com>
Cc: <akpm@...ux-foundation.org>, <fan.yu9@....com.cn>, <corbet@....net>,
        <bsingharora@...il.com>, <yang.yang29@....com.cn>,
        <linux-kernel@...r.kernel.org>, <linux-doc@...r.kernel.org>,
        <mingo@...hat.com>, <peterz@...radead.org>, <juri.lelli@...hat.com>,
        <vincent.guittot@...aro.org>, <dietmar.eggemann@....com>,
        <rostedt@...dmis.org>, <bsegall@...gle.com>, <mgorman@...e.de>,
        <vschneid@...hat.com>, <jiang.kun2@....com.cn>, <xu.xin16@....com.cn>,
        <tiozhang@...iglobal.com>, <zyhtheonly@...il.com>,
        <zyhtheonly@...h.net>
Subject: Re: [PATCH v2] delayaccy/sched: add SOFTIRQ delay

> Taking the difference of two successive readings of a given
> counter (say cpu_delay_total) for a task will give the delay
>@@ -123,6 +124,8 @@ Get sum and peak of delays, since system boot, for all pids with tgid 242::
>                   156       11215873          0.072ms     0.207403ms     0.033913ms
>     IRQ         count    delay total  delay average      delay max      delay min
>                     0              0          0.000ms     0.000000ms     0.000000ms
>+    SOFTIRQ     count    delay total  delay average      delay max      delay min
>+                    0              0          0.000ms     0.000000ms     0.000000ms
>
> Get IO accounting for pid 1, it works only with -p::

If possible, you can construct some abnormal scenarios to assign values to IRQ and SOFTIRQ,
highlighting the differences between them. Additionally, this should cover the testing of
new features. If the delay info is entirely zero, it may fail to demonstrate such differences.

>+    /*
>+     * We only account softirq time when we are called by
>+     * account_softirq_enter{,exit}
>+     * and we do not account ksoftirqd here.
>+     */
>+    if (curr != this_cpu_ksoftirqd() &&
>+        ((offset & SOFTIRQ_OFFSET) || (pc & SOFTIRQ_OFFSET))) {
>+        delta_soft = cpu_clock - irqtime->soft_start_time;
>+        irqtime->soft_start_time += delta_soft;
>+    }
>+
>     /*
>      * We do not account for softirq time from ksoftirqd here.
>      * We want to continue accounting softirq time to ksoftirqd thread
>@@ -75,9 +88,9 @@ void irqtime_account_irq(struct task_struct *curr, unsigned int offset)
>      * that do not consume any time, but still wants to run.
>      */
>     if (pc & HARDIRQ_MASK)
>-        irqtime_account_delta(irqtime, delta, CPUTIME_IRQ);
>+        irqtime_account_delta(irqtime, delta, delta_soft, CPUTIME_IRQ);
>     else if ((pc & SOFTIRQ_OFFSET) && curr != this_cpu_ksoftirqd())
>-        irqtime_account_delta(irqtime, delta, CPUTIME_SOFTIRQ);
>+        irqtime_account_delta(irqtime, delta, delta_soft, CPUTIME_SOFTIRQ);
> }
>
> static u64 irqtime_tick_accounted(u64 maxtime)

As you mentioned, delta_soft represents SOFTIRQ, but it appears to have been accumulated once
under the condition of (pc & HARDIRQ_MASK). Is there a potential double-counting issue?
replacing delta_soft with 0 when in HARDIRQ might make it more intuitive.

Thanks,
Yaxin