[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <202212072030084707211@zte.com.cn>
Date:   Wed, 7 Dec 2022 20:30:08 +0800 (CST)
From:   <yang.yang29@....com.cn>
To:     <edumazet@...gle.com>, <kuba@...nel.org>
Cc:     <kuba@...nel.org>, <davem@...emloft.net>, <pabeni@...hat.com>,
        <bigeasy@...utronix.de>, <imagedong@...cent.com>,
        <kuniyu@...zon.com>, <petrm@...dia.com>, <liu3101@...due.edu>,
        <wujianguo@...natelecom.cn>, <netdev@...r.kernel.org>,
        <linux-kernel@...r.kernel.org>, <tedheadster@...il.com>
Subject: Re: [PATCH linux-next] net: record times of netdev_budget exhausted
> Presumably, modern tracing techniques can let you do what you want
> without adding new counters.
By the way, should we add a tracepoint likes trace_napi_poll() to make
it easier? Something likes:
        if (unlikely(budget <= 0 ||
                 time_after_eq(jiffies, time_limit))) {
            sd->time_squeeze++;
+            trace_napi_poll(budget, jiffies, time_limit);
            break;
        }
Powered by blists - more mailing lists
 
