[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <36cb15b2-a8e0-1beb-3462-16a79d41c40d@huawei.com>
Date: Thu, 16 Jun 2022 09:31:47 +0800
From: Yang Jihong <yangjihong1@...wei.com>
To: Namhyung Kim <namhyung@...nel.org>
CC: Peter Zijlstra <peterz@...radead.org>,
Ingo Molnar <mingo@...hat.com>,
Arnaldo Carvalho de Melo <acme@...nel.org>,
Mark Rutland <mark.rutland@....com>,
Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
Jiri Olsa <jolsa@...nel.org>,
linux-kernel <linux-kernel@...r.kernel.org>,
linux-perf-users <linux-perf-users@...r.kernel.org>
Subject: Re: [RFC 06/13] perf kwork: Implement perf kwork report
Hello,
On 2022/6/16 6:01, Namhyung Kim wrote:
> On Tue, Jun 14, 2022 at 8:12 PM Yang Jihong <yangjihong1@...wei.com> wrote:
>>
>> Hello,
>>
>> On 2022/6/15 5:49, Namhyung Kim wrote:
>>> On Mon, Jun 13, 2022 at 2:48 AM Yang Jihong <yangjihong1@...wei.com> wrote:
>>>>
>>>> Implements framework of perf kwork report, which is used to report time
>>>> properties such as run time and frequency:
>>>>
>>>> test case:
>>>>
>>>> Kwork Name | Cpu | Avg delay | Frequency | Max delay | Max delay start | Max delay end |
>>>> ---------------------------------------------------------------------------------------------------------------------------
>>>> ---------------------------------------------------------------------------------------------------------------------------
>>>>
>>>> Since there are no report supported events, the output is empty.
>>>>
>>>> Briefly describe the data structure:
>>>> 1. "class" indicates event type. For example, irq and softiq correspond
>>>> to different types.
>>>> 2. "cluster" refers to a specific event corresponding to a type. For
>>>> example, RCU and TIMER in softirq correspond to different clusters,
>>>> which contains three types of events: raise, entry, and exit.
>>>
>>> I'm not good at naming but how about calling it "work" instead?
>>> It's all about kernel works anyway..
>>>
>> How about calling it "work_atoms" ?
>> Otherwise, "work" feels a little similar to kwork_class in the first step.
>
> Didn't you already have "atom"? I think we can have a hierarchy like
> class (IRQ) - work (eth0) - atom.
OK, I'll modify it in next version according to this hierarchy.
Thanks,
Jihong
.
Powered by blists - more mailing lists