linux-kernel - Re: [PATCH] rcu-scale: change rcu-scale report.

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: <405e1757-45c7-583c-6e1f-1ed54cd935ac@windriver.com>
Date:   Tue, 22 Jun 2021 18:56:24 +0800
From:   "Han, Jiangong" <Jiangong.Han@...driver.com>
To:     paulmck@...nel.org
Cc:     dave@...olabs.net, josh@...htriplett.org, rostedt@...dmis.org,
        mathieu.desnoyers@...icios.com, jiangshanlai@...il.com,
        joel@...lfernandes.org, linux-kernel@...r.kernel.org,
        rcu@...r.kernel.org
Subject: Re: [PATCH] rcu-scale: change rcu-scale report.

and the new edition for reviewing is here: 
https://lkml.org/lkml/2021/6/22/327

Thanks for you time.
Br,
Jiangong.


在 2021/6/22 下午6:48, Han, Jiangong 写道:
> Sorry, my the comment is wrong, and the correct comment should be like 
> below, and I had modify this, and I had resend one more applying 
> review patch email for this.
>
> Thanks for you time.
>
> Br,
>
> Jiangong.
>
> ----------------------------------
>     rcu-scale: rcu-scale returns one less than the real number of gps 
> in the dmesg report.
>
>     The dmesg report on rcu-scale shows there are N grace periods, and 
> gps
>     are listed from 0 to N.
>     This commit make writer_n_durations stores the counts of gps,
>     and shows there are N+1 gps, change the listed gps index begin 
> from 0 to N.
>
>     From
>         [ 8306.087880] rcu-scale: writer 0 gps: 133
>         ......
>         [ 8307.864630] rcu-scale:    0 writer-duration:     0 44003961
>         [ 8307.935711] rcu-scale:    0 writer-duration:     1 32003582
>         ......
>         [ 8316.472860] rcu-scale:    0 writer-duration:   132 28004391
>         [ 8316.538498] rcu-scale:    0 writer-duration:   133 27996410
>
>     to
>         [ 8306.087880] rcu-scale: writer 0 gps: 134
>         ......
>         [ 8307.864630] rcu-scale:    0 writer-duration:     0 44003961
>         [ 8307.935711] rcu-scale:    0 writer-duration:     1 32003582
>         ......
>         [ 8316.472860] rcu-scale:    0 writer-duration:   132 28004391
>         [ 8316.538498] rcu-scale:    0 writer-duration:   133 27996410
>
> ----------------------------------
>
>
>
>  2021/6/10 上午4:37, Paul E. McKenney 写道:
>> [Please note: This e-mail is from an EXTERNAL e-mail address]
>>
>> On Sat, Jun 05, 2021 at 03:00:19PM +0800, Jiangong.Han wrote:
>>> The report on rcu-scale shows there are N grace periods, and gps
>>> are listed from 0 to N-1.
>>> This commit make writer_n_durations stores the counts of gps,
>>> change the listed gps index begin from 1 to N.
>>>
>>> From
>>>      [ 8306.087880] rcu-scale: writer 0 gps: 133
>>>      ......
>>>      [ 8307.864630] rcu-scale:    0 writer-duration:     0 44003961
>>>      [ 8307.935711] rcu-scale:    0 writer-duration:     1 32003582
>>>      ......
>>>      [ 8316.472860] rcu-scale:    0 writer-duration:   131 28004391
>>>      [ 8316.538498] rcu-scale:    0 writer-duration:   132 27996410
>>>
>>> to
>>>      [ 8306.087880] rcu-scale: writer 0 gps: 133
>>>      ......
>>>      [ 8307.864630] rcu-scale:    0 writer-duration:     1 44003961
>>>      [ 8307.935711] rcu-scale:    0 writer-duration:     2 32003582
>>>      ......
>>>      [ 8316.472860] rcu-scale:    0 writer-duration:   132 28004391
>>>      [ 8316.538498] rcu-scale:    0 writer-duration:   133 27996410
>>>
>>> Signed-off-by: Jiangong.Han <jiangong.han@...driver.com>
>> You lost me on this one.  Why is this helpful?  And how does the change
>> shown below actually result in the output shown above, given that
>> rcu_scale_cleanup() still starts j at zero?
>>
>>                                                          Thanx, Paul
>>
>>> ---
>>>   kernel/rcu/rcuscale.c | 4 ++--
>>>   1 file changed, 2 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/kernel/rcu/rcuscale.c b/kernel/rcu/rcuscale.c
>>> index dca51fe9c73f..2cc34a22a506 100644
>>> --- a/kernel/rcu/rcuscale.c
>>> +++ b/kernel/rcu/rcuscale.c
>>> @@ -487,7 +487,7 @@ rcu_scale_writer(void *arg)
>>>        if (gp_async) {
>>>                cur_ops->gp_barrier();
>>>        }
>>> -     writer_n_durations[me] = i_max;
>>> +     writer_n_durations[me] = i_max + 1;
>>>        torture_kthread_stopping("rcu_scale_writer");
>>>        return 0;
>>>   }
>>> @@ -561,7 +561,7 @@ rcu_scale_cleanup(void)
>>>                        wdpp = writer_durations[i];
>>>                        if (!wdpp)
>>>                                continue;
>>> -                     for (j = 0; j <= writer_n_durations[i]; j++) {
>>> +                     for (j = 0; j < writer_n_durations[i]; j++) {
>>>                                wdp = &wdpp[j];
>>>                                pr_alert("%s%s %4d writer-duration: 
>>> %5d %llu\n",
>>>                                        scale_type, SCALE_FLAG,
>>> -- 
>>> 2.17.1
>>>