linux-kernel - Re: [PATCH] docs: lockdep-design: correct the notation for writer

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

Date:   Mon, 24 May 2021 09:42:20 -0400
From:   Waiman Long <llong@...hat.com>
To:     Boqun Feng <boqun.feng@...il.com>,
        Xiongwei Song <sxwjean@...il.com>
Cc:     Waiman Long <llong@...hat.com>, Xiongwei Song <sxwjean@...com>,
        peterz@...radead.org, mingo@...hat.com, will@...nel.org,
        corbet@....net,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        linux-doc@...r.kernel.org
Subject: Re: [PATCH] docs: lockdep-design: correct the notation for writer

On 5/24/21 6:32 AM, Boqun Feng wrote:
> On Mon, May 24, 2021 at 12:24:00PM +0800, Xiongwei Song wrote:
>> On Fri, May 21, 2021 at 11:17 PM Waiman Long <llong@...hat.com> wrote:
>>> On 5/21/21 2:29 AM, Xiongwei Song wrote:
>>>> From: Xiongwei Song <sxwjean@...il.com>
>>>>
>>>> The block condition matrix is using 'E' as the writer noation here, so it
>>>> would be better to use 'E' as the reminder rather than 'W'.
>>>>
>>>> Signed-off-by: Xiongwei Song <sxwjean@...il.com>
>>>> ---
>>>>    Documentation/locking/lockdep-design.rst | 2 +-
>>>>    1 file changed, 1 insertion(+), 1 deletion(-)
>>>>
>>>> diff --git a/Documentation/locking/lockdep-design.rst b/Documentation/locking/lockdep-design.rst
>>>> index 9f3cfca..c3b923a 100644
>>>> --- a/Documentation/locking/lockdep-design.rst
>>>> +++ b/Documentation/locking/lockdep-design.rst
>>>> @@ -462,7 +462,7 @@ Block condition matrix, Y means the row blocks the column, and N means otherwise
>>>>        | R | Y | Y | N |
>>>>        +---+---+---+---+
>>>>
>>>> -     (W: writers, r: non-recursive readers, R: recursive readers)
>>>> +     (E: writers, r: non-recursive readers, R: recursive readers)
>>>>
>>>>
>>>>    acquired recursively. Unlike non-recursive read locks, recursive read locks
>>> I would say it should be the other way around. Both W and E refer to the
>>> same type of lockers. W emphasizes writer aspect of it and E for
>>> exclusive. I think we should change the block condition matrix to use W
>>> instead of E.
>> The doc uses 'E'  to describe dependency egdes too. Should we change them
>> to 'W'? Personally,  both 'W' and 'E' are fine.
>>
> I also think Waiman's suggestion is solid, there are two ways to
> classify locks:
>
> 1.	W (Writers), R (Recursive Readers), r (Non-recursive Readers)
>
> 2.	E (Exclusive locks), S (Shared locks), R (Recursive Readers),
> 	N (Non-recursive locks)
>
> And the relations between them are as follow:
>
> 	E = W
> 	R = R
> 	N = W \/ r
> 	S = R \/ r
>
> , where "\/" is the set union.
>
> The story is that I used the way #1 at first, and later on realized way
> #2 is better for BFS implementation, also for reasoning, so here came
> this leftover..
>
My suggestion was based on the fact that it is harder to associate E 
with writer. So from a readability perspective, it is better to change 
the block condition matrix to use 'W' to make it more readable.

Cheers,
Longman