[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAGudoHEmKLkPHMegbuqDOdS-HvojWYJ-KKVjUs=N4e3DLMg-zA@mail.gmail.com>
Date: Wed, 23 Aug 2023 19:12:15 +0200
From: Mateusz Guzik <mjguzik@...il.com>
To: Jan Kara <jack@...e.cz>
Cc: Dennis Zhou <dennis@...nel.org>, linux-kernel@...r.kernel.org,
tj@...nel.org, cl@...ux.com, akpm@...ux-foundation.org,
shakeelb@...gle.com, linux-mm@...ck.org
Subject: Re: [PATCH 0/2] execve scalability issues, part 1
On 8/23/23, Jan Kara <jack@...e.cz> wrote:
> On Wed 23-08-23 18:10:29, Mateusz Guzik wrote:
>> So the question is how much does this matter. My personal approach is
>> that avoidable slowdowns (like atomics here) only facilitate further
>> avoidable slowdowns as people can claim there is a minuscule change in
>> % to baseline. But if the baseline is already slow....
>
> I get your point but over the years I've also learned that premature
> optimization isn't good either as we will be dragging the maintenance
> burden for a long time ;) It's a balance.
>
Mate, your proposal is not maintenance burden-free either. ;)
I claim mine is simpler and faster for single threaded case, but is
not generic so should another consumer show up with the single vs
multithreaded need, there may or may not be more work to accommodate
it.
--
Mateusz Guzik <mjguzik gmail.com>
Powered by blists - more mailing lists