Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20200724152424.GC17209@redhat.com> <CAHk-=whuG+5pUeUqdiW4gk0prvqu7GZSMo-6oWv5PdDC5dBr=A@mail.gmail.com>
 <CAHk-=wjYHvbOs9i39EnUsC6VEJiuJ2e_5gZB5-J5CRKxq80B_Q@mail.gmail.com>
 <20200725101445.GB3870@redhat.com> <CAHk-=whSJbODMVmxxDs64f7BaESKWuMqOxWGpjUSDn6Jzqa71g@mail.gmail.com>
 <alpine.LSU.2.11.2007251343370.3804@eggly.anvils> <alpine.LSU.2.11.2007252100230.5376@eggly.anvils>
 <alpine.LSU.2.11.2007261246530.6812@eggly.anvils> <alpine.LSU.2.11.2008052105040.8716@eggly.anvils>
 <CAHk-=whf7wCUV2oTDUg0eeNafhhk_OhJBT2SbHZXwgtmAzNeTg@mail.gmail.com> <20200806180024.GB17456@casper.infradead.org>
In-Reply-To: <20200806180024.GB17456@casper.infradead.org>
From:   Linus Torvalds <torvalds@linux-foundation.org>
Date:   Thu, 6 Aug 2020 11:32:20 -0700
Message-ID: <CAHk-=wihTRHMm1LC4AfidZptT9ZuT-wBjE2VhYzKBy66e4iwQw@mail.gmail.com>
Subject: Re: [RFC PATCH] mm: silence soft lockups from unlock_page
To:     Matthew Wilcox <willy@infradead.org>
Cc:     Hugh Dickins <hughd@google.com>, Oleg Nesterov <oleg@redhat.com>,
        Michal Hocko <mhocko@kernel.org>,
        Linux-MM <linux-mm@kvack.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Tim Chen <tim.c.chen@linux.intel.com>,
        Michal Hocko <mhocko@suse.com>,
        Greg KH <gregkh@linuxfoundation.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, Aug 6, 2020 at 11:00 AM Matthew Wilcox <willy@infradead.org> wrote:
>
> It wasn't clear to me whether Hugh thought it was an improvement or
> not that trylock was now less likely to jump the queue.  There're
> the usual "fair is the opposite of throughput" kind of arguments.

Yeah, it could go either way. But on the whole, if the lock bit is
getting any contention, I think we'd rather have it be fair for
latency reasons.

That said, I'm not convinced about my patch, and I actually threw it
away without even testing it (sometimes I keep patches around in my
private tree for testing, and they can live there for months or even
years when I wonder if they are worth it, but this time I didn't
bother to go to the trouble).

If somebody is interested in pursuing this, I think that patch might
be a good starting point (and it _might_ even work), but it seemed to
be too subtle to really worry about unless somebody finds an actual
acute reason for it.

I think the existing patch narrowing the window is good, and it
clearly didn't hurt throughput (although that was almost certainly for
other reasons).

                Linus