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;
IronPort-SDR: /CWxTVl4jK6AL1c7zLUXB6EOIyiW4v7Rh/LUnxTiV47KjtaF+zj4VwZmnthSJWAYhgLUgR6gcz
 cXSpTKp3r0Dw==
From:   SeongJae Park <sjpark@amazon.com>
To:     Greg KH <gregkh@linuxfoundation.org>
CC:     SeongJae Park <sjpark@amazon.com>, <davem@davemloft.net>,
        <viro@zeniv.linux.org.uk>, <kuba@kernel.org>,
        <edumazet@google.com>, <netdev@vger.kernel.org>,
        <linux-kernel@vger.kernel.org>, SeongJae Park <sjpark@amazon.de>
Subject: Re: Re: [PATCH net 1/2] Revert "coallocate socket_wq with socket itself"
Date:   Tue, 5 May 2020 10:04:34 +0200
Message-ID: <20200505080434.5651-1-sjpark@amazon.com>
In-Reply-To: <20200505074511.GA4054974@kroah.com> (raw)
MIME-Version: 1.0
Content-Type: text/plain
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, 5 May 2020 09:45:11 +0200 Greg KH <gregkh@linuxfoundation.org> wrote:

> On Tue, May 05, 2020 at 09:28:40AM +0200, SeongJae Park wrote:
> > From: SeongJae Park <sjpark@amazon.de>
> > 
> > This reverts commit 333f7909a8573145811c4ab7d8c9092301707721.
> > 
> > The commit 6d7855c54e1e ("sockfs: switch to ->free_inode()") made the
> > deallocation of 'socket_alloc' to be done asynchronously using RCU, as
> > same to 'sock.wq'.  And the following commit 333f7909a857 ("coallocate
> > socket_sq with socket itself") made those to have same life cycle.
> > 
> > The changes made the code much more simple, but also made 'socket_alloc'
> > live longer than before.  For the reason, user programs intensively
> > repeating allocations and deallocations of sockets could cause memory
> > pressure on recent kernels.
> > 
> > To avoid the problem, this commit separates the life cycle of
> > 'socket_alloc' and 'sock.wq' again.  The following commit will make the
> > deallocation of 'socket_alloc' to be done synchronously again.
> > ---
> 
> No signed-off-by?
> No "Fixes:"?

Oops, my mistake.  I will post next version right now.


Thanks,
SeongJae Park

> 
> :(