[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CA+CK2bBvKpWFtPz+OmjX5SJuAfL4hAaLbHmfRok4Ueo2toeTiA@mail.gmail.com>
Date: Wed, 27 Jan 2021 11:12:34 -0500
From: Pavel Tatashin <pasha.tatashin@...een.com>
To: Jens Axboe <axboe@...nel.dk>
Cc: Tyler Hicks <tyhicks@...ux.microsoft.com>,
linux-block@...r.kernel.org, LKML <linux-kernel@...r.kernel.org>,
Sasha Levin <sashal@...nel.org>,
James Morris <jmorris@...ei.org>, lukas.bulwahn@...il.com,
hch@....de, Petr Vorel <pvorel@...e.cz>, ming.lei@...hat.com,
mzxreary@...inter.de, mcgrof@...nel.org, zhengbin13@...wei.com,
maco@...roid.com, Colin King <colin.king@...onical.com>,
evgreen@...omium.org
Subject: Re: [PATCH v4 1/1] loop: scale loop device by introducing per device lock
On Tue, Jan 26, 2021 at 3:09 PM Jens Axboe <axboe@...nel.dk> wrote:
>
> On 1/26/21 7:46 AM, Pavel Tatashin wrote:
> > Currently, loop device has only one global lock: loop_ctl_mutex.
> >
> > This becomes hot in scenarios where many loop devices are used.
> >
> > Scale it by introducing per-device lock: lo_mutex that protects
> > modifications of all fields in struct loop_device.
> >
> > Keep loop_ctl_mutex to protect global data: loop_index_idr, loop_lookup,
> > loop_add.
> >
> > The new lock ordering requirement is that loop_ctl_mutex must be taken
> > before lo_mutex.
>
> Applied, thanks.
Great, thank you!
Pasha
>
> --
> Jens Axboe
>
Powered by blists - more mailing lists