linux-kernel - Re: [RFC v2 5/5] virtiofs: Disable notifications more aggresively

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]

Message-ID: <CAJaqyWcxWvYrFoocHD6e4ccZxQp9J=rQ_Ou+s3zcwmXQt4xMYA@mail.gmail.com>
Date: Mon, 24 Feb 2025 07:44:16 +0100
From: Eugenio Perez Martin <eperezma@...hat.com>
To: Jason Wang <jasowang@...hat.com>
Cc: Xuan Zhuo <xuanzhuo@...ux.alibaba.com>, Stefan Hajnoczi <stefanha@...hat.com>, 
	Hanna Reitz <hreitz@...hat.com>, linux-kernel@...r.kernel.org, 
	German Maglione <gmaglione@...hat.com>, virtualization@...ts.linux.dev, 
	Stefano Garzarella <sgarzare@...hat.com>, yama@...hat.com, Vivek Goyal <vgoyal@...hat.com>, 
	Miklos Szeredi <miklos@...redi.hu>, mst@...hat.com
Subject: Re: [RFC v2 5/5] virtiofs: Disable notifications more aggresively

On Mon, Feb 24, 2025 at 3:01 AM Jason Wang <jasowang@...hat.com> wrote:
>
> On Sat, Feb 22, 2025 at 1:07 AM Eugenio Pérez <eperezma@...hat.com> wrote:
> >
> > Disable notifications right before scheduling, instead of waiting until
> > the work is running.
> >
> > After applying this patch, fio read test goes from 1191MiB/s to
> > 1263.14Mib/s
>
> Let's describe more about the testing. E.g FIO parameters, test setups.
>
> >
> > Signed-off-by: Eugenio Pérez <eperezma@...hat.com>
> > ---
> >  fs/fuse/virtio_fs.c | 2 ++
> >  1 file changed, 2 insertions(+)
> >
> > diff --git a/fs/fuse/virtio_fs.c b/fs/fuse/virtio_fs.c
> > index e19c78f2480e..3d6981c0f3c3 100644
> > --- a/fs/fuse/virtio_fs.c
> > +++ b/fs/fuse/virtio_fs.c
> > @@ -915,6 +915,8 @@ static void virtio_fs_vq_done(struct virtqueue *vq)
> >
> >         dev_dbg(&vq->vdev->dev, "%s %s\n", __func__, fsvq->name);
> >
> > +       /* Shceduled, don't send more notifications */
>
> Typo, and I think we can just drop this commnet.

Agree.

>
> > +       virtqueue_disable_cb(vq);
>
> Do we need to tweak the virtio_fs_requests_done_work() as well?

I don't think so, as virtqueue_disable_cb can be called multiple times
without consequences.

To control it in virtio_fs_requests_done_work imply to create a new
boolean somewhere and add a conditional there, redundant with the
conditional in virtio_fs_requests_done_work.