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: <CACGkMEuUa+6_uaa7H2CSvUnfNzBr-rdoQ+cp8eZD+Ay1CZ=A-g@mail.gmail.com>
Date: Wed, 4 Dec 2024 10:49:02 +0800
From: Jason Wang <jasowang@...hat.com>
To: Koichiro Den <koichiro.den@...onical.com>
Cc: virtualization@...ts.linux.dev, mst@...hat.com, xuanzhuo@...ux.alibaba.com, 
	eperezma@...hat.com, andrew+netdev@...n.ch, davem@...emloft.net, 
	edumazet@...gle.com, kuba@...nel.org, pabeni@...hat.com, jiri@...nulli.us, 
	netdev@...r.kernel.org, linux-kernel@...r.kernel.org, stable@...r.kernel.org
Subject: Re: [PATCH net-next v2 4/5] virtio_ring: add 'flushed' as an argument
 to virtqueue_reset()

On Tue, Dec 3, 2024 at 3:31 PM Koichiro Den <koichiro.den@...onical.com> wrote:
>
> When virtqueue_reset() has actually recycled all unused buffers,
> additional work may be required in some cases. Relying solely on its
> return status is fragile, so introduce a new argument 'flushed' to
> explicitly indicate whether it has really occurred.
>
> Signed-off-by: Koichiro Den <koichiro.den@...onical.com>
> ---
>  drivers/net/virtio_net.c     | 6 ++++--
>  drivers/virtio/virtio_ring.c | 6 +++++-
>  include/linux/virtio.h       | 3 ++-
>  3 files changed, 11 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index 0103d7990e44..d5240a03b7d6 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -5695,6 +5695,7 @@ static int virtnet_rq_bind_xsk_pool(struct virtnet_info *vi, struct receive_queu
>                                     struct xsk_buff_pool *pool)
>  {
>         int err, qindex;
> +       bool flushed;
>
>         qindex = rq - vi->rq;
>
> @@ -5713,7 +5714,7 @@ static int virtnet_rq_bind_xsk_pool(struct virtnet_info *vi, struct receive_queu
>
>         virtnet_rx_pause(vi, rq);
>
> -       err = virtqueue_reset(rq->vq, virtnet_rq_unmap_free_buf);
> +       err = virtqueue_reset(rq->vq, virtnet_rq_unmap_free_buf, &flushed);
>         if (err) {
>                 netdev_err(vi->dev, "reset rx fail: rx queue index: %d err: %d\n", qindex, err);
>
> @@ -5737,12 +5738,13 @@ static int virtnet_sq_bind_xsk_pool(struct virtnet_info *vi,
>                                     struct xsk_buff_pool *pool)
>  {
>         int err, qindex;
> +       bool flushed;
>
>         qindex = sq - vi->sq;
>
>         virtnet_tx_pause(vi, sq);
>
> -       err = virtqueue_reset(sq->vq, virtnet_sq_free_unused_buf);
> +       err = virtqueue_reset(sq->vq, virtnet_sq_free_unused_buf, &flushed);
>         if (err) {
>                 netdev_err(vi->dev, "reset tx fail: tx queue index: %d err: %d\n", qindex, err);
>                 pool = NULL;
> diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
> index 34a068d401ec..b522ef798946 100644
> --- a/drivers/virtio/virtio_ring.c
> +++ b/drivers/virtio/virtio_ring.c
> @@ -2828,6 +2828,7 @@ EXPORT_SYMBOL_GPL(virtqueue_resize);
>   * virtqueue_reset - detach and recycle all unused buffers
>   * @_vq: the struct virtqueue we're talking about.
>   * @recycle: callback to recycle unused buffers
> + * @flushed: whether or not unused buffers are all flushed
>   *
>   * Caller must ensure we don't call this with other virtqueue operations
>   * at the same time (except where noted).
> @@ -2839,14 +2840,17 @@ EXPORT_SYMBOL_GPL(virtqueue_resize);
>   * -EPERM: Operation not permitted
>   */
>  int virtqueue_reset(struct virtqueue *_vq,
> -                   void (*recycle)(struct virtqueue *vq, void *buf))
> +                   void (*recycle)(struct virtqueue *vq, void *buf),
> +                   bool *flushed)
>  {
>         struct vring_virtqueue *vq = to_vvq(_vq);
>         int err;
>
> +       *flushed = false;
>         err = virtqueue_disable_and_recycle(_vq, recycle);
>         if (err)
>                 return err;
> +       *flushed = true;
>

This makes me think if it would be easier if we just find a way to
reset the tx queue inside virtqueue_disable_and_recycle().

For example, introducing a recycle_done callback?

Thanks


Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ