| 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
| ||
|
Message-ID: <1726127409.3427224-4-xuanzhuo@linux.alibaba.com>
Date: Thu, 12 Sep 2024 15:50:09 +0800
From: Xuan Zhuo <xuanzhuo@...ux.alibaba.com>
To: Jason Wang <jasowang@...hat.com>
Cc: netdev@...r.kernel.org,
"Michael S. Tsirkin" <mst@...hat.com>,
Eugenio Pérez <eperezma@...hat.com>,
"David S. Miller" <davem@...emloft.net>,
Eric Dumazet <edumazet@...gle.com>,
Jakub Kicinski <kuba@...nel.org>,
Paolo Abeni <pabeni@...hat.com>,
Alexei Starovoitov <ast@...nel.org>,
Daniel Borkmann <daniel@...earbox.net>,
Jesper Dangaard Brouer <hawk@...nel.org>,
John Fastabend <john.fastabend@...il.com>,
virtualization@...ts.linux.dev,
bpf@...r.kernel.org
Subject: Re: [PATCH net-next 07/13] virtio_net: refactor the xmit type
On Wed, 11 Sep 2024 12:04:16 +0800, Jason Wang <jasowang@...hat.com> wrote:
> On Tue, Aug 20, 2024 at 3:33 PM Xuan Zhuo <xuanzhuo@...ux.alibaba.com> wrote:
> >
> > Because the af-xdp will introduce a new xmit type, so I refactor the
> > xmit type mechanism first.
> >
> > We use the last two bits of the pointer to distinguish the xmit type,
> > so we can distinguish four xmit types. Now we have three types: skb,
> > orphan and xdp.
>
> And if I was not wrong, we do not anymore use bitmasks. If yes, let's
> explain the reason here.
In general, pointers are aligned to 4 or 8 bytes. If it is aligned to 4 bytes,
then only two bits are free for a pointer. So we can only use two bits.
But there are 4 types here, so we can't use bits to distinguish them.
b00 for skb
b01 for SKB_ORPHAN
b10 for XDP
b11 for af-xdp tx
>
> >
> > Signed-off-by: Xuan Zhuo <xuanzhuo@...ux.alibaba.com>
> > ---
> > drivers/net/virtio_net.c | 90 +++++++++++++++++++++++-----------------
> > 1 file changed, 51 insertions(+), 39 deletions(-)
> >
> > diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> > index 41aaea3b90fd..96abee36738b 100644
> > --- a/drivers/net/virtio_net.c
> > +++ b/drivers/net/virtio_net.c
> > @@ -45,9 +45,6 @@ module_param(napi_tx, bool, 0644);
> > #define VIRTIO_XDP_TX BIT(0)
> > #define VIRTIO_XDP_REDIR BIT(1)
> >
> > -#define VIRTIO_XDP_FLAG BIT(0)
> > -#define VIRTIO_ORPHAN_FLAG BIT(1)
> > -
> > /* RX packet size EWMA. The average packet size is used to determine the packet
> > * buffer size when refilling RX rings. As the entire RX ring may be refilled
> > * at once, the weight is chosen so that the EWMA will be insensitive to short-
> > @@ -509,34 +506,35 @@ static struct sk_buff *virtnet_skb_append_frag(struct sk_buff *head_skb,
> > struct page *page, void *buf,
> > int len, int truesize);
> >
> > -static bool is_xdp_frame(void *ptr)
> > -{
> > - return (unsigned long)ptr & VIRTIO_XDP_FLAG;
> > -}
> > +enum virtnet_xmit_type {
> > + VIRTNET_XMIT_TYPE_SKB,
> > + VIRTNET_XMIT_TYPE_ORPHAN,
>
> Let's rename this to SKB_ORPHAN?
>
> > + VIRTNET_XMIT_TYPE_XDP,
> > +};
> >
> > -static void *xdp_to_ptr(struct xdp_frame *ptr)
> > -{
> > - return (void *)((unsigned long)ptr | VIRTIO_XDP_FLAG);
> > -}
> > +#define VIRTNET_XMIT_TYPE_MASK (VIRTNET_XMIT_TYPE_SKB | VIRTNET_XMIT_TYPE_ORPHAN \
> > + | VIRTNET_XMIT_TYPE_XDP)
>
Maybe I should define VIRTNET_XMIT_TYPE_MASK to 0x3 directly with some comments.
Thanks.
> I may miss something but it seems not a correct bitmask definition as
> each member is not a bit actually?
>
> >
> > -static struct xdp_frame *ptr_to_xdp(void *ptr)
> > +static enum virtnet_xmit_type virtnet_xmit_ptr_strip(void **ptr)
> > {
> > - return (struct xdp_frame *)((unsigned long)ptr & ~VIRTIO_XDP_FLAG);
> > -}
> > + unsigned long p = (unsigned long)*ptr;
> >
> > -static bool is_orphan_skb(void *ptr)
> > -{
> > - return (unsigned long)ptr & VIRTIO_ORPHAN_FLAG;
> > + *ptr = (void *)(p & ~VIRTNET_XMIT_TYPE_MASK);
> > +
> > + return p & VIRTNET_XMIT_TYPE_MASK;
> > }
> >
> > -static void *skb_to_ptr(struct sk_buff *skb, bool orphan)
> > +static void *virtnet_xmit_ptr_mix(void *ptr, enum virtnet_xmit_type type)
> > {
> > - return (void *)((unsigned long)skb | (orphan ? VIRTIO_ORPHAN_FLAG : 0));
> > + return (void *)((unsigned long)ptr | type);
> > }
> >
> > -static struct sk_buff *ptr_to_skb(void *ptr)
> > +static int virtnet_add_outbuf(struct send_queue *sq, int num, void *data,
> > + enum virtnet_xmit_type type)
> > {
> > - return (struct sk_buff *)((unsigned long)ptr & ~VIRTIO_ORPHAN_FLAG);
> > + return virtqueue_add_outbuf(sq->vq, sq->sg, num,
> > + virtnet_xmit_ptr_mix(data, type),
> > + GFP_ATOMIC);
> > }
> >
> > static void sg_fill_dma(struct scatterlist *sg, dma_addr_t addr, u32 len)
> > @@ -549,29 +547,37 @@ static void sg_fill_dma(struct scatterlist *sg, dma_addr_t addr, u32 len)
> > static void __free_old_xmit(struct send_queue *sq, struct netdev_queue *txq,
> > bool in_napi, struct virtnet_sq_free_stats *stats)
> > {
> > + struct xdp_frame *frame;
> > + struct sk_buff *skb;
> > unsigned int len;
> > void *ptr;
> >
> > while ((ptr = virtqueue_get_buf(sq->vq, &len)) != NULL) {
> > - if (!is_xdp_frame(ptr)) {
> > - struct sk_buff *skb = ptr_to_skb(ptr);
> > + switch (virtnet_xmit_ptr_strip(&ptr)) {
> > + case VIRTNET_XMIT_TYPE_SKB:
> > + skb = ptr;
> >
> > pr_debug("Sent skb %p\n", skb);
> > + stats->napi_packets++;
> > + stats->napi_bytes += skb->len;
> > + napi_consume_skb(skb, in_napi);
> > + break;
> >
> > - if (is_orphan_skb(ptr)) {
> > - stats->packets++;
> > - stats->bytes += skb->len;
> > - } else {
> > - stats->napi_packets++;
> > - stats->napi_bytes += skb->len;
> > - }
> > + case VIRTNET_XMIT_TYPE_ORPHAN:
> > + skb = ptr;
> > +
> > + stats->packets++;
> > + stats->bytes += skb->len;
> > napi_consume_skb(skb, in_napi);
> > - } else {
> > - struct xdp_frame *frame = ptr_to_xdp(ptr);
> > + break;
> > +
> > + case VIRTNET_XMIT_TYPE_XDP:
> > + frame = ptr;
> >
> > stats->packets++;
> > stats->bytes += xdp_get_frame_len(frame);
> > xdp_return_frame(frame);
> > + break;
> > }
> > }
> > netdev_tx_completed_queue(txq, stats->napi_packets, stats->napi_bytes);
> > @@ -1421,8 +1427,7 @@ static int __virtnet_xdp_xmit_one(struct virtnet_info *vi,
> > skb_frag_size(frag), skb_frag_off(frag));
> > }
> >
> > - err = virtqueue_add_outbuf(sq->vq, sq->sg, nr_frags + 1,
> > - xdp_to_ptr(xdpf), GFP_ATOMIC);
> > + err = virtnet_add_outbuf(sq, nr_frags + 1, xdpf, VIRTNET_XMIT_TYPE_XDP);
> > if (unlikely(err))
> > return -ENOSPC; /* Caller handle free/refcnt */
> >
> > @@ -3028,8 +3033,9 @@ static int xmit_skb(struct send_queue *sq, struct sk_buff *skb, bool orphan)
> > return num_sg;
> > num_sg++;
> > }
> > - return virtqueue_add_outbuf(sq->vq, sq->sg, num_sg,
> > - skb_to_ptr(skb, orphan), GFP_ATOMIC);
> > +
> > + return virtnet_add_outbuf(sq, num_sg, skb,
> > + orphan ? VIRTNET_XMIT_TYPE_ORPHAN : VIRTNET_XMIT_TYPE_SKB);
> > }
> >
> > static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
> > @@ -5906,10 +5912,16 @@ static void free_receive_page_frags(struct virtnet_info *vi)
> >
> > static void virtnet_sq_free_unused_buf(struct virtqueue *vq, void *buf)
> > {
> > - if (!is_xdp_frame(buf))
> > + switch (virtnet_xmit_ptr_strip(&buf)) {
> > + case VIRTNET_XMIT_TYPE_SKB:
> > + case VIRTNET_XMIT_TYPE_ORPHAN:
> > dev_kfree_skb(buf);
> > - else
> > - xdp_return_frame(ptr_to_xdp(buf));
> > + break;
> > +
> > + case VIRTNET_XMIT_TYPE_XDP:
> > + xdp_return_frame(buf);
> > + break;
> > + }
> > }
>
> Others look fine.
>
> Thanks
>
> >
> > static void free_unused_bufs(struct virtnet_info *vi)
> > --
> > 2.32.0.3.g01195cf9f
> >
>
Powered by blists - more mailing lists