From: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
To: netdev@vger.kernel.org
Cc: "Michael S. Tsirkin" <mst@redhat.com>,
Jason Wang <jasowang@redhat.com>,
"David S. Miller" <davem@davemloft.net>,
Eric Dumazet <edumazet@google.com>,
Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
Alexei Starovoitov <ast@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
Jesper Dangaard Brouer <hawk@kernel.org>,
John Fastabend <john.fastabend@gmail.com>,
virtualization@lists.linux-foundation.org, bpf@vger.kernel.org
Subject: [PATCH net-next v2 02/14] virtio_net: introduce mergeable_xdp_prepare()
Date: Tue, 18 Apr 2023 14:53:15 +0800 [thread overview]
Message-ID: <20230418065327.72281-3-xuanzhuo@linux.alibaba.com> (raw)
In-Reply-To: <20230418065327.72281-1-xuanzhuo@linux.alibaba.com>
Separating the logic of preparation for xdp from receive_mergeable.
The purpose of this is to simplify the logic of execution of XDP.
The main logic here is that when headroom is insufficient, we need to
allocate a new page and calculate offset. It should be noted that if
there is new page, the variable page will refer to the new page.
Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
---
drivers/net/virtio_net.c | 135 +++++++++++++++++++++++----------------
1 file changed, 79 insertions(+), 56 deletions(-)
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 42435e762d72..12559062ffb6 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1162,6 +1162,81 @@ static int virtnet_build_xdp_buff_mrg(struct net_device *dev,
return 0;
}
+static void *mergeable_xdp_prepare(struct virtnet_info *vi,
+ struct receive_queue *rq,
+ struct bpf_prog *xdp_prog,
+ void *ctx,
+ unsigned int *frame_sz,
+ int *num_buf,
+ struct page **page,
+ int offset,
+ unsigned int *len,
+ struct virtio_net_hdr_mrg_rxbuf *hdr)
+{
+ unsigned int truesize = mergeable_ctx_to_truesize(ctx);
+ unsigned int headroom = mergeable_ctx_to_headroom(ctx);
+ struct page *xdp_page;
+ unsigned int xdp_room;
+
+ /* Transient failure which in theory could occur if
+ * in-flight packets from before XDP was enabled reach
+ * the receive path after XDP is loaded.
+ */
+ if (unlikely(hdr->hdr.gso_type))
+ return NULL;
+
+ /* Now XDP core assumes frag size is PAGE_SIZE, but buffers
+ * with headroom may add hole in truesize, which
+ * make their length exceed PAGE_SIZE. So we disabled the
+ * hole mechanism for xdp. See add_recvbuf_mergeable().
+ */
+ *frame_sz = truesize;
+
+ /* This happens when headroom is not enough because
+ * of the buffer was prefilled before XDP is set.
+ * This should only happen for the first several packets.
+ * In fact, vq reset can be used here to help us clean up
+ * the prefilled buffers, but many existing devices do not
+ * support it, and we don't want to bother users who are
+ * using xdp normally.
+ */
+ if (!xdp_prog->aux->xdp_has_frags &&
+ (*num_buf > 1 || headroom < virtnet_get_headroom(vi))) {
+ /* linearize data for XDP */
+ xdp_page = xdp_linearize_page(rq, num_buf,
+ *page, offset,
+ VIRTIO_XDP_HEADROOM,
+ len);
+ *frame_sz = PAGE_SIZE;
+
+ if (!xdp_page)
+ return NULL;
+ offset = VIRTIO_XDP_HEADROOM;
+
+ put_page(*page);
+ *page = xdp_page;
+ } else if (unlikely(headroom < virtnet_get_headroom(vi))) {
+ xdp_room = SKB_DATA_ALIGN(VIRTIO_XDP_HEADROOM +
+ sizeof(struct skb_shared_info));
+ if (*len + xdp_room > PAGE_SIZE)
+ return NULL;
+
+ xdp_page = alloc_page(GFP_ATOMIC);
+ if (!xdp_page)
+ return NULL;
+
+ memcpy(page_address(xdp_page) + VIRTIO_XDP_HEADROOM,
+ page_address(*page) + offset, *len);
+ *frame_sz = PAGE_SIZE;
+ offset = VIRTIO_XDP_HEADROOM;
+
+ put_page(*page);
+ *page = xdp_page;
+ }
+
+ return page_address(*page) + offset;
+}
+
static struct sk_buff *receive_mergeable(struct net_device *dev,
struct virtnet_info *vi,
struct receive_queue *rq,
@@ -1181,7 +1256,7 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
unsigned int headroom = mergeable_ctx_to_headroom(ctx);
unsigned int tailroom = headroom ? sizeof(struct skb_shared_info) : 0;
unsigned int room = SKB_DATA_ALIGN(headroom + tailroom);
- unsigned int frame_sz, xdp_room;
+ unsigned int frame_sz;
int err;
head_skb = NULL;
@@ -1211,63 +1286,11 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
u32 act;
int i;
- /* Transient failure which in theory could occur if
- * in-flight packets from before XDP was enabled reach
- * the receive path after XDP is loaded.
- */
- if (unlikely(hdr->hdr.gso_type))
+ data = mergeable_xdp_prepare(vi, rq, xdp_prog, ctx, &frame_sz,
+ &num_buf, &page, offset, &len, hdr);
+ if (unlikely(!data))
goto err_xdp;
- /* Now XDP core assumes frag size is PAGE_SIZE, but buffers
- * with headroom may add hole in truesize, which
- * make their length exceed PAGE_SIZE. So we disabled the
- * hole mechanism for xdp. See add_recvbuf_mergeable().
- */
- frame_sz = truesize;
-
- /* This happens when headroom is not enough because
- * of the buffer was prefilled before XDP is set.
- * This should only happen for the first several packets.
- * In fact, vq reset can be used here to help us clean up
- * the prefilled buffers, but many existing devices do not
- * support it, and we don't want to bother users who are
- * using xdp normally.
- */
- if (!xdp_prog->aux->xdp_has_frags &&
- (num_buf > 1 || headroom < virtnet_get_headroom(vi))) {
- /* linearize data for XDP */
- xdp_page = xdp_linearize_page(rq, &num_buf,
- page, offset,
- VIRTIO_XDP_HEADROOM,
- &len);
- frame_sz = PAGE_SIZE;
-
- if (!xdp_page)
- goto err_xdp;
- offset = VIRTIO_XDP_HEADROOM;
-
- put_page(page);
- page = xdp_page;
- } else if (unlikely(headroom < virtnet_get_headroom(vi))) {
- xdp_room = SKB_DATA_ALIGN(VIRTIO_XDP_HEADROOM +
- sizeof(struct skb_shared_info));
- if (len + xdp_room > PAGE_SIZE)
- goto err_xdp;
-
- xdp_page = alloc_page(GFP_ATOMIC);
- if (!xdp_page)
- goto err_xdp;
-
- memcpy(page_address(xdp_page) + VIRTIO_XDP_HEADROOM,
- page_address(page) + offset, len);
- frame_sz = PAGE_SIZE;
- offset = VIRTIO_XDP_HEADROOM;
-
- put_page(page);
- page = xdp_page;
- }
-
- data = page_address(page) + offset;
err = virtnet_build_xdp_buff_mrg(dev, vi, rq, &xdp, data, len, frame_sz,
&num_buf, &xdp_frags_truesz, stats);
if (unlikely(err))
--
2.32.0.3.g01195cf9f
next prev parent reply other threads:[~2023-04-18 6:53 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-18 6:53 [PATCH net-next v2 00/14] virtio_net: refactor xdp codes Xuan Zhuo
2023-04-18 6:53 ` [PATCH net-next v2 01/14] virtio_net: mergeable xdp: put old page immediately Xuan Zhuo
2023-04-20 5:58 ` Jason Wang
2023-04-18 6:53 ` Xuan Zhuo [this message]
2023-04-20 5:58 ` [PATCH net-next v2 02/14] virtio_net: introduce mergeable_xdp_prepare() Jason Wang
2023-04-18 6:53 ` [PATCH net-next v2 03/14] virtio_net: optimize mergeable_xdp_prepare() Xuan Zhuo
2023-04-20 5:59 ` Jason Wang
2023-04-18 6:53 ` [PATCH net-next v2 04/14] virtio_net: introduce virtnet_xdp_handler() to seprate the logic of run xdp Xuan Zhuo
2023-04-20 5:59 ` Jason Wang
2023-04-18 6:53 ` [PATCH net-next v2 05/14] virtio_net: introduce xdp res enums Xuan Zhuo
2023-04-20 5:59 ` Jason Wang
2023-04-21 7:00 ` Michael S. Tsirkin
2023-04-21 7:24 ` Xuan Zhuo
2023-04-21 11:54 ` Michael S. Tsirkin
2023-04-23 1:57 ` Xuan Zhuo
2023-04-23 5:28 ` Jason Wang
2023-04-23 6:27 ` Xuan Zhuo
2023-04-18 6:53 ` [PATCH net-next v2 06/14] virtio_net: separate the logic of freeing xdp shinfo Xuan Zhuo
2023-04-18 6:53 ` [PATCH net-next v2 07/14] virtio_net: separate the logic of freeing the rest mergeable buf Xuan Zhuo
2023-04-18 6:53 ` [PATCH net-next v2 08/14] virtio_net: auto release xdp shinfo Xuan Zhuo
2023-04-20 5:59 ` Jason Wang
2023-04-20 9:10 ` Xuan Zhuo
2023-04-18 6:53 ` [PATCH net-next v2 09/14] virtio_net: introduce receive_mergeable_xdp() Xuan Zhuo
2023-04-20 6:01 ` Jason Wang
2023-04-18 6:53 ` [PATCH net-next v2 10/14] virtio_net: merge: remove skip_xdp Xuan Zhuo
2023-04-20 6:10 ` Jason Wang
2023-04-18 6:53 ` [PATCH net-next v2 11/14] virtio_net: introduce receive_small_xdp() Xuan Zhuo
2023-04-20 6:23 ` Jason Wang
2023-04-18 6:53 ` [PATCH net-next v2 12/14] virtio_net: small: optimize code Xuan Zhuo
2023-04-20 6:28 ` Jason Wang
2023-04-18 6:53 ` [PATCH net-next v2 13/14] " Xuan Zhuo
2023-04-20 6:32 ` Jason Wang
2023-04-20 8:56 ` Xuan Zhuo
2023-04-18 6:53 ` [PATCH net-next v2 14/14] virtio_net: small: remove skip_xdp Xuan Zhuo
2023-04-18 11:49 ` [PATCH net-next v2 00/14] virtio_net: refactor xdp codes Michael S. Tsirkin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230418065327.72281-3-xuanzhuo@linux.alibaba.com \
--to=xuanzhuo@linux.alibaba.com \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=hawk@kernel.org \
--cc=jasowang@redhat.com \
--cc=john.fastabend@gmail.com \
--cc=kuba@kernel.org \
--cc=mst@redhat.com \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=virtualization@lists.linux-foundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox