[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20250626021445.49068-2-jasowang@redhat.com>
Date: Thu, 26 Jun 2025 10:14:45 +0800
From: Jason Wang <jasowang@...hat.com>
To: willemdebruijn.kernel@...il.com,
jasowang@...hat.com,
andrew+netdev@...n.ch,
davem@...emloft.net,
edumazet@...gle.com,
kuba@...nel.org,
pabeni@...hat.com
Cc: mst@...hat.com,
eperezma@...hat.com,
netdev@...r.kernel.org,
linux-kernel@...r.kernel.org,
kvm@...r.kernel.org,
virtualization@...ts.linux.dev,
Willem de Bruijn <willemb@...gle.com>
Subject: [PATCH V2 net-next 2/2] vhost-net: reduce one userspace copy when building XDP buff
We used to do twice copy_from_iter() to copy virtio-net and packet
separately. This introduce overheads for userspace access hardening as
well as SMAP (for x86 it's stac/clac). So this patch tries to use one
copy_from_iter() to copy them once and move the virtio-net header
afterwards to reduce overheads.
Testpmd + vhost_net shows 10% improvement from 5.45Mpps to 6.0Mpps.
Acked-by: Willem de Bruijn <willemb@...gle.com>
Signed-off-by: Jason Wang <jasowang@...hat.com>
---
Changes since V1:
- Add a comment to explain no overlapping when using memcpy
---
drivers/vhost/net.c | 14 +++++---------
1 file changed, 5 insertions(+), 9 deletions(-)
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 777eb6193985..a33a32a1e488 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -690,13 +690,13 @@ static int vhost_net_build_xdp(struct vhost_net_virtqueue *nvq,
if (unlikely(!buf))
return -ENOMEM;
- copied = copy_from_iter(buf, sock_hlen, from);
- if (copied != sock_hlen) {
+ copied = copy_from_iter(buf + pad - sock_hlen, len, from);
+ if (copied != len) {
ret = -EFAULT;
goto err;
}
- gso = buf;
+ gso = buf + pad - sock_hlen;
if (!sock_hlen)
memset(buf, 0, pad);
@@ -715,12 +715,8 @@ static int vhost_net_build_xdp(struct vhost_net_virtqueue *nvq,
}
}
- len -= sock_hlen;
- copied = copy_from_iter(buf + pad, len, from);
- if (copied != len) {
- ret = -EFAULT;
- goto err;
- }
+ /* pad contains sock_hlen */
+ memcpy(buf, buf + pad - sock_hlen, sock_hlen);
xdp_init_buff(xdp, buflen, NULL);
xdp_prepare_buff(xdp, buf, pad, len, true);
--
2.34.1
Powered by blists - more mailing lists