Skip to contentSkip to conversation

VSOCK: support fill mergeable rx buffer in guest

PATCH In review jiangyiwen 2 participants 5 messages Dec 12, 2018 – Dec 13, 2018

Subsystems (6)
Networking Network drivers KVM virtio core +2 more subsystems
lore
Share & tools
lore

jiangyiwen

author
#0 lore raw

PATCH v2 1/5 VSOCK: support fill mergeable rx buffer in guest

In driver probing, if virtio has VIRTIO_VSOCK_F_MRG_RXBUF feature, it will fill mergeable rx buffer, support for host send mergeable rx buffer. It will fill a fixed size(PAGE_SIZE) everytime to compact with small packet and big packet.

In addition, it also add one optimizations copied from virtio-net.c.
- Skb_page_frag_refill() which can use high order page and reduce the stress of page allocator.

Signed-off-by Yiwen Jiang <[email protected]>
---
 include/linux/virtio_vsock.h     |   3 ++
 net/vmw_vsock/virtio_transport.c | 112 +++++++++++++++++++++++++++++++--------
 2 files changed, 92 insertions(+), 23 deletions(-)
include/linux/virtio_vsock.h +3 -0
index e223e26..bf84418 100644
@@ -14,6 +14,9 @@
 #define VIRTIO_VSOCK_MAX_BUF_SIZE		0xFFFFFFFFUL
 #define VIRTIO_VSOCK_MAX_PKT_BUF_SIZE		(1024 * 64)
 
+/* Virtio-vsock feature */
+#define VIRTIO_VSOCK_F_MRG_RXBUF 0 /* Host can merge receive buffers. */
+
 enum {
 	VSOCK_VQ_RX     = 0, /* for host to guest data */
 	VSOCK_VQ_TX     = 1, /* for guest to host data */
net/vmw_vsock/virtio_transport.c +89 -23

154 lines not on this page · raw message

Signature · 1 line
1.8.3.1
maintainer
#1 lore raw

From: jiangyiwen <[email protected]>
Date: Wed, 12 Dec 2018 17:28:16 +0800

net/vmw_vsock/virtio_transport.c static int virtio_transport_send_pkt_loopback(struct virtio_vsock *vsock, · line 275

Quoted 11 lines
net/vmw_vsock/virtio_transport.c +11 -0
+static int fill_mergeable_rx_buff(struct virtio_vsock *vsock,
+		struct virtqueue *vq)
+{
+	struct page_frag *alloc_frag = &vsock->alloc_frag;
+	struct scatterlist sg;
+	/* Currently we don't use ewma len, use PAGE_SIZE instead, because too
+	 * small size can't fill one full packet, sadly we only 128 vq num now.
+	 */
+	unsigned int len = PAGE_SIZE, hole;
+	void *buf;
+	int err;

Please don't break up a set of local variable declarations with a comment like this. The comment seems to be about the initialization of 'len', so move that initialization into the code below the variable declarations and bring the comment along for the ride as well.

author
#2 lore raw
Quoted 20 lines · David Miller

From: jiangyiwen <[email protected]>
Date: Wed, 12 Dec 2018 17:28:16 +0800

net/vmw_vsock/virtio_transport.c static int virtio_transport_send_pkt_loopback(struct virtio_vsock *vsock, · line 275

Quoted 11 lines
net/vmw_vsock/virtio_transport.c +11 -0
+static int fill_mergeable_rx_buff(struct virtio_vsock *vsock,
+		struct virtqueue *vq)
+{
+	struct page_frag *alloc_frag = &vsock->alloc_frag;
+	struct scatterlist sg;
+	/* Currently we don't use ewma len, use PAGE_SIZE instead, because too
+	 * small size can't fill one full packet, sadly we only 128 vq num now.
+	 */
+	unsigned int len = PAGE_SIZE, hole;
+	void *buf;
+	int err;

Please don't break up a set of local variable declarations with a comment like this. The comment seems to be about the initialization of 'len', so move that initialization into the code below the variable declarations and bring the comment along for the ride as well.

Quoted 2 lines

.

Hi David,

Thanks your suggestions, if maintainers approve use this series of patches other than "vsock over virtio-net" idea, I will send to next version and fix it. Otherwise, I hope it can give maintainers the motivation that aggregate the vsock(virtio_transport related) and virtio-net.

Thanks,
Yiwen.

author
#3 lore raw
In reply to David Miller
Quoted 20 lines · David Miller

From: jiangyiwen <[email protected]>
Date: Wed, 12 Dec 2018 17:28:16 +0800

net/vmw_vsock/virtio_transport.c static int virtio_transport_send_pkt_loopback(struct virtio_vsock *vsock, · line 275

Quoted 11 lines
net/vmw_vsock/virtio_transport.c +11 -0
+static int fill_mergeable_rx_buff(struct virtio_vsock *vsock,
+		struct virtqueue *vq)
+{
+	struct page_frag *alloc_frag = &vsock->alloc_frag;
+	struct scatterlist sg;
+	/* Currently we don't use ewma len, use PAGE_SIZE instead, because too
+	 * small size can't fill one full packet, sadly we only 128 vq num now.
+	 */
+	unsigned int len = PAGE_SIZE, hole;
+	void *buf;
+	int err;

Please don't break up a set of local variable declarations with a comment like this. The comment seems to be about the initialization of 'len', so move that initialization into the code below the variable declarations and bring the comment along for the ride as well.

Quoted 2 lines

.

Hi David,

Thanks your suggestions, if maintainers approve use this series of patches other than "vsock over virtio-net" idea, I will send to next version and fix it. Otherwise, I hope it can give maintainers the motivation that aggregate the vsock(virtio_transport related) and virtio-net.

Thanks,
Yiwen.

maintainer
#4 lore raw
In reply to jiangyiwen

From: jiangyiwen <[email protected]>
Date: Wed, 12 Dec 2018 17:28:16 +0800

net/vmw_vsock/virtio_transport.c static int virtio_transport_send_pkt_loopback(struct virtio_vsock *vsock, · line 275

Quoted 11 lines
net/vmw_vsock/virtio_transport.c +11 -0
+static int fill_mergeable_rx_buff(struct virtio_vsock *vsock,
+		struct virtqueue *vq)
+{
+	struct page_frag *alloc_frag = &vsock->alloc_frag;
+	struct scatterlist sg;
+	/* Currently we don't use ewma len, use PAGE_SIZE instead, because too
+	 * small size can't fill one full packet, sadly we only 128 vq num now.
+	 */
+	unsigned int len = PAGE_SIZE, hole;
+	void *buf;
+	int err;

Please don't break up a set of local variable declarations with a comment like this. The comment seems to be about the initialization of 'len', so move that initialization into the code below the variable declarations and bring the comment along for the ride as well.