All of lore.kernel.org
 help / color / mirror / Atom feed
From: Will Deacon <will@kernel.org>
To: linux-kernel@vger.kernel.org
Cc: "Will Deacon" <will@kernel.org>, "Keir Fraser" <keirf@google.com>,
	"Steven Moreland" <smoreland@google.com>,
	"Frederick Mayle" <fmayle@google.com>,
	"Stefan Hajnoczi" <stefanha@redhat.com>,
	"Stefano Garzarella" <sgarzare@redhat.com>,
	"Michael S. Tsirkin" <mst@redhat.com>,
	"Jason Wang" <jasowang@redhat.com>,
	"Eugenio Pérez" <eperezma@redhat.com>,
	netdev@vger.kernel.org, virtualization@lists.linux.dev
Subject: [PATCH v4 4/9] vsock/virtio: Resize receive buffers so that each SKB fits in a 4K page
Date: Thu, 17 Jul 2025 10:01:11 +0100	[thread overview]
Message-ID: <20250717090116.11987-5-will@kernel.org> (raw)
In-Reply-To: <20250717090116.11987-1-will@kernel.org>

When allocating receive buffers for the vsock virtio RX virtqueue, an
SKB is allocated with a 4140 data payload (the 44-byte packet header +
VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE). Even when factoring in the SKB
overhead, the resulting 8KiB allocation thanks to the rounding in
kmalloc_reserve() is wasteful (~3700 unusable bytes) and results in a
higher-order page allocation on systems with 4KiB pages just for the
sake of a few hundred bytes of packet data.

Limit the vsock virtio RX buffers to 4KiB per SKB, resulting in much
better memory utilisation and removing the need to allocate higher-order
pages entirely.

Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Will Deacon <will@kernel.org>
---
 include/linux/virtio_vsock.h     | 7 ++++++-
 net/vmw_vsock/virtio_transport.c | 2 +-
 2 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/include/linux/virtio_vsock.h b/include/linux/virtio_vsock.h
index 97465f378ade..879f1dfa7d3a 100644
--- a/include/linux/virtio_vsock.h
+++ b/include/linux/virtio_vsock.h
@@ -106,7 +106,12 @@ static inline size_t virtio_vsock_skb_len(struct sk_buff *skb)
 	return (size_t)(skb_end_pointer(skb) - skb->head);
 }
 
-#define VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE	(1024 * 4)
+/* Dimension the RX SKB so that the entire thing fits exactly into
+ * a single 4KiB page. This avoids wasting memory due to alloc_skb()
+ * rounding up to the next page order and also means that we
+ * don't leave higher-order pages sitting around in the RX queue.
+ */
+#define VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE	SKB_WITH_OVERHEAD(1024 * 4)
 #define VIRTIO_VSOCK_MAX_BUF_SIZE		0xFFFFFFFFUL
 #define VIRTIO_VSOCK_MAX_PKT_BUF_SIZE		(1024 * 64)
 
diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c
index 0166919f8705..39f346890f7f 100644
--- a/net/vmw_vsock/virtio_transport.c
+++ b/net/vmw_vsock/virtio_transport.c
@@ -307,7 +307,7 @@ virtio_transport_cancel_pkt(struct vsock_sock *vsk)
 
 static void virtio_vsock_rx_fill(struct virtio_vsock *vsock)
 {
-	int total_len = VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE + VIRTIO_VSOCK_SKB_HEADROOM;
+	int total_len = VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE;
 	struct scatterlist pkt, *p;
 	struct virtqueue *vq;
 	struct sk_buff *skb;
-- 
2.50.0.727.gbf7dc18ff4-goog


  parent reply	other threads:[~2025-07-17  9:01 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-07-17  9:01 [PATCH v4 0/9] vsock/virtio: SKB allocation improvements Will Deacon
2025-07-17  9:01 ` [PATCH v4 1/9] vhost/vsock: Avoid allocating arbitrarily-sized SKBs Will Deacon
2025-07-17  9:10   ` Jason Wang
2025-07-17  9:25     ` Stefano Garzarella
2025-07-17  9:01 ` [PATCH v4 2/9] vsock/virtio: Validate length in packet header before skb_put() Will Deacon
2025-07-17  9:39   ` Stefano Garzarella
2025-07-17  9:01 ` [PATCH v4 3/9] vsock/virtio: Move length check to callers of virtio_vsock_skb_rx_put() Will Deacon
2025-07-17  9:01 ` Will Deacon [this message]
2026-01-08 16:33   ` [PATCH v4 4/9] vsock/virtio: Resize receive buffers so that each SKB fits in a 4K page David Woodhouse
2026-01-12 14:48     ` Stefano Garzarella
2026-01-12 15:22       ` Will Deacon
2025-07-17  9:01 ` [PATCH v4 5/9] vsock/virtio: Rename virtio_vsock_alloc_skb() Will Deacon
2025-07-17  9:01 ` [PATCH v4 6/9] vsock/virtio: Move SKB allocation lower-bound check to callers Will Deacon
2025-07-17  9:01 ` [PATCH v4 7/9] vhost/vsock: Allocate nonlinear SKBs for handling large receive buffers Will Deacon
2025-07-17  9:01 ` [PATCH v4 8/9] vsock/virtio: Rename virtio_vsock_skb_rx_put() Will Deacon
2025-07-17  9:01 ` [PATCH v4 9/9] vsock/virtio: Allocate nonlinear SKBs for handling large transmit buffers Will Deacon
2025-08-13  8:41   ` Michael S. Tsirkin
2025-08-13 13:25     ` Hillf Danton
2025-08-15 10:22       ` Michael S. Tsirkin
2025-08-15 12:07         ` Hillf Danton
2025-08-16 10:38           ` Michael S. Tsirkin
2025-07-17 12:31 ` [PATCH v4 0/9] vsock/virtio: SKB allocation improvements Michael S. Tsirkin
2025-07-17 12:36   ` Stefano Garzarella
2025-07-17 12:35 ` 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=20250717090116.11987-5-will@kernel.org \
    --to=will@kernel.org \
    --cc=eperezma@redhat.com \
    --cc=fmayle@google.com \
    --cc=jasowang@redhat.com \
    --cc=keirf@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mst@redhat.com \
    --cc=netdev@vger.kernel.org \
    --cc=sgarzare@redhat.com \
    --cc=smoreland@google.com \
    --cc=stefanha@redhat.com \
    --cc=virtualization@lists.linux.dev \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.