netdev - [RFC bpf-next v3 08/14] net: veth: Implement devtx tx checksum

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

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Fri,  7 Jul 2023 12:30:00 -0700
From: Stanislav Fomichev <sdf@...gle.com>
To: bpf@...r.kernel.org
Cc: ast@...nel.org, daniel@...earbox.net, andrii@...nel.org, 
	martin.lau@...ux.dev, song@...nel.org, yhs@...com, john.fastabend@...il.com, 
	kpsingh@...nel.org, sdf@...gle.com, haoluo@...gle.com, jolsa@...nel.org, 
	kuba@...nel.org, toke@...nel.org, willemb@...gle.com, dsahern@...nel.org, 
	magnus.karlsson@...el.com, bjorn@...nel.org, maciej.fijalkowski@...el.com, 
	hawk@...nel.org, netdev@...r.kernel.org, xdp-hints@...-project.net
Subject: [RFC bpf-next v3 08/14] net: veth: Implement devtx tx checksum

Implement tx checksum kfunc for veth by checksumming the packet
in software (since there is nothing to offload). The change mostly
exists to make it possible to have software-based selftest.

Probably should instead set csum_start/csum_offset on the skb
itself?

Signed-off-by: Stanislav Fomichev <sdf@...gle.com>
---
 drivers/net/veth.c | 29 +++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index 5af4b15e107c..6f97511a545b 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -1814,6 +1814,34 @@ static int veth_devtx_tx_timestamp(const struct devtx_ctx *_ctx, u64 *timestamp)
 	return 0;
 }
 
+static int veth_devtx_request_l4_csum(const struct devtx_ctx *_ctx,
+				      u16 csum_start, u16 csum_offset)
+{
+	struct veth_devtx_ctx *ctx = (struct veth_devtx_ctx *)_ctx;
+	struct sk_buff *skb = ctx->skb;
+	__wsum csum;
+	int ret;
+
+	if (!skb)
+		return -EINVAL;
+
+	if (skb_transport_header_was_set(skb))
+		return -EINVAL;
+
+	if (csum_start >= skb->len)
+		return -EINVAL;
+
+	ret = skb_ensure_writable(skb, csum_offset + sizeof(__sum16));
+	if (ret)
+		return ret;
+
+	csum = csum_partial(skb->data + csum_start, skb->len - csum_start, 0);
+	*(__sum16 *)(skb->data + csum_offset) = csum_fold(csum) ?: CSUM_MANGLED_0;
+	skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+	return 0;
+}
+
 static const struct net_device_ops veth_netdev_ops = {
 	.ndo_init            = veth_dev_init,
 	.ndo_open            = veth_open,
@@ -1840,6 +1868,7 @@ static const struct xdp_metadata_ops veth_xdp_metadata_ops = {
 	.xmo_rx_hash			= veth_xdp_rx_hash,
 	.xmo_request_tx_timestamp	= veth_devtx_request_tx_timestamp,
 	.xmo_tx_timestamp		= veth_devtx_tx_timestamp,
+	.xmo_request_l4_checksum	= veth_devtx_request_l4_csum,
 };
 
 #define VETH_FEATURES (NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_HW_CSUM | \
-- 
2.41.0.255.g8b1d071c50-goog