openwrt/target/linux/generic/backport-6.6/611-01-v6.11-udp-Allow-GSO-transmit-from-devices-with-no-checksum.patch
John Audia 12f12df569 kernel: bump 6.6 to 6.6.55
Changelog: https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.6.55

Added the following default ksym to target/linux/generic/config-6.6:
	CONFIG_PROC_MEM_ALWAYS_FORCE=y
	# CONFIG_PROC_MEM_FORCE_PTRACE is not set
	# CONFIG_PROC_MEM_NO_FORCE is not set

Removed upstreamed:
	generic/backport-6.6/780-23-v6.12-r8169-Fix-spelling-mistake-tx_underun-tx_underrun.patch[1]
	generic/backport-6.6/780-25-v6.12-r8169-add-tally-counter-fields-added-with-RTL8125.patch[2]
	generic/pending-6.6/684-gso-fix-gso-fraglist-segmentation-after-pull-from-fr.patch[3]
	lantiq/patches-6.6/0025-v6.12-net-ethernet-lantiq_etop-fix-memory-disclosure.patch[4]

Manually rebased:
	bcm27xx/patches-6.6/950-0086-Main-bcm2708-bcm2709-linux-port.patch
	bcm27xx/patches-6.6/950-0998-i2c-designware-Add-support-for-bus-clear-feature.patch

All other patches automatically rebased.

1. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.6.56&id=f02fcb7283b1c25f7e3ae07d7a2c830e06eb1a62
2. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.6.56&id=1c723d785adb711496bc64c24240f952f4faaabf
3. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.6.56&id=af3122f5fdc0d00581d6e598a668df6bf54c9daa
4. https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.6.56&id=e66e38d07b31e177ca430758ed97fbc79f27d966

Build system: x86/64
Build-tested: x86/64/AMD Cezanne, flogic/xiaomi_redmi-router-ax6000-ubootmod, ramips/tplink_archer-a6-v3
Run-tested: x86/64/AMD Cezanne, flogic/xiaomi_redmi-router-ax6000-ubootmod, ramips/tplink_archer-a6-v3

Signed-off-by: John Audia <therealgraysky@proton.me>
Link: https://github.com/openwrt/openwrt/pull/16655
Signed-off-by: Nick Hainke <vincent@systemli.org>
2024-10-11 19:21:41 +02:00

95 lines
3.4 KiB
Diff

From: Jakub Sitnicki <jakub@cloudflare.com>
Date: Wed, 26 Jun 2024 19:51:26 +0200
Subject: [PATCH] udp: Allow GSO transmit from devices with no checksum offload
Today sending a UDP GSO packet from a TUN device results in an EIO error:
import fcntl, os, struct
from socket import *
TUNSETIFF = 0x400454CA
IFF_TUN = 0x0001
IFF_NO_PI = 0x1000
UDP_SEGMENT = 103
tun_fd = os.open("/dev/net/tun", os.O_RDWR)
ifr = struct.pack("16sH", b"tun0", IFF_TUN | IFF_NO_PI)
fcntl.ioctl(tun_fd, TUNSETIFF, ifr)
os.system("ip addr add 192.0.2.1/24 dev tun0")
os.system("ip link set dev tun0 up")
s = socket(AF_INET, SOCK_DGRAM)
s.setsockopt(SOL_UDP, UDP_SEGMENT, 1200)
s.sendto(b"x" * 3000, ("192.0.2.2", 9)) # EIO
This is due to a check in the udp stack if the egress device offers
checksum offload. While TUN/TAP devices, by default, don't advertise this
capability because it requires support from the TUN/TAP reader.
However, the GSO stack has a software fallback for checksum calculation,
which we can use. This way we don't force UDP_SEGMENT users to handle the
EIO error and implement a segmentation fallback.
Lift the restriction so that UDP_SEGMENT can be used with any egress
device. We also need to adjust the UDP GSO code to match the GSO stack
expectation about ip_summed field, as set in commit 8d63bee643f1 ("net:
avoid skb_warn_bad_offload false positives on UFO"). Otherwise we will hit
the bad offload check.
Users should, however, expect a potential performance impact when
batch-sending packets with UDP_SEGMENT without checksum offload on the
egress device. In such case the packet payload is read twice: first during
the sendmsg syscall when copying data from user memory, and then in the GSO
stack for checksum computation. This double memory read can be less
efficient than a regular sendmsg where the checksum is calculated during
the initial data copy from user memory.
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240626-linux-udpgso-v2-1-422dfcbd6b48@cloudflare.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -942,8 +942,7 @@ static int udp_send_skb(struct sk_buff *
kfree_skb(skb);
return -EINVAL;
}
- if (skb->ip_summed != CHECKSUM_PARTIAL || is_udplite ||
- dst_xfrm(skb_dst(skb))) {
+ if (is_udplite || dst_xfrm(skb_dst(skb))) {
kfree_skb(skb);
return -EIO;
}
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -380,6 +380,14 @@ struct sk_buff *__udp_gso_segment(struct
else
uh->check = gso_make_checksum(seg, ~check) ? : CSUM_MANGLED_0;
+ /* On the TX path, CHECKSUM_NONE and CHECKSUM_UNNECESSARY have the same
+ * meaning. However, check for bad offloads in the GSO stack expects the
+ * latter, if the checksum was calculated in software. To vouch for the
+ * segment skbs we actually need to set it on the gso_skb.
+ */
+ if (gso_skb->ip_summed == CHECKSUM_NONE)
+ gso_skb->ip_summed = CHECKSUM_UNNECESSARY;
+
/* update refcount for the packet */
if (copy_dtor) {
int delta = sum_truesize - gso_skb->truesize;
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1261,8 +1261,7 @@ static int udp_v6_send_skb(struct sk_buf
kfree_skb(skb);
return -EINVAL;
}
- if (skb->ip_summed != CHECKSUM_PARTIAL || is_udplite ||
- dst_xfrm(skb_dst(skb))) {
+ if (is_udplite || dst_xfrm(skb_dst(skb))) {
kfree_skb(skb);
return -EIO;
}