This reverts commit 30bb896b98fce7d823a96fc02cd69be30384a5cc.
The patch currently broke the bpf selftest test_tc_dtime because
uapi field __sk_buff->tstamp_type depends on skb->mono_delivery_time which
does not necessarily mean mono with the original fix as the bit was re-used
for userspace timestamp as well to avoid tstamp reset in the forwarding
path. To solve this we need to keep mono_delivery_time as ease and
introduce another bit called user_delivery_time and fall back to the
initial proposal of setting the user_delivery_time bit based on
sk_clockid set from userspace.
Link: https://lore.kernel.org/netdev/[email protected]/
Signed-off-by: Abhishek Chauhan <[email protected]>
---
Changes since v1
- Took care of Jakub's comment to explain more about the
revert commit
- Added Link to the discussion of the problem found in the
original commit.
include/linux/skbuff.h | 6 +++---
net/ipv4/ip_output.c | 1 -
net/ipv4/raw.c | 1 -
net/ipv6/ip6_output.c | 2 +-
net/ipv6/raw.c | 2 +-
net/packet/af_packet.c | 4 +---
6 files changed, 6 insertions(+), 10 deletions(-)
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 4726298d4ed4..2dde34c29203 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -817,9 +817,9 @@ typedef unsigned char *sk_buff_data_t;
* @decrypted: Decrypted SKB
* @slow_gro: state present at GRO time, slower prepare step required
* @mono_delivery_time: When set, skb->tstamp has the
- * delivery_time in mono clock base (i.e., EDT) or a clock base chosen
- * by SO_TXTIME. If zero, skb->tstamp has the (rcv) timestamp at
- * ingress.
+ * delivery_time in mono clock base (i.e. EDT). Otherwise, the
+ * skb->tstamp has the (rcv) timestamp at ingress and
+ * delivery_time at egress.
* @napi_id: id of the NAPI struct this skb came from
* @sender_cpu: (aka @napi_id) source CPU in XPS
* @alloc_cpu: CPU which did the skb allocation.
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index ff1df64c5697..5b5a0adb927f 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -1455,7 +1455,6 @@ struct sk_buff *__ip_make_skb(struct sock *sk,
skb->priority = (cork->tos != -1) ? cork->priority: READ_ONCE(sk->sk_priority);
skb->mark = cork->mark;
skb->tstamp = cork->transmit_time;
- skb->mono_delivery_time = !!skb->tstamp;
/*
* Steal rt from cork.dst to avoid a pair of atomic_inc/atomic_dec
* on dst refcount
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index c4c29fc5b73f..aea89326c697 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -353,7 +353,6 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4,
skb->priority = READ_ONCE(sk->sk_priority);
skb->mark = sockc->mark;
skb->tstamp = sockc->transmit_time;
- skb->mono_delivery_time = !!skb->tstamp;
skb_dst_set(skb, &rt->dst);
*rtp = NULL;
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 2fc1d03dc07d..a722a43dd668 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -1922,7 +1922,7 @@ struct sk_buff *__ip6_make_skb(struct sock *sk,
skb->priority = READ_ONCE(sk->sk_priority);
skb->mark = cork->base.mark;
skb->tstamp = cork->base.transmit_time;
- skb->mono_delivery_time = !!skb->tstamp;
+
ip6_cork_steal_dst(skb, cork);
IP6_INC_STATS(net, rt->rt6i_idev, IPSTATS_MIB_OUTREQUESTS);
if (proto == IPPROTO_ICMPV6) {
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 13f54f8eea35..03dbb874c363 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -616,7 +616,7 @@ static int rawv6_send_hdrinc(struct sock *sk, struct msghdr *msg, int length,
skb->priority = READ_ONCE(sk->sk_priority);
skb->mark = sockc->mark;
skb->tstamp = sockc->transmit_time;
- skb->mono_delivery_time = !!skb->tstamp;
+
skb_put(skb, length);
skb_reset_network_header(skb);
iph = ipv6_hdr(skb);
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 0db31ca4982d..c9bbc2686690 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -2057,7 +2057,7 @@ static int packet_sendmsg_spkt(struct socket *sock, struct msghdr *msg,
skb->priority = READ_ONCE(sk->sk_priority);
skb->mark = READ_ONCE(sk->sk_mark);
skb->tstamp = sockc.transmit_time;
- skb->mono_delivery_time = !!skb->tstamp;
+
skb_setup_tx_timestamp(skb, sockc.tsflags);
if (unlikely(extra_len == 4))
@@ -2586,7 +2586,6 @@ static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb,
skb->priority = READ_ONCE(po->sk.sk_priority);
skb->mark = READ_ONCE(po->sk.sk_mark);
skb->tstamp = sockc->transmit_time;
- skb->mono_delivery_time = !!skb->tstamp;
skb_setup_tx_timestamp(skb, sockc->tsflags);
skb_zcopy_set_nouarg(skb, ph.raw);
@@ -3065,7 +3064,6 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
skb->priority = READ_ONCE(sk->sk_priority);
skb->mark = sockc.mark;
skb->tstamp = sockc.transmit_time;
- skb->mono_delivery_time = !!skb->tstamp;
if (unlikely(extra_len == 4))
skb->no_fcs = 1;
--
2.25.1
Abhishek Chauhan wrote:
> This reverts commit 30bb896b98fce7d823a96fc02cd69be30384a5cc.
Upstream SHA1 is 885c36e59f46375c138de18ff1692f18eff67b7f ?
> The patch currently broke the bpf selftest test_tc_dtime because
> uapi field __sk_buff->tstamp_type depends on skb->mono_delivery_time which
> does not necessarily mean mono with the original fix as the bit was re-used
> for userspace timestamp as well to avoid tstamp reset in the forwarding
> path. To solve this we need to keep mono_delivery_time as ease and
minor typo: as ease -> as is
> introduce another bit called user_delivery_time and fall back to the
> initial proposal of setting the user_delivery_time bit based on
> sk_clockid set from userspace.
>
> Link: https://lore.kernel.org/netdev/[email protected]/
> Signed-off-by: Abhishek Chauhan <[email protected]>
Reviewed-by: Willem de Bruijn <[email protected]>
once the SHA1 is verified/fixed.
On Thu, 14 Mar 2024 05:54:14 -0400 Willem de Bruijn wrote:
> Abhishek Chauhan wrote:
> > This reverts commit 30bb896b98fce7d823a96fc02cd69be30384a5cc.
>
> Upstream SHA1 is 885c36e59f46375c138de18ff1692f18eff67b7f ?
Yes, and please also throw in a Fixes tag pointing at that
commit.