From 173b6d1cdf582e7438b3ab4ef2f40e6833579490 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 12 Dec 2023 11:06:08 +0000 Subject: [PATCH] docs: networking: timestamping: mention MSG_EOR flag TCP got MSG_EOR support in linux-4.7. This is a canonical way of making sure no coalescing will be performed on the skb, even if it could not be immediately sent. Signed-off-by: Eric Dumazet Cc: Martin KaFai Lau Acked-by: Willem de Bruijn Link: https://lore.kernel.org/r/20231212110608.3673677-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- Documentation/networking/timestamping.rst | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/Documentation/networking/timestamping.rst b/Documentation/networking/timestamping.rst index f17c01834a12..5e93cd71f99f 100644 --- a/Documentation/networking/timestamping.rst +++ b/Documentation/networking/timestamping.rst @@ -357,7 +357,8 @@ enabling SOF_TIMESTAMPING_OPT_ID and comparing the byte offset at send time with the value returned for each timestamp. It can prevent the situation by always flushing the TCP stack in between requests, for instance by enabling TCP_NODELAY and disabling TCP_CORK and -autocork. +autocork. After linux-4.7, a better way to prevent coalescing is +to use MSG_EOR flag at sendmsg() time. These precautions ensure that the timestamp is generated only when all bytes have passed a timestamp point, assuming that the network stack