From: Weiping Zhang <zhangweiping@didiglobal.com>
To: <edumazet@google.com>, <davem@davemloft.net>,
<yoshfuji@linux-ipv6.org>, <dsahern@kernel.org>,
<kuba@kernel.org>, <pabeni@redhat.com>
Cc: <netdev@vger.kernel.org>, <zwp10758@gmail.com>
Subject: [RFC PATCH] tcp: correct srtt and mdev_us calculation
Date: Tue, 6 Dec 2022 02:07:36 +0800 [thread overview]
Message-ID: <Y44zaD2DHdPENblB@zwp-5820-Tower> (raw)
From the comments we can see that, rtt = 7/8 rtt + 1/8 new,
but there is an mistake,
m -= (srtt >> 3);
srtt += m;
explain:
m -= (srtt >> 3); //use t stands for new m
t = m - srtt/8;
srtt = srtt + t
= srtt + m - srtt/8
= srtt 7/8 + m
Test code:
#include<stdio.h>
#define u32 unsigned int
static void test_old(u32 srtt, long mrtt_us)
{
long m = mrtt_us;
u32 old = srtt;
m -= (srtt >> 3);
srtt += m;
printf("%s old_srtt: %u mrtt_us: %ld new_srtt: %u\n", __func__, old, mrtt_us, srtt);
}
static void test_new(u32 srtt, long mrtt_us)
{
long m = mrtt_us;
u32 old = srtt;
m = ((m - srtt) >> 3);
srtt += m;
printf("%s old_srtt: %u mrtt_us: %ld new_srtt: %u\n", __func__, old, mrtt_us, srtt);
}
int main(int argc, char **argv)
{
u32 srtt = 100;
long mrtt_us = 90;
test_old(srtt, mrtt_us);
test_new(srtt, mrtt_us);
return 0;
}
./a.out
test_old old_srtt: 100 mrtt_us: 90 new_srtt: 178
test_new old_srtt: 100 mrtt_us: 90 new_srtt: 98
Signed-off-by: Weiping Zhang <zhangweiping@didiglobal.com>
---
net/ipv4/tcp_input.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 0640453fce54..0242bb31e1ce 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -848,7 +848,7 @@ static void tcp_rtt_estimator(struct sock *sk, long mrtt_us)
* that VJ failed to avoid. 8)
*/
if (srtt != 0) {
- m -= (srtt >> 3); /* m is now error in rtt est */
+ m = (m - srtt >> 3); /* m is now error in rtt est */
srtt += m; /* rtt = 7/8 rtt + 1/8 new */
if (m < 0) {
m = -m; /* m is now abs(error) */
@@ -864,7 +864,7 @@ static void tcp_rtt_estimator(struct sock *sk, long mrtt_us)
if (m > 0)
m >>= 3;
} else {
- m -= (tp->mdev_us >> 2); /* similar update on mdev */
+ m = (m - tp->mdev_us >> 2); /* similar update on mdev */
}
tp->mdev_us += m; /* mdev = 3/4 mdev + 1/4 new */
if (tp->mdev_us > tp->mdev_max_us) {
--
2.34.1
next reply other threads:[~2022-12-05 18:07 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-12-05 18:07 Weiping Zhang [this message]
2022-12-06 0:46 ` [RFC PATCH] tcp: correct srtt and mdev_us calculation kernel test robot
-- strict thread matches above, loose matches on Subject: below --
2022-12-05 17:59 Weiping Zhang
2022-12-05 18:27 ` Eric Dumazet
2022-12-05 19:15 ` Neal Cardwell
2022-12-06 9:11 ` Jakub Sitnicki
2022-12-10 13:35 ` Weiping Zhang
2022-12-10 17:45 ` Eric Dumazet
2022-12-11 17:04 ` Stephen Hemminger
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=Y44zaD2DHdPENblB@zwp-5820-Tower \
--to=zhangweiping@didiglobal.com \
--cc=davem@davemloft.net \
--cc=dsahern@kernel.org \
--cc=edumazet@google.com \
--cc=kuba@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=yoshfuji@linux-ipv6.org \
--cc=zwp10758@gmail.com \
/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.