netdev - ingress bandwidth limitation for ipv6 is inaccurate

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-next>] [day] [month] [year] [list]

Message-ID: <CAPaK2r_T28BqvOkjDSD=SR-5sKeD_HgHu1tvB+b1jR20FuU0WQ@mail.gmail.com>
Date:   Tue, 8 Oct 2019 13:53:13 +0800
From:   yue longguang <yuelongguang@...il.com>
To:     kuznet@....inr.ac.ru, stephen@...workplumber.org,
        netdev@...r.kernel.org
Subject: ingress bandwidth limitation for ipv6 is inaccurate

Firstly, thank you  stephen.

Hi, kuznet:

1.  according to my test,  bandwidth limitation is inaccurate for
ingress ipv6 .  The accuracy is affected by burst.   if rate is less
than 100mbit , set burst to 1MB or 2MB, the result is almost
acceptable.  but when rate is bigger , the result is approximately 1/3
of rate.
command:  tc filter add dev qr-869fbdc2-1e parent ffff: protocol ipv6
u32 match ip6 src any police rate 500mbit burst 1M drop
so except for using ifb, what should be done to get a accurate result.

2. can flowid option of  ingess's filter belong to egress's class.
for example    tc filter add dev qr-869fbdc2-1e parent ffff: protocol
ipv6 u32 match ip6 src any police rate 500mbit burst 1M flowid 1:10
 (1:10 classid  is egress's qdisc's class )

thanks