netdev - Re: Fwd: u32 ht filters

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

Message-ID: <20180207070148.GA2149@nanopsycho.orion>
Date:   Wed, 7 Feb 2018 08:01:48 +0100
From:   Jiri Pirko <jiri@...nulli.us>
To:     Cong Wang <xiyou.wangcong@...il.com>
Cc:     Jiri Pirko <jiri@...lanox.com>,
        Linux Kernel Network Developers <netdev@...r.kernel.org>,
        Paweł Staszewski <pstaszewski@...are.pl>
Subject: Re: Fwd: u32 ht filters

Wed, Feb 07, 2018 at 06:09:15AM CET, xiyou.wangcong@...il.com wrote:
>Hi, Jiri
>
>Your  commit 7fa9d974f3c2a016b9accb18f4ee2ed2a738585c
>breaks the tc script by Paweł. Please find below for details.

Did you do the bisection?
The commit just uses block struct instead of q, but since they
are in 1:1 relation, that should be equvivalent. So basically you still
have per-qdisc hashtables for u32.


>
>
>commit 7fa9d974f3c2a016b9accb18f4ee2ed2a738585c
>Author: Jiri Pirko <jiri@...lanox.com>
>Date:   Fri Oct 13 14:01:02 2017 +0200
>
>    net: sched: cls_u32: use block instead of q in tc_u_common
>
>    tc_u_common is now per-q. With blocks, it has to be converted to be
>    per-block.
>
>    Signed-off-by: Jiri Pirko <jiri@...lanox.com>
>    Signed-off-by: David S. Miller <davem@...emloft.net>
>
>Before this commit, u32 hashtables are per-qdisc, after this commit
>it becomes per-block or per-class... this is why the script below is broken.
>
>
>---------- Forwarded message ----------
>From: Paweł Staszewski <pstaszewski@...are.pl>
>Date: Tue, Feb 6, 2018 at 8:05 AM
>Subject: u32 ht filters
>To: Cong Wang <xiyou.wangcong@...il.com>
>
>
>Hi
>
>
>Is there something changed in kernek 4.15 that makes problem with old
>configuration of tc filters with hashing filters ?
>
>for example this :
>
>tc qdisc del root dev ifb1
>
>tc qdisc add dev ifb1 root handle 1:0 hfsc default 8000
>tc filter add dev ifb1 parent 1:0 protocol ip u32
>tc class add dev ifb1 parent 1:0 classid 1:1 hfsc ls m2 10000Mbit ul
>m2 10000Mbit
>tc class add dev ifb1 parent 1:1 classid 1:2 hfsc ls m2 10000Mbit ul
>m2 10000Mbit
>tc class add dev ifb1 parent 1:1 classid 1:3 hfsc ls m2 5000Mbit ul m2 5000Mbit
>tc class add dev ifb1 parent 1:2 classid 1:8000 hfsc ls m2 10000Mbit
>ul m2 10000Mbit
>tc qdisc add dev ifb1 parent 1:8000 handle 8000: sfq perturb 60
>tc qdisc add dev ifb1 parent 1:3 handle 3: pfifo limit 10000
>
>
>tc filter add dev ifb1 protocol ip parent 1:0 handle 9: u32 divisor 256
>tc filter add dev ifb1 protocol ip parent 1:0 u32 ht 800:: match ip
>dst 192.168.0.0/24 hashkey mask 0x000000ff at 16 link 9:
>tc class add dev ifb1 parent 1:2 classid 1:60 hfsc ls m2 8kbit ul m2 51200kbit
>echo 1
>tc filter add dev ifb1 parent 1:2 protocol ip u32 ht 9:22 match ip dst
>192.168.0.34 flowid 1:60
>echo 2
>tc qdisc add dev ifb1 parent 1:60 handle 60: pfifo limit 8192
>
>
>Is working with 4.13
>
>
>But it is not working with 4.15
>
>error is when adding:
>
>tc filter add dev ifb1 protocol ip parent 1:2 prio 4 u32 ht 9:0x22
>match ip dst 192.168.0.34 flowid 1:60
>RTNETLINK answers: Invalid argument
>We have an error talking to the kernel
>
>
>
>
>Thanks
>
>Paweł Staszewski