netdev - Re: [PATCH 2/2] net_sched: optimize tcf_match

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: <CAM_iQpWA2qjjEV97SFajg0Y_0WnWAY2qPY9_7vSuFVuYq5tH6Q@mail.gmail.com>
Date:	Thu, 19 Dec 2013 22:36:01 -0800
From:	Cong Wang <xiyou.wangcong@...il.com>
To:	Eric Dumazet <eric.dumazet@...il.com>
Cc:	Linux Kernel Network Developers <netdev@...r.kernel.org>,
	Jamal Hadi Salim <jhs@...atatu.com>,
	"David S. Miller" <davem@...emloft.net>
Subject: Re: [PATCH 2/2] net_sched: optimize tcf_match_indev()

On Thu, Dec 19, 2013 at 10:22 PM, Eric Dumazet <eric.dumazet@...il.com> wrote:
> On Thu, 2013-12-19 at 17:34 -0800, Cong Wang wrote:
>> tcf_match_indev() is called in fast path, it is not wise to
>> search for a netdev by ifindex and then compare by its name,
>> just compare the ifindex.
>>
>> This will also save some bytes from the core struct of u32.
>>
>
> This seems very suspect to me. How was it tested ?

Why? I tested it with 1000 u32 filters with mirred actions
to a dummy interface, it is the test case I used from the beginning.

>
> BTW, latest net-next doesn't work anymore.
>
> [   76.591023] BUG: unable to handle kernel paging request at ffffffffffffffff^M
> [   76.598025] IP: [<ffffffffa008d882>] mirred_cleanup_module+0x13a/0x2a [act_mirred]^M


All of my kernel modules are built-in, this is why I can't catch this bug.
So why not share your test case with me?

> I wonder how you tested "net_sched: init struct tcf_hashinfo at register time",
> it is completely buggy.
>

Since I compile everything as built-in, so just boot and shutdown the VM,
and of course, run some test cases like above one.

> static inline int tcf_hashinfo_init(struct tcf_hashinfo *hf, unsigned int mask)
>
> mask is a mask, yet all callers pass 'MASK+1'
>

Hmm, off-by-one error, stupid me.

Again, it is always welcome if you can share your test case.

Thanks.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html