netdev - Re: [PATCH] dev: use ifindex hash for dev_seq

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: <CAOMsUMKWzKNNR-cDK7eSK-+F-hhCc8BhkzrpLHRs2tsmjXxdXw@mail.gmail.com>
Date:	Fri, 14 Oct 2011 15:20:40 +0300
From:	Mihai Maruseac <mihai.maruseac@...il.com>
To:	Stephen Hemminger <shemminger@...tta.com>
Cc:	davem@...emloft.net, eric.dumazet@...il.com,
	mirq-linux@...e.qmqm.pl, therbert@...gle.com, jpirko@...hat.com,
	netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
	dbaluta@...acom.com, Mihai Maruseac <mmaruseac@...acom.com>
Subject: Re: [PATCH] dev: use ifindex hash for dev_seq_ops

On Wed, Oct 12, 2011 at 6:50 PM, Stephen Hemminger
<shemminger@...tta.com> wrote:
> On Wed, 12 Oct 2011 12:57:26 +0300
> Mihai Maruseac <mihai.maruseac@...il.com> wrote:
>
> It looks like you lost the ability to seek back and read the header
> (start token).  How is the header handled, is possible to rewind
> the file and read it over again?

We tested with a simple program reading a part of the /proc file and
then doing a seek to the start and rereading. It worked with latest
submitted patch.

>> +static inline struct net_device *next_dev(struct seq_file *seq, loff_t *pos)
>> +{
>> +     struct dev_iter_state *state = seq->private;
>> +     struct net *net = seq_file_net(seq);
>> +     struct net_device *dev = NULL;
>> +     loff_t off;
>> +
>> +     ++*pos;
>> +     dev = dev_get_by_index_rcu(net, state->ifindex);
>
> Looks good a couple of minor nits.
>
> 1. The function should not be inline, since it is in no way performance
>   critical. The compiler will probably inline it anyway.
>
> 2. dev does not have to be initialized since it is assigned a few
>   lines later.  Most programmers are trained now to always initialize
>   variables, but often it is unnecessary.
>
> 3. The name next_dev() is a little generic; maybe a better name.
>

Fixed all of them,

Thanks,
-- 
Mihai
--
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