netdev - Re: [PATCH] bridge: Fix format string for %ul

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: <D79B6506-31BF-4B39-A81C-F3900F062164@linuxhacker.ru>
Date:   Sat, 27 Aug 2016 12:36:49 -0400
From:   Oleg Drokin <green@...uxhacker.ru>
To:     Sergei Shtylyov <sergei.shtylyov@...entembedded.com>
Cc:     Stephen Hemminger <stephen@...workplumber.org>,
        "David S. Miller" <davem@...emloft.net>,
        bridge@...ts.linux-foundation.org, netdev@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] bridge: Fix format string for %ul


On Aug 27, 2016, at 12:18 PM, Sergei Shtylyov wrote:

> Hello.
> 
> On 8/27/2016 6:58 PM, Oleg Drokin wrote:
> 
>>>> %ul would print an unsigned value and a letter l,
>>>> likely it was %lu that was meant to print the long int,
>>>> but in reality the values printed there are just regular signed
>>> 
>>>  Signed? Then you need probably "%d" or "%i"…
>> 
>> They are signed in the struct definition, but in reality they
>> designate time, so could not be negative, I imagine?
> 
>   That doesn't matter. If the type is signed, it should be printed as signed.
> Doesn't gcc complain about the format specifiers not matching the values passed?

only if the size differs.

I don't care either way.
I can change the struct to unsigned too, but if we want super correctness,
the problem lies deeper.

E.g. if we look how those times are derived, the age comes as:
static inline int br_get_ticks(const unsigned char *src)
{
        unsigned long ticks = get_unaligned_be16(src);

        return DIV_ROUND_UP(ticks * HZ, STP_HZ);
}

So we divide unsigned value by positive value and get a signed result ;)

there's a struct net_bridge_port with similar members in this are that are
unsigned long, so I guess we should convert to unsigned long in the
struct br_config_bpdu, do you want a patch for that separately or as part of this one?

> 
>>>> ints, so just dropping the l altogether.
>>>> 
>>>> Signed-off-by: Oleg Drokin <green@...uxhacker.ru>
>>> [...]
> 
> MBR, Sergei