netdev - Re: [PATCH]ipv6:use ipv6_addr_type instead of __ipv6_addr

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]

Date:	Fri, 08 May 2009 16:13:33 +0800
From:	Yang Hongyang <yanghy@...fujitsu.com>
To:	David Miller <davem@...emloft.net>
CC:	netdev@...r.kernel.org, yoshfuji@...ux-ipv6.org
Subject: Re: [PATCH]ipv6:use ipv6_addr_type instead of __ipv6_addr_type

David Miller wrote:
> From: Yang Hongyang <yanghy@...fujitsu.com>
> Date: Fri, 08 May 2009 09:18:11 +0800
> 
>> David Miller wrote:
>>> From: Yang Hongyang <yanghy@...fujitsu.com>
>>> Date: Thu, 07 May 2009 15:47:38 +0800
>>>
>>>> Although the function of these two functions are the same,we should use the
>>>> encapsulation function of __ipv6_addr_type
>>>>
>>>> Signed-off-by: Yang Hongyang<yanghy@...fujitsu.com>
>>> Are you sure?  ipv6_addr_type() masks out the upper 16-bits
>>> of __ipv6_addr_type()'s return value, which is the scope.
>>>
>>> I'm not convinced that is what we want to do here.
>> My advice here is that the function which has __ maybe shouldn't be directly used.
> 
> You're using vague terms like "advice" and "maybe" and this
> only decreases my confidence in your change.

Yes,I said maybe because I judge this change by search the code.There are many
places that use ipv6_addr_type() but only three places out of ipv6.h use
__ipv6_addr_type().Now, there are two methods can  to do the right things,
one is ipv6_addr_type(),another is __ipv6_addr_type().There's no doult that
These three places can be converted to use ipv6_addr_type().I said converted
because it's not just change __ipv6_addr_type() to ipv6_addr_type().

Of course you can say that this dose not imply that we *should* do this change.
I agree,so I just said this is a advice:)

> 
> You're also not addressing my questions at all.
> 
> Two "_" characters at the beginning of the function's name says
> nothing about whether these call sites want the scope bits present in
> the return value they receive.

I used to think Two "_" means the funtion shouldn't(or should better not)
be called outside of the  file.You teach me,thank you!

> 
> This seems like a completely pointless and mindless change,
> and in fact might introduce bugs.

If you think this is a pointless change,you can just ingore it,
So sorry for the inconvenience:(But I don't think it's mindless,Because
before you teach me, I really misunderstand  Two "_" means .

> 
> There is no way I'm applying this.  You can't even explain if the
> patch is correct or not.  In fact, I doubt you even know.
> 

Yes,this single patch is not correct.As I said,this patch should together with
another patch to covert use of __ipv6_addr_src_scope to ipv6_addr_src_scope.

-- 
Regards
Yang Hongyang
--
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