linux-kernel - Re: [PATCH] kbuild: Allow to combine multiple W= levels

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 for Android: free password hash cracker in your pocket

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

Date:	Mon, 2 May 2011 13:34:54 -0400
From:	Arnaud Lacombe <lacombar@...il.com>
To:	Sam Ravnborg <sam@...nborg.org>
Cc:	Américo Wang <xiyou.wangcong@...il.com>,
	Michal Marek <mmarek@...e.cz>, Valdis.Kletnieks@...edu,
	bp@...en8.de, linux-kbuild@...r.kernel.org,
	linux-kernel@...r.kernel.org, davej@...hat.com
Subject: Re: [PATCH] kbuild: Allow to combine multiple W= levels

Hi,

On Mon, May 2, 2011 at 1:07 PM, Sam Ravnborg <sam@...nborg.org> wrote:
> On Tue, May 03, 2011 at 12:16:15AM +0800, Américo Wang wrote:
>> On Fri, Apr 29, 2011 at 9:31 PM, Michal Marek <mmarek@...e.cz> wrote:
>> > Add support for make W=12, make W=123 and so on, to enable warnings from
>> > multiple W= levels. Normally, make W=<level> does not include warnings
>> > from the previous level.
>> >
>>
>> This interface is not friendly, at least not as normal as we often see.
>> W=x+1 is supposed to include warnings of W=x.
>>
>> Please refine the interface.
>
> Until we see that several people have had benefit if W=...
> we should leave it as is now.
>
> Then when people really start to use if we can refine it.
>
> IMO it is much more important to find the right sub-set
> of warnings to keep on W=1 level than how we see additional warnings.
>
> There may well be warnings where we say that the benefit of it is
> zero - or it is plain wrong in the kernel.
> Lets try to focus on this.
>
Because you are not even sure of what you submitted is good for the kernel ?

I think this has been done completely wrong, first all the extra
warnings got in (without a detailed impact implied by each one), then
it was decided to be split in several level (which I proposed on Feb
20th, but was originally rejected by Borislav Petkov), then, well,
some may need to be removed because they are not good for Linux. I
would rather have provided the framework first, clean, natural
interface, then added new warnings gradually when we were sure had a
positive impact on the kernel, with the associated patch fixing them.
Right now, +80k warnings will be just impossible to fix.

Now, the drawback of your comment is that when it will be time to
change the interface, some will object because it will have been
started to be used by third party scripts, and as authors of third
party script, it is a PITA to have to check the kernel version to know
if I should W=1,2,3, or W=123 or if W=3 includes W=1 ...

my 0.2c,

 - Arnaud
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/