linux-kernel - Re: [PATCH] make gconfig: fix the "(NEW)" string

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] [day] [month] [year] [list]

Date:	Wed, 09 Dec 2009 02:40:46 +0100
From:	Matteo Cortese <matteo_cortese@...twebnet.it>
To:	Randy Dunlap <randy.dunlap@...cle.com>
Cc:	linux-kbuild@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] make gconfig: fix the "(NEW)" string

On 04/12/2009 11:02 -0800, Randy Dunlap wrote:

> Did you just notice that thru source code inspection or via testing of
> gconfig?

I noticed the problem with the following test case (with the patch NOT
applied). 

$ rm -f .config
$ make clean && make gconfig

Select File->Save to save a "clean" version of .config and exit.
Now let's target a specific option, e.g. CONFIG_MODULES, and check that
it is present, either set to 'y' or commented out (mine is set to 'y'):

$ grep 'CONFIG_MODULES' .config
CONFIG_MODULES=y
$ make gconfig

At this point I see that module support has the NEW flag, but it
shouldn't! Exit, then manually remove CONFIG_MODULES from the .config
file:

$ sed -i '/CONFIG_MODULES/d' .config
$ make clean && make gconfig

Now I see that module support is not flagged as NEW, but it should.

I concluded that the logic was inverted.
I looked up the string NEW in the source and found the piece of code
cited below. Then I tried and figured out what that code was supposed to
do, and thought that it made more sense that the NEW flag was added when
an option (sym) had not been assigned a value yet (!sym_has_value).
So I worked out this simple patch. I've been using it for some time, and
it works for me, I mean, if I try the above test case, it gives the
expected results.

> Well, for me, with just some random config file that has
> CONFIG_MODULES=y
> CONFIG_BLOCK=y
> 
> gconfig with this patch says:
> 
>   Enable loadable module support
>   Enable the block layer (NEW)
> 
> and gconfig without this patch says:
> 
>   Enable loadable module support (NEW)
>   Enable the block layer
> 
> 
> I'm confuzed.  Go figure.

Yes, it makes no sense. Unfortunately I have no experience with .config
files generated in any other way than written by the GTK configurator...

Please bear in mind that I'm not a kernel developer, I just recompile my
kernel when a new version is released. I usually copy my .config file
from the previous kernel directory and then use the "NEW" flag to tell
which options were added since my last compile and thus deserve my
attention.
---
Matteo

--
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/