linux-kernel - Re: [PATCH] KGDB: refactor kconfig menu

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: <479CCA28.90806@web.de>
Date:	Sun, 27 Jan 2008 19:15:04 +0100
From:	Jan Kiszka <jan.kiszka@....de>
To:	Jason Wessel <jason.wessel@...driver.com>
CC:	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Ingo Molnar <mingo@...e.hu>
Subject: Re: [PATCH] KGDB: refactor kconfig menu

Jason Wessel wrote:
> Jan Kiszka wrote:
>> This is an attempt to improve the so far, well, unfortunate Kconfig menu
>> organization of KGDB. Most notably, it pushes all sub-entries into their
>> own menuconfig, removes the (IMHO) unneeded "Method for KGDB
>> communication" choice, and ensures everything is indented properly. This
>> should keep the original semantic while reducing the number of knobs the
>> user is confronted with.
>>
>>   
> 
> While it is probably reasonable to put the kgdb options into a sub
> entry,  you cannot get rid of the "Method for KGDB communication"
> choice, unless you re-implement it another way.
> 
> It is designed to implement the following:
> 1) You can chose to build 1 and only 1 KGDB I/O module into the kernel
> 2) Or you select modules only and can you can build as many additional
>     KGDB I/O modules as you like.
> 3) Additional I/O modules can be built and you can load one at run time
>     if you want to use another I/O method vs #1

I see no reason why these cases cannot be enforced later on (when there
are >1 I/O modules) via Kconfig depends-on rules. But maybe I'm
overseeing something, please correct me.

> 
> KGDB's Kconfig options are designed to allow you to have alternate ways
> to select static or run time configuration of the I/O connection to the
> KGDB core, IE: over serial, over ethernet, over USB or what ever module
> you would like to implement.

Well, I'm looking at this from the user perspective now, and as a user I
don't understand why I shouldn't be able to configure my setup via the
tristates of the involved I/O drivers (+ some underlying dependencies).

Jan


Download attachment "signature.asc" of type "application/pgp-signature" (255 bytes)