linux-kernel - Re: [PATCH v3 20/21] dyndbg: add user-flag, negating-flags, and filtering on flags

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]

Message-ID: <20200619074555.GF3617@alley>
Date:   Fri, 19 Jun 2020 09:45:55 +0200
From:   Petr Mladek <pmladek@...e.com>
To:     jim.cromie@...il.com
Cc:     Jason Baron <jbaron@...mai.com>,
        LKML <linux-kernel@...r.kernel.org>, akpm@...uxfoundation.org,
        Greg KH <gregkh@...uxfoundation.org>,
        Rasmus Villemoes <linux@...musvillemoes.dk>,
        Jonathan Corbet <corbet@....net>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Will Deacon <will@...nel.org>,
        Orson Zhai <orson.zhai@...soc.com>,
        Linux Documentation List <linux-doc@...r.kernel.org>
Subject: Re: [PATCH v3 20/21] dyndbg: add user-flag, negating-flags, and
 filtering on flags

On Thu 2020-06-18 13:11:05, jim.cromie@...il.com wrote:
> On Thu, Jun 18, 2020 at 12:17 PM Jason Baron <jbaron@...mai.com> wrote:
> > Yes, I'm wondering as well if people are really going to use the
> > new flags and filter flags - I mentioned that here:
> > https://lkml.org/lkml/2020/6/12/732
> 
> yes, I saw, and replied there.

No, the repply only explains how the interface might be used. There is
no prove that people would actually use it.

> but since that was v1, and we're on v3, we should refresh.
> 
> the central use-case is above, 1-liner version summarized here:
> 
> 1- enable sites as you chase a problem with +up
> 2- examine them with grep =pu
> 3- change the set to suit, either by adding or subtracting callsites.
> 4- continue debugging, and changing callsites to suit
> 5- grep =pu control > ~/debugging-session-task1-callsites
> 6- echo up-p >control   # disable for now, leave u-set for later
> 7- do other stuff
> 8 echo uP+p >control # reactivate useful debug-state and resume

In short, this feature allows repeatedly enable/disable some
slowly growing maze of debug messages. Who need this, please? !!!

If I am debugging then I add/remove debug messages. But I never
enable/disable all of them repeatedly.

Also this is far from the original problem. It was about debugging
a single driver (venus, drm). In this case, people need something
easy to use. The following is the easy way:

   drm.debug = area_of_interest
   venus.debug = level_of_interest

   echo module=drm group=area_of_interest +p >control  [*]
   echo module=venus group=level_of_interes +p >control

Anyway, why filtering and 'u' flag would be necessary to debug these drivers?
Is anyone going to use it?

I would really like to hear the motivation for these features.
Has anyone asked for them?
Or are these just some "interesting" ideas from some brainstorming?


[*] Well, I wonder if the dyndbg interface would even be useful for drm
    because it it is actually split into many modules. So it might require
    creating/maintaining several filters.


Best Regards,
Petr

PS: This is probably my last mail in this thread. It goes in cycle.
    You repeatedly explain how many possibilities the new features
    allow. I repeatedly doubt that they are worth it.

    I also proposed another solution for the original problem (venus)
    but it has never been commented.

    I just hope that these features will not get merged without
    a clear interest in them.