linux-kernel - Re: [PATCH 00/16] dynamic_debug: cleanups, 2 features

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]

Message-ID: <CAJfuBxy0YZuPHRhEn4rG4poKv3te9vJRDpJY1AvtMzGH7O9gMw@mail.gmail.com>
Date:   Fri, 12 Jun 2020 20:19:16 -0600
From:   jim.cromie@...il.com
To:     Jason Baron <jbaron@...mai.com>
Cc:     LKML <linux-kernel@...r.kernel.org>, akpm@...uxfoundation.org,
        Greg KH <gregkh@...uxfoundation.org>,
        Rasmus Villemoes <linux@...musvillemoes.dk>
Subject: Re: [PATCH 00/16] dynamic_debug: cleanups, 2 features

On Fri, Jun 12, 2020 at 3:31 PM Jason Baron <jbaron@...mai.com> wrote:
>
>
>
> On 6/5/20 12:26 PM, Jim Cromie wrote:
> > Patchset starts with 7 "cleanups";
> > - it changes section name from vague "__verbose" to "__dyndbg"
> > - cleaner docs, drop obsolete comment & useless debug prints, refine
> >   verbosity, fix a BUG_ON, ram reporting miscounts.
> >
> > It adds a few query parsing conveniences;
> > accept combined file:line & file:func forms
> >
> >   file inode.c:100-200                # file & line-range
> >   file inode.c:start_*                # file & function
> >
>
> So I like the shortened notation there.
>
> > Then it expands flags:
> >
> > Adds 'u' user flag, allowing user to compose an arbitrary set of
> > callsites by marking them with 'u', without altering current
> > print-modifying flags.
> >
> > Adds 'PFMLTU' flags, which negate their lower-case counterparts.
> >
> > Extends flags-spec with filter-flags, which select callsites for
> > modification based upon their current flags.  This lets user activate
> > the set of callsites marked with 'u' in a batch.
> >
> >   echo 'u+p' > control
> >
>
> I'm wondering if users are really going to use these and how much they
> simplify things? Do you find them useful while debugging issues?
>
> Especially now that now that we are looking to let people define
> groupings.
>
> Thanks,
>
> -Jason

so we have
1- u flag - in modflags, allows marking of sets
2- filterflags - constrain matching sites to those marked
    plus any other subcondition you might want on your marked set

3- negating flags

in filters, they allow complete match to all the bits.
dont want to touch callsites marked with Pt ?
(marked with t, without printing)  now you can.

filters let you use current flagstate to select subsets of " +p >control"
+negations gives complete selectivity on that flagstate

in modflags theyre more convenience,
but setting and clearing 2+ bits atomically
lets you both use the u bit to enable printing
(inc module name filename) and retire that use of u,
or to change the logging output subtly (by dropping threads display,
or module-name)

I cant say its essential, but its so cheap (last patch) once negating
flags are in place,
which are needed for full selectivity filtering.

thanks
Jim