linux-kernel - Re: [PATCH 06/13] kdb: Remove "mdW" and "mdWcN" handling of "W" == 0

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]

Date: Tue, 18 Jun 2024 07:42:31 -0700
From: Doug Anderson <dianders@...omium.org>
To: Daniel Thompson <daniel.thompson@...aro.org>
Cc: kgdb-bugreport@...ts.sourceforge.net, 
	Christophe JAILLET <christophe.jaillet@...adoo.fr>, Jason Wessel <jason.wessel@...driver.com>, 
	Thorsten Blum <thorsten.blum@...lux.com>, Yuran Pereira <yuran.pereira@...mail.com>, 
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 06/13] kdb: Remove "mdW" and "mdWcN" handling of "W" == 0

Hi,

On Tue, Jun 18, 2024 at 4:38 AM Daniel Thompson
<daniel.thompson@...aro.org> wrote:
>
> On Mon, Jun 17, 2024 at 05:34:40PM -0700, Douglas Anderson wrote:
> > The "mdW" and "mdWcN" generally lets the user control more carefully
> > what word size we display memory in and exactly how many words should
> > be displayed. Specifically, "md4" says to display memory w/ 4
> > bytes-per word and "md4c6" says to display 6 words of memory w/
> > 4-bytes-per word.
> >
> > The kdb "md" implementation has a special rule for when "W" is 0. In
> > this case:
> > * If you run with "W" == 0 and you've never run a kdb "md" command
> >   this reboot then it will pick 4 bytes-per-word, ignoring the normal
> >   default from the environment.
> > * If you run with "W" == 0 and you've run a kdb "md" command this
> >   reboot then it will pick up the bytes per word of the last command.
> >
> > As an example:
> >   [1]kdb> md2 0xffffff80c8e2b280 1
> >   0xffffff80c8e2b280 0200 0000 0000 0000 e000 8235 0000 0000   ...
> >   [1]kdb> md0 0xffffff80c8e2b280 1
> >   0xffffff80c8e2b280 0200 0000 0000 0000 e000 8235 0000 0000   ...
> >   [1]kdb> md 0xffffff80c8e2b280 1
> >   0xffffff80c8e2b280 0000000000000200 000000008235e000   ...
> >   [1]kdb> md0 0xffffff80c8e2b280 1
> >   0xffffff80c8e2b280 0000000000000200 000000008235e000   ...
> >
> > This doesn't seem like particularly useful behavior and adds a bunch
> > of complexity to the arg parsing. Remove it.
> >
> > Signed-off-by: Douglas Anderson <dianders@...omium.org>
> > ---
> >
> >  kernel/debug/kdb/kdb_main.c | 5 -----
> >  1 file changed, 5 deletions(-)
> >
> > diff --git a/kernel/debug/kdb/kdb_main.c b/kernel/debug/kdb/kdb_main.c
> > index c013b014a7d3..700b4e355545 100644
> > --- a/kernel/debug/kdb/kdb_main.c
> > +++ b/kernel/debug/kdb/kdb_main.c
> > @@ -1611,11 +1611,6 @@ static int kdb_md(int argc, const char **argv)
> >
> >       if (isdigit(argv[0][2])) {
> >               bytesperword = (int)(argv[0][2] - '0');
> > -             if (bytesperword == 0) {
> > -                     bytesperword = last_bytesperword;
> > -                     if (bytesperword == 0)
> > -                             bytesperword = 4;
> > -             }
> >               last_bytesperword = bytesperword;
> >               repeat = mdcount * 16 / bytesperword;
>
> Isn't this now a divide-by-zero?

Dang, you're right. It goes away in a later patch, though, since we
stop re-calculating everything until the end when things are
validated. I'll plan to reorder this patch to be after the patch
("kdb: In kdb_md() make `repeat` and `mdcount` calculations more
obvious").