linux-kernel - Re: [PATCH v2 0/7] kdb: Refactor and fix bugs in kdb

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: <20240423111957.GD1567803@aspen.lan>
Date: Tue, 23 Apr 2024 12:19:57 +0100
From: Daniel Thompson <daniel.thompson@...aro.org>
To: Justin Stitt <justinstitt@...gle.com>
Cc: Jason Wessel <jason.wessel@...driver.com>,
	Douglas Anderson <dianders@...omium.org>,
	kgdb-bugreport@...ts.sourceforge.net, linux-kernel@...r.kernel.org,
	stable@...r.kernel.org
Subject: Re: [PATCH v2 0/7] kdb: Refactor and fix bugs in kdb_read()

On Mon, Apr 22, 2024 at 10:49:29PM +0000, Justin Stitt wrote:
> Hi,
>
> On Mon, Apr 22, 2024 at 05:35:53PM +0100, Daniel Thompson wrote:
> > Inspired by a patch from [Justin][1] I took a closer look at kdb_read().
> >
> > Despite Justin's patch being a (correct) one-line manipulation it was a
> > tough patch to review because the surrounding code was hard to read and
> > it looked like there were unfixed problems.
> >
> > This series isn't enough to make kdb_read() beautiful but it does make
> > it shorter, easier to reason about and fixes two buffer overflows and a
> > screen redraw problem!
> >
> > [1]: https://lore.kernel.org/all/20240403-strncpy-kernel-debug-kdb-kdb_io-c-v1-1-7f78a08e9ff4@google.com/
> >
> > Signed-off-by: Daniel Thompson <daniel.thompson@...aro.org>
>
> Seems to work nicely.
>
> There is some weird behavior which was present before your patch and is
> still present with it (let >< represent cursor position):
>
> [0]kdb> test_ap>< (now press TAB)
>
> [0]kdb> test_aperfmperf>< (so far so good, we got our autocomplete)
>
> [0]kdb> test_ap><erfmperf (now, let's move the cursor back and press TAB again)
>
> [0]kdb> test_aperfmperf><erfmperf
>
> This is because the autocomplete engine is not considering the
> characters after the cursor position. To be clear, this isn't really a
> bug but rather a decision to be made about which functionality is
> desired.
>
> For example, my shell (zsh) will just simply move the cursor back to
> the end of the complete match instead of re-writing stuff.

Interesting observation. I hadn't realized zsh does that. FWIW default
settings for both bash and gdb complete the same way as kdb. Overall
that makes me OK with the current kdb behaviour.

However I was curious about this and found "skip-completed-text" in
the GNU Readline documentation. I think that would give you zsh-like
completion in gdb if you ever want it!


> At any rate,
> Tested-by: Justin Stitt <justinstitt@...gle.com>

Thanks.


Daniel.