linux-kernel - Re: [PATCH] sched/fair: Replace zero-length array with flexible-array member

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:   Thu, 13 Feb 2020 18:05:13 -0800
From:   Joe Perches <joe@...ches.com>
To:     Valentin Schneider <valentin.schneider@....com>,
        Peter Zijlstra <peterz@...radead.org>,
        "Gustavo A. R. Silva" <gustavo@...eddedor.com>
Cc:     Ingo Molnar <mingo@...hat.com>, Juri Lelli <juri.lelli@...hat.com>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Dietmar Eggemann <dietmar.eggemann@....com>,
        Steven Rostedt <rostedt@...dmis.org>,
        Ben Segall <bsegall@...gle.com>, Mel Gorman <mgorman@...e.de>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] sched/fair: Replace zero-length array with
 flexible-array member

On Fri, 2020-02-14 at 00:25 +0000, Valentin Schneider wrote:
> On 13/02/2020 22:02, Joe Perches wrote:
> > That might be a somewhat difficult thing to add to checkpatch
> > as it is effectively a per-line scanner:
> > 
> > Try something like:
> > 
> > $ git grep -P -A1 '^\s*(?!return)(\w+\s+){1,3}\w+\[0\];' -- '*.[ch]'
> > 
> > and look at the results.
> > 
> > In checkpatch that could be something like:
> > 
> > 	if ($line =~ /^.\s*$Type\s+$Ident\s*\[\s*0\s*\]\s*;/) {
> > 		warn...
> > 	}
> > 
> 
> So FWIW I felt like doing some coccinelle and ended up with this:
> 
> This patches up valid ZLAs:
>   $ spatch -D patch zero_length_array.cocci kernel/sched/fair.c
> 
> This prints out the location of invalid ZLAs:
>   $ spatch -D report zero_length_array.cocci kernel/sched/fair.c
> 
> ---
> virtual patch
> virtual report
> 
> @valid_zla depends on patch@
> identifier struct_name;
> type T;
> identifier zla;
> position pos;
> @@
> struct struct_name {
>        ...
>        T zla@pos
> - [0];
> + [];
> };
> 
> @invalid_zla depends on report@
> identifier struct_name;
> type T1;
> identifier zla;
> type T2;
> identifier tail;
> position pos;
> @@
> struct struct_name {
>        ...
>        T1 zla[0]@pos;
>        T2 tail;
>        ...
> };
> 
> @script:python depends on invalid_zla@
> pos << invalid_zla.pos;
> @@
> coccilib.report.print_report(pos[0], "Invalid ZLA!");
> ---

Nice.
It would miss a few forms like:

	typedef struct tagfoo {
		...
		type t[0];
	} foo;

and

	struct {
		...
		type t[0];
	} foo;

and

	struct foo {
		...
		type t[0];
	} *foo;

etc...