linux-kernel - Re: [PATCH] coccinelle: api: add vma

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]

Message-ID: <CAM41TOuWHZpnZS8zbDeDnHr1+VZUusCBEiNAx4DHeppoFhT1LQ@mail.gmail.com>
Date:	Sun, 17 May 2015 19:20:28 +0300
From:	Dmitry Kalinkin <dmitry.kalinkin@...il.com>
To:	Julia Lawall <julia.lawall@...6.fr>
Cc:	linux-kernel@...r.kernel.org, cocci@...teme.lip6.fr,
	Gilles Muller <Gilles.Muller@...6.fr>,
	Nicolas Palix <nicolas.palix@...g.fr>,
	Michal Marek <mmarek@...e.cz>
Subject: Re: [PATCH] coccinelle: api: add vma_pages.cocci

On Sun, May 17, 2015 at 4:39 PM, Julia Lawall <julia.lawall@...6.fr> wrote:
>> +@...atch2 depends on !context && patch && !org@
>> +struct vm_area_struct *vma;
>> +@@
>> +
>> +- (vma->vm_end - vma->vm_start) >> PAGE_SHIFT
>> ++ vma_pages(vma)
>
> This rule is not needed.  An isomorphism will allow tha case with ()
> around the whole thing to match the case where those parentheses are
> absent.
So coccinelle is a real semantic tool and sees the difference between
(vma->vm_end - vma->vm_start) >> PAGE_SHIFT + 3
and
(vma->vm_end - vma->vm_start) >> PAGE_SHIFT & 3
Cool. Didn't realize that.

> It may be good to put below the keywords:
>
> // Comments:
> // Options: --all-includes
>
> This rule relies on type information, ie knowing whether something has
> type struct vm_area_struct *, so having more include files available may
> make that more apparent.  On the other hand, if you think the referenced
> thing will always be a local variable (as opposed to a structure field),
> then just using the current C file should be good enough, and it will be
> faster without the --all-includes option.
There probably won't be any benefit in doing that. The statement is long
enough already long enough as it is. So it is unlikely that any code like
(dev->priv->vma->vm_end@p - dev->priv->vma->vm_start) >> PAGE_SHIFT
is to appear.
I also just checked and saw that using --all-includes doesn't produce
additional triggers anywhere on the current code (except for false fire
on mm.h at vma_pages() itself).
>
> I tried the rule replacing struct vm_area_struct *vma; by expression vma
> and got a couple of results on other types.  One could wonder if that code
> could be improved in some way.
Indeed, there are such cases. I checked a couple manually and they turned
out to be some different third party structures having fields called vm_end
and vm_start used for vma-related accounting. These may or may not be a
justifiable uses. We could report them, but not propose an automatic patch

Kind regards,
Dmitry
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/