linux-kernel - Re: [PATCH] coccinelle: semantic patch for missing put

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>] [day] [month] [year] [list]

Date:   Sat, 9 Feb 2019 20:10:53 +0100 (CET)
From:   Julia Lawall <julia.lawall@...6.fr>
To:     Wen Yang <yellowriver2010@...mail.com>
cc:     Gilles Muller <Gilles.Muller@...6.fr>,
        Nicolas Palix <nicolas.palix@...g.fr>,
        Michal Marek <michal.lkml@...kovi.net>,
        "cocci@...teme.lip6.fr" <cocci@...teme.lip6.fr>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] coccinelle: semantic patch for missing put_device()



On Sat, 9 Feb 2019, Wen Yang wrote:

> The of_find_device_by_node() takes a reference to the underlying device
> structure, we should release that reference.
> Use this semantic patch, we've found some object reference leaks,
> such as:
> 11907e9d3533 ("ASoC: fsl-asoc-card: fix object reference leaks in fsl_asoc_card_probe")
> a12085d13997 ("mtd: rawnand: atmel: fix possible object reference leak")
> 11493f26856a ("mtd: rawnand: jz4780: fix possible object reference leak")
> There are still dozens of reference leaks in the kernel code.
>
> Signed-off-by: Wen Yang <yellowriver2010@...mail.com>
> ---
>  scripts/coccinelle/free/put_device.cocci | 54 ++++++++++++++++++++++++++++++++
>  1 file changed, 54 insertions(+)
>  create mode 100644 scripts/coccinelle/free/put_device.cocci
>
> diff --git a/scripts/coccinelle/free/put_device.cocci b/scripts/coccinelle/free/put_device.cocci
> new file mode 100644
> index 0000000..d3e3ba9
> --- /dev/null
> +++ b/scripts/coccinelle/free/put_device.cocci
> @@ -0,0 +1,54 @@
> +/// Find missing put_device for every of_find_device_by_node.
> +///
> +// Confidence: Moderate
> +// Copyright: (C) 2018-2019 Wen Yang, ZTE.  GPLv2.
> +// Comments:
> +// Options: --no-includes --include-headers
> +
> +virtual report
> +virtual org
> +
> +@...rch@

Just put exists after search, and then drop the when exists below.

> +local idexpression id;
> +expression x,e;
> +position p1,p2;
> +type T;
> +@@
> +
> +id = of_find_device_by_node@p1(x)
> +... when != e = id

This should not use the same e as in the when's below.  Make a new
matavariavle e1 and use that here.

> +if (id == NULL || ...) { ... return ...; }
> +... when != put_device(&id->dev)

I have the impression that you should also consider platform_device_put.

> +    when != of_dev_put(&id)
> +    when != if (id) { ... put_device(&id->dev) ... }
> +    when != e = (T)id
> +    when != e = &id->dev
> +    when != e = gen_pool_get(&id->dev, ...)
> +    when != e = platform_get_drvdata(id)
> +    when exists
> +(
> +return id;
> +|
> +return dev_get_drvdata(&id->dev);
> +|
> +return platform_get_drvdata(id)

Make a new type metavariable T1 amd use it to put a cast on the result of
platform_get_drvdata.  For dev_get_drvdata, I didn't actually find any
examples with a cast, but it wouldn't hurt to put a cast there too

Thanks,
julia

> +|
> +return@p2 ...;
> +)
> +
> +@...ipt:python depends on report@
> +p1 << search.p1;
> +p2 << search.p2;
> +@@
> +
> +msg = "ERROR: missing put_device; of_find_device_by_node on line %s and return without releasing on line %s" % (p1[0].line,p2[0].line)
> +coccilib.report.print_report(p2[0],msg)
> +
> +@...ipt:python depends on org@
> +p1 << search.p1;
> +p2 << search.p2;
> +@@
> +
> +msg = "ERROR: missing put_device; of_find_device_by_node on line %s and return without releasing on line %s" % (p1[0].line,p2[0].line)
> +cocci.print_main(msg,p1)
> +cocci.print_secs("",p2)
> --
> 2.7.4
>
>