linux-kernel - Re: [PATCH] [perf] convert_variable_type does not correctly check type of arrays

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: <87bof9onxl.fsf@sejong.aot.lge.com>
Date:	Wed, 07 Nov 2012 16:46:30 +0900
From:	Namhyung Kim <namhyung@...nel.org>
To:	linux-kernel@...r.kernel.org,
	Hannes Frederic Sowa <hannes@...essinduktion.org>
Cc:	a.p.zijlstra@...llo.nl, paulus@...ba.org, mingo@...hat.com,
	acme@...stprotocols.net, hyc.lee@...il.com
Subject: Re: [PATCH] [perf] convert_variable_type does not correctly check type of arrays

Hi Hannes,

On Mon, 5 Nov 2012 23:49:16 +0100, Hannes Frederic Sowa wrote:
> While casting an array of (unsigned) chars to a string, perf does not
> check the containing type but only the opaque type and is bailing out:
>
>   $ perf probe -v -a 'neigh_destroy:22 dev->name:string'
>   probe-definition(0): neigh_destroy:22 dev->name:string
>   symbol:neigh_destroy file:(null) line:22 offset:0 return:0 lazy:(null)
>   parsing arg: dev->name:string into type:string dev, name(1)
>   1 arguments
>   Use vmlinux: /home/hannes/linux/vmlinux
>   Using /home/hannes/linux/vmlinux for symbols
>   Probe point found: neigh_destroy+115
>   Searching 'dev' variable in context.
>   Converting variable dev into trace event.
>   converting name in dev
>   name type is (null).
>   Failed to cast into string: name is not (unsigned) char *.
>   Failed to find 'dev' in this function.
>   An error occurred in debuginfo analysis (-22).
>     Error: Failed to add events. (-22)
>
> After the code flow ensures that type could only be a pointer or
> array type, call die_get_real_type unconditionally again to fetch the
> containing type and have further validation been done on that Die.

Hyeoncheol posted a same fix before, but not merged yet, Arnaldo?

  https://lkml.org/lkml/2012/9/20/3

Thanks,
Namhyung

--
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/