linux-kernel - Re: [PATCH -next] optprobes: fix kconfig depends/selects

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:	Mon, 13 Sep 2010 19:14:13 +0900
From:	Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>
To:	Felipe Contreras <felipe.contreras@...il.com>
Cc:	Randy Dunlap <randy.dunlap@...cle.com>,
	lkml <linux-kernel@...r.kernel.org>, linux-next@...r.kernel.org,
	Ananth N Mavinakayanahalli <ananth@...ibm.com>,
	Anil S Keshavamurthy <anil.s.keshavamurthy@...el.com>,
	"David S. Miller" <davem@...emloft.net>,
	akpm <akpm@...ux-foundation.org>,
	2nddept-manager@....hitachi.co.jp
Subject: Re: [PATCH -next] optprobes: fix kconfig depends/selects

(2010/09/13 18:41), Felipe Contreras wrote:
> On Mon, Sep 13, 2010 at 10:15 AM, Masami Hiramatsu
> <masami.hiramatsu.pt@...achi.com> wrote:
>> (2010/09/13 10:06), Masami Hiramatsu wrote:
>>> (2010/09/12 19:49), Felipe Contreras wrote:
>>>> On Thu, Aug 12, 2010 at 10:26 PM, Randy Dunlap <randy.dunlap@...cle.com> wrote:
>>>>> From: Randy Dunlap <randy.dunlap@...cle.com>
>>>>>
>>>>> KALLSYMS and KALLSYMS_ALL depend on DEBUG_KERNEL, so add that dependency.
>>>>> KALLSYMS_ALL depends on KALLSYMS, so select both of them, not just KALLSYMS_ALL.
>>>>>
>>>>> warning: (OPTPROBES && KPROBES && HAVE_OPTPROBES && !PREEMPT || MTD_UBI_DEBUG && MTD && SYSFS && MTD_UBI || UBIFS_FS_DEBUG && MISC_FILESYSTEMS && UBIFS_FS || LOCKDEP && DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT || LATENCYTOP && HAVE_LATENCYTOP_SUPPORT) selects KALLSYMS_ALL which has unmet direct dependencies (DEBUG_KERNEL && KALLSYMS)
>>>>>
>>>>> Signed-off-by: Randy Dunlap <randy.dunlap@...cle.com>
>>>>> Cc:     Ananth N Mavinakayanahalli <ananth@...ibm.com>
>>>>> Cc:     Anil S Keshavamurthy <anil.s.keshavamurthy@...el.com>
>>>>> Cc:     "David S. Miller" <davem@...emloft.net>
>>>>> Cc:     Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>
>>>>> ---
>>>>>  arch/Kconfig |    2 ++
>>>>>  1 file changed, 2 insertions(+)
>>>>>
>>>>> --- linux-next-20100810.orig/arch/Kconfig
>>>>> +++ linux-next-20100810/arch/Kconfig
>>>>> @@ -45,6 +45,8 @@ config OPTPROBES
>>>>>        def_bool y
>>>>>        depends on KPROBES && HAVE_OPTPROBES
>>>>>        depends on !PREEMPT
>>>>> +       depends on DEBUG_KERNEL
>>>>> +       select KALLSYMS
>>>>>        select KALLSYMS_ALL
>>>>>
>>>>>  config HAVE_EFFICIENT_UNALIGNED_ACCESS
>>>>> --
>>>>
>>>> This causes another problem:
>>>>
>>>> arch/Kconfig:33:error: recursive dependency detected!
>>>> arch/Kconfig:33:     symbol KPROBES depends on KALLSYMS
>>>> init/Kconfig:825:    symbol KALLSYMS is selected by OPTPROBES
>>>> arch/Kconfig:44:     symbol OPTPROBES depends on KPROBES
>>>>
>>>> The 'select KALLSYMS' is redundant.
>>>>
>>>> Otherwise: Signed-off-by: Felipe Contreras <felipe.contreras@...il.com>
>>>>
>>>
>>> Thanks, let me clear the dependency;
>>> Kprobes uses kallsyms_lookup for finding target function and checking
>>> instruction boundary, thus CONFIG_KPROBES should select CONFIG_KALLSYMS
>>> and CONFIG_KALLSYMS_ALL.
>>> Optprobe is a optional feature which is supported on x86 arch, and
>>> it also uses kallsyms_lookup for checking instructions in the target
>>> function. Anyway, since optprobe itself is depends on kprobes, it doesn't
>>> need to select both of KALLSYMS/KALLSYMS_ALL.
>>
>> Hmm, wait, I might misunderstand something here. I thought
>> KALLSYMS_ALL added symbols of static functions.
>> But, does KALLSYMS_ALL just add symbols of variables?
>> If so, both of KPROBES and OPTPROBES require only KALLSYMS.
>> It means that they don't need to care about CONFIG_DEBUG_KERNEL.
> 
> So:
> -       select KALLSYMS_ALL
> +       depends on KALLSYMS_ALL
> 
> Because KALLSYMS_ALL depends on DEBUG_KERNEL && KALLSYMS.

No, what I meant is;
-       select KALLSYMS_ALL

:-)



-- 
Masami HIRAMATSU
2nd Dept. Linux Technology Center
Hitachi, Ltd., Systems Development Laboratory
E-mail: masami.hiramatsu.pt@...achi.com
--
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/