linux-kernel - Re: [PATCH] mm/slub: fix help comment of SLUB_DEBUG[

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: <b0ceb6a5-5949-2e0b-0634-e9d991c43188@suse.cz>
Date:   Mon, 13 Mar 2023 17:26:19 +0100
From:   Vlastimil Babka <vbabka@...e.cz>
To:     Hyeonggon Yoo <42.hyeyoo@...il.com>,
        Vernon Yang <vernon2gm@...il.com>
Cc:     akpm@...ux-foundation.org, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] mm/slub: fix help comment of SLUB_DEBUG[_ON]

On 2/23/23 02:03, Hyeonggon Yoo wrote:
> On Thu, Feb 23, 2023 at 07:20:58AM +0800, Vernon Yang wrote:
>> Since commit 081248de0a02 ("kset: move /sys/slab to /sys/kernel/slab"),
>> the SYSFS path of slab was modified, so fix up the help comment of
>> SLUB_DEBUG to use the correct one.
>> 
>> And now the "slub_debug" parameter on boot that is support for more
>> fine grained debug control, so fix up help comment of SLUB_DEBUG_ON
>> for "no support" to "support".
>> 
>> Signed-off-by: Vernon Yang <vernon2gm@...il.com>
>> ---
>>  mm/Kconfig.debug | 12 ++++++------
>>  1 file changed, 6 insertions(+), 6 deletions(-)
>> 
>> diff --git a/mm/Kconfig.debug b/mm/Kconfig.debug
>> index fca699ad1fb0..732043bf56c6 100644
>> --- a/mm/Kconfig.debug
>> +++ b/mm/Kconfig.debug
>> @@ -61,8 +61,8 @@ config SLUB_DEBUG
>>  	help
>>  	  SLUB has extensive debug support features. Disabling these can
>>  	  result in significant savings in code size. This also disables
>> -	  SLUB sysfs support. /sys/slab will not exist and there will be
>> -	  no support for cache validation etc.
>> +	  SLUB sysfs support. /sys/kernel/slab will not exist and there
>> +	  will be no support for cache validation etc.
> 
> After commit ab4d5ed5eeda ("slub: Enable sysfs support for !CONFIG_SLUB_DEBUG"),
> "Disabling SLUB_DEBUG also disables SLUB sysfs support" is not true anymore.

Yeah we could say that /sys/kernel/slab will exist but will not provide e.g.
cache validation.

>>  config SLUB_DEBUG_ON
>>  	bool "SLUB debugging on by default"
>> @@ -73,10 +73,10 @@ config SLUB_DEBUG_ON
>>  	  Boot with debugging on by default. SLUB boots by default with
>>  	  the runtime debug capabilities switched off. Enabling this is
>>  	  equivalent to specifying the "slub_debug" parameter on boot.
>> -	  There is no support for more fine grained debug control like
>> -	  possible with slub_debug=xxx. SLUB debugging may be switched
>> -	  off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying
>> -	  "slub_debug=-".
>> +	  There is support for more fine grained debug control like
>> +	  possible with slub_debug=xxx, details in Documentation/mm/slub.rst.
>> +	  SLUB debugging may be switched off in a kernel built with
>> +	  CONFIG_SLUB_DEBUG_ON by specifying "slub_debug=-".
> 
> I think this is a misunderstanding. SLUB_DEBUG_ON is equivalent to
> passing boot parameter slub_debug=FUPZ, and what the help text says is that
> 
> "You can enable slub debugging by default using this config option but
>  there is no support for fine-grained control like what slub_debug
>  boot parameter provides."
> 
>  IOW SLUB_DEBUG_ON cannot be used to have same effect as passing
>  something like slub_debug=F.

Agree with both comments. Vernon, will you send an updated patch?

>>  
>>  config PAGE_OWNER
>>  	bool "Track page owner"
>> -- 
>> 2.34.1
>> 
>>