linux-kernel - Re: [PATCH 1/3] PM / Sleep: Make [un]lock_system

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: <4EDD0D35.1020907@linux.vnet.ibm.com>
Date:	Mon, 05 Dec 2011 23:58:05 +0530
From:	"Srivatsa S. Bhat" <srivatsa.bhat@...ux.vnet.ibm.com>
To:	Tejun Heo <tj@...nel.org>
CC:	rjw@...k.pl, pavel@....cz, len.brown@...el.com,
	ebiederm@...ssion.com, rdunlap@...otime.net,
	linux-doc@...r.kernel.org, kexec@...ts.infradead.org,
	linux-kernel@...r.kernel.org, linux-pm@...r.kernel.org,
	Oleg Nesterov <oleg@...hat.com>
Subject: Re: [PATCH 1/3] PM / Sleep: Make [un]lock_system_sleep() generic

On 12/05/2011 11:11 PM, Srivatsa S. Bhat wrote:

> On 12/05/2011 11:00 PM, Tejun Heo wrote:
> 
>> (cc'ing Oleg)
>>
>> On Mon, Dec 05, 2011 at 10:55:51PM +0530, Srivatsa S. Bhat wrote:
>>> I wanted these APIs to be generic, not restricted to work only for userspace
>>> processes. Both freezer_do_not_count() and freezer_count() are effective only
>>> when current->mm is non-NULL (ie., only for userspace ones).
>>> I think I have documented this in the patch which added these things to the
>>> 2 APIs. See commit 6a76b7a in linux-pm/linux-next.
>>
>> I see.  Oleg was curious about the ->mm condition too and IIRC there's
>> no reason for that restriction.  Maybe removing that in another patch
>> and using the count functions is better?
>>
> 
> 
> Oh well, then yes, that sounds like a better idea. Will send patches for
> that. Thank you.
> 


I looked up in git and found that commit ba96a0c by Rafael introduced the
count functions, to handle the vfork case. But now, we seem to have more
uses than that. So I think we can remove that userspace restriction in the
count functions, and in kernel/fork.c, do something like:

if (current->mm)
	freezer_do_not_count();
...
if (current->mm)
	freezer_count();

This way, it wouldn't break anything, since functionality-wise it is equivalent
to the existing code; And we get what we want: ability to directly use the
count functions elsewhere.

Regards,
Srivatsa S. Bhat

--
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/