linux-kernel - Re: [RFC][PATCH] SCHED

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: <4AC396CA.6070302@nortel.com>
Date:	Wed, 30 Sep 2009 11:35:06 -0600
From:	"Chris Friesen" <cfriesen@...tel.com>
To:	Raistlin <raistlin@...ux.it>
CC:	Henrik Austad <henrik@...tad.us>,
	Peter Zijlstra <peterz@...radead.org>, claudio@...dence.eu.com,
	michael@...dence.eu.com, mingo@...e.hu,
	linux-kernel@...r.kernel.org, tglx@...utronix.de,
	johan.eker@...csson.com, p.faure@...tech.ch,
	Fabio Checconi <fabio@...dalf.sssup.it>,
	Dhaval Giani <dhaval.giani@...il.com>,
	Steven Rostedt <rostedt@...dmis.org>,
	Tommaso Cucinotta <tommaso.cucinotta@...up.it>
Subject: Re: [RFC][PATCH] SCHED_EDF scheduling class

On 09/30/2009 09:58 AM, Raistlin wrote:
> On Tue, 2009-09-29 at 11:34 -0600, Chris Friesen wrote:

>> a) The child should get an identical bandwidth guarantee as the parent
>> and if that can't be guaranteed then the fork() should fail, maybe with
>> an errno of EBUSY.
>>
> Again, this could be done, pretty easily actually. :-)
> 
>> b) The child should start out with no guarantees (SCHED_OTHER nice 0
>> maybe?) and should have to request a bandwidth guarantee.  This could
>> complicate things in some circumstances because if it can't get the
>> guarantee then it needs to inform the parent somehow.
>>
> Ok, I see and agree, again, to many extents.

> Maybe, since I'm adding (in the next patch I'm going to send
> soon) a flag field in the sched_param_ex structure, we can also use some
> of the bits for deciding how the fork will behave... The main problem
> would be the code will get more complicated, and we thus would have to
> decide if it is worth...

For now it might be best to keep it simple...it can always be extended
later on.  Personally I prefer option "a" above as it makes applications
easier to code.

The only problem that I see is that it will refuse to fork() a task that
has a bandwidth of more than 50% of the system.  I wouldn't expect this
to be a common occurrence, but I could be wrong.

Chris
--
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/