linux-kernel - Re: [PATCH 1/1] mlock: split the shmlock_user_lock spinlock into per user

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:	Thu, 10 May 2012 18:30:38 -0400
From:	Rik van Riel <riel@...hat.com>
To:	rajman mekaco <rajman.mekaco@...il.com>
CC:	Ingo Molnar <mingo@...hat.com>,
	Peter Zijlstra <peterz@...radead.org>,
	Paul Gortmaker <paul.gortmaker@...driver.com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>,
	Minchan Kim <minchan.kim@...il.com>,
	Christoph Lameter <cl@...two.org>,
	linux-kernel@...r.kernel.org, linux-mm@...ck.org
Subject: Re: [PATCH 1/1] mlock: split the shmlock_user_lock spinlock into
 per user_struct spinlock

On 05/10/2012 11:39 AM, rajman mekaco wrote:
> On Thu, May 10, 2012 at 8:24 PM, Rik van Riel<riel@...hat.com>  wrote:
>> On 05/10/2012 09:34 AM, rajman mekaco wrote:
>>
>>> Any updates on this ?
>>
>>
>> There is still no usecase to demonstrate a problem, so no real
>> justification to merge the patch.  Coming up with such a usecase
>> is up to the submitter of the patch.
>
> Maybe you didn't read my last email:
> If 2 different user-mode processes executing on 2 CPUs under 2 different
> users want to access the same shared memory through the
> shmctl(SHM_LOCK) / shmget(SHM_HUGETLB) / usr_shm_lock
> primitives, they could compete/spin even though their user_structs
> are different.
>
> Can you please correct me if I am missing some crucial point of understanding ?

Mlock is a very very expensive operation.

Updating the mlock statistics is a very cheap operation.

Does this spinlock ever show up contention wise?

-- 
All rights reversed
--
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/