linux-kernel - Re: [Linaro-mm-sig] [PATCHv17 0/11] Contiguous Memory Allocator

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:	Fri, 18 Nov 2011 15:30:49 -0800
From:	sandeep patil <psandeep.s@...il.com>
To:	Michal Nazarewicz <mina86@...a86.com>
Cc:	Marek Szyprowski <m.szyprowski@...sung.com>,
	linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
	linux-media@...r.kernel.org, linux-mm@...ck.org,
	linaro-mm-sig@...ts.linaro.org,
	Daniel Walker <dwalker@...eaurora.org>,
	Russell King <linux@....linux.org.uk>,
	Arnd Bergmann <arnd@...db.de>,
	Jonathan Corbet <corbet@....net>, Mel Gorman <mel@....ul.ie>,
	Dave Hansen <dave@...ux.vnet.ibm.com>,
	Jesse Barker <jesse.barker@...aro.org>,
	Kyungmin Park <kyungmin.park@...sung.com>,
	Ankita Garg <ankita@...ibm.com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>
Subject: Re: [Linaro-mm-sig] [PATCHv17 0/11] Contiguous Memory Allocator

2011/11/18 Michal Nazarewicz <mina86@...a86.com>:
> On Fri, 18 Nov 2011 22:20:48 +0100, sandeep patil <psandeep.s@...il.com>
> wrote:
>>
>> I am running a simple test to allocate contiguous regions and write a log
>> on
>> in a file on sdcard simultaneously. I can reproduce this migration failure
>> 100%
>> times with it.
>> when I tracked the pages that failed to migrate, I found them on the
>> buffer head lru
>> list with a reference held on the buffer_head in the page, which
>> causes drop_buffers()
>> to fail.
>>
>> So, i guess my question is, until all the migration failures are
>> tracked down and fixed,
>> is there a plan to retry the contiguous allocation from a new range in
>> the CMA region?
>
> No.  Current CMA implementation will stick to the same range of pages also
> on consequent allocations of the same size.
>

Doesn't that mean the drivers that fail to allocate from contiguous DMA region
will fail, if the migration fails?

~ sandeep
--
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/