[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20141230050038.GD4588@js1304-P5Q-DELUXE>
Date: Tue, 30 Dec 2014 14:00:38 +0900
From: Joonsoo Kim <iamjoonsoo.kim@....com>
To: Hui Zhu <zhuhui@...omi.com>
Cc: m.szyprowski@...sung.com, mina86@...a86.com,
akpm@...ux-foundation.org, aneesh.kumar@...ux.vnet.ibm.com,
pintu.k@...sung.com, weijie.yang@...sung.com, mgorman@...e.de,
hannes@...xchg.org, riel@...hat.com, vbabka@...e.cz,
laurent.pinchart+renesas@...asonboard.com, rientjes@...gle.com,
sasha.levin@...cle.com, liuweixing@...omi.com,
linux-kernel@...r.kernel.org, linux-mm@...ck.org,
teawater@...il.com
Subject: Re: [PATCH 3/3] CMA: Add cma_alloc_counter to make cma_alloc work
better if it meet busy range
On Thu, Dec 25, 2014 at 05:43:28PM +0800, Hui Zhu wrote:
> In [1], Joonsoo said that cma_alloc_counter is useless because pageblock
> is isolated.
> But if alloc_contig_range meet a busy range, it will undo_isolate_page_range
> before goto try next range. At this time, __rmqueue_cma can begin allocd
> CMA memory from the range.
Is there any real issue from this?
When failed, we will quickly re-isolate pageblock for adjacent page
so there is no big problem I guess.
If there is real issue, how about doing start_isolation/undo_isolation
in cma_alloc()? It would reduce useless do/undo isolation due to
failed trial.
Thanks.
--
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/
Powered by blists - more mailing lists