[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <alpine.DEB.2.10.1602221647520.4688@chino.kir.corp.google.com>
Date: Mon, 22 Feb 2016 16:50:09 -0800 (PST)
From: David Rientjes <rientjes@...gle.com>
To: Xishi Qiu <qiuxishi@...wei.com>
cc: "Figo.zhang" <figo1802@...il.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
arve@...roid.com, riandrews@...roid.com,
devel@...verdev.osuosl.org, zhong jiang <zhongjiang@...wei.com>,
LKML <linux-kernel@...r.kernel.org>,
Linux MM <linux-mm@...ck.org>
Subject: Re: [PATCH] mm: add MM_SWAPENTS and page table when calculate tasksize
in lowmem_scan()
On Thu, 18 Feb 2016, Xishi Qiu wrote:
> How about kill more processes at one time?
>
> Usually loading camera will alloc 300-500M memory immediately, so call lmk
> repeatedly is a waste of time.
>
> And can we reclaim memory at one time instead of reclaim-alloc-reclaim-alloc...
> in this situation? e.g. use try_to_free_pages(), set nr_to_reclaim=300M
>
I don't use the lmk myself and it's never been used on my phone, so I
can't speak for the usecase. However, killing more than one process at a
time is generally a bad idea because it can allow processes to deplete
memory reserves which may be small on these systems. The lmk cannot be
considered a hotpath, so waiting for a process to exit and free its memory
for a small amount of time is generally not a bad trade-off when the
alternative is to kill many processes (perhaps unnecessarily), open
yourself up to livelock, and free memory that is potentially unneeded.
Powered by blists - more mailing lists