[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <49681805-b521-71c9-23c0-9bb35d25ba58@intel.com>
Date: Fri, 18 Jan 2019 07:20:43 -0800
From: Dave Hansen <dave.hansen@...el.com>
To: Yanmin Zhang <yanmin_zhang@...ux.intel.com>,
Dave Hansen <dave.hansen@...ux.intel.com>, dave@...1.net
Cc: dan.j.williams@...el.com, dave.jiang@...el.com, zwisler@...nel.org,
vishal.l.verma@...el.com, thomas.lendacky@....com,
akpm@...ux-foundation.org, mhocko@...e.com,
linux-nvdimm@...ts.01.org, linux-kernel@...r.kernel.org,
linux-mm@...ck.org, ying.huang@...el.com, fengguang.wu@...el.com,
bp@...e.de, bhelgaas@...gle.com, baiyaowei@...s.chinamobile.com,
tiwai@...e.de
Subject: Re: [PATCH 4/4] dax: "Hotplug" persistent memory for use like normal
RAM
On 1/17/19 11:47 PM, Yanmin Zhang wrote:
> a chance for kernel to allocate PMEM as DMA buffer.
> Some super speed devices like 10Giga NIC, USB (SSIC connecting modem),
> might not work well if DMA buffer is in PMEM as it's slower than DRAM.
>
> Should your patchset consider it?
No, I don't think so.
They can DMA to persistent memory whether this patch set exists or not.
So, if the hardware falls over, that's a separate problem.
If an app wants memory that performs in a particular way, then I would
suggest those app find the NUMA nodes on the system that match their
needs with these patches:
> http://lkml.kernel.org/r/20190116175804.30196-1-keith.busch@intel.com
and use the existing NUMA APIs to select that memory.
Powered by blists - more mailing lists