linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Dave Hansen <dave@sr71.net>
To: Jeff Moyer <jmoyer@redhat.com>,
	Dave Hansen <dave.hansen@linux.intel.com>
Cc: thomas.lendacky@amd.com, mhocko@suse.com,
	linux-nvdimm@lists.01.org, tiwai@suse.de, ying.huang@intel.com,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org, bp@suse.de,
	baiyaowei@cmss.chinamobile.com, zwisler@kernel.org,
	bhelgaas@google.com, fengguang.wu@intel.com,
	akpm@linux-foundation.org
Subject: Re: [PATCH 0/4] Allow persistent memory to be used like normal RAM
Date: Thu, 17 Jan 2019 14:43:56 -0800	[thread overview]
Message-ID: <360f77dc-fe8e-c7c4-84a0-852ef3c4a152@sr71.net> (raw)
In-Reply-To: <x49sgxr9rjd.fsf@segfault.boston.devel.redhat.com>

On 1/17/19 8:29 AM, Jeff Moyer wrote:
>> Persistent memory is cool.  But, currently, you have to rewrite
>> your applications to use it.  Wouldn't it be cool if you could
>> just have it show up in your system like normal RAM and get to
>> it like a slow blob of memory?  Well... have I got the patch
>> series for you!
> So, isn't that what memory mode is for?
>   https://itpeernetwork.intel.com/intel-optane-dc-persistent-memory-operating-modes/
> 
> Why do we need this code in the kernel?

So, my bad for not mentioning memory mode.  This patch set existed
before we could talk about it publicly, so it simply ignores its
existence.  It's a pretty glaring omissions at this point, sorry.

I'll add this to the patches, but here are a few reasons you might want
this instead of memory mode:
1. Memory mode is all-or-nothing.  Either 100% of your persistent memory
   is used for memory mode, or nothing is.  With this set, you can
   (theoretically) have very granular (128MB) assignment of PMEM to
   either volatile or persistent uses.  We have a few practical matters
   to fix to get us down to that 128MB value, but we can get there.
2. The capacity of memory mode is the size of your persistent memory.
   DRAM capacity is "lost" because it is used for cache.  With this,
   you get PMEM+DRAM capacity for memory.
3. DRAM acts as a cache with memory mode, and caches can lead to
   unpredictable latencies.  Since memory mode is all-or-nothing, your
   entire memory space is exposed to these unpredictable latencies.
   This solution lets you guarantee DRAM latencies if you need them.
4. The new "tier" of memory is exposed to software.  That means that you
   can build tiered applications or infrastructure.  A cloud provider
   could sell cheaper VMs that use more PMEM and more expensive ones
   that use DRAM.  That's impossible with memory mode.

Don't take this as criticism of memory mode.  Memory mode is awesome,
and doesn't strictly require *any* software changes (we have software
changes proposed for optimizing it though).  It has tons of other
advantages over *this* approach.  Basically, they are complementary
enough that we think both can live side-by-side.

      parent reply	other threads:[~2019-01-17 22:50 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-01-16 18:18 [PATCH 0/4] Allow persistent memory to be used like normal RAM Dave Hansen
2019-01-16 18:19 ` [PATCH 1/4] mm/resource: return real error codes from walk failures Dave Hansen
2019-01-16 20:38   ` Bjorn Helgaas
2019-01-16 18:19 ` [PATCH 2/4] mm/memory-hotplug: allow memory resources to be children Dave Hansen
2019-01-16 19:16   ` Jerome Glisse
2019-01-16 23:01     ` Dave Hansen
2019-01-16 23:38       ` Jerome Glisse
2019-01-23 20:03         ` Dave Hansen
2019-01-23 20:15           ` Jerome Glisse
2019-01-18 19:58     ` Dave Hansen
2019-01-18 20:26       ` Jerome Glisse
2019-01-23 17:05   ` Michal Hocko
2019-01-16 18:19 ` [PATCH 3/4] dax/kmem: let walk_system_ram_range() search child resources Dave Hansen
2019-01-16 18:19 ` [PATCH 4/4] dax: "Hotplug" persistent memory for use like normal RAM Dave Hansen
2019-01-16 21:16   ` Bjorn Helgaas
2019-01-16 21:40     ` Dave Hansen
2019-01-16 21:44       ` Bjorn Helgaas
2019-01-16 22:06       ` Dan Williams
2019-01-16 21:53     ` Dave Hansen
2019-01-16 21:59       ` Bjorn Helgaas
2019-01-16 21:31   ` Dan Williams
2019-01-17  5:21   ` Du, Fan
2019-01-17 16:56     ` Dan Williams
2019-01-17  8:19   ` Yanmin Zhang
2019-01-17 15:17     ` Dave Hansen
2019-01-18  7:47       ` Yanmin Zhang
2019-01-18 15:20         ` Dave Hansen
2019-01-17 16:29 ` [PATCH 0/4] Allow persistent memory to be used " Jeff Moyer
2019-01-17 16:47   ` Keith Busch
2019-01-17 17:20     ` Jeff Moyer
2019-01-17 19:34       ` Keith Busch
2019-01-17 21:57         ` Jeff Moyer
2019-01-18 11:48       ` Fengguang Wu
2019-01-17 16:50   ` Dan Williams
2019-01-17 22:43   ` Dave Hansen [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=360f77dc-fe8e-c7c4-84a0-852ef3c4a152@sr71.net \
    --to=dave@sr71.net \
    --cc=akpm@linux-foundation.org \
    --cc=baiyaowei@cmss.chinamobile.com \
    --cc=bhelgaas@google.com \
    --cc=bp@suse.de \
    --cc=dave.hansen@linux.intel.com \
    --cc=fengguang.wu@intel.com \
    --cc=jmoyer@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-nvdimm@lists.01.org \
    --cc=mhocko@suse.com \
    --cc=thomas.lendacky@amd.com \
    --cc=tiwai@suse.de \
    --cc=ying.huang@intel.com \
    --cc=zwisler@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).