linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: David Hildenbrand <david@redhat.com>
To: Gregory Price <gourry@gourry.net>, lsf-pc@lists.linux-foundation.org
Cc: linux-mm@kvack.org, linux-cxl@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: Re: CXL Boot to Bash - Section 3: Memory (block) Hotplug
Date: Tue, 18 Feb 2025 17:24:30 +0100	[thread overview]
Message-ID: <2f759351-a01f-4b1a-bf37-793a42a67c69@redhat.com> (raw)
In-Reply-To: <Z7OWmDXEYhT0BB0X@gourry-fedora-PF4VCD3F>

> ---------------------
> Step 9: Memory Zones.
> ---------------------
> We've alluded to "Memory Zones" in prior sections, with really the only
> detail about these concepts being that there are "Kernel-allocation
> compatible" and "Movable" zones, as well as some relationship between
> memory blocks and memory zones.
> 
> The two zones we really care about are `ZONE_NORMAL` and `ZONE_MOVABLE`.
> 
> For the purpose of this reading we'll consider two basic use-cases:
> - memory block hot-unplug
> - kernel resource allocation
> 
> You can (for the most part) consider these cases incompatible.  If the
> kernel allocates `struct page` memory from a block, then that block cannot
> be hot-unplugged.  This memory is typically unmovable (cannot be migrated),
> and its pages unlikely to be removed from the memory map.
> 
> There are other scenarios, such as page pinning, that can block hot-unplug.
> The individual mechanisms preventing hot-unplug are less important than
> their relationship to memory zones.
> 
> ZONE_NORMAL basically allows any allocations, including things like page
> tables, struct pages, and pinned memory.
> 
> ZONE_MOVABLE, under normal conditions, disallows most kernel allocations.
> 

In essence, only movable allocations (some kernel allcoations are movable).

> ZONE_MOVABLE does NOT make a *strong* guarantee of hut-unplug-ability.
> The kernel and privileged users can cause long-term pinning to occur -
> even in ZONE_MOVABLE.  It should be seen as a best-attempt at providing
> hot-unplug-ability under normal conditions.

Yes and no; actual long-term pinning is disallowed (FOLL_LONGTERM), but 
we have a bunch of cases that need fixing. [1]

Of course, new cases keep popping up. It's a constant fight to make 
hot-unplug as reliable as possible. So yes, we cannot give "strong" 
guarantees, but make it as reliable as possible in sane configurations.

[1] 
https://lkml.kernel.org/r/882b566c-34d6-4e68-9447-6c74a0693f18@redhat.com


-- 
Cheers,

David / dhildenb



  reply	other threads:[~2025-02-18 16:24 UTC|newest]

Thread overview: 81+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-12-26 20:19 [LSF/MM] Linux management of volatile CXL memory devices - boot to bash Gregory Price
2025-02-05  2:17 ` [LSF/MM] CXL Boot to Bash - Section 1: BIOS, EFI, and Early Boot Gregory Price
2025-02-18 10:12   ` Yuquan Wang
2025-02-18 16:11     ` Gregory Price
2025-02-20 16:30   ` Jonathan Cameron
2025-02-20 16:52     ` Gregory Price
2025-03-04  0:32   ` Gregory Price
2025-03-13 16:12     ` Jonathan Cameron
2025-03-13 17:20       ` Gregory Price
2025-03-10 10:45   ` Yuquan Wang
2025-03-10 14:19     ` Gregory Price
2025-02-05 16:06 ` CXL Boot to Bash - Section 2: The Drivers Gregory Price
2025-02-06  0:47   ` Dan Williams
2025-02-06 15:59     ` Gregory Price
2025-03-04  1:32   ` Gregory Price
2025-03-06 23:56   ` CXL Boot to Bash - Section 2a (Drivers): CXL Decoder Programming Gregory Price
2025-03-07  0:57     ` Zhijian Li (Fujitsu)
2025-03-07 15:07       ` Gregory Price
2025-03-11  2:48         ` Zhijian Li (Fujitsu)
2025-04-02  6:45     ` Zhijian Li (Fujitsu)
2025-04-02 14:18       ` Gregory Price
2025-04-08  3:10         ` Zhijian Li (Fujitsu)
2025-04-08  4:14           ` Gregory Price
2025-04-08  5:37             ` Zhijian Li (Fujitsu)
2025-02-17 20:05 ` CXL Boot to Bash - Section 3: Memory (block) Hotplug Gregory Price
2025-02-18 16:24   ` David Hildenbrand [this message]
2025-02-18 17:03     ` Gregory Price
2025-02-18 17:49   ` Yang Shi
2025-02-18 18:04     ` Gregory Price
2025-02-18 19:25       ` David Hildenbrand
2025-02-18 20:25         ` Gregory Price
2025-02-18 20:57           ` David Hildenbrand
2025-02-19  1:10             ` Gregory Price
2025-02-19  8:53               ` David Hildenbrand
2025-02-19 16:14                 ` Gregory Price
2025-02-20 17:50             ` Yang Shi
2025-02-20 18:43               ` Gregory Price
2025-02-20 19:26                 ` David Hildenbrand
2025-02-20 19:35                   ` Gregory Price
2025-02-20 19:44                     ` David Hildenbrand
2025-02-20 20:06                       ` Gregory Price
2025-03-11 14:53                   ` Zi Yan
2025-03-11 15:58                     ` Gregory Price
2025-03-11 16:08                       ` Zi Yan
2025-03-11 16:15                         ` Gregory Price
2025-03-11 16:35                         ` Oscar Salvador
2025-03-05 22:20 ` [LSF/MM] CXL Boot to Bash - Section 0: ACPI and Linux Resources Gregory Price
2025-03-05 22:44   ` Dave Jiang
2025-03-05 23:34     ` Gregory Price
2025-03-05 23:41       ` Dave Jiang
2025-03-06  0:09         ` Gregory Price
2025-03-06  1:37   ` Yuquan Wang
2025-03-06 17:08     ` Gregory Price
2025-03-07  2:20       ` Yuquan Wang
2025-03-07 15:12         ` Gregory Price
2025-03-13 17:00           ` Jonathan Cameron
2025-03-08  3:23   ` [LSF/MM] CXL Boot to Bash - Section 0a: CFMWS and NUMA Flexiblity Gregory Price
2025-03-13 17:20     ` Jonathan Cameron
2025-03-13 18:17       ` Gregory Price
2025-03-14 11:09         ` Jonathan Cameron
2025-03-14 13:46           ` Gregory Price
2025-03-13 16:55   ` [LSF/MM] CXL Boot to Bash - Section 0: ACPI and Linux Resources Jonathan Cameron
2025-03-13 17:30     ` Gregory Price
2025-03-14 11:14       ` Jonathan Cameron
2025-03-27  9:34     ` Yuquan Wang
2025-03-27 12:36       ` Gregory Price
2025-03-27 13:21         ` Dan Williams
2025-03-27 16:36           ` Gregory Price
2025-03-31 23:49             ` [Lsf-pc] " Dan Williams
2025-03-12  0:09 ` [LSF/MM] CXL Boot to Bash - Section 4: Interleave Gregory Price
2025-03-13  8:31   ` Yuquan Wang
2025-03-13 16:48     ` Gregory Price
2025-03-26  9:28   ` Yuquan Wang
2025-03-26 12:53     ` Gregory Price
2025-03-27  2:20       ` Yuquan Wang
2025-03-27  2:51         ` [Lsf-pc] " Dan Williams
2025-03-27  6:29           ` Yuquan Wang
2025-03-14  3:21 ` [LSF/MM] CXL Boot to Bash - Section 6: Page allocation Gregory Price
2025-03-18 17:09 ` [LSFMM] Updated: Linux Management of Volatile CXL Memory Devices Gregory Price
2025-04-02  4:49   ` Gregory Price
     [not found]     ` <CGME20250407161445uscas1p19322b476cafd59f9d7d6e1877f3148b8@uscas1p1.samsung.com>
2025-04-07 16:14       ` Adam Manzanares

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=2f759351-a01f-4b1a-bf37-793a42a67c69@redhat.com \
    --to=david@redhat.com \
    --cc=gourry@gourry.net \
    --cc=linux-cxl@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=lsf-pc@lists.linux-foundation.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