[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <aSWr-frCzG8_z3S3@gourry-fedora-PF4VCD3F>
Date: Tue, 25 Nov 2025 08:15:37 -0500
From: Gregory Price <gourry@...rry.net>
To: "David Hildenbrand (Red Hat)" <david@...nel.org>
Cc: linux-mm@...ck.org, akpm@...ux-foundation.org, kernel-team@...a.com,
vbabka@...e.cz, surenb@...gle.com, mhocko@...e.com,
jackmanb@...gle.com, hannes@...xchg.org, ziy@...dia.com,
linux-kernel@...r.kernel.org, Mel Gorman <mgorman@...e.de>,
Alexandru Moise <00moses.alexander00@...il.com>,
David Rientjes <rientjes@...gle.com>
Subject: Re: [PATCH v2] mm, hugetlb: implement movable_gigantic_pages sysctl
On Tue, Nov 25, 2025 at 11:24:24AM +0100, David Hildenbrand (Red Hat) wrote:
> On 11/21/25 20:27, Gregory Price wrote:
>
> Can we also reference this parameter here from
> Documentation/admin-guide/mm/memory-hotplug.rst?
>
Good call
> In particular, there we also talk about gigantic pages and the interaction
> with ZONE_MOVABLE:
>
> "
> - Gigantic pages are unmovable, resulting in user space consuming a
> lot of unmovable memory.
>
> - Huge pages are unmovable when an architectures does not support huge
> page migration, resulting in a similar issue as with gigantic pages.
> "
>
> And then later, we talk about "Even with ZONE_MOVABLE, there are some corner
> cases where offlining a memory
> block might fail:" where this should be spelled out as well.
>
I will take a look, thank you!
> --
> Cheers
>
> David
Powered by blists - more mailing lists