linux-kernel - Re: [PATCH 0/1] mm/fake-numa: allow later numa node hotplug

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <2f19432c-cc3a-4860-9aa3-e6bc1f60de3d@redhat.com>
Date: Tue, 7 Jan 2025 11:08:03 +0100
From: David Hildenbrand <david@...hat.com>
To: Bruno Faccini <bfaccini@...dia.com>, linux-kernel@...r.kernel.org
Cc: linux-mm@...ck.org, akpm@...ux-foundation.org, rppt@...nel.org,
 ziy@...dia.com, jhubbard@...dia.com, mrusiniak@...dia.com
Subject: Re: [PATCH 0/1] mm/fake-numa: allow later numa node hotplug

Hi,

> 
> With recent M.Rapoport set of fake-numa patches in mm-everything
> and using numa=fake=4 boot parameter :
> $ numactl --hardware
> available: 4 nodes (0-3)
> node 0 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
> 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
> 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
> 65 66 67 68 69 70 71
> node 0 size: 122518 MB
> node 0 free: 117141 MB
> node 1 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
> 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
> 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
> 65 66 67 68 69 70 71
> node 1 size: 219911 MB
> node 1 free: 219751 MB
> node 2 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
> 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
> 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
> 65 66 67 68 69 70 71
> node 2 size: 122599 MB
> node 2 free: 122541 MB
> node 3 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
> 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
> 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
> 65 66 67 68 69 70 71
> node 3 size: 122479 MB
> node 3 free: 122408 MB

Why are all CPUs indicated as belonging to all nodes? Is that expected 
or a BUG?

I would have thought, just like memory, that one resource only belongs 
to one NUMA node.


> 
> With recent M.Rapoport set of fake-numa patches in mm-everything,
> this patch on top, using numa=fake=4 boot parameter :
> # numactl —hardware
> available: 12 nodes (0-11)
> node 0 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
> 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
> 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
> 65 66 67 68 69 70 71
> node 0 size: 122518 MB
> node 0 free: 116429 MB
> node 1 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
> 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
> 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
> 65 66 67 68 69 70 71
> node 1 size: 122631 MB
> node 1 free: 122576 MB
> node 2 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
> 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
> 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
> 65 66 67 68 69 70 71
> node 2 size: 122599 MB
> node 2 free: 122544 MB
> node 3 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
> 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
> 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
> 65 66 67 68 69 70 71
> node 3 size: 122479 MB
> node 3 free: 122419 MB
> node 4 cpus:
> node 4 size: 97280 MB
> node 4 free: 97279 MB


^ Is this where your driver hotplugged a single node and hotplugged memory?


-- 
Cheers,

David / dhildenb