linux-kernel - Re: [PATCH v2 1/3] mm,slub: Do not special case N_NORMAL nodes for slab

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: <0dbcbb75-9d5d-48ee-9bbb-02eacd856aaa@redhat.com>
Date: Wed, 30 Apr 2025 11:02:22 +0200
From: David Hildenbrand <david@...hat.com>
To: Vlastimil Babka <vbabka@...e.cz>, Oscar Salvador <osalvador@...e.de>
Cc: Harry Yoo <harry.yoo@...cle.com>,
 Andrew Morton <akpm@...ux-foundation.org>, linux-mm@...ck.org,
 linux-kernel@...r.kernel.org, Jonathan Cameron
 <Jonathan.Cameron@...wei.com>, linux-cxl@...r.kernel.org
Subject: Re: [PATCH v2 1/3] mm,slub: Do not special case N_NORMAL nodes for
 slab_nodes

On 30.04.25 10:57, Vlastimil Babka wrote:
> On 4/30/25 10:47, Oscar Salvador wrote:
>> On Tue, Apr 08, 2025 at 08:18:32PM +0200, David Hildenbrand wrote:
>>> We could by grabbing the mutex in MEM_GOING_ONLINE and putting it in
>>> MEM_CANCEL_ONLINE / MEM_ONLINE when a node is going online.
>>
>> Hi guys,
>>
>> After a few busy days I have been revisiting this.
>>
>> I checked the proposal of slab_mutex spanning GOING_ONLINE <-> ONLINE,
>> and I did not see any issue.
>>
>> The only concern I had is that we might be calling some slab function from
>> {online,offline}_pages() that also takes the mutex.
>> I am not aware of any though, and quickly checking did not reveal
>> anything either.
>>
>> If there is any will be quickly revealed though :-).
>>
>> So, unless there is an opposition, I can move forward and see how it
>> looks.
>>
>> Thoughts?
> 
> I feel a bit uneasy about it, while maintaining slab_nodes doesn't seem like
> a huge issue to me, so dunno. Maybe David has a better idea now :)

Yeah, let's avoid the slab_mutex change for now.

-- 
Cheers,

David / dhildenb