linux-kernel - Re: [PATCH v9 11/17] mm: replace vm_lock and detached flag with a reference count

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

Message-ID: <20250115153844.GF8385@noisy.programming.kicks-ass.net>
Date: Wed, 15 Jan 2025 16:38:44 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: Suren Baghdasaryan <surenb@...gle.com>
Cc: Mateusz Guzik <mjguzik@...il.com>, akpm@...ux-foundation.org,
	willy@...radead.org, liam.howlett@...cle.com,
	lorenzo.stoakes@...cle.com, david.laight.linux@...il.com,
	mhocko@...e.com, vbabka@...e.cz, hannes@...xchg.org,
	oliver.sang@...el.com, mgorman@...hsingularity.net,
	david@...hat.com, peterx@...hat.com, oleg@...hat.com,
	dave@...olabs.net, paulmck@...nel.org, brauner@...nel.org,
	dhowells@...hat.com, hdanton@...a.com, hughd@...gle.com,
	lokeshgidra@...gle.com, minchan@...gle.com, jannh@...gle.com,
	shakeel.butt@...ux.dev, souravpanda@...gle.com,
	pasha.tatashin@...een.com, klarasmodin@...il.com,
	richard.weiyang@...il.com, corbet@....net,
	linux-doc@...r.kernel.org, linux-mm@...ck.org,
	linux-kernel@...r.kernel.org, kernel-team@...roid.com
Subject: Re: [PATCH v9 11/17] mm: replace vm_lock and detached flag with a
 reference count

On Wed, Jan 15, 2025 at 04:35:07PM +0100, Peter Zijlstra wrote:

> Consider:
> 
>     CPU0				CPU1
> 
>     rcu_read_lock();
>     vma = vma_lookup(mm, vaddr);
> 
>     ... cpu goes sleep for a *long time* ...
> 
>     					__vma_exit_locked();
> 					vma_area_free()
> 					..
> 					vma = vma_area_alloc();
> 					vma_mark_attached();
> 
>     ... comes back once vma is re-used ...
> 
>     vma_start_read()
>       vm_refcount_inc(); // success!!
> 
> At which point we need to validate vma is for mm and covers vaddr, which
> is what patch 15 does, no?

Also, critically, we want these reads to happen *after* the refcount
increment.