linux-kernel - Re: [RFC PATCH] mm/mmap: Fix uprobe anon page be overwritten when expanding vma during mremap

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: <20250530180920.GC25160@redhat.com>
Date: Fri, 30 May 2025 20:09:20 +0200
From: Oleg Nesterov <oleg@...hat.com>
To: David Hildenbrand <david@...hat.com>
Cc: Pu Lehui <pulehui@...weicloud.com>, lorenzo.stoakes@...cle.com,
	mhiramat@...nel.org, peterz@...radead.org, Liam.Howlett@...cle.com,
	akpm@...ux-foundation.org, vbabka@...e.cz, jannh@...gle.com,
	pfalcato@...e.de, linux-mm@...ck.org, linux-kernel@...r.kernel.org,
	pulehui@...wei.com
Subject: Re: [RFC PATCH] mm/mmap: Fix uprobe anon page be overwritten when
 expanding vma during mremap

Well, let me say this again ;) I can't really comment, I don't understand
this code enough.

That said...

On 05/30, David Hildenbrand wrote:
>
> I wonder if there might be a clean way to move the uprobe_mmap() out of
> vma_complete().

Me too.

Not only the uprobe_mmap() calls in vma_complete() doesn't look right
"in general" (at least to me).

To remind, vma_complete/uprobe_mmap/install_breakpoint is not even called
in, say, this case when VMA grows and moves. See
https://lore.kernel.org/all/20250526173845.GC4156@redhat.com/
I guess we don't really care, but still...

But just in case... I agree with Lehui and Lorenzo in that we need a short
term fix, and the last patch from Lehui seems to fix the immediate problem.

Oleg.