linux-kernel - Re: [RFC PATCH 3/3] x86/boot: Use alternatives based selector for 5-level paging constants

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: <aBpJV7fJNyfb7tSx@gmail.com>
Date: Tue, 6 May 2025 19:39:35 +0200
From: Ingo Molnar <mingo@...nel.org>
To: Ard Biesheuvel <ardb@...nel.org>
Cc: Linus Torvalds <torvalds@...ux-foundation.org>,
	Ard Biesheuvel <ardb+git@...gle.com>, linux-kernel@...r.kernel.org,
	x86@...nel.org
Subject: Re: [RFC PATCH 3/3] x86/boot: Use alternatives based selector for
 5-level paging constants


* Ard Biesheuvel <ardb@...nel.org> wrote:

> > All of those change depending on
> > whether we do 5-level page tables or not, so the whole argument that
> > "pgtable_l5_enabled() is special" is just wrong to begin with.
> >
> 
> In my original patch, which is the one Ingo objected to, 
> pgtable_l5_enabled() is unambiguously based on whether CR4.LA57 is 
> set.

So I didn't really object to the simplification aspect - I was 
criticizing the current state of LA57 handling, regardless of your 
patch. In fact in that thread I supported the simplification aspect:

  > > Anyway, I'm not against Ard's simplification patch as a first step, and
  > > any optimizations can be layered on top of that.

But in hindsight I can see how my first reply came away as 
disagreement...

> In the light of the above, care to comment on the previous approach?
> 
> https://lore.kernel.org/all/20250504095230.2932860-28-ardb+git@google.com/
> 
> That also uses the ALTERNATIVE_TERNARY() so the CR4 access gets
> patched away, and I'm happy to take suggestions how to improve that.

I still think we should introduce a LA57_ENABLED synthethic cpufeature 
flag or so for the MM constants and all the late facilities, and go 
from there.

Thanks,

	Ingo