linux-kernel - Re: [PATCH v5 2/7] x86/mm: Use a single cache hot per-CPU variable to record pgdir

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: <CAMj1kXE+2P6_y0SnmtmD=J42pe67itnr5jQs6NxjMTvV7HHp0A@mail.gmail.com>
Date: Tue, 20 May 2025 13:28:52 +0200
From: Ard Biesheuvel <ardb@...nel.org>
To: "Kirill A. Shutemov" <kirill@...temov.name>
Cc: Ard Biesheuvel <ardb+git@...gle.com>, linux-kernel@...r.kernel.org, x86@...nel.org, 
	Ingo Molnar <mingo@...nel.org>, Linus Torvalds <torvalds@...ux-foundation.org>, 
	Brian Gerst <brgerst@...il.com>, Borislav Petkov <bp@...en8.de>
Subject: Re: [PATCH v5 2/7] x86/mm: Use a single cache hot per-CPU variable to
 record pgdir_shift

On Tue, 20 May 2025 at 13:03, Kirill A. Shutemov <kirill@...temov.name> wrote:
>
> On Tue, May 20, 2025 at 12:41:41PM +0200, Ard Biesheuvel wrote:
> > From: Ard Biesheuvel <ardb@...nel.org>
> >
> > Shrink the global variable 'pgdir_shift' to a single byte, and move it
> > to the cache hot per-CPU variable section so that it can be accessed
> > virtually for free.
>
> Hm. I am not convinced about per-CPU being useful here. Read-only cache
> lines can be shared between cores anyway.
>
> Putting it to __ro_after_init should do the trick, no?
>

Are you saying we shouldn't optimize prematurely? You must be new here :-)

But seriously, whatever works is fine with me, as long as the
observable return value of pgtable_l5_enabled() never changes.