Received-SPF: pass (google.com: domain of linux-kernel+bounces-96329-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
Date: Thu, 7 Mar 2024 22:30:12 +0000
In-Reply-To: <b7f0e16c-17e0-46ae-a777-7e4dca0bbdb3@intel.com>
Precedence: bulk
Mime-Version: 1.0
References: <20240307133916.3782068-1-yosryahmed@google.com>
 <20240307133916.3782068-3-yosryahmed@google.com> <tvbgstojgmb5jerhrdwbhknuhhbwmq2pqkb4lxyo3htdynioee@poos2yt52xg2>
 <034466a4-0917-47c4-934b-9549c3076624@intel.com> <Zeos1arCH11X_sXv@google.com>
 <b7f0e16c-17e0-46ae-a777-7e4dca0bbdb3@intel.com>
Message-ID: <Zeo_9IJgcVKYUzrm@google.com>
Subject: Re: [RFC PATCH 2/3] x86/mm: make sure LAM is up-to-date during
 context switching
From: Yosry Ahmed <yosryahmed@google.com>
To: Dave Hansen <dave.hansen@intel.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>, 
	Andrew Morton <akpm@linux-foundation.org>, Thomas Gleixner <tglx@linutronix.de>, 
	Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>, Peter Zijlstra <peterz@infradead.org>, 
	Andy Lutomirski <luto@kernel.org>, x86@kernel.org, linux-mm@kvack.org, 
	linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="us-ascii"

On Thu, Mar 07, 2024 at 01:48:28PM -0800, Dave Hansen wrote:
> On 3/7/24 13:08, Yosry Ahmed wrote:
> > CPU 1                                   CPU 2
> > /* user thread running */
> > context_switch() /* to kthread */
> >                                         /* user thread enables LAM */
> >                                         context_switch()
> > context_switch() /* to user thread */
> > 
> > In this case, there are no races, but the second context switch on CPU 1
> > may not write CR3 (if TLB is up-to-date), in which case we will run the
> > user thread with CR3 having the wrong LAM mask. This could cause bigger
> > problems, right?
> 
> Yes, but this is precisely the kind of thing that we should solve with
> mm generations.  Its sole purpose is to thwart optimization where the
> mm_prev==mm_next and might not rewrite CR3.

I think it's clearer to have an explicit check for oudated LAM in
switch_mm_irqs_off(), as I mentioned in my other reply.