Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Wed, 7 Apr 2021 16:47:34 +0200
From:   Peter Zijlstra <peterz@infradead.org>
To:     Michel Lespinasse <michel@lespinasse.org>
Cc:     Linux-MM <linux-mm@kvack.org>,
        Laurent Dufour <ldufour@linux.ibm.com>,
        Michal Hocko <mhocko@suse.com>,
        Matthew Wilcox <willy@infradead.org>,
        Rik van Riel <riel@surriel.com>,
        Paul McKenney <paulmck@kernel.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Suren Baghdasaryan <surenb@google.com>,
        Joel Fernandes <joelaf@google.com>,
        Rom Lemarchand <romlem@google.com>,
        Linux-Kernel <linux-kernel@vger.kernel.org>
Subject: Re: [RFC PATCH 09/37] mm: add per-mm mmap sequence counter for
 speculative page fault handling.
Message-ID: <YG3GBhvVt2cQkZVz@hirez.programming.kicks-ass.net>
References: <20210407014502.24091-1-michel@lespinasse.org>
 <20210407014502.24091-10-michel@lespinasse.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20210407014502.24091-10-michel@lespinasse.org>
Precedence: bulk

On Tue, Apr 06, 2021 at 06:44:34PM -0700, Michel Lespinasse wrote:
> The counter's write side is hooked into the existing mmap locking API:
> mmap_write_lock() increments the counter to the next (odd) value, and
> mmap_write_unlock() increments it again to the next (even) value.
> 
> The counter's speculative read side is supposed to be used as follows:
> 
> seq = mmap_seq_read_start(mm);
> if (seq & 1)
> 	goto fail;
> .... speculative handling here ....
> if (!mmap_seq_read_check(mm, seq)
> 	goto fail;
> 
> This API guarantees that, if none of the "fail" tests abort
> speculative execution, the speculative code section did not run
> concurrently with any mmap writer.

So this is obviously safe, but it's also super excessive. Any change,
anywhere, will invalidate and abort a SPF.

Since you make a complete copy of the vma, you could memcmp it in its
entirety instead of this.