Received-SPF: pass (google.com: domain of linux-kernel+bounces-185946-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Date: Wed, 22 May 2024 08:42:16 +0000
From: Brendan Jackman <jackmanb@google.com>
To: David Hildenbrand <david@redhat.com>,
	Oscar Salvador <osalvador@suse.de>,
	Andrew Morton <akpm@linux-foundation.org>,
	Mike Rapoport <rppt@kernel.org>
Cc: Michal Hocko <mhocko@suse.com>,
	Anshuman Khandual <anshuman.khandual@arm.com>,
	Vlastimil Babka <vbabka@suse.cz>,
	Pavel Tatashin <pasha.tatashin@soleen.com>, linux-mm@kvack.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH 2/2] mm,memory_hotplug: {READ,WRITE}_ONCE unsynchronized
 zone data
Message-ID: <Zk2v6LJPC0HKtBsO@google.com>
References: <20240521-mm-hotplug-sync-v1-0-6d53706c1ba8@google.com>
 <20240521-mm-hotplug-sync-v1-2-6d53706c1ba8@google.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20240521-mm-hotplug-sync-v1-2-6d53706c1ba8@google.com>

On Tue, May 21, 2024 at 12:57:19PM +0000, Brendan Jackman wrote:
> These fields are written by memory hotplug under mem_hotplug_lock but
> read without any lock. It seems like reader code is robust against the
> value being stale or "from the future", but we also need to account
> for:
> 
> 1. Load/store tearing (according to Linus[1], this really happens,
>    even when everything is aligned as you would hope).
> 
> 2. Invented loads[2] - the compiler can spill and re-read these fields
>    ([2] calls this "invented loads") and assume that they have not
>    changed.
> 
> Note we don't need READ_ONCE in paths that have the mem_hotplug_lock
> for write, but we still need WRITE_ONCE to prevent store-tearing.
> 
> [1] https://lore.kernel.org/all/CAHk-=wj2t+GK+DGQ7Xy6U7zMf72e7Jkxn4_-kGyfH3WFEoH+YQ@mail.gmail.com/T/#u
>     As discovered via the original big-bad article[2]
> [2] https://lwn.net/Articles/793253/
> 
> Signed-off-by: Brendan Jackman <jackmanb@google.com>

Oh, from a quick look it seems cma_pages would need this too.
present_early_pages seems fine.

I'll wait a few days in case anyone points out this whole thing is
garbage, then check more carefully and send a v2.