Received-SPF: pass (google.com: domain of linux-kernel+bounces-40612-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Feedback-ID: i56a14606:Fastmail
User-Agent: Cyrus-JMAP/3.11.0-alpha0-119-ga8b98d1bd8-fm-20240108.001-ga8b98d1b
Precedence: bulk
MIME-Version: 1.0
Message-Id: <c54d0ede-4f41-4fcf-8fe7-d3f9e1bb63a4@app.fastmail.com>
In-Reply-To: <e523b29c-0fd0-4b7c-bf8c-d3424ee2c031@efficios.com>
References: <e523b29c-0fd0-4b7c-bf8c-d3424ee2c031@efficios.com>
Date: Fri, 26 Jan 2024 21:14:15 +0100
From: "Arnd Bergmann" <arnd@arndb.de>
To: "Mathieu Desnoyers" <mathieu.desnoyers@efficios.com>,
 "Dan Williams" <dan.j.williams@intel.com>,
 "Vishal Verma" <vishal.l.verma@intel.com>,
 "Dave Jiang" <dave.jiang@intel.com>, "Matthew Wilcox" <willy@infradead.org>,
 "Andrew Morton" <akpm@linux-foundation.org>,
 "Linus Torvalds" <torvalds@linux-foundation.org>,
 linux-mm <linux-mm@kvack.org>, Linux-Arch <linux-arch@vger.kernel.org>,
 nvdimm@lists.linux.dev, linux-cxl@vger.kernel.org,
 linux-kernel <linux-kernel@vger.kernel.org>,
 "Russell King" <linux@armlinux.org.uk>
Subject: Re: [REGRESSION] v5.13: FS_DAX unavailable on 32-bit ARM
Content-Type: text/plain

On Fri, Jan 26, 2024, at 20:33, Mathieu Desnoyers wrote:
>
> A) I have prepared a patch series introducing cache_is_aliasing() with 
> new Kconfig
>     options:
>
>    * ARCH_HAS_CACHE_ALIASING
>    * ARCH_HAS_CACHE_ALIASING_DYNAMIC
>
> and implemented it for all architectures. The "DYNAMIC" implementation
> implements cache_is_aliasing() as a runtime check, which is what is needed
> on architectures like 32-bit ARM.
>
> With this we can basically narrow down the list of architectures which are
> unsupported by DAX to those which are really affected, without actually solving
> the issue for architectures with virtually aliased dcaches.

The dynamic option should only be required when building for
ARMv6, which is really rare. On an ARMv7-only configuration,
we know that the dcache is non-aliasing, so the compile-time
check should be sufficient.

Even on ARMv6, this could be done as a compile-time choice
by platform, since we mostly know what the chips can do:
bcm2835, imx3, wm8750 and s3c64xx are non-aliasing because
they are limited to 16KB L1 caches, while omap2 and as2500
are aliasing with 32KB caches. With realview/integrator it
depends on the exact CPU that was installed.

     Arnd