Date:   Tue, 24 Jan 2023 17:15:18 +0000
From:   Conor Dooley <conor@kernel.org>
To:     Pierre Gondois <pierre.gondois@arm.com>
Cc:     linux-kernel@vger.kernel.org,
        Geert Uytterhoeven <geert@linux-m68k.org>,
        Conor Dooley <conor.dooley@microchip.com>,
        Dan Carpenter <error27@gmail.com>,
        Catalin Marinas <catalin.marinas@arm.com>,
        Will Deacon <will@kernel.org>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        "Rafael J. Wysocki" <rafael@kernel.org>,
        Sudeep Holla <sudeep.holla@arm.com>,
        Akihiko Odaki <akihiko.odaki@daynix.com>,
        Palmer Dabbelt <palmer@rivosinc.com>,
        Gavin Shan <gshan@redhat.com>,
        linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH -next v2 2/3] cacheinfo: Make default
 acpi_get_cache_info() return an error
Message-ID: <Y9ASJiKjIhpMk/8n@spud>
References: <20230124154053.355376-1-pierre.gondois@arm.com>
 <20230124154053.355376-3-pierre.gondois@arm.com>
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha256;
        protocol="application/pgp-signature"; boundary="SHrM4lZTmcwll7Yv"
Content-Disposition: inline
In-Reply-To: <20230124154053.355376-3-pierre.gondois@arm.com>
Precedence: bulk


--SHrM4lZTmcwll7Yv
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Jan 24, 2023 at 04:40:47PM +0100, Pierre Gondois wrote:
> commit bd500361a937 ("ACPI: PPTT: Update acpi_find_last_cache_level()
> to acpi_get_cache_info()")
> updates the prototype of acpi_get_cache_info(). The cache 'levels'
> is update through a pointer and not the return value of the function.
>=20
> If CONFIG_ACPI_PPTT is not defined, acpi_get_cache_info() doesn't
> update its *levels and *split_levels parameters and returns 0.
> This can lead to a faulty behaviour.
>=20
> Make acpi_get_cache_info() return an error code if CONFIG_ACPI_PPTT
> is not defined.
> Also,
>=20
> In init_cache_level(), if no PPTT is present or CONFIG_ACPI_PPTT is
> not defined, instead of aborting if acpi_get_cache_info() returns an
> error code, just continue. This allows to try fetching the cache
> information from clidr_el1.

Again, dunno jack about clidr_el1. But the change here seems sane once
more? I was going to suggest throwing away ret entirely and just
initialising fw_level to zero, since the function inside
acpi_get_cache_level() that actually modifies it comes after the last
error return anyway - but that probably would just leave you exposed to
a change in the core code that doesn't also update the callsites.
Plus this looks more deliberate IMO. On that basis, LGTM.

> Signed-off-by: Pierre Gondois <pierre.gondois@arm.com>
> ---
>  arch/arm64/kernel/cacheinfo.c | 2 +-
>  include/linux/cacheinfo.h     | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
>=20
> diff --git a/arch/arm64/kernel/cacheinfo.c b/arch/arm64/kernel/cacheinfo.c
> index d3fe9542c370..bf348b8d321f 100644
> --- a/arch/arm64/kernel/cacheinfo.c
> +++ b/arch/arm64/kernel/cacheinfo.c
> @@ -59,7 +59,7 @@ int init_cache_level(unsigned int cpu)
>  	} else {
>  		ret =3D acpi_get_cache_info(cpu, &fw_level, NULL);
>  		if (ret < 0)
> -			return ret;
> +			fw_level =3D 0;
>  	}
> =20
>  	if (fw_level < 0)
> diff --git a/include/linux/cacheinfo.h b/include/linux/cacheinfo.h
> index dfef57077cd0..908e19d17f49 100644
> --- a/include/linux/cacheinfo.h
> +++ b/include/linux/cacheinfo.h
> @@ -100,7 +100,7 @@ static inline
>  int acpi_get_cache_info(unsigned int cpu,
>  			unsigned int *levels, unsigned int *split_levels)
>  {
> -	return 0;
> +	return -ENOENT;
>  }
>  #else
>  int acpi_get_cache_info(unsigned int cpu,
> --=20
> 2.25.1
>=20

--SHrM4lZTmcwll7Yv
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iHUEABYIAB0WIQRh246EGq/8RLhDjO14tDGHoIJi0gUCY9ASJgAKCRB4tDGHoIJi
0kCAAPwM3UDd3wrW0Ki90PSY0KOZP5nvSwQFHugxZ9qRKiq+rgEA8Ps9R0lLiliq
mzMeShNEiuc5jE9G3Fkdi+l/pLHCAgA=
=42pH
-----END PGP SIGNATURE-----

--SHrM4lZTmcwll7Yv--