Received-SPF: pass (google.com: domain of linux-kernel+bounces-167933-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Date: Fri, 3 May 2024 17:59:33 +0100
From: Conor Dooley <conor@kernel.org>
To: Charlie Jenkins <charlie@rivosinc.com>
Cc: Rob Herring <robh@kernel.org>,
	Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
	Paul Walmsley <paul.walmsley@sifive.com>,
	Palmer Dabbelt <palmer@dabbelt.com>,
	Albert Ou <aou@eecs.berkeley.edu>, Guo Ren <guoren@kernel.org>,
	Conor Dooley <conor+dt@kernel.org>, Chen-Yu Tsai <wens@csie.org>,
	Jernej Skrabec <jernej.skrabec@gmail.com>,
	Samuel Holland <samuel@sholland.org>,
	Conor Dooley <conor.dooley@microchip.com>,
	Evan Green <evan@rivosinc.com>,
	=?iso-8859-1?Q?Cl=E9ment_L=E9ger?= <cleger@rivosinc.com>,
	Jonathan Corbet <corbet@lwn.net>, Shuah Khan <shuah@kernel.org>,
	linux-riscv@lists.infradead.org, devicetree@vger.kernel.org,
	linux-kernel@vger.kernel.org, Palmer Dabbelt <palmer@rivosinc.com>,
	linux-arm-kernel@lists.infradead.org, linux-sunxi@lists.linux.dev,
	linux-doc@vger.kernel.org, linux-kselftest@vger.kernel.org
Subject: Re: [PATCH v5 03/17] riscv: vector: Use vlenb from DT
Message-ID: <20240503-zippy-skeletal-e5f63c9f17c1@spud>
References: <20240502-dev-charlie-support_thead_vector_6_9-v5-0-d1b5c013a966@rivosinc.com>
 <20240502-dev-charlie-support_thead_vector_6_9-v5-3-d1b5c013a966@rivosinc.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha256;
	protocol="application/pgp-signature"; boundary="Jm1skXhSfPF6Cfdj"
Content-Disposition: inline
In-Reply-To: <20240502-dev-charlie-support_thead_vector_6_9-v5-3-d1b5c013a966@rivosinc.com>


--Jm1skXhSfPF6Cfdj
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Thu, May 02, 2024 at 09:46:38PM -0700, Charlie Jenkins wrote:
> If vlenb is provided in the device tree, prefer that over reading the
> vlenb csr.
>=20
> Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
> ---
>  arch/riscv/include/asm/cpufeature.h |  2 ++
>  arch/riscv/kernel/cpufeature.c      | 43 +++++++++++++++++++++++++++++++=
++++++
>  arch/riscv/kernel/vector.c          | 12 ++++++++++-
>  3 files changed, 56 insertions(+), 1 deletion(-)
>=20
> diff --git a/arch/riscv/include/asm/cpufeature.h b/arch/riscv/include/asm=
/cpufeature.h
> index 347805446151..0c4f08577015 100644
> --- a/arch/riscv/include/asm/cpufeature.h
> +++ b/arch/riscv/include/asm/cpufeature.h
> @@ -31,6 +31,8 @@ DECLARE_PER_CPU(struct riscv_cpuinfo, riscv_cpuinfo);
>  /* Per-cpu ISA extensions. */
>  extern struct riscv_isainfo hart_isa[NR_CPUS];
> =20
> +extern u32 riscv_vlenb_of;
> +
>  void riscv_user_isa_enable(void);
> =20
>  #if defined(CONFIG_RISCV_MISALIGNED)
> diff --git a/arch/riscv/kernel/cpufeature.c b/arch/riscv/kernel/cpufeatur=
e.c
> index 3ed2359eae35..12c79db0b0bb 100644
> --- a/arch/riscv/kernel/cpufeature.c
> +++ b/arch/riscv/kernel/cpufeature.c
> @@ -35,6 +35,8 @@ static DECLARE_BITMAP(riscv_isa, RISCV_ISA_EXT_MAX) __r=
ead_mostly;
>  /* Per-cpu ISA extensions. */
>  struct riscv_isainfo hart_isa[NR_CPUS];
> =20
> +u32 riscv_vlenb_of;
> +
>  /**
>   * riscv_isa_extension_base() - Get base extension word
>   *
> @@ -648,6 +650,42 @@ static int __init riscv_isa_fallback_setup(char *__u=
nused)
>  early_param("riscv_isa_fallback", riscv_isa_fallback_setup);
>  #endif
> =20
> +static int has_riscv_homogeneous_vlenb(void)
> +{
> +	int cpu;
> +	u32 prev_vlenb =3D 0;
> +	u32 vlenb;
> +
> +	for_each_possible_cpu(cpu) {
> +		struct device_node *cpu_node;
> +
> +		cpu_node =3D of_cpu_device_node_get(cpu);
> +		if (!cpu_node) {
> +			pr_warn("Unable to find cpu node\n");
> +			return -ENOENT;
> +		}
> +
> +		if (of_property_read_u32(cpu_node, "riscv,vlenb", &vlenb)) {
> +			of_node_put(cpu_node);
> +
> +			if (prev_vlenb)
> +				return -ENOENT;
> +			continue;
> +		}
> +
> +		if (prev_vlenb && vlenb !=3D prev_vlenb) {
> +			of_node_put(cpu_node);
> +			return -ENOENT;
> +		}
> +
> +		prev_vlenb =3D vlenb;
> +		of_node_put(cpu_node);
> +	}
> +
> +	riscv_vlenb_of =3D vlenb;
> +	return 0;
> +}
> +
>  void __init riscv_fill_hwcap(void)
>  {
>  	char print_str[NUM_ALPHA_EXTS + 1];
> @@ -671,6 +709,11 @@ void __init riscv_fill_hwcap(void)
>  			pr_info("Falling back to deprecated \"riscv,isa\"\n");
>  			riscv_fill_hwcap_from_isa_string(isa2hwcap);
>  		}
> +
> +		if (elf_hwcap & COMPAT_HWCAP_ISA_V && has_riscv_homogeneous_vlenb() < =
0) {

I still think this isn't quite right, as it will emit a warning when
RISCV_ISA_V is disabled. The simplest thing to do probably is just
add an `if (IS_ENABLED(CONFIG_RISCV_ISA_V) return 0` shortcut the to
function? It'll get disabled a few lines later so I think a zero is
safe.

> +			pr_warn("Unsupported heterogeneous vlen detected, vector extension di=
sabled.\n");
> +			elf_hwcap &=3D ~COMPAT_HWCAP_ISA_V;
> +		}
>  	}
> =20
>  	/*
> diff --git a/arch/riscv/kernel/vector.c b/arch/riscv/kernel/vector.c
> index 6727d1d3b8f2..e04586cdb7f0 100644
> --- a/arch/riscv/kernel/vector.c
> +++ b/arch/riscv/kernel/vector.c
> @@ -33,7 +33,17 @@ int riscv_v_setup_vsize(void)
>  {
>  	unsigned long this_vsize;
> =20
> -	/* There are 32 vector registers with vlenb length. */
> +	/*
> +	 * There are 32 vector registers with vlenb length.
> +	 *
> +	 * If the riscv,vlenb property was provided by the firmware, use that
> +	 * instead of probing the CSRs.
> +	 */
> +	if (riscv_vlenb_of) {
> +		this_vsize =3D riscv_vlenb_of * 32;
> +		return 0;
> +	}
> +
>  	riscv_v_enable();
>  	this_vsize =3D csr_read(CSR_VLENB) * 32;
>  	riscv_v_disable();
>=20
> --=20
> 2.44.0
>=20

--Jm1skXhSfPF6Cfdj
Content-Type: application/pgp-signature; name="signature.asc"

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

iHUEABYIAB0WIQRh246EGq/8RLhDjO14tDGHoIJi0gUCZjUX9QAKCRB4tDGHoIJi
0rXxAQDDUnlsnbkBc4xorvMXBorwh8UuxTwps60RTP8U5kKaKwD/RTwTniatfEGc
LN74K3N4cTgYcOwsuNu6rNRBb5lk5Ao=
=d5rs
-----END PGP SIGNATURE-----

--Jm1skXhSfPF6Cfdj--