public inbox for linux-arm-kernel@lists.infradead.org
 help / color / mirror / Atom feed
From: Jean-Philippe Brucker <jpb@kernel.org>
To: Mark Brown <broonie@kernel.org>
Cc: Marc Zyngier <maz@kernel.org>, Joey Gouly <joey.gouly@arm.com>,
	Catalin Marinas <catalin.marinas@arm.com>,
	Suzuki K Poulose <suzuki.poulose@arm.com>,
	Will Deacon <will@kernel.org>,
	Paolo Bonzini <pbonzini@redhat.com>,
	Jonathan Corbet <corbet@lwn.net>, Shuah Khan <shuah@kernel.org>,
	Oliver Upton <oupton@kernel.org>,
	Dave Martin <Dave.Martin@arm.com>, Fuad Tabba <tabba@google.com>,
	Mark Rutland <mark.rutland@arm.com>,
	Ben Horgan <ben.horgan@arm.com>,
	linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev,
	linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
	linux-doc@vger.kernel.org, linux-kselftest@vger.kernel.org,
	Peter Maydell <peter.maydell@linaro.org>,
	Eric Auger <eric.auger@redhat.com>
Subject: Re: [PATCH v10 04/30] arm64/fpsimd: Determine maximum virtualisable SME vector length
Date: Wed, 18 Mar 2026 17:29:33 +0000	[thread overview]
Message-ID: <20260318172933.GB2390801@myrica> (raw)
In-Reply-To: <20260306-kvm-arm64-sme-v10-4-43f7683a0fb7@kernel.org>

On Fri, Mar 06, 2026 at 05:00:56PM +0000, Mark Brown wrote:
> As with SVE we can only virtualise SME vector lengths that are supported by
> all CPUs in the system, implement similar checks to those for SVE. Since
> unlike SVE there are no specific vector lengths that are architecturally
> required the handling is subtly different, we report a system where this
> happens with a maximum vector length of SME_VQ_INVALID.
> 
> Signed-off-by: Mark Brown <broonie@kernel.org>
> ---
>  arch/arm64/include/asm/fpsimd.h |  2 ++
>  arch/arm64/kernel/fpsimd.c      | 21 ++++++++++++++++++++-
>  2 files changed, 22 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/arm64/include/asm/fpsimd.h b/arch/arm64/include/asm/fpsimd.h
> index e97729aa3b2f..0cd8a866e844 100644
> --- a/arch/arm64/include/asm/fpsimd.h
> +++ b/arch/arm64/include/asm/fpsimd.h
> @@ -69,6 +69,8 @@ static inline void cpacr_restore(unsigned long cpacr)
>  #define ARCH_SVE_VQ_MAX ((ZCR_ELx_LEN_MASK >> ZCR_ELx_LEN_SHIFT) + 1)
>  #define SME_VQ_MAX	((SMCR_ELx_LEN_MASK >> SMCR_ELx_LEN_SHIFT) + 1)
>  
> +#define SME_VQ_INVALID	(SME_VQ_MAX + 1)
> +
>  struct task_struct;
>  
>  extern void fpsimd_save_state(struct user_fpsimd_state *state);
> diff --git a/arch/arm64/kernel/fpsimd.c b/arch/arm64/kernel/fpsimd.c
> index 2af0e0c5b9f4..49c050ef6db9 100644
> --- a/arch/arm64/kernel/fpsimd.c
> +++ b/arch/arm64/kernel/fpsimd.c
> @@ -1218,7 +1218,8 @@ void cpu_enable_sme(const struct arm64_cpu_capabilities *__always_unused p)
>  void __init sme_setup(void)
>  {
>  	struct vl_info *info = &vl_info[ARM64_VEC_SME];
> -	int min_bit, max_bit;
> +	DECLARE_BITMAP(tmp_map, SVE_VQ_MAX);
> +	int min_bit, max_bit, b;
>  
>  	if (!system_supports_sme())
>  		return;
> @@ -1249,12 +1250,30 @@ void __init sme_setup(void)
>  	 */
>  	set_sme_default_vl(find_supported_vector_length(ARM64_VEC_SME, 32));
>  
> +	bitmap_andnot(tmp_map, info->vq_partial_map, info->vq_map,
> +		      SVE_VQ_MAX);
> +
> +	b = find_last_bit(tmp_map, SVE_VQ_MAX);
> +	if (b >= SVE_VQ_MAX)
> +		/* All VLs virtualisable */
> +		info->max_virtualisable_vl = sve_vl_from_vq(ARCH_SVE_VQ_MAX);
> +	else if (b == SVE_VQ_MAX - 1)
> +		/* No virtualisable VLs */
> +		info->max_virtualisable_vl = sve_vl_from_vq(SME_VQ_INVALID);
> +	else
> +		info->max_virtualisable_vl = sve_vl_from_vq(__bit_to_vq(b +  1));

nit: "b + 1"

Reviewed-by: Jean-Philippe Brucker <jpb@kernel.org>

> +
>  	pr_info("SME: minimum available vector length %u bytes per vector\n",
>  		info->min_vl);
>  	pr_info("SME: maximum available vector length %u bytes per vector\n",
>  		info->max_vl);
>  	pr_info("SME: default vector length %u bytes per vector\n",
>  		get_sme_default_vl());
> +
> +	/* KVM decides whether to support mismatched systems. Just warn here: */
> +	if (info->max_virtualisable_vl < info->max_vl ||
> +	    info->max_virtualisable_vl == sve_vl_from_vq(SME_VQ_INVALID))
> +		pr_warn("SME: unvirtualisable vector lengths present\n");
>  }
>  
>  void sme_suspend_exit(void)
> 
> -- 
> 2.47.3
> 
> 


  parent reply	other threads:[~2026-03-18 17:29 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-06 17:00 [PATCH v10 00/30] KVM: arm64: Implement support for SME Mark Brown
2026-03-06 17:00 ` [PATCH v10 01/30] arm64/sysreg: Update SMIDR_EL1 to DDI0601 2025-06 Mark Brown
2026-03-16 16:34   ` Catalin Marinas
2026-03-06 17:00 ` [PATCH v10 02/30] arm64/fpsimd: Update FA64 and ZT0 enables when loading SME state Mark Brown
2026-03-16 17:37   ` Catalin Marinas
2026-03-06 17:00 ` [PATCH v10 03/30] arm64/fpsimd: Decide to save ZT0 and streaming mode FFR at bind time Mark Brown
2026-03-16 17:42   ` Catalin Marinas
2026-03-06 17:00 ` [PATCH v10 04/30] arm64/fpsimd: Determine maximum virtualisable SME vector length Mark Brown
2026-03-16 17:44   ` Catalin Marinas
2026-03-18 17:29   ` Jean-Philippe Brucker [this message]
2026-03-06 17:00 ` [PATCH v10 05/30] KVM: arm64: Pay attention to FFR parameter in SVE save and load Mark Brown
2026-03-18 17:30   ` Jean-Philippe Brucker
2026-03-06 17:00 ` [PATCH v10 06/30] KVM: arm64: Pull ctxt_has_ helpers to start of sysreg-sr.h Mark Brown
2026-03-18 17:31   ` Jean-Philippe Brucker
2026-03-06 17:00 ` [PATCH v10 07/30] KVM: arm64: Move SVE state access macros after feature test macros Mark Brown
2026-03-18 17:32   ` Jean-Philippe Brucker
2026-03-06 17:01 ` [PATCH v10 08/30] KVM: arm64: Rename SVE finalization constants to be more general Mark Brown
2026-03-18 17:33   ` Jean-Philippe Brucker
2026-03-06 17:01 ` [PATCH v10 09/30] KVM: arm64: Define internal features for SME Mark Brown
2026-03-18 17:44   ` Jean-Philippe Brucker
2026-03-18 17:50     ` Mark Brown
2026-03-06 17:01 ` [PATCH v10 10/30] KVM: arm64: Rename sve_state_reg_region Mark Brown
2026-03-18 17:46   ` Jean-Philippe Brucker
2026-03-06 17:01 ` [PATCH v10 11/30] KVM: arm64: Store vector lengths in an array Mark Brown
2026-03-18 17:48   ` Jean-Philippe Brucker
2026-03-06 17:01 ` [PATCH v10 12/30] KVM: arm64: Factor SVE code out of fpsimd_lazy_switch_to_host() Mark Brown
2026-03-18 17:49   ` Jean-Philippe Brucker
2026-03-06 17:01 ` [PATCH v10 13/30] KVM: arm64: Document the KVM ABI for SME Mark Brown
2026-03-18 17:51   ` Jean-Philippe Brucker
2026-03-06 17:01 ` [PATCH v10 14/30] KVM: arm64: Implement SME vector length configuration Mark Brown
2026-03-18 17:53   ` Jean-Philippe Brucker
2026-03-06 17:01 ` [PATCH v10 15/30] KVM: arm64: Support SME control registers Mark Brown
2026-03-18 17:54   ` Jean-Philippe Brucker
2026-03-06 17:01 ` [PATCH v10 16/30] KVM: arm64: Support TPIDR2_EL0 Mark Brown
2026-03-18 17:55   ` Jean-Philippe Brucker
2026-03-06 17:01 ` [PATCH v10 17/30] KVM: arm64: Support SME identification registers for guests Mark Brown
2026-03-18 17:27   ` Jean-Philippe Brucker
2026-03-06 17:01 ` [PATCH v10 18/30] KVM: arm64: Support SME priority registers Mark Brown
2026-03-06 17:01 ` [PATCH v10 19/30] KVM: arm64: Provide assembly for SME register access Mark Brown
2026-03-06 17:01 ` [PATCH v10 20/30] KVM: arm64: Support userspace access to streaming mode Z and P registers Mark Brown
2026-03-06 17:01 ` [PATCH v10 21/30] KVM: arm64: Flush register state on writes to SVCR.SM and SVCR.ZA Mark Brown
2026-03-06 17:01 ` [PATCH v10 22/30] KVM: arm64: Expose SME specific state to userspace Mark Brown
2026-03-06 17:01 ` [PATCH v10 23/30] KVM: arm64: Context switch SME state for guests Mark Brown
2026-03-06 17:01 ` [PATCH v10 24/30] KVM: arm64: Handle SME exceptions Mark Brown
2026-03-06 17:01 ` [PATCH v10 25/30] KVM: arm64: Expose SME to nested guests Mark Brown
2026-03-06 17:01 ` [PATCH v10 26/30] KVM: arm64: Provide interface for configuring and enabling SME for guests Mark Brown
2026-03-06 17:01 ` [PATCH v10 27/30] KVM: arm64: selftests: Remove spurious check for single bit safe values Mark Brown
2026-03-06 17:01 ` [PATCH v10 28/30] KVM: arm64: selftests: Skip impossible invalid value tests Mark Brown
2026-03-24 14:54   ` Ben Horgan
2026-03-24 14:56     ` Mark Brown
2026-03-06 17:01 ` [PATCH v10 29/30] KVM: arm64: selftests: Add SME system registers to get-reg-list Mark Brown
2026-03-06 17:01 ` [PATCH v10 30/30] KVM: arm64: selftests: Add SME to set_id_regs test Mark Brown

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260318172933.GB2390801@myrica \
    --to=jpb@kernel.org \
    --cc=Dave.Martin@arm.com \
    --cc=ben.horgan@arm.com \
    --cc=broonie@kernel.org \
    --cc=catalin.marinas@arm.com \
    --cc=corbet@lwn.net \
    --cc=eric.auger@redhat.com \
    --cc=joey.gouly@arm.com \
    --cc=kvm@vger.kernel.org \
    --cc=kvmarm@lists.linux.dev \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-kselftest@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=maz@kernel.org \
    --cc=oupton@kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=shuah@kernel.org \
    --cc=suzuki.poulose@arm.com \
    --cc=tabba@google.com \
    --cc=will@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox