Re: [PATCH v8 07/14] riscv: misaligned: use on_each_cpu() for scalar misaligned access probing

From: Charlie Jenkins <charlie@rivosinc.com>
To: "Clément Léger" <cleger@rivosinc.com>
Cc: Paul Walmsley <paul.walmsley@sifive.com>,
	Palmer Dabbelt <palmer@dabbelt.com>,
	Anup Patel <anup@brainfault.org>,
	Atish Patra <atishp@atishpatra.org>,
	Shuah Khan <shuah@kernel.org>, Jonathan Corbet <corbet@lwn.net>,
	linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org,
	linux-doc@vger.kernel.org, kvm@vger.kernel.org,
	kvm-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org,
	Samuel Holland <samuel.holland@sifive.com>,
	Andrew Jones <ajones@ventanamicro.com>,
	Deepak Gupta <debug@rivosinc.com>
Subject: Re: [PATCH v8 07/14] riscv: misaligned: use on_each_cpu() for scalar misaligned access probing
Date: Fri, 23 May 2025 11:37:14 -0700	[thread overview]
Message-ID: <aDDAWmNlr_mzrdLm@ghost> (raw)
In-Reply-To: <20250523101932.1594077-8-cleger@rivosinc.com>

On Fri, May 23, 2025 at 12:19:24PM +0200, Clément Léger wrote:
> schedule_on_each_cpu() was used without any good reason while documented
> as very slow. This call was in the boot path, so better use
> on_each_cpu() for scalar misaligned checking. Vector misaligned check
> still needs to use schedule_on_each_cpu() since it requires irqs to be
> enabled but that's less of a problem since this code is ran in a kthread.
> Add a comment to explicit that.
> 
> Signed-off-by: Clément Léger <cleger@rivosinc.com>
> Reviewed-by: Andrew Jones <ajones@ventanamicro.com>

Reviewed-by: Charlie Jenkins <charlie@rivosinc.com>
Tested-by: Charlie Jenkins <charlie@rivosinc.com>

> ---
>  arch/riscv/kernel/traps_misaligned.c | 8 ++++++--
>  1 file changed, 6 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/riscv/kernel/traps_misaligned.c b/arch/riscv/kernel/traps_misaligned.c
> index 592b1a28e897..34b4a4e9dfca 100644
> --- a/arch/riscv/kernel/traps_misaligned.c
> +++ b/arch/riscv/kernel/traps_misaligned.c
> @@ -627,6 +627,10 @@ bool __init check_vector_unaligned_access_emulated_all_cpus(void)
>  {
>  	int cpu;
>  
> +	/*
> +	 * While being documented as very slow, schedule_on_each_cpu() is used since
> +	 * kernel_vector_begin() expects irqs to be enabled or it will panic()
> +	 */
>  	schedule_on_each_cpu(check_vector_unaligned_access_emulated);
>  
>  	for_each_online_cpu(cpu)
> @@ -647,7 +651,7 @@ bool __init check_vector_unaligned_access_emulated_all_cpus(void)
>  
>  static bool unaligned_ctl __read_mostly;
>  
> -static void check_unaligned_access_emulated(struct work_struct *work __always_unused)
> +static void check_unaligned_access_emulated(void *arg __always_unused)
>  {
>  	int cpu = smp_processor_id();
>  	long *mas_ptr = per_cpu_ptr(&misaligned_access_speed, cpu);
> @@ -688,7 +692,7 @@ bool __init check_unaligned_access_emulated_all_cpus(void)
>  	 * accesses emulated since tasks requesting such control can run on any
>  	 * CPU.
>  	 */
> -	schedule_on_each_cpu(check_unaligned_access_emulated);
> +	on_each_cpu(check_unaligned_access_emulated, NULL, 1);
>  
>  	for_each_online_cpu(cpu)
>  		if (per_cpu(misaligned_access_speed, cpu)
> -- 
> 2.49.0
> 

-- 
kvm-riscv mailing list
kvm-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kvm-riscv