linux-kernel - Re: [PATCH] riscv: Add riscv_force_qspinlock for early

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <330c7596-cd93-4ae5-b00d-789c72a944aa@ghiti.fr>
Date: Mon, 16 Dec 2024 10:14:57 +0100
From: Alexandre Ghiti <alex@...ti.fr>
To: guoren@...nel.org, conor@...nel.org, alexghiti@...osinc.com
Cc: linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org,
 paul.walmsley@...ive.com, palmer@...belt.com, bjorn@...osinc.com,
 leobras@...hat.com, peterz@...radead.org, parri.andrea@...il.com,
 longman@...hat.com, boqun.feng@...il.com, arnd@...db.de, will@...nel.org,
 corbet@....net, Guo Ren <guoren@...ux.alibaba.com>
Subject: Re: [PATCH] riscv: Add riscv_force_qspinlock for early_param

Hi Guo,

On 14/12/2024 05:35, guoren@...nel.org wrote:
> From: Guo Ren <guoren@...ux.alibaba.com>
>
> When CONFIG_RISCV_COMBO_SPINLOCKS is enabled, permit qspinlock
> force enabled. See the Kconfig entry for RISCV_COMBO_SPINLOCKS.
>
> Signed-off-by: Guo Ren <guoren@...ux.alibaba.com>
> Signed-off-by: Guo Ren <guoren@...nel.org>
> ---
>   Documentation/admin-guide/kernel-parameters.txt |  5 +++++
>   arch/riscv/kernel/setup.c                       | 13 ++++++++++++-
>   2 files changed, 17 insertions(+), 1 deletion(-)
>
> diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
> index 3872bc6ec49d..43d0df2922b2 100644
> --- a/Documentation/admin-guide/kernel-parameters.txt
> +++ b/Documentation/admin-guide/kernel-parameters.txt
> @@ -5887,6 +5887,11 @@
>   			[KNL] Disable ring 3 MONITOR/MWAIT feature on supported
>   			CPUs.
>   
> +	riscv_force_qspinlock [RISCV, EARLY]
> +			When CONFIG_RISCV_COMBO_SPINLOCKS is enabled, permit
> +			qspinlock force enabled. See the Kconfig entry for
> +			RISCV_COMBO_SPINLOCKS.
> +
>   	riscv_isa_fallback [RISCV,EARLY]
>   			When CONFIG_RISCV_ISA_FALLBACK is not enabled, permit
>   			falling back to detecting extension support by parsing
> diff --git a/arch/riscv/kernel/setup.c b/arch/riscv/kernel/setup.c
> index 45010e71df86..74b13bc64c9c 100644
> --- a/arch/riscv/kernel/setup.c
> +++ b/arch/riscv/kernel/setup.c
> @@ -247,6 +247,15 @@ static void __init parse_dtb(void)
>   #if defined(CONFIG_RISCV_COMBO_SPINLOCKS)
>   DEFINE_STATIC_KEY_TRUE(qspinlock_key);
>   EXPORT_SYMBOL(qspinlock_key);
> +
> +static bool force_qspinlock;
> +
> +static int __init riscv_force_qspinlock(char *p)
> +{
> +	force_qspinlock = true;
> +	return 0;
> +}
> +early_param("riscv_force_qspinlock", riscv_force_qspinlock);
>   #endif
>   
>   static void __init riscv_spinlock_init(void)
> @@ -267,7 +276,9 @@ static void __init riscv_spinlock_init(void)
>   		using_ext = "using Ziccrse";
>   	}
>   #if defined(CONFIG_RISCV_COMBO_SPINLOCKS)
> -	else {
> +	else if (force_qspinlock) {
> +		using_ext = "force";
> +	} else {
>   		static_branch_disable(&qspinlock_key);
>   		pr_info("Ticket spinlock: enabled\n");
>   		return;


What's the use case for this early param? To me that implies that a 
platform may have another extension which would allow the usage of 
qspinlock, so why not listing it in riscv_spinlock_init() instead?

Thanks,

Alex