linux-kernel - Re: [PATCH 1/8] riscv: move riscv_noncoherent

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <4842439.0VBMTVartN@diego>
Date:   Fri, 07 Oct 2022 11:18:45 +0200
From:   Heiko Stübner <heiko@...ech.de>
To:     Paul Walmsley <paul.walmsley@...ive.com>,
        Palmer Dabbelt <palmer@...belt.com>,
        Albert Ou <aou@...s.berkeley.edu>,
        linux-riscv@...ts.infradead.org
Cc:     linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org,
        Jisheng Zhang <jszhang@...nel.org>
Subject: Re: [PATCH 1/8] riscv: move riscv_noncoherent_supported() out of ZICBOM probe

Hi,

Am Donnerstag, 6. Oktober 2022, 09:08:11 CEST schrieb Jisheng Zhang:
> It's a bit wired to call riscv_noncoherent_supported() once when
> insmod a module. Move the calling out of feature patch func.
> 
> Signed-off-by: Jisheng Zhang <jszhang@...nel.org>
> ---
>  arch/riscv/kernel/cpufeature.c | 7 +------
>  arch/riscv/kernel/setup.c      | 4 ++++
>  2 files changed, 5 insertions(+), 6 deletions(-)
> 
> diff --git a/arch/riscv/kernel/cpufeature.c b/arch/riscv/kernel/cpufeature.c
> index 3b5583db9d80..03611b3ef45e 100644
> --- a/arch/riscv/kernel/cpufeature.c
> +++ b/arch/riscv/kernel/cpufeature.c
> @@ -272,12 +272,7 @@ static bool __init_or_module cpufeature_probe_zicbom(unsigned int stage)
>  	case RISCV_ALTERNATIVES_EARLY_BOOT:
>  		return false;
>  	default:
> -		if (riscv_isa_extension_available(NULL, ZICBOM)) {
> -			riscv_noncoherent_supported();
> -			return true;
> -		} else {
> -			return false;
> -		}
> +		return riscv_isa_extension_available(NULL, ZICBOM);
>  	}
>  #endif
>  
> diff --git a/arch/riscv/kernel/setup.c b/arch/riscv/kernel/setup.c
> index 2dfc463b86bb..1a055c3f5d9d 100644
> --- a/arch/riscv/kernel/setup.c
> +++ b/arch/riscv/kernel/setup.c
> @@ -299,6 +299,10 @@ void __init setup_arch(char **cmdline_p)
>  	riscv_init_cbom_blocksize();
>  	riscv_fill_hwcap();
>  	apply_boot_alternatives();
> +#ifdef CONFIG_RISCV_DMA_NONCOHERENT
> +	if (riscv_isa_extension_available(NULL, ZICBOM))
> +		riscv_noncoherent_supported();
> +#endif

The nice thing about doing this in cpufeature_probe_zicbom is that
you keep all the "ifs" in one place, where now you have 2 places that
check the existence of the extension.

The overhead is one "x = true" setting on each call to _probe() and with
this change things are now also handled differently between the main
implementation and the deviants.

Though I guess, I'll let others do the judgement call on what is the
desired way to go ;-) .


Heiko