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] [day] [month] [year] [list]
Message-ID: <ZyS2fZeIuK2Zuu2l@collins>
Date: Fri, 1 Nov 2024 12:07:41 +0100
From: Paul Kocialkowski <paulk@...-base.io>
To: linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org
Cc: Russell King <linux@...linux.org.uk>, Ingo Molnar <mingo@...nel.org>,
	Arnd Bergmann <arnd@...db.de>, Maxime Ripard <mripard@...nel.org>,
	Paul Kocialkowski <contact@...lk.fr>
Subject: Re: [PATCH] ARM: topology: Allow missing CPU clock-frequency
 device-tree property

Hi!

Le Sun 29 Sep 24, 20:19, Paul Kocialkowski a écrit :
> Allow the fallback mechanism to continue by assuming the same nominal
> frequency for all CPU cores, while still benefiting from the static
> coefficient provided by the compatible-driven table entries.
> This is similar to what is done in the common arch topology code when
> it fails to find a clock to get the frequency from.

Any thoughts about this patch?

Thanks!

Paul

> The ranging mechanism (using the middle capacity) is unaffected by
> the use of a unit frequency and still returns values in the requested
> range.
> 
> Also add a comment to clarify what is going on.
> 
> Signed-off-by: Paul Kocialkowski <contact@...lk.fr>
> ---
>  arch/arm/kernel/topology.c | 22 ++++++++++++++++------
>  1 file changed, 16 insertions(+), 6 deletions(-)
> 
> diff --git a/arch/arm/kernel/topology.c b/arch/arm/kernel/topology.c
> index 2336ee2aa44a..0eb743c65166 100644
> --- a/arch/arm/kernel/topology.c
> +++ b/arch/arm/kernel/topology.c
> @@ -119,13 +119,23 @@ static void __init parse_dt_topology(void)
>  		if (cpu_eff->compatible == NULL)
>  			continue;
>  
> +		/*
> +		 * Use the legacy clock-frequency property (representing the
> +		 * maximum achievable clock frequency) as an efficiency
> +		 * coefficient (divided by 2^20, roughly 1 MHz) to the table
> +		 * value. If no such property is available, use the table value
> +		 * directly and assume all CPUs are running at the same
> +		 * nominal frequency.
> +		 *
> +		 * It is assumed that clock-frequency is either provided for all
> +		 * CPUs or for none of them.
> +		 */
>  		rate = of_get_property(cn, "clock-frequency", &len);
> -		if (!rate || len != 4) {
> -			pr_err("%pOF missing clock-frequency property\n", cn);
> -			continue;
> -		}
> -
> -		capacity = ((be32_to_cpup(rate)) >> 20) * cpu_eff->efficiency;
> +		if (rate && len == 4)
> +			capacity = ((be32_to_cpup(rate)) >> 20) *
> +				   cpu_eff->efficiency;
> +		else
> +			capacity = cpu_eff->efficiency;
>  
>  		/* Save min capacity of the system */
>  		if (capacity < min_capacity)
> -- 
> 2.46.2
> 
> 

-- 
Paul Kocialkowski,

Independent contractor - sys-base - https://www.sys-base.io/
Free software developer - https://www.paulk.fr/

Specialist in multimedia, graphics and embedded hardware support with Linux.

Download attachment "signature.asc" of type "application/pgp-signature" (834 bytes)

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ