linux-kernel - Re: [PATCH 2/5] ARM: smp_scu: Add the enable speculative linefills operation

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]

Date:	Sat, 28 Jun 2014 17:02:16 +0200
From:	Thomas Petazzoni <thomas.petazzoni@...e-electrons.com>
To:	Gregory CLEMENT <gregory.clement@...e-electrons.com>
Cc:	Jason Cooper <jason@...edaemon.net>, Andrew Lunn <andrew@...n.ch>,
	Sebastian Hesselbarth <sebastian.hesselbarth@...il.com>,
	Russell King <linux@....linux.org.uk>,
	Shawn Guo <shawn.guo@...escale.com>,
	Sascha Hauer <kernel@...gutronix.de>,
	Lior Amsalem <alior@...vell.com>,
	Tawfik Bayouk <tawfik@...vell.com>,
	linux-kernel@...r.kernel.org, Nadav Haklai <nadavh@...vell.com>,
	Ezequiel Garcia <ezequiel.garcia@...e-electrons.com>,
	linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH 2/5] ARM: smp_scu: Add the enable speculative linefills
 operation

Dear Gregory CLEMENT,

On Fri, 27 Jun 2014 00:43:25 +0200, Gregory CLEMENT wrote:
> Quotin the ARM datasheet "When set, coherent linefill requests are

Quotin -> Quoting

> sent speculatively to the L2C-310 in parallel with the tag look-up. If
> the tag look-up misses, the confirmed linefill is sent to the L2C-310
> and gets RDATA earlier because the data request was already initiated
> by the speculative request. "
> 
> Some SoC (such as the Armada 375/38x) can benefit of this feature. As

benefit of -> benefit from.

>  #if defined(CONFIG_SMP) && defined(CONFIG_HAVE_ARM_SCU)
> diff --git a/arch/arm/kernel/smp_scu.c b/arch/arm/kernel/smp_scu.c
> index cfea41b41ad0..3fd21a495028 100644
> --- a/arch/arm/kernel/smp_scu.c
> +++ b/arch/arm/kernel/smp_scu.c
> @@ -18,6 +18,7 @@
>  
>  #define SCU_CTRL		0x00
>  #define SCU_CTRL_ENABLE		    BIT(1)
> +#define SCU_CTRL_SPEC_LINEFILLS	    BIT(3)
>  #define SCU_CONFIG		0x04
>  #define SCU_CPU_STATUS		0x08
>  #define SCU_INVALIDATE		0x0c
> @@ -88,3 +89,24 @@ int scu_power_mode(void __iomem *scu_base, unsigned int mode)
>  
>  	return 0;
>  }
> +
> +/*
> + * When enabled, coherent linefill requests are sent speculatively to
> + * the L2C-310 in parallel with the tag look-up
> + *
> + */
> +void scu_spec_linefills_enable(void __iomem *scu_base, bool enable)
> +{
> +	u32 scu_ctrl;
> +
> +	scu_ctrl = readl_relaxed(scu_base + SCU_CTRL);
> +	/* already enabled? */

Comment not needed, since SCU_CTRL_ENABLE already documents what's
happening. Or a more useful comment would be: "We cannot change the SCU
configuration while it is enabled".

> +	if (scu_ctrl & SCU_CTRL_ENABLE)
> +		return;

Return an error in this case maybe?

> +	if (enable)
> +		scu_ctrl |= SCU_CTRL_SPEC_LINEFILLS;
> +	else
> +		scu_ctrl &= ~SCU_CTRL_SPEC_LINEFILLS;
> +
> +	writel_relaxed(scu_ctrl, scu_base + SCU_CTRL);
> +}

Instead of having a separate function to do this (and the standby
operation), what about doing that directly in scu_enable() ? Either
unconditionally if that is fine for all SCU users, or through a flags
argument?

Best regards,

Thomas Petazzoni
-- 
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/