From mboxrd@z Thu Jan  1 00:00:00 1970
From: arnd@arndb.de (Arnd Bergmann)
Date: Tue, 25 Aug 2015 22:36:22 +0200
Subject: [PATCH] ARM: stm32: Reduce FORCE_MAX_ZONEORDER to 9
In-Reply-To: <1440533237-5831-1-git-send-email-mcoquelin.stm32@gmail.com>
References: <1440533237-5831-1-git-send-email-mcoquelin.stm32@gmail.com>
Message-ID: <3551937.qa4lRhD8e3@wuerfel>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

On Tuesday 25 August 2015 22:07:17 Maxime Coquelin wrote:
> diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
> index a750c14..57d53af 100644
> --- a/arch/arm/Kconfig
> +++ b/arch/arm/Kconfig
> @@ -1734,7 +1734,7 @@ config FORCE_MAX_ZONEORDER
>         int "Maximum zone order" if ARCH_SHMOBILE_LEGACY
>         range 11 64 if ARCH_SHMOBILE_LEGACY
>         default "12" if SOC_AM33XX
> -       default "9" if SA1111 || ARCH_EFM32
> +       default "9" if SA1111 || ARCH_EFM32 || ARCH_STM32
>         default "11"
> 

This looks wrong now that we can enable multiple ARMV7M platforms
together. In practice it doesn't matter much, because you wouldn't
do that for a real-life system, but out of principle this seems
like something you'd better put into the defconfig file for
both ARCH_EFM32 and ARCH_STM32. We could also think about changing
the default to "9" for any ARM_SINGLE_ARMV7M config and increasing
it again in defconfig for any platform that actually has a lot of
RAM (if any).

The SOC_AM33XX is slightly different because it increases the default
rather than reducing it, presumably because the larger allocation is
actually required (for what?).

	Arnd