[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180706170617.GB4961@arm.com>
Date: Fri, 6 Jul 2018 18:06:17 +0100
From: Will Deacon <will.deacon@....com>
To: Nikunj Kela <nkela@...co.com>
Cc: Catalin Marinas <catalin.marinas@....com>,
xe-kernel@...ernal.cisco.com, linux-arm-kernel@...ts.infradead.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] arm64: add ARM64-specific support for flatmem
On Fri, Jul 06, 2018 at 10:02:15AM -0700, Nikunj Kela wrote:
> Flatmem is useful in reducing kernel memory usage.
> One usecase is in kdump kernel. We are able to save
> ~14M by moving to flatmem scheme.
>
> Cc: xe-kernel@...ernal.cisco.com
> Cc: Nikunj Kela <nkela@...co.com>
> Signed-off-by: Nikunj Kela <nkela@...co.com>
> ---
> arch/arm64/Kconfig | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> index 42c090c..96ab9a7 100644
> --- a/arch/arm64/Kconfig
> +++ b/arch/arm64/Kconfig
> @@ -775,6 +775,10 @@ config ARCH_SPARSEMEM_DEFAULT
> config ARCH_SELECT_MEMORY_MODEL
> def_bool ARCH_SPARSEMEM_ENABLE
>
> +config ARCH_FLATMEM_ENABLE
> + def_bool y
> + depends on !NUMA
I still don't understand what this has to do with NUMA. SPARSEMEM is still
going to be what you want on a non-NUMA system with large holes in the
physical memory map, no?
Will
Powered by blists - more mailing lists