Re: [PATCH 4/7] [RFC] enable L1_WRITETHROUGH mode for BG/P
From: Eric Van Hensbergen <hidden>
Date: 2011-05-19 12:53:37
Also in:
lkml
On Thu, May 19, 2011 at 5:43 AM, Josh Boyer [off-list ref] wr= ote:
On Wed, May 18, 2011 at 04:24:52PM -0500, Eric Van Hensbergen wrote:quoted
+config L1_WRITETHROUGH + =A0 =A0 =A0bool "Blue Gene/P enabled writethrough mode" + =A0 =A0 =A0depends on BGP + =A0 =A0 =A0default yYou add this config option here, named generically, but then make it depend on BGP. =A0It sees it should be named BGP_L1_WRITETHROUGH, and the=
n
just selected by the BGP platform. =A0But then....quoted
endmenudiff --git a/arch/powerpc/platforms/Kconfig.cputype b/arch/powerpc/platfo=
rms/Kconfig.cputype
quoted
index 111138c..3a3c711 100644--- a/arch/powerpc/platforms/Kconfig.cputype +++ b/arch/powerpc/platforms/Kconfig.cputype@@ -329,9 +329,13 @@ config NOT_COHERENT_CACHE=A0 =A0 =A0 bool =A0 =A0 =A0 depends on 4xx || 8xx || E200 || PPC_MPC512x || GAMECUBE_COM=
MON
quoted
=A0 =A0 =A0 default n if PPC_47x + =A0 =A0 =A0default n if BGP =A0 =A0 =A0 default y config CHECK_CACHE_COHERENCY =A0 =A0 =A0 bool +config L1_WRITETHROUGH + =A0 =A0 =A0boolYou add an identical option down here. =A0Confused.
Yeah, this was copied from the original patches and it confused me as
well, but I had never
modified Kconfig.cputype before so I wasn't sure if there were some
weird rules. I'm happy
to remove the vestigial one and make the changes you suggest to make
the naming BGP
specific.
-eric