From mboxrd@z Thu Jan 1 00:00:00 1970 From: will.deacon@arm.com (Will Deacon) Date: Thu, 17 Aug 2017 14:31:54 +0100 Subject: [PATCH 6/6] arm64: Remove the CONFIG_ARM64_HW_AFDBM option In-Reply-To: <20170725135308.18173-7-catalin.marinas@arm.com> References: <20170725135308.18173-1-catalin.marinas@arm.com> <20170725135308.18173-7-catalin.marinas@arm.com> Message-ID: <20170817133154.GB29194@arm.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Tue, Jul 25, 2017 at 02:53:08PM +0100, Catalin Marinas wrote: > Since the pte handling for hardware AF/DBM works even when the hardware > feature is not present, make the implementation permanent and remove the > Kconfig option. > > Cc: Will Deacon > Cc: Marc Zyngier > Cc: Christoffer Dall > Signed-off-by: Catalin Marinas > --- > arch/arm64/Kconfig | 17 ----------------- > arch/arm64/include/asm/pgtable.h | 9 +-------- > arch/arm64/kvm/hyp/s2-setup.c | 2 +- > arch/arm64/mm/fault.c | 2 -- > arch/arm64/mm/proc.S | 3 +-- > 5 files changed, 3 insertions(+), 30 deletions(-) Largely in favour of this, but I'd prefer to keep the option so that we can build kernels where the enabling: > diff --git a/arch/arm64/mm/proc.S b/arch/arm64/mm/proc.S > index 877d42fb0df6..ba82f8bf3cf1 100644 > --- a/arch/arm64/mm/proc.S > +++ b/arch/arm64/mm/proc.S > @@ -234,7 +234,7 @@ ENTRY(__cpu_setup) > */ > mrs x9, ID_AA64MMFR0_EL1 > bfi x10, x9, #32, #3 > -#ifdef CONFIG_ARM64_HW_AFDBM > + > /* > * Hardware update of the Access and Dirty bits. > */ > @@ -246,7 +246,6 @@ ENTRY(__cpu_setup) > orr x10, x10, #TCR_HD // hardware Dirty flag update > 1: orr x10, x10, #TCR_HA // hardware Access flag update > 2: > -#endif /* CONFIG_ARM64_HW_AFDBM */ > msr tcr_el1, x10 > ret // return to head.S > ENDPROC(__cpu_setup) is still configurable. That might help for debugging suspected races in this area. Will