linux-kernel - Re: [PATCH v3 18/20] kvm: arm64: Add support for handling 52bit IPA

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:   Mon, 2 Jul 2018 14:43:14 +0100
From:   Marc Zyngier <marc.zyngier@....com>
To:     Suzuki K Poulose <suzuki.poulose@....com>,
        linux-arm-kernel@...ts.infradead.org
Cc:     linux-kernel@...r.kernel.org, kvm@...r.kernel.org,
        kvmarm@...ts.cs.columbia.edu, james.morse@....com,
        cdall@...nel.org, eric.auger@...hat.com, julien.grall@....com,
        will.deacon@....com, catalin.marinas@....com,
        punit.agrawal@....com, qemu-devel@...gnu.org,
        Kristina Martsenko <kristina.martsenko@....com>
Subject: Re: [PATCH v3 18/20] kvm: arm64: Add support for handling 52bit IPA

On 29/06/18 12:15, Suzuki K Poulose wrote:
> Add support for handling the 52bit IPA. 52bit IPA
> support needs changes to the following :
> 
>  1) Page-table entries - We use kernel page table helpers for setting
>  up the stage2. Hence we don't explicit changes here
> 
>  2) VTTBR:BADDR - This is already supported with :
>    commit 529c4b05a3cb2f324aa ("arm64: handle 52-bit addresses in TTBR")
> 
>  3) VGIC support for 52bit: Supported with a patch in this series.
> 
> That leaves us with the handling for PAR and HPAR. This patch adds

HPFAR?

> support for handling the 52bit addresses in PAR and HPFAR,
> which are used while handling the permission faults in stage1.

Overall, this is a pretty confusing commit message. Can you just call it:

KVM/arm64: Add 52bit support for PAR to HPFAR conversion

and just describe that it now uses PHYS_MASK_SHIFT instead of a
hardcoded constant?

> 
> Cc: Marc Zyngier <marc.zyngier@....com>
> Cc: Kristina Martsenko <kristina.martsenko@....com>
> Cc: Christoffer Dall <cdall@...nel.org>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@....com>
> ---
>  arch/arm64/include/asm/kvm_arm.h | 7 +++++++
>  arch/arm64/kvm/hyp/switch.c      | 2 +-
>  2 files changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
> index 2e90942..cb6a2ee 100644
> --- a/arch/arm64/include/asm/kvm_arm.h
> +++ b/arch/arm64/include/asm/kvm_arm.h
> @@ -301,6 +301,13 @@
>  
>  /* Hyp Prefetch Fault Address Register (HPFAR/HDFAR) */
>  #define HPFAR_MASK	(~UL(0xf))
> +/*
> + * We have
> + *	PAR	[PA_Shift - 1	: 12] = PA	[PA_Shift - 1 : 12]
> + *	HPFAR	[PA_Shift - 9	: 4]  = FIPA	[PA_Shift - 1 : 12]
> + */
> +#define PAR_TO_HPFAR(par)		\
> +	(((par) & GENMASK_ULL(PHYS_MASK_SHIFT - 1, 12)) >> 8)
>  
>  #define kvm_arm_exception_type	\
>  	{0, "IRQ" }, 		\
> diff --git a/arch/arm64/kvm/hyp/switch.c b/arch/arm64/kvm/hyp/switch.c
> index 355fb25..fb66320 100644
> --- a/arch/arm64/kvm/hyp/switch.c
> +++ b/arch/arm64/kvm/hyp/switch.c
> @@ -260,7 +260,7 @@ static bool __hyp_text __translate_far_to_hpfar(u64 far, u64 *hpfar)
>  		return false; /* Translation failed, back to guest */
>  
>  	/* Convert PAR to HPFAR format */
> -	*hpfar = ((tmp >> 12) & ((1UL << 36) - 1)) << 4;
> +	*hpfar = PAR_TO_HPFAR(tmp);
>  	return true;
>  }
>  
> 

Otherwise:

Acked-by: Marc Zyngier <marc.zyngier@....com>

	M.
-- 
Jazz is not dead. It just smells funny...