linux-kernel - Re: [PATCH v3 1/3] arm64: add FEAT_MTE_TAGGED

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]

Message-ID: <aBT9CA7vnlfszTDw@arm.com>
Date: Fri, 2 May 2025 18:12:40 +0100
From: Catalin Marinas <catalin.marinas@....com>
To: Yeoreum Yun <yeoreum.yun@....com>
Cc: will@...nel.org, broonie@...nel.org, anshuman.khandual@....com,
	joey.gouly@....com, yury.khrustalev@....com, maz@...nel.org,
	oliver.upton@...ux.dev, frederic@...nel.org,
	shmeerali.kolothum.thodi@...wei.com, james.morse@....com,
	mark.rutland@....com, huangxiaojia2@...wei.com,
	akpm@...ux-foundation.org, surenb@...gle.com, robin.murphy@....com,
	linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
	linux-doc@...r.kernel.org, nd@....com,
	Peter Collingbourne <pcc@...gle.com>
Subject: Re: [PATCH v3 1/3] arm64: add FEAT_MTE_TAGGED_FAR feature

On Thu, Apr 10, 2025 at 08:47:19AM +0100, Yeoreum Yun wrote:
> Add FEAT_MTE_TAGGED_FAR cpucap which makes FAR_ELx report
> all non-address bits on a synchronous MTE tag check fault since Armv8.9
> 
> Signed-off-by: Yeoreum Yun <yeoreum.yun@....com>
> Acked-by: Yury Khrustalev <yury.khrustalev@....com>
[...]
> @@ -3191,6 +3199,7 @@ static const struct arm64_cpu_capabilities arm64_elf_hwcaps[] = {
>  #ifdef CONFIG_ARM64_MTE
>  	HWCAP_CAP(ID_AA64PFR1_EL1, MTE, MTE2, CAP_HWCAP, KERNEL_HWCAP_MTE),
>  	HWCAP_CAP(ID_AA64PFR1_EL1, MTE, MTE3, CAP_HWCAP, KERNEL_HWCAP_MTE3),
> +	HWCAP_CAP(ID_AA64PFR2_EL1, MTEFAR, IMP, CAP_HWCAP, KERNEL_HWCAP_MTE_FAR),

Nitpick: from a theoretical bisectability perspective, I'd move the
hwcap exposing to the following patch. The rest, feature detection etc
can stay in this file. Otherwise, one can apply this patch, see the
hwcap but not get the actual behaviour.

>  #endif /* CONFIG_ARM64_MTE */
>  	HWCAP_CAP(ID_AA64MMFR0_EL1, ECV, IMP, CAP_HWCAP, KERNEL_HWCAP_ECV),
>  	HWCAP_CAP(ID_AA64MMFR1_EL1, AFP, IMP, CAP_HWCAP, KERNEL_HWCAP_AFP),
> diff --git a/arch/arm64/kernel/cpuinfo.c b/arch/arm64/kernel/cpuinfo.c
> index 285d7d538342..e2b13454e38a 100644
> --- a/arch/arm64/kernel/cpuinfo.c
> +++ b/arch/arm64/kernel/cpuinfo.c
> @@ -160,6 +160,7 @@ static const char *const hwcap_str[] = {
>  	[KERNEL_HWCAP_SME_SFEXPA]	= "smesfexpa",
>  	[KERNEL_HWCAP_SME_STMOP]	= "smestmop",
>  	[KERNEL_HWCAP_SME_SMOP4]	= "smesmop4",
> +	[KERNEL_HWCAP_MTE_FAR]		= "mte_far",

Personal preference: "mtefar" (no underscore), mostly because there's no
other underscore in the hwcap strings yet ;).

-- 
Catalin