linux-kernel - Re: [PATCH v4] arm64: errata: Work around AmpereOne's erratum AC04_CPU

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: <87wma7e5jd.wl-maz@kernel.org>
Date: Fri, 23 May 2025 16:00:06 +0100
From: Marc Zyngier <maz@...nel.org>
To: Mark Brown <broonie@...nel.org>
Cc: D Scott Phillips <scott@...amperecomputing.com>,
	Catalin Marinas <catalin.marinas@....com>,
	James Clark <james.clark@...aro.org>,
	James Morse <james.morse@....com>,
	Joey Gouly <joey.gouly@....com>,
	Kevin Brodsky <kevin.brodsky@....com>,
	Mark Rutland <mark.rutland@....com>,
	Oliver Upton <oliver.upton@...ux.dev>,
	"Rob Herring  (Arm)" <robh@...nel.org>,
	Shameer Kolothum <shameerali.kolothum.thodi@...wei.com>,
	Shiqi Liu <shiqiliu@...t.edu.cn>,
	Will Deacon <will@...nel.org>,
	Yicong Yang <yangyicong@...ilicon.com>,
	kvmarm@...ts.linux.dev,
	linux-arm-kernel@...ts.infradead.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH v4] arm64: errata: Work around AmpereOne's erratum AC04_CPU_23

On Fri, 23 May 2025 15:15:53 +0100,
Mark Brown <broonie@...nel.org> wrote:
> 
> On Tue, May 13, 2025 at 11:45:14AM -0700, D Scott Phillips wrote:
> > On AmpereOne AC04, updates to HCR_EL2 can rarely corrupt simultaneous
> > translations for data addresses initiated by load/store instructions.
> > Only instruction initiated translations are vulnerable, not translations
> > from prefetches for example. A DSB before the store to HCR_EL2 is
> > sufficient to prevent older instructions from hitting the window for
> > corruption, and an ISB after is sufficient to prevent younger
> > instructions from hitting the window for corruption.
> 
> This patch, which is in -next as fed55f49fad181be9dfb93c0, breaks the
> build of at least the vDSO selftests:
> 
> $ make -C tools/testing/selftests ARCH=arm64 LLVM=1 TARGETS=vDSO
> 
>   CC       vdso_test_chacha
> In file included from vgetrandom-chacha.S:9:
> In file included from ./../../../../arch/arm64/kernel/vdso/vgetrandom-chacha.S:5:
> In file included from /home/broonie/git/bisect/tools/testing/selftests/../../../arch/arm64/include/asm/assembler.h:21:
> In file included from /home/broonie/git/bisect/tools/testing/selftests/../../../arch/arm64/include/asm/cpufeature.h:13:
> /home/broonie/git/bisect/tools/testing/selftests/../../../arch/arm64/include/asm/sysreg.h:1097:5: error: function-like macro 'IS_ENABLED' is not defined
>  1097 | #if IS_ENABLED(CONFIG_AMPERE_ERRATUM_AC04_CPU_23)
>       |     ^
> 1 error generated.

This:

diff --git a/arch/arm64/include/asm/assembler.h b/arch/arm64/include/asm/assembler.h
index ad63457a05c5b..4389d3916426c 100644
--- a/arch/arm64/include/asm/assembler.h
+++ b/arch/arm64/include/asm/assembler.h
@@ -13,6 +13,7 @@
 #define __ASM_ASSEMBLER_H
 
 #include <linux/export.h>
+#include <linux/kconfig.h>
 
 #include <asm/alternative.h>
 #include <asm/asm-bug.h>

should solve it.

But it also outlines that the vdso is getting built using stuff that
is not meant for userspace code.

	M.

-- 
Jazz isn't dead. It just smells funny.