[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <86ecwsgxnh.fsf@scott-ph-mail.amperecomputing.com>
Date: Tue, 13 May 2025 11:42:42 -0700
From: D Scott Phillips <scott@...amperecomputing.com>
To: Oliver Upton <oliver.upton@...ux.dev>
Cc: 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>, Marc Zyngier
<maz@...nel.org>, Mark Brown <broonie@...nel.org>, Mark Rutland
<mark.rutland@....com>, "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 v3] arm64: errata: Work around AmpereOne's erratum
AC04_CPU_23
Oliver Upton <oliver.upton@...ux.dev> writes:
> Hey D Scott,
>
> On Thu, May 08, 2025 at 02:00:09PM -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.
>>
>> Signed-off-by: D Scott Phillips <scott@...amperecomputing.com>
>
> Overall looks good, still needs an entry in Documentation/arch/arm64/silicon-errata.rst
> which Marc noted in v2.
Ah, sorry for missing that and making you repeat yourself. I'll fix
that.
Powered by blists - more mailing lists