[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CACT4Y+Z3suKV=M1zLEfU5SWAmYdLbRcorvsY2UK-CTvYb1bNPA@mail.gmail.com>
Date: Thu, 6 Feb 2025 19:35:54 +0100
From: Dmitry Vyukov <dvyukov@...gle.com>
To: Dave Hansen <dave.hansen@...el.com>
Cc: aruna.ramakrishna@...cle.com, mathieu.desnoyers@...icios.com,
peterz@...radead.org, paulmck@...nel.org, boqun.feng@...il.com,
dave.hansen@...ux.intel.com, jannh@...gle.com, jeffxu@...omium.org,
jorgelo@...omium.org, keescook@...omium.org, keith.lucas@...cle.com,
linux-kernel@...r.kernel.org, linux-mm@...ck.org, mingo@...nel.org,
rick.p.edgecombe@...el.com, sroettger@...gle.com, tglx@...utronix.de,
x86@...nel.org
Subject: Re: [PATCH v8 3/5] x86/pkeys: Update PKRU to enable all pkeys before XSAVE
On Thu, 6 Feb 2025 at 19:21, Dave Hansen <dave.hansen@...el.com> wrote:
>
> On 2/4/25 02:01, Dmitry Vyukov wrote:
> >> But this updated PKRU value is also pushed onto the sigframe, which
> >> means the register value restored from sigcontext will be different from
> >> the user-defined one, which is unexpected. Fix that by overwriting the
> >> PKRU value on the sigframe with the original, user-defined PKRU.
> >
> > This unfortunatly seems to be broken for rseq user-space writes.
>
> Hi Dmitry,
>
> Are you saying that Aruna's patch caused a regression in the rseq code?
> Or that the rseq code has a separate but similar issue to the one that
> Aruna's patch fixed?
No, no regression. Just another issue for real PKEY uses.
Powered by blists - more mailing lists