Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Fri, 20 Jan 2023 20:59:58 -0800
From:   "H. Peter Anvin" <hpa@zytor.com>
To:     Dave Hansen <dave.hansen@intel.com>,
        "Li, Xin3" <xin3.li@intel.com>,
        "tglx@linutronix.de" <tglx@linutronix.de>,
        "mingo@redhat.com" <mingo@redhat.com>,
        "bp@alien8.de" <bp@alien8.de>,
        "peterz@infradead.org" <peterz@infradead.org>,
        "dave.hansen@linux.intel.com" <dave.hansen@linux.intel.com>
CC:     "x86@kernel.org" <x86@kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: the x86 sysret_rip test fails on the Intel FRED architecture
User-Agent: K-9 Mail for Android
In-Reply-To: <5d4ad3e3-034f-c7da-d141-9c001c2343af@intel.com>
References: <SA1PR11MB6734FA9139B9C9F6CC2ED123A8C59@SA1PR11MB6734.namprd11.prod.outlook.com> <5d4ad3e3-034f-c7da-d141-9c001c2343af@intel.com>
Message-ID: <18B5DB6D-AEBD-4A67-A7B3-CE64940819B7@zytor.com>
MIME-Version: 1.0
Content-Type: text/plain;
 charset=utf-8
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On January 20, 2023 9:45:26 AM PST, Dave Hansen <dave=2Ehansen@intel=2Ecom>=
 wrote:
>On 1/19/23 23:49, Li, Xin3 wrote:
>> The x86 sysret_rip test has the following assertion:
>>=20
>>         /* R11 and EFLAGS should already match=2E */
>>         assert(ctx->uc_mcontext=2Egregs[REG_EFL] =3D=3D
>>                ctx->uc_mcontext=2Egregs[REG_R11]);
>>=20
>> This is being tested to avoid kernel state leak due to sysret vs iret,
>> but that on FRED r11 is *always* preserved, and the test just fails=2E
>
>Let's figure out the reason that FRED acts differently, first=2E  Right
>now, the SDM says:
>
>	SYSCALL also saves RFLAGS into R11
>
>so that behavior of SYSCALL _looks_ architectural to me=2E  Was this
>change in SYSCALL behavior with FRED intentional?
>
>If not intentional, it might be something that can still be fixed=2E  If
>it is intentional and is going to be with us for a while we have a few
>options=2E  If userspace is _really_ depending on this behavior, we could
>just clobber r11 ourselves in the FRED entry path=2E  If not, we can
>remove the assertion in the selftest=2E

We can't clobber it in the FRED entry path, since it is common for all eve=
nts, but we could do it in the syscall dispatch=2E

However, it doesn't seem to make sense to do so to me=2E The current behav=
ior is much more of an artifact than desired behavior=2E