linux-kernel - Re: the x86 sysret_rip test fails on the Intel FRED architecture

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-next>] [day] [month] [year] [list]

Message-ID: <5d4ad3e3-034f-c7da-d141-9c001c2343af@intel.com>
Date:   Fri, 20 Jan 2023 09:45:26 -0800
From:   Dave Hansen <dave.hansen@...el.com>
To:     "Li, Xin3" <xin3.li@...el.com>,
        "tglx@...utronix.de" <tglx@...utronix.de>,
        "mingo@...hat.com" <mingo@...hat.com>,
        "bp@...en8.de" <bp@...en8.de>,
        "peterz@...radead.org" <peterz@...radead.org>,
        "dave.hansen@...ux.intel.com" <dave.hansen@...ux.intel.com>,
        "H.Peter Anvin" <hpa@...or.com>
Cc:     "x86@...nel.org" <x86@...nel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: the x86 sysret_rip test fails on the Intel FRED architecture

On 1/19/23 23:49, Li, Xin3 wrote:
> The x86 sysret_rip test has the following assertion:
> 
>         /* R11 and EFLAGS should already match. */
>         assert(ctx->uc_mcontext.gregs[REG_EFL] ==
>                ctx->uc_mcontext.gregs[REG_R11]);
> 
> 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.

Let's figure out the reason that FRED acts differently, first.  Right
now, the SDM says:

	SYSCALL also saves RFLAGS into R11

so that behavior of SYSCALL _looks_ architectural to me.  Was this
change in SYSCALL behavior with FRED intentional?

If not intentional, it might be something that can still be fixed.  If
it is intentional and is going to be with us for a while we have a few
options.  If userspace is _really_ depending on this behavior, we could
just clobber r11 ourselves in the FRED entry path.  If not, we can
remove the assertion in the selftest.