linux-kernel - Re: [PATCH] x86/xen: avoid warning in Xen pv guest with CONFIG_AMD_MEM

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] [day] [month] [year] [list]

Message-ID: <aebac64f-a807-f3c6-1738-f14a4470efe3@suse.com>
Date:   Tue, 26 Jan 2021 06:12:42 +0100
From:   Jürgen Groß <jgross@...e.com>
To:     Andrew Cooper <andrew.cooper3@...rix.com>,
        xen-devel@...ts.xenproject.org, x86@...nel.org,
        linux-kernel@...r.kernel.org
Cc:     Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        "H. Peter Anvin" <hpa@...or.com>,
        Boris Ostrovsky <boris.ostrovsky@...cle.com>,
        Stefano Stabellini <sstabellini@...nel.org>,
        stable@...r.kernel.org
Subject: Re: [PATCH] x86/xen: avoid warning in Xen pv guest with
 CONFIG_AMD_MEM_ENCRYPT enabled

On 25.01.21 18:26, Andrew Cooper wrote:
> On 25/01/2021 14:00, Juergen Gross wrote:
>> diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c
>> index 4409306364dc..82948251f57b 100644
>> --- a/arch/x86/xen/enlighten_pv.c
>> +++ b/arch/x86/xen/enlighten_pv.c
>> @@ -583,6 +583,14 @@ DEFINE_IDTENTRY_RAW(xenpv_exc_debug)
>>   		exc_debug(regs);
>>   }
>>   
>> +#ifdef CONFIG_AMD_MEM_ENCRYPT
>> +DEFINE_IDTENTRY_RAW(xenpv_exc_vmm_communication)
>> +{
>> +	/* This should never happen and there is no way to handle it. */
>> +	panic("X86_TRAP_VC in Xen PV mode.");
> 
> Honestly, exactly the same is true of #VE, #HV and #SX.
> 
> What we do in the hypervisor is wire up one handler for all unknown
> exceptions (to avoid potential future #DF issues) leading to a panic.
> Wouldn't it be better to do this unconditionally, especially as #GP/#NP
> doesn't work for PV guests for unregistered callbacks, rather than
> fixing up piecewise like this?

I agree it would be better to have a "catch all unknown" handler.

I'll have a try how this would look like.


Juergen


Download attachment "OpenPGP_0xB0DE9DD628BF132F.asc" of type "application/pgp-keys" (3092 bytes)

Download attachment "OpenPGP_signature" of type "application/pgp-signature" (496 bytes)