linux-kernel - Re: [PATCH v13 10/13] x86/sgx: Add sgx

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

Date:   Tue, 4 Sep 2018 18:26:05 +0300
From:   Jarkko Sakkinen <jarkko.sakkinen@...ux.intel.com>
To:     "Huang, Kai" <kai.huang@...el.com>
Cc:     "Christopherson, Sean J" <sean.j.christopherson@...el.com>,
        "platform-driver-x86@...r.kernel.org" 
        <platform-driver-x86@...r.kernel.org>,
        "x86@...nel.org" <x86@...nel.org>,
        "nhorman@...hat.com" <nhorman@...hat.com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "tglx@...utronix.de" <tglx@...utronix.de>,
        "suresh.b.siddha@...el.com" <suresh.b.siddha@...el.com>,
        "Ayoun, Serge" <serge.ayoun@...el.com>,
        "hpa@...or.com" <hpa@...or.com>,
        "npmccallum@...hat.com" <npmccallum@...hat.com>,
        "mingo@...hat.com" <mingo@...hat.com>,
        "linux-sgx@...r.kernel.org" <linux-sgx@...r.kernel.org>,
        "Hansen, Dave" <dave.hansen@...el.com>
Subject: Re: [PATCH v13 10/13] x86/sgx: Add sgx_einit() for initializing
 enclaves

On Mon, Sep 03, 2018 at 11:45:14PM +0000, Huang, Kai wrote:
> But INVALID_TOKEN is not only returned when MSRs are mismatched, so do
> you plan to check to rule out other cases that cause INVALID_TOKEN
> before retrying EINIT, or unconditionally  retry EINIT? And we should
> only retry once?

In the case of this error we will do wrmsrs and retry einit once.
This is how I understood it at least.

Q: Do we have to care about VMM sleep anyway? I mean if VMM always traps
EINIT it can write the MSRs with guest values if it has been sleeping.

If the answer is no, then retrying once should be a complete solution
and we don't need pm_cnt.

/Jarkko