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;
Message-ID: <33c2a495-13ad-97ac-f2c2-4096cf8f5b58@molgen.mpg.de>
Date:   Wed, 24 Aug 2022 00:33:07 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.1.2
Subject: Re: WARNING: CPU: 1 PID: 83 at arch/x86/kernel/cpu/sgx/main.c:446
 ksgxd+0x1b7/0x1d0
Content-Language: en-US
To:     Dave Hansen <dave.hansen@intel.com>,
        Jarkko Sakkinen <jarkko@kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
        x86@kernel.org, Reinette Chatre <reinette.chatre@intel.com>
Cc:     linux-sgx@vger.kernel.org, LKML <linux-kernel@vger.kernel.org>
References: <ce0b4d26-3a6e-7c5a-5f66-44cba05f9f35@molgen.mpg.de>
 <4253695b-85aa-a2fb-fbf6-718db8b6c20c@molgen.mpg.de>
 <46e3483b-a5ab-2a05-8a28-f9ea87e881c3@intel.com>
 <04c9d5fa-5861-bbc3-3e2f-e18a73866645@molgen.mpg.de>
 <63a60042-4a4a-3bc3-5fa1-4495d80cc06c@molgen.mpg.de>
 <beabe16c-b28b-cd65-f6b6-4242bc74926d@intel.com>
From:   Paul Menzel <pmenzel@molgen.mpg.de>
In-Reply-To: <beabe16c-b28b-cd65-f6b6-4242bc74926d@intel.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk

Dear Dave,


Thank you for your reply.

Am 23.08.22 um 18:32 schrieb Dave Hansen:
> On 8/23/22 06:48, Paul Menzel wrote:
>>>> I'm suspecting either a BIOS problem.  Reinette (cc'd) also thought this
>>>> might be a case of the SGX initialization getting a bit too far along
>>>> when it should have been disabled.
>>>>
>>>> We had some bugs where we didn't stop fast enough after spitting out the
>>>> "SGX Launch Control is locked..." errors.
>>
>> Let’s hope it’s something known to you.
> 
> Thanks for the extra debug info.  Unfortunately, nothing is really
> sticking out as an obvious problem.
> 
> The EREMOVE return codes would be interesting to know, as well as an
> idea what the physical addresses are that fail and the _counts_ of how
> many pages get sanitized versus fail.

Is there a knob to print out this information? Or way to get this 
information using ftrace? I’d like to avoid rebuilding the Linux kernel.

> But, I don't really have a theory about what could be going on yet.

Kind regards,

Paul