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, 18 Mar 2022 12:29:23 +0100
From:   Borislav Petkov <bp@alien8.de>
To:     Joerg Roedel <joro@8bytes.org>
Cc:     x86@kernel.org, Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>,
        Dave Hansen <dave.hansen@linux.intel.com>, hpa@zytor.com,
        Joerg Roedel <jroedel@suse.de>,
        Tom Lendacky <thomas.lendacky@amd.com>,
        Brijesh Singh <brijesh.singh@amd.com>,
        linux-kernel@vger.kernel.org, stable@vger.kernel.org
Subject: Re: [PATCH] x86/sev: Unroll string mmio with
 CC_ATTR_GUEST_UNROLL_STRING_IO
Message-ID: <YjRtE30Usxy8PurP@zn.tnic>
References: <20220310112615.31133-1-joro@8bytes.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <20220310112615.31133-1-joro@8bytes.org>
Precedence: bulk

On Thu, Mar 10, 2022 at 12:26:15PM +0100, Joerg Roedel wrote:
> From: Joerg Roedel <jroedel@suse.de>
> 
> The io specific memcpy/memset functions use string mmio accesses to do
> their work. Under SEV the hypervisor can't emulate these instructions,
> because they read/write directly from/to encrypted memory.
> 
> KVM will inject a page fault exception into the guest when it is asked
> to emulate string mmio instructions for an SEV guest:
> 
> 	BUG: unable to handle page fault for address: ffffc90000065068
> 	#PF: supervisor read access in kernel mode
> 	#PF: error_code(0x0000) - not-present page
> 	PGD 8000100000067 P4D 8000100000067 PUD 80001000fb067 PMD 80001000fc067 PTE 80000000fed40173
> 	Oops: 0000 [#1] PREEMPT SMP NOPTI
> 	CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.17.0-rc7 #3
> 
> As string mmio for an SEV guest can not be supported by the
> hypervisor, unroll the instructions for CC_ATTR_GUEST_UNROLL_STRING_IO
> enabled kernels.

What I'm missing in this description is why wasn't it a problem until now?

You mentioned something about libvirt adding TPMs and that causing this
but I'm still unclear as to why exactly this is causing the issue. I'm
guessing SEV guests didn't do string IO but libvirt adding a TPM is
somehow causing them to use them now...

> @@ -56,9 +55,8 @@ void memcpy_toio(volatile void __iomem *to, const void *from, size_t n)
>  	}
>  	rep_movs((void *)to, (const void *) from, n);
>  }
> -EXPORT_SYMBOL(memcpy_toio);
>  
> -void memset_io(volatile void __iomem *a, int b, size_t c)
> +static void string_memset_io(volatile void __iomem *a, int b, size_t c)

You can simply remove that wrapper and use memset() at the callsite.

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette