linux-kernel - Re: [PATCH v6 05/10] x86/tdx: Handle port I/O

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:   Thu, 23 Sep 2021 10:24:20 -0700
From:   "Kuppuswamy, Sathyanarayanan" 
        <sathyanarayanan.kuppuswamy@...ux.intel.com>
To:     Tom Lendacky <thomas.lendacky@....com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        x86@...nel.org, "H . Peter Anvin" <hpa@...or.com>
Cc:     Dave Hansen <dave.hansen@...el.com>,
        Tony Luck <tony.luck@...el.com>,
        Dan Williams <dan.j.williams@...el.com>,
        Andi Kleen <ak@...ux.intel.com>,
        Kirill Shutemov <kirill.shutemov@...ux.intel.com>,
        Sean Christopherson <seanjc@...gle.com>,
        Kuppuswamy Sathyanarayanan <knsathya@...nel.org>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v6 05/10] x86/tdx: Handle port I/O



On 9/23/21 9:32 AM, Tom Lendacky wrote:
> On 9/22/21 5:52 PM, Kuppuswamy Sathyanarayanan wrote:
>> From: "Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>
>>
>> TDX hypervisors cannot emulate instructions directly. This includes
>> port IO which is normally emulated in the hypervisor. All port IO
>> instructions inside TDX trigger the #VE exception in the guest and
>> would be normally emulated there.
>>
>> Also string I/O is not supported in TDX guest. So, unroll the string
>> I/O operation into a loop operating on one element at a time. This
>> method is similar to AMD SEV, so just extend the support for TDX guest
>> platform.
>>
>> Add a new confidential guest flag CC_ATTR_GUEST_UNROLL_STRING_IO to
>> add string unroll support in asm/io.h
>>
>> Co-developed-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@...ux.intel.com>
>> Signed-off-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@...ux.intel.com>
>> Signed-off-by: Kirill A. Shutemov <kirill.shutemov@...ux.intel.com>
>> Reviewed-by: Andi Kleen <ak@...ux.intel.com>
>> Reviewed-by: Dan Williams <dan.j.williams@...el.com>
>> ---
>>
>> Changes since v5:
>>   * Changed prot_guest_has() to cc_platform_has().
>>
>> Changes since v4:
>>   * Changed order of variable declaration in tdx_handle_io().
>>   * Changed tdg_* prefix with tdx_*.
>>
>> Changes since v3:
>>   * Included PATTR_GUEST_UNROLL_STRING_IO protected guest flag
>>     addition change in this patch.
>>   * Rebased on top of Tom Lendacks protected guest change.
>>
>> Changes since v2:
>>   * None
>>
>> Changes since v1:
>>   * Fixed comments for tdg_handle_io().
>>   * Used _tdx_hypercall() instead of __tdx_hypercall() in tdg_handle_io().
>>
>>   arch/x86/include/asm/io.h   |  7 +++++--
>>   arch/x86/kernel/cpu/intel.c |  1 +
>>   arch/x86/kernel/tdx.c       | 35 +++++++++++++++++++++++++++++++++++
>>   include/linux/cc_platform.h | 11 +++++++++++
>>   4 files changed, 52 insertions(+), 2 deletions(-)
>>
>> diff --git a/arch/x86/include/asm/io.h b/arch/x86/include/asm/io.h
>> index fa6aa43e5dc3..67e0c4a0a0f4 100644
>> --- a/arch/x86/include/asm/io.h
>> +++ b/arch/x86/include/asm/io.h
>> @@ -40,6 +40,7 @@
>>   #include <linux/string.h>
>>   #include <linux/compiler.h>
>> +#include <linux/cc_platform.h>
>>   #include <asm/page.h>
>>   #include <asm/tdx.h>
>>   #include <asm/early_ioremap.h>
>> @@ -310,7 +311,8 @@ static inline unsigned type in##bwl##_p(int port)            \
>>                                       \
>>   static inline void outs##bwl(int port, const void *addr, unsigned long count) \
>>   {                                    \
>> -    if (sev_key_active()) {                        \ > +    if (sev_key_active() 
>> ||                        \
>> +        cc_platform_has(CC_ATTR_GUEST_UNROLL_STRING_IO)) {        \
> 
> Would it make sense to make sev_key_active() and sev_enable_key generic and just re-use those 
> instead of adding CC_ATTR_GUEST_UNROLL_STRING_IO and having multiple conditions here?
> 
> You can set the key in the TDX init routine just like SEV does.

Any reason for using sev_enable_key over CC attribute? IMO, CC attribute exist
to generalize the common feature code. My impression is SEV is specific to AMD
code.

> 
> Thanks,
> Tom
> 

-- 
Sathyanarayanan Kuppuswamy
Linux Kernel Developer