linux-kernel - Re: [PATCH v1 05/11] x86/tdx: Add __tdx_module_call() and __tdx

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 for Android: free password hash cracker in your pocket

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

Date:   Mon, 14 Jun 2021 14:37:32 -0700
From:   "Kuppuswamy, Sathyanarayanan" 
        <sathyanarayanan.kuppuswamy@...ux.intel.com>
To:     Borislav Petkov <bp@...en8.de>
Cc:     Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Andy Lutomirski <luto@...nel.org>,
        Peter H Anvin <hpa@...or.com>,
        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>,
        Kuppuswamy Sathyanarayanan <knsathya@...nel.org>,
        Sean Christopherson <seanjc@...gle.com>,
        linux-kernel@...r.kernel.org, x86@...nel.org
Subject: Re: [PATCH v1 05/11] x86/tdx: Add __tdx_module_call() and
 __tdx_hypercall() helper functions



On 6/14/21 1:11 PM, Borislav Petkov wrote:
> On Mon, Jun 14, 2021 at 12:45:45PM -0700, Kuppuswamy, Sathyanarayanan wrote:
>> May be I should define a macro for it and use Mov to keep it uniform
>> with other register updates.
> 
> Macro?
> 
> There's the, well, *MOV* instruction, if you insist on keeping it
> uniform. But this is not about keeping it uniform - it is about having
> the code as clear as understandable as possible:
> 
> 
> 	/* Set RAX to TDCALL leaf function 0 */
> 	xor %eax, %eax
> 
> Plain and simple and clear why the XORing is done.

Ok. I will fix the comment.


>> With the trace support, they should be able to see the flow before making
>> the tdx_*_call(). That should be enough clue for debug right?
> 
> Are you expecting all those cloud users to trace their guests just to
> figure that out? I'm sceptical they will...
> 
> Rather, I'd try to allocate a special error value that
> do_tdx_hypercall() returns in %eax and then have the wrapper which will
> puts %r10 on the stack, check that error value and panic with a nice
> error message.
> 

I will add r10 to struct tdx_hypercall_output and return it to callers to
check it.


> 
> Btw, where is that function used? Gurgling, it shows it in some MMIO
> patch, I'm guessing that's still coming.
> 
> As to how to do it properly, you pass in
> 
> 	struct tdx_hypercall_output *out
> 
> as a function parameter and caller can pick out whatever it wants from
> that struct.

It looks like it is used only in MMIO use case now. I think we don't need
it anymore. I will remove it.

> 
> Thx.
> 

-- 
Sathyanarayanan Kuppuswamy
Linux Kernel Developer