linux-kernel - Re: [PATCH 17/20] x86/tdx: Convert VM_RD/VM

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: Fri, 17 May 2024 09:07:54 -0700
From: Dave Hansen <dave.hansen@...el.com>
To: "Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>,
 Sean Christopherson <seanjc@...gle.com>, Paolo Bonzini
 <pbonzini@...hat.com>, Dave Hansen <dave.hansen@...ux.intel.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>, "K. Y. Srinivasan" <kys@...rosoft.com>,
 Haiyang Zhang <haiyangz@...rosoft.com>, Wei Liu <wei.liu@...nel.org>,
 Dexuan Cui <decui@...rosoft.com>, Josh Poimboeuf <jpoimboe@...nel.org>,
 Peter Zijlstra <peterz@...radead.org>
Cc: linux-coco@...ts.linux.dev, linux-kernel@...r.kernel.org,
 linux-hyperv@...r.kernel.org
Subject: Re: [PATCH 17/20] x86/tdx: Convert VM_RD/VM_WR tdcalls to use new
 TDCALL macros

Let's say you're debugging tdg_vm_rd().  You suspect someone read the
spec wrong.  You pull up the spec:

	https://sr71.net/~dave/intel/tdg.vm.rd.png

On 5/17/24 07:19, Kirill A. Shutemov wrote:
>  static inline u64 tdg_vm_rd(u64 field, u64 *value)
>  {
> -	struct tdx_module_args args = {
> -		.rdx = field,
> -	};

RDX is assigned 'field'.  Makes sense based on the input operands.

> -	u64 ret;
> -
> -	ret = __tdcall_ret(TDG_VM_RD, &args)> -	*value = args.r8;

'value' is set to r8.  Also matches the spec.  It's obvious that this is
a 'two return values' pattern.

> -	return ret;

This is also obviously correct.

Compare that to:

> +	return TDCALL_1(TDG_VM_RD, 0, field, 0, 0, value);
>  }

Where it's 100% opaque which registers thing to into or that 'value' is
an output, not an input.

So, yeah, this is fewer lines of C code.  But it's *WAY* less
self-documenting.  It's harder to audit.  It's harder to understand and
it's more opaque.

While the goals here are laudable, I'm not a big fan of the end result.