Re: [RFC PATCH] x86/sev: x86/sev: enforce PC-relative addressing in clang

From: Kevin Loughlin <kevinloughlin@google.com>
To: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
	 Dave Hansen <dave.hansen@linux.intel.com>,
	x86@kernel.org,  "H. Peter Anvin" <hpa@zytor.com>,
	Andy Lutomirski <luto@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	 Nathan Chancellor <nathan@kernel.org>,
	Nick Desaulniers <ndesaulniers@google.com>,
	 Bill Wendling <morbo@google.com>,
	Justin Stitt <justinstitt@google.com>,
	 Rick Edgecombe <rick.p.edgecombe@intel.com>,
	Kees Cook <keescook@chromium.org>,
	 "Masami Hiramatsu (Google)" <mhiramat@kernel.org>,
	Ze Gao <zegao2021@gmail.com>,
	 Josh Poimboeuf <jpoimboe@kernel.org>,
	Pengfei Xu <pengfei.xu@intel.com>,
	 Brijesh Singh <brijesh.singh@amd.com>,
	Michael Roth <michael.roth@amd.com>,
	 Ashish Kalra <ashish.kalra@amd.com>,
	Tom Lendacky <thomas.lendacky@amd.com>,
	 Joerg Roedel <jroedel@suse.de>,
	linux-kernel@vger.kernel.org, llvm@lists.linux.dev,
	 linux-coco@lists.linux.dev, Adam Dunlap <acdunlap@google.com>,
	 Peter Gonda <pgonda@google.com>, Jacob Xu <jacobhxu@google.com>,
	 Sidharth Telang <sidtelang@google.com>
Subject: Re: [RFC PATCH] x86/sev: x86/sev: enforce PC-relative addressing in clang
Date: Wed, 10 Jan 2024 09:28:21 -0800	[thread overview]
Message-ID: <CAGdbjmK3oTE=3PizXrvux2XT1GLMPp=Fto-2ySfSf_Em4Tjnsw@mail.gmail.com> (raw)
In-Reply-To: <20240110133657.vbpzplchgaim3bya@box>

On Wed, Jan 10, 2024 at 5:37 AM Kirill A. Shutemov
<kirill.shutemov@linux.intel.com> wrote:
>
> Similar issues was fixed before with fixup_pointer() tricks. Have you
> tried looking this direction.
>
> Relevant thread starting with:
>
> https://lore.kernel.org/all/20210920192341.maue7db4lcbdn46x@box.shutemov.name

Yes, I considered using `fixup_pointer()` as well, and I agree that it
remains a valid option to solve this problem. I just now mentioned the
primary downside to using `fixup_pointer()` in my response to Andi
(https://lore.kernel.org/lkml/CAGdbjmJcQeLWroBkbRC-cFQ6OuMHONsGLzpUrrA5k3uc8Rzkmw@mail.gmail.com/).
Namely, we would have to pass around `physaddr` from `__startup64()`
to any functions wanting to use `fixup_pointer()`). However, we should
be able to get around needing `physaddr` by using something like the
`GET_RIP_RELATIVE_PTR()` macro that I proposed in the same reply
(forthcoming in v2 of this RFC PATCH).