linux-kernel - Re: [RFC] x86/alternative: Support relocations in alternatives

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:   Sat, 4 Feb 2023 00:25:49 +0100
From:   Borislav Petkov <bp@...en8.de>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     x86@...nel.org, Masami Hiramatsu <mhiramat@...nel.org>,
        "Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>,
        Andrew Cooper <Andrew.Cooper3@...rix.com>,
        Josh Poimboeuf <jpoimboe@...hat.com>,
        linux-kernel@...r.kernel.org
Subject: Re: [RFC] x86/alternative: Support relocations in alternatives

On Fri, Feb 03, 2023 at 05:46:47PM +0100, Borislav Petkov wrote:
> On Fri, Feb 03, 2023 at 05:04:35PM +0100, Borislav Petkov wrote:
> > Rest on IRC. :)
> 
> Here's what I'm thinking. It still fails somewhere while booting so it
> is not good yet but the idea is to show what I mean:

Yeah, forget it. I need both next_rip at the place we're patching and
next_rip in the .altinstr_replacement section. And by the time I do
that, it won't get any prettier.

And I think yours solves that more elegantly but please document the
math transformation to compute the new offset.

Also, pls do this:

	/*
	 * Do not recompute the offset if the target is within the
	 * patched insn block.
	 */
	if (target < repl || target > repl + repl_len)

to hint that you don't have to replace the offsets which are already
correct when a whole set of insns is being patched in.

FILL_RETURN_BUFFER was one example. :)

Thx.

-- 
Regards/Gruss,
    Boris.

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