linux-kernel - Re: [PATCH 5/7] x86/bugs: Only harden syscalls when needed

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]

Message-ID: <ZhkH9aFKw+Y1utPV@gmail.com>
Date: Fri, 12 Apr 2024 12:07:49 +0200
From: Ingo Molnar <mingo@...nel.org>
To: Josh Poimboeuf <jpoimboe@...nel.org>
Cc: Pawan Gupta <pawan.kumar.gupta@...ux.intel.com>, x86@...nel.org,
	linux-kernel@...r.kernel.org,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Daniel Sneddon <daniel.sneddon@...ux.intel.com>,
	Thomas Gleixner <tglx@...utronix.de>,
	Alexandre Chartre <alexandre.chartre@...cle.com>,
	Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>,
	Peter Zijlstra <peterz@...radead.org>,
	Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	Sean Christopherson <seanjc@...gle.com>,
	Andrew Cooper <andrew.cooper3@...rix.com>,
	Dave Hansen <dave.hansen@...ux.intel.com>,
	Nikolay Borisov <nik.borisov@...e.com>,
	KP Singh <kpsingh@...nel.org>, Waiman Long <longman@...hat.com>,
	Borislav Petkov <bp@...en8.de>
Subject: Re: [PATCH 5/7] x86/bugs: Only harden syscalls when needed


* Josh Poimboeuf <jpoimboe@...nel.org> wrote:

> > [...]
> > > @@ -1720,6 +1744,7 @@ static void __init spectre_v2_select_mitigation(void)
> > >  
> > >  	case SPECTRE_V2_CMD_RETPOLINE_LFENCE:
> > >  		pr_err(SPECTRE_V2_LFENCE_MSG);
> > > +		setup_clear_cpu_cap(X86_FEATURE_INDIRECT_SAFE);
> > 
> > I don't know if it intentional, this seems to be the duplicate of
> > X86_FEATURE_INDIRECT_SAFE clear later in SPECTRE_V2_LFENCE mode. Also it
> > seems a bit odd to do this here in SPECTRE_V2_CMD handling.
> 
> Yeah, I accidentally left that in from an earlier implementation.  It's 
> harmless but I'll clean that up too with a new patch unless Ingo wants to 
> remove that line.

Lemme remove it entirely from x86/urgent, so that you can submit an updated 
patch with all feedback included.

In addition to the above line, Pawan's suggestion of doing it in C via 
cpu_feature_enabled() looks quite a bit simpler and easier to read & argue 
about, right?

Thanks,

	Ingo