linux-kernel - Re: [PATCH final?] x86/bugs: KVM: Add support for SRSO_MSR

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] [day] [month] [year] [list]

Message-ID: <20250303135914.GGZ8W1skZjxqhYiJa8@fat_crate.local>
Date: Mon, 3 Mar 2025 14:59:14 +0100
From: Borislav Petkov <bp@...en8.de>
To: Yosry Ahmed <yosry.ahmed@...ux.dev>
Cc: Patrick Bellasi <derkling@...gle.com>,
	Sean Christopherson <seanjc@...gle.com>,
	Paolo Bonzini <pbonzini@...hat.com>,
	Josh Poimboeuf <jpoimboe@...hat.com>,
	Pawan Gupta <pawan.kumar.gupta@...ux.intel.com>, x86@...nel.org,
	kvm@...r.kernel.org, linux-kernel@...r.kernel.org,
	Patrick Bellasi <derkling@...bug.net>,
	Brendan Jackman <jackmanb@...gle.com>,
	David Kaplan <David.Kaplan@....com>
Subject: Re: [PATCH final?] x86/bugs: KVM: Add support for SRSO_MSR_FIX

On Wed, Feb 26, 2025 at 07:51:08PM +0000, Yosry Ahmed wrote:
> To add more details about this, we are using ASI as our main mitigation
> for SRSO. However, it's likely that bp-spec-reduce is cheaper, so we
> basically want to always use bp-spec-reduce if available, if not, we
> don't want the ibpb-on-vmexit or safe-ret as they are a lot more
> expensive than ASI.
> 
> So we want the cmdline option to basically say only use bp-spec-reduce
> if it's available, but don't fallback if it isn't.

Yap, that should also be a part of the commit message.

> On the other hand we are enlighting ASI to skip mitigating SRSO if
> X86_FEATURE_SRSO_BP_SPEC_REDUCE is enabled

Yap, makes sense.

-- 
Regards/Gruss,
    Boris.

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