linux-kernel - Re: [PATCH v2 2/3] x86/vmscape: Replace IBPB with branch history clear on exit to userspace

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <aPZe6Xc2H2P-iNQe@google.com>
Date: Mon, 20 Oct 2025 09:10:17 -0700
From: Sean Christopherson <seanjc@...gle.com>
To: Pawan Gupta <pawan.kumar.gupta@...ux.intel.com>
Cc: x86@...nel.org, "H. Peter Anvin" <hpa@...or.com>, Josh Poimboeuf <jpoimboe@...nel.org>, 
	David Kaplan <david.kaplan@....com>, Paolo Bonzini <pbonzini@...hat.com>, linux-kernel@...r.kernel.org, 
	kvm@...r.kernel.org, Asit Mallick <asit.k.mallick@...el.com>, 
	Tao Zhang <tao1.zhang@...el.com>
Subject: Re: [PATCH v2 2/3] x86/vmscape: Replace IBPB with branch history
 clear on exit to userspace

On Wed, Oct 15, 2025, Pawan Gupta wrote:
> diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h
> index 49707e563bdf71bdd05d3827f10dd2b8ac6bca2c..00730cc22c2e7115f6dbb38a1ed8d10383ada5c0 100644
> --- a/arch/x86/include/asm/nospec-branch.h
> +++ b/arch/x86/include/asm/nospec-branch.h
> @@ -534,7 +534,7 @@ void alternative_msr_write(unsigned int msr, u64 val, unsigned int feature)
>  		: "memory");
>  }
>  
> -DECLARE_PER_CPU(bool, x86_ibpb_exit_to_user);
> +DECLARE_PER_CPU(bool, x86_pred_flush_pending);

Rather than "flush pending", what about using "need" in the name to indicate that
a flush is necessary?  That makes it more obvious that e.g. KVM is marking the
CPU as needing a flush by some other code, as opposed to implying that KVM itself
has a pending flush.

And maybe spell out "prediction"?  Without the context of features being checked,
I don't know that I would be able to guess "prediction".

E.g. x86_need_prediction_flush?

Or x86_prediction_flush_exit_to_user if we would prefer to clarify when the flush
needs to occur?