Re: [PATCH v2 11/35] x86/bugs: Restructure spectre_v1 mitigation

[Pawan Gupta]

On Thu, Nov 14, 2024 at 03:36:44PM +0000, Kaplan, David wrote:
> [AMD Official Use Only - AMD Internal Distribution Only]
> 
> > -----Original Message-----
> > From: Pawan Gupta <pawan.kumar.gupta@xxxxxxxxxxxxxxx>
> > Sent: Thursday, November 14, 2024 12:57 AM
> > To: Kaplan, David <David.Kaplan@xxxxxxx>
> > Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>; Borislav Petkov <bp@xxxxxxxxx>; Peter
> > Zijlstra <peterz@xxxxxxxxxxxxx>; Josh Poimboeuf <jpoimboe@xxxxxxxxxx>; Ingo
> > Molnar <mingo@xxxxxxxxxx>; Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>;
> > x86@xxxxxxxxxx; H . Peter Anvin <hpa@xxxxxxxxx>; linux-kernel@xxxxxxxxxxxxxxx
> > Subject: Re: [PATCH v2 11/35] x86/bugs: Restructure spectre_v1 mitigation
> >
> > Caution: This message originated from an External Source. Use proper caution
> > when opening attachments, clicking links, or responding.
> >
> >
> > On Tue, Nov 05, 2024 at 03:54:31PM -0600, David Kaplan wrote:
> > >  static void __init spectre_v1_select_mitigation(void)
> > >  {
> > > -     if (!boot_cpu_has_bug(X86_BUG_SPECTRE_V1) || cpu_mitigations_off()) {
> > > +     if (!boot_cpu_has_bug(X86_BUG_SPECTRE_V1) ||
> > > + cpu_mitigations_off())
> > >               spectre_v1_mitigation = SPECTRE_V1_MITIGATION_NONE;
> > > +}
> > > +
> > > +static void __init spectre_v1_apply_mitigation(void) {
> > > +     if (!boot_cpu_has_bug(X86_BUG_SPECTRE_V1) ||
> > > +cpu_mitigations_off())
> >
> > We probably don't need to repeat this check, is this okay:
> >
> >         if (spectre_v1_mitigation == SPECTRE_V1_MITIGATION_NONE)
> > >               return;
> > > -     }
> > >
> > >       if (spectre_v1_mitigation == SPECTRE_V1_MITIGATION_AUTO) {
> 
> I don't think so.  That would stop us from printing the message about the
> system being vulnerable at the end of the function.

Sorry it wasn't clear, my comment was not about the return, but about
simplifying the check:

diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c
index 22fdaaac2d21..e8c481c7a590 100644
--- a/arch/x86/kernel/cpu/bugs.c
+++ b/arch/x86/kernel/cpu/bugs.c
@@ -1115,7 +1115,7 @@ static void __init spectre_v1_select_mitigation(void)
 
 static void __init spectre_v1_apply_mitigation(void)
 {
-	if (!boot_cpu_has_bug(X86_BUG_SPECTRE_V1) || cpu_mitigations_off())
+	if (spectre_v1_mitigation == SPECTRE_V1_MITIGATION_NONE)
 		return;
 
 	if (spectre_v1_mitigation == SPECTRE_V1_MITIGATION_AUTO) {

Since we already set spectre_v1_mitigation to SPECTRE_V1_MITIGATION_NONE
for that exact condition.

> We should only not print the message I believe if the CPU is actually not
> vulnerable or mitigations are globally disabled.  Although now I realize
> my patches may not be suppressing the print statements always if
> cpu_mitigations_off(), so I need to go and fix that.