Re: [PATCH 1/3] cpu/bugs: Fix selecting Automatic IBRS using spectre_v2=eibrs
From: Jim Mattson
Date: Tue Feb 24 2026 - 13:30:43 EST
On Tue, Feb 24, 2026 at 10:23 AM Dave Hansen <dave.hansen@xxxxxxxxx> wrote:
>
> On 2/24/26 10:01, Kim Phillips wrote:
> > @@ -2136,7 +2136,8 @@ static void __init spectre_v2_select_mitigation(void)
> > if ((spectre_v2_cmd == SPECTRE_V2_CMD_EIBRS ||
> > spectre_v2_cmd == SPECTRE_V2_CMD_EIBRS_LFENCE ||
> > spectre_v2_cmd == SPECTRE_V2_CMD_EIBRS_RETPOLINE) &&
> > - !boot_cpu_has(X86_FEATURE_IBRS_ENHANCED)) {
> > + !(boot_cpu_has(X86_FEATURE_IBRS_ENHANCED) ||
> > + boot_cpu_has(X86_FEATURE_AUTOIBRS))) {
> > pr_err("EIBRS selected but CPU doesn't have Enhanced or Automatic IBRS. Switching to AUTO select\n");
> > spectre_v2_cmd = SPECTRE_V2_CMD_AUTO;
> > }
>
> Didn't we agree to just use the "Intel feature" name?
Aren't they quite different? IIRC, IBRS_ENHANCED protects host
userspace from guest indirect branch steering (i.e. VMSCAPE style
attacks), but AUTOIBRS does not.