[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [PATCH v2 2/2] x86/spec: adjust logic to logic that elides lfence
It's currently too restrictive by just checking whether there's a BHB clearing sequence selected. It should instead check whether BHB clearing is used on entry from PV or HVM specifically. Switch to use opt_bhb_entry_{pv,hvm} instead, and then remove cpu_has_bhb_seq since it no longer has any users. Reported-by: Jan Beulich <jbeulich@xxxxxxxx> Fixes: 954c983abcee ('x86/spec-ctrl: Software BHB-clearing sequences') Signed-off-by: Roger Pau Monné <roger.pau@xxxxxxxxxx> --- Changes since v1: - New in this version. There (possibly) still a bit of overhead for dom0 if BHB clearing is not used for dom0, as Xen would still add the lfence if domUs require it. --- xen/arch/x86/include/asm/cpufeature.h | 3 --- xen/arch/x86/spec_ctrl.c | 6 +++--- 2 files changed, 3 insertions(+), 6 deletions(-) diff --git a/xen/arch/x86/include/asm/cpufeature.h b/xen/arch/x86/include/asm/cpufeature.h index 743f11f98940..9bc553681f4a 100644 --- a/xen/arch/x86/include/asm/cpufeature.h +++ b/xen/arch/x86/include/asm/cpufeature.h @@ -235,9 +235,6 @@ static inline bool boot_cpu_has(unsigned int feat) #define cpu_bug_fpu_ptrs boot_cpu_has(X86_BUG_FPU_PTRS) #define cpu_bug_null_seg boot_cpu_has(X86_BUG_NULL_SEG) -#define cpu_has_bhb_seq (boot_cpu_has(X86_SPEC_BHB_TSX) || \ - boot_cpu_has(X86_SPEC_BHB_LOOPS)) - enum _cache_type { CACHE_TYPE_NULL = 0, CACHE_TYPE_DATA = 1, diff --git a/xen/arch/x86/spec_ctrl.c b/xen/arch/x86/spec_ctrl.c index 1e831c1c108e..40f6ae017010 100644 --- a/xen/arch/x86/spec_ctrl.c +++ b/xen/arch/x86/spec_ctrl.c @@ -2328,7 +2328,7 @@ void __init init_speculation_mitigations(void) * unconditional WRMSR. If we do have it, or we're not using any * prior conditional block, then it's safe to drop the LFENCE. */ - if ( !cpu_has_bhb_seq && + if ( !opt_bhb_entry_pv && (boot_cpu_has(X86_FEATURE_SC_MSR_PV) || !boot_cpu_has(X86_FEATURE_IBPB_ENTRY_PV)) ) setup_force_cpu_cap(X86_SPEC_NO_LFENCE_ENTRY_PV); @@ -2344,7 +2344,7 @@ void __init init_speculation_mitigations(void) * active in the block that is skipped when interrupting guest * context, then it's safe to drop the LFENCE. */ - if ( !cpu_has_bhb_seq && + if ( !opt_bhb_entry_pv && (boot_cpu_has(X86_FEATURE_SC_MSR_PV) || (!boot_cpu_has(X86_FEATURE_IBPB_ENTRY_PV) && !boot_cpu_has(X86_FEATURE_SC_RSB_PV))) ) @@ -2356,7 +2356,7 @@ void __init init_speculation_mitigations(void) * A BHB sequence, if used, is the only conditional action, so if we * don't have it, we don't need the safety LFENCE. */ - if ( !cpu_has_bhb_seq ) + if ( !opt_bhb_entry_hvm ) setup_force_cpu_cap(X86_SPEC_NO_LFENCE_ENTRY_VMX); } -- 2.44.0
|
Lists.xenproject.org is hosted with RackSpace, monitoring our |