| Message ID | 20240429151625.977884-4-andrew.cooper3@citrix.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | x86: AMD CPUID handling improvements | expand |
On 29.04.2024 17:16, Andrew Cooper wrote: > Now that the SVM feature leaf has been included in normal feature handling, it > is available early enough for init_speculation_mitigations() to use. > > No functional change. > > Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com> Reviewed-by: Jan Beulich <jbeulich@suse.com> even if ... > --- a/xen/arch/x86/include/asm/cpufeature.h > +++ b/xen/arch/x86/include/asm/cpufeature.h > @@ -217,6 +217,9 @@ static inline bool boot_cpu_has(unsigned int feat) > #define cpu_has_rfds_no boot_cpu_has(X86_FEATURE_RFDS_NO) > #define cpu_has_rfds_clear boot_cpu_has(X86_FEATURE_RFDS_CLEAR) > > +/* CPUID level 0x8000000a.edx */ > +#define cpu_has_v_spec_ctrl boot_cpu_has(X86_FEATURE_V_SPEC_CTRL) ... the names here were to change (see comment on the earlier patch). In fact ... > @@ -1896,12 +1895,8 @@ void __init init_speculation_mitigations(void) > * > * No need for SCF_ist_sc_msr because Xen's value is restored > * atomically WRT NMIs in the VMExit path. > - * > - * TODO: Adjust cpu_has_svm_spec_ctrl to be usable earlier on boot. > */ > - if ( opt_msr_sc_hvm && > - (boot_cpu_data.extended_cpuid_level >= 0x8000000aU) && > - (cpuid_edx(0x8000000aU) & (1u << SVM_FEATURE_SPEC_CTRL)) ) > + if ( opt_msr_sc_hvm && cpu_has_v_spec_ctrl ) > setup_force_cpu_cap(X86_FEATURE_SC_MSR_HVM); > } ... the use here demonstrates my earlier point quite well: It being AMD's feature is completely invisible here when not considering the code being replaced. But yes, when looking at the entire comment / block, it still is visible. Jan
diff --git a/xen/arch/x86/include/asm/cpufeature.h b/xen/arch/x86/include/asm/cpufeature.h index 9bc553681f4a..77cfd900cb56 100644 --- a/xen/arch/x86/include/asm/cpufeature.h +++ b/xen/arch/x86/include/asm/cpufeature.h @@ -217,6 +217,9 @@ static inline bool boot_cpu_has(unsigned int feat) #define cpu_has_rfds_no boot_cpu_has(X86_FEATURE_RFDS_NO) #define cpu_has_rfds_clear boot_cpu_has(X86_FEATURE_RFDS_CLEAR) +/* CPUID level 0x8000000a.edx */ +#define cpu_has_v_spec_ctrl boot_cpu_has(X86_FEATURE_V_SPEC_CTRL) + /* Synthesized. */ #define cpu_has_arch_perfmon boot_cpu_has(X86_FEATURE_ARCH_PERFMON) #define cpu_has_cpuid_faulting boot_cpu_has(X86_FEATURE_CPUID_FAULTING) diff --git a/xen/arch/x86/spec_ctrl.c b/xen/arch/x86/spec_ctrl.c index 40f6ae017010..0bda9d01def5 100644 --- a/xen/arch/x86/spec_ctrl.c +++ b/xen/arch/x86/spec_ctrl.c @@ -11,7 +11,6 @@ #include <xen/warning.h> #include <asm/amd.h> -#include <asm/hvm/svm/svm.h> #include <asm/intel-family.h> #include <asm/microcode.h> #include <asm/msr.h> @@ -1896,12 +1895,8 @@ void __init init_speculation_mitigations(void) * * No need for SCF_ist_sc_msr because Xen's value is restored * atomically WRT NMIs in the VMExit path. - * - * TODO: Adjust cpu_has_svm_spec_ctrl to be usable earlier on boot. */ - if ( opt_msr_sc_hvm && - (boot_cpu_data.extended_cpuid_level >= 0x8000000aU) && - (cpuid_edx(0x8000000aU) & (1u << SVM_FEATURE_SPEC_CTRL)) ) + if ( opt_msr_sc_hvm && cpu_has_v_spec_ctrl ) setup_force_cpu_cap(X86_FEATURE_SC_MSR_HVM); }
Now that the SVM feature leaf has been included in normal feature handling, it is available early enough for init_speculation_mitigations() to use. No functional change. Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com> --- CC: Jan Beulich <JBeulich@suse.com> CC: Roger Pau Monné <roger.pau@citrix.com> CC: Stefano Stabellini <sstabellini@kernel.org> CC: Xenia Ragiadakou <xenia.ragiadakou@amd.com> CC: Sergiy Kibrik <Sergiy_Kibrik@epam.com> CC: George Dunlap <george.dunlap@citrix.com> CC: Andrei Semenov <andrei.semenov@vates.fr> CC: Vaishali Thakkar <vaishali.thakkar@vates.tech> --- xen/arch/x86/include/asm/cpufeature.h | 3 +++ xen/arch/x86/spec_ctrl.c | 7 +------ 2 files changed, 4 insertions(+), 6 deletions(-)