| Message ID | 20220316091315.19281-3-jane.malalane@citrix.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | xen: Report and use hardware APIC virtualization capabilities | expand |
On Wed, Mar 16, 2022 at 09:13:15AM +0000, Jane Malalane wrote: > Introduce a new per-domain creation x86 specific flag to > select whether hardware assisted virtualization should be used for > x{2}APIC. > > A per-domain option is added to xl in order to select the usage of > x{2}APIC hardware assisted virtualization, as well as a global > configuration option. > > Having all APIC interaction exit to Xen for emulation is slow and can > induce much overhead. Hardware can speed up x{2}APIC by decoding the > APIC access and providing a VM exit with a more specific exit reason > than a regular EPT fault or by altogether avoiding a VM exit. > > On the other hand, being able to disable x{2}APIC hardware assisted > virtualization can be useful for testing and debugging purposes. > > Note: vmx_install_vlapic_mapping doesn't require modifications > regardless of whether the guest has "Virtualize APIC accesses" enabled > or not, i.e., setting the APIC_ACCESS_ADDR VMCS field is fine so long > as virtualize_apic_accesses is supported by the CPU. Have you tested migration of guests with this patch applied? We need to be careful so that a guest that doesn't have assisted_x{2}apic set in the config file can be migrated between hosts that have different support for hardware assisted x{2}APIC virtualization. Ie: we need to make sure the selection of arch_x86.assisted_x{2}apic is only present in the migration stream when explicitly set in the configuration file. > diff --git a/tools/libs/light/libxl_x86.c b/tools/libs/light/libxl_x86.c > index e0a06ecfe3..46d4de22d1 100644 > --- a/tools/libs/light/libxl_x86.c > +++ b/tools/libs/light/libxl_x86.c > @@ -23,6 +23,15 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc, > if (libxl_defbool_val(d_config->b_info.arch_x86.msr_relaxed)) > config->arch.misc_flags |= XEN_X86_MSR_RELAXED; > > + if (d_config->c_info.type != LIBXL_DOMAIN_TYPE_PV) > + { > + if (libxl_defbool_val(d_config->b_info.arch_x86.assisted_xapic)) > + config->arch.misc_flags |= XEN_X86_ASSISTED_XAPIC; > + > + if (libxl_defbool_val(d_config->b_info.arch_x86.assisted_x2apic)) > + config->arch.misc_flags |= XEN_X86_ASSISTED_X2APIC; > + } > + > return 0; > } > > @@ -819,11 +828,26 @@ void libxl__arch_domain_create_info_setdefault(libxl__gc *gc, > { > } > > -void libxl__arch_domain_build_info_setdefault(libxl__gc *gc, > - libxl_domain_build_info *b_info) > +int libxl__arch_domain_build_info_setdefault(libxl__gc *gc, > + libxl_domain_build_info *b_info, > + const libxl_physinfo *physinfo) > { > libxl_defbool_setdefault(&b_info->acpi, true); > libxl_defbool_setdefault(&b_info->arch_x86.msr_relaxed, false); > + > + if (b_info->type != LIBXL_DOMAIN_TYPE_PV) { > + libxl_defbool_setdefault(&b_info->arch_x86.assisted_xapic, > + physinfo->cap_assisted_xapic); > + libxl_defbool_setdefault(&b_info->arch_x86.assisted_x2apic, > + physinfo->cap_assisted_x2apic); Nit: the split lines would better be adjusted to match the indentation of the first parameter. libxl_defbool_setdefault(&b_info->arch_x86.assisted_xapic, physinfo->cap_assisted_xapic); > + } > + else if (!libxl_defbool_is_default(b_info->arch_x86.assisted_xapic) || > + !libxl_defbool_is_default(b_info->arch_x86.assisted_x2apic)) { > + LOG(ERROR, "Interrupt Controller Virtualization not supported for PV"); > + return ERROR_INVAL; > + } > + > + return 0; > } > > int libxl__arch_passthrough_mode_setdefault(libxl__gc *gc, > diff --git a/tools/ocaml/libs/xc/xenctrl.ml b/tools/ocaml/libs/xc/xenctrl.ml > index 712456e098..32f3028828 100644 > --- a/tools/ocaml/libs/xc/xenctrl.ml > +++ b/tools/ocaml/libs/xc/xenctrl.ml > @@ -50,6 +50,8 @@ type x86_arch_emulation_flags = > > type x86_arch_misc_flags = > | X86_MSR_RELAXED > + | X86_ASSISTED_XAPIC > + | X86_ASSISTED_X2APIC > > type xen_x86_arch_domainconfig = > { > diff --git a/tools/ocaml/libs/xc/xenctrl.mli b/tools/ocaml/libs/xc/xenctrl.mli > index b034434f68..d0fcbc8866 100644 > --- a/tools/ocaml/libs/xc/xenctrl.mli > +++ b/tools/ocaml/libs/xc/xenctrl.mli > @@ -44,6 +44,8 @@ type x86_arch_emulation_flags = > > type x86_arch_misc_flags = > | X86_MSR_RELAXED > + | X86_ASSISTED_XAPIC > + | X86_ASSISTED_X2APIC > > type xen_x86_arch_domainconfig = { > emulation_flags: x86_arch_emulation_flags list; > diff --git a/tools/ocaml/libs/xc/xenctrl_stubs.c b/tools/ocaml/libs/xc/xenctrl_stubs.c > index 7e9c32ad1b..5df8aaa58f 100644 > --- a/tools/ocaml/libs/xc/xenctrl_stubs.c > +++ b/tools/ocaml/libs/xc/xenctrl_stubs.c > @@ -239,7 +239,7 @@ CAMLprim value stub_xc_domain_create(value xch, value wanted_domid, value config > > cfg.arch.misc_flags = ocaml_list_to_c_bitmap > /* ! x86_arch_misc_flags X86_ none */ > - /* ! XEN_X86_ XEN_X86_MSR_RELAXED all */ > + /* ! XEN_X86_ XEN_X86_ASSISTED_X2APIC max */ In order to simplify adding new options we would generally introduce a: #define XEN_X86_MISC_FLAGS_MAX XEN_X86_ASSISTED_X2APIC In include/public/arch-x86/xen.h so that adding new flags doesn't require changing the Ocaml code here. This wasn't done before because there was a single flag. > (VAL_MISC_FLAGS); > > #undef VAL_MISC_FLAGS > diff --git a/tools/xl/xl.c b/tools/xl/xl.c > index 2d1ec18ea3..31eb223309 100644 > --- a/tools/xl/xl.c > +++ b/tools/xl/xl.c > @@ -57,6 +57,8 @@ int max_grant_frames = -1; > int max_maptrack_frames = -1; > int max_grant_version = LIBXL_MAX_GRANT_DEFAULT; > libxl_domid domid_policy = INVALID_DOMID; > +int assisted_xapic = -1; > +int assisted_x2apic = -1; > > xentoollog_level minmsglevel = minmsglevel_default; > > @@ -201,6 +203,12 @@ static void parse_global_config(const char *configfile, > if (!xlu_cfg_get_long (config, "claim_mode", &l, 0)) > claim_mode = l; > > + if (!xlu_cfg_get_long (config, "assisted_xapic", &l, 0)) > + assisted_xapic = l; > + > + if (!xlu_cfg_get_long (config, "assisted_x2apic", &l, 0)) > + assisted_x2apic = l; > + > xlu_cfg_replace_string (config, "remus.default.netbufscript", > &default_remus_netbufscript, 0); > xlu_cfg_replace_string (config, "colo.default.proxyscript", > diff --git a/tools/xl/xl.h b/tools/xl/xl.h > index c5c4bedbdd..528deb3feb 100644 > --- a/tools/xl/xl.h > +++ b/tools/xl/xl.h > @@ -286,6 +286,8 @@ extern libxl_bitmap global_vm_affinity_mask; > extern libxl_bitmap global_hvm_affinity_mask; > extern libxl_bitmap global_pv_affinity_mask; > extern libxl_domid domid_policy; > +extern int assisted_xapic; > +extern int assisted_x2apic; > > enum output_format { > OUTPUT_FORMAT_JSON, > diff --git a/tools/xl/xl_parse.c b/tools/xl/xl_parse.c > index 117fcdcb2b..f118dc7e97 100644 > --- a/tools/xl/xl_parse.c > +++ b/tools/xl/xl_parse.c > @@ -2761,6 +2761,24 @@ skip_usbdev: > > xlu_cfg_get_defbool(config, "vpmu", &b_info->vpmu, 0); > > + if (b_info->type != LIBXL_DOMAIN_TYPE_PV) { > + e = xlu_cfg_get_long(config, "assisted_xapic", &l , 0); > + if ((e == ESRCH && assisted_xapic != -1)) /* use global default if present */ > + libxl_defbool_set(&b_info->arch_x86.assisted_xapic, assisted_xapic); > + else if (!e) > + libxl_defbool_set(&b_info->arch_x86.assisted_xapic, l); > + else > + exit(1); > + > + e = xlu_cfg_get_long(config, "assisted_x2apic", &l, 0); > + if ((e == ESRCH && assisted_x2apic != -1)) /* use global default if present */ > + libxl_defbool_set(&b_info->arch_x86.assisted_x2apic, assisted_x2apic); > + else if (!e) > + libxl_defbool_set(&b_info->arch_x86.assisted_x2apic, l); > + else > + exit(1); Indentation seems wrong in the line above. Thanks, Roger.
On Thu, Mar 17, 2022 at 09:58:15AM +0100, Roger Pau Monné wrote: > On Wed, Mar 16, 2022 at 09:13:15AM +0000, Jane Malalane wrote: > > Introduce a new per-domain creation x86 specific flag to > > select whether hardware assisted virtualization should be used for > > x{2}APIC. > > > > A per-domain option is added to xl in order to select the usage of > > x{2}APIC hardware assisted virtualization, as well as a global > > configuration option. > > > > Having all APIC interaction exit to Xen for emulation is slow and can > > induce much overhead. Hardware can speed up x{2}APIC by decoding the > > APIC access and providing a VM exit with a more specific exit reason > > than a regular EPT fault or by altogether avoiding a VM exit. > > > > On the other hand, being able to disable x{2}APIC hardware assisted > > virtualization can be useful for testing and debugging purposes. > > > > Note: vmx_install_vlapic_mapping doesn't require modifications > > regardless of whether the guest has "Virtualize APIC accesses" enabled > > or not, i.e., setting the APIC_ACCESS_ADDR VMCS field is fine so long > > as virtualize_apic_accesses is supported by the CPU. > > Have you tested migration of guests with this patch applied? > > We need to be careful so that a guest that doesn't have > assisted_x{2}apic set in the config file can be migrated between hosts > that have different support for hardware assisted x{2}APIC > virtualization. > > Ie: we need to make sure the selection of arch_x86.assisted_x{2}apic > is only present in the migration stream when explicitly set in the > configuration file. I've tested this myself, and the behavior seems to be correct. The selection is only present on the stream when explicitly set by the user. > > diff --git a/tools/xl/xl.h b/tools/xl/xl.h > > index c5c4bedbdd..528deb3feb 100644 > > --- a/tools/xl/xl.h > > +++ b/tools/xl/xl.h > > @@ -286,6 +286,8 @@ extern libxl_bitmap global_vm_affinity_mask; > > extern libxl_bitmap global_hvm_affinity_mask; > > extern libxl_bitmap global_pv_affinity_mask; > > extern libxl_domid domid_policy; > > +extern int assisted_xapic; > > +extern int assisted_x2apic; > > > > enum output_format { > > OUTPUT_FORMAT_JSON, > > diff --git a/tools/xl/xl_parse.c b/tools/xl/xl_parse.c > > index 117fcdcb2b..f118dc7e97 100644 > > --- a/tools/xl/xl_parse.c > > +++ b/tools/xl/xl_parse.c > > @@ -2761,6 +2761,24 @@ skip_usbdev: > > > > xlu_cfg_get_defbool(config, "vpmu", &b_info->vpmu, 0); > > > > + if (b_info->type != LIBXL_DOMAIN_TYPE_PV) { > > + e = xlu_cfg_get_long(config, "assisted_xapic", &l , 0); > > + if ((e == ESRCH && assisted_xapic != -1)) /* use global default if present */ > > + libxl_defbool_set(&b_info->arch_x86.assisted_xapic, assisted_xapic); > > + else if (!e) > > + libxl_defbool_set(&b_info->arch_x86.assisted_xapic, l); > > + else > > + exit(1); > > + > > + e = xlu_cfg_get_long(config, "assisted_x2apic", &l, 0); > > + if ((e == ESRCH && assisted_x2apic != -1)) /* use global default if present */ > > + libxl_defbool_set(&b_info->arch_x86.assisted_x2apic, assisted_x2apic); > > + else if (!e) > > + libxl_defbool_set(&b_info->arch_x86.assisted_x2apic, l); > > + else > > + exit(1); > > Indentation seems wrong in the line above. The chunk above is also logically wrong, because it will exit(1) if no assisted_x{2}apic option is provided on either the guest or the global config files. I think: e = xlu_cfg_get_long(config, "assisted_xapic", &l , 0); if (!e) libxl_defbool_set(&b_info->arch_x86.assisted_xapic, l); else if (e != ESRCH) exit(1); else if (assisted_xapic != -1) /* use global default if present */ libxl_defbool_set(&b_info->arch_x86.assisted_xapic, assisted_xapic); e = xlu_cfg_get_long(config, "assisted_x2apic", &l, 0); if (!e) libxl_defbool_set(&b_info->arch_x86.assisted_x2apic, l); else if (e != ESRCH) exit(1); else if (assisted_x2apic != -1) /* use global default if present */ libxl_defbool_set(&b_info->arch_x86.assisted_x2apic, assisted_x2apic); Is better. Thanks, Roger.
diff --git a/docs/man/xl.cfg.5.pod.in b/docs/man/xl.cfg.5.pod.in index b98d161398..6d98d73d76 100644 --- a/docs/man/xl.cfg.5.pod.in +++ b/docs/man/xl.cfg.5.pod.in @@ -1862,6 +1862,21 @@ firmware tables when using certain older guest Operating Systems. These tables have been superseded by newer constructs within the ACPI tables. +=item B<assisted_xapic=BOOLEAN> + +B<(x86 only)> Enables or disables hardware assisted virtualization for +xAPIC. With this option enabled, a memory-mapped APIC access will be +decoded by hardware and either issue a more specific VM exit than just +a p2m fault, or altogether avoid a VM exit. The +default is settable via L<xl.conf(5)>. + +=item B<assisted_x2apic=BOOLEAN> + +B<(x86 only)> Enables or disables hardware assisted virtualization for +x2APIC. With this option enabled, certain accesses to MSR APIC +registers will avoid a VM exit into the hypervisor. The default is +settable via L<xl.conf(5)>. + =item B<nx=BOOLEAN> B<(x86 only)> Hides or exposes the No-eXecute capability. This allows a guest diff --git a/docs/man/xl.conf.5.pod.in b/docs/man/xl.conf.5.pod.in index df20c08137..95d136d1ea 100644 --- a/docs/man/xl.conf.5.pod.in +++ b/docs/man/xl.conf.5.pod.in @@ -107,6 +107,18 @@ Sets the default value for the C<max_grant_version> domain config value. Default: maximum grant version supported by the hypervisor. +=item B<assisted_xapic=BOOLEAN> + +If enabled, domains will use xAPIC hardware assisted virtualization by default. + +Default: enabled if supported. + +=item B<assisted_x2apic=BOOLEAN> + +If enabled, domains will use x2APIC hardware assisted virtualization by default. + +Default: enabled if supported. + =item B<vif.default.script="PATH"> Configures the default hotplug script used by virtual network devices. diff --git a/tools/golang/xenlight/helpers.gen.go b/tools/golang/xenlight/helpers.gen.go index dd4e6c9f14..dece545ee0 100644 --- a/tools/golang/xenlight/helpers.gen.go +++ b/tools/golang/xenlight/helpers.gen.go @@ -1120,6 +1120,12 @@ x.ArchArm.Vuart = VuartType(xc.arch_arm.vuart) if err := x.ArchX86.MsrRelaxed.fromC(&xc.arch_x86.msr_relaxed);err != nil { return fmt.Errorf("converting field ArchX86.MsrRelaxed: %v", err) } +if err := x.ArchX86.AssistedXapic.fromC(&xc.arch_x86.assisted_xapic);err != nil { +return fmt.Errorf("converting field ArchX86.AssistedXapic: %v", err) +} +if err := x.ArchX86.AssistedX2Apic.fromC(&xc.arch_x86.assisted_x2apic);err != nil { +return fmt.Errorf("converting field ArchX86.AssistedX2Apic: %v", err) +} x.Altp2M = Altp2MMode(xc.altp2m) x.VmtraceBufKb = int(xc.vmtrace_buf_kb) if err := x.Vpmu.fromC(&xc.vpmu);err != nil { @@ -1605,6 +1611,12 @@ xc.arch_arm.vuart = C.libxl_vuart_type(x.ArchArm.Vuart) if err := x.ArchX86.MsrRelaxed.toC(&xc.arch_x86.msr_relaxed); err != nil { return fmt.Errorf("converting field ArchX86.MsrRelaxed: %v", err) } +if err := x.ArchX86.AssistedXapic.toC(&xc.arch_x86.assisted_xapic); err != nil { +return fmt.Errorf("converting field ArchX86.AssistedXapic: %v", err) +} +if err := x.ArchX86.AssistedX2Apic.toC(&xc.arch_x86.assisted_x2apic); err != nil { +return fmt.Errorf("converting field ArchX86.AssistedX2Apic: %v", err) +} xc.altp2m = C.libxl_altp2m_mode(x.Altp2M) xc.vmtrace_buf_kb = C.int(x.VmtraceBufKb) if err := x.Vpmu.toC(&xc.vpmu); err != nil { diff --git a/tools/golang/xenlight/types.gen.go b/tools/golang/xenlight/types.gen.go index 87be46c745..253c9ad93d 100644 --- a/tools/golang/xenlight/types.gen.go +++ b/tools/golang/xenlight/types.gen.go @@ -520,6 +520,8 @@ Vuart VuartType } ArchX86 struct { MsrRelaxed Defbool +AssistedXapic Defbool +AssistedX2Apic Defbool } Altp2M Altp2MMode VmtraceBufKb int diff --git a/tools/include/libxl.h b/tools/include/libxl.h index 94e6355822..cdcccd6d01 100644 --- a/tools/include/libxl.h +++ b/tools/include/libxl.h @@ -535,6 +535,13 @@ #define LIBXL_HAVE_PHYSINFO_ASSISTED_APIC 1 /* + * LIBXL_HAVE_ASSISTED_APIC indicates that libxl_domain_build_info has + * assisted_xapic and assisted_x2apic fields for enabling hardware + * assisted virtualization for x{2}apic per domain. + */ +#define LIBXL_HAVE_ASSISTED_APIC 1 + +/* * libxl ABI compatibility * * The only guarantee which libxl makes regarding ABI compatibility diff --git a/tools/libs/light/libxl_arch.h b/tools/libs/light/libxl_arch.h index 207ceac6a1..03b89929e6 100644 --- a/tools/libs/light/libxl_arch.h +++ b/tools/libs/light/libxl_arch.h @@ -71,8 +71,9 @@ void libxl__arch_domain_create_info_setdefault(libxl__gc *gc, libxl_domain_create_info *c_info); _hidden -void libxl__arch_domain_build_info_setdefault(libxl__gc *gc, - libxl_domain_build_info *b_info); +int libxl__arch_domain_build_info_setdefault(libxl__gc *gc, + libxl_domain_build_info *b_info, + const libxl_physinfo *physinfo); _hidden int libxl__arch_passthrough_mode_setdefault(libxl__gc *gc, diff --git a/tools/libs/light/libxl_arm.c b/tools/libs/light/libxl_arm.c index 39fdca1b49..7dee2afd4b 100644 --- a/tools/libs/light/libxl_arm.c +++ b/tools/libs/light/libxl_arm.c @@ -1384,14 +1384,15 @@ void libxl__arch_domain_create_info_setdefault(libxl__gc *gc, } } -void libxl__arch_domain_build_info_setdefault(libxl__gc *gc, - libxl_domain_build_info *b_info) +int libxl__arch_domain_build_info_setdefault(libxl__gc *gc, + libxl_domain_build_info *b_info, + const libxl_physinfo *physinfo) { /* ACPI is disabled by default */ libxl_defbool_setdefault(&b_info->acpi, false); if (b_info->type != LIBXL_DOMAIN_TYPE_PV) - return; + return 0; LOG(DEBUG, "Converting build_info to PVH"); @@ -1399,6 +1400,8 @@ void libxl__arch_domain_build_info_setdefault(libxl__gc *gc, memset(&b_info->u, '\0', sizeof(b_info->u)); b_info->type = LIBXL_DOMAIN_TYPE_INVALID; libxl_domain_build_info_init_type(b_info, LIBXL_DOMAIN_TYPE_PVH); + + return 0; } int libxl__arch_passthrough_mode_setdefault(libxl__gc *gc, diff --git a/tools/libs/light/libxl_create.c b/tools/libs/light/libxl_create.c index 15ed021f41..88d08d7277 100644 --- a/tools/libs/light/libxl_create.c +++ b/tools/libs/light/libxl_create.c @@ -75,6 +75,7 @@ int libxl__domain_build_info_setdefault(libxl__gc *gc, libxl_domain_build_info *b_info) { int i, rc; + libxl_physinfo info; if (b_info->type != LIBXL_DOMAIN_TYPE_HVM && b_info->type != LIBXL_DOMAIN_TYPE_PV && @@ -264,7 +265,18 @@ int libxl__domain_build_info_setdefault(libxl__gc *gc, if (!b_info->event_channels) b_info->event_channels = 1023; - libxl__arch_domain_build_info_setdefault(gc, b_info); + rc = libxl_get_physinfo(CTX, &info); + if (rc) { + LOG(ERROR, "failed to get hypervisor info"); + return rc; + } + + rc = libxl__arch_domain_build_info_setdefault(gc, b_info, &info); + if (rc) { + LOG(ERROR, "unable to set domain arch build info defaults"); + return rc; + } + libxl_defbool_setdefault(&b_info->dm_restrict, false); if (b_info->iommu_memkb == LIBXL_MEMKB_DEFAULT) @@ -457,14 +469,6 @@ int libxl__domain_build_info_setdefault(libxl__gc *gc, } if (b_info->max_grant_version == LIBXL_MAX_GRANT_DEFAULT) { - libxl_physinfo info; - - rc = libxl_get_physinfo(CTX, &info); - if (rc) { - LOG(ERROR, "failed to get hypervisor info"); - return rc; - } - if (info.cap_gnttab_v2) b_info->max_grant_version = 2; else if (info.cap_gnttab_v1) diff --git a/tools/libs/light/libxl_types.idl b/tools/libs/light/libxl_types.idl index 42ac6c357b..db5eb0a0b3 100644 --- a/tools/libs/light/libxl_types.idl +++ b/tools/libs/light/libxl_types.idl @@ -648,6 +648,8 @@ libxl_domain_build_info = Struct("domain_build_info",[ ("vuart", libxl_vuart_type), ])), ("arch_x86", Struct(None, [("msr_relaxed", libxl_defbool), + ("assisted_xapic", libxl_defbool), + ("assisted_x2apic", libxl_defbool), ])), # Alternate p2m is not bound to any architecture or guest type, as it is # supported by x86 HVM and ARM support is planned. diff --git a/tools/libs/light/libxl_x86.c b/tools/libs/light/libxl_x86.c index e0a06ecfe3..46d4de22d1 100644 --- a/tools/libs/light/libxl_x86.c +++ b/tools/libs/light/libxl_x86.c @@ -23,6 +23,15 @@ int libxl__arch_domain_prepare_config(libxl__gc *gc, if (libxl_defbool_val(d_config->b_info.arch_x86.msr_relaxed)) config->arch.misc_flags |= XEN_X86_MSR_RELAXED; + if (d_config->c_info.type != LIBXL_DOMAIN_TYPE_PV) + { + if (libxl_defbool_val(d_config->b_info.arch_x86.assisted_xapic)) + config->arch.misc_flags |= XEN_X86_ASSISTED_XAPIC; + + if (libxl_defbool_val(d_config->b_info.arch_x86.assisted_x2apic)) + config->arch.misc_flags |= XEN_X86_ASSISTED_X2APIC; + } + return 0; } @@ -819,11 +828,26 @@ void libxl__arch_domain_create_info_setdefault(libxl__gc *gc, { } -void libxl__arch_domain_build_info_setdefault(libxl__gc *gc, - libxl_domain_build_info *b_info) +int libxl__arch_domain_build_info_setdefault(libxl__gc *gc, + libxl_domain_build_info *b_info, + const libxl_physinfo *physinfo) { libxl_defbool_setdefault(&b_info->acpi, true); libxl_defbool_setdefault(&b_info->arch_x86.msr_relaxed, false); + + if (b_info->type != LIBXL_DOMAIN_TYPE_PV) { + libxl_defbool_setdefault(&b_info->arch_x86.assisted_xapic, + physinfo->cap_assisted_xapic); + libxl_defbool_setdefault(&b_info->arch_x86.assisted_x2apic, + physinfo->cap_assisted_x2apic); + } + else if (!libxl_defbool_is_default(b_info->arch_x86.assisted_xapic) || + !libxl_defbool_is_default(b_info->arch_x86.assisted_x2apic)) { + LOG(ERROR, "Interrupt Controller Virtualization not supported for PV"); + return ERROR_INVAL; + } + + return 0; } int libxl__arch_passthrough_mode_setdefault(libxl__gc *gc, diff --git a/tools/ocaml/libs/xc/xenctrl.ml b/tools/ocaml/libs/xc/xenctrl.ml index 712456e098..32f3028828 100644 --- a/tools/ocaml/libs/xc/xenctrl.ml +++ b/tools/ocaml/libs/xc/xenctrl.ml @@ -50,6 +50,8 @@ type x86_arch_emulation_flags = type x86_arch_misc_flags = | X86_MSR_RELAXED + | X86_ASSISTED_XAPIC + | X86_ASSISTED_X2APIC type xen_x86_arch_domainconfig = { diff --git a/tools/ocaml/libs/xc/xenctrl.mli b/tools/ocaml/libs/xc/xenctrl.mli index b034434f68..d0fcbc8866 100644 --- a/tools/ocaml/libs/xc/xenctrl.mli +++ b/tools/ocaml/libs/xc/xenctrl.mli @@ -44,6 +44,8 @@ type x86_arch_emulation_flags = type x86_arch_misc_flags = | X86_MSR_RELAXED + | X86_ASSISTED_XAPIC + | X86_ASSISTED_X2APIC type xen_x86_arch_domainconfig = { emulation_flags: x86_arch_emulation_flags list; diff --git a/tools/ocaml/libs/xc/xenctrl_stubs.c b/tools/ocaml/libs/xc/xenctrl_stubs.c index 7e9c32ad1b..5df8aaa58f 100644 --- a/tools/ocaml/libs/xc/xenctrl_stubs.c +++ b/tools/ocaml/libs/xc/xenctrl_stubs.c @@ -239,7 +239,7 @@ CAMLprim value stub_xc_domain_create(value xch, value wanted_domid, value config cfg.arch.misc_flags = ocaml_list_to_c_bitmap /* ! x86_arch_misc_flags X86_ none */ - /* ! XEN_X86_ XEN_X86_MSR_RELAXED all */ + /* ! XEN_X86_ XEN_X86_ASSISTED_X2APIC max */ (VAL_MISC_FLAGS); #undef VAL_MISC_FLAGS diff --git a/tools/xl/xl.c b/tools/xl/xl.c index 2d1ec18ea3..31eb223309 100644 --- a/tools/xl/xl.c +++ b/tools/xl/xl.c @@ -57,6 +57,8 @@ int max_grant_frames = -1; int max_maptrack_frames = -1; int max_grant_version = LIBXL_MAX_GRANT_DEFAULT; libxl_domid domid_policy = INVALID_DOMID; +int assisted_xapic = -1; +int assisted_x2apic = -1; xentoollog_level minmsglevel = minmsglevel_default; @@ -201,6 +203,12 @@ static void parse_global_config(const char *configfile, if (!xlu_cfg_get_long (config, "claim_mode", &l, 0)) claim_mode = l; + if (!xlu_cfg_get_long (config, "assisted_xapic", &l, 0)) + assisted_xapic = l; + + if (!xlu_cfg_get_long (config, "assisted_x2apic", &l, 0)) + assisted_x2apic = l; + xlu_cfg_replace_string (config, "remus.default.netbufscript", &default_remus_netbufscript, 0); xlu_cfg_replace_string (config, "colo.default.proxyscript", diff --git a/tools/xl/xl.h b/tools/xl/xl.h index c5c4bedbdd..528deb3feb 100644 --- a/tools/xl/xl.h +++ b/tools/xl/xl.h @@ -286,6 +286,8 @@ extern libxl_bitmap global_vm_affinity_mask; extern libxl_bitmap global_hvm_affinity_mask; extern libxl_bitmap global_pv_affinity_mask; extern libxl_domid domid_policy; +extern int assisted_xapic; +extern int assisted_x2apic; enum output_format { OUTPUT_FORMAT_JSON, diff --git a/tools/xl/xl_parse.c b/tools/xl/xl_parse.c index 117fcdcb2b..f118dc7e97 100644 --- a/tools/xl/xl_parse.c +++ b/tools/xl/xl_parse.c @@ -2761,6 +2761,24 @@ skip_usbdev: xlu_cfg_get_defbool(config, "vpmu", &b_info->vpmu, 0); + if (b_info->type != LIBXL_DOMAIN_TYPE_PV) { + e = xlu_cfg_get_long(config, "assisted_xapic", &l , 0); + if ((e == ESRCH && assisted_xapic != -1)) /* use global default if present */ + libxl_defbool_set(&b_info->arch_x86.assisted_xapic, assisted_xapic); + else if (!e) + libxl_defbool_set(&b_info->arch_x86.assisted_xapic, l); + else + exit(1); + + e = xlu_cfg_get_long(config, "assisted_x2apic", &l, 0); + if ((e == ESRCH && assisted_x2apic != -1)) /* use global default if present */ + libxl_defbool_set(&b_info->arch_x86.assisted_x2apic, assisted_x2apic); + else if (!e) + libxl_defbool_set(&b_info->arch_x86.assisted_x2apic, l); + else + exit(1); + } + xlu_cfg_destroy(config); } diff --git a/xen/arch/x86/domain.c b/xen/arch/x86/domain.c index a5048ed654..279936a016 100644 --- a/xen/arch/x86/domain.c +++ b/xen/arch/x86/domain.c @@ -50,6 +50,7 @@ #include <asm/cpuidle.h> #include <asm/mpspec.h> #include <asm/ldt.h> +#include <asm/hvm/domain.h> #include <asm/hvm/hvm.h> #include <asm/hvm/nestedhvm.h> #include <asm/hvm/support.h> @@ -619,6 +620,8 @@ int arch_sanitise_domain_config(struct xen_domctl_createdomain *config) bool hvm = config->flags & XEN_DOMCTL_CDF_hvm; bool hap = config->flags & XEN_DOMCTL_CDF_hap; bool nested_virt = config->flags & XEN_DOMCTL_CDF_nested_virt; + bool assisted_xapic = config->arch.misc_flags & XEN_X86_ASSISTED_XAPIC; + bool assisted_x2apic = config->arch.misc_flags & XEN_X86_ASSISTED_X2APIC; unsigned int max_vcpus; if ( hvm ? !hvm_enabled : !IS_ENABLED(CONFIG_PV) ) @@ -685,13 +688,31 @@ int arch_sanitise_domain_config(struct xen_domctl_createdomain *config) } } - if ( config->arch.misc_flags & ~XEN_X86_MSR_RELAXED ) + if ( config->arch.misc_flags & ~(XEN_X86_MSR_RELAXED | + XEN_X86_ASSISTED_XAPIC | + XEN_X86_ASSISTED_X2APIC) ) { dprintk(XENLOG_INFO, "Invalid arch misc flags %#x\n", config->arch.misc_flags); return -EINVAL; } + if ( (assisted_xapic || assisted_x2apic) && !hvm ) + { + dprintk(XENLOG_INFO, + "Interrupt Controller Virtualization not supported for PV\n"); + return -EINVAL; + } + + if ( (assisted_xapic && !assisted_xapic_available) || + (assisted_x2apic && !assisted_x2apic_available) ) + { + dprintk(XENLOG_INFO, + "Hardware assisted x%sAPIC requested but not available\n", + assisted_xapic && !assisted_xapic_available ? "" : "2"); + return -ENODEV; + } + return 0; } @@ -864,6 +885,12 @@ int arch_domain_create(struct domain *d, d->arch.msr_relaxed = config->arch.misc_flags & XEN_X86_MSR_RELAXED; + d->arch.hvm.assisted_xapic = + config->arch.misc_flags & XEN_X86_ASSISTED_XAPIC; + + d->arch.hvm.assisted_x2apic = + config->arch.misc_flags & XEN_X86_ASSISTED_X2APIC; + return 0; fail: diff --git a/xen/arch/x86/hvm/vmx/vmcs.c b/xen/arch/x86/hvm/vmx/vmcs.c index 77ce0b2121..47c27740d3 100644 --- a/xen/arch/x86/hvm/vmx/vmcs.c +++ b/xen/arch/x86/hvm/vmx/vmcs.c @@ -1157,6 +1157,10 @@ static int construct_vmcs(struct vcpu *v) __vmwrite(PLE_WINDOW, ple_window); } + if ( !has_assisted_xapic(d) ) + v->arch.hvm.vmx.secondary_exec_control &= + ~SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES; + if ( cpu_has_vmx_secondary_exec_control ) __vmwrite(SECONDARY_VM_EXEC_CONTROL, v->arch.hvm.vmx.secondary_exec_control); diff --git a/xen/arch/x86/hvm/vmx/vmx.c b/xen/arch/x86/hvm/vmx/vmx.c index c075370f64..949ddd684c 100644 --- a/xen/arch/x86/hvm/vmx/vmx.c +++ b/xen/arch/x86/hvm/vmx/vmx.c @@ -3344,16 +3344,11 @@ static void vmx_install_vlapic_mapping(struct vcpu *v) void vmx_vlapic_msr_changed(struct vcpu *v) { - int virtualize_x2apic_mode; struct vlapic *vlapic = vcpu_vlapic(v); unsigned int msr; - virtualize_x2apic_mode = ( (cpu_has_vmx_apic_reg_virt || - cpu_has_vmx_virtual_intr_delivery) && - cpu_has_vmx_virtualize_x2apic_mode ); - - if ( !cpu_has_vmx_virtualize_apic_accesses && - !virtualize_x2apic_mode ) + if ( !has_assisted_xapic(v->domain) && + !has_assisted_x2apic(v->domain) ) return; vmx_vmcs_enter(v); @@ -3363,7 +3358,7 @@ void vmx_vlapic_msr_changed(struct vcpu *v) if ( !vlapic_hw_disabled(vlapic) && (vlapic_base_address(vlapic) == APIC_DEFAULT_PHYS_BASE) ) { - if ( virtualize_x2apic_mode && vlapic_x2apic_mode(vlapic) ) + if ( has_assisted_x2apic(v->domain) && vlapic_x2apic_mode(vlapic) ) { v->arch.hvm.vmx.secondary_exec_control |= SECONDARY_EXEC_VIRTUALIZE_X2APIC_MODE; @@ -3384,7 +3379,7 @@ void vmx_vlapic_msr_changed(struct vcpu *v) vmx_clear_msr_intercept(v, MSR_X2APIC_SELF, VMX_MSR_W); } } - else + else if ( has_assisted_xapic(v->domain) ) v->arch.hvm.vmx.secondary_exec_control |= SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES; } diff --git a/xen/arch/x86/include/asm/hvm/domain.h b/xen/arch/x86/include/asm/hvm/domain.h index 698455444e..92bf53483c 100644 --- a/xen/arch/x86/include/asm/hvm/domain.h +++ b/xen/arch/x86/include/asm/hvm/domain.h @@ -117,6 +117,12 @@ struct hvm_domain { bool is_s3_suspended; + /* xAPIC hardware assisted virtualization. */ + bool assisted_xapic; + + /* x2APIC hardware assisted virtualization. */ + bool assisted_x2apic; + /* hypervisor intercepted msix table */ struct list_head msixtbl_list; diff --git a/xen/arch/x86/include/asm/hvm/hvm.h b/xen/arch/x86/include/asm/hvm/hvm.h index e0d9348878..6ecbe22cc9 100644 --- a/xen/arch/x86/include/asm/hvm/hvm.h +++ b/xen/arch/x86/include/asm/hvm/hvm.h @@ -376,6 +376,9 @@ int hvm_get_param(struct domain *d, uint32_t index, uint64_t *value); extern bool assisted_xapic_available; extern bool assisted_x2apic_available; +#define has_assisted_xapic(d) ((d)->arch.hvm.assisted_xapic) +#define has_assisted_x2apic(d) ((d)->arch.hvm.assisted_x2apic) + #define hvm_get_guest_time(v) hvm_get_guest_time_fixed(v, 0) #define hvm_paging_enabled(v) \ @@ -878,6 +881,8 @@ static inline void hvm_set_reg(struct vcpu *v, unsigned int reg, uint64_t val) #define assisted_xapic_available false #define assisted_x2apic_available false +#define has_assisted_xapic(d) ((void)(d), false) +#define has_assisted_x2apic(d) ((void)(d), false) #define hvm_paging_enabled(v) ((void)(v), false) #define hvm_wp_enabled(v) ((void)(v), false) #define hvm_pcid_enabled(v) ((void)(v), false) diff --git a/xen/arch/x86/traps.c b/xen/arch/x86/traps.c index a2278d9499..a8dba88916 100644 --- a/xen/arch/x86/traps.c +++ b/xen/arch/x86/traps.c @@ -1121,7 +1121,8 @@ void cpuid_hypervisor_leaves(const struct vcpu *v, uint32_t leaf, if ( !is_hvm_domain(d) || subleaf != 0 ) break; - if ( cpu_has_vmx_apic_reg_virt ) + if ( cpu_has_vmx_apic_reg_virt && + has_assisted_xapic(d) ) res->a |= XEN_HVM_CPUID_APIC_ACCESS_VIRT; /* @@ -1130,7 +1131,7 @@ void cpuid_hypervisor_leaves(const struct vcpu *v, uint32_t leaf, * and wrmsr in the guest will run without VMEXITs (see * vmx_vlapic_msr_changed()). */ - if ( cpu_has_vmx_virtualize_x2apic_mode && + if ( has_assisted_x2apic(d) && cpu_has_vmx_apic_reg_virt && cpu_has_vmx_virtual_intr_delivery ) res->a |= XEN_HVM_CPUID_X2APIC_VIRT; diff --git a/xen/include/public/arch-x86/xen.h b/xen/include/public/arch-x86/xen.h index 7acd94c8eb..9da32c6239 100644 --- a/xen/include/public/arch-x86/xen.h +++ b/xen/include/public/arch-x86/xen.h @@ -317,6 +317,8 @@ struct xen_arch_domainconfig { * doesn't allow the guest to read or write to the underlying MSR. */ #define XEN_X86_MSR_RELAXED (1u << 0) +#define XEN_X86_ASSISTED_XAPIC (1u << 1) +#define XEN_X86_ASSISTED_X2APIC (1u << 2) uint32_t misc_flags; };
Introduce a new per-domain creation x86 specific flag to select whether hardware assisted virtualization should be used for x{2}APIC. A per-domain option is added to xl in order to select the usage of x{2}APIC hardware assisted virtualization, as well as a global configuration option. Having all APIC interaction exit to Xen for emulation is slow and can induce much overhead. Hardware can speed up x{2}APIC by decoding the APIC access and providing a VM exit with a more specific exit reason than a regular EPT fault or by altogether avoiding a VM exit. On the other hand, being able to disable x{2}APIC hardware assisted virtualization can be useful for testing and debugging purposes. Note: vmx_install_vlapic_mapping doesn't require modifications regardless of whether the guest has "Virtualize APIC accesses" enabled or not, i.e., setting the APIC_ACCESS_ADDR VMCS field is fine so long as virtualize_apic_accesses is supported by the CPU. Suggested-by: Andrew Cooper <andrew.cooper3@citrix.com> Signed-off-by: Jane Malalane <jane.malalane@citrix.com> --- CC: Wei Liu <wl@xen.org> CC: Anthony PERARD <anthony.perard@citrix.com> CC: Juergen Gross <jgross@suse.com> CC: Andrew Cooper <andrew.cooper3@citrix.com> CC: George Dunlap <george.dunlap@citrix.com> CC: Jan Beulich <jbeulich@suse.com> CC: Julien Grall <julien@xen.org> CC: Stefano Stabellini <sstabellini@kernel.org> CC: Christian Lindig <christian.lindig@citrix.com> CC: David Scott <dave@recoil.org> CC: Volodymyr Babchuk <Volodymyr_Babchuk@epam.com> CC: "Roger Pau Monné" <roger.pau@citrix.com> v8: * Widen assisted_x{2}apic parsing to PVH guests in parse_config_data() v7: * Fix void return in libxl__arch_domain_build_info_setdefault * Fix style issues * Use EINVAL when rejecting assisted_x{2}apic for PV guests and ENODEV otherwise, when assisted_x{2}apic isn't supported * Define has_assisted_x{2}apic macros for when !CONFIG_HVM * Replace "EPT" fault reference with "p2m" fault since the former is Intel-specific v6: * Use ENODEV instead of EINVAL when rejecting assisted_x{2}apic for PV guests * Move has_assisted_x{2}apic macros out of an Intel specific header * Remove references to Intel specific features in documentation v5: * Revert v4 changes in vmx_vlapic_msr_changed(), preserving the use of the has_assisted_x{2}apic macros * Following changes in assisted_x{2}apic_available definitions in patch 1, retighten conditionals for setting XEN_HVM_CPUID_APIC_ACCESS_VIRT and XEN_HVM_CPUID_X2APIC_VIRT in cpuid_hypervisor_leaves() v4: * Add has_assisted_x{2}apic macros and use them where appropriate * Replace CPU checks with per-domain assisted_x{2}apic control options in vmx_vlapic_msr_changed() and cpuid_hypervisor_leaves(), following edits to assisted_x{2}apic_available definitions in patch 1 Note: new assisted_x{2}apic_available definitions make later cpu_has_vmx_apic_reg_virt and cpu_has_vmx_virtual_intr_delivery checks redundant in vmx_vlapic_msr_changed() v3: * Change info in xl.cfg to better express reality and fix capitalization of x{2}apic * Move "physinfo" variable definition to the beggining of libxl__domain_build_info_setdefault() * Reposition brackets in if statement to match libxl coding style * Shorten logic in libxl__arch_domain_build_info_setdefault() * Correct dprintk message in arch_sanitise_domain_config() * Make appropriate changes in vmx_vlapic_msr_changed() and cpuid_hypervisor_leaves() for amended "assisted_x2apic" bit * Remove unneeded parantheses v2: * Add a LIBXL_HAVE_ASSISTED_APIC macro * Pass xcpyshinfo as a pointer in libxl__arch_get_physinfo * Add a return statement in now "int" libxl__arch_domain_build_info_setdefault() * Preserve libxl__arch_domain_build_info_setdefault 's location in libxl_create.c * Correct x{2}apic default setting logic in libxl__arch_domain_prepare_config() * Correct logic for parsing assisted_x{2}apic host/guest options in xl_parse.c and initialize them to -1 in xl.c * Use guest options directly in vmx_vlapic_msr_changed * Fix indentation of bool assisted_x{2}apic in struct hvm_domain * Add a change in xenctrl_stubs.c to pass xenctrl ABI checks --- docs/man/xl.cfg.5.pod.in | 15 +++++++++++++++ docs/man/xl.conf.5.pod.in | 12 ++++++++++++ tools/golang/xenlight/helpers.gen.go | 12 ++++++++++++ tools/golang/xenlight/types.gen.go | 2 ++ tools/include/libxl.h | 7 +++++++ tools/libs/light/libxl_arch.h | 5 +++-- tools/libs/light/libxl_arm.c | 9 ++++++--- tools/libs/light/libxl_create.c | 22 +++++++++++++--------- tools/libs/light/libxl_types.idl | 2 ++ tools/libs/light/libxl_x86.c | 28 ++++++++++++++++++++++++++-- tools/ocaml/libs/xc/xenctrl.ml | 2 ++ tools/ocaml/libs/xc/xenctrl.mli | 2 ++ tools/ocaml/libs/xc/xenctrl_stubs.c | 2 +- tools/xl/xl.c | 8 ++++++++ tools/xl/xl.h | 2 ++ tools/xl/xl_parse.c | 18 ++++++++++++++++++ xen/arch/x86/domain.c | 29 ++++++++++++++++++++++++++++- xen/arch/x86/hvm/vmx/vmcs.c | 4 ++++ xen/arch/x86/hvm/vmx/vmx.c | 13 ++++--------- xen/arch/x86/include/asm/hvm/domain.h | 6 ++++++ xen/arch/x86/include/asm/hvm/hvm.h | 5 +++++ xen/arch/x86/traps.c | 5 +++-- xen/include/public/arch-x86/xen.h | 2 ++ 23 files changed, 183 insertions(+), 29 deletions(-)