Merge remote-tracking branch 'remotes/ehabkost/tags/x86-pull-request' into staging

x86 queue, 2016-10-17

# gpg: Signature made Mon 17 Oct 2016 18:51:07 BST
# gpg:                using RSA key 0x2807936F984DC5A6
# gpg: Good signature from "Eduardo Habkost <ehabkost@redhat.com>"
# Primary key fingerprint: 5A32 2FD5 ABC4 D3DB ACCF  D1AA 2807 936F 984D C5A6

* remotes/ehabkost/tags/x86-pull-request: (21 commits)
  target-i386: Don't use cpu->migratable when filtering features
  target-i386: Return runnability information on query-cpu-definitions
  target-i386: x86_cpu_load_features() function
  target-i386: Unset cannot_destroy_with_object_finalize_yet
  target-i386/kvm: cache the return value of kvm_enable_x2apic()
  intel_iommu: reject broken EIM
  intel_iommu: add OnOffAuto intr_eim as "eim" property
  intel_iommu: redo configuraton check in realize
  intel_iommu: pass whole remapped addresses to apic
  apic: add send_msi() to APICCommonClass
  apic: add global apic_get_class()
  target-i386: Move warning code outside x86_cpu_filter_features()
  qmp: Add runnability information to query-cpu-definitions
  target-i386: xsave: Add FP and SSE bits to x86_ext_save_areas
  target-i386: Register properties for feature aliases manually
  target-i386: Remove underscores from feat_names arrays
  target-i386: Make plus_features/minus_features QOM-based
  target-i386: Register aliases for feature names with underscores
  target-i386: Disable VME by default with TCG
  target-i386: List CPU models using subclass list
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

hw/i386/intel_iommu.c

@@ -21,6 +21,7 @@
 
 #include "qemu/osdep.h"
 #include "qemu/error-report.h"
+#include "qapi/error.h"
 #include "hw/sysbus.h"
 #include "exec/address-spaces.h"
 #include "intel_iommu_internal.h"
@@ -32,6 +33,8 @@
 #include "hw/i386/x86-iommu.h"
 #include "hw/pci-host/q35.h"
 #include "sysemu/kvm.h"
+#include "hw/i386/apic_internal.h"
+#include "kvm_i386.h"
 
 /*#define DEBUG_INTEL_IOMMU*/
 #ifdef DEBUG_INTEL_IOMMU
@@ -280,18 +283,17 @@ static void vtd_update_iotlb(IntelIOMMUState *s, uint16_t source_id,
 static void vtd_generate_interrupt(IntelIOMMUState *s, hwaddr mesg_addr_reg,
                                    hwaddr mesg_data_reg)
 {
-    hwaddr addr;
-    uint32_t data;
+    MSIMessage msi;
 
     assert(mesg_data_reg < DMAR_REG_SIZE);
     assert(mesg_addr_reg < DMAR_REG_SIZE);
 
-    addr = vtd_get_long_raw(s, mesg_addr_reg);
-    data = vtd_get_long_raw(s, mesg_data_reg);
+    msi.address = vtd_get_long_raw(s, mesg_addr_reg);
+    msi.data = vtd_get_long_raw(s, mesg_data_reg);
 
-    VTD_DPRINTF(FLOG, "msi: addr 0x%"PRIx64 " data 0x%"PRIx32, addr, data);
-    address_space_stl_le(&address_space_memory, addr, data,
-                         MEMTXATTRS_UNSPECIFIED, NULL);
+    VTD_DPRINTF(FLOG, "msi: addr 0x%"PRIx64 " data 0x%"PRIx32,
+                msi.address, msi.data);
+    apic_get_class()->send_msi(&msi);
 }
 
 /* Generate a fault event to software via MSI if conditions are met.
@@ -2012,6 +2014,9 @@ static const MemoryRegionOps vtd_mem_ops = {
 
 static Property vtd_properties[] = {
     DEFINE_PROP_UINT32("version", IntelIOMMUState, version, 0),
+    DEFINE_PROP_ON_OFF_AUTO("eim", IntelIOMMUState, intr_eim,
+                            ON_OFF_AUTO_AUTO),
+    DEFINE_PROP_BOOL("x-buggy-eim", IntelIOMMUState, buggy_eim, false),
     DEFINE_PROP_END_OF_LIST(),
 };
 
@@ -2134,6 +2139,7 @@ static void vtd_generate_msi_message(VTDIrq *irq, MSIMessage *msg_out)
     msg.dest_mode = irq->dest_mode;
     msg.redir_hint = irq->redir_hint;
     msg.dest = irq->dest;
+    msg.__addr_hi = irq->dest & 0xffffff00;
     msg.__addr_head = cpu_to_le32(0xfee);
     /* Keep this from original MSI address bits */
     msg.__not_used = irq->msi_addr_last_bits;
@@ -2293,11 +2299,7 @@ static MemTxResult vtd_mem_ir_write(void *opaque, hwaddr addr,
                 " for device sid 0x%04x",
                 to.address, to.data, sid);
 
-    if (dma_memory_write(&address_space_memory, to.address,
-                         &to.data, size)) {
-        VTD_DPRINTF(GENERAL, "error: fail to write 0x%"PRIx64
-                    " value 0x%"PRIx32, to.address, to.data);
-    }
+    apic_get_class()->send_msi(&to);
 
     return MEMTX_OK;
 }
@@ -2382,7 +2384,11 @@ static void vtd_init(IntelIOMMUState *s)
     s->ecap = VTD_ECAP_QI | VTD_ECAP_IRO;
 
     if (x86_iommu->intr_supported) {
-        s->ecap |= VTD_ECAP_IR | VTD_ECAP_EIM | VTD_ECAP_MHMV;
+        s->ecap |= VTD_ECAP_IR | VTD_ECAP_MHMV;
+        if (s->intr_eim == ON_OFF_AUTO_ON) {
+            s->ecap |= VTD_ECAP_EIM;
+        }
+        assert(s->intr_eim != ON_OFF_AUTO_AUTO);
     }
 
     vtd_reset_context_cache(s);
@@ -2463,6 +2469,42 @@ static AddressSpace *vtd_host_dma_iommu(PCIBus *bus, void *opaque, int devfn)
     return &vtd_as->as;
 }
 
+static bool vtd_decide_config(IntelIOMMUState *s, Error **errp)
+{
+    X86IOMMUState *x86_iommu = X86_IOMMU_DEVICE(s);
+
+    /* Currently Intel IOMMU IR only support "kernel-irqchip={off|split}" */
+    if (x86_iommu->intr_supported && kvm_irqchip_in_kernel() &&
+        !kvm_irqchip_is_split()) {
+        error_setg(errp, "Intel Interrupt Remapping cannot work with "
+                         "kernel-irqchip=on, please use 'split|off'.");
+        return false;
+    }
+    if (s->intr_eim == ON_OFF_AUTO_ON && !x86_iommu->intr_supported) {
+        error_setg(errp, "eim=on cannot be selected without intremap=on");
+        return false;
+    }
+
+    if (s->intr_eim == ON_OFF_AUTO_AUTO) {
+        s->intr_eim = (kvm_irqchip_in_kernel() || s->buggy_eim)
+                      && x86_iommu->intr_supported ?
+                                              ON_OFF_AUTO_ON : ON_OFF_AUTO_OFF;
+    }
+    if (s->intr_eim == ON_OFF_AUTO_ON && !s->buggy_eim) {
+        if (!kvm_irqchip_in_kernel()) {
+            error_setg(errp, "eim=on requires accel=kvm,kernel-irqchip=split");
+            return false;
+        }
+        if (!kvm_enable_x2apic()) {
+            error_setg(errp, "eim=on requires support on the KVM side"
+                             "(X2APIC_API, first shipped in v4.7)");
+            return false;
+        }
+    }
+
+    return true;
+}
+
 static void vtd_realize(DeviceState *dev, Error **errp)
 {
     PCMachineState *pcms = PC_MACHINE(qdev_get_machine());
@@ -2472,6 +2514,11 @@ static void vtd_realize(DeviceState *dev, Error **errp)
 
     VTD_DPRINTF(GENERAL, "");
     x86_iommu->type = TYPE_INTEL;
+
+    if (!vtd_decide_config(s, errp)) {
+        return;
+    }
+
     memset(s->vtd_as_by_bus_num, 0, sizeof(s->vtd_as_by_bus_num));
     memory_region_init_io(&s->csrmem, OBJECT(s), &vtd_mem_ops, s,
                           "intel_iommu", DMAR_REG_SIZE);
@@ -2486,14 +2533,6 @@ static void vtd_realize(DeviceState *dev, Error **errp)
     pci_setup_iommu(bus, vtd_host_dma_iommu, dev);
     /* Pseudo address space under root PCI bus. */
     pcms->ioapic_as = vtd_host_dma_iommu(bus, s, Q35_PSEUDO_DEVFN_IOAPIC);
-
-    /* Currently Intel IOMMU IR only support "kernel-irqchip={off|split}" */
-    if (x86_iommu->intr_supported && kvm_irqchip_in_kernel() &&
-        !kvm_irqchip_is_split()) {
-        error_report("Intel Interrupt Remapping cannot work with "
-                     "kernel-irqchip=on, please use 'split|off'.");
-        exit(1);
-    }
 }
 
 static void vtd_class_init(ObjectClass *klass, void *data)

hw/i386/kvm/apic.c

@@ -169,6 +169,17 @@ static void kvm_apic_external_nmi(APICCommonState *s)
     run_on_cpu(CPU(s->cpu), do_inject_external_nmi, s);
 }
 
+static void kvm_send_msi(MSIMessage *msg)
+{
+    int ret;
+
+    ret = kvm_irqchip_send_msi(kvm_state, *msg);
+    if (ret < 0) {
+        fprintf(stderr, "KVM: injection failed, MSI lost (%s)\n",
+                strerror(-ret));
+    }
+}
+
 static uint64_t kvm_apic_mem_read(void *opaque, hwaddr addr,
                                   unsigned size)
 {
@@ -179,13 +190,8 @@ static void kvm_apic_mem_write(void *opaque, hwaddr addr,
                                uint64_t data, unsigned size)
 {
     MSIMessage msg = { .address = addr, .data = data };
-    int ret;
 
-    ret = kvm_irqchip_send_msi(kvm_state, msg);
-    if (ret < 0) {
-        fprintf(stderr, "KVM: injection failed, MSI lost (%s)\n",
-                strerror(-ret));
-    }
+    kvm_send_msi(&msg);
 }
 
 static const MemoryRegionOps kvm_apic_io_ops = {
@@ -232,6 +238,7 @@ static void kvm_apic_class_init(ObjectClass *klass, void *data)
     k->enable_tpr_reporting = kvm_apic_enable_tpr_reporting;
     k->vapic_base_update = kvm_apic_vapic_base_update;
     k->external_nmi = kvm_apic_external_nmi;
+    k->send_msi = kvm_send_msi;
 }
 
 static const TypeInfo kvm_apic_info = {

hw/i386/xen/xen_apic.c

@@ -68,6 +68,11 @@ static void xen_apic_external_nmi(APICCommonState *s)
 {
 }
 
+static void xen_send_msi(MSIMessage *msi)
+{
+    xen_hvm_inject_msi(msi->address, msi->data);
+}
+
 static void xen_apic_class_init(ObjectClass *klass, void *data)
 {
     APICCommonClass *k = APIC_COMMON_CLASS(klass);
@@ -78,6 +83,7 @@ static void xen_apic_class_init(ObjectClass *klass, void *data)
     k->get_tpr = xen_apic_get_tpr;
     k->vapic_base_update = xen_apic_vapic_base_update;
     k->external_nmi = xen_apic_external_nmi;
+    k->send_msi = xen_send_msi;
 }
 
 static const TypeInfo xen_apic_info = {

hw/intc/apic.c

@@ -740,8 +740,10 @@ static uint32_t apic_mem_readl(void *opaque, hwaddr addr)
     return val;
 }
 
-static void apic_send_msi(hwaddr addr, uint32_t data)
+static void apic_send_msi(MSIMessage *msi)
 {
+    uint64_t addr = msi->address;
+    uint32_t data = msi->data;
     uint8_t dest = (addr & MSI_ADDR_DEST_ID_MASK) >> MSI_ADDR_DEST_ID_SHIFT;
     uint8_t vector = (data & MSI_DATA_VECTOR_MASK) >> MSI_DATA_VECTOR_SHIFT;
     uint8_t dest_mode = (addr >> MSI_ADDR_DEST_MODE_SHIFT) & 0x1;
@@ -762,7 +764,8 @@ static void apic_mem_writel(void *opaque, hwaddr addr, uint32_t val)
          * APIC is connected directly to the CPU.
          * Mapping them on the global bus happens to work because
          * MSI registers are reserved in APIC MMIO and vice versa. */
-        apic_send_msi(addr, val);
+        MSIMessage msi = { .address = addr, .data = val };
+        apic_send_msi(&msi);
         return;
     }
 
@@ -913,6 +916,7 @@ static void apic_class_init(ObjectClass *klass, void *data)
     k->external_nmi = apic_external_nmi;
     k->pre_save = apic_pre_save;
     k->post_load = apic_post_load;
+    k->send_msi = apic_send_msi;
 }
 
 static const TypeInfo apic_info = {

hw/intc/apic_common.c

@@ -18,6 +18,7 @@
  * License along with this library; if not, see <http://www.gnu.org/licenses/>
  */
 #include "qemu/osdep.h"
+#include "qemu/error-report.h"
 #include "qapi/error.h"
 #include "qemu-common.h"
 #include "cpu.h"

include/hw/compat.h

@@ -14,6 +14,10 @@
         .driver   = "ioapic",\
         .property = "version",\
         .value    = "0x11",\
+    },{\
+        .driver   = "intel-iommu",\
+        .property = "x-buggy-eim",\
+        .value    = "true",\
     },
 
 #define HW_COMPAT_2_6 \

include/hw/i386/apic_internal.h

@@ -146,6 +146,10 @@ typedef struct APICCommonClass
     void (*pre_save)(APICCommonState *s);
     void (*post_load)(APICCommonState *s);
     void (*reset)(APICCommonState *s);
+    /* send_msi emulates an APIC bus and its proper place would be in a new
+     * device, but it's convenient to have it here for now.
+     */
+    void (*send_msi)(MSIMessage *msi);
 } APICCommonClass;
 
 struct APICCommonState {
@@ -222,4 +226,6 @@ static inline int apic_get_bit(uint32_t *tab, int index)
     return !!(tab[i] & mask);
 }
 
+APICCommonClass *apic_get_class(void);
+
 #endif /* QEMU_APIC_INTERNAL_H */

include/hw/i386/intel_iommu.h

@@ -289,6 +289,8 @@ struct IntelIOMMUState {
     dma_addr_t intr_root;           /* Interrupt remapping table pointer */
     uint32_t intr_size;             /* Number of IR table entries */
     bool intr_eime;                 /* Extended interrupt mode enabled */
+    OnOffAuto intr_eim;             /* Toggle for EIM cabability */
+    bool buggy_eim;                 /* Force buggy EIM unless eim=off */
 };
 
 /* Find the VTD Address space associated with the given bus pointer,

qapi-schema.json

@@ -3101,10 +3101,31 @@
 #          QEMU version, machine type, machine options and accelerator options.
 #          A static model is always migration-safe. (since 2.8)
 #
+# @unavailable-features: #optional List of properties that prevent
+#                        the CPU model from running in the current
+#                        host. (since 2.8)
+#
+# @unavailable-features is a list of QOM property names that
+# represent CPU model attributes that prevent the CPU from running.
+# If the QOM property is read-only, that means there's no known
+# way to make the CPU model run in the current host. Implementations
+# that choose not to provide specific information return the
+# property name "type".
+# If the property is read-write, it means that it MAY be possible
+# to run the CPU model in the current host if that property is
+# changed. Management software can use it as hints to suggest or
+# choose an alternative for the user, or just to generate meaningful
+# error messages explaining why the CPU model can't be used.
+# If @unavailable-features is an empty list, the CPU model is
+# runnable using the current host and machine-type.
+# If @unavailable-features is not present, runnability
+# information for the CPU is not available.
+#
 # Since: 1.2.0
 ##
 { 'struct': 'CpuDefinitionInfo',
-  'data': { 'name': 'str', '*migration-safe': 'bool', 'static': 'bool' } }
+  'data': { 'name': 'str', '*migration-safe': 'bool', 'static': 'bool',
+            '*unavailable-features': [ 'str' ] } }
 
 ##
 # @query-cpu-definitions:

target-i386/cpu-qom.h

@@ -63,6 +63,10 @@ typedef struct X86CPUClass {
 
     bool kvm_required;
 
+    /* Optional description of CPU model.
+     * If unavailable, cpu_def->model_id is used */
+    const char *model_description;
+
     DeviceRealize parent_realize;
     void (*parent_reset)(CPUState *cpu);
 } X86CPUClass;

target-i386/kvm-stub.c

@@ -25,6 +25,11 @@ bool kvm_has_smm(void)
     return 1;
 }
 
+bool kvm_enable_x2apic(void)
+{
+    return false;
+}
+
 /* This function is only called inside conditionals which we
  * rely on the compiler to optimize out when CONFIG_KVM is not
  * defined.

target-i386/kvm.c

@@ -122,6 +122,32 @@ bool kvm_allows_irq0_override(void)
     return !kvm_irqchip_in_kernel() || kvm_has_gsi_routing();
 }
 
+static bool kvm_x2apic_api_set_flags(uint64_t flags)
+{
+    KVMState *s = KVM_STATE(current_machine->accelerator);
+
+    return !kvm_vm_enable_cap(s, KVM_CAP_X2APIC_API, 0, flags);
+}
+
+#define MEMORIZE(fn) \
+    ({ \
+        static typeof(fn) _result; \
+        static bool _memorized; \
+        \
+        if (_memorized) { \
+            return _result; \
+        } \
+        _memorized = true; \
+        _result = fn; \
+    })
+
+bool kvm_enable_x2apic(void)
+{
+    return MEMORIZE(
+             kvm_x2apic_api_set_flags(KVM_X2APIC_API_USE_32BIT_IDS |
+                                      KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK));
+}
+
 static int kvm_get_tsc(CPUState *cs)
 {
     X86CPU *cpu = X86_CPU(cs);

target-i386/kvm_i386.h

@@ -43,4 +43,5 @@ int kvm_device_msix_deassign(KVMState *s, uint32_t dev_id);
 
 void kvm_put_apicbase(X86CPU *cpu, uint64_t value);
 
+bool kvm_enable_x2apic(void);
 #endif

tests/test-x86-cpuid-compat.c

@@ -3,6 +3,7 @@
 #include "qapi/qmp/qlist.h"
 #include "qapi/qmp/qdict.h"
 #include "qapi/qmp/qint.h"
+#include "qapi/qmp/qbool.h"
 #include "libqtest.h"
 
 static char *get_cpu0_qom_path(void)
@@ -34,6 +35,15 @@ static QObject *qom_get(const char *path, const char *prop)
     return ret;
 }
 
+static bool qom_get_bool(const char *path, const char *prop)
+{
+    QBool *value = qobject_to_qbool(qom_get(path, prop));
+    bool b = qbool_get_bool(value);
+
+    QDECREF(value);
+    return b;
+}
+
 typedef struct CpuidTestArgs {
     const char *cmdline;
     const char *property;
@@ -66,10 +76,44 @@ static void add_cpuid_test(const char *name, const char *cmdline,
     qtest_add_data_func(name, args, test_cpuid_prop);
 }
 
+static void test_plus_minus(void)
+{
+    char *path;
+
+    /* Rules:
+     * 1)"-foo" overrides "+foo"
+     * 2) "[+-]foo" overrides "foo=..."
+     * 3) Old feature names with underscores (e.g. "sse4_2")
+     *    should keep working
+     *
+     * Note: rules 1 and 2 are planned to be removed soon, but we
+     * need to keep compatibility for a while until we start
+     * warning users about it.
+     */
+    qtest_start("-cpu pentium,-fpu,+fpu,-mce,mce=on,+cx8,cx8=off,+sse4_1,sse4_2=on");
+    path = get_cpu0_qom_path();
+
+    g_assert_false(qom_get_bool(path, "fpu"));
+    g_assert_false(qom_get_bool(path, "mce"));
+    g_assert_true(qom_get_bool(path, "cx8"));
+
+    /* Test both the original and the alias feature names: */
+    g_assert_true(qom_get_bool(path, "sse4-1"));
+    g_assert_true(qom_get_bool(path, "sse4.1"));
+
+    g_assert_true(qom_get_bool(path, "sse4-2"));
+    g_assert_true(qom_get_bool(path, "sse4.2"));
+
+    qtest_end();
+    g_free(path);
+}
+
 int main(int argc, char **argv)
 {
     g_test_init(&argc, &argv, NULL);
 
+    qtest_add_func("x86/cpuid/parsing-plus-minus", test_plus_minus);
+
     /* Original level values for CPU models: */
     add_cpuid_test("x86/cpuid/phenom/level",
                    "-cpu phenom", "level", 5);