Merge remote-tracking branch 'qemu-kvm/uq/master' into staging

* qemu-kvm/uq/master:
  kvm: i386: Add classic PCI device assignment
  kvm: i386: Add services required for PCI device assignment
  kvm: Introduce kvm_has_intx_set_mask
  kvm: Introduce kvm_irqchip_update_msi_route
  kvm: Clean up irqfd API
  qemu: Use valgrind annotations to mark kvm guest memory as defined

configure

@@ -2971,11 +2971,12 @@ if compile_prog "-Werror" "" ; then
 fi
 
 ########################################
-# check if we have valgrind/valgrind.h
+# check if we have valgrind/valgrind.h and valgrind/memcheck.h
 
 valgrind_h=no
 cat > $TMPC << EOF
 #include <valgrind/valgrind.h>
+#include <valgrind/memcheck.h>
 int main(void) {
   return 0;
 }

hw/kvm/Makefile.objs

@@ -1 +1 @@
-obj-$(CONFIG_KVM) += clock.o apic.o i8259.o ioapic.o i8254.o
+obj-$(CONFIG_KVM) += clock.o apic.o i8259.o ioapic.o i8254.o pci-assign.o

hw/qdev-monitor.c

@@ -44,6 +44,7 @@ static const QDevAlias qdev_alias_table[] = {
     { "virtio-serial-s390", "virtio-serial", QEMU_ARCH_S390X },
     { "lsi53c895a", "lsi" },
     { "ich9-ahci", "ahci" },
+    { "kvm-pci-assign", "pci-assign" },
     { }
 };
 

hw/virtio-pci.c

@@ -517,7 +517,7 @@ static int kvm_virtio_pci_vq_vector_use(VirtIOPCIProxy *proxy,
     }
     irqfd->users++;
 
-    ret = kvm_irqchip_add_irq_notifier(kvm_state, n, irqfd->virq);
+    ret = kvm_irqchip_add_irqfd_notifier(kvm_state, n, irqfd->virq);
     if (ret < 0) {
         if (--irqfd->users == 0) {
             kvm_irqchip_release_virq(kvm_state, irqfd->virq);
@@ -538,7 +538,7 @@ static void kvm_virtio_pci_vq_vector_release(VirtIOPCIProxy *proxy,
     VirtIOIRQFD *irqfd = &proxy->vector_irqfd[vector];
     int ret;
 
-    ret = kvm_irqchip_remove_irq_notifier(kvm_state, n, irqfd->virq);
+    ret = kvm_irqchip_remove_irqfd_notifier(kvm_state, n, irqfd->virq);
     assert(ret == 0);
 
     if (--irqfd->users == 0) {

kvm-all.c

@@ -39,6 +39,10 @@
 #include <sys/eventfd.h>
 #endif
 
+#ifdef CONFIG_VALGRIND_H
+#include <valgrind/memcheck.h>
+#endif
+
 /* KVM uses PAGE_SIZE in its definition of COALESCED_MMIO_MAX */
 #define PAGE_SIZE TARGET_PAGE_SIZE
 
@@ -84,6 +88,7 @@ struct KVMState
     int pit_state2;
     int xsave, xcrs;
     int many_ioeventfds;
+    int intx_set_mask;
     /* The man page (and posix) say ioctl numbers are signed int, but
      * they're not.  Linux, glibc and *BSD all treat ioctl numbers as
      * unsigned, and treating them as signed here can break things */
@@ -959,6 +964,30 @@ static void kvm_add_routing_entry(KVMState *s,
     kvm_irqchip_commit_routes(s);
 }
 
+static int kvm_update_routing_entry(KVMState *s,
+                                    struct kvm_irq_routing_entry *new_entry)
+{
+    struct kvm_irq_routing_entry *entry;
+    int n;
+
+    for (n = 0; n < s->irq_routes->nr; n++) {
+        entry = &s->irq_routes->entries[n];
+        if (entry->gsi != new_entry->gsi) {
+            continue;
+        }
+
+        entry->type = new_entry->type;
+        entry->flags = new_entry->flags;
+        entry->u = new_entry->u;
+
+        kvm_irqchip_commit_routes(s);
+
+        return 0;
+    }
+
+    return -ESRCH;
+}
+
 void kvm_irqchip_add_irq_route(KVMState *s, int irq, int irqchip, int pin)
 {
     struct kvm_irq_routing_entry e;
@@ -1121,6 +1150,24 @@ int kvm_irqchip_add_msi_route(KVMState *s, MSIMessage msg)
     return virq;
 }
 
+int kvm_irqchip_update_msi_route(KVMState *s, int virq, MSIMessage msg)
+{
+    struct kvm_irq_routing_entry kroute;
+
+    if (!kvm_irqchip_in_kernel()) {
+        return -ENOSYS;
+    }
+
+    kroute.gsi = virq;
+    kroute.type = KVM_IRQ_ROUTING_MSI;
+    kroute.flags = 0;
+    kroute.u.msi.address_lo = (uint32_t)msg.address;
+    kroute.u.msi.address_hi = msg.address >> 32;
+    kroute.u.msi.data = msg.data;
+
+    return kvm_update_routing_entry(s, &kroute);
+}
+
 static int kvm_irqchip_assign_irqfd(KVMState *s, int fd, int virq, bool assign)
 {
     struct kvm_irqfd irqfd = {
@@ -1162,24 +1209,14 @@ static int kvm_irqchip_assign_irqfd(KVMState *s, int fd, int virq, bool assign)
 }
 #endif /* !KVM_CAP_IRQ_ROUTING */
 
-int kvm_irqchip_add_irqfd(KVMState *s, int fd, int virq)
+int kvm_irqchip_add_irqfd_notifier(KVMState *s, EventNotifier *n, int virq)
 {
-    return kvm_irqchip_assign_irqfd(s, fd, virq, true);
+    return kvm_irqchip_assign_irqfd(s, event_notifier_get_fd(n), virq, true);
 }
 
-int kvm_irqchip_add_irq_notifier(KVMState *s, EventNotifier *n, int virq)
+int kvm_irqchip_remove_irqfd_notifier(KVMState *s, EventNotifier *n, int virq)
 {
-    return kvm_irqchip_add_irqfd(s, event_notifier_get_fd(n), virq);
-}
-
-int kvm_irqchip_remove_irqfd(KVMState *s, int fd, int virq)
-{
-    return kvm_irqchip_assign_irqfd(s, fd, virq, false);
-}
-
-int kvm_irqchip_remove_irq_notifier(KVMState *s, EventNotifier *n, int virq)
-{
-    return kvm_irqchip_remove_irqfd(s, event_notifier_get_fd(n), virq);
+    return kvm_irqchip_assign_irqfd(s, event_notifier_get_fd(n), virq, false);
 }
 
 static int kvm_irqchip_create(KVMState *s)
@@ -1350,6 +1387,8 @@ int kvm_init(void)
     s->direct_msi = (kvm_check_extension(s, KVM_CAP_SIGNAL_MSI) > 0);
 #endif
 
+    s->intx_set_mask = kvm_check_extension(s, KVM_CAP_PCI_2_3);
+
     ret = kvm_arch_init(s);
     if (ret < 0) {
         goto err;
@@ -1704,6 +1743,11 @@ int kvm_has_gsi_routing(void)
 #endif
 }
 
+int kvm_has_intx_set_mask(void)
+{
+    return kvm_state->intx_set_mask;
+}
+
 void *kvm_vmalloc(ram_addr_t size)
 {
 #ifdef TARGET_S390X
@@ -1719,6 +1763,9 @@ void *kvm_vmalloc(ram_addr_t size)
 
 void kvm_setup_guest_memory(void *start, size_t size)
 {
+#ifdef CONFIG_VALGRIND_H
+    VALGRIND_MAKE_MEM_DEFINED(start, size);
+#endif
     if (!kvm_has_sync_mmu()) {
         int ret = qemu_madvise(start, size, QEMU_MADV_DONTFORK);
 

kvm-stub.c

@@ -141,22 +141,12 @@ void kvm_irqchip_release_virq(KVMState *s, int virq)
 {
 }
 
-int kvm_irqchip_add_irqfd(KVMState *s, int fd, int virq)
+int kvm_irqchip_add_irqfd_notifier(KVMState *s, EventNotifier *n, int virq)
 {
     return -ENOSYS;
 }
 
-int kvm_irqchip_add_irq_notifier(KVMState *s, EventNotifier *n, int virq)
-{
-    return -ENOSYS;
-}
-
-int kvm_irqchip_remove_irqfd(KVMState *s, int fd, int virq)
-{
-    return -ENOSYS;
-}
-
-int kvm_irqchip_remove_irq_notifier(KVMState *s, EventNotifier *n, int virq)
+int kvm_irqchip_remove_irqfd_notifier(KVMState *s, EventNotifier *n, int virq)
 {
     return -ENOSYS;
 }

kvm.h

@@ -117,6 +117,7 @@ int kvm_has_xcrs(void);
 int kvm_has_pit_state2(void);
 int kvm_has_many_ioeventfds(void);
 int kvm_has_gsi_routing(void);
+int kvm_has_intx_set_mask(void);
 
 #ifdef NEED_CPU_H
 int kvm_init_vcpu(CPUArchState *env);
@@ -270,10 +271,9 @@ int kvm_set_ioeventfd_mmio(int fd, uint32_t adr, uint32_t val, bool assign,
 int kvm_set_ioeventfd_pio_word(int fd, uint16_t adr, uint16_t val, bool assign);
 
 int kvm_irqchip_add_msi_route(KVMState *s, MSIMessage msg);
+int kvm_irqchip_update_msi_route(KVMState *s, int virq, MSIMessage msg);
 void kvm_irqchip_release_virq(KVMState *s, int virq);
 
-int kvm_irqchip_add_irqfd(KVMState *s, int fd, int virq);
-int kvm_irqchip_remove_irqfd(KVMState *s, int fd, int virq);
-int kvm_irqchip_add_irq_notifier(KVMState *s, EventNotifier *n, int virq);
-int kvm_irqchip_remove_irq_notifier(KVMState *s, EventNotifier *n, int virq);
+int kvm_irqchip_add_irqfd_notifier(KVMState *s, EventNotifier *n, int virq);
+int kvm_irqchip_remove_irqfd_notifier(KVMState *s, EventNotifier *n, int virq);
 #endif

target-i386/kvm.c

@@ -31,6 +31,7 @@
 #include "hw/apic.h"
 #include "ioport.h"
 #include "hyperv.h"
+#include "hw/pci.h"
 
 //#define DEBUG_KVM
 
@@ -2068,3 +2069,143 @@ void kvm_arch_init_irq_routing(KVMState *s)
     kvm_msi_via_irqfd_allowed = true;
     kvm_gsi_routing_allowed = true;
 }
+
+/* Classic KVM device assignment interface. Will remain x86 only. */
+int kvm_device_pci_assign(KVMState *s, PCIHostDeviceAddress *dev_addr,
+                          uint32_t flags, uint32_t *dev_id)
+{
+    struct kvm_assigned_pci_dev dev_data = {
+        .segnr = dev_addr->domain,
+        .busnr = dev_addr->bus,
+        .devfn = PCI_DEVFN(dev_addr->slot, dev_addr->function),
+        .flags = flags,
+    };
+    int ret;
+
+    dev_data.assigned_dev_id =
+        (dev_addr->domain << 16) | (dev_addr->bus << 8) | dev_data.devfn;
+
+    ret = kvm_vm_ioctl(s, KVM_ASSIGN_PCI_DEVICE, &dev_data);
+    if (ret < 0) {
+        return ret;
+    }
+
+    *dev_id = dev_data.assigned_dev_id;
+
+    return 0;
+}
+
+int kvm_device_pci_deassign(KVMState *s, uint32_t dev_id)
+{
+    struct kvm_assigned_pci_dev dev_data = {
+        .assigned_dev_id = dev_id,
+    };
+
+    return kvm_vm_ioctl(s, KVM_DEASSIGN_PCI_DEVICE, &dev_data);
+}
+
+static int kvm_assign_irq_internal(KVMState *s, uint32_t dev_id,
+                                   uint32_t irq_type, uint32_t guest_irq)
+{
+    struct kvm_assigned_irq assigned_irq = {
+        .assigned_dev_id = dev_id,
+        .guest_irq = guest_irq,
+        .flags = irq_type,
+    };
+
+    if (kvm_check_extension(s, KVM_CAP_ASSIGN_DEV_IRQ)) {
+        return kvm_vm_ioctl(s, KVM_ASSIGN_DEV_IRQ, &assigned_irq);
+    } else {
+        return kvm_vm_ioctl(s, KVM_ASSIGN_IRQ, &assigned_irq);
+    }
+}
+
+int kvm_device_intx_assign(KVMState *s, uint32_t dev_id, bool use_host_msi,
+                           uint32_t guest_irq)
+{
+    uint32_t irq_type = KVM_DEV_IRQ_GUEST_INTX |
+        (use_host_msi ? KVM_DEV_IRQ_HOST_MSI : KVM_DEV_IRQ_HOST_INTX);
+
+    return kvm_assign_irq_internal(s, dev_id, irq_type, guest_irq);
+}
+
+int kvm_device_intx_set_mask(KVMState *s, uint32_t dev_id, bool masked)
+{
+    struct kvm_assigned_pci_dev dev_data = {
+        .assigned_dev_id = dev_id,
+        .flags = masked ? KVM_DEV_ASSIGN_MASK_INTX : 0,
+    };
+
+    return kvm_vm_ioctl(s, KVM_ASSIGN_SET_INTX_MASK, &dev_data);
+}
+
+static int kvm_deassign_irq_internal(KVMState *s, uint32_t dev_id,
+                                     uint32_t type)
+{
+    struct kvm_assigned_irq assigned_irq = {
+        .assigned_dev_id = dev_id,
+        .flags = type,
+    };
+
+    return kvm_vm_ioctl(s, KVM_DEASSIGN_DEV_IRQ, &assigned_irq);
+}
+
+int kvm_device_intx_deassign(KVMState *s, uint32_t dev_id, bool use_host_msi)
+{
+    return kvm_deassign_irq_internal(s, dev_id, KVM_DEV_IRQ_GUEST_INTX |
+        (use_host_msi ? KVM_DEV_IRQ_HOST_MSI : KVM_DEV_IRQ_HOST_INTX));
+}
+
+int kvm_device_msi_assign(KVMState *s, uint32_t dev_id, int virq)
+{
+    return kvm_assign_irq_internal(s, dev_id, KVM_DEV_IRQ_HOST_MSI |
+                                              KVM_DEV_IRQ_GUEST_MSI, virq);
+}
+
+int kvm_device_msi_deassign(KVMState *s, uint32_t dev_id)
+{
+    return kvm_deassign_irq_internal(s, dev_id, KVM_DEV_IRQ_GUEST_MSI |
+                                                KVM_DEV_IRQ_HOST_MSI);
+}
+
+bool kvm_device_msix_supported(KVMState *s)
+{
+    /* The kernel lacks a corresponding KVM_CAP, so we probe by calling
+     * KVM_ASSIGN_SET_MSIX_NR with an invalid parameter. */
+    return kvm_vm_ioctl(s, KVM_ASSIGN_SET_MSIX_NR, NULL) == -EFAULT;
+}
+
+int kvm_device_msix_init_vectors(KVMState *s, uint32_t dev_id,
+                                 uint32_t nr_vectors)
+{
+    struct kvm_assigned_msix_nr msix_nr = {
+        .assigned_dev_id = dev_id,
+        .entry_nr = nr_vectors,
+    };
+
+    return kvm_vm_ioctl(s, KVM_ASSIGN_SET_MSIX_NR, &msix_nr);
+}
+
+int kvm_device_msix_set_vector(KVMState *s, uint32_t dev_id, uint32_t vector,
+                               int virq)
+{
+    struct kvm_assigned_msix_entry msix_entry = {
+        .assigned_dev_id = dev_id,
+        .gsi = virq,
+        .entry = vector,
+    };
+
+    return kvm_vm_ioctl(s, KVM_ASSIGN_SET_MSIX_ENTRY, &msix_entry);
+}
+
+int kvm_device_msix_assign(KVMState *s, uint32_t dev_id)
+{
+    return kvm_assign_irq_internal(s, dev_id, KVM_DEV_IRQ_HOST_MSIX |
+                                              KVM_DEV_IRQ_GUEST_MSIX, 0);
+}
+
+int kvm_device_msix_deassign(KVMState *s, uint32_t dev_id)
+{
+    return kvm_deassign_irq_internal(s, dev_id, KVM_DEV_IRQ_GUEST_MSIX |
+                                                KVM_DEV_IRQ_HOST_MSIX);
+}

target-i386/kvm_i386.h

@@ -11,6 +11,28 @@
 #ifndef QEMU_KVM_I386_H
 #define QEMU_KVM_I386_H
 
+#include "kvm.h"
+
 bool kvm_allows_irq0_override(void);
 
+int kvm_device_pci_assign(KVMState *s, PCIHostDeviceAddress *dev_addr,
+                          uint32_t flags, uint32_t *dev_id);
+int kvm_device_pci_deassign(KVMState *s, uint32_t dev_id);
+
+int kvm_device_intx_assign(KVMState *s, uint32_t dev_id,
+                           bool use_host_msi, uint32_t guest_irq);
+int kvm_device_intx_set_mask(KVMState *s, uint32_t dev_id, bool masked);
+int kvm_device_intx_deassign(KVMState *s, uint32_t dev_id, bool use_host_msi);
+
+int kvm_device_msi_assign(KVMState *s, uint32_t dev_id, int virq);
+int kvm_device_msi_deassign(KVMState *s, uint32_t dev_id);
+
+bool kvm_device_msix_supported(KVMState *s);
+int kvm_device_msix_init_vectors(KVMState *s, uint32_t dev_id,
+                                 uint32_t nr_vectors);
+int kvm_device_msix_set_vector(KVMState *s, uint32_t dev_id, uint32_t vector,
+                               int virq);
+int kvm_device_msix_assign(KVMState *s, uint32_t dev_id);
+int kvm_device_msix_deassign(KVMState *s, uint32_t dev_id);
+
 #endif