Merge remote branch 'mst/for_anthony' into staging

docs/specs/acpi_pci_hotplug.txt

@@ -0,0 +1,37 @@
+QEMU<->ACPI BIOS PCI hotplug interface
+--------------------------------------
+
+QEMU supports PCI hotplug via ACPI, for PCI bus 0. This document
+describes the interface between QEMU and the ACPI BIOS.
+
+ACPI GPE block (IO ports 0xafe0-0xafe3, byte access):
+-----------------------------------------
+
+Generic ACPI GPE block. Bit 1 (GPE.1) used to notify PCI hotplug/eject
+event to ACPI BIOS, via SCI interrupt.
+
+PCI slot injection notification pending (IO port 0xae00-0xae03, 4-byte access):
+---------------------------------------------------------------
+Slot injection notification pending. One bit per slot.
+
+Read by ACPI BIOS GPE.1 handler to notify OS of injection
+events.
+
+PCI slot removal notification (IO port 0xae04-0xae07, 4-byte access):
+-----------------------------------------------------
+Slot removal notification pending. One bit per slot.
+
+Read by ACPI BIOS GPE.1 handler to notify OS of removal
+events.
+
+PCI device eject (IO port 0xae08-0xae0b, 4-byte access):
+----------------------------------------
+
+Used by ACPI BIOS _EJ0 method to request device removal. One bit per slot.
+Reads return 0.
+
+PCI removability status (IO port 0xae0c-0xae0f, 4-byte access):
+-----------------------------------------------
+
+Used by ACPI BIOS _RMV method to indicate removability status to OS. One
+bit per slot.

hw/acpi_piix4.c

@@ -37,6 +37,7 @@
 #define GPE_BASE 0xafe0
 #define PCI_BASE 0xae00
 #define PCI_EJ_BASE 0xae08
+#define PCI_RMV_BASE 0xae0c
 
 #define PIIX4_PCI_HOTPLUG_STATUS 2
 
@@ -73,6 +74,7 @@ typedef struct PIIX4PMState {
     /* for pci hotplug */
     struct gpe_regs gpe;
     struct pci_status pci0_status;
+    uint32_t pci0_hotplug_enable;
 } PIIX4PMState;
 
 static void piix4_acpi_system_hot_add_init(PCIBus *bus, PIIX4PMState *s);
@@ -322,6 +324,25 @@ static const VMStateDescription vmstate_acpi = {
     }
 };
 
+static void piix4_update_hotplug(PIIX4PMState *s)
+{
+    PCIDevice *dev = &s->dev;
+    BusState *bus = qdev_get_parent_bus(&dev->qdev);
+    DeviceState *qdev, *next;
+
+    s->pci0_hotplug_enable = ~0;
+
+    QLIST_FOREACH_SAFE(qdev, &bus->children, sibling, next) {
+        PCIDeviceInfo *info = container_of(qdev->info, PCIDeviceInfo, qdev);
+        PCIDevice *pdev = DO_UPCAST(PCIDevice, qdev, qdev);
+        int slot = PCI_SLOT(pdev->devfn);
+
+        if (info->no_hotplug) {
+            s->pci0_hotplug_enable &= ~(1 << slot);
+        }
+    }
+}
+
 static void piix4_reset(void *opaque)
 {
     PIIX4PMState *s = opaque;
@@ -336,6 +357,7 @@ static void piix4_reset(void *opaque)
         /* Mark SMM as already inited (until KVM supports SMM). */
         pci_conf[0x5B] = 0x02;
     }
+    piix4_update_hotplug(s);
 }
 
 static void piix4_powerdown(void *opaque, int irq, int power_failing)
@@ -576,6 +598,18 @@ static void pciej_write(void *opaque, uint32_t addr, uint32_t val)
     PIIX4_DPRINTF("pciej write %x <== %d\n", addr, val);
 }
 
+static uint32_t pcirmv_read(void *opaque, uint32_t addr)
+{
+    PIIX4PMState *s = opaque;
+
+    return s->pci0_hotplug_enable;
+}
+
+static void pcirmv_write(void *opaque, uint32_t addr, uint32_t val)
+{
+    return;
+}
+
 static int piix4_device_hotplug(DeviceState *qdev, PCIDevice *dev,
                                 PCIHotplugState state);
 
@@ -592,6 +626,9 @@ static void piix4_acpi_system_hot_add_init(PCIBus *bus, PIIX4PMState *s)
     register_ioport_write(PCI_EJ_BASE, 4, 4, pciej_write, bus);
     register_ioport_read(PCI_EJ_BASE, 4, 4,  pciej_read, bus);
 
+    register_ioport_write(PCI_RMV_BASE, 4, 4, pcirmv_write, s);
+    register_ioport_read(PCI_RMV_BASE, 4, 4,  pcirmv_read, s);
+
     pci_bus_hotplug(bus, piix4_device_hotplug, &s->dev.qdev);
 }
 

hw/msi.c

@@ -255,7 +255,6 @@ void msi_write_config(PCIDevice *dev, uint32_t addr, uint32_t val, int len)
     uint8_t log_max_vecs;
     unsigned int vector;
     uint32_t pending;
-    int i;
 
     if (!ranges_overlap(addr, len, dev->msi_cap, msi_cap_sizeof(flags))) {
         return;
@@ -296,9 +295,7 @@ void msi_write_config(PCIDevice *dev, uint32_t addr, uint32_t val, int len)
      *   from using its INTx# pin (if implemented) to request
      *   service (MSI, MSI-X, and INTx# are mutually exclusive).
      */
-    for (i = 0; i < PCI_NUM_PINS; ++i) {
+    pci_device_deassert_intx(dev);
-        qemu_set_irq(dev->irq[i], 0);
-    }
 
     /*
      * nr_vectors might be set bigger than capable. So clamp it.

hw/msix.c

@@ -159,7 +159,6 @@ void msix_write_config(PCIDevice *dev, uint32_t addr,
 {
     unsigned enable_pos = dev->msix_cap + MSIX_CONTROL_OFFSET;
     int vector;
-    int i;
 
     if (!range_covers_byte(addr, len, enable_pos)) {
         return;
@@ -169,9 +168,7 @@ void msix_write_config(PCIDevice *dev, uint32_t addr,
         return;
     }
 
-    for (i = 0; i < PCI_NUM_PINS; ++i) {
+    pci_device_deassert_intx(dev);
-        qemu_set_irq(dev->irq[i], 0);
-    }
 
     if (msix_function_masked(dev)) {
         return;

hw/pci.c

@@ -137,6 +137,14 @@ static void pci_update_irq_status(PCIDevice *dev)
     }
 }
 
+void pci_device_deassert_intx(PCIDevice *dev)
+{
+    int i;
+    for (i = 0; i < PCI_NUM_PINS; ++i) {
+        qemu_set_irq(dev->irq[i], 0);
+    }
+}
+
 /*
  * This function is called on #RST and FLR.
  * FLR if PCI_EXP_DEVCTL_BCR_FLR is set
@@ -152,6 +160,7 @@ void pci_device_reset(PCIDevice *dev)
 
     dev->irq_state = 0;
     pci_update_irq_status(dev);
+    pci_device_deassert_intx(dev);
     /* Clear all writeable bits */
     pci_word_test_and_clear_mask(dev->config + PCI_COMMAND,
                                  pci_get_word(dev->wmask + PCI_COMMAND) |
@@ -2032,10 +2041,13 @@ static char *pcibus_get_dev_path(DeviceState *dev)
      * domain:Bus:Slot.Func for systems without nested PCI bridges.
      * Slot.Function list specifies the slot and function numbers for all
      * devices on the path from root to the specific device. */
-    int domain_len = strlen("DDDD:00");
+    char domain[] = "DDDD:00";
-    int slot_len = strlen(":SS.F");
+    char slot[] = ":SS.F";
+    int domain_len = sizeof domain - 1 /* For '\0' */;
+    int slot_len = sizeof slot - 1 /* For '\0' */;
     int path_len;
     char *path, *p;
+    int s;
 
     /* Calculate # of slots on path between device and root. */;
     slot_depth = 0;
@@ -2046,18 +2058,23 @@ static char *pcibus_get_dev_path(DeviceState *dev)
     path_len = domain_len + slot_len * slot_depth;
 
     /* Allocate memory, fill in the terminating null byte. */
-    path = malloc(path_len + 1 /* For '\0' */);
+    path = qemu_malloc(path_len + 1 /* For '\0' */);
     path[path_len] = '\0';
 
     /* First field is the domain. */
-    snprintf(path, domain_len, "%04x:00", pci_find_domain(d->bus));
+    s = snprintf(domain, sizeof domain, "%04x:00", pci_find_domain(d->bus));
+    assert(s == domain_len);
+    memcpy(path, domain, domain_len);
 
     /* Fill in slot numbers. We walk up from device to root, so need to print
      * them in the reverse order, last to first. */
     p = path + path_len;
     for (t = d; t; t = t->bus->parent_dev) {
         p -= slot_len;
-        snprintf(p, slot_len, ":%02x.%x", PCI_SLOT(t->devfn), PCI_FUNC(d->devfn));
+        s = snprintf(slot, sizeof slot, ":%02x.%x",
+                     PCI_SLOT(t->devfn), PCI_FUNC(d->devfn));
+        assert(s == slot_len);
+        memcpy(p, slot, slot_len);
     }
 
     return path;

hw/pci.h

@@ -264,6 +264,8 @@ void do_pci_info_print(Monitor *mon, const QObject *data);
 void do_pci_info(Monitor *mon, QObject **ret_data);
 void pci_bridge_update_mappings(PCIBus *b);
 
+void pci_device_deassert_intx(PCIDevice *dev);
+
 static inline void
 pci_set_byte(uint8_t *config, uint8_t val)
 {

hw/virtio-serial-bus.c

@@ -769,10 +769,16 @@ VirtIODevice *virtio_serial_init(DeviceState *dev, uint32_t max_nr_ports)
     /* Add a queue for guest to host transfers for port 0 (backward compat) */
     vser->ovqs[0] = virtio_add_queue(vdev, 128, handle_output);
 
+    /* TODO: host to guest notifications can get dropped
+     * if the queue fills up. Implement queueing in host,
+     * this might also make it possible to reduce the control
+     * queue size: as guest preposts buffers there,
+     * this will save 4Kbyte of guest memory per entry. */
+
     /* control queue: host to guest */
-    vser->c_ivq = virtio_add_queue(vdev, 16, control_in);
+    vser->c_ivq = virtio_add_queue(vdev, 32, control_in);
     /* control queue: guest to host */
-    vser->c_ovq = virtio_add_queue(vdev, 16, control_out);
+    vser->c_ovq = virtio_add_queue(vdev, 32, control_out);
 
     for (i = 1; i < vser->bus->max_nr_ports; i++) {
         /* Add a per-port queue for host to guest transfers */

migration.c

@@ -88,6 +88,10 @@ int do_migrate(Monitor *mon, const QDict *qdict, QObject **ret_data)
         return -1;
     }
 
+    if (qemu_savevm_state_blocked(mon)) {
+        return -1;
+    }
+
     if (strstart(uri, "tcp:", &p)) {
         s = tcp_start_outgoing_migration(mon, p, max_throttle, detach,
                                          blk, inc);

savevm.c

@@ -1401,19 +1401,13 @@ static int vmstate_load(QEMUFile *f, SaveStateEntry *se, int version_id)
     return vmstate_load_state(f, se->vmsd, se->opaque, version_id);
 }
 
-static int vmstate_save(QEMUFile *f, SaveStateEntry *se)
+static void vmstate_save(QEMUFile *f, SaveStateEntry *se)
 {
-    if (se->no_migrate) {
-        return -1;
-    }
-
     if (!se->vmsd) {         /* Old style */
         se->save_state(f, se->opaque);
-        return 0;
+        return;
     }
     vmstate_save_state(f,se->vmsd, se->opaque);
-
-    return 0;
 }
 
 #define QEMU_VM_FILE_MAGIC           0x5145564d
@@ -1427,6 +1421,20 @@ static int vmstate_save(QEMUFile *f, SaveStateEntry *se)
 #define QEMU_VM_SECTION_FULL         0x04
 #define QEMU_VM_SUBSECTION           0x05
 
+bool qemu_savevm_state_blocked(Monitor *mon)
+{
+    SaveStateEntry *se;
+
+    QTAILQ_FOREACH(se, &savevm_handlers, entry) {
+        if (se->no_migrate) {
+            monitor_printf(mon, "state blocked by non-migratable device '%s'\n",
+                           se->idstr);
+            return true;
+        }
+    }
+    return false;
+}
+
 int qemu_savevm_state_begin(Monitor *mon, QEMUFile *f, int blk_enable,
                             int shared)
 {
@@ -1508,7 +1516,6 @@ int qemu_savevm_state_iterate(Monitor *mon, QEMUFile *f)
 int qemu_savevm_state_complete(Monitor *mon, QEMUFile *f)
 {
     SaveStateEntry *se;
-    int r;
 
     cpu_synchronize_all_states();
 
@@ -1541,11 +1548,7 @@ int qemu_savevm_state_complete(Monitor *mon, QEMUFile *f)
         qemu_put_be32(f, se->instance_id);
         qemu_put_be32(f, se->version_id);
 
-        r = vmstate_save(f, se);
+        vmstate_save(f, se);
-        if (r < 0) {
-            monitor_printf(mon, "cannot migrate with device '%s'\n", se->idstr);
-            return r;
-        }
     }
 
     qemu_put_byte(f, QEMU_VM_EOF);
@@ -1575,6 +1578,11 @@ static int qemu_savevm_state(Monitor *mon, QEMUFile *f)
     saved_vm_running = vm_running;
     vm_stop(0);
 
+    if (qemu_savevm_state_blocked(mon)) {
+        ret = -EINVAL;
+        goto out;
+    }
+
     ret = qemu_savevm_state_begin(mon, f, 0, 0);
     if (ret < 0)
         goto out;
@@ -1692,6 +1700,10 @@ int qemu_loadvm_state(QEMUFile *f)
     unsigned int v;
     int ret;
 
+    if (qemu_savevm_state_blocked(default_mon)) {
+        return -EINVAL;
+    }
+
     v = qemu_get_be32(f);
     if (v != QEMU_VM_FILE_MAGIC)
         return -EINVAL;

sysemu.h

@@ -75,6 +75,7 @@ void qemu_announce_self(void);
 
 void main_loop_wait(int nonblocking);
 
+bool qemu_savevm_state_blocked(Monitor *mon);
 int qemu_savevm_state_begin(Monitor *mon, QEMUFile *f, int blk_enable,
                             int shared);
 int qemu_savevm_state_iterate(Monitor *mon, QEMUFile *f);