#include "virtio-net.h"
#include "virtio-serial.h"
#include "virtio-scsi.h"
-#include "pci.h"
-#include "qemu-error.h"
-#include "msi.h"
-#include "msix.h"
-#include "net.h"
+#include "pci/pci.h"
+#include "qemu/error-report.h"
+#include "pci/msi.h"
+#include "pci/msix.h"
#include "loader.h"
-#include "kvm.h"
-#include "blockdev.h"
+#include "sysemu/kvm.h"
+#include "sysemu/blockdev.h"
#include "virtio-pci.h"
-#include "range.h"
+#include "qemu/range.h"
/* from Linux's linux/virtio_pci.h */
bool virtio_is_big_endian(void);
/* virtio device */
+/* DeviceState to VirtIOPCIProxy. For use off data-path. TODO: use QOM. */
+static inline VirtIOPCIProxy *to_virtio_pci_proxy(DeviceState *d)
+{
+ return container_of(d, VirtIOPCIProxy, pci_dev.qdev);
+}
-static void virtio_pci_notify(void *opaque, uint16_t vector)
+/* DeviceState to VirtIOPCIProxy. Note: used on datapath,
+ * be careful and test performance if you change this.
+ */
+static inline VirtIOPCIProxy *to_virtio_pci_proxy_fast(DeviceState *d)
{
- VirtIOPCIProxy *proxy = opaque;
+ return container_of(d, VirtIOPCIProxy, pci_dev.qdev);
+}
+
+static void virtio_pci_notify(DeviceState *d, uint16_t vector)
+{
+ VirtIOPCIProxy *proxy = to_virtio_pci_proxy_fast(d);
if (msix_enabled(&proxy->pci_dev))
msix_notify(&proxy->pci_dev, vector);
else
qemu_set_irq(proxy->pci_dev.irq[0], proxy->vdev->isr & 1);
}
-static void virtio_pci_save_config(void * opaque, QEMUFile *f)
+static void virtio_pci_save_config(DeviceState *d, QEMUFile *f)
{
- VirtIOPCIProxy *proxy = opaque;
+ VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
pci_device_save(&proxy->pci_dev, f);
msix_save(&proxy->pci_dev, f);
if (msix_present(&proxy->pci_dev))
qemu_put_be16(f, proxy->vdev->config_vector);
}
-static void virtio_pci_save_queue(void * opaque, int n, QEMUFile *f)
+static void virtio_pci_save_queue(DeviceState *d, int n, QEMUFile *f)
{
- VirtIOPCIProxy *proxy = opaque;
+ VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
if (msix_present(&proxy->pci_dev))
qemu_put_be16(f, virtio_queue_vector(proxy->vdev, n));
}
-static int virtio_pci_load_config(void * opaque, QEMUFile *f)
+static int virtio_pci_load_config(DeviceState *d, QEMUFile *f)
{
- VirtIOPCIProxy *proxy = opaque;
+ VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
int ret;
ret = pci_device_load(&proxy->pci_dev, f);
if (ret) {
return ret;
}
+ msix_unuse_all_vectors(&proxy->pci_dev);
msix_load(&proxy->pci_dev, f);
if (msix_present(&proxy->pci_dev)) {
qemu_get_be16s(f, &proxy->vdev->config_vector);
return 0;
}
-static int virtio_pci_load_queue(void * opaque, int n, QEMUFile *f)
+static int virtio_pci_load_queue(DeviceState *d, int n, QEMUFile *f)
{
- VirtIOPCIProxy *proxy = opaque;
+ VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
uint16_t vector;
if (msix_present(&proxy->pci_dev)) {
qemu_get_be16s(f, &vector);
}
static int virtio_pci_set_host_notifier_internal(VirtIOPCIProxy *proxy,
- int n, bool assign)
+ int n, bool assign, bool set_handler)
{
VirtQueue *vq = virtio_get_queue(proxy->vdev, n);
EventNotifier *notifier = virtio_queue_get_host_notifier(vq);
__func__, r);
return r;
}
+ virtio_queue_set_host_notifier_fd_handler(vq, true, set_handler);
memory_region_add_eventfd(&proxy->bar, VIRTIO_PCI_QUEUE_NOTIFY, 2,
- true, n, event_notifier_get_fd(notifier));
+ true, n, notifier);
} else {
memory_region_del_eventfd(&proxy->bar, VIRTIO_PCI_QUEUE_NOTIFY, 2,
- true, n, event_notifier_get_fd(notifier));
- /* Handle the race condition where the guest kicked and we deassigned
- * before we got around to handling the kick.
- */
- if (event_notifier_test_and_clear(notifier)) {
- virtio_queue_notify_vq(vq);
- }
-
+ true, n, notifier);
+ virtio_queue_set_host_notifier_fd_handler(vq, false, false);
event_notifier_cleanup(notifier);
}
return r;
}
-static void virtio_pci_host_notifier_read(void *opaque)
-{
- VirtQueue *vq = opaque;
- EventNotifier *n = virtio_queue_get_host_notifier(vq);
- if (event_notifier_test_and_clear(n)) {
- virtio_queue_notify_vq(vq);
- }
-}
-
-static void virtio_pci_set_host_notifier_fd_handler(VirtIOPCIProxy *proxy,
- int n, bool assign)
-{
- VirtQueue *vq = virtio_get_queue(proxy->vdev, n);
- EventNotifier *notifier = virtio_queue_get_host_notifier(vq);
- if (assign) {
- qemu_set_fd_handler(event_notifier_get_fd(notifier),
- virtio_pci_host_notifier_read, NULL, vq);
- } else {
- qemu_set_fd_handler(event_notifier_get_fd(notifier),
- NULL, NULL, NULL);
- }
-}
-
static void virtio_pci_start_ioeventfd(VirtIOPCIProxy *proxy)
{
int n, r;
continue;
}
- r = virtio_pci_set_host_notifier_internal(proxy, n, true);
+ r = virtio_pci_set_host_notifier_internal(proxy, n, true, true);
if (r < 0) {
goto assign_error;
}
-
- virtio_pci_set_host_notifier_fd_handler(proxy, n, true);
}
proxy->ioeventfd_started = true;
return;
continue;
}
- virtio_pci_set_host_notifier_fd_handler(proxy, n, false);
- r = virtio_pci_set_host_notifier_internal(proxy, n, false);
+ r = virtio_pci_set_host_notifier_internal(proxy, n, false, false);
assert(r >= 0);
}
proxy->ioeventfd_started = false;
continue;
}
- virtio_pci_set_host_notifier_fd_handler(proxy, n, false);
- r = virtio_pci_set_host_notifier_internal(proxy, n, false);
+ r = virtio_pci_set_host_notifier_internal(proxy, n, false, false);
assert(r >= 0);
}
proxy->ioeventfd_started = false;
void virtio_pci_reset(DeviceState *d)
{
- VirtIOPCIProxy *proxy = container_of(d, VirtIOPCIProxy, pci_dev.qdev);
+ VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
virtio_pci_stop_ioeventfd(proxy);
virtio_reset(proxy->vdev);
+ msix_unuse_all_vectors(&proxy->pci_dev);
proxy->flags &= ~VIRTIO_PCI_FLAG_BUS_MASTER_BUG;
}
{
VirtIOPCIProxy *proxy = opaque;
VirtIODevice *vdev = proxy->vdev;
- target_phys_addr_t pa;
+ hwaddr pa;
switch (addr) {
case VIRTIO_PCI_GUEST_FEATURES:
virtio_set_features(vdev, val);
break;
case VIRTIO_PCI_QUEUE_PFN:
- pa = (target_phys_addr_t)val << VIRTIO_PCI_QUEUE_ADDR_SHIFT;
+ pa = (hwaddr)val << VIRTIO_PCI_QUEUE_ADDR_SHIFT;
if (pa == 0) {
virtio_pci_stop_ioeventfd(proxy);
virtio_reset(proxy->vdev);
return ret;
}
-static uint32_t virtio_pci_config_readb(void *opaque, uint32_t addr)
-{
- VirtIOPCIProxy *proxy = opaque;
- uint32_t config = VIRTIO_PCI_CONFIG(&proxy->pci_dev);
- if (addr < config)
- return virtio_ioport_read(proxy, addr);
- addr -= config;
- return virtio_config_readb(proxy->vdev, addr);
-}
-
-static uint32_t virtio_pci_config_readw(void *opaque, uint32_t addr)
-{
- VirtIOPCIProxy *proxy = opaque;
- uint32_t config = VIRTIO_PCI_CONFIG(&proxy->pci_dev);
- uint16_t val;
- if (addr < config)
- return virtio_ioport_read(proxy, addr);
- addr -= config;
- val = virtio_config_readw(proxy->vdev, addr);
- if (virtio_is_big_endian()) {
- /*
- * virtio is odd, ioports are LE but config space is target native
- * endian. However, in qemu, all PIO is LE, so we need to re-swap
- * on BE targets
- */
- val = bswap16(val);
- }
- return val;
-}
-
-static uint32_t virtio_pci_config_readl(void *opaque, uint32_t addr)
-{
- VirtIOPCIProxy *proxy = opaque;
- uint32_t config = VIRTIO_PCI_CONFIG(&proxy->pci_dev);
- uint32_t val;
- if (addr < config)
- return virtio_ioport_read(proxy, addr);
- addr -= config;
- val = virtio_config_readl(proxy->vdev, addr);
- if (virtio_is_big_endian()) {
- val = bswap32(val);
- }
- return val;
-}
-
-static void virtio_pci_config_writeb(void *opaque, uint32_t addr, uint32_t val)
+static uint64_t virtio_pci_config_read(void *opaque, hwaddr addr,
+ unsigned size)
{
VirtIOPCIProxy *proxy = opaque;
uint32_t config = VIRTIO_PCI_CONFIG(&proxy->pci_dev);
+ uint64_t val = 0;
if (addr < config) {
- virtio_ioport_write(proxy, addr, val);
- return;
+ return virtio_ioport_read(proxy, addr);
}
addr -= config;
- virtio_config_writeb(proxy->vdev, addr, val);
-}
-static void virtio_pci_config_writew(void *opaque, uint32_t addr, uint32_t val)
-{
- VirtIOPCIProxy *proxy = opaque;
- uint32_t config = VIRTIO_PCI_CONFIG(&proxy->pci_dev);
- if (addr < config) {
- virtio_ioport_write(proxy, addr, val);
- return;
- }
- addr -= config;
- if (virtio_is_big_endian()) {
- val = bswap16(val);
+ switch (size) {
+ case 1:
+ val = virtio_config_readb(proxy->vdev, addr);
+ break;
+ case 2:
+ val = virtio_config_readw(proxy->vdev, addr);
+ if (virtio_is_big_endian()) {
+ val = bswap16(val);
+ }
+ break;
+ case 4:
+ val = virtio_config_readl(proxy->vdev, addr);
+ if (virtio_is_big_endian()) {
+ val = bswap32(val);
+ }
+ break;
}
- virtio_config_writew(proxy->vdev, addr, val);
+ return val;
}
-static void virtio_pci_config_writel(void *opaque, uint32_t addr, uint32_t val)
+static void virtio_pci_config_write(void *opaque, hwaddr addr,
+ uint64_t val, unsigned size)
{
VirtIOPCIProxy *proxy = opaque;
uint32_t config = VIRTIO_PCI_CONFIG(&proxy->pci_dev);
return;
}
addr -= config;
- if (virtio_is_big_endian()) {
- val = bswap32(val);
+ /*
+ * Virtio-PCI is odd. Ioports are LE but config space is target native
+ * endian.
+ */
+ switch (size) {
+ case 1:
+ virtio_config_writeb(proxy->vdev, addr, val);
+ break;
+ case 2:
+ if (virtio_is_big_endian()) {
+ val = bswap16(val);
+ }
+ virtio_config_writew(proxy->vdev, addr, val);
+ break;
+ case 4:
+ if (virtio_is_big_endian()) {
+ val = bswap32(val);
+ }
+ virtio_config_writel(proxy->vdev, addr, val);
+ break;
}
- virtio_config_writel(proxy->vdev, addr, val);
}
-static const MemoryRegionPortio virtio_portio[] = {
- { 0, 0x10000, 1, .write = virtio_pci_config_writeb, },
- { 0, 0x10000, 2, .write = virtio_pci_config_writew, },
- { 0, 0x10000, 4, .write = virtio_pci_config_writel, },
- { 0, 0x10000, 1, .read = virtio_pci_config_readb, },
- { 0, 0x10000, 2, .read = virtio_pci_config_readw, },
- { 0, 0x10000, 4, .read = virtio_pci_config_readl, },
- PORTIO_END_OF_LIST()
-};
-
static const MemoryRegionOps virtio_pci_config_ops = {
- .old_portio = virtio_portio,
+ .read = virtio_pci_config_read,
+ .write = virtio_pci_config_write,
+ .impl = {
+ .min_access_size = 1,
+ .max_access_size = 4,
+ },
.endianness = DEVICE_LITTLE_ENDIAN,
};
}
}
-static unsigned virtio_pci_get_features(void *opaque)
+static unsigned virtio_pci_get_features(DeviceState *d)
{
- VirtIOPCIProxy *proxy = opaque;
+ VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
return proxy->host_features;
}
-static void virtio_pci_guest_notifier_read(void *opaque)
-{
- VirtQueue *vq = opaque;
- EventNotifier *n = virtio_queue_get_guest_notifier(vq);
- if (event_notifier_test_and_clear(n)) {
- virtio_irq(vq);
- }
-}
-
static int kvm_virtio_pci_vq_vector_use(VirtIOPCIProxy *proxy,
unsigned int queue_no,
unsigned int vector,
MSIMessage msg)
{
- VirtQueue *vq = virtio_get_queue(proxy->vdev, queue_no);
VirtIOIRQFD *irqfd = &proxy->vector_irqfd[vector];
- int fd, ret;
-
- fd = event_notifier_get_fd(virtio_queue_get_guest_notifier(vq));
+ int ret;
if (irqfd->users == 0) {
ret = kvm_irqchip_add_msi_route(kvm_state, msg);
irqfd->virq = ret;
}
irqfd->users++;
+ return 0;
+}
- ret = kvm_irqchip_add_irqfd(kvm_state, fd, irqfd->virq);
- if (ret < 0) {
- if (--irqfd->users == 0) {
- kvm_irqchip_release_virq(kvm_state, irqfd->virq);
- }
- return ret;
+static void kvm_virtio_pci_vq_vector_release(VirtIOPCIProxy *proxy,
+ unsigned int vector)
+{
+ VirtIOIRQFD *irqfd = &proxy->vector_irqfd[vector];
+ if (--irqfd->users == 0) {
+ kvm_irqchip_release_virq(kvm_state, irqfd->virq);
}
+}
- qemu_set_fd_handler(fd, NULL, NULL, NULL);
+static int kvm_virtio_pci_irqfd_use(VirtIOPCIProxy *proxy,
+ unsigned int queue_no,
+ unsigned int vector)
+{
+ VirtIOIRQFD *irqfd = &proxy->vector_irqfd[vector];
+ VirtQueue *vq = virtio_get_queue(proxy->vdev, queue_no);
+ EventNotifier *n = virtio_queue_get_guest_notifier(vq);
+ int ret;
+ ret = kvm_irqchip_add_irqfd_notifier(kvm_state, n, irqfd->virq);
+ return ret;
+}
+static void kvm_virtio_pci_irqfd_release(VirtIOPCIProxy *proxy,
+ unsigned int queue_no,
+ unsigned int vector)
+{
+ VirtQueue *vq = virtio_get_queue(proxy->vdev, queue_no);
+ EventNotifier *n = virtio_queue_get_guest_notifier(vq);
+ VirtIOIRQFD *irqfd = &proxy->vector_irqfd[vector];
+ int ret;
+
+ ret = kvm_irqchip_remove_irqfd_notifier(kvm_state, n, irqfd->virq);
+ assert(ret == 0);
+}
+
+static int kvm_virtio_pci_vector_use(VirtIOPCIProxy *proxy, int nvqs)
+{
+ PCIDevice *dev = &proxy->pci_dev;
+ VirtIODevice *vdev = proxy->vdev;
+ unsigned int vector;
+ int ret, queue_no;
+ MSIMessage msg;
+
+ for (queue_no = 0; queue_no < nvqs; queue_no++) {
+ if (!virtio_queue_get_num(vdev, queue_no)) {
+ break;
+ }
+ vector = virtio_queue_vector(vdev, queue_no);
+ if (vector >= msix_nr_vectors_allocated(dev)) {
+ continue;
+ }
+ msg = msix_get_message(dev, vector);
+ ret = kvm_virtio_pci_vq_vector_use(proxy, queue_no, vector, msg);
+ if (ret < 0) {
+ goto undo;
+ }
+ /* If guest supports masking, set up irqfd now.
+ * Otherwise, delay until unmasked in the frontend.
+ */
+ if (proxy->vdev->guest_notifier_mask) {
+ ret = kvm_virtio_pci_irqfd_use(proxy, queue_no, vector);
+ if (ret < 0) {
+ kvm_virtio_pci_vq_vector_release(proxy, vector);
+ goto undo;
+ }
+ }
+ }
return 0;
+
+undo:
+ while (--queue_no >= 0) {
+ vector = virtio_queue_vector(vdev, queue_no);
+ if (vector >= msix_nr_vectors_allocated(dev)) {
+ continue;
+ }
+ if (proxy->vdev->guest_notifier_mask) {
+ kvm_virtio_pci_irqfd_release(proxy, vector, queue_no);
+ }
+ kvm_virtio_pci_vq_vector_release(proxy, vector);
+ }
+ return ret;
}
-static void kvm_virtio_pci_vq_vector_release(VirtIOPCIProxy *proxy,
- unsigned int queue_no,
- unsigned int vector)
+static void kvm_virtio_pci_vector_release(VirtIOPCIProxy *proxy, int nvqs)
+{
+ PCIDevice *dev = &proxy->pci_dev;
+ VirtIODevice *vdev = proxy->vdev;
+ unsigned int vector;
+ int queue_no;
+
+ for (queue_no = 0; queue_no < nvqs; queue_no++) {
+ if (!virtio_queue_get_num(vdev, queue_no)) {
+ break;
+ }
+ vector = virtio_queue_vector(vdev, queue_no);
+ if (vector >= msix_nr_vectors_allocated(dev)) {
+ continue;
+ }
+ /* If guest supports masking, clean up irqfd now.
+ * Otherwise, it was cleaned when masked in the frontend.
+ */
+ if (proxy->vdev->guest_notifier_mask) {
+ kvm_virtio_pci_irqfd_release(proxy, vector, queue_no);
+ }
+ kvm_virtio_pci_vq_vector_release(proxy, vector);
+ }
+}
+
+static int kvm_virtio_pci_vq_vector_unmask(VirtIOPCIProxy *proxy,
+ unsigned int queue_no,
+ unsigned int vector,
+ MSIMessage msg)
{
VirtQueue *vq = virtio_get_queue(proxy->vdev, queue_no);
+ EventNotifier *n = virtio_queue_get_guest_notifier(vq);
VirtIOIRQFD *irqfd = &proxy->vector_irqfd[vector];
- int fd, ret;
-
- fd = event_notifier_get_fd(virtio_queue_get_guest_notifier(vq));
+ int ret;
- ret = kvm_irqchip_remove_irqfd(kvm_state, fd, irqfd->virq);
- assert(ret == 0);
+ if (irqfd->msg.data != msg.data || irqfd->msg.address != msg.address) {
+ ret = kvm_irqchip_update_msi_route(kvm_state, irqfd->virq, msg);
+ if (ret < 0) {
+ return ret;
+ }
+ }
- if (--irqfd->users == 0) {
- kvm_irqchip_release_virq(kvm_state, irqfd->virq);
+ /* If guest supports masking, irqfd is already setup, unmask it.
+ * Otherwise, set it up now.
+ */
+ if (proxy->vdev->guest_notifier_mask) {
+ proxy->vdev->guest_notifier_mask(proxy->vdev, queue_no, false);
+ /* Test after unmasking to avoid losing events. */
+ if (proxy->vdev->guest_notifier_pending &&
+ proxy->vdev->guest_notifier_pending(proxy->vdev, queue_no)) {
+ event_notifier_set(n);
+ }
+ } else {
+ ret = kvm_virtio_pci_irqfd_use(proxy, queue_no, vector);
}
+ return ret;
+}
- qemu_set_fd_handler(fd, virtio_pci_guest_notifier_read, NULL, vq);
+static void kvm_virtio_pci_vq_vector_mask(VirtIOPCIProxy *proxy,
+ unsigned int queue_no,
+ unsigned int vector)
+{
+ /* If guest supports masking, keep irqfd but mask it.
+ * Otherwise, clean it up now.
+ */
+ if (proxy->vdev->guest_notifier_mask) {
+ proxy->vdev->guest_notifier_mask(proxy->vdev, queue_no, true);
+ } else {
+ kvm_virtio_pci_irqfd_release(proxy, vector, queue_no);
+ }
}
-static int kvm_virtio_pci_vector_use(PCIDevice *dev, unsigned vector,
+static int kvm_virtio_pci_vector_unmask(PCIDevice *dev, unsigned vector,
MSIMessage msg)
{
VirtIOPCIProxy *proxy = container_of(dev, VirtIOPCIProxy, pci_dev);
VirtIODevice *vdev = proxy->vdev;
int ret, queue_no;
- for (queue_no = 0; queue_no < VIRTIO_PCI_QUEUE_MAX; queue_no++) {
+ for (queue_no = 0; queue_no < proxy->nvqs_with_notifiers; queue_no++) {
if (!virtio_queue_get_num(vdev, queue_no)) {
break;
}
if (virtio_queue_vector(vdev, queue_no) != vector) {
continue;
}
- ret = kvm_virtio_pci_vq_vector_use(proxy, queue_no, vector, msg);
+ ret = kvm_virtio_pci_vq_vector_unmask(proxy, queue_no, vector, msg);
if (ret < 0) {
goto undo;
}
if (virtio_queue_vector(vdev, queue_no) != vector) {
continue;
}
- kvm_virtio_pci_vq_vector_release(proxy, queue_no, vector);
+ kvm_virtio_pci_vq_vector_mask(proxy, queue_no, vector);
}
return ret;
}
-static void kvm_virtio_pci_vector_release(PCIDevice *dev, unsigned vector)
+static void kvm_virtio_pci_vector_mask(PCIDevice *dev, unsigned vector)
{
VirtIOPCIProxy *proxy = container_of(dev, VirtIOPCIProxy, pci_dev);
VirtIODevice *vdev = proxy->vdev;
int queue_no;
- for (queue_no = 0; queue_no < VIRTIO_PCI_QUEUE_MAX; queue_no++) {
+ for (queue_no = 0; queue_no < proxy->nvqs_with_notifiers; queue_no++) {
if (!virtio_queue_get_num(vdev, queue_no)) {
break;
}
if (virtio_queue_vector(vdev, queue_no) != vector) {
continue;
}
- kvm_virtio_pci_vq_vector_release(proxy, queue_no, vector);
+ kvm_virtio_pci_vq_vector_mask(proxy, queue_no, vector);
}
}
-static int virtio_pci_set_guest_notifier(void *opaque, int n, bool assign)
+static void kvm_virtio_pci_vector_poll(PCIDevice *dev,
+ unsigned int vector_start,
+ unsigned int vector_end)
{
- VirtIOPCIProxy *proxy = opaque;
+ VirtIOPCIProxy *proxy = container_of(dev, VirtIOPCIProxy, pci_dev);
+ VirtIODevice *vdev = proxy->vdev;
+ int queue_no;
+ unsigned int vector;
+ EventNotifier *notifier;
+ VirtQueue *vq;
+
+ for (queue_no = 0; queue_no < proxy->nvqs_with_notifiers; queue_no++) {
+ if (!virtio_queue_get_num(vdev, queue_no)) {
+ break;
+ }
+ vector = virtio_queue_vector(vdev, queue_no);
+ if (vector < vector_start || vector >= vector_end ||
+ !msix_is_masked(dev, vector)) {
+ continue;
+ }
+ vq = virtio_get_queue(vdev, queue_no);
+ notifier = virtio_queue_get_guest_notifier(vq);
+ if (vdev->guest_notifier_pending) {
+ if (vdev->guest_notifier_pending(vdev, queue_no)) {
+ msix_set_pending(dev, vector);
+ }
+ } else if (event_notifier_test_and_clear(notifier)) {
+ msix_set_pending(dev, vector);
+ }
+ }
+}
+
+static int virtio_pci_set_guest_notifier(DeviceState *d, int n, bool assign,
+ bool with_irqfd)
+{
+ VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
VirtQueue *vq = virtio_get_queue(proxy->vdev, n);
EventNotifier *notifier = virtio_queue_get_guest_notifier(vq);
if (r < 0) {
return r;
}
- qemu_set_fd_handler(event_notifier_get_fd(notifier),
- virtio_pci_guest_notifier_read, NULL, vq);
+ virtio_queue_set_guest_notifier_fd_handler(vq, true, with_irqfd);
} else {
- qemu_set_fd_handler(event_notifier_get_fd(notifier),
- NULL, NULL, NULL);
- /* Test and clear notifier before closing it,
- * in case poll callback didn't have time to run. */
- virtio_pci_guest_notifier_read(vq);
+ virtio_queue_set_guest_notifier_fd_handler(vq, false, with_irqfd);
event_notifier_cleanup(notifier);
}
return 0;
}
-static bool virtio_pci_query_guest_notifiers(void *opaque)
+static bool virtio_pci_query_guest_notifiers(DeviceState *d)
{
- VirtIOPCIProxy *proxy = opaque;
+ VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
return msix_enabled(&proxy->pci_dev);
}
-static int virtio_pci_set_guest_notifiers(void *opaque, bool assign)
+static int virtio_pci_set_guest_notifiers(DeviceState *d, int nvqs, bool assign)
{
- VirtIOPCIProxy *proxy = opaque;
+ VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
VirtIODevice *vdev = proxy->vdev;
int r, n;
+ bool with_irqfd = msix_enabled(&proxy->pci_dev) &&
+ kvm_msi_via_irqfd_enabled();
+
+ nvqs = MIN(nvqs, VIRTIO_PCI_QUEUE_MAX);
+
+ /* When deassigning, pass a consistent nvqs value
+ * to avoid leaking notifiers.
+ */
+ assert(assign || nvqs == proxy->nvqs_with_notifiers);
+
+ proxy->nvqs_with_notifiers = nvqs;
/* Must unset vector notifier while guest notifier is still assigned */
- if (kvm_irqchip_in_kernel() && !assign) {
+ if (proxy->vector_irqfd && !assign) {
msix_unset_vector_notifiers(&proxy->pci_dev);
+ kvm_virtio_pci_vector_release(proxy, nvqs);
g_free(proxy->vector_irqfd);
proxy->vector_irqfd = NULL;
}
- for (n = 0; n < VIRTIO_PCI_QUEUE_MAX; n++) {
+ for (n = 0; n < nvqs; n++) {
if (!virtio_queue_get_num(vdev, n)) {
break;
}
- r = virtio_pci_set_guest_notifier(opaque, n, assign);
+ r = virtio_pci_set_guest_notifier(d, n, assign,
+ kvm_msi_via_irqfd_enabled());
if (r < 0) {
goto assign_error;
}
}
/* Must set vector notifier after guest notifier has been assigned */
- if (kvm_irqchip_in_kernel() && assign) {
+ if (with_irqfd && assign) {
proxy->vector_irqfd =
g_malloc0(sizeof(*proxy->vector_irqfd) *
msix_nr_vectors_allocated(&proxy->pci_dev));
- r = msix_set_vector_notifiers(&proxy->pci_dev,
- kvm_virtio_pci_vector_use,
- kvm_virtio_pci_vector_release);
+ r = kvm_virtio_pci_vector_use(proxy, nvqs);
if (r < 0) {
goto assign_error;
}
+ r = msix_set_vector_notifiers(&proxy->pci_dev,
+ kvm_virtio_pci_vector_unmask,
+ kvm_virtio_pci_vector_mask,
+ kvm_virtio_pci_vector_poll);
+ if (r < 0) {
+ goto notifiers_error;
+ }
}
return 0;
+notifiers_error:
+ assert(assign);
+ kvm_virtio_pci_vector_release(proxy, nvqs);
+
assign_error:
/* We get here on assignment failure. Recover by undoing for VQs 0 .. n. */
assert(assign);
while (--n >= 0) {
- virtio_pci_set_guest_notifier(opaque, n, !assign);
+ virtio_pci_set_guest_notifier(d, n, !assign, with_irqfd);
}
return r;
}
-static int virtio_pci_set_host_notifier(void *opaque, int n, bool assign)
+static int virtio_pci_set_host_notifier(DeviceState *d, int n, bool assign)
{
- VirtIOPCIProxy *proxy = opaque;
+ VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
/* Stop using ioeventfd for virtqueue kick if the device starts using host
* notifiers. This makes it easy to avoid stepping on each others' toes.
* currently only stops on status change away from ok,
* reset, vmstop and such. If we do add code to start here,
* need to check vmstate, device state etc. */
- return virtio_pci_set_host_notifier_internal(proxy, n, assign);
+ return virtio_pci_set_host_notifier_internal(proxy, n, assign, false);
}
-static void virtio_pci_vmstate_change(void *opaque, bool running)
+static void virtio_pci_vmstate_change(DeviceState *d, bool running)
{
- VirtIOPCIProxy *proxy = opaque;
+ VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
if (running) {
/* Try to find out if the guest has bus master disabled, but is
pci_set_word(config + PCI_SUBSYSTEM_ID, vdev->device_id);
config[PCI_INTERRUPT_PIN] = 1;
- memory_region_init(&proxy->msix_bar, "virtio-msix", 4096);
- if (vdev->nvectors && !msix_init(&proxy->pci_dev, vdev->nvectors,
- &proxy->msix_bar, 1, 0)) {
- pci_register_bar(&proxy->pci_dev, 1, PCI_BASE_ADDRESS_SPACE_MEMORY,
- &proxy->msix_bar);
- } else
+ if (vdev->nvectors &&
+ msix_init_exclusive_bar(&proxy->pci_dev, vdev->nvectors, 1)) {
vdev->nvectors = 0;
+ }
proxy->pci_dev.config_write = virtio_write_config;
proxy->flags &= ~VIRTIO_PCI_FLAG_USE_IOEVENTFD;
}
- virtio_bind_device(vdev, &virtio_pci_bindings, proxy);
+ virtio_bind_device(vdev, &virtio_pci_bindings, DEVICE(proxy));
proxy->host_features |= 0x1 << VIRTIO_F_NOTIFY_ON_EMPTY;
proxy->host_features |= 0x1 << VIRTIO_F_BAD_FEATURE;
proxy->host_features = vdev->get_features(vdev, proxy->host_features);
return 0;
}
-static int virtio_exit_pci(PCIDevice *pci_dev)
+static void virtio_exit_pci(PCIDevice *pci_dev)
{
VirtIOPCIProxy *proxy = DO_UPCAST(VirtIOPCIProxy, pci_dev, pci_dev);
- int r;
memory_region_destroy(&proxy->bar);
- r = msix_uninit(pci_dev, &proxy->msix_bar);
- memory_region_destroy(&proxy->msix_bar);
- return r;
+ msix_uninit_exclusive_bar(pci_dev);
}
-static int virtio_blk_exit_pci(PCIDevice *pci_dev)
+static void virtio_blk_exit_pci(PCIDevice *pci_dev)
{
VirtIOPCIProxy *proxy = DO_UPCAST(VirtIOPCIProxy, pci_dev, pci_dev);
virtio_pci_stop_ioeventfd(proxy);
virtio_blk_exit(proxy->vdev);
- return virtio_exit_pci(pci_dev);
+ virtio_exit_pci(pci_dev);
}
static int virtio_serial_init_pci(PCIDevice *pci_dev)
return 0;
}
-static int virtio_serial_exit_pci(PCIDevice *pci_dev)
+static void virtio_serial_exit_pci(PCIDevice *pci_dev)
{
VirtIOPCIProxy *proxy = DO_UPCAST(VirtIOPCIProxy, pci_dev, pci_dev);
virtio_pci_stop_ioeventfd(proxy);
virtio_serial_exit(proxy->vdev);
- return virtio_exit_pci(pci_dev);
+ virtio_exit_pci(pci_dev);
}
static int virtio_net_init_pci(PCIDevice *pci_dev)
return 0;
}
-static int virtio_net_exit_pci(PCIDevice *pci_dev)
+static void virtio_net_exit_pci(PCIDevice *pci_dev)
{
VirtIOPCIProxy *proxy = DO_UPCAST(VirtIOPCIProxy, pci_dev, pci_dev);
virtio_pci_stop_ioeventfd(proxy);
virtio_net_exit(proxy->vdev);
- return virtio_exit_pci(pci_dev);
+ virtio_exit_pci(pci_dev);
}
static int virtio_balloon_init_pci(PCIDevice *pci_dev)
return 0;
}
-static int virtio_balloon_exit_pci(PCIDevice *pci_dev)
+static void virtio_balloon_exit_pci(PCIDevice *pci_dev)
{
VirtIOPCIProxy *proxy = DO_UPCAST(VirtIOPCIProxy, pci_dev, pci_dev);
virtio_pci_stop_ioeventfd(proxy);
virtio_balloon_exit(proxy->vdev);
- return virtio_exit_pci(pci_dev);
+ virtio_exit_pci(pci_dev);
+}
+
+static int virtio_rng_init_pci(PCIDevice *pci_dev)
+{
+ VirtIOPCIProxy *proxy = DO_UPCAST(VirtIOPCIProxy, pci_dev, pci_dev);
+ VirtIODevice *vdev;
+
+ if (proxy->rng.rng == NULL) {
+ proxy->rng.default_backend = RNG_RANDOM(object_new(TYPE_RNG_RANDOM));
+
+ object_property_add_child(OBJECT(pci_dev),
+ "default-backend",
+ OBJECT(proxy->rng.default_backend),
+ NULL);
+
+ object_property_set_link(OBJECT(pci_dev),
+ OBJECT(proxy->rng.default_backend),
+ "rng", NULL);
+ }
+
+ vdev = virtio_rng_init(&pci_dev->qdev, &proxy->rng);
+ if (!vdev) {
+ return -1;
+ }
+ virtio_init_pci(proxy, vdev);
+ return 0;
+}
+
+static void virtio_rng_exit_pci(PCIDevice *pci_dev)
+{
+ VirtIOPCIProxy *proxy = DO_UPCAST(VirtIOPCIProxy, pci_dev, pci_dev);
+
+ virtio_pci_stop_ioeventfd(proxy);
+ virtio_rng_exit(proxy->vdev);
+ virtio_exit_pci(pci_dev);
}
static Property virtio_blk_properties[] = {
#ifdef __linux__
DEFINE_PROP_BIT("scsi", VirtIOPCIProxy, blk.scsi, 0, true),
#endif
+ DEFINE_PROP_BIT("config-wce", VirtIOPCIProxy, blk.config_wce, 0, true),
DEFINE_PROP_BIT("ioeventfd", VirtIOPCIProxy, flags, VIRTIO_PCI_FLAG_USE_IOEVENTFD_BIT, true),
+#ifdef CONFIG_VIRTIO_BLK_DATA_PLANE
+ DEFINE_PROP_BIT("x-data-plane", VirtIOPCIProxy, blk.data_plane, 0, false),
+#endif
DEFINE_PROP_UINT32("vectors", VirtIOPCIProxy, nvectors, 2),
DEFINE_VIRTIO_BLK_FEATURES(VirtIOPCIProxy, host_features),
DEFINE_PROP_END_OF_LIST(),
dc->props = virtio_blk_properties;
}
-static TypeInfo virtio_blk_info = {
+static const TypeInfo virtio_blk_info = {
.name = "virtio-blk-pci",
.parent = TYPE_PCI_DEVICE,
.instance_size = sizeof(VirtIOPCIProxy),
dc->props = virtio_net_properties;
}
-static TypeInfo virtio_net_info = {
+static const TypeInfo virtio_net_info = {
.name = "virtio-net-pci",
.parent = TYPE_PCI_DEVICE,
.instance_size = sizeof(VirtIOPCIProxy),
dc->props = virtio_serial_properties;
}
-static TypeInfo virtio_serial_info = {
+static const TypeInfo virtio_serial_info = {
.name = "virtio-serial-pci",
.parent = TYPE_PCI_DEVICE,
.instance_size = sizeof(VirtIOPCIProxy),
dc->props = virtio_balloon_properties;
}
-static TypeInfo virtio_balloon_info = {
+static const TypeInfo virtio_balloon_info = {
.name = "virtio-balloon-pci",
.parent = TYPE_PCI_DEVICE,
.instance_size = sizeof(VirtIOPCIProxy),
.class_init = virtio_balloon_class_init,
};
+static void virtio_rng_initfn(Object *obj)
+{
+ PCIDevice *pci_dev = PCI_DEVICE(obj);
+ VirtIOPCIProxy *proxy = DO_UPCAST(VirtIOPCIProxy, pci_dev, pci_dev);
+
+ object_property_add_link(obj, "rng", TYPE_RNG_BACKEND,
+ (Object **)&proxy->rng.rng, NULL);
+}
+
+static Property virtio_rng_properties[] = {
+ DEFINE_VIRTIO_COMMON_FEATURES(VirtIOPCIProxy, host_features),
+ /* Set a default rate limit of 2^47 bytes per minute or roughly 2TB/s. If
+ you have an entropy source capable of generating more entropy than this
+ and you can pass it through via virtio-rng, then hats off to you. Until
+ then, this is unlimited for all practical purposes.
+ */
+ DEFINE_PROP_UINT64("max-bytes", VirtIOPCIProxy, rng.max_bytes, INT64_MAX),
+ DEFINE_PROP_UINT32("period", VirtIOPCIProxy, rng.period_ms, 1 << 16),
+ DEFINE_PROP_END_OF_LIST(),
+};
+
+static void virtio_rng_class_init(ObjectClass *klass, void *data)
+{
+ DeviceClass *dc = DEVICE_CLASS(klass);
+ PCIDeviceClass *k = PCI_DEVICE_CLASS(klass);
+
+ k->init = virtio_rng_init_pci;
+ k->exit = virtio_rng_exit_pci;
+ k->vendor_id = PCI_VENDOR_ID_REDHAT_QUMRANET;
+ k->device_id = PCI_DEVICE_ID_VIRTIO_RNG;
+ k->revision = VIRTIO_PCI_ABI_VERSION;
+ k->class_id = PCI_CLASS_OTHERS;
+ dc->reset = virtio_pci_reset;
+ dc->props = virtio_rng_properties;
+}
+
+static const TypeInfo virtio_rng_info = {
+ .name = "virtio-rng-pci",
+ .parent = TYPE_PCI_DEVICE,
+ .instance_size = sizeof(VirtIOPCIProxy),
+ .instance_init = virtio_rng_initfn,
+ .class_init = virtio_rng_class_init,
+};
+
static int virtio_scsi_init_pci(PCIDevice *pci_dev)
{
VirtIOPCIProxy *proxy = DO_UPCAST(VirtIOPCIProxy, pci_dev, pci_dev);
return -EINVAL;
}
- vdev->nvectors = proxy->nvectors;
+ vdev->nvectors = proxy->nvectors == DEV_NVECTORS_UNSPECIFIED
+ ? proxy->scsi.num_queues + 3
+ : proxy->nvectors;
virtio_init_pci(proxy, vdev);
/* make the actual value visible */
return 0;
}
-static int virtio_scsi_exit_pci(PCIDevice *pci_dev)
+static void virtio_scsi_exit_pci(PCIDevice *pci_dev)
{
VirtIOPCIProxy *proxy = DO_UPCAST(VirtIOPCIProxy, pci_dev, pci_dev);
virtio_scsi_exit(proxy->vdev);
- return virtio_exit_pci(pci_dev);
+ virtio_exit_pci(pci_dev);
}
static Property virtio_scsi_properties[] = {
- DEFINE_PROP_UINT32("vectors", VirtIOPCIProxy, nvectors, 2),
+ DEFINE_PROP_BIT("ioeventfd", VirtIOPCIProxy, flags, VIRTIO_PCI_FLAG_USE_IOEVENTFD_BIT, true),
+ DEFINE_PROP_UINT32("vectors", VirtIOPCIProxy, nvectors, DEV_NVECTORS_UNSPECIFIED),
DEFINE_VIRTIO_SCSI_PROPERTIES(VirtIOPCIProxy, host_features, scsi),
DEFINE_PROP_END_OF_LIST(),
};
dc->props = virtio_scsi_properties;
}
-static TypeInfo virtio_scsi_info = {
+static const TypeInfo virtio_scsi_info = {
.name = "virtio-scsi-pci",
.parent = TYPE_PCI_DEVICE,
.instance_size = sizeof(VirtIOPCIProxy),
type_register_static(&virtio_serial_info);
type_register_static(&virtio_balloon_info);
type_register_static(&virtio_scsi_info);
+ type_register_static(&virtio_rng_info);
}
type_init(virtio_pci_register_types)