[PATCH v14 01/11] virtio-iommu: Add skeleton

[Eric Auger <eric.auger@redhat.com>]

This patchs adds the skeleton for the virtio-iommu device.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>

---

v13 -> v14:
- use device_class_set_props
- updated copyright's year

v12 -> v13
- removed IOMMU_PCI_BUS_MAX and IOMMU_PCI_DEVFN_MAX

v11 -> v12:
- remove s_by_bus_num
- drop set_features (rely on default implementation) and
  acked_features

v9 -> v10:
- mutex initialized here
- initialize tail
- included hw/qdev-properties.h
- removed g_memdup
- removed s->config.domain_range.start = 0;

v9 -> v10:
- expose VIRTIO_IOMMU_F_MMIO feature
- s/domain_bits/domain_range struct
- change error codes
- enforce unmigratable
- Kconfig

v7 -> v8:
- expose VIRTIO_IOMMU_F_BYPASS and VIRTIO_F_VERSION_1
  features
- set_config dummy implementation + tracing
- add trace in get_features
- set the features on realize() and store the acked ones
- remove inclusion of linux/virtio_iommu.h

v6 -> v7:
- removed qapi-event.h include
- add primary_bus and associated property

v4 -> v5:
- use the new v0.5 terminology (domain, endpoint)
- add the event virtqueue

v3 -> v4:
- use page_size_mask instead of page_sizes
- added set_features()
- added some traces (reset, set_status, set_features)
- empty virtio_iommu_set_config() as the driver MUST NOT
  write to device configuration fields
- add get_config trace

v2 -> v3:
- rebase on 2.10-rc0, ie. use IOMMUMemoryRegion and remove
  iommu_ops.
- advertise VIRTIO_IOMMU_F_MAP_UNMAP feature
- page_sizes set to TARGET_PAGE_SIZE

Conflicts:
	hw/virtio/trace-events
---
 hw/virtio/Kconfig                |   5 +
 hw/virtio/Makefile.objs          |   1 +
 hw/virtio/trace-events           |   7 +
 hw/virtio/virtio-iommu.c         | 265 +++++++++++++++++++++++++++++++
 include/hw/virtio/virtio-iommu.h |  57 +++++++
 5 files changed, 335 insertions(+)
 create mode 100644 hw/virtio/virtio-iommu.c
 create mode 100644 include/hw/virtio/virtio-iommu.h

diff --git a/hw/virtio/Kconfig b/hw/virtio/Kconfig
index f87def27a6..d29525b36f 100644
--- a/hw/virtio/Kconfig
+++ b/hw/virtio/Kconfig
@@ -9,6 +9,11 @@ config VIRTIO_RNG
     default y
     depends on VIRTIO
 
+config VIRTIO_IOMMU
+    bool
+    default y
+    depends on VIRTIO
+
 config VIRTIO_PCI
     bool
     default y if PCI_DEVICES
diff --git a/hw/virtio/Makefile.objs b/hw/virtio/Makefile.objs
index de0f5fc39b..2fd9da7410 100644
--- a/hw/virtio/Makefile.objs
+++ b/hw/virtio/Makefile.objs
@@ -16,6 +16,7 @@ obj-$(call land,$(CONFIG_VIRTIO_CRYPTO),$(CONFIG_VIRTIO_PCI)) += virtio-crypto-p
 obj-$(CONFIG_VIRTIO_PMEM) += virtio-pmem.o
 common-obj-$(call land,$(CONFIG_VIRTIO_PMEM),$(CONFIG_VIRTIO_PCI)) += virtio-pmem-pci.o
 obj-$(call land,$(CONFIG_VHOST_USER_FS),$(CONFIG_VIRTIO_PCI)) += vhost-user-fs-pci.o
+obj-$(CONFIG_VIRTIO_IOMMU) += virtio-iommu.o
 obj-$(CONFIG_VHOST_VSOCK) += vhost-vsock.o
 
 ifeq ($(CONFIG_VIRTIO_PCI),y)
diff --git a/hw/virtio/trace-events b/hw/virtio/trace-events
index e28ba48da6..02d93d7f63 100644
--- a/hw/virtio/trace-events
+++ b/hw/virtio/trace-events
@@ -53,3 +53,10 @@ virtio_mmio_write_offset(uint64_t offset, uint64_t value) "virtio_mmio_write off
 virtio_mmio_guest_page(uint64_t size, int shift) "guest page size 0x%" PRIx64 " shift %d"
 virtio_mmio_queue_write(uint64_t value, int max_size) "mmio_queue write 0x%" PRIx64 " max %d"
 virtio_mmio_setting_irq(int level) "virtio_mmio setting IRQ %d"
+
+# hw/virtio/virtio-iommu.c
+virtio_iommu_device_reset(void) "reset!"
+virtio_iommu_get_features(uint64_t features) "device supports features=0x%"PRIx64
+virtio_iommu_device_status(uint8_t status) "driver status = %d"
+virtio_iommu_get_config(uint64_t page_size_mask, uint64_t start, uint64_t end, uint32_t domain_range, uint32_t probe_size) "page_size_mask=0x%"PRIx64" start=0x%"PRIx64" end=0x%"PRIx64" domain_range=%d probe_size=0x%x"
+virtio_iommu_set_config(uint64_t page_size_mask, uint64_t start, uint64_t end, uint32_t domain_range, uint32_t probe_size) "page_size_mask=0x%"PRIx64" start=0x%"PRIx64" end=0x%"PRIx64" domain_bits=%d probe_size=0x%x"
diff --git a/hw/virtio/virtio-iommu.c b/hw/virtio/virtio-iommu.c
new file mode 100644
index 0000000000..30579267d5
--- /dev/null
+++ b/hw/virtio/virtio-iommu.c
@@ -0,0 +1,265 @@
+/*
+ * virtio-iommu device
+ *
+ * Copyright (c) 2020 Red Hat, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2 or later, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+
+#include "qemu/osdep.h"
+#include "qemu/iov.h"
+#include "qemu-common.h"
+#include "hw/qdev-properties.h"
+#include "hw/virtio/virtio.h"
+#include "sysemu/kvm.h"
+#include "trace.h"
+
+#include "standard-headers/linux/virtio_ids.h"
+
+#include "hw/virtio/virtio-bus.h"
+#include "hw/virtio/virtio-access.h"
+#include "hw/virtio/virtio-iommu.h"
+
+/* Max size */
+#define VIOMMU_DEFAULT_QUEUE_SIZE 256
+
+static int virtio_iommu_handle_attach(VirtIOIOMMU *s,
+                                      struct iovec *iov,
+                                      unsigned int iov_cnt)
+{
+    return VIRTIO_IOMMU_S_UNSUPP;
+}
+static int virtio_iommu_handle_detach(VirtIOIOMMU *s,
+                                      struct iovec *iov,
+                                      unsigned int iov_cnt)
+{
+    return VIRTIO_IOMMU_S_UNSUPP;
+}
+static int virtio_iommu_handle_map(VirtIOIOMMU *s,
+                                   struct iovec *iov,
+                                   unsigned int iov_cnt)
+{
+    return VIRTIO_IOMMU_S_UNSUPP;
+}
+static int virtio_iommu_handle_unmap(VirtIOIOMMU *s,
+                                     struct iovec *iov,
+                                     unsigned int iov_cnt)
+{
+    return VIRTIO_IOMMU_S_UNSUPP;
+}
+
+static void virtio_iommu_handle_command(VirtIODevice *vdev, VirtQueue *vq)
+{
+    VirtIOIOMMU *s = VIRTIO_IOMMU(vdev);
+    struct virtio_iommu_req_head head;
+    struct virtio_iommu_req_tail tail = {};
+    VirtQueueElement *elem;
+    unsigned int iov_cnt;
+    struct iovec *iov;
+    size_t sz;
+
+    for (;;) {
+        elem = virtqueue_pop(vq, sizeof(VirtQueueElement));
+        if (!elem) {
+            return;
+        }
+
+        if (iov_size(elem->in_sg, elem->in_num) < sizeof(tail) ||
+            iov_size(elem->out_sg, elem->out_num) < sizeof(head)) {
+            virtio_error(vdev, "virtio-iommu bad head/tail size");
+            virtqueue_detach_element(vq, elem, 0);
+            g_free(elem);
+            break;
+        }
+
+        iov_cnt = elem->out_num;
+        iov = elem->out_sg;
+        sz = iov_to_buf(iov, iov_cnt, 0, &head, sizeof(head));
+        if (unlikely(sz != sizeof(head))) {
+            tail.status = VIRTIO_IOMMU_S_DEVERR;
+            goto out;
+        }
+        qemu_mutex_lock(&s->mutex);
+        switch (head.type) {
+        case VIRTIO_IOMMU_T_ATTACH:
+            tail.status = virtio_iommu_handle_attach(s, iov, iov_cnt);
+            break;
+        case VIRTIO_IOMMU_T_DETACH:
+            tail.status = virtio_iommu_handle_detach(s, iov, iov_cnt);
+            break;
+        case VIRTIO_IOMMU_T_MAP:
+            tail.status = virtio_iommu_handle_map(s, iov, iov_cnt);
+            break;
+        case VIRTIO_IOMMU_T_UNMAP:
+            tail.status = virtio_iommu_handle_unmap(s, iov, iov_cnt);
+            break;
+        default:
+            tail.status = VIRTIO_IOMMU_S_UNSUPP;
+        }
+        qemu_mutex_unlock(&s->mutex);
+
+out:
+        sz = iov_from_buf(elem->in_sg, elem->in_num, 0,
+                          &tail, sizeof(tail));
+        assert(sz == sizeof(tail));
+
+        virtqueue_push(vq, elem, sizeof(tail));
+        virtio_notify(vdev, vq);
+        g_free(elem);
+    }
+}
+
+static void virtio_iommu_get_config(VirtIODevice *vdev, uint8_t *config_data)
+{
+    VirtIOIOMMU *dev = VIRTIO_IOMMU(vdev);
+    struct virtio_iommu_config *config = &dev->config;
+
+    trace_virtio_iommu_get_config(config->page_size_mask,
+                                  config->input_range.start,
+                                  config->input_range.end,
+                                  config->domain_range.end,
+                                  config->probe_size);
+    memcpy(config_data, &dev->config, sizeof(struct virtio_iommu_config));
+}
+
+static void virtio_iommu_set_config(VirtIODevice *vdev,
+                                      const uint8_t *config_data)
+{
+    struct virtio_iommu_config config;
+
+    memcpy(&config, config_data, sizeof(struct virtio_iommu_config));
+    trace_virtio_iommu_set_config(config.page_size_mask,
+                                  config.input_range.start,
+                                  config.input_range.end,
+                                  config.domain_range.end,
+                                  config.probe_size);
+}
+
+static uint64_t virtio_iommu_get_features(VirtIODevice *vdev, uint64_t f,
+                                          Error **errp)
+{
+    VirtIOIOMMU *dev = VIRTIO_IOMMU(vdev);
+
+    f |= dev->features;
+    trace_virtio_iommu_get_features(f);
+    return f;
+}
+
+/*
+ * Migration is not yet supported: most of the state consists
+ * of balanced binary trees which are not yet ready for getting
+ * migrated
+ */
+static const VMStateDescription vmstate_virtio_iommu_device = {
+    .name = "virtio-iommu-device",
+    .unmigratable = 1,
+};
+
+static void virtio_iommu_device_realize(DeviceState *dev, Error **errp)
+{
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtIOIOMMU *s = VIRTIO_IOMMU(dev);
+
+    virtio_init(vdev, "virtio-iommu", VIRTIO_ID_IOMMU,
+                sizeof(struct virtio_iommu_config));
+
+    s->req_vq = virtio_add_queue(vdev, VIOMMU_DEFAULT_QUEUE_SIZE,
+                             virtio_iommu_handle_command);
+    s->event_vq = virtio_add_queue(vdev, VIOMMU_DEFAULT_QUEUE_SIZE, NULL);
+
+    s->config.page_size_mask = TARGET_PAGE_MASK;
+    s->config.input_range.end = -1UL;
+    s->config.domain_range.end = 32;
+
+    virtio_add_feature(&s->features, VIRTIO_RING_F_EVENT_IDX);
+    virtio_add_feature(&s->features, VIRTIO_RING_F_INDIRECT_DESC);
+    virtio_add_feature(&s->features, VIRTIO_F_VERSION_1);
+    virtio_add_feature(&s->features, VIRTIO_IOMMU_F_INPUT_RANGE);
+    virtio_add_feature(&s->features, VIRTIO_IOMMU_F_DOMAIN_RANGE);
+    virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MAP_UNMAP);
+    virtio_add_feature(&s->features, VIRTIO_IOMMU_F_BYPASS);
+    virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MMIO);
+
+    qemu_mutex_init(&s->mutex);
+}
+
+static void virtio_iommu_device_unrealize(DeviceState *dev, Error **errp)
+{
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+
+    virtio_cleanup(vdev);
+}
+
+static void virtio_iommu_device_reset(VirtIODevice *vdev)
+{
+    trace_virtio_iommu_device_reset();
+}
+
+static void virtio_iommu_set_status(VirtIODevice *vdev, uint8_t status)
+{
+    trace_virtio_iommu_device_status(status);
+}
+
+static void virtio_iommu_instance_init(Object *obj)
+{
+}
+
+static const VMStateDescription vmstate_virtio_iommu = {
+    .name = "virtio-iommu",
+    .minimum_version_id = 1,
+    .version_id = 1,
+    .fields = (VMStateField[]) {
+        VMSTATE_VIRTIO_DEVICE,
+        VMSTATE_END_OF_LIST()
+    },
+};
+
+static Property virtio_iommu_properties[] = {
+    DEFINE_PROP_LINK("primary-bus", VirtIOIOMMU, primary_bus, "PCI", PCIBus *),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void virtio_iommu_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
+
+    device_class_set_props(dc, virtio_iommu_properties);
+    dc->vmsd = &vmstate_virtio_iommu;
+
+    set_bit(DEVICE_CATEGORY_MISC, dc->categories);
+    vdc->realize = virtio_iommu_device_realize;
+    vdc->unrealize = virtio_iommu_device_unrealize;
+    vdc->reset = virtio_iommu_device_reset;
+    vdc->get_config = virtio_iommu_get_config;
+    vdc->set_config = virtio_iommu_set_config;
+    vdc->get_features = virtio_iommu_get_features;
+    vdc->set_status = virtio_iommu_set_status;
+    vdc->vmsd = &vmstate_virtio_iommu_device;
+}
+
+static const TypeInfo virtio_iommu_info = {
+    .name = TYPE_VIRTIO_IOMMU,
+    .parent = TYPE_VIRTIO_DEVICE,
+    .instance_size = sizeof(VirtIOIOMMU),
+    .instance_init = virtio_iommu_instance_init,
+    .class_init = virtio_iommu_class_init,
+};
+
+static void virtio_register_types(void)
+{
+    type_register_static(&virtio_iommu_info);
+}
+
+type_init(virtio_register_types)
diff --git a/include/hw/virtio/virtio-iommu.h b/include/hw/virtio/virtio-iommu.h
new file mode 100644
index 0000000000..d24ba63305
--- /dev/null
+++ b/include/hw/virtio/virtio-iommu.h
@@ -0,0 +1,57 @@
+/*
+ * virtio-iommu device
+ *
+ * Copyright (c) 2020 Red Hat, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2 or later, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+
+#ifndef QEMU_VIRTIO_IOMMU_H
+#define QEMU_VIRTIO_IOMMU_H
+
+#include "standard-headers/linux/virtio_iommu.h"
+#include "hw/virtio/virtio.h"
+#include "hw/pci/pci.h"
+
+#define TYPE_VIRTIO_IOMMU "virtio-iommu-device"
+#define VIRTIO_IOMMU(obj) \
+        OBJECT_CHECK(VirtIOIOMMU, (obj), TYPE_VIRTIO_IOMMU)
+
+typedef struct IOMMUDevice {
+    void         *viommu;
+    PCIBus       *bus;
+    int           devfn;
+    IOMMUMemoryRegion  iommu_mr;
+    AddressSpace  as;
+} IOMMUDevice;
+
+typedef struct IOMMUPciBus {
+    PCIBus       *bus;
+    IOMMUDevice  *pbdev[0]; /* Parent array is sparse, so dynamically alloc */
+} IOMMUPciBus;
+
+typedef struct VirtIOIOMMU {
+    VirtIODevice parent_obj;
+    VirtQueue *req_vq;
+    VirtQueue *event_vq;
+    struct virtio_iommu_config config;
+    uint64_t features;
+    GHashTable *as_by_busptr;
+    PCIBus *primary_bus;
+    GTree *domains;
+    QemuMutex mutex;
+    GTree *endpoints;
+} VirtIOIOMMU;
+
+#endif
-- 
2.20.1