vhost-scsi: new device supporting the tcm_vhost Linux kernel module

The WWPN specified in configfs is passed to "-device vhost-scsi-pci".
The tgpt field of the SET_ENDPOINT ioctl is obsolete now, so it is not
available from the QEMU command-line.  Instead, I hardcode it to zero.

Changes in Patch-v2:
   - Add vhost_scsi_get_features() in order to determine feature bits
     supports by host kernel (mst + nab)
   - Re-enable usage of DEFINE_VIRTIO_COMMON_FEATURES, and allow
     EVENT_IDX to be disabled by host in vhost_scsi_get_features()
   - Drop unused hotplug bit in DEFINE_VHOST_SCSI_PROPERTIES

Changes in Patch-v1:
   - Set event_idx=off by default (nab, thanks asias)
   - Disable hotplug feature bit for v3.9 tcm_vhost kernel code, need to
     re-enable in v3.10 (nab)
   - Update to latest qemu.git/master HEAD

Changes in WIP-V3:
   - Drop ioeventfd vhost_scsi_properties (asias, thanks stefanha)
   - Add CONFIG_VHOST_SCSI (asias, thanks stefanha)
   - Add hotplug feature bit

Changes in WIP-V2:
   - Add backend guest masking support (nab)
   - Bump ABI_VERSION to 1 (nab)
   - Set up set_guest_notifiers (asias)
   - Set up vs->dev.vq_index (asias)
   - Drop vs->vs.vdev.{set,clear}_vhost_endpoint (asias)
   - Drop VIRTIO_CONFIG_S_DRIVER check in vhost_scsi_set_status (asias)

Howto:
   Use the latest seabios, at least commit b44a7be17b
   git clone git://git.seabios.org/seabios.git
   make
   cp out/bios.bin /usr/share/qemu/bios.bin
   qemu -device vhost-scsi-pci,wwpn=naa.6001405bd4e8476d,event_idx=off
...

Cc: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: Asias He <asias@redhat.com>
[ Rebase on top of VirtIOSCSICommon patch, fix bugs in feature
  negotiation and irqfd masking - Paolo ]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

5 files changed, 378 insertions, 1 deletions

diff --git a/configure b/configure
index ed49f919c8..51a6c5621c 100755
--- a/configure
+++ b/configure
@@ -179,6 +179,7 @@ libattr=""
 xfs=""
 
 vhost_net="no"
+vhost_scsi="no"
 kvm="no"
 gprof="no"
 debug_tcg="no"
@@ -543,6 +544,7 @@ Haiku)
   usb="linux"
   kvm="yes"
   vhost_net="yes"
+  vhost_scsi="yes"
   if [ "$cpu" = "i386" -o "$cpu" = "x86_64" ] ; then
     audio_possible_drivers="$audio_possible_drivers fmod"
   fi
@@ -870,6 +872,10 @@ for opt do
   ;;
   --enable-vhost-net) vhost_net="yes"
   ;;
+  --disable-vhost-scsi) vhost_scsi="no"
+  ;;
+  --enable-vhost-scsi) vhost_scsi="yes"
+  ;;
   --disable-glx) glx="no"
   ;;
   --enable-glx) glx="yes"
@@ -3553,6 +3559,7 @@ echo "sigev_thread_id   $sigev_thread_id"
 echo "uuid support      $uuid"
 echo "libcap-ng support $cap_ng"
 echo "vhost-net support $vhost_net"
+echo "vhost-scsi support $vhost_scsi"
 echo "Trace backend     $trace_backend"
 echo "Trace output file $trace_file-<pid>"
 echo "spice support     $spice ($spice_protocol_version/$spice_server_version)"
@@ -3836,6 +3843,9 @@ fi
 if test "$virtfs" = "yes" ; then
   echo "CONFIG_VIRTFS=y" >> $config_host_mak
 fi
+if test "$vhost_scsi" = "yes" ; then
+  echo "CONFIG_VHOST_SCSI=y" >> $config_host_mak
+fi
 if test "$blobs" = "yes" ; then
   echo "INSTALL_BLOBS=yes" >> $config_host_mak
 fi
diff --git a/hw/scsi/Makefile.objs b/hw/scsi/Makefile.objs
index eaec6c87ef..121ddc5cf4 100644
--- a/hw/scsi/Makefile.objs
+++ b/hw/scsi/Makefile.objs
@@ -6,4 +6,8 @@ common-obj-$(CONFIG_VMW_PVSCSI_SCSI_PCI) += vmw_pvscsi.o
 common-obj-$(CONFIG_ESP) += esp.o
 common-obj-$(CONFIG_ESP_PCI) += esp-pci.o
 obj-$(CONFIG_PSERIES) += spapr_vscsi.o
-obj-$(CONFIG_VIRTIO) += virtio-scsi.o
+
+ifeq ($(CONFIG_VIRTIO),y)
+obj-y += virtio-scsi.o
+obj-$(CONFIG_VHOST_SCSI) += vhost-scsi.o
+endif
diff --git a/hw/scsi/vhost-scsi.c b/hw/scsi/vhost-scsi.c
new file mode 100644
index 0000000000..3dd1a0f2ed
--- /dev/null
+++ b/hw/scsi/vhost-scsi.c
@@ -0,0 +1,288 @@
+/*
+ * vhost_scsi host device
+ *
+ * Copyright IBM, Corp. 2011
+ *
+ * Authors:
+ *  Stefan Hajnoczi   <stefanha@linux.vnet.ibm.com>
+ *
+ * Changes for QEMU mainline + tcm_vhost kernel upstream:
+ *  Nicholas Bellinger <nab@risingtidesystems.com>
+ *
+ * This work is licensed under the terms of the GNU LGPL, version 2 or later.
+ * See the COPYING.LIB file in the top-level directory.
+ *
+ */
+
+#include <sys/ioctl.h>
+#include "config.h"
+#include "qemu/queue.h"
+#include "monitor/monitor.h"
+#include "migration/migration.h"
+#include "hw/virtio/vhost-scsi.h"
+#include "hw/virtio/vhost.h"
+#include "hw/virtio/virtio-scsi.h"
+
+static int vhost_scsi_set_endpoint(VHostSCSI *s)
+{
+    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(s);
+    struct vhost_scsi_target backend;
+    int ret;
+
+    memset(&backend, 0, sizeof(backend));
+    pstrcpy(backend.vhost_wwpn, sizeof(backend.vhost_wwpn), vs->conf.wwpn);
+    ret = ioctl(s->dev.control, VHOST_SCSI_SET_ENDPOINT, &backend);
+    if (ret < 0) {
+        return -errno;
+    }
+    return 0;
+}
+
+static void vhost_scsi_clear_endpoint(VHostSCSI *s)
+{
+    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(s);
+    struct vhost_scsi_target backend;
+
+    memset(&backend, 0, sizeof(backend));
+    pstrcpy(backend.vhost_wwpn, sizeof(backend.vhost_wwpn), vs->conf.wwpn);
+    ioctl(s->dev.control, VHOST_SCSI_CLEAR_ENDPOINT, &backend);
+}
+
+static int vhost_scsi_start(VHostSCSI *s)
+{
+    int ret, abi_version, i;
+    VirtIODevice *vdev = VIRTIO_DEVICE(s);
+
+    if (!vdev->binding->set_guest_notifiers) {
+        error_report("binding does not support guest notifiers");
+        return -ENOSYS;
+    }
+
+    ret = ioctl(s->dev.control, VHOST_SCSI_GET_ABI_VERSION, &abi_version);
+    if (ret < 0) {
+        return -errno;
+    }
+    if (abi_version > VHOST_SCSI_ABI_VERSION) {
+        error_report("vhost-scsi: The running tcm_vhost kernel abi_version:"
+                     " %d is greater than vhost_scsi userspace supports: %d, please"
+                     " upgrade your version of QEMU\n", abi_version,
+                     VHOST_SCSI_ABI_VERSION);
+        return -ENOSYS;
+    }
+
+    ret = vhost_dev_enable_notifiers(&s->dev, vdev);
+    if (ret < 0) {
+        return ret;
+    }
+
+    s->dev.acked_features = vdev->guest_features;
+    ret = vhost_dev_start(&s->dev, vdev);
+    if (ret < 0) {
+        error_report("Error start vhost dev");
+        goto err_notifiers;
+    }
+
+    ret = vhost_scsi_set_endpoint(s);
+    if (ret < 0) {
+        error_report("Error set vhost-scsi endpoint");
+        goto err_vhost_stop;
+    }
+
+    ret = vdev->binding->set_guest_notifiers(vdev->binding_opaque, s->dev.nvqs, true);
+    if (ret < 0) {
+        error_report("Error binding guest notifier");
+        goto err_endpoint;
+    }
+
+    /* guest_notifier_mask/pending not used yet, so just unmask
+     * everything here.  virtio-pci will do the right thing by
+     * enabling/disabling irqfd.
+     */
+    for (i = 0; i < s->dev.nvqs; i++) {
+        vhost_virtqueue_mask(&s->dev, vdev, i, false);
+    }
+
+    return ret;
+
+err_endpoint:
+    vhost_scsi_clear_endpoint(s);
+err_vhost_stop:
+    vhost_dev_stop(&s->dev, vdev);
+err_notifiers:
+    vhost_dev_disable_notifiers(&s->dev, vdev);
+    return ret;
+}
+
+static void vhost_scsi_stop(VHostSCSI *s)
+{
+    VirtIODevice *vdev = VIRTIO_DEVICE(s);
+    int ret = 0;
+
+    if (!vdev->binding->set_guest_notifiers) {
+        ret = vdev->binding->set_guest_notifiers(vdev->binding_opaque,
+                                                 s->dev.nvqs, false);
+        if (ret < 0) {
+                error_report("vhost guest notifier cleanup failed: %d\n", ret);
+        }
+    }
+    assert(ret >= 0);
+
+    vhost_scsi_clear_endpoint(s);
+    vhost_dev_stop(&s->dev, vdev);
+    vhost_dev_disable_notifiers(&s->dev, vdev);
+}
+
+static uint32_t vhost_scsi_get_features(VirtIODevice *vdev,
+                                        uint32_t features)
+{
+    VHostSCSI *s = VHOST_SCSI(vdev);
+
+    /* Clear features not supported by host kernel. */
+    if (!(s->dev.features & (1 << VIRTIO_F_NOTIFY_ON_EMPTY))) {
+        features &= ~(1 << VIRTIO_F_NOTIFY_ON_EMPTY);
+    }
+    if (!(s->dev.features & (1 << VIRTIO_RING_F_INDIRECT_DESC))) {
+        features &= ~(1 << VIRTIO_RING_F_INDIRECT_DESC);
+    }
+    if (!(s->dev.features & (1 << VIRTIO_RING_F_EVENT_IDX))) {
+        features &= ~(1 << VIRTIO_RING_F_EVENT_IDX);
+    }
+    if (!(s->dev.features & (1 << VIRTIO_SCSI_F_HOTPLUG))) {
+        features &= ~(1 << VIRTIO_SCSI_F_HOTPLUG);
+    }
+
+    return features;
+}
+
+static void vhost_scsi_set_config(VirtIODevice *vdev,
+                                  const uint8_t *config)
+{
+    VirtIOSCSIConfig *scsiconf = (VirtIOSCSIConfig *)config;
+    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(vdev);
+
+    if ((uint32_t) ldl_raw(&scsiconf->sense_size) != vs->sense_size ||
+        (uint32_t) ldl_raw(&scsiconf->cdb_size) != vs->cdb_size) {
+        error_report("vhost-scsi does not support changing the sense data and CDB sizes");
+        exit(1);
+    }
+}
+
+static void vhost_scsi_set_status(VirtIODevice *vdev, uint8_t val)
+{
+    VHostSCSI *s = (VHostSCSI *)vdev;
+    bool start = (val & VIRTIO_CONFIG_S_DRIVER_OK);
+
+    if (s->dev.started == start) {
+        return;
+    }
+
+    if (start) {
+        int ret;
+
+        ret = vhost_scsi_start(s);
+        if (ret < 0) {
+            error_report("virtio-scsi: unable to start vhost: %s\n",
+                         strerror(-ret));
+
+            /* There is no userspace virtio-scsi fallback so exit */
+            exit(1);
+        }
+    } else {
+        vhost_scsi_stop(s);
+    }
+}
+
+static int vhost_scsi_init(VirtIODevice *vdev)
+{
+    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(vdev);
+    VHostSCSI *s = VHOST_SCSI(vdev);
+    int vhostfd = -1;
+    int ret;
+
+    if (!vs->conf.wwpn) {
+        error_report("vhost-scsi: missing wwpn\n");
+        return -EINVAL;
+    }
+
+    if (vs->conf.vhostfd) {
+        vhostfd = monitor_handle_fd_param(cur_mon, vs->conf.vhostfd);
+        if (vhostfd == -1) {
+            error_report("vhost-scsi: unable to parse vhostfd\n");
+            return -EINVAL;
+        }
+    }
+
+    ret = virtio_scsi_common_init(vs);
+    if (ret < 0) {
+        return ret;
+    }
+
+    vdev->get_features = vhost_scsi_get_features;
+    vdev->set_config = vhost_scsi_set_config;
+    vdev->set_status = vhost_scsi_set_status;
+
+    s->dev.nvqs = VHOST_SCSI_VQ_NUM_FIXED + vs->conf.num_queues;
+    s->dev.vqs = g_new(struct vhost_virtqueue, s->dev.nvqs);
+    s->dev.vq_index = 0;
+
+    ret = vhost_dev_init(&s->dev, vhostfd, "/dev/vhost-scsi", true);
+    if (ret < 0) {
+        error_report("vhost-scsi: vhost initialization failed: %s\n",
+                strerror(-ret));
+        return ret;
+    }
+    s->dev.backend_features = 0;
+
+    error_setg(&s->migration_blocker,
+            "vhost-scsi does not support migration");
+    migrate_add_blocker(s->migration_blocker);
+
+    return 0;
+}
+
+static int vhost_scsi_exit(DeviceState *qdev)
+{
+    VirtIODevice *vdev = VIRTIO_DEVICE(qdev);
+    VHostSCSI *s = VHOST_SCSI(qdev);
+    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(qdev);
+
+    migrate_del_blocker(s->migration_blocker);
+    error_free(s->migration_blocker);
+
+    /* This will stop vhost backend. */
+    vhost_scsi_set_status(vdev, 0);
+
+    g_free(s->dev.vqs);
+    return virtio_scsi_common_exit(vs);
+}
+
+static Property vhost_scsi_properties[] = {
+    DEFINE_VHOST_SCSI_PROPERTIES(VHostSCSI, parent_obj.conf),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void vhost_scsi_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
+    dc->exit = vhost_scsi_exit;
+    dc->props = vhost_scsi_properties;
+    vdc->init = vhost_scsi_init;
+    vdc->get_features = vhost_scsi_get_features;
+    vdc->set_config = vhost_scsi_set_config;
+    vdc->set_status = vhost_scsi_set_status;
+}
+
+static const TypeInfo vhost_scsi_info = {
+    .name = TYPE_VHOST_SCSI,
+    .parent = TYPE_VIRTIO_SCSI_COMMON,
+    .instance_size = sizeof(VHostSCSI),
+    .class_init = vhost_scsi_class_init,
+};
+
+static void virtio_register_types(void)
+{
+    type_register_static(&vhost_scsi_info);
+}
+
+type_init(virtio_register_types)
diff --git a/include/hw/virtio/vhost-scsi.h b/include/hw/virtio/vhost-scsi.h
new file mode 100644
index 0000000000..85cc031281
--- /dev/null
+++ b/include/hw/virtio/vhost-scsi.h
@@ -0,0 +1,73 @@
+/*
+ * vhost_scsi host device
+ *
+ * Copyright IBM, Corp. 2011
+ *
+ * Authors:
+ *  Stefan Hajnoczi   <stefanha@linux.vnet.ibm.com>
+ *
+ * This work is licensed under the terms of the GNU LGPL, version 2 or later.
+ * See the COPYING.LIB file in the top-level directory.
+ *
+ */
+
+#ifndef VHOST_SCSI_H
+#define VHOST_SCSI_H
+
+#include "qemu-common.h"
+#include "hw/qdev.h"
+#include "hw/virtio/virtio-scsi.h"
+#include "hw/virtio/vhost.h"
+
+/*
+ * Used by QEMU userspace to ensure a consistent vhost-scsi ABI.
+ *
+ * ABI Rev 0: July 2012 version starting point for v3.6-rc merge candidate +
+ *            RFC-v2 vhost-scsi userspace.  Add GET_ABI_VERSION ioctl usage
+ * ABI Rev 1: January 2013. Ignore vhost_tpgt filed in struct vhost_scsi_target.
+ * 	      All the targets under vhost_wwpn can be seen and used by guest.
+ */
+
+#define VHOST_SCSI_ABI_VERSION 1
+
+/* TODO #include <linux/vhost.h> properly */
+/* For VHOST_SCSI_SET_ENDPOINT/VHOST_SCSI_CLEAR_ENDPOINT ioctl */
+struct vhost_scsi_target {
+    int abi_version;
+    char vhost_wwpn[224];
+    unsigned short vhost_tpgt;
+    unsigned short reserved;
+};
+
+enum vhost_scsi_vq_list {
+    VHOST_SCSI_VQ_CONTROL = 0,
+    VHOST_SCSI_VQ_EVENT = 1,
+    VHOST_SCSI_VQ_NUM_FIXED = 2,
+};
+
+#define VHOST_VIRTIO 0xAF
+#define VHOST_SCSI_SET_ENDPOINT _IOW(VHOST_VIRTIO, 0x40, struct vhost_scsi_target)
+#define VHOST_SCSI_CLEAR_ENDPOINT _IOW(VHOST_VIRTIO, 0x41, struct vhost_scsi_target)
+#define VHOST_SCSI_GET_ABI_VERSION _IOW(VHOST_VIRTIO, 0x42, int)
+
+#define TYPE_VHOST_SCSI "vhost-scsi"
+#define VHOST_SCSI(obj) \
+        OBJECT_CHECK(VHostSCSI, (obj), TYPE_VHOST_SCSI)
+
+typedef struct VHostSCSI {
+    VirtIOSCSICommon parent_obj;
+
+    Error *migration_blocker;
+
+    struct vhost_dev dev;
+} VHostSCSI;
+
+#define DEFINE_VHOST_SCSI_PROPERTIES(_state, _conf_field) \
+    DEFINE_PROP_STRING("vhostfd", _state, _conf_field.vhostfd), \
+    DEFINE_PROP_STRING("wwpn", _state, _conf_field.wwpn), \
+    DEFINE_PROP_UINT32("num_queues", _state, _conf_field.num_queues, 1), \
+    DEFINE_PROP_UINT32("max_sectors", _state, _conf_field.max_sectors, 0xFFFF), \
+    DEFINE_PROP_UINT32("cmd_per_lun", _state, _conf_field.cmd_per_lun, 128)
+
+
+#endif
diff --git a/include/hw/virtio/virtio-scsi.h b/include/hw/virtio/virtio-scsi.h
index 4a961b9c5a..4db346b546 100644
--- a/include/hw/virtio/virtio-scsi.h
+++ b/include/hw/virtio/virtio-scsi.h
@@ -151,6 +151,8 @@ struct VirtIOSCSIConf {
     uint32_t num_queues;
     uint32_t max_sectors;
     uint32_t cmd_per_lun;
+    char *vhostfd;
+    char *wwpn;
 };
 
 typedef struct VirtIOSCSICommon {