[Qemu-devel] [PULL v2 00/14] virtio,vhost,pc: fixes and updates

[Qemu-devel] [PULL v2 00/14] virtio,vhost,pc: fixes and updates

[Michael S. Tsirkin]

The following changes since commit c2a57aae9a1c3dd7de77daf5478df10379aeeebf:

  Merge remote-tracking branch 'remotes/famz/tags/docker-pull-request' into staging (2016-09-09 12:49:41 +0100)

are available in the git repository at:

  git://git.kernel.org/pub/scm/virt/kvm/mst/qemu.git tags/for_upstream

for you to fetch changes up to fc0b9b0e1cbb49017ea882758634cf876be17bc3:

  vhost-vsock: add virtio sockets device (2016-09-10 00:28:08 +0300)

----------------------------------------------------------------
virtio,vhost,pc: fixes and updates

balloon fixes wrt migration
virtio-vsock device support

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

----------------------------------------------------------------
Greg Kurz (1):
      virtio-pci: error out when both legacy and modern modes are disabled

Jason Wang (1):
      vhost: don't set vring call if no vector

Ladi Prosek (2):
      virtio-balloon: discard virtqueue element on reset
      virtio-balloon: fix stats vq migration

Longpeng(Mike) (2):
      pc: Add 2.8 machine
      target-i386: present virtual L3 cache info for vcpus

Marcel Apfelbaum (2):
      virtio-pci: reduce modern_mem_bar size
      tests/acpi: speedup acpi tests

Michael S. Tsirkin (2):
      virtio-pci: use size from correct structure
      virtio-pci: minor refactoring

Stefan Hajnoczi (3):
      virtio: zero vq->inuse in virtio_reset()
      virtio: add virtqueue_rewind()
      vhost-vsock: add virtio sockets device

Thomas Huth (1):
      virtio: Tell the user what went wrong when event_notifier_init failed

 configure                         |  10 +
 hw/s390x/virtio-ccw.h             |  15 ++
 hw/virtio/virtio-pci.h            |  23 +++
 include/hw/compat.h               |   6 +-
 include/hw/i386/pc.h              |  12 ++
 include/hw/pci/pci.h              |   1 +
 include/hw/virtio/vhost-backend.h |   5 +
 include/hw/virtio/vhost-vsock.h   |  41 ++++
 include/hw/virtio/virtio.h        |   1 +
 target-i386/cpu.h                 |   6 +
 hw/i386/pc_piix.c                 |  16 +-
 hw/i386/pc_q35.c                  |  13 +-
 hw/s390x/virtio-ccw.c             |  54 +++++
 hw/virtio/vhost-backend.c         |  17 ++
 hw/virtio/vhost-vsock.c           | 417 ++++++++++++++++++++++++++++++++++++++
 hw/virtio/vhost.c                 |  14 ++
 hw/virtio/virtio-balloon.c        |  14 ++
 hw/virtio/virtio-bus.c            |   3 +-
 hw/virtio/virtio-pci.c            |  83 +++++++-
 hw/virtio/virtio.c                |  23 +++
 target-i386/cpu.c                 |  49 ++++-
 tests/bios-tables-test.c          |  39 ++--
 hw/virtio/Makefile.objs           |   2 +
 23 files changed, 825 insertions(+), 39 deletions(-)
 create mode 100644 include/hw/virtio/vhost-vsock.h
 create mode 100644 hw/virtio/vhost-vsock.c

[Qemu-devel] [PULL v2 01/14] virtio: Tell the user what went wrong when event_notifier_init failed

[Michael S. Tsirkin]

From: Thomas Huth <thuth@redhat.com>

event_notifier_init() can fail in real life, for example when there
are not enough open file handles available (EMFILE) when using a lot
of devices. So instead of leaving the average user with a cryptic
error number only, print out a proper error message with strerror()
instead, so that the user has a better way to figure out what is
going on and that using "ulimit -n" might help here for example.

Signed-off-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 hw/virtio/virtio-bus.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/hw/virtio/virtio-bus.c b/hw/virtio/virtio-bus.c
index a85b7c8..1492793 100644
--- a/hw/virtio/virtio-bus.c
+++ b/hw/virtio/virtio-bus.c
@@ -164,7 +164,8 @@ static int set_host_notifier_internal(DeviceState *proxy, VirtioBusState *bus,
     if (assign) {
         r = event_notifier_init(notifier, 1);
         if (r < 0) {
-            error_report("%s: unable to init event notifier: %d", __func__, r);
+            error_report("%s: unable to init event notifier: %s (%d)",
+                         __func__, strerror(-r), r);
             return r;
         }
         virtio_queue_set_host_notifier_fd_handler(vq, true, set_handler);
-- 
MST

[Qemu-devel] [PULL v2 02/14] virtio-pci: use size from correct structure

[Michael S. Tsirkin]

PIO MR registration should use size from the correct notify struct.
Doesn't affect any visible behaviour because the field values are the
same (both are 4).

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 hw/virtio/virtio-pci.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/hw/virtio/virtio-pci.c b/hw/virtio/virtio-pci.c
index 755f921..268fd8e 100644
--- a/hw/virtio/virtio-pci.c
+++ b/hw/virtio/virtio-pci.c
@@ -1520,7 +1520,7 @@ static void virtio_pci_modern_regions_init(VirtIOPCIProxy *proxy)
                           &notify_pio_ops,
                           virtio_bus_get_device(&proxy->bus),
                           "virtio-pci-notify-pio",
-                          proxy->notify.size);
+                          proxy->notify_pio.size);
 }
 
 static void virtio_pci_modern_region_map(VirtIOPCIProxy *proxy,
-- 
MST

[Qemu-devel] [PULL v2 03/14] pc: Add 2.8 machine

[Michael S. Tsirkin]

From: "Longpeng(Mike)" <longpeng2@huawei.com>

This will used by the next patch.

Signed-off-by: Longpeng(Mike) <longpeng2@huawei.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 include/hw/i386/pc.h |  3 +++
 hw/i386/pc_piix.c    | 16 +++++++++++++---
 hw/i386/pc_q35.c     | 13 +++++++++++--
 3 files changed, 27 insertions(+), 5 deletions(-)

diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index c4316c3..8ad6f15 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -368,6 +368,9 @@ int e820_add_entry(uint64_t, uint64_t, uint32_t);
 int e820_get_num_entries(void);
 bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
 
+#define PC_COMPAT_2_7 \
+    HW_COMPAT_2_7
+
 #define PC_COMPAT_2_6 \
     HW_COMPAT_2_6 \
     {\
diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
index 2af8888..a54a468 100644
--- a/hw/i386/pc_piix.c
+++ b/hw/i386/pc_piix.c
@@ -437,13 +437,25 @@ static void pc_i440fx_machine_options(MachineClass *m)
     m->default_display = "std";
 }
 
-static void pc_i440fx_2_7_machine_options(MachineClass *m)
+static void pc_i440fx_2_8_machine_options(MachineClass *m)
 {
     pc_i440fx_machine_options(m);
     m->alias = "pc";
     m->is_default = 1;
 }
 
+DEFINE_I440FX_MACHINE(v2_8, "pc-i440fx-2.8", NULL,
+                      pc_i440fx_2_8_machine_options);
+
+
+static void pc_i440fx_2_7_machine_options(MachineClass *m)
+{
+    pc_i440fx_2_8_machine_options(m);
+    m->is_default = 0;
+    m->alias = NULL;
+    SET_MACHINE_COMPAT(m, PC_COMPAT_2_7);
+}
+
 DEFINE_I440FX_MACHINE(v2_7, "pc-i440fx-2.7", NULL,
                       pc_i440fx_2_7_machine_options);
 
@@ -452,8 +464,6 @@ static void pc_i440fx_2_6_machine_options(MachineClass *m)
 {
     PCMachineClass *pcmc = PC_MACHINE_CLASS(m);
     pc_i440fx_2_7_machine_options(m);
-    m->is_default = 0;
-    m->alias = NULL;
     pcmc->legacy_cpu_hotplug = true;
     SET_MACHINE_COMPAT(m, PC_COMPAT_2_6);
 }
diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c
index 3cbcbb0..0b214f2 100644
--- a/hw/i386/pc_q35.c
+++ b/hw/i386/pc_q35.c
@@ -293,12 +293,22 @@ static void pc_q35_machine_options(MachineClass *m)
     m->has_dynamic_sysbus = true;
 }
 
-static void pc_q35_2_7_machine_options(MachineClass *m)
+static void pc_q35_2_8_machine_options(MachineClass *m)
 {
     pc_q35_machine_options(m);
     m->alias = "q35";
 }
 
+DEFINE_Q35_MACHINE(v2_8, "pc-q35-2.8", NULL,
+                   pc_q35_2_8_machine_options);
+
+static void pc_q35_2_7_machine_options(MachineClass *m)
+{
+    pc_q35_2_8_machine_options(m);
+    m->alias = NULL;
+    SET_MACHINE_COMPAT(m, PC_COMPAT_2_7);
+}
+
 DEFINE_Q35_MACHINE(v2_7, "pc-q35-2.7", NULL,
                    pc_q35_2_7_machine_options);
 
@@ -306,7 +316,6 @@ static void pc_q35_2_6_machine_options(MachineClass *m)
 {
     PCMachineClass *pcmc = PC_MACHINE_CLASS(m);
     pc_q35_2_7_machine_options(m);
-    m->alias = NULL;
     pcmc->legacy_cpu_hotplug = true;
     SET_MACHINE_COMPAT(m, PC_COMPAT_2_6);
 }
-- 
MST

[Qemu-devel] [PULL v2 04/14] target-i386: present virtual L3 cache info for vcpus

[Michael S. Tsirkin]

From: "Longpeng(Mike)" <longpeng2@huawei.com>

Some software algorithms are based on the hardware's cache info, for example,
for x86 linux kernel, when cpu1 want to wakeup a task on cpu2, cpu1 will trigger
a resched IPI and told cpu2 to do the wakeup if they don't share low level
cache. Oppositely, cpu1 will access cpu2's runqueue directly if they share llc.
The relevant linux-kernel code as bellow:

	static void ttwu_queue(struct task_struct *p, int cpu)
	{
		struct rq *rq = cpu_rq(cpu);
		......
		if (... && !cpus_share_cache(smp_processor_id(), cpu)) {
			......
			ttwu_queue_remote(p, cpu); /* will trigger RES IPI */
			return;
		}
		......
		ttwu_do_activate(rq, p, 0); /* access target's rq directly */
		......
	}

In real hardware, the cpus on the same socket share L3 cache, so one won't
trigger a resched IPIs when wakeup a task on others. But QEMU doesn't present a
virtual L3 cache info for VM, then the linux guest will trigger lots of RES IPIs
under some workloads even if the virtual cpus belongs to the same virtual socket.

For KVM, there will be lots of vmexit due to guest send IPIs.
The workload is a SAP HANA's testsuite, we run it one round(about 40 minuates)
and observe the (Suse11sp3)Guest's amounts of RES IPIs which triggering during
the period:
        No-L3           With-L3(applied this patch)
cpu0:	363890		44582
cpu1:	373405		43109
cpu2:	340783		43797
cpu3:	333854		43409
cpu4:	327170		40038
cpu5:	325491		39922
cpu6:	319129		42391
cpu7:	306480		41035
cpu8:	161139		32188
cpu9:	164649		31024
cpu10:	149823		30398
cpu11:	149823		32455
cpu12:	164830		35143
cpu13:	172269		35805
cpu14:	179979		33898
cpu15:	194505		32754
avg:	268963.6	40129.8

The VM's topology is "1*socket 8*cores 2*threads".
After present virtual L3 cache info for VM, the amounts of RES IPIs in guest
reduce 85%.

For KVM, vcpus send IPIs will cause vmexit which is expensive, so it can cause
severe performance degradation. We had tested the overall system performance if
vcpus actually run on sparate physical socket. With L3 cache, the performance
improves 7.2%~33.1%(avg:15.7%).

Signed-off-by: Longpeng(Mike) <longpeng2@huawei.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 include/hw/i386/pc.h |  9 +++++++++
 target-i386/cpu.h    |  6 ++++++
 target-i386/cpu.c    | 49 ++++++++++++++++++++++++++++++++++++++++++++-----
 3 files changed, 59 insertions(+), 5 deletions(-)

diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index 8ad6f15..ebba151 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -368,7 +368,16 @@ int e820_add_entry(uint64_t, uint64_t, uint32_t);
 int e820_get_num_entries(void);
 bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
 
+#define PC_COMPAT_2_8 \
+    {\
+        .driver   = TYPE_X86_CPU,\
+        .property = "l3-cache",\
+        .value    = "off",\
+    },
+
+
 #define PC_COMPAT_2_7 \
+    PC_COMPAT_2_8 \
     HW_COMPAT_2_7
 
 #define PC_COMPAT_2_6 \
diff --git a/target-i386/cpu.h b/target-i386/cpu.h
index cf14bcb..bb3ffda 100644
--- a/target-i386/cpu.h
+++ b/target-i386/cpu.h
@@ -1207,6 +1207,12 @@ struct X86CPU {
      */
     bool enable_lmce;
 
+    /* Compatibility bits for old machine types.
+     * If true present virtual l3 cache for VM, the vcpus in the same virtual
+     * socket share an virtual l3 cache.
+     */
+    bool enable_l3_cache;
+
     /* Compatibility bits for old machine types: */
     bool enable_cpuid_0xb;
 
diff --git a/target-i386/cpu.c b/target-i386/cpu.c
index ec674dc..5a5299a 100644
--- a/target-i386/cpu.c
+++ b/target-i386/cpu.c
@@ -57,6 +57,7 @@
 #define CPUID_2_L1D_32KB_8WAY_64B 0x2c
 #define CPUID_2_L1I_32KB_8WAY_64B 0x30
 #define CPUID_2_L2_2MB_8WAY_64B   0x7d
+#define CPUID_2_L3_16MB_16WAY_64B 0x4d
 
 
 /* CPUID Leaf 4 constants: */
@@ -131,11 +132,18 @@
 #define L2_LINES_PER_TAG       1
 #define L2_SIZE_KB_AMD       512
 
-/* No L3 cache: */
+/* Level 3 unified cache: */
 #define L3_SIZE_KB             0 /* disabled */
 #define L3_ASSOCIATIVITY       0 /* disabled */
 #define L3_LINES_PER_TAG       0 /* disabled */
 #define L3_LINE_SIZE           0 /* disabled */
+#define L3_N_LINE_SIZE         64
+#define L3_N_ASSOCIATIVITY     16
+#define L3_N_SETS           16384
+#define L3_N_PARTITIONS         1
+#define L3_N_DESCRIPTOR CPUID_2_L3_16MB_16WAY_64B
+#define L3_N_LINES_PER_TAG      1
+#define L3_N_SIZE_KB_AMD    16384
 
 /* TLB definitions: */
 
@@ -2279,6 +2287,7 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
 {
     X86CPU *cpu = x86_env_get_cpu(env);
     CPUState *cs = CPU(cpu);
+    uint32_t pkg_offset;
 
     /* test if maximum index reached */
     if (index & 0x80000000) {
@@ -2332,7 +2341,11 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
         }
         *eax = 1; /* Number of CPUID[EAX=2] calls required */
         *ebx = 0;
-        *ecx = 0;
+        if (!cpu->enable_l3_cache) {
+            *ecx = 0;
+        } else {
+            *ecx = L3_N_DESCRIPTOR;
+        }
         *edx = (L1D_DESCRIPTOR << 16) | \
                (L1I_DESCRIPTOR <<  8) | \
                (L2_DESCRIPTOR);
@@ -2378,6 +2391,25 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
                 *ecx = L2_SETS - 1;
                 *edx = CPUID_4_NO_INVD_SHARING;
                 break;
+            case 3: /* L3 cache info */
+                if (!cpu->enable_l3_cache) {
+                    *eax = 0;
+                    *ebx = 0;
+                    *ecx = 0;
+                    *edx = 0;
+                    break;
+                }
+                *eax |= CPUID_4_TYPE_UNIFIED | \
+                        CPUID_4_LEVEL(3) | \
+                        CPUID_4_SELF_INIT_LEVEL;
+                pkg_offset = apicid_pkg_offset(cs->nr_cores, cs->nr_threads);
+                *eax |= ((1 << pkg_offset) - 1) << 14;
+                *ebx = (L3_N_LINE_SIZE - 1) | \
+                       ((L3_N_PARTITIONS - 1) << 12) | \
+                       ((L3_N_ASSOCIATIVITY - 1) << 22);
+                *ecx = L3_N_SETS - 1;
+                *edx = CPUID_4_INCLUSIVE | CPUID_4_COMPLEX_IDX;
+                break;
             default: /* end of info */
                 *eax = 0;
                 *ebx = 0;
@@ -2589,9 +2621,15 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
         *ecx = (L2_SIZE_KB_AMD << 16) | \
                (AMD_ENC_ASSOC(L2_ASSOCIATIVITY) << 12) | \
                (L2_LINES_PER_TAG << 8) | (L2_LINE_SIZE);
-        *edx = ((L3_SIZE_KB/512) << 18) | \
-               (AMD_ENC_ASSOC(L3_ASSOCIATIVITY) << 12) | \
-               (L3_LINES_PER_TAG << 8) | (L3_LINE_SIZE);
+        if (!cpu->enable_l3_cache) {
+            *edx = ((L3_SIZE_KB / 512) << 18) | \
+                   (AMD_ENC_ASSOC(L3_ASSOCIATIVITY) << 12) | \
+                   (L3_LINES_PER_TAG << 8) | (L3_LINE_SIZE);
+        } else {
+            *edx = ((L3_N_SIZE_KB_AMD / 512) << 18) | \
+                   (AMD_ENC_ASSOC(L3_N_ASSOCIATIVITY) << 12) | \
+                   (L3_N_LINES_PER_TAG << 8) | (L3_N_LINE_SIZE);
+        }
         break;
     case 0x80000007:
         *eax = 0;
@@ -3368,6 +3406,7 @@ static Property x86_cpu_properties[] = {
     DEFINE_PROP_STRING("hv-vendor-id", X86CPU, hyperv_vendor_id),
     DEFINE_PROP_BOOL("cpuid-0xb", X86CPU, enable_cpuid_0xb, true),
     DEFINE_PROP_BOOL("lmce", X86CPU, enable_lmce, false),
+    DEFINE_PROP_BOOL("l3-cache", X86CPU, enable_l3_cache, true),
     DEFINE_PROP_END_OF_LIST()
 };
 
-- 
MST

[Qemu-devel] [PULL v2 05/14] virtio-pci: reduce modern_mem_bar size

[Michael S. Tsirkin]

From: Marcel Apfelbaum <marcel@redhat.com>

Currently each VQ Notification Virtio Capability is allocated
on a different page. The idea is to enable split drivers within
guests, however there are no known plans to do that.
The allocation will result in a 8MB BAR, more than various
guest firmwares pre-allocates for PCI Bridges hotplug process.

Reserve 4 bytes per VQ by default and add a new parameter
"page-per-vq" to be used with split drivers.

Signed-off-by: Marcel Apfelbaum <marcel@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 hw/virtio/virtio-pci.h |  5 +++++
 include/hw/compat.h    |  6 +++++-
 hw/virtio/virtio-pci.c | 22 +++++++++++++++-------
 3 files changed, 25 insertions(+), 8 deletions(-)

diff --git a/hw/virtio/virtio-pci.h b/hw/virtio/virtio-pci.h
index 25fbf8a..a745512 100644
--- a/hw/virtio/virtio-pci.h
+++ b/hw/virtio/virtio-pci.h
@@ -64,6 +64,7 @@ enum {
     VIRTIO_PCI_FLAG_MIGRATE_EXTRA_BIT,
     VIRTIO_PCI_FLAG_MODERN_PIO_NOTIFY_BIT,
     VIRTIO_PCI_FLAG_DISABLE_PCIE_BIT,
+    VIRTIO_PCI_FLAG_PAGE_PER_VQ_BIT,
 };
 
 /* Need to activate work-arounds for buggy guests at vmstate load. */
@@ -84,6 +85,10 @@ enum {
 #define VIRTIO_PCI_FLAG_MODERN_PIO_NOTIFY \
     (1 << VIRTIO_PCI_FLAG_MODERN_PIO_NOTIFY_BIT)
 
+/* page per vq flag to be used by split drivers within guests */
+#define VIRTIO_PCI_FLAG_PAGE_PER_VQ \
+    (1 << VIRTIO_PCI_FLAG_PAGE_PER_VQ_BIT)
+
 typedef struct {
     MSIMessage msg;
     int virq;
diff --git a/include/hw/compat.h b/include/hw/compat.h
index 08dd4fb..a1d6694 100644
--- a/include/hw/compat.h
+++ b/include/hw/compat.h
@@ -2,7 +2,11 @@
 #define HW_COMPAT_H
 
 #define HW_COMPAT_2_7 \
-        /* empty */
+    {\
+        .driver   = "virtio-pci",\
+        .property = "page-per-vq",\
+        .value    = "on",\
+    },
 
 #define HW_COMPAT_2_6 \
     {\
diff --git a/hw/virtio/virtio-pci.c b/hw/virtio/virtio-pci.c
index 268fd8e..89bc196 100644
--- a/hw/virtio/virtio-pci.c
+++ b/hw/virtio/virtio-pci.c
@@ -294,6 +294,12 @@ static void virtio_pci_ioeventfd_set_disabled(DeviceState *d, bool disabled)
 
 #define QEMU_VIRTIO_PCI_QUEUE_MEM_MULT 0x1000
 
+static inline int virtio_pci_queue_mem_mult(struct VirtIOPCIProxy *proxy)
+{
+    return (proxy->flags & VIRTIO_PCI_FLAG_PAGE_PER_VQ) ?
+        QEMU_VIRTIO_PCI_QUEUE_MEM_MULT : 4;
+}
+
 static int virtio_pci_ioeventfd_assign(DeviceState *d, EventNotifier *notifier,
                                        int n, bool assign)
 {
@@ -307,7 +313,7 @@ static int virtio_pci_ioeventfd_assign(DeviceState *d, EventNotifier *notifier,
     MemoryRegion *modern_mr = &proxy->notify.mr;
     MemoryRegion *modern_notify_mr = &proxy->notify_pio.mr;
     MemoryRegion *legacy_mr = &proxy->bar;
-    hwaddr modern_addr = QEMU_VIRTIO_PCI_QUEUE_MEM_MULT *
+    hwaddr modern_addr = virtio_pci_queue_mem_mult(proxy) *
                          virtio_get_queue_index(vq);
     hwaddr legacy_addr = VIRTIO_PCI_QUEUE_NOTIFY;
 
@@ -1370,7 +1376,8 @@ static void virtio_pci_notify_write(void *opaque, hwaddr addr,
                                     uint64_t val, unsigned size)
 {
     VirtIODevice *vdev = opaque;
-    unsigned queue = addr / QEMU_VIRTIO_PCI_QUEUE_MEM_MULT;
+    VirtIOPCIProxy *proxy = VIRTIO_PCI(DEVICE(vdev)->parent_bus->parent);
+    unsigned queue = addr / virtio_pci_queue_mem_mult(proxy);
 
     if (queue < VIRTIO_QUEUE_MAX) {
         virtio_queue_notify(vdev, queue);
@@ -1609,7 +1616,7 @@ static void virtio_pci_device_plugged(DeviceState *d, Error **errp)
         struct virtio_pci_notify_cap notify = {
             .cap.cap_len = sizeof notify,
             .notify_off_multiplier =
-                cpu_to_le32(QEMU_VIRTIO_PCI_QUEUE_MEM_MULT),
+                cpu_to_le32(virtio_pci_queue_mem_mult(proxy)),
         };
         struct virtio_pci_cfg_cap cfg = {
             .cap.cap_len = sizeof cfg,
@@ -1744,8 +1751,7 @@ static void virtio_pci_realize(PCIDevice *pci_dev, Error **errp)
     proxy->device.type = VIRTIO_PCI_CAP_DEVICE_CFG;
 
     proxy->notify.offset = 0x3000;
-    proxy->notify.size =
-        QEMU_VIRTIO_PCI_QUEUE_MEM_MULT * VIRTIO_QUEUE_MAX;
+    proxy->notify.size = virtio_pci_queue_mem_mult(proxy) * VIRTIO_QUEUE_MAX;
     proxy->notify.type = VIRTIO_PCI_CAP_NOTIFY_CFG;
 
     proxy->notify_pio.offset = 0x0;
@@ -1754,8 +1760,8 @@ static void virtio_pci_realize(PCIDevice *pci_dev, Error **errp)
 
     /* subclasses can enforce modern, so do this unconditionally */
     memory_region_init(&proxy->modern_bar, OBJECT(proxy), "virtio-pci",
-                       2 * QEMU_VIRTIO_PCI_QUEUE_MEM_MULT *
-                       VIRTIO_QUEUE_MAX);
+                       /* PCI BAR regions must be powers of 2 */
+                       pow2ceil(proxy->notify.offset + proxy->notify.size));
 
     memory_region_init_alias(&proxy->modern_cfg,
                              OBJECT(proxy),
@@ -1833,6 +1839,8 @@ static Property virtio_pci_properties[] = {
                     VIRTIO_PCI_FLAG_MODERN_PIO_NOTIFY_BIT, false),
     DEFINE_PROP_BIT("x-disable-pcie", VirtIOPCIProxy, flags,
                     VIRTIO_PCI_FLAG_DISABLE_PCIE_BIT, false),
+    DEFINE_PROP_BIT("page-per-vq", VirtIOPCIProxy, flags,
+                    VIRTIO_PCI_FLAG_PAGE_PER_VQ_BIT, false),
     DEFINE_PROP_END_OF_LIST(),
 };
 
-- 
MST

[Qemu-devel] [PULL v2 06/14] virtio: zero vq->inuse in virtio_reset()

[Michael S. Tsirkin]

From: Stefan Hajnoczi <stefanha@redhat.com>

vq->inuse must be zeroed upon device reset like most other virtqueue
fields.

In theory, virtio_reset() just needs assert(vq->inuse == 0) since
devices must clean up in-flight requests during reset (requests cannot
not be leaked!).

In practice, it is difficult to achieve vq->inuse == 0 across reset
because balloon, blk, 9p, etc implement various different strategies for
cleaning up requests.  Most devices call g_free(elem) directly without
telling virtio.c that the VirtQueueElement is cleaned up.  Therefore
vq->inuse is not decremented during reset.

This patch zeroes vq->inuse and trusts that devices are not leaking
VirtQueueElements across reset.

I will send a follow-up series that refactors request life-cycle across
all devices and converts vq->inuse = 0 into assert(vq->inuse == 0) but
this more invasive approach is not appropriate for stable trees.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Cc: qemu-stable <qemu-stable@nongnu.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Ladi Prosek <lprosek@redhat.com>
---
 hw/virtio/virtio.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
index 74c085c..e8a13a5 100644
--- a/hw/virtio/virtio.c
+++ b/hw/virtio/virtio.c
@@ -822,6 +822,7 @@ void virtio_reset(void *opaque)
         vdev->vq[i].signalled_used_valid = false;
         vdev->vq[i].notification = true;
         vdev->vq[i].vring.num = vdev->vq[i].vring.num_default;
+        vdev->vq[i].inuse = 0;
     }
 }
 
-- 
MST

[Qemu-devel] [PULL v2 07/14] virtio-balloon: discard virtqueue element on reset

[Michael S. Tsirkin]

From: Ladi Prosek <lprosek@redhat.com>

The one pending element is being freed but not discarded on device
reset, which causes svq->inuse to creep up, eventually hitting the
"Virtqueue size exceeded" error.

Properly discarding the element on device reset makes sure that its
buffers are unmapped and the inuse counter stays balanced.

Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Roman Kagan <rkagan@virtuozzo.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 hw/virtio/virtio-balloon.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/hw/virtio/virtio-balloon.c b/hw/virtio/virtio-balloon.c
index 5af429a..ad4189a 100644
--- a/hw/virtio/virtio-balloon.c
+++ b/hw/virtio/virtio-balloon.c
@@ -463,6 +463,7 @@ static void virtio_balloon_device_reset(VirtIODevice *vdev)
     VirtIOBalloon *s = VIRTIO_BALLOON(vdev);
 
     if (s->stats_vq_elem != NULL) {
+        virtqueue_discard(s->svq, s->stats_vq_elem, 0);
         g_free(s->stats_vq_elem);
         s->stats_vq_elem = NULL;
     }
-- 
MST

[Qemu-devel] [PULL v2 08/14] virtio: add virtqueue_rewind()

[Michael S. Tsirkin]

From: Stefan Hajnoczi <stefanha@redhat.com>

virtqueue_discard() requires a VirtQueueElement but virtio-balloon does
not migrate its in-use element.  Introduce a new function that is
similar to virtqueue_discard() but doesn't require a VirtQueueElement.

This will allow virtio-balloon to access element again after migration
with the usual proviso that the guest may have modified the vring since
last time.

Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Roman Kagan <rkagan@virtuozzo.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 include/hw/virtio/virtio.h |  1 +
 hw/virtio/virtio.c         | 22 ++++++++++++++++++++++
 2 files changed, 23 insertions(+)

diff --git a/include/hw/virtio/virtio.h b/include/hw/virtio/virtio.h
index d2490c1..f05559d 100644
--- a/include/hw/virtio/virtio.h
+++ b/include/hw/virtio/virtio.h
@@ -154,6 +154,7 @@ void virtqueue_push(VirtQueue *vq, const VirtQueueElement *elem,
 void virtqueue_flush(VirtQueue *vq, unsigned int count);
 void virtqueue_discard(VirtQueue *vq, const VirtQueueElement *elem,
                        unsigned int len);
+bool virtqueue_rewind(VirtQueue *vq, unsigned int num);
 void virtqueue_fill(VirtQueue *vq, const VirtQueueElement *elem,
                     unsigned int len, unsigned int idx);
 
diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
index e8a13a5..fcf3358 100644
--- a/hw/virtio/virtio.c
+++ b/hw/virtio/virtio.c
@@ -272,6 +272,28 @@ void virtqueue_discard(VirtQueue *vq, const VirtQueueElement *elem,
     virtqueue_unmap_sg(vq, elem, len);
 }
 
+/* virtqueue_rewind:
+ * @vq: The #VirtQueue
+ * @num: Number of elements to push back
+ *
+ * Pretend that elements weren't popped from the virtqueue.  The next
+ * virtqueue_pop() will refetch the oldest element.
+ *
+ * Use virtqueue_discard() instead if you have a VirtQueueElement.
+ *
+ * Returns: true on success, false if @num is greater than the number of in use
+ * elements.
+ */
+bool virtqueue_rewind(VirtQueue *vq, unsigned int num)
+{
+    if (num > vq->inuse) {
+        return false;
+    }
+    vq->last_avail_idx -= num;
+    vq->inuse -= num;
+    return true;
+}
+
 void virtqueue_fill(VirtQueue *vq, const VirtQueueElement *elem,
                     unsigned int len, unsigned int idx)
 {
-- 
MST

[Qemu-devel] [PULL v2 09/14] virtio-balloon: fix stats vq migration

[Michael S. Tsirkin]

From: Ladi Prosek <lprosek@redhat.com>

The statistics virtqueue is not migrated properly because virtio-balloon
does not include s->stats_vq_elem in the migration stream.

After migration the statistics virtqueue hangs because the host never
completes the last element (s->stats_vq_elem is NULL on the destination
QEMU).  Therefore the guest never submits new elements and the virtqueue
is hung.

Instead of changing the migration stream format in an incompatible way,
detect the migration case and rewind the virtqueue so the last element
can be completed.

Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Roman Kagan <rkagan@virtuozzo.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Suggested-by: Roman Kagan <rkagan@virtuozzo.com>
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 hw/virtio/virtio-balloon.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/hw/virtio/virtio-balloon.c b/hw/virtio/virtio-balloon.c
index ad4189a..49a2f4a 100644
--- a/hw/virtio/virtio-balloon.c
+++ b/hw/virtio/virtio-balloon.c
@@ -469,6 +469,18 @@ static void virtio_balloon_device_reset(VirtIODevice *vdev)
     }
 }
 
+static void virtio_balloon_set_status(VirtIODevice *vdev, uint8_t status)
+{
+    VirtIOBalloon *s = VIRTIO_BALLOON(vdev);
+
+    if (!s->stats_vq_elem && vdev->vm_running &&
+        (status & VIRTIO_CONFIG_S_DRIVER_OK) && virtqueue_rewind(s->svq, 1)) {
+        /* poll stats queue for the element we have discarded when the VM
+         * was stopped */
+        virtio_balloon_receive_stats(vdev, s->svq);
+    }
+}
+
 static void virtio_balloon_instance_init(Object *obj)
 {
     VirtIOBalloon *s = VIRTIO_BALLOON(obj);
@@ -506,6 +518,7 @@ static void virtio_balloon_class_init(ObjectClass *klass, void *data)
     vdc->get_features = virtio_balloon_get_features;
     vdc->save = virtio_balloon_save_device;
     vdc->load = virtio_balloon_load_device;
+    vdc->set_status = virtio_balloon_set_status;
 }
 
 static const TypeInfo virtio_balloon_info = {
-- 
MST

[Qemu-devel] [PULL v2 10/14] virtio-pci: error out when both legacy and modern modes are disabled

[Michael S. Tsirkin]

From: Greg Kurz <gkurz@linux.vnet.ibm.com>

Without presuming if we got there because of a user mistake or some
more subtle bug in the tooling, it really does not make sense to
implement a non-functional device.

Signed-off-by: Greg Kurz <gkurz@linux.vnet.ibm.com>
Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 hw/virtio/virtio-pci.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/hw/virtio/virtio-pci.c b/hw/virtio/virtio-pci.c
index 89bc196..d70c986 100644
--- a/hw/virtio/virtio-pci.c
+++ b/hw/virtio/virtio-pci.c
@@ -1776,6 +1776,14 @@ static void virtio_pci_realize(PCIDevice *pci_dev, Error **errp)
         proxy->disable_legacy = pcie_port ? ON_OFF_AUTO_ON : ON_OFF_AUTO_OFF;
     }
 
+    if (!(virtio_pci_modern(proxy) || virtio_pci_legacy(proxy))) {
+        error_setg(errp, "device cannot work as neither modern nor legacy mode"
+                   " is enabled");
+        error_append_hint(errp, "Set either disable-modern or disable-legacy"
+                          " to off\n");
+        return;
+    }
+
     if (pcie_port && pci_is_express(pci_dev)) {
         int pos;
 
-- 
MST

[Qemu-devel] [PULL v2 11/14] vhost: don't set vring call if no vector

[Michael S. Tsirkin]

From: Jason Wang <jasowang@redhat.com>

We used to set vring call fd unconditionally even if guest driver does
not use MSIX for this vritqueue at all. This will cause lots of
unnecessary userspace access and other checks for drivers does not use
interrupt at all (e.g virtio-net pmd). So check and clean vring call
fd if guest does not use any vector for this virtqueue at
all.

Perf diffs (on rx) shows lots of cpus wasted on vhost_signal() were saved:

#
    28.12%  -27.82%  [vhost]           [k] vhost_signal
    14.44%   -1.69%  [kernel.vmlinux]  [k] copy_user_generic_string
     7.05%   +1.53%  [kernel.vmlinux]  [k] __free_page_frag
     6.51%   +5.53%  [vhost]           [k] vhost_get_vq_desc
...

Pktgen tests shows 15.8% improvement on rx pps and 6.5% on tx pps.

Before: RX 2.08Mpps TX 1.35Mpps
After:  RX 2.41Mpps TX 1.44Mpps

Signed-off-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 hw/virtio/vhost.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
index 3d0c807..bd051ab 100644
--- a/hw/virtio/vhost.c
+++ b/hw/virtio/vhost.c
@@ -822,6 +822,9 @@ static int vhost_virtqueue_start(struct vhost_dev *dev,
                                 struct vhost_virtqueue *vq,
                                 unsigned idx)
 {
+    BusState *qbus = BUS(qdev_get_parent_bus(DEVICE(vdev)));
+    VirtioBusState *vbus = VIRTIO_BUS(qbus);
+    VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(vbus);
     hwaddr s, l, a;
     int r;
     int vhost_vq_index = dev->vhost_ops->vhost_get_vq_index(dev, idx);
@@ -912,8 +915,19 @@ static int vhost_virtqueue_start(struct vhost_dev *dev,
         vhost_virtqueue_mask(dev, vdev, idx, false);
     }
 
+    if (k->query_guest_notifiers &&
+        k->query_guest_notifiers(qbus->parent) &&
+        virtio_queue_vector(vdev, idx) == VIRTIO_NO_VECTOR) {
+        file.fd = -1;
+        r = dev->vhost_ops->vhost_set_vring_call(dev, &file);
+        if (r) {
+            goto fail_vector;
+        }
+    }
+
     return 0;
 
+fail_vector:
 fail_kick:
 fail_alloc:
     cpu_physical_memory_unmap(vq->ring, virtio_queue_get_ring_size(vdev, idx),
-- 
MST

[Qemu-devel] [PULL v2 12/14] virtio-pci: minor refactoring

[Michael S. Tsirkin]

!legacy && !modern is shorter than !(legacy || modern).
I also perfer this (less ()s) as a matter of taste.

Cc: Greg Kurz <gkurz@linux.vnet.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 hw/virtio/virtio-pci.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/hw/virtio/virtio-pci.c b/hw/virtio/virtio-pci.c
index d70c986..bf19808 100644
--- a/hw/virtio/virtio-pci.c
+++ b/hw/virtio/virtio-pci.c
@@ -1776,7 +1776,7 @@ static void virtio_pci_realize(PCIDevice *pci_dev, Error **errp)
         proxy->disable_legacy = pcie_port ? ON_OFF_AUTO_ON : ON_OFF_AUTO_OFF;
     }
 
-    if (!(virtio_pci_modern(proxy) || virtio_pci_legacy(proxy))) {
+    if (!virtio_pci_modern(proxy) && !virtio_pci_legacy(proxy)) {
         error_setg(errp, "device cannot work as neither modern nor legacy mode"
                    " is enabled");
         error_append_hint(errp, "Set either disable-modern or disable-legacy"
-- 
MST

[Qemu-devel] [PULL v2 13/14] tests/acpi: speedup acpi tests

[Michael S. Tsirkin]

From: Marcel Apfelbaum <marcel@redhat.com>

Use kvm acceleration if available.
Disable kernel-irqchip and use qemu64 cpu
for both kvm and tcg cases.

Using kvm acceleration saves about a second
and disabling kernel-irqchip has no visible
performance impact.

Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Marcel Apfelbaum <marcel@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 tests/bios-tables-test.c | 39 ++++++++++++++++++++-------------------
 1 file changed, 20 insertions(+), 19 deletions(-)

diff --git a/tests/bios-tables-test.c b/tests/bios-tables-test.c
index de4019e..7e27ea9 100644
--- a/tests/bios-tables-test.c
+++ b/tests/bios-tables-test.c
@@ -711,9 +711,12 @@ static void test_acpi_one(const char *params, test_data *data)
 {
     char *args;
 
-    args = g_strdup_printf("-net none -display none %s "
+    /* Disable kernel irqchip to be able to override apic irq0. */
+    args = g_strdup_printf("-machine %s,accel=%s,kernel-irqchip=off "
+                           "-net none -display none %s "
                            "-drive id=hd0,if=none,file=%s,format=raw "
                            "-device ide-hd,drive=hd0 ",
+                           data->machine, "kvm:tcg",
                            params ? params : "", disk);
 
     qtest_start(args);
@@ -758,7 +761,7 @@ static void test_acpi_piix4_tcg(void)
     data.machine = MACHINE_PC;
     data.required_struct_types = base_required_struct_types;
     data.required_struct_types_len = ARRAY_SIZE(base_required_struct_types);
-    test_acpi_one("-machine accel=tcg", &data);
+    test_acpi_one(NULL, &data);
     free_test_data(&data);
 }
 
@@ -771,7 +774,7 @@ static void test_acpi_piix4_tcg_bridge(void)
     data.variant = ".bridge";
     data.required_struct_types = base_required_struct_types;
     data.required_struct_types_len = ARRAY_SIZE(base_required_struct_types);
-    test_acpi_one("-machine accel=tcg -device pci-bridge,chassis_nr=1", &data);
+    test_acpi_one("-device pci-bridge,chassis_nr=1", &data);
     free_test_data(&data);
 }
 
@@ -783,7 +786,7 @@ static void test_acpi_q35_tcg(void)
     data.machine = MACHINE_Q35;
     data.required_struct_types = base_required_struct_types;
     data.required_struct_types_len = ARRAY_SIZE(base_required_struct_types);
-    test_acpi_one("-machine q35,accel=tcg", &data);
+    test_acpi_one(NULL, &data);
     free_test_data(&data);
 }
 
@@ -796,7 +799,7 @@ static void test_acpi_q35_tcg_bridge(void)
     data.variant = ".bridge";
     data.required_struct_types = base_required_struct_types;
     data.required_struct_types_len = ARRAY_SIZE(base_required_struct_types);
-    test_acpi_one("-machine q35,accel=tcg -device pci-bridge,chassis_nr=1",
+    test_acpi_one("-device pci-bridge,chassis_nr=1",
                   &data);
     free_test_data(&data);
 }
@@ -808,8 +811,7 @@ static void test_acpi_piix4_tcg_cphp(void)
     memset(&data, 0, sizeof(data));
     data.machine = MACHINE_PC;
     data.variant = ".cphp";
-    test_acpi_one("-machine accel=tcg"
-                  " -smp 2,cores=3,sockets=2,maxcpus=6",
+    test_acpi_one("-smp 2,cores=3,sockets=2,maxcpus=6",
                   &data);
     free_test_data(&data);
 }
@@ -821,8 +823,7 @@ static void test_acpi_q35_tcg_cphp(void)
     memset(&data, 0, sizeof(data));
     data.machine = MACHINE_Q35;
     data.variant = ".cphp";
-    test_acpi_one("-machine q35,accel=tcg"
-                  " -smp 2,cores=3,sockets=2,maxcpus=6",
+    test_acpi_one(" -smp 2,cores=3,sockets=2,maxcpus=6",
                   &data);
     free_test_data(&data);
 }
@@ -840,7 +841,7 @@ static void test_acpi_q35_tcg_ipmi(void)
     data.variant = ".ipmibt";
     data.required_struct_types = ipmi_required_struct_types;
     data.required_struct_types_len = ARRAY_SIZE(ipmi_required_struct_types);
-    test_acpi_one("-machine q35,accel=tcg -device ipmi-bmc-sim,id=bmc0"
+    test_acpi_one("-device ipmi-bmc-sim,id=bmc0"
                   " -device isa-ipmi-bt,bmc=bmc0",
                   &data);
     free_test_data(&data);
@@ -858,7 +859,7 @@ static void test_acpi_piix4_tcg_ipmi(void)
     data.variant = ".ipmikcs";
     data.required_struct_types = ipmi_required_struct_types;
     data.required_struct_types_len = ARRAY_SIZE(ipmi_required_struct_types);
-    test_acpi_one("-machine accel=tcg -device ipmi-bmc-sim,id=bmc0"
+    test_acpi_one("-device ipmi-bmc-sim,id=bmc0"
                   " -device isa-ipmi-kcs,irq=0,bmc=bmc0",
                   &data);
     free_test_data(&data);
@@ -876,14 +877,14 @@ int main(int argc, char *argv[])
     g_test_init(&argc, &argv, NULL);
 
     if (strcmp(arch, "i386") == 0 || strcmp(arch, "x86_64") == 0) {
-        qtest_add_func("acpi/piix4/tcg", test_acpi_piix4_tcg);
-        qtest_add_func("acpi/piix4/tcg/bridge", test_acpi_piix4_tcg_bridge);
-        qtest_add_func("acpi/q35/tcg", test_acpi_q35_tcg);
-        qtest_add_func("acpi/q35/tcg/bridge", test_acpi_q35_tcg_bridge);
-        qtest_add_func("acpi/piix4/tcg/ipmi", test_acpi_piix4_tcg_ipmi);
-        qtest_add_func("acpi/q35/tcg/ipmi", test_acpi_q35_tcg_ipmi);
-        qtest_add_func("acpi/piix4/tcg/cpuhp", test_acpi_piix4_tcg_cphp);
-        qtest_add_func("acpi/q35/tcg/cpuhp", test_acpi_q35_tcg_cphp);
+        qtest_add_func("acpi/piix4", test_acpi_piix4_tcg);
+        qtest_add_func("acpi/piix4/bridge", test_acpi_piix4_tcg_bridge);
+        qtest_add_func("acpi/q35", test_acpi_q35_tcg);
+        qtest_add_func("acpi/q35/bridge", test_acpi_q35_tcg_bridge);
+        qtest_add_func("acpi/piix4/ipmi", test_acpi_piix4_tcg_ipmi);
+        qtest_add_func("acpi/q35/ipmi", test_acpi_q35_tcg_ipmi);
+        qtest_add_func("acpi/piix4/cpuhp", test_acpi_piix4_tcg_cphp);
+        qtest_add_func("acpi/q35/cpuhp", test_acpi_q35_tcg_cphp);
     }
     ret = g_test_run();
     boot_sector_cleanup(disk);
-- 
MST

[Qemu-devel] [PULL v2 14/14] vhost-vsock: add virtio sockets device

[Michael S. Tsirkin]

From: Stefan Hajnoczi <stefanha@redhat.com>

Implement the new virtio sockets device for host<->guest communication
using the Sockets API.  Most of the work is done in a vhost kernel
driver so that virtio-vsock can hook into the AF_VSOCK address family.
The QEMU vhost-vsock device handles configuration and live migration
while the rx/tx happens in the vhost_vsock.ko Linux kernel driver.

The vsock device must be given a CID (host-wide unique address):

  # qemu -device vhost-vsock-pci,id=vhost-vsock-pci0,guest-cid=3 ...

For more information see:
http://qemu-project.org/Features/VirtioVsock

[Endianness fixes and virtio-ccw support by Claudio Imbrenda
<imbrenda@linux.vnet.ibm.com>]

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
[mst: rebase to master]
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 configure                         |  10 +
 hw/s390x/virtio-ccw.h             |  15 ++
 hw/virtio/virtio-pci.h            |  18 ++
 include/hw/pci/pci.h              |   1 +
 include/hw/virtio/vhost-backend.h |   5 +
 include/hw/virtio/vhost-vsock.h   |  41 ++++
 hw/s390x/virtio-ccw.c             |  54 +++++
 hw/virtio/vhost-backend.c         |  17 ++
 hw/virtio/vhost-vsock.c           | 417 ++++++++++++++++++++++++++++++++++++++
 hw/virtio/virtio-pci.c            |  51 +++++
 hw/virtio/Makefile.objs           |   2 +
 11 files changed, 631 insertions(+)
 create mode 100644 include/hw/virtio/vhost-vsock.h
 create mode 100644 hw/virtio/vhost-vsock.c

diff --git a/configure b/configure
index 5a9bda1..331c36f 100755
--- a/configure
+++ b/configure
@@ -229,6 +229,7 @@ xfs=""
 
 vhost_net="no"
 vhost_scsi="no"
+vhost_vsock="no"
 kvm="no"
 rdma=""
 gprof="no"
@@ -674,6 +675,7 @@ Haiku)
   kvm="yes"
   vhost_net="yes"
   vhost_scsi="yes"
+  vhost_vsock="yes"
   QEMU_INCLUDES="-I\$(SRC_PATH)/linux-headers -I$(pwd)/linux-headers $QEMU_INCLUDES"
 ;;
 esac
@@ -1017,6 +1019,10 @@ for opt do
   ;;
   --enable-vhost-scsi) vhost_scsi="yes"
   ;;
+  --disable-vhost-vsock) vhost_vsock="no"
+  ;;
+  --enable-vhost-vsock) vhost_vsock="yes"
+  ;;
   --disable-opengl) opengl="no"
   ;;
   --enable-opengl) opengl="yes"
@@ -4883,6 +4889,7 @@ echo "uuid support      $uuid"
 echo "libcap-ng support $cap_ng"
 echo "vhost-net support $vhost_net"
 echo "vhost-scsi support $vhost_scsi"
+echo "vhost-vsock support $vhost_vsock"
 echo "Trace backends    $trace_backends"
 if have_backend "simple"; then
 echo "Trace output file $trace_file-<pid>"
@@ -5264,6 +5271,9 @@ fi
 if test "$vhost_net" = "yes" ; then
   echo "CONFIG_VHOST_NET_USED=y" >> $config_host_mak
 fi
+if test "$vhost_vsock" = "yes" ; then
+  echo "CONFIG_VHOST_VSOCK=y" >> $config_host_mak
+fi
 if test "$blobs" = "yes" ; then
   echo "INSTALL_BLOBS=yes" >> $config_host_mak
 fi
diff --git a/hw/s390x/virtio-ccw.h b/hw/s390x/virtio-ccw.h
index 1c6bc86..904e357 100644
--- a/hw/s390x/virtio-ccw.h
+++ b/hw/s390x/virtio-ccw.h
@@ -23,6 +23,9 @@
 #include "hw/virtio/virtio-balloon.h"
 #include "hw/virtio/virtio-rng.h"
 #include "hw/virtio/virtio-bus.h"
+#ifdef CONFIG_VHOST_VSOCK
+#include "hw/virtio/vhost-vsock.h"
+#endif /* CONFIG_VHOST_VSOCK */
 
 #include "hw/s390x/s390_flic.h"
 #include "hw/s390x/css.h"
@@ -197,4 +200,16 @@ typedef struct V9fsCCWState {
 
 #endif /* CONFIG_VIRTFS */
 
+#ifdef CONFIG_VHOST_VSOCK
+#define TYPE_VHOST_VSOCK_CCW "vhost-vsock-ccw"
+#define VHOST_VSOCK_CCW(obj) \
+    OBJECT_CHECK(VHostVSockCCWState, (obj), TYPE_VHOST_VSOCK_CCW)
+
+typedef struct VHostVSockCCWState {
+    VirtioCcwDevice parent_obj;
+    VHostVSock vdev;
+} VHostVSockCCWState;
+
+#endif /* CONFIG_VHOST_VSOCK */
+
 #endif
diff --git a/hw/virtio/virtio-pci.h b/hw/virtio/virtio-pci.h
index a745512..0698157 100644
--- a/hw/virtio/virtio-pci.h
+++ b/hw/virtio/virtio-pci.h
@@ -31,6 +31,9 @@
 #ifdef CONFIG_VHOST_SCSI
 #include "hw/virtio/vhost-scsi.h"
 #endif
+#ifdef CONFIG_VHOST_VSOCK
+#include "hw/virtio/vhost-vsock.h"
+#endif
 
 typedef struct VirtIOPCIProxy VirtIOPCIProxy;
 typedef struct VirtIOBlkPCI VirtIOBlkPCI;
@@ -44,6 +47,7 @@ typedef struct VirtIOInputPCI VirtIOInputPCI;
 typedef struct VirtIOInputHIDPCI VirtIOInputHIDPCI;
 typedef struct VirtIOInputHostPCI VirtIOInputHostPCI;
 typedef struct VirtIOGPUPCI VirtIOGPUPCI;
+typedef struct VHostVSockPCI VHostVSockPCI;
 
 /* virtio-pci-bus */
 
@@ -329,6 +333,20 @@ struct VirtIOGPUPCI {
     VirtIOGPU vdev;
 };
 
+#ifdef CONFIG_VHOST_VSOCK
+/*
+ * vhost-vsock-pci: This extends VirtioPCIProxy.
+ */
+#define TYPE_VHOST_VSOCK_PCI "vhost-vsock-pci"
+#define VHOST_VSOCK_PCI(obj) \
+        OBJECT_CHECK(VHostVSockPCI, (obj), TYPE_VHOST_VSOCK_PCI)
+
+struct VHostVSockPCI {
+    VirtIOPCIProxy parent_obj;
+    VHostVSock vdev;
+};
+#endif
+
 /* Virtio ABI version, if we increment this, we break the guest driver. */
 #define VIRTIO_PCI_ABI_VERSION          0
 
diff --git a/include/hw/pci/pci.h b/include/hw/pci/pci.h
index 929ec2f..e8b83bb 100644
--- a/include/hw/pci/pci.h
+++ b/include/hw/pci/pci.h
@@ -79,6 +79,7 @@
 #define PCI_DEVICE_ID_VIRTIO_SCSI        0x1004
 #define PCI_DEVICE_ID_VIRTIO_RNG         0x1005
 #define PCI_DEVICE_ID_VIRTIO_9P          0x1009
+#define PCI_DEVICE_ID_VIRTIO_VSOCK       0x1012
 
 #define PCI_VENDOR_ID_REDHAT             0x1b36
 #define PCI_DEVICE_ID_REDHAT_BRIDGE      0x0001
diff --git a/include/hw/virtio/vhost-backend.h b/include/hw/virtio/vhost-backend.h
index cf7f0b5..6e90703 100644
--- a/include/hw/virtio/vhost-backend.h
+++ b/include/hw/virtio/vhost-backend.h
@@ -73,6 +73,9 @@ typedef int (*vhost_migration_done_op)(struct vhost_dev *dev,
 typedef bool (*vhost_backend_can_merge_op)(struct vhost_dev *dev,
                                            uint64_t start1, uint64_t size1,
                                            uint64_t start2, uint64_t size2);
+typedef int (*vhost_vsock_set_guest_cid_op)(struct vhost_dev *dev,
+                                            uint64_t guest_cid);
+typedef int (*vhost_vsock_set_running_op)(struct vhost_dev *dev, int start);
 
 typedef struct VhostOps {
     VhostBackendType backend_type;
@@ -102,6 +105,8 @@ typedef struct VhostOps {
     vhost_requires_shm_log_op vhost_requires_shm_log;
     vhost_migration_done_op vhost_migration_done;
     vhost_backend_can_merge_op vhost_backend_can_merge;
+    vhost_vsock_set_guest_cid_op vhost_vsock_set_guest_cid;
+    vhost_vsock_set_running_op vhost_vsock_set_running;
 } VhostOps;
 
 extern const VhostOps user_ops;
diff --git a/include/hw/virtio/vhost-vsock.h b/include/hw/virtio/vhost-vsock.h
new file mode 100644
index 0000000..7b9205f
--- /dev/null
+++ b/include/hw/virtio/vhost-vsock.h
@@ -0,0 +1,41 @@
+/*
+ * Vhost vsock virtio device
+ *
+ * Copyright 2015 Red Hat, Inc.
+ *
+ * Authors:
+ *  Stefan Hajnoczi <stefanha@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or
+ * (at your option) any later version.  See the COPYING file in the
+ * top-level directory.
+ */
+
+#ifndef _QEMU_VHOST_VSOCK_H
+#define _QEMU_VHOST_VSOCK_H
+
+#include "hw/virtio/virtio.h"
+#include "hw/virtio/vhost.h"
+
+#define TYPE_VHOST_VSOCK "vhost-vsock-device"
+#define VHOST_VSOCK(obj) \
+        OBJECT_CHECK(VHostVSock, (obj), TYPE_VHOST_VSOCK)
+
+typedef struct {
+    uint64_t guest_cid;
+    char *vhostfd;
+} VHostVSockConf;
+
+typedef struct {
+    /*< private >*/
+    VirtIODevice parent;
+    VHostVSockConf conf;
+    struct vhost_virtqueue vhost_vqs[2];
+    struct vhost_dev vhost_dev;
+    VirtQueue *event_vq;
+    QEMUTimer *post_load_timer;
+
+    /*< public >*/
+} VHostVSock;
+
+#endif /* _QEMU_VHOST_VSOCK_H */
diff --git a/hw/s390x/virtio-ccw.c b/hw/s390x/virtio-ccw.c
index a554a24..9678956 100644
--- a/hw/s390x/virtio-ccw.c
+++ b/hw/s390x/virtio-ccw.c
@@ -1658,6 +1658,57 @@ static const TypeInfo virtio_ccw_9p_info = {
 };
 #endif
 
+#ifdef CONFIG_VHOST_VSOCK
+
+static Property vhost_vsock_ccw_properties[] = {
+    DEFINE_PROP_CSS_DEV_ID("devno", VirtioCcwDevice, parent_obj.bus_id),
+    DEFINE_PROP_UINT32("max_revision", VirtioCcwDevice, max_rev,
+                       VIRTIO_CCW_MAX_REV),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void vhost_vsock_ccw_realize(VirtioCcwDevice *ccw_dev, Error **errp)
+{
+    VHostVSockCCWState *dev = VHOST_VSOCK_CCW(ccw_dev);
+    DeviceState *vdev = DEVICE(&dev->vdev);
+    Error *err = NULL;
+
+    qdev_set_parent_bus(vdev, BUS(&ccw_dev->bus));
+    object_property_set_bool(OBJECT(vdev), true, "realized", &err);
+    if (err) {
+        error_propagate(errp, err);
+    }
+}
+
+static void vhost_vsock_ccw_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    VirtIOCCWDeviceClass *k = VIRTIO_CCW_DEVICE_CLASS(klass);
+
+    k->realize = vhost_vsock_ccw_realize;
+    k->exit = virtio_ccw_exit;
+    set_bit(DEVICE_CATEGORY_MISC, dc->categories);
+    dc->props = vhost_vsock_ccw_properties;
+    dc->reset = virtio_ccw_reset;
+}
+
+static void vhost_vsock_ccw_instance_init(Object *obj)
+{
+    VHostVSockCCWState *dev = VHOST_VSOCK_CCW(obj);
+
+    virtio_instance_init_common(obj, &dev->vdev, sizeof(dev->vdev),
+                                TYPE_VHOST_VSOCK);
+}
+
+static const TypeInfo vhost_vsock_ccw_info = {
+    .name          = TYPE_VHOST_VSOCK_CCW,
+    .parent        = TYPE_VIRTIO_CCW_DEVICE,
+    .instance_size = sizeof(VHostVSockCCWState),
+    .instance_init = vhost_vsock_ccw_instance_init,
+    .class_init    = vhost_vsock_ccw_class_init,
+};
+#endif
+
 static void virtio_ccw_register(void)
 {
     type_register_static(&virtio_ccw_bus_info);
@@ -1674,6 +1725,9 @@ static void virtio_ccw_register(void)
 #ifdef CONFIG_VIRTFS
     type_register_static(&virtio_ccw_9p_info);
 #endif
+#ifdef CONFIG_VHOST_VSOCK
+    type_register_static(&vhost_vsock_ccw_info);
+#endif
 }
 
 type_init(virtio_ccw_register)
diff --git a/hw/virtio/vhost-backend.c b/hw/virtio/vhost-backend.c
index 7681f15..272a5ec 100644
--- a/hw/virtio/vhost-backend.c
+++ b/hw/virtio/vhost-backend.c
@@ -172,6 +172,19 @@ static int vhost_kernel_get_vq_index(struct vhost_dev *dev, int idx)
     return idx - dev->vq_index;
 }
 
+#ifdef CONFIG_VHOST_VSOCK
+static int vhost_kernel_vsock_set_guest_cid(struct vhost_dev *dev,
+                                            uint64_t guest_cid)
+{
+    return vhost_kernel_call(dev, VHOST_VSOCK_SET_GUEST_CID, &guest_cid);
+}
+
+static int vhost_kernel_vsock_set_running(struct vhost_dev *dev, int start)
+{
+    return vhost_kernel_call(dev, VHOST_VSOCK_SET_RUNNING, &start);
+}
+#endif /* CONFIG_VHOST_VSOCK */
+
 static const VhostOps kernel_ops = {
         .backend_type = VHOST_BACKEND_TYPE_KERNEL,
         .vhost_backend_init = vhost_kernel_init,
@@ -197,6 +210,10 @@ static const VhostOps kernel_ops = {
         .vhost_set_owner = vhost_kernel_set_owner,
         .vhost_reset_device = vhost_kernel_reset_device,
         .vhost_get_vq_index = vhost_kernel_get_vq_index,
+#ifdef CONFIG_VHOST_VSOCK
+        .vhost_vsock_set_guest_cid = vhost_kernel_vsock_set_guest_cid,
+        .vhost_vsock_set_running = vhost_kernel_vsock_set_running,
+#endif /* CONFIG_VHOST_VSOCK */
 };
 
 int vhost_set_backend_type(struct vhost_dev *dev, VhostBackendType backend_type)
diff --git a/hw/virtio/vhost-vsock.c b/hw/virtio/vhost-vsock.c
new file mode 100644
index 0000000..bde2456
--- /dev/null
+++ b/hw/virtio/vhost-vsock.c
@@ -0,0 +1,417 @@
+/*
+ * Virtio vsock device
+ *
+ * Copyright 2015 Red Hat, Inc.
+ *
+ * Authors:
+ *  Stefan Hajnoczi <stefanha@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or
+ * (at your option) any later version.  See the COPYING file in the
+ * top-level directory.
+ */
+
+#include <sys/ioctl.h>
+#include "qemu/osdep.h"
+#include "standard-headers/linux/virtio_vsock.h"
+#include "qapi/error.h"
+#include "hw/virtio/virtio-bus.h"
+#include "hw/virtio/virtio-access.h"
+#include "migration/migration.h"
+#include "qemu/error-report.h"
+#include "hw/virtio/vhost-vsock.h"
+#include "qemu/iov.h"
+#include "monitor/monitor.h"
+
+enum {
+    VHOST_VSOCK_SAVEVM_VERSION = 0,
+
+    VHOST_VSOCK_QUEUE_SIZE = 128,
+};
+
+static void vhost_vsock_get_config(VirtIODevice *vdev, uint8_t *config)
+{
+    VHostVSock *vsock = VHOST_VSOCK(vdev);
+    struct virtio_vsock_config vsockcfg = {};
+
+    virtio_stq_p(vdev, &vsockcfg.guest_cid, vsock->conf.guest_cid);
+    memcpy(config, &vsockcfg, sizeof(vsockcfg));
+}
+
+static int vhost_vsock_set_guest_cid(VHostVSock *vsock)
+{
+    const VhostOps *vhost_ops = vsock->vhost_dev.vhost_ops;
+    int ret;
+
+    if (!vhost_ops->vhost_vsock_set_guest_cid) {
+        return -ENOSYS;
+    }
+
+    ret = vhost_ops->vhost_vsock_set_guest_cid(&vsock->vhost_dev,
+                                               vsock->conf.guest_cid);
+    if (ret < 0) {
+        return -errno;
+    }
+    return 0;
+}
+
+static int vhost_vsock_set_running(VHostVSock *vsock, int start)
+{
+    const VhostOps *vhost_ops = vsock->vhost_dev.vhost_ops;
+    int ret;
+
+    if (!vhost_ops->vhost_vsock_set_running) {
+        return -ENOSYS;
+    }
+
+    ret = vhost_ops->vhost_vsock_set_running(&vsock->vhost_dev, start);
+    if (ret < 0) {
+        return -errno;
+    }
+    return 0;
+}
+
+static void vhost_vsock_start(VirtIODevice *vdev)
+{
+    VHostVSock *vsock = VHOST_VSOCK(vdev);
+    BusState *qbus = BUS(qdev_get_parent_bus(DEVICE(vdev)));
+    VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(qbus);
+    int ret;
+    int i;
+
+    if (!k->set_guest_notifiers) {
+        error_report("binding does not support guest notifiers");
+        return;
+    }
+
+    ret = vhost_dev_enable_notifiers(&vsock->vhost_dev, vdev);
+    if (ret < 0) {
+        error_report("Error enabling host notifiers: %d", -ret);
+        return;
+    }
+
+    ret = k->set_guest_notifiers(qbus->parent, vsock->vhost_dev.nvqs, true);
+    if (ret < 0) {
+        error_report("Error binding guest notifier: %d", -ret);
+        goto err_host_notifiers;
+    }
+
+    vsock->vhost_dev.acked_features = vdev->guest_features;
+    ret = vhost_dev_start(&vsock->vhost_dev, vdev);
+    if (ret < 0) {
+        error_report("Error starting vhost: %d", -ret);
+        goto err_guest_notifiers;
+    }
+
+    ret = vhost_vsock_set_running(vsock, 1);
+    if (ret < 0) {
+        error_report("Error starting vhost vsock: %d", -ret);
+        goto err_dev_start;
+    }
+
+    /* 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 < vsock->vhost_dev.nvqs; i++) {
+        vhost_virtqueue_mask(&vsock->vhost_dev, vdev, i, false);
+    }
+
+    return;
+
+err_dev_start:
+    vhost_dev_stop(&vsock->vhost_dev, vdev);
+err_guest_notifiers:
+    k->set_guest_notifiers(qbus->parent, vsock->vhost_dev.nvqs, false);
+err_host_notifiers:
+    vhost_dev_disable_notifiers(&vsock->vhost_dev, vdev);
+}
+
+static void vhost_vsock_stop(VirtIODevice *vdev)
+{
+    VHostVSock *vsock = VHOST_VSOCK(vdev);
+    BusState *qbus = BUS(qdev_get_parent_bus(DEVICE(vdev)));
+    VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(qbus);
+    int ret;
+
+    if (!k->set_guest_notifiers) {
+        return;
+    }
+
+    ret = vhost_vsock_set_running(vsock, 0);
+    if (ret < 0) {
+        error_report("vhost vsock set running failed: %d", ret);
+        return;
+    }
+
+    vhost_dev_stop(&vsock->vhost_dev, vdev);
+
+    ret = k->set_guest_notifiers(qbus->parent, vsock->vhost_dev.nvqs, false);
+    if (ret < 0) {
+        error_report("vhost guest notifier cleanup failed: %d", ret);
+        return;
+    }
+
+    vhost_dev_disable_notifiers(&vsock->vhost_dev, vdev);
+}
+
+static void vhost_vsock_set_status(VirtIODevice *vdev, uint8_t status)
+{
+    VHostVSock *vsock = VHOST_VSOCK(vdev);
+    bool should_start = status & VIRTIO_CONFIG_S_DRIVER_OK;
+
+    if (!vdev->vm_running) {
+        should_start = false;
+    }
+
+    if (vsock->vhost_dev.started == should_start) {
+        return;
+    }
+
+    if (should_start) {
+        vhost_vsock_start(vdev);
+    } else {
+        vhost_vsock_stop(vdev);
+    }
+}
+
+static uint64_t vhost_vsock_get_features(VirtIODevice *vdev,
+                                         uint64_t requested_features,
+                                         Error **errp)
+{
+    /* No feature bits used yet */
+    return requested_features;
+}
+
+static void vhost_vsock_handle_output(VirtIODevice *vdev, VirtQueue *vq)
+{
+    /* Do nothing */
+}
+
+static void vhost_vsock_guest_notifier_mask(VirtIODevice *vdev, int idx,
+                                            bool mask)
+{
+    VHostVSock *vsock = VHOST_VSOCK(vdev);
+
+    vhost_virtqueue_mask(&vsock->vhost_dev, vdev, idx, mask);
+}
+
+static bool vhost_vsock_guest_notifier_pending(VirtIODevice *vdev, int idx)
+{
+    VHostVSock *vsock = VHOST_VSOCK(vdev);
+
+    return vhost_virtqueue_pending(&vsock->vhost_dev, idx);
+}
+
+static void vhost_vsock_send_transport_reset(VHostVSock *vsock)
+{
+    VirtQueueElement *elem;
+    VirtQueue *vq = vsock->event_vq;
+    struct virtio_vsock_event event = {
+        .id = cpu_to_le32(VIRTIO_VSOCK_EVENT_TRANSPORT_RESET),
+    };
+
+    elem = virtqueue_pop(vq, sizeof(VirtQueueElement));
+    if (!elem) {
+        error_report("vhost-vsock missed transport reset event");
+        return;
+    }
+
+    if (elem->out_num) {
+        error_report("invalid vhost-vsock event virtqueue element with "
+                     "out buffers");
+        goto out;
+    }
+
+    if (iov_from_buf(elem->in_sg, elem->in_num, 0,
+                     &event, sizeof(event)) != sizeof(event)) {
+        error_report("vhost-vsock event virtqueue element is too short");
+        goto out;
+    }
+
+    virtqueue_push(vq, elem, sizeof(event));
+    virtio_notify(VIRTIO_DEVICE(vsock), vq);
+
+out:
+    g_free(elem);
+}
+
+static void vhost_vsock_save(QEMUFile *f, void *opaque, size_t size)
+{
+    VHostVSock *vsock = opaque;
+    VirtIODevice *vdev = VIRTIO_DEVICE(vsock);
+
+    /* At this point, backend must be stopped, otherwise
+     * it might keep writing to memory. */
+    assert(!vsock->vhost_dev.started);
+    virtio_save(vdev, f);
+}
+
+static void vhost_vsock_post_load_timer_cleanup(VHostVSock *vsock)
+{
+    if (!vsock->post_load_timer) {
+        return;
+    }
+
+    timer_del(vsock->post_load_timer);
+    timer_free(vsock->post_load_timer);
+    vsock->post_load_timer = NULL;
+}
+
+static void vhost_vsock_post_load_timer_cb(void *opaque)
+{
+    VHostVSock *vsock = opaque;
+
+    vhost_vsock_post_load_timer_cleanup(vsock);
+    vhost_vsock_send_transport_reset(vsock);
+}
+
+static int vhost_vsock_load(QEMUFile *f, void *opaque, size_t size)
+{
+    VHostVSock *vsock = opaque;
+    VirtIODevice *vdev = VIRTIO_DEVICE(vsock);
+    int ret;
+
+    ret = virtio_load(vdev, f, VHOST_VSOCK_SAVEVM_VERSION);
+    if (ret) {
+        return ret;
+    }
+
+    if (virtio_queue_get_addr(vdev, 2)) {
+        /* Defer transport reset event to a vm clock timer so that virtqueue
+         * changes happen after migration has completed.
+         */
+        assert(!vsock->post_load_timer);
+        vsock->post_load_timer =
+            timer_new_ns(QEMU_CLOCK_VIRTUAL,
+                         vhost_vsock_post_load_timer_cb,
+                         vsock);
+        timer_mod(vsock->post_load_timer, 1);
+    }
+
+    return 0;
+}
+
+VMSTATE_VIRTIO_DEVICE(vhost_vsock, VHOST_VSOCK_SAVEVM_VERSION,
+                      vhost_vsock_load, vhost_vsock_save);
+
+static void vhost_vsock_device_realize(DeviceState *dev, Error **errp)
+{
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VHostVSock *vsock = VHOST_VSOCK(dev);
+    int vhostfd;
+    int ret;
+
+    /* Refuse to use reserved CID numbers */
+    if (vsock->conf.guest_cid <= 2) {
+        error_setg(errp, "guest-cid property must be greater than 2");
+        return;
+    }
+
+    if (vsock->conf.guest_cid > UINT32_MAX) {
+        error_setg(errp, "guest-cid property must be a 32-bit number");
+        return;
+    }
+
+    if (vsock->conf.vhostfd) {
+        vhostfd = monitor_fd_param(cur_mon, vsock->conf.vhostfd, errp);
+        if (vhostfd == -1) {
+            error_prepend(errp, "vhost-vsock: unable to parse vhostfd: ");
+            return;
+        }
+    } else {
+        vhostfd = open("/dev/vhost-vsock", O_RDWR);
+        if (vhostfd < 0) {
+            error_setg_errno(errp, -errno,
+                             "vhost-vsock: failed to open vhost device");
+            return;
+        }
+    }
+
+    virtio_init(vdev, "vhost-vsock", VIRTIO_ID_VSOCK,
+                sizeof(struct virtio_vsock_config));
+
+    /* Receive and transmit queues belong to vhost */
+    virtio_add_queue(vdev, VHOST_VSOCK_QUEUE_SIZE, vhost_vsock_handle_output);
+    virtio_add_queue(vdev, VHOST_VSOCK_QUEUE_SIZE, vhost_vsock_handle_output);
+
+    /* The event queue belongs to QEMU */
+    vsock->event_vq = virtio_add_queue(vdev, VHOST_VSOCK_QUEUE_SIZE,
+                                       vhost_vsock_handle_output);
+
+    vsock->vhost_dev.nvqs = ARRAY_SIZE(vsock->vhost_vqs);
+    vsock->vhost_dev.vqs = vsock->vhost_vqs;
+    ret = vhost_dev_init(&vsock->vhost_dev, (void *)(uintptr_t)vhostfd,
+                         VHOST_BACKEND_TYPE_KERNEL, 0);
+    if (ret < 0) {
+        error_setg_errno(errp, -ret, "vhost-vsock: vhost_dev_init failed");
+        goto err_virtio;
+    }
+
+    ret = vhost_vsock_set_guest_cid(vsock);
+    if (ret < 0) {
+        error_setg_errno(errp, -ret, "vhost-vsock: unable to set guest cid");
+        goto err_vhost_dev;
+    }
+
+    vsock->post_load_timer = NULL;
+    return;
+
+err_vhost_dev:
+    vhost_dev_cleanup(&vsock->vhost_dev);
+err_virtio:
+    virtio_cleanup(vdev);
+    close(vhostfd);
+    return;
+}
+
+static void vhost_vsock_device_unrealize(DeviceState *dev, Error **errp)
+{
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VHostVSock *vsock = VHOST_VSOCK(dev);
+
+    vhost_vsock_post_load_timer_cleanup(vsock);
+
+    /* This will stop vhost backend if appropriate. */
+    vhost_vsock_set_status(vdev, 0);
+
+    vhost_dev_cleanup(&vsock->vhost_dev);
+    virtio_cleanup(vdev);
+}
+
+static Property vhost_vsock_properties[] = {
+    DEFINE_PROP_UINT64("guest-cid", VHostVSock, conf.guest_cid, 0),
+    DEFINE_PROP_STRING("vhostfd", VHostVSock, conf.vhostfd),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void vhost_vsock_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
+
+    dc->props = vhost_vsock_properties;
+    dc->vmsd = &vmstate_virtio_vhost_vsock;
+    set_bit(DEVICE_CATEGORY_MISC, dc->categories);
+    vdc->realize = vhost_vsock_device_realize;
+    vdc->unrealize = vhost_vsock_device_unrealize;
+    vdc->get_features = vhost_vsock_get_features;
+    vdc->get_config = vhost_vsock_get_config;
+    vdc->set_status = vhost_vsock_set_status;
+    vdc->guest_notifier_mask = vhost_vsock_guest_notifier_mask;
+    vdc->guest_notifier_pending = vhost_vsock_guest_notifier_pending;
+}
+
+static const TypeInfo vhost_vsock_info = {
+    .name = TYPE_VHOST_VSOCK,
+    .parent = TYPE_VIRTIO_DEVICE,
+    .instance_size = sizeof(VHostVSock),
+    .class_init = vhost_vsock_class_init,
+};
+
+static void vhost_vsock_register_types(void)
+{
+    type_register_static(&vhost_vsock_info);
+}
+
+type_init(vhost_vsock_register_types)
diff --git a/hw/virtio/virtio-pci.c b/hw/virtio/virtio-pci.c
index bf19808..dde71a5 100644
--- a/hw/virtio/virtio-pci.c
+++ b/hw/virtio/virtio-pci.c
@@ -2071,6 +2071,54 @@ static const TypeInfo vhost_scsi_pci_info = {
 };
 #endif
 
+/* vhost-vsock-pci */
+
+#ifdef CONFIG_VHOST_VSOCK
+static Property vhost_vsock_pci_properties[] = {
+    DEFINE_PROP_UINT32("vectors", VirtIOPCIProxy, nvectors, 3),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void vhost_vsock_pci_realize(VirtIOPCIProxy *vpci_dev, Error **errp)
+{
+    VHostVSockPCI *dev = VHOST_VSOCK_PCI(vpci_dev);
+    DeviceState *vdev = DEVICE(&dev->vdev);
+
+    qdev_set_parent_bus(vdev, BUS(&vpci_dev->bus));
+    object_property_set_bool(OBJECT(vdev), true, "realized", errp);
+}
+
+static void vhost_vsock_pci_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    VirtioPCIClass *k = VIRTIO_PCI_CLASS(klass);
+    PCIDeviceClass *pcidev_k = PCI_DEVICE_CLASS(klass);
+    k->realize = vhost_vsock_pci_realize;
+    set_bit(DEVICE_CATEGORY_MISC, dc->categories);
+    dc->props = vhost_vsock_pci_properties;
+    pcidev_k->vendor_id = PCI_VENDOR_ID_REDHAT_QUMRANET;
+    pcidev_k->device_id = PCI_DEVICE_ID_VIRTIO_VSOCK;
+    pcidev_k->revision = 0x00;
+    pcidev_k->class_id = PCI_CLASS_COMMUNICATION_OTHER;
+}
+
+static void vhost_vsock_pci_instance_init(Object *obj)
+{
+    VHostVSockPCI *dev = VHOST_VSOCK_PCI(obj);
+
+    virtio_instance_init_common(obj, &dev->vdev, sizeof(dev->vdev),
+                                TYPE_VHOST_VSOCK);
+}
+
+static const TypeInfo vhost_vsock_pci_info = {
+    .name          = TYPE_VHOST_VSOCK_PCI,
+    .parent        = TYPE_VIRTIO_PCI,
+    .instance_size = sizeof(VHostVSockPCI),
+    .instance_init = vhost_vsock_pci_instance_init,
+    .class_init    = vhost_vsock_pci_class_init,
+};
+#endif
+
 /* virtio-balloon-pci */
 
 static Property virtio_balloon_pci_properties[] = {
@@ -2501,6 +2549,9 @@ static void virtio_pci_register_types(void)
 #ifdef CONFIG_VHOST_SCSI
     type_register_static(&vhost_scsi_pci_info);
 #endif
+#ifdef CONFIG_VHOST_VSOCK
+    type_register_static(&vhost_vsock_pci_info);
+#endif
 }
 
 type_init(virtio_pci_register_types)
diff --git a/hw/virtio/Makefile.objs b/hw/virtio/Makefile.objs
index 3e2b175..e716308 100644
--- a/hw/virtio/Makefile.objs
+++ b/hw/virtio/Makefile.objs
@@ -5,3 +5,5 @@ common-obj-y += virtio-mmio.o
 
 obj-y += virtio.o virtio-balloon.o 
 obj-$(CONFIG_LINUX) += vhost.o vhost-backend.o vhost-user.o
+
+obj-$(CONFIG_VHOST_VSOCK) += vhost-vsock.o
-- 
MST

Re: [Qemu-devel] [PULL v2 00/14] virtio, vhost, pc: fixes and updates

[Peter Maydell]

On 9 September 2016 at 22:47, Michael S. Tsirkin <mst@redhat.com> wrote:
> The following changes since commit c2a57aae9a1c3dd7de77daf5478df10379aeeebf:
>
>   Merge remote-tracking branch 'remotes/famz/tags/docker-pull-request' into staging (2016-09-09 12:49:41 +0100)
>
> are available in the git repository at:
>
>   git://git.kernel.org/pub/scm/virt/kvm/mst/qemu.git tags/for_upstream
>
> for you to fetch changes up to fc0b9b0e1cbb49017ea882758634cf876be17bc3:
>
>   vhost-vsock: add virtio sockets device (2016-09-10 00:28:08 +0300)
>
> ----------------------------------------------------------------
> virtio,vhost,pc: fixes and updates
>
> balloon fixes wrt migration
> virtio-vsock device support
>
> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
>

Applied, thanks.

-- PMM