[Patch 0/3][v2] Userspace for MSI-X

[Patch 0/3][v2] Userspace for MSI-X

[Sheng Yang]

Change from v2:
Move MMIO intercepting to userspace, and add two new ioctls.

For kernel space MSI-X depends on lots of related patch I sent before, this
time I kept it and wait for others to be checked in.

Please help to review, thanks.

--
regards
Yang, Sheng

[PATCH 1/3] Add MSI-X related macro to pci.c

[Sheng Yang]

Signed-off-by: Sheng Yang <sheng@linux.intel.com>
---
 qemu/hw/pci.h |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/qemu/hw/pci.h b/qemu/hw/pci.h
index f2a622c..22c5de1 100644
--- a/qemu/hw/pci.h
+++ b/qemu/hw/pci.h
@@ -87,6 +87,7 @@ typedef struct PCIIORegion {
 #define PCI_CAPABILITY_CONFIG_MAX_LENGTH 0x60
 #define PCI_CAPABILITY_CONFIG_DEFAULT_START_ADDR 0x40
 #define PCI_CAPABILITY_CONFIG_MSI_LENGTH 0x10
+#define PCI_CAPABILITY_CONFIG_MSIX_LENGTH 0x10
 
 struct PCIDevice {
     /* PCI config space */
-- 
1.5.4.5

[PATCH 2/3] kvm: add ioctl KVM_SET_MSIX_ENTRY_NR and KVM_SET_MSIX_ENTRY

[Sheng Yang]

Signed-off-by: Sheng Yang <sheng@linux.intel.com>
---
 libkvm/libkvm.c |   26 ++++++++++++++++++++++++++
 libkvm/libkvm.h |    6 ++++++
 2 files changed, 32 insertions(+), 0 deletions(-)

diff --git a/libkvm/libkvm.c b/libkvm/libkvm.c
index ddcc929..ad218e1 100644
--- a/libkvm/libkvm.c
+++ b/libkvm/libkvm.c
@@ -1171,3 +1171,29 @@ int kvm_free_gsi_msg(kvm_context_t kvm, uint32_t gsi)
 }
 
 #endif
+
+#ifdef KVM_CAP_DEVICE_MSIX
+int kvm_set_msix_nr(kvm_context_t kvm,
+                    struct kvm_assigned_msix_nr *msix_nr)
+{
+        int ret;
+
+        ret = ioctl(kvm->vm_fd, KVM_SET_MSIX_NR, msix_nr);
+        if (ret < 0)
+                return -errno;
+
+        return ret;
+}
+
+int kvm_set_msix_entry(kvm_context_t kvm,
+                       struct kvm_assigned_msix_entry *entry)
+{
+        int ret;
+
+        ret = ioctl(kvm->vm_fd, KVM_SET_MSIX_ENTRY, entry);
+        if (ret < 0)
+                return -errno;
+
+        return ret;
+}
+#endif
diff --git a/libkvm/libkvm.h b/libkvm/libkvm.h
index 53e57c5..af704f7 100644
--- a/libkvm/libkvm.h
+++ b/libkvm/libkvm.h
@@ -725,4 +725,10 @@ int kvm_request_gsi_msg(kvm_context_t kvm,
 int kvm_free_gsi_msg(kvm_context_t kvm, uint32_t gsi);
 #endif
 
+#ifdef KVM_CAP_DEVICE_MSIX
+int kvm_set_msix_nr(kvm_context_t kvm,
+                    struct kvm_assigned_msix_nr *msix_nr);
+int kvm_set_msix_entry(kvm_context_t kvm,
+                       struct kvm_assigned_msix_entry *entry);
+#endif
 #endif
-- 
1.5.4.5

[PATCH 3/3] kvm: enable MSI-X capabilty for assigned device

[Sheng Yang]

The most important part here, is we emulate a page of MMIO region using a
page of memory. That's because MSI-X table was put in the region and we have to
intercept it.

Signed-off-by: Sheng Yang <sheng@linux.intel.com>
---
 qemu/hw/device-assignment.c |  275 ++++++++++++++++++++++++++++++++++++++++++-
 qemu/hw/device-assignment.h |    6 +
 2 files changed, 276 insertions(+), 5 deletions(-)

diff --git a/qemu/hw/device-assignment.c b/qemu/hw/device-assignment.c
index 2d3e67e..dc2020a 100644
--- a/qemu/hw/device-assignment.c
+++ b/qemu/hw/device-assignment.c
@@ -146,6 +146,7 @@ static void assigned_dev_iomem_map(PCIDevice *pci_dev, int region_num,
 {
     AssignedDevice *r_dev = container_of(pci_dev, AssignedDevice, dev);
     AssignedDevRegion *region = &r_dev->v_addrs[region_num];
+    PCIRegion *real_region = &r_dev->real_device.regions[region_num];
     uint32_t old_ephys = region->e_physbase;
     uint32_t old_esize = region->e_size;
     int first_map = (region->e_size == 0);
@@ -164,10 +165,27 @@ static void assigned_dev_iomem_map(PCIDevice *pci_dev, int region_num,
                                  TARGET_PAGE_ALIGN(old_esize));
     }
 
-    if (e_size > 0)
+    if (e_size > 0) {
+        /* deal with MSI-X MMIO page */
+        if (real_region->base_addr <= r_dev->msix_table_addr &&
+                real_region->base_addr + real_region->size >=
+                r_dev->msix_table_addr) {
+            int offset = r_dev->msix_table_addr - real_region->base_addr;
+            ret = munmap(region->u.r_virtbase + offset, TARGET_PAGE_SIZE);
+            if (ret == 0)
+                DEBUG("munmap done, virt_base 0x%p\n",
+                        region->u.r_virtbase + offset);
+            else {
+                fprintf(stderr, "%s: fail munmap msix table!\n", __func__);
+                exit(1);
+            }
+            cpu_register_physical_memory(e_phys + offset,
+                    TARGET_PAGE_SIZE, r_dev->mmio_index);
+        }
 	ret = kvm_register_phys_mem(kvm_context, e_phys,
                                     region->u.r_virtbase,
                                     TARGET_PAGE_ALIGN(e_size), 0);
+    }
 
     if (ret != 0) {
 	fprintf(stderr, "%s: Error: create new mapping failed\n", __func__);
@@ -570,7 +588,9 @@ void assigned_dev_update_irq(PCIDevice *d)
     }
 }
 
-#if defined(KVM_CAP_DEVICE_MSI) && defined (KVM_CAP_GSI_MSG)
+#ifdef KVM_CAP_GSI_MSG
+
+#ifdef KVM_CAP_DEVICE_MSI
 static void assigned_dev_update_msi(PCIDevice *pci_dev, unsigned int ctrl_pos)
 {
     struct kvm_assigned_irq assigned_irq_data;
@@ -610,14 +630,140 @@ static void assigned_dev_update_msi(PCIDevice *pci_dev, unsigned int ctrl_pos)
 }
 #endif
 
-void assigned_device_pci_cap_write_config(PCIDevice *pci_dev, uint32_t address,
+#ifdef KVM_CAP_DEVICE_MSIX
+static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
+{
+    AssignedDevice *adev = container_of(pci_dev, AssignedDevice, dev);
+    u16 entries_nr = 0, entries_max_nr;
+    int pos = 0, i, r = 0;
+    u32 msg_addr, msg_upper_addr, msg_data, msg_ctrl;
+    struct kvm_assigned_msix_nr msix_nr;
+    struct kvm_assigned_msix_entry msix_entry;
+    struct kvm_assigned_gsi_msg gsi_msg;
+    void *va = adev->msix_table_page;
+
+    if (adev->cap.available & ASSIGNED_DEVICE_CAP_MSI)
+        pos = PCI_CAPABILITY_CONFIG_MSI_LENGTH;
+    entries_max_nr = pci_dev->cap.config[pos + 2];
+    entries_max_nr &= PCI_MSIX_TABSIZE;
+
+    /* Get the usable entry number for allocating */
+    for (i = 0; i < entries_max_nr; i++) {
+        memcpy(&msg_ctrl, va + i * 16 + 12, 4);
+        /* 0x1 is mask bit for per vector */
+        if (msg_ctrl & 0x1)
+            continue;
+        memcpy(&msg_data, va + i * 16 + 8, 4);
+        /* Ignore unused entry even it's unmasked */
+        if (msg_data == 0)
+            continue;
+        entries_nr ++;
+    }
+
+    msix_nr.assigned_dev_id = calc_assigned_dev_id(adev->h_busnr,
+                                          (uint8_t)adev->h_devfn);
+    msix_nr.entry_nr = entries_nr;
+    r = kvm_set_msix_nr(kvm_context, &msix_nr);
+    if (r != 0) {
+        fprintf(stderr, "fail to set MSI-X entry number for MSIX! %s\n",
+			strerror(-r));
+        return r;
+    }
+
+    msix_entry.assigned_dev_id = msix_nr.assigned_dev_id;
+    entries_nr = 0;
+    for (i = 0; i < entries_max_nr; i++) {
+        if (entries_nr >= msix_nr.entry_nr)
+            break;
+        memcpy(&msg_ctrl, va + i * 16 + 12, 4);
+        if (msg_ctrl & 0x1)
+            continue;
+        memcpy(&msg_data, va + i * 16 + 8, 4);
+        if (msg_data == 0)
+            continue;
+
+        memcpy(&msg_addr, va + i * 16, 4);
+        memcpy(&msg_upper_addr, va + i * 16 + 4, 4);
+
+        gsi_msg.gsi = 0;
+        gsi_msg.msg.addr_lo = msg_addr;
+        gsi_msg.msg.addr_hi = msg_upper_addr;
+        gsi_msg.msg.data = msg_data;
+        r = kvm_request_gsi_msg(kvm_context, &gsi_msg);
+        if (r) {
+            fprintf(stderr, "fail to request gsi msg for MSIX! %s\n",
+			    strerror(-r));
+            break;
+        }
+        DEBUG("MSI-X data 0x%x, MSI-X addr_lo 0x%x\n!", msg_data, msg_addr);
+
+        msix_entry.gsi = gsi_msg.gsi;
+        msix_entry.entry = i;
+        msix_entry.pos = entries_nr;
+        r = kvm_set_msix_entry(kvm_context, &msix_entry);
+        if (r) {
+            fprintf(stderr, "fail to set MSI-X entry! %s\n", strerror(-r));
+            break;
+        }
+        DEBUG("MSI-X entry gsi 0x%x, entry %d, pos %d\n!",
+                msix_entry.gsi, msix_entry.entry, msix_entry.pos);
+        entries_nr ++;
+    }
+
+    return r;
+}
+
+static void assigned_dev_update_msix(PCIDevice *pci_dev, unsigned int ctrl_pos)
+{
+    struct kvm_assigned_irq assigned_irq_data;
+    AssignedDevice *assigned_dev = container_of(pci_dev, AssignedDevice, dev);
+    uint16_t *ctrl_word = (uint16_t *)(pci_dev->cap.config + ctrl_pos);
+
+    memset(&assigned_irq_data, 0, sizeof assigned_irq_data);
+    assigned_irq_data.assigned_dev_id  =
+            calc_assigned_dev_id(assigned_dev->h_busnr,
+                    (uint8_t)assigned_dev->h_devfn);
+
+    assigned_irq_data.flags = KVM_DEV_IRQ_ASSIGN_MSIX_ACTION;
+    if (*ctrl_word & PCI_MSIX_ENABLE) {
+        assigned_irq_data.flags |= KVM_DEV_IRQ_ASSIGN_ENABLE_MSIX;
+        if (assigned_dev_update_msix_mmio(pci_dev) < 0) {
+            perror("assigned_dev_update_msix_mmio");
+        }
+    }
+    if (*ctrl_word & PCI_MSIX_MASK)
+        assigned_irq_data.flags |= KVM_DEV_IRQ_ASSIGN_MASK_MSIX;
+
+    if (kvm_assign_irq(kvm_context, &assigned_irq_data) < 0)
+        perror("assigned_dev_enable_msi");
+
+    if (assigned_irq_data.flags & KVM_DEV_IRQ_ASSIGN_ENABLE_MSIX) {
+        assigned_dev->cap.state |= ASSIGNED_DEVICE_MSIX_ENABLED;
+        *ctrl_word |= PCI_MSIX_ENABLE;
+    } else {
+        assigned_dev->cap.state &= ~ASSIGNED_DEVICE_MSIX_ENABLED;
+        *ctrl_word &= ~PCI_MSIX_ENABLE;
+    }
+    if (assigned_irq_data.flags & KVM_DEV_IRQ_ASSIGN_MASK_MSIX) {
+        assigned_dev->cap.state |= ASSIGNED_DEVICE_MSIX_MASKED;
+        *ctrl_word |= PCI_MSIX_MASK;
+    } else {
+        assigned_dev->cap.state &= ~ASSIGNED_DEVICE_MSIX_MASKED;
+        *ctrl_word &= ~PCI_MSIX_MASK;
+    }
+}
+#endif
+#endif
+
+static void assigned_device_pci_cap_write_config(PCIDevice *pci_dev, uint32_t address,
                                           uint32_t val, int len)
 {
     AssignedDevice *assigned_dev = container_of(pci_dev, AssignedDevice, dev);
     unsigned int pos = pci_dev->cap.start, ctrl_pos;
 
     pci_default_cap_write_config(pci_dev, address, val, len);
-#if defined(KVM_CAP_DEVICE_MSI) && defined (KVM_CAP_GSI_MSG)
+#ifdef KVM_CAP_GSI_MSG
+#ifdef KVM_CAP_DEVICE_MSI
     if (assigned_dev->cap.available & ASSIGNED_DEVICE_CAP_MSI) {
         ctrl_pos = pos + PCI_MSI_FLAGS;
         if (address <= ctrl_pos && address + len > ctrl_pos)
@@ -625,6 +771,17 @@ void assigned_device_pci_cap_write_config(PCIDevice *pci_dev, uint32_t address,
         pos += PCI_CAPABILITY_CONFIG_MSI_LENGTH;
     }
 #endif
+#ifdef KVM_CAP_DEVICE_MSIX
+    if (assigned_dev->cap.available & ASSIGNED_DEVICE_CAP_MSI) {
+        ctrl_pos = pos + 3;
+        if (address <= ctrl_pos && address + len > ctrl_pos) {
+            ctrl_pos--; /* control is word long */
+            assigned_dev_update_msix(pci_dev, ctrl_pos - pci_dev->cap.start);
+	}
+        pos += PCI_CAPABILITY_CONFIG_MSIX_LENGTH;
+    }
+#endif
+#endif
     return;
 }
 
@@ -643,7 +800,9 @@ static void assigned_device_pci_cap_init(PCIDevice *pci_dev)
     pci_init(pacc);
     dev->pdev = pci_get_dev(pacc, 0, h_bus, h_dev, h_func);
     pci_cleanup(pacc);
-#if defined(KVM_CAP_DEVICE_MSI) && defined (KVM_CAP_GSI_MSG)
+    pci_fill_info(dev->pdev, PCI_FILL_BASES);
+#ifdef KVM_CAP_GSI_MSG
+#ifdef KVM_CAP_DEVICE_MSI
     /* Expose MSI capability
      * MSI capability is the 1st capability in cap.config */
     if (pci_find_cap_offset(dev->pdev, PCI_CAP_ID_MSI)) {
@@ -653,6 +812,107 @@ static void assigned_device_pci_cap_init(PCIDevice *pci_dev)
         next_cap_pt = 1;
     }
 #endif
+#ifdef KVM_CAP_DEVICE_MSIX
+    /* Expose MSI-X capability */
+    if (pci_find_cap_offset(dev->pdev, PCI_CAP_ID_MSIX)) {
+        int pos, entry_nr, bar_nr;
+        u32 msix_table_entry;
+        dev->cap.available |= ASSIGNED_DEVICE_CAP_MSIX;
+        pos = pci_find_cap_offset(dev->pdev, PCI_CAP_ID_MSIX);
+        entry_nr = pci_read_word(dev->pdev, pos + 2) & PCI_MSIX_TABSIZE;
+        pci_dev->cap.config[pci_dev->cap.length] = 0x11;
+        pci_dev->cap.config[pci_dev->cap.length + 2] = entry_nr;
+        msix_table_entry = pci_read_long(dev->pdev, pos + PCI_MSIX_TABLE);
+        *(uint32_t *)(pci_dev->cap.config +
+                      pci_dev->cap.length + PCI_MSIX_TABLE) = msix_table_entry;
+        *(uint32_t *)(pci_dev->cap.config +
+                      pci_dev->cap.length + PCI_MSIX_PBA) =
+                    pci_read_long(dev->pdev, pos + PCI_MSIX_PBA);
+        bar_nr = msix_table_entry & PCI_MSIX_BIR;
+        msix_table_entry &= ~PCI_MSIX_BIR;
+        dev->msix_table_addr = dev->pdev->base_addr[bar_nr] + msix_table_entry;
+        pci_dev->cap.config[next_cap_pt] =
+                pci_dev->cap.start + pci_dev->cap.length;
+        pci_dev->cap.length += PCI_CAPABILITY_CONFIG_MSIX_LENGTH;
+        next_cap_pt += PCI_CAPABILITY_CONFIG_MSIX_LENGTH;
+    }
+#endif
+#endif
+}
+
+static uint32_t msix_mmio_readl(void *opaque, target_phys_addr_t addr)
+{
+    AssignedDevice *adev = opaque;
+    unsigned int offset = addr & 0xfff;
+    void *page = adev->msix_table_page;
+    uint32_t val = 0;
+
+    DEBUG("read from MSI-X entry table mmio 0x%lx", addr);
+    memcpy(&val, (void *)((char *)page + offset), 4);
+
+    return val;
+}
+
+static uint32_t msix_mmio_readb(void *opaque, target_phys_addr_t addr)
+{
+    return ((msix_mmio_readl(opaque, addr & ~3)) >>
+            (8 * (addr & 3))) & 0xff;
+}
+
+static uint32_t msix_mmio_readw(void *opaque, target_phys_addr_t addr)
+{
+    return ((msix_mmio_readl(opaque, addr & ~3)) >>
+            (8 * (addr & 3))) & 0xffff;
+}
+
+static void msix_mmio_writel(void *opaque,
+                             target_phys_addr_t addr, uint32_t val)
+{
+    AssignedDevice *adev = opaque;
+    unsigned int offset = addr & 0xfff;
+    void *page = adev->msix_table_page;
+
+    DEBUG("write to MSI-X entry table mmio offset 0x%lx, val 0x%lx\n",
+		    addr, val);
+    memcpy((void *)((char *)page + offset), &val, 4);
+}
+
+static void msix_mmio_writew(void *opaque,
+                             target_phys_addr_t addr, uint32_t val)
+{
+    msix_mmio_writel(opaque, addr & ~3,
+                     (val & 0xffff) << (8*(addr & 3)));
+}
+
+static void msix_mmio_writeb(void *opaque,
+                             target_phys_addr_t addr, uint32_t val)
+{
+    msix_mmio_writel(opaque, addr & ~3,
+                     (val & 0xff) << (8*(addr & 3)));
+}
+
+static CPUWriteMemoryFunc *msix_mmio_write[] = {
+    msix_mmio_writeb,	msix_mmio_writew,	msix_mmio_writel
+};
+
+static CPUReadMemoryFunc *msix_mmio_read[] = {
+    msix_mmio_readb,	msix_mmio_readw,	msix_mmio_readl
+};
+
+static int assigned_dev_register_msix_mmio(AssignedDevice *dev)
+{
+    /* handle exception */
+    dev->msix_table_page = mmap(NULL, 0x10000,
+                                PROT_READ|PROT_WRITE,
+                                MAP_ANONYMOUS|MAP_PRIVATE, 0, 0);
+    if (dev->msix_table_page == MAP_FAILED) {
+        fprintf(stderr, "fail allocate msix_table_page! %s\n", strerror(errno));
+        return -EFAULT;
+    }
+
+    dev->mmio_index = cpu_register_io_memory(0, msix_mmio_read,
+                                    msix_mmio_write, dev);
+    return 0;
 }
 
 struct PCIDevice *init_assigned_device(AssignedDevInfo *adev, PCIBus *bus)
@@ -726,6 +986,11 @@ struct PCIDevice *init_assigned_device(AssignedDevInfo *adev, PCIBus *bus)
                     assigned_device_pci_cap_write_config,
                     assigned_device_pci_cap_init);
 
+    /* intercept MSI-X entry page in the MMIO */
+    if (dev->cap.available & ASSIGNED_DEVICE_CAP_MSIX)
+        if (assigned_dev_register_msix_mmio(dev))
+            return NULL;
+
     return &dev->dev;
 }
 
diff --git a/qemu/hw/device-assignment.h b/qemu/hw/device-assignment.h
index ea26de5..0db371a 100644
--- a/qemu/hw/device-assignment.h
+++ b/qemu/hw/device-assignment.h
@@ -84,10 +84,16 @@ typedef struct {
     struct pci_dev *pdev;
     struct {
 #define ASSIGNED_DEVICE_CAP_MSI (1 << 0)
+#define ASSIGNED_DEVICE_CAP_MSIX (1 << 1)
         uint32_t available;
 #define ASSIGNED_DEVICE_MSI_ENABLED (1 << 0)
+#define ASSIGNED_DEVICE_MSIX_ENABLED (1 << 1)
+#define ASSIGNED_DEVICE_MSIX_MASKED (1 << 2)
         uint32_t state;
     } cap;
+    void *msix_table_page;
+    target_phys_addr_t msix_table_addr;
+    int mmio_index;
 } AssignedDevice;
 
 typedef struct AssignedDevInfo AssignedDevInfo;
-- 
1.5.4.5

Re: [PATCH 3/3] kvm: enable MSI-X capabilty for assigned device

[Sheng Yang]

On Tuesday 30 December 2008 14:29:32 Sheng Yang wrote:
> The most important part here, is we emulate a page of MMIO region using a
> page of memory. That's because MSI-X table was put in the region and we
> have to intercept it.
>

Found some issues, would update this later.

-- 
regards
Yang, Sheng

> Signed-off-by: Sheng Yang <sheng@linux.intel.com>
> ---
>  qemu/hw/device-assignment.c |  275
> ++++++++++++++++++++++++++++++++++++++++++- qemu/hw/device-assignment.h |  
>  6 +
>  2 files changed, 276 insertions(+), 5 deletions(-)
>
> diff --git a/qemu/hw/device-assignment.c b/qemu/hw/device-assignment.c
> index 2d3e67e..dc2020a 100644
> --- a/qemu/hw/device-assignment.c
> +++ b/qemu/hw/device-assignment.c
> @@ -146,6 +146,7 @@ static void assigned_dev_iomem_map(PCIDevice *pci_dev,
> int region_num, {
>      AssignedDevice *r_dev = container_of(pci_dev, AssignedDevice, dev);
>      AssignedDevRegion *region = &r_dev->v_addrs[region_num];
> +    PCIRegion *real_region = &r_dev->real_device.regions[region_num];
>      uint32_t old_ephys = region->e_physbase;
>      uint32_t old_esize = region->e_size;
>      int first_map = (region->e_size == 0);
> @@ -164,10 +165,27 @@ static void assigned_dev_iomem_map(PCIDevice
> *pci_dev, int region_num, TARGET_PAGE_ALIGN(old_esize));
>      }
>
> -    if (e_size > 0)
> +    if (e_size > 0) {
> +        /* deal with MSI-X MMIO page */
> +        if (real_region->base_addr <= r_dev->msix_table_addr &&
> +                real_region->base_addr + real_region->size >=
> +                r_dev->msix_table_addr) {
> +            int offset = r_dev->msix_table_addr - real_region->base_addr;
> +            ret = munmap(region->u.r_virtbase + offset, TARGET_PAGE_SIZE);
> +            if (ret == 0)
> +                DEBUG("munmap done, virt_base 0x%p\n",
> +                        region->u.r_virtbase + offset);
> +            else {
> +                fprintf(stderr, "%s: fail munmap msix table!\n",
> __func__); +                exit(1);
> +            }
> +            cpu_register_physical_memory(e_phys + offset,
> +                    TARGET_PAGE_SIZE, r_dev->mmio_index);
> +        }
>  	ret = kvm_register_phys_mem(kvm_context, e_phys,
>                                      region->u.r_virtbase,
>                                      TARGET_PAGE_ALIGN(e_size), 0);
> +    }
>
>      if (ret != 0) {
>  	fprintf(stderr, "%s: Error: create new mapping failed\n", __func__);
> @@ -570,7 +588,9 @@ void assigned_dev_update_irq(PCIDevice *d)
>      }
>  }
>
> -#if defined(KVM_CAP_DEVICE_MSI) && defined (KVM_CAP_GSI_MSG)
> +#ifdef KVM_CAP_GSI_MSG
> +
> +#ifdef KVM_CAP_DEVICE_MSI
>  static void assigned_dev_update_msi(PCIDevice *pci_dev, unsigned int
> ctrl_pos) {
>      struct kvm_assigned_irq assigned_irq_data;
> @@ -610,14 +630,140 @@ static void assigned_dev_update_msi(PCIDevice
> *pci_dev, unsigned int ctrl_pos) }
>  #endif
>
> -void assigned_device_pci_cap_write_config(PCIDevice *pci_dev, uint32_t
> address, +#ifdef KVM_CAP_DEVICE_MSIX
> +static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
> +{
> +    AssignedDevice *adev = container_of(pci_dev, AssignedDevice, dev);
> +    u16 entries_nr = 0, entries_max_nr;
> +    int pos = 0, i, r = 0;
> +    u32 msg_addr, msg_upper_addr, msg_data, msg_ctrl;
> +    struct kvm_assigned_msix_nr msix_nr;
> +    struct kvm_assigned_msix_entry msix_entry;
> +    struct kvm_assigned_gsi_msg gsi_msg;
> +    void *va = adev->msix_table_page;
> +
> +    if (adev->cap.available & ASSIGNED_DEVICE_CAP_MSI)
> +        pos = PCI_CAPABILITY_CONFIG_MSI_LENGTH;
> +    entries_max_nr = pci_dev->cap.config[pos + 2];
> +    entries_max_nr &= PCI_MSIX_TABSIZE;
> +
> +    /* Get the usable entry number for allocating */
> +    for (i = 0; i < entries_max_nr; i++) {
> +        memcpy(&msg_ctrl, va + i * 16 + 12, 4);
> +        /* 0x1 is mask bit for per vector */
> +        if (msg_ctrl & 0x1)
> +            continue;
> +        memcpy(&msg_data, va + i * 16 + 8, 4);
> +        /* Ignore unused entry even it's unmasked */
> +        if (msg_data == 0)
> +            continue;
> +        entries_nr ++;
> +    }
> +
> +    msix_nr.assigned_dev_id = calc_assigned_dev_id(adev->h_busnr,
> +                                          (uint8_t)adev->h_devfn);
> +    msix_nr.entry_nr = entries_nr;
> +    r = kvm_set_msix_nr(kvm_context, &msix_nr);
> +    if (r != 0) {
> +        fprintf(stderr, "fail to set MSI-X entry number for MSIX! %s\n",
> +			strerror(-r));
> +        return r;
> +    }
> +
> +    msix_entry.assigned_dev_id = msix_nr.assigned_dev_id;
> +    entries_nr = 0;
> +    for (i = 0; i < entries_max_nr; i++) {
> +        if (entries_nr >= msix_nr.entry_nr)
> +            break;
> +        memcpy(&msg_ctrl, va + i * 16 + 12, 4);
> +        if (msg_ctrl & 0x1)
> +            continue;
> +        memcpy(&msg_data, va + i * 16 + 8, 4);
> +        if (msg_data == 0)
> +            continue;
> +
> +        memcpy(&msg_addr, va + i * 16, 4);
> +        memcpy(&msg_upper_addr, va + i * 16 + 4, 4);
> +
> +        gsi_msg.gsi = 0;
> +        gsi_msg.msg.addr_lo = msg_addr;
> +        gsi_msg.msg.addr_hi = msg_upper_addr;
> +        gsi_msg.msg.data = msg_data;
> +        r = kvm_request_gsi_msg(kvm_context, &gsi_msg);
> +        if (r) {
> +            fprintf(stderr, "fail to request gsi msg for MSIX! %s\n",
> +			    strerror(-r));
> +            break;
> +        }
> +        DEBUG("MSI-X data 0x%x, MSI-X addr_lo 0x%x\n!", msg_data,
> msg_addr); +
> +        msix_entry.gsi = gsi_msg.gsi;
> +        msix_entry.entry = i;
> +        msix_entry.pos = entries_nr;
> +        r = kvm_set_msix_entry(kvm_context, &msix_entry);
> +        if (r) {
> +            fprintf(stderr, "fail to set MSI-X entry! %s\n",
> strerror(-r)); +            break;
> +        }
> +        DEBUG("MSI-X entry gsi 0x%x, entry %d, pos %d\n!",
> +                msix_entry.gsi, msix_entry.entry, msix_entry.pos);
> +        entries_nr ++;
> +    }
> +
> +    return r;
> +}
> +
> +static void assigned_dev_update_msix(PCIDevice *pci_dev, unsigned int
> ctrl_pos) +{
> +    struct kvm_assigned_irq assigned_irq_data;
> +    AssignedDevice *assigned_dev = container_of(pci_dev, AssignedDevice,
> dev); +    uint16_t *ctrl_word = (uint16_t *)(pci_dev->cap.config +
> ctrl_pos); +
> +    memset(&assigned_irq_data, 0, sizeof assigned_irq_data);
> +    assigned_irq_data.assigned_dev_id  =
> +            calc_assigned_dev_id(assigned_dev->h_busnr,
> +                    (uint8_t)assigned_dev->h_devfn);
> +
> +    assigned_irq_data.flags = KVM_DEV_IRQ_ASSIGN_MSIX_ACTION;
> +    if (*ctrl_word & PCI_MSIX_ENABLE) {
> +        assigned_irq_data.flags |= KVM_DEV_IRQ_ASSIGN_ENABLE_MSIX;
> +        if (assigned_dev_update_msix_mmio(pci_dev) < 0) {
> +            perror("assigned_dev_update_msix_mmio");
> +        }
> +    }
> +    if (*ctrl_word & PCI_MSIX_MASK)
> +        assigned_irq_data.flags |= KVM_DEV_IRQ_ASSIGN_MASK_MSIX;
> +
> +    if (kvm_assign_irq(kvm_context, &assigned_irq_data) < 0)
> +        perror("assigned_dev_enable_msi");
> +
> +    if (assigned_irq_data.flags & KVM_DEV_IRQ_ASSIGN_ENABLE_MSIX) {
> +        assigned_dev->cap.state |= ASSIGNED_DEVICE_MSIX_ENABLED;
> +        *ctrl_word |= PCI_MSIX_ENABLE;
> +    } else {
> +        assigned_dev->cap.state &= ~ASSIGNED_DEVICE_MSIX_ENABLED;
> +        *ctrl_word &= ~PCI_MSIX_ENABLE;
> +    }
> +    if (assigned_irq_data.flags & KVM_DEV_IRQ_ASSIGN_MASK_MSIX) {
> +        assigned_dev->cap.state |= ASSIGNED_DEVICE_MSIX_MASKED;
> +        *ctrl_word |= PCI_MSIX_MASK;
> +    } else {
> +        assigned_dev->cap.state &= ~ASSIGNED_DEVICE_MSIX_MASKED;
> +        *ctrl_word &= ~PCI_MSIX_MASK;
> +    }
> +}
> +#endif
> +#endif
> +
> +static void assigned_device_pci_cap_write_config(PCIDevice *pci_dev,
> uint32_t address, uint32_t val, int len)
>  {
>      AssignedDevice *assigned_dev = container_of(pci_dev, AssignedDevice,
> dev); unsigned int pos = pci_dev->cap.start, ctrl_pos;
>
>      pci_default_cap_write_config(pci_dev, address, val, len);
> -#if defined(KVM_CAP_DEVICE_MSI) && defined (KVM_CAP_GSI_MSG)
> +#ifdef KVM_CAP_GSI_MSG
> +#ifdef KVM_CAP_DEVICE_MSI
>      if (assigned_dev->cap.available & ASSIGNED_DEVICE_CAP_MSI) {
>          ctrl_pos = pos + PCI_MSI_FLAGS;
>          if (address <= ctrl_pos && address + len > ctrl_pos)
> @@ -625,6 +771,17 @@ void assigned_device_pci_cap_write_config(PCIDevice
> *pci_dev, uint32_t address, pos += PCI_CAPABILITY_CONFIG_MSI_LENGTH;
>      }
>  #endif
> +#ifdef KVM_CAP_DEVICE_MSIX
> +    if (assigned_dev->cap.available & ASSIGNED_DEVICE_CAP_MSI) {
> +        ctrl_pos = pos + 3;
> +        if (address <= ctrl_pos && address + len > ctrl_pos) {
> +            ctrl_pos--; /* control is word long */
> +            assigned_dev_update_msix(pci_dev, ctrl_pos -
> pci_dev->cap.start); +	}
> +        pos += PCI_CAPABILITY_CONFIG_MSIX_LENGTH;
> +    }
> +#endif
> +#endif
>      return;
>  }
>
> @@ -643,7 +800,9 @@ static void assigned_device_pci_cap_init(PCIDevice
> *pci_dev) pci_init(pacc);
>      dev->pdev = pci_get_dev(pacc, 0, h_bus, h_dev, h_func);
>      pci_cleanup(pacc);
> -#if defined(KVM_CAP_DEVICE_MSI) && defined (KVM_CAP_GSI_MSG)
> +    pci_fill_info(dev->pdev, PCI_FILL_BASES);
> +#ifdef KVM_CAP_GSI_MSG
> +#ifdef KVM_CAP_DEVICE_MSI
>      /* Expose MSI capability
>       * MSI capability is the 1st capability in cap.config */
>      if (pci_find_cap_offset(dev->pdev, PCI_CAP_ID_MSI)) {
> @@ -653,6 +812,107 @@ static void assigned_device_pci_cap_init(PCIDevice
> *pci_dev) next_cap_pt = 1;
>      }
>  #endif
> +#ifdef KVM_CAP_DEVICE_MSIX
> +    /* Expose MSI-X capability */
> +    if (pci_find_cap_offset(dev->pdev, PCI_CAP_ID_MSIX)) {
> +        int pos, entry_nr, bar_nr;
> +        u32 msix_table_entry;
> +        dev->cap.available |= ASSIGNED_DEVICE_CAP_MSIX;
> +        pos = pci_find_cap_offset(dev->pdev, PCI_CAP_ID_MSIX);
> +        entry_nr = pci_read_word(dev->pdev, pos + 2) & PCI_MSIX_TABSIZE;
> +        pci_dev->cap.config[pci_dev->cap.length] = 0x11;
> +        pci_dev->cap.config[pci_dev->cap.length + 2] = entry_nr;
> +        msix_table_entry = pci_read_long(dev->pdev, pos + PCI_MSIX_TABLE);
> +        *(uint32_t *)(pci_dev->cap.config +
> +                      pci_dev->cap.length + PCI_MSIX_TABLE) =
> msix_table_entry; +        *(uint32_t *)(pci_dev->cap.config +
> +                      pci_dev->cap.length + PCI_MSIX_PBA) =
> +                    pci_read_long(dev->pdev, pos + PCI_MSIX_PBA);
> +        bar_nr = msix_table_entry & PCI_MSIX_BIR;
> +        msix_table_entry &= ~PCI_MSIX_BIR;
> +        dev->msix_table_addr = dev->pdev->base_addr[bar_nr] +
> msix_table_entry; +        pci_dev->cap.config[next_cap_pt] =
> +                pci_dev->cap.start + pci_dev->cap.length;
> +        pci_dev->cap.length += PCI_CAPABILITY_CONFIG_MSIX_LENGTH;
> +        next_cap_pt += PCI_CAPABILITY_CONFIG_MSIX_LENGTH;
> +    }
> +#endif
> +#endif
> +}
> +
> +static uint32_t msix_mmio_readl(void *opaque, target_phys_addr_t addr)
> +{
> +    AssignedDevice *adev = opaque;
> +    unsigned int offset = addr & 0xfff;
> +    void *page = adev->msix_table_page;
> +    uint32_t val = 0;
> +
> +    DEBUG("read from MSI-X entry table mmio 0x%lx", addr);
> +    memcpy(&val, (void *)((char *)page + offset), 4);
> +
> +    return val;
> +}
> +
> +static uint32_t msix_mmio_readb(void *opaque, target_phys_addr_t addr)
> +{
> +    return ((msix_mmio_readl(opaque, addr & ~3)) >>
> +            (8 * (addr & 3))) & 0xff;
> +}
> +
> +static uint32_t msix_mmio_readw(void *opaque, target_phys_addr_t addr)
> +{
> +    return ((msix_mmio_readl(opaque, addr & ~3)) >>
> +            (8 * (addr & 3))) & 0xffff;
> +}
> +
> +static void msix_mmio_writel(void *opaque,
> +                             target_phys_addr_t addr, uint32_t val)
> +{
> +    AssignedDevice *adev = opaque;
> +    unsigned int offset = addr & 0xfff;
> +    void *page = adev->msix_table_page;
> +
> +    DEBUG("write to MSI-X entry table mmio offset 0x%lx, val 0x%lx\n",
> +		    addr, val);
> +    memcpy((void *)((char *)page + offset), &val, 4);
> +}
> +
> +static void msix_mmio_writew(void *opaque,
> +                             target_phys_addr_t addr, uint32_t val)
> +{
> +    msix_mmio_writel(opaque, addr & ~3,
> +                     (val & 0xffff) << (8*(addr & 3)));
> +}
> +
> +static void msix_mmio_writeb(void *opaque,
> +                             target_phys_addr_t addr, uint32_t val)
> +{
> +    msix_mmio_writel(opaque, addr & ~3,
> +                     (val & 0xff) << (8*(addr & 3)));
> +}
> +
> +static CPUWriteMemoryFunc *msix_mmio_write[] = {
> +    msix_mmio_writeb,	msix_mmio_writew,	msix_mmio_writel
> +};
> +
> +static CPUReadMemoryFunc *msix_mmio_read[] = {
> +    msix_mmio_readb,	msix_mmio_readw,	msix_mmio_readl
> +};
> +
> +static int assigned_dev_register_msix_mmio(AssignedDevice *dev)
> +{
> +    /* handle exception */
> +    dev->msix_table_page = mmap(NULL, 0x10000,
> +                                PROT_READ|PROT_WRITE,
> +                                MAP_ANONYMOUS|MAP_PRIVATE, 0, 0);
> +    if (dev->msix_table_page == MAP_FAILED) {
> +        fprintf(stderr, "fail allocate msix_table_page! %s\n",
> strerror(errno)); +        return -EFAULT;
> +    }
> +
> +    dev->mmio_index = cpu_register_io_memory(0, msix_mmio_read,
> +                                    msix_mmio_write, dev);
> +    return 0;
>  }
>
>  struct PCIDevice *init_assigned_device(AssignedDevInfo *adev, PCIBus *bus)
> @@ -726,6 +986,11 @@ struct PCIDevice *init_assigned_device(AssignedDevInfo
> *adev, PCIBus *bus) assigned_device_pci_cap_write_config,
>                      assigned_device_pci_cap_init);
>
> +    /* intercept MSI-X entry page in the MMIO */
> +    if (dev->cap.available & ASSIGNED_DEVICE_CAP_MSIX)
> +        if (assigned_dev_register_msix_mmio(dev))
> +            return NULL;
> +
>      return &dev->dev;
>  }
>
> diff --git a/qemu/hw/device-assignment.h b/qemu/hw/device-assignment.h
> index ea26de5..0db371a 100644
> --- a/qemu/hw/device-assignment.h
> +++ b/qemu/hw/device-assignment.h
> @@ -84,10 +84,16 @@ typedef struct {
>      struct pci_dev *pdev;
>      struct {
>  #define ASSIGNED_DEVICE_CAP_MSI (1 << 0)
> +#define ASSIGNED_DEVICE_CAP_MSIX (1 << 1)
>          uint32_t available;
>  #define ASSIGNED_DEVICE_MSI_ENABLED (1 << 0)
> +#define ASSIGNED_DEVICE_MSIX_ENABLED (1 << 1)
> +#define ASSIGNED_DEVICE_MSIX_MASKED (1 << 2)
>          uint32_t state;
>      } cap;
> +    void *msix_table_page;
> +    target_phys_addr_t msix_table_addr;
> +    int mmio_index;
>  } AssignedDevice;
>
>  typedef struct AssignedDevInfo AssignedDevInfo;