MSI-X mask supporting for assigned device(QEmu)

PATCH In review Sheng Yang 2 participants 20 messages Nov 4, 2010 – Nov 11, 2010 KVM

Share & tools

Sheng Yang

author

2010-11-04 06:18 UTC #0 lore raw

cover MSI-X mask supporting for assigned device(QEmu)

Sheng Yang (3):
qemu-kvm: Ioctl for in-kernel mask support
qemu-kvm: device assignment: Some clean up about MSI-X code
qemu-kvm: device assignment: emulate MSI-X mask bits

 hw/device-assignment.c |  236 ++++++++++++++++++++++++++++++++++++++++++------
 qemu-kvm.c             |   15 +++
 qemu-kvm.h             |    6 ++
 3 files changed, 227 insertions(+), 30 deletions(-)

Sheng Yang

author

2010-11-04 06:18 UTC #1 lore raw

PATCH 1/3 qemu-kvm: Ioctl for in-kernel mask support

Signed-off-by Sheng Yang <[email protected]>

---
 qemu-kvm.c |   15 +++++++++++++++
 qemu-kvm.h |    6 ++++++
 2 files changed, 21 insertions(+), 0 deletions(-)

qemu-kvm.c +15 -0

index 733d0a9..ba6db51 100644
@@ -1092,6 +1092,21 @@ int kvm_assign_set_msix_entry(kvm_context_t kvm,
 {
     return kvm_vm_ioctl(kvm_state, KVM_ASSIGN_SET_MSIX_ENTRY, entry);
 }
+
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+int kvm_assign_get_msix_entry(kvm_context_t kvm,
+                              struct kvm_assigned_msix_entry *entry)
+{
+    return kvm_vm_ioctl(kvm_state, KVM_ASSIGN_GET_MSIX_ENTRY, entry);
+}
+
+int kvm_assign_reg_msix_mmio(kvm_context_t kvm,
+                           struct kvm_assigned_msix_mmio *msix_mmio)
+{
+    return kvm_vm_ioctl(kvm_state, KVM_ASSIGN_REG_MSIX_MMIO, msix_mmio);
+}
+#endif
+
 #endif
 
 #if defined(KVM_CAP_IRQFD) && defined(CONFIG_EVENTFD)

qemu-kvm.h +6 -0

index 9c08ab4..1afdd42 100644
@@ -743,6 +743,12 @@ int kvm_assign_set_msix_nr(kvm_context_t kvm,
                            struct kvm_assigned_msix_nr *msix_nr);
 int kvm_assign_set_msix_entry(kvm_context_t kvm,
                               struct kvm_assigned_msix_entry *entry);
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+int kvm_assign_get_msix_entry(kvm_context_t kvm,
+                              struct kvm_assigned_msix_entry *entry);
+int kvm_assign_reg_msix_mmio(kvm_context_t kvm,
+                           struct kvm_assigned_msix_mmio *msix_mmio);
+#endif
 #endif
 
 #else                           /* !CONFIG_KVM */

Signature · 1 line

1.7.0.1

Michael S. Tsirkin

2010-11-04 09:48 UTC #2 lore raw

Re: [PATCH 1/3] qemu-kvm: Ioctl for in-kernel mask support

Quoted 52 lines · Sheng Yang

Signed-off-by Sheng Yang <[email protected]>

---
 qemu-kvm.c |   15 +++++++++++++++
 qemu-kvm.h |    6 ++++++
 2 files changed, 21 insertions(+), 0 deletions(-)

qemu-kvm.c +15 -0

index 733d0a9..ba6db51 100644
@@ -1092,6 +1092,21 @@ int kvm_assign_set_msix_entry(kvm_context_t kvm,
 {
     return kvm_vm_ioctl(kvm_state, KVM_ASSIGN_SET_MSIX_ENTRY, entry);
 }
+
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+int kvm_assign_get_msix_entry(kvm_context_t kvm,
+                              struct kvm_assigned_msix_entry *entry)
+{
+    return kvm_vm_ioctl(kvm_state, KVM_ASSIGN_GET_MSIX_ENTRY, entry);
+}
+
+int kvm_assign_reg_msix_mmio(kvm_context_t kvm,
+                           struct kvm_assigned_msix_mmio *msix_mmio)
+{
+    return kvm_vm_ioctl(kvm_state, KVM_ASSIGN_REG_MSIX_MMIO, msix_mmio);
+}
+#endif
+
 #endif
 
 #if defined(KVM_CAP_IRQFD) && defined(CONFIG_EVENTFD)

qemu-kvm.h +6 -0

index 9c08ab4..1afdd42 100644
@@ -743,6 +743,12 @@ int kvm_assign_set_msix_nr(kvm_context_t kvm,
                            struct kvm_assigned_msix_nr *msix_nr);
 int kvm_assign_set_msix_entry(kvm_context_t kvm,
                               struct kvm_assigned_msix_entry *entry);
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+int kvm_assign_get_msix_entry(kvm_context_t kvm,
+                              struct kvm_assigned_msix_entry *entry);
+int kvm_assign_reg_msix_mmio(kvm_context_t kvm,
+                           struct kvm_assigned_msix_mmio *msix_mmio);
+#endif
 #endif
 
 #else                           /* !CONFIG_KVM */

Signature · 1 line

1.7.0.1

We are trying to move away from using ifdefs. Stub these out instead?

Signature · 1 line

MST

Sheng Yang

author

2010-11-05 02:59 UTC #3 lore raw

Re: [PATCH 1/3] qemu-kvm: Ioctl for in-kernel mask support

Quoted 59 lines · On Thursday 04 November 2010 17:47:58 Michael S. Tsirkin wrote:

Quoted 58 lines · Sheng Yang

Signed-off-by Sheng Yang <[email protected]>

---

 qemu-kvm.c |   15 +++++++++++++++
 qemu-kvm.h |    6 ++++++
 2 files changed, 21 insertions(+), 0 deletions(-)

qemu-kvm.c +15 -0

index 733d0a9..ba6db51 100644
@@ -1092,6 +1092,21 @@ int kvm_assign_set_msix_entry(kvm_context_t kvm,
 
 {
 
     return kvm_vm_ioctl(kvm_state, KVM_ASSIGN_SET_MSIX_ENTRY, entry);
 
 }
 
+
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+int kvm_assign_get_msix_entry(kvm_context_t kvm,
+                              struct kvm_assigned_msix_entry *entry)
+{
+    return kvm_vm_ioctl(kvm_state, KVM_ASSIGN_GET_MSIX_ENTRY, entry);
+}
+
+int kvm_assign_reg_msix_mmio(kvm_context_t kvm,
+                           struct kvm_assigned_msix_mmio *msix_mmio)
+{
+    return kvm_vm_ioctl(kvm_state, KVM_ASSIGN_REG_MSIX_MMIO, msix_mmio);
+}
+#endif
+
 
 #endif
 
 #if defined(KVM_CAP_IRQFD) && defined(CONFIG_EVENTFD)

qemu-kvm.h +6 -0

index 9c08ab4..1afdd42 100644
@@ -743,6 +743,12 @@ int kvm_assign_set_msix_nr(kvm_context_t kvm,
 
                            struct kvm_assigned_msix_nr *msix_nr);
 
 int kvm_assign_set_msix_entry(kvm_context_t kvm,
 
                               struct kvm_assigned_msix_entry *entry);
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+int kvm_assign_get_msix_entry(kvm_context_t kvm,
+                              struct kvm_assigned_msix_entry *entry);
+int kvm_assign_reg_msix_mmio(kvm_context_t kvm,
+                           struct kvm_assigned_msix_mmio *msix_mmio);
+#endif

Quoted 5 lines

Quoted 3 lines
#endif
#else /* !CONFIG_KVM */
We are trying to move away from using ifdefs. Stub these out instead?

Example?

--
regards
Yang, Sheng

Michael S. Tsirkin

2010-11-05 08:52 UTC #4 lore raw

Re: [PATCH 1/3] qemu-kvm: Ioctl for in-kernel mask support

Quoted 3 lines · Sheng Yang

Quoted 1 line
We are trying to move away from using ifdefs. Stub these out instead?
Example?

37859f054986fab6b4094cd5950266ae56a6ca6a and 889e30cc18e21f2091b77267dca8096d7dd34f8b that depends on it.

Quoted 3 lines

--
regards
Yang, Sheng

Sheng Yang

author

2010-11-11 06:20 UTC #5 lore raw

Re: [PATCH 1/3] qemu-kvm: Ioctl for in-kernel mask support

Quoted 7 lines · On Friday 05 November 2010 16:52:47 Michael S. Tsirkin wrote:

Quoted 3 lines · Sheng Yang
Quoted 1 line
We are trying to move away from using ifdefs. Stub these out instead?
Example?
37859f054986fab6b4094cd5950266ae56a6ca6a and 889e30cc18e21f2091b77267dca8096d7dd34f8b that depends on it.

Gave it a try, but the parameter of the function depends on certain version of KVM header file which still require #ifdef...

I think we can leave it a bit later.

--
regards
Yang, Sheng

Quoted 4 lines

Quoted 4 lines
--
regards
Yang, Sheng

Sheng Yang

author

2010-11-04 06:18 UTC #6 lore raw

PATCH 2/3 qemu-kvm: device assignment: Some clean up about MSI-X code

In reply to Sheng Yang

Signed-off-by Sheng Yang <[email protected]>

---
 hw/device-assignment.c |   77 ++++++++++++++++++++++++++++++++---------------
 1 files changed, 52 insertions(+), 25 deletions(-)

hw/device-assignment.c +52 ~~-25~~

index 2605bd1..8a98876 100644
@@ -1106,15 +1106,10 @@ static void assigned_dev_update_msi(PCIDevice *pci_dev, unsigned int ctrl_pos)
 #endif
 
 #ifdef KVM_CAP_DEVICE_MSIX
-static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
+static int get_msix_entries_max_nr(AssignedDevice *adev)
 {
-    AssignedDevice *adev = container_of(pci_dev, AssignedDevice, dev);
-    uint16_t entries_nr = 0, entries_max_nr;
-    int pos = 0, i, r = 0;
-    uint32_t msg_addr, msg_upper_addr, msg_data, msg_ctrl;
-    struct kvm_assigned_msix_nr msix_nr;
-    struct kvm_assigned_msix_entry msix_entry;
-    void *va = adev->msix_table_page;
+    int pos, entries_max_nr;
+    PCIDevice *pci_dev = &adev->dev;
 
     if (adev->cap.available & ASSIGNED_DEVICE_CAP_MSI)
         pos = pci_dev->cap.start + PCI_CAPABILITY_CONFIG_MSI_LENGTH;

@@ -1125,6 +1120,17 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
     entries_max_nr &= PCI_MSIX_TABSIZE;
     entries_max_nr += 1;
 
+    return entries_max_nr;
+}
+
+static int get_msix_valid_entries_nr(AssignedDevice *adev,
+				     uint16_t entries_max_nr)
+{
+    void *va = adev->msix_table_page;
+    uint32_t msg_data, msg_ctrl;
+    uint16_t entries_nr = 0;
+    int i;
+
     /* Get the usable entry number for allocating */
     for (i = 0; i < entries_max_nr; i++) {
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);

@@ -1134,11 +1140,20 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
             continue;
         entries_nr ++;
     }
+    return entries_nr;
+}
+
+static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
+                                         uint16_t entries_nr,
+                                         uint16_t entries_max_nr)
+{
+    AssignedDevice *adev = container_of(pci_dev, AssignedDevice, dev);
+    int i, r = 0;
+    uint32_t msg_addr, msg_upper_addr, msg_data, msg_ctrl;
+    struct kvm_assigned_msix_nr msix_nr;
+    struct kvm_assigned_msix_entry msix_entry;
+    void *va = adev->msix_table_page;
 
-    if (entries_nr == 0) {
-        fprintf(stderr, "MSI-X entry number is zero!\n");
-        return -EINVAL;
-    }
     msix_nr.assigned_dev_id = calc_assigned_dev_id(adev->h_segnr, adev->h_busnr,
                                           (uint8_t)adev->h_devfn);
     msix_nr.entry_nr = entries_nr;

@@ -1180,8 +1195,8 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
         adev->entry[entries_nr].u.msi.address_lo = msg_addr;
         adev->entry[entries_nr].u.msi.address_hi = msg_upper_addr;
         adev->entry[entries_nr].u.msi.data = msg_data;
-        DEBUG("MSI-X data 0x%x, MSI-X addr_lo 0x%x\n!", msg_data, msg_addr);
-	kvm_add_routing_entry(&adev->entry[entries_nr]);
+        DEBUG("MSI-X data 0x%x, MSI-X addr_lo 0x%x!\n", msg_data, msg_addr);
+        kvm_add_routing_entry(&adev->entry[entries_nr]);
 
         msix_entry.gsi = adev->entry[entries_nr].gsi;
         msix_entry.entry = i;

@@ -1190,7 +1205,7 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
             fprintf(stderr, "fail to set MSI-X entry! %s\n", strerror(-r));
             break;
         }
-        DEBUG("MSI-X entry gsi 0x%x, entry %d\n!",
+        DEBUG("MSI-X entry gsi 0x%x, entry %d!\n",
                 msix_entry.gsi, msix_entry.entry);
         entries_nr ++;
     }

@@ -1203,12 +1218,12 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
     return r;
 }
 
-static void assigned_dev_update_msix(PCIDevice *pci_dev, unsigned int ctrl_pos)
+static void assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix)
 {
     struct kvm_assigned_irq assigned_irq_data;
     AssignedDevice *assigned_dev = container_of(pci_dev, AssignedDevice, dev);
-    uint16_t *ctrl_word = (uint16_t *)(pci_dev->config + ctrl_pos);
     int r;
+    uint16_t entries_nr, entries_max_nr;
 
     memset(&assigned_irq_data, 0, sizeof assigned_irq_data);
     assigned_irq_data.assigned_dev_id  =

@@ -1219,8 +1234,7 @@ static void assigned_dev_update_msix(PCIDevice *pci_dev, unsigned int ctrl_pos)
      * try to catch this by only deassigning irqs if the guest is using
      * MSIX or intends to start. */
     if ((assigned_dev->irq_requested_type & KVM_DEV_IRQ_GUEST_MSIX) ||
-        (*ctrl_word & PCI_MSIX_ENABLE)) {
-
+            enable_msix) {
         assigned_irq_data.flags = assigned_dev->irq_requested_type;
         free_dev_irq_entries(assigned_dev);
         r = kvm_deassign_irq(kvm_context, &assigned_irq_data);

@@ -1231,14 +1245,25 @@ static void assigned_dev_update_msix(PCIDevice *pci_dev, unsigned int ctrl_pos)
         assigned_dev->irq_requested_type = 0;
     }
 
-    if (*ctrl_word & PCI_MSIX_ENABLE) {
-        assigned_irq_data.flags = KVM_DEV_IRQ_HOST_MSIX |
-                                  KVM_DEV_IRQ_GUEST_MSIX;
-
-        if (assigned_dev_update_msix_mmio(pci_dev) < 0) {
+    entries_max_nr = get_msix_entries_max_nr(assigned_dev);
+    if (entries_max_nr == 0) {
+        fprintf(stderr, "assigned_dev_update_msix: MSI-X entries_max_nr == 0");
+        return;
+    }
+    entries_nr = get_msix_valid_entries_nr(assigned_dev, entries_max_nr);
+    if (entries_nr == 0) {
+        if (enable_msix)
+            fprintf(stderr, "MSI-X entry number is zero!\n");
+        return;
+    }
+    if (enable_msix) {
+        if (assigned_dev_update_msix_mmio(pci_dev,
+                    entries_nr, entries_max_nr) < 0) {
             perror("assigned_dev_update_msix_mmio");
             return;
         }
+        assigned_irq_data.flags = KVM_DEV_IRQ_HOST_MSIX |
+                                  KVM_DEV_IRQ_GUEST_MSIX;
         if (kvm_assign_irq(kvm_context, &assigned_irq_data) < 0) {
             perror("assigned_dev_enable_msix: assign irq");
             return;

@@ -1254,6 +1279,7 @@ static void assigned_device_pci_cap_write_config(PCIDevice *pci_dev, uint32_t ad
 {
     AssignedDevice *assigned_dev = container_of(pci_dev, AssignedDevice, dev);
     unsigned int pos = pci_dev->cap.start, ctrl_pos;
+    uint16_t *ctrl_word;
 
     pci_default_cap_write_config(pci_dev, address, val, len);
 #ifdef KVM_CAP_IRQ_ROUTING

@@ -1270,7 +1296,8 @@ static void assigned_device_pci_cap_write_config(PCIDevice *pci_dev, uint32_t ad
         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);
+            ctrl_word = (uint16_t *)(pci_dev->config + ctrl_pos);
+            assigned_dev_update_msix(pci_dev, (*ctrl_word & PCI_MSIX_ENABLE));
 	}
         pos += PCI_CAPABILITY_CONFIG_MSIX_LENGTH;
     }

Signature · 1 line

1.7.0.1

Michael S. Tsirkin

2010-11-04 09:47 UTC #7 lore raw

Re: [PATCH 2/3] qemu-kvm: device assignment: Some clean up about MSI-X code

Quoted 2 lines · Sheng Yang

Signed-off-by Sheng Yang <[email protected]>

If you have the energy for cleanups, how about
removing most of msi/msix code in device-assignment and using msi.c and msix.c instead?

Quoted 166 lines

---
 hw/device-assignment.c |   77 ++++++++++++++++++++++++++++++++---------------
 1 files changed, 52 insertions(+), 25 deletions(-)

hw/device-assignment.c +52 ~~-25~~

index 2605bd1..8a98876 100644
@@ -1106,15 +1106,10 @@ static void assigned_dev_update_msi(PCIDevice *pci_dev, unsigned int ctrl_pos)
 #endif
 
 #ifdef KVM_CAP_DEVICE_MSIX
-static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
+static int get_msix_entries_max_nr(AssignedDevice *adev)
 {
-    AssignedDevice *adev = container_of(pci_dev, AssignedDevice, dev);
-    uint16_t entries_nr = 0, entries_max_nr;
-    int pos = 0, i, r = 0;
-    uint32_t msg_addr, msg_upper_addr, msg_data, msg_ctrl;
-    struct kvm_assigned_msix_nr msix_nr;
-    struct kvm_assigned_msix_entry msix_entry;
-    void *va = adev->msix_table_page;
+    int pos, entries_max_nr;
+    PCIDevice *pci_dev = &adev->dev;
 
     if (adev->cap.available & ASSIGNED_DEVICE_CAP_MSI)
         pos = pci_dev->cap.start + PCI_CAPABILITY_CONFIG_MSI_LENGTH;

@@ -1125,6 +1120,17 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
     entries_max_nr &= PCI_MSIX_TABSIZE;
     entries_max_nr += 1;
 
+    return entries_max_nr;
+}
+
+static int get_msix_valid_entries_nr(AssignedDevice *adev,
+				     uint16_t entries_max_nr)
+{
+    void *va = adev->msix_table_page;
+    uint32_t msg_data, msg_ctrl;
+    uint16_t entries_nr = 0;
+    int i;
+
     /* Get the usable entry number for allocating */
     for (i = 0; i < entries_max_nr; i++) {
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);

@@ -1134,11 +1140,20 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
             continue;
         entries_nr ++;
     }
+    return entries_nr;
+}
+
+static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
+                                         uint16_t entries_nr,
+                                         uint16_t entries_max_nr)
+{
+    AssignedDevice *adev = container_of(pci_dev, AssignedDevice, dev);
+    int i, r = 0;
+    uint32_t msg_addr, msg_upper_addr, msg_data, msg_ctrl;
+    struct kvm_assigned_msix_nr msix_nr;
+    struct kvm_assigned_msix_entry msix_entry;
+    void *va = adev->msix_table_page;
 
-    if (entries_nr == 0) {
-        fprintf(stderr, "MSI-X entry number is zero!\n");
-        return -EINVAL;
-    }
     msix_nr.assigned_dev_id = calc_assigned_dev_id(adev->h_segnr, adev->h_busnr,
                                           (uint8_t)adev->h_devfn);
     msix_nr.entry_nr = entries_nr;

@@ -1180,8 +1195,8 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
         adev->entry[entries_nr].u.msi.address_lo = msg_addr;
         adev->entry[entries_nr].u.msi.address_hi = msg_upper_addr;
         adev->entry[entries_nr].u.msi.data = msg_data;
-        DEBUG("MSI-X data 0x%x, MSI-X addr_lo 0x%x\n!", msg_data, msg_addr);
-	kvm_add_routing_entry(&adev->entry[entries_nr]);
+        DEBUG("MSI-X data 0x%x, MSI-X addr_lo 0x%x!\n", msg_data, msg_addr);
+        kvm_add_routing_entry(&adev->entry[entries_nr]);
 
         msix_entry.gsi = adev->entry[entries_nr].gsi;
         msix_entry.entry = i;

@@ -1190,7 +1205,7 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
             fprintf(stderr, "fail to set MSI-X entry! %s\n", strerror(-r));
             break;
         }
-        DEBUG("MSI-X entry gsi 0x%x, entry %d\n!",
+        DEBUG("MSI-X entry gsi 0x%x, entry %d!\n",
                 msix_entry.gsi, msix_entry.entry);
         entries_nr ++;
     }

@@ -1203,12 +1218,12 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
     return r;
 }
 
-static void assigned_dev_update_msix(PCIDevice *pci_dev, unsigned int ctrl_pos)
+static void assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix)
 {
     struct kvm_assigned_irq assigned_irq_data;
     AssignedDevice *assigned_dev = container_of(pci_dev, AssignedDevice, dev);
-    uint16_t *ctrl_word = (uint16_t *)(pci_dev->config + ctrl_pos);
     int r;
+    uint16_t entries_nr, entries_max_nr;
 
     memset(&assigned_irq_data, 0, sizeof assigned_irq_data);
     assigned_irq_data.assigned_dev_id  =

@@ -1219,8 +1234,7 @@ static void assigned_dev_update_msix(PCIDevice *pci_dev, unsigned int ctrl_pos)
      * try to catch this by only deassigning irqs if the guest is using
      * MSIX or intends to start. */
     if ((assigned_dev->irq_requested_type & KVM_DEV_IRQ_GUEST_MSIX) ||
-        (*ctrl_word & PCI_MSIX_ENABLE)) {
-
+            enable_msix) {
         assigned_irq_data.flags = assigned_dev->irq_requested_type;
         free_dev_irq_entries(assigned_dev);
         r = kvm_deassign_irq(kvm_context, &assigned_irq_data);

@@ -1231,14 +1245,25 @@ static void assigned_dev_update_msix(PCIDevice *pci_dev, unsigned int ctrl_pos)
         assigned_dev->irq_requested_type = 0;
     }
 
-    if (*ctrl_word & PCI_MSIX_ENABLE) {
-        assigned_irq_data.flags = KVM_DEV_IRQ_HOST_MSIX |
-                                  KVM_DEV_IRQ_GUEST_MSIX;
-
-        if (assigned_dev_update_msix_mmio(pci_dev) < 0) {
+    entries_max_nr = get_msix_entries_max_nr(assigned_dev);
+    if (entries_max_nr == 0) {
+        fprintf(stderr, "assigned_dev_update_msix: MSI-X entries_max_nr == 0");
+        return;
+    }
+    entries_nr = get_msix_valid_entries_nr(assigned_dev, entries_max_nr);
+    if (entries_nr == 0) {
+        if (enable_msix)
+            fprintf(stderr, "MSI-X entry number is zero!\n");
+        return;
+    }
+    if (enable_msix) {
+        if (assigned_dev_update_msix_mmio(pci_dev,
+                    entries_nr, entries_max_nr) < 0) {
             perror("assigned_dev_update_msix_mmio");
             return;
         }
+        assigned_irq_data.flags = KVM_DEV_IRQ_HOST_MSIX |
+                                  KVM_DEV_IRQ_GUEST_MSIX;
         if (kvm_assign_irq(kvm_context, &assigned_irq_data) < 0) {
             perror("assigned_dev_enable_msix: assign irq");
             return;

@@ -1254,6 +1279,7 @@ static void assigned_device_pci_cap_write_config(PCIDevice *pci_dev, uint32_t ad
 {
     AssignedDevice *assigned_dev = container_of(pci_dev, AssignedDevice, dev);
     unsigned int pos = pci_dev->cap.start, ctrl_pos;
+    uint16_t *ctrl_word;
 
     pci_default_cap_write_config(pci_dev, address, val, len);
 #ifdef KVM_CAP_IRQ_ROUTING

@@ -1270,7 +1296,8 @@ static void assigned_device_pci_cap_write_config(PCIDevice *pci_dev, uint32_t ad
         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);
+            ctrl_word = (uint16_t *)(pci_dev->config + ctrl_pos);
+            assigned_dev_update_msix(pci_dev, (*ctrl_word & PCI_MSIX_ENABLE));
 	}
         pos += PCI_CAPABILITY_CONFIG_MSIX_LENGTH;
     }

Signature · 1 line

1.7.0.1

Sheng Yang

author

2010-11-05 03:08 UTC #8 lore raw

Re: [PATCH 2/3] qemu-kvm: device assignment: Some clean up about MSI-X code

Quoted 6 lines · On Thursday 04 November 2010 17:47:16 Michael S. Tsirkin wrote:

Quoted 1 line · Sheng Yang
Signed-off-by Sheng Yang <[email protected]>
If you have the energy for cleanups, how about
removing most of msi/msix code in device-assignment and using msi.c and msix.c instead?

Ah, this one is only used to reduce the review difficulty. I won't mind if you want to merge it with the last one...

Use msi.c/msix.c totally another big clean up work to do. Let put it later...

--
regards
Yang, Sheng

Quoted 241 lines

---

 hw/device-assignment.c |   77

++++++++++++++++++++++++++++++++--------------- 1 files changed, 52 insertions(+), 25 deletions(-)

hw/device-assignment.c +14 -8

index 2605bd1..8a98876 100644
@@ -1106,15 +1106,10 @@ static void assigned_dev_update_msi(PCIDevice
 *pci_dev, unsigned int ctrl_pos)
 
 #endif
 
 #ifdef KVM_CAP_DEVICE_MSIX
 
-static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
+static int get_msix_entries_max_nr(AssignedDevice *adev)
 
 {
 
-    AssignedDevice *adev = container_of(pci_dev, AssignedDevice, dev);
-    uint16_t entries_nr = 0, entries_max_nr;
-    int pos = 0, i, r = 0;
-    uint32_t msg_addr, msg_upper_addr, msg_data, msg_ctrl;
-    struct kvm_assigned_msix_nr msix_nr;
-    struct kvm_assigned_msix_entry msix_entry;
-    void *va = adev->msix_table_page;
+    int pos, entries_max_nr;
+    PCIDevice *pci_dev = &adev->dev;
 
     if (adev->cap.available & ASSIGNED_DEVICE_CAP_MSI)
 
         pos = pci_dev->cap.start + PCI_CAPABILITY_CONFIG_MSI_LENGTH;

@@ -1125,6 +1120,17 @@ static int assigned_dev_update_msix_mmio(PCIDevice
 *pci_dev)
 
     entries_max_nr &= PCI_MSIX_TABSIZE;
     entries_max_nr += 1;
 
+    return entries_max_nr;
+}
+
+static int get_msix_valid_entries_nr(AssignedDevice *adev,
+				     uint16_t entries_max_nr)
+{
+    void *va = adev->msix_table_page;
+    uint32_t msg_data, msg_ctrl;
+    uint16_t entries_nr = 0;
+    int i;
+

     /* Get the usable entry number for allocating */
     for (i = 0; i < entries_max_nr; i++) {
     
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);

hw/device-assignment.c +14 -5

@@ -1134,11 +1140,20 @@ static int
 assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
 
             continue;
 
         entries_nr ++;
 
     }
 
+    return entries_nr;
+}
+
+static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
+                                         uint16_t entries_nr,
+                                         uint16_t entries_max_nr)
+{
+    AssignedDevice *adev = container_of(pci_dev, AssignedDevice, dev);
+    int i, r = 0;
+    uint32_t msg_addr, msg_upper_addr, msg_data, msg_ctrl;
+    struct kvm_assigned_msix_nr msix_nr;
+    struct kvm_assigned_msix_entry msix_entry;
+    void *va = adev->msix_table_page;
 
-    if (entries_nr == 0) {
-        fprintf(stderr, "MSI-X entry number is zero!\n");
-        return -EINVAL;
-    }
 
     msix_nr.assigned_dev_id = calc_assigned_dev_id(adev->h_segnr,
     adev->h_busnr,
 
                                           (uint8_t)adev->h_devfn);
 
     msix_nr.entry_nr = entries_nr;

@@ -1180,8 +1195,8 @@ static int assigned_dev_update_msix_mmio(PCIDevice
 *pci_dev)
 
         adev->entry[entries_nr].u.msi.address_lo = msg_addr;
         adev->entry[entries_nr].u.msi.address_hi = msg_upper_addr;
         adev->entry[entries_nr].u.msi.data = msg_data;
 
-        DEBUG("MSI-X data 0x%x, MSI-X addr_lo 0x%x\n!", msg_data,
 msg_addr); -	kvm_add_routing_entry(&adev->entry[entries_nr]);
+        DEBUG("MSI-X data 0x%x, MSI-X addr_lo 0x%x!\n", msg_data,

msg_addr); + kvm_add_routing_entry(&adev->entry[entries_nr]);

         msix_entry.gsi = adev->entry[entries_nr].gsi;
         msix_entry.entry = i;

hw/device-assignment.c +1 -2

@@ -1190,7 +1205,7 @@ static int assigned_dev_update_msix_mmio(PCIDevice
 *pci_dev)
 
             fprintf(stderr, "fail to set MSI-X entry! %s\n",
             strerror(-r)); break;
 
         }
 
-        DEBUG("MSI-X entry gsi 0x%x, entry %d\n!",
+        DEBUG("MSI-X entry gsi 0x%x, entry %d!\n",
 
                 msix_entry.gsi, msix_entry.entry);
 
         entries_nr ++;
 
     }

@@ -1203,12 +1218,12 @@ static int
 assigned_dev_update_msix_mmio(PCIDevice *pci_dev)
 
     return r;
 
 }
 
-static void assigned_dev_update_msix(PCIDevice *pci_dev, unsigned int

ctrl_pos) +static void assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix)

{

     struct kvm_assigned_irq assigned_irq_data;
     AssignedDevice *assigned_dev = container_of(pci_dev, AssignedDevice,
     dev);

hw/device-assignment.c +16 -8

-    uint16_t *ctrl_word = (uint16_t *)(pci_dev->config + ctrl_pos);
 
     int r;
 
+    uint16_t entries_nr, entries_max_nr;
 
     memset(&assigned_irq_data, 0, sizeof assigned_irq_data);
     assigned_irq_data.assigned_dev_id  =

@@ -1219,8 +1234,7 @@ static void assigned_dev_update_msix(PCIDevice
 *pci_dev, unsigned int ctrl_pos)
 
      * try to catch this by only deassigning irqs if the guest is using
      * MSIX or intends to start. */
 
     if ((assigned_dev->irq_requested_type & KVM_DEV_IRQ_GUEST_MSIX) ||
 
-        (*ctrl_word & PCI_MSIX_ENABLE)) {
-
+            enable_msix) {
 
         assigned_irq_data.flags = assigned_dev->irq_requested_type;
         free_dev_irq_entries(assigned_dev);
         r = kvm_deassign_irq(kvm_context, &assigned_irq_data);

@@ -1231,14 +1245,25 @@ static void assigned_dev_update_msix(PCIDevice
 *pci_dev, unsigned int ctrl_pos)
 
         assigned_dev->irq_requested_type = 0;
 
     }
 
-    if (*ctrl_word & PCI_MSIX_ENABLE) {
-        assigned_irq_data.flags = KVM_DEV_IRQ_HOST_MSIX |
-                                  KVM_DEV_IRQ_GUEST_MSIX;
-
-        if (assigned_dev_update_msix_mmio(pci_dev) < 0) {
+    entries_max_nr = get_msix_entries_max_nr(assigned_dev);
+    if (entries_max_nr == 0) {
+        fprintf(stderr, "assigned_dev_update_msix: MSI-X entries_max_nr
 == 0"); +        return;
+    }
+    entries_nr = get_msix_valid_entries_nr(assigned_dev,
 entries_max_nr); +    if (entries_nr == 0) {
+        if (enable_msix)
+            fprintf(stderr, "MSI-X entry number is zero!\n");
+        return;
+    }
+    if (enable_msix) {
+        if (assigned_dev_update_msix_mmio(pci_dev,
+                    entries_nr, entries_max_nr) < 0) {
 
             perror("assigned_dev_update_msix_mmio");
             return;
 
         }
 
+        assigned_irq_data.flags = KVM_DEV_IRQ_HOST_MSIX |
+                                  KVM_DEV_IRQ_GUEST_MSIX;
 
         if (kvm_assign_irq(kvm_context, &assigned_irq_data) < 0) {
 
             perror("assigned_dev_enable_msix: assign irq");
             return;

@@ -1254,6 +1279,7 @@ static void

assigned_device_pci_cap_write_config(PCIDevice *pci_dev, uint32_t ad

{

AssignedDevice *assigned_dev = container_of(pci_dev, AssignedDevice, dev); unsigned int pos = pci_dev->cap.start, ctrl_pos;

hw/device-assignment.c +1 -0

+    uint16_t *ctrl_word;
 
     pci_default_cap_write_config(pci_dev, address, val, len);
 
 #ifdef KVM_CAP_IRQ_ROUTING

@@ -1270,7 +1296,8 @@ static void

assigned_device_pci_cap_write_config(PCIDevice *pci_dev, uint32_t ad

         ctrl_pos = pos + 3;
         if (address <= ctrl_pos && address + len > ctrl_pos) {
         
             ctrl_pos--; /* control is word long */

hw/device-assignment.c +2 -1

-            assigned_dev_update_msix(pci_dev, ctrl_pos);
+            ctrl_word = (uint16_t *)(pci_dev->config + ctrl_pos);
+            assigned_dev_update_msix(pci_dev, (*ctrl_word &

PCI_MSIX_ENABLE));

}

         pos += PCI_CAPABILITY_CONFIG_MSIX_LENGTH;
     
     }

Sheng Yang

author

2010-11-04 06:18 UTC #9 lore raw

PATCH 3/3 qemu-kvm: device assignment: emulate MSI-X mask bits

In reply to Sheng Yang

This patch emulated MSI-X per vector mask bit on assigned device.

Signed-off-by Sheng Yang <[email protected]>

---
 hw/device-assignment.c |  161 ++++++++++++++++++++++++++++++++++++++++++++++--
 1 files changed, 155 insertions(+), 6 deletions(-)

hw/device-assignment.c +155 -6

index 8a98876..639aa0b 100644
@@ -62,6 +62,11 @@ static void assigned_dev_load_option_rom(AssignedDevice *dev);
 
 static void assigned_dev_unregister_msix_mmio(AssignedDevice *dev);
 
+static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus, uint8_t devfn)
+{
+    return (uint32_t)seg << 16 | (uint32_t)bus << 8 | (uint32_t)devfn;
+}
+
 static uint32_t assigned_dev_ioport_rw(AssignedDevRegion *dev_region,
                                        uint32_t addr, int len, uint32_t *val)
 {

@@ -264,6 +269,9 @@ static void assigned_dev_iomem_map(PCIDevice *pci_dev, int region_num,
     AssignedDevRegion *region = &r_dev->v_addrs[region_num];
     PCIRegion *real_region = &r_dev->real_device.regions[region_num];
     int ret = 0;
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    struct kvm_assigned_msix_mmio msix_mmio;
+#endif
 
     DEBUG("e_phys=%08" FMT_PCIBUS " r_virt=%p type=%d len=%08" FMT_PCIBUS " region_num=%d \n",
           e_phys, region->u.r_virtbase, type, e_size, region_num);

@@ -282,6 +290,16 @@ static void assigned_dev_iomem_map(PCIDevice *pci_dev, int region_num,
 
             cpu_register_physical_memory(e_phys + offset,
                     TARGET_PAGE_SIZE, r_dev->mmio_index);
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+	    memset(&msix_mmio, 0, sizeof(struct kvm_assigned_msix_mmio));
+	    msix_mmio.assigned_dev_id = calc_assigned_dev_id(r_dev->h_segnr,
+			    r_dev->h_busnr, r_dev->h_devfn);
+	    msix_mmio.base_addr = e_phys + offset;
+            /* We required kernel MSI-X support */
+	    ret = kvm_assign_reg_msix_mmio(kvm_context, &msix_mmio);
+	    if (ret)
+                fprintf(stderr, "fail to register in-kernel msix_mmio!\n");
+#endif
         }
     }

@@ -824,11 +842,6 @@ static void free_assigned_device(AssignedDevice *dev)
     }
 }
 
-static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus, uint8_t devfn)
-{
-    return (uint32_t)seg << 16 | (uint32_t)bus << 8 | (uint32_t)devfn;
-}
-
 static void assign_failed_examine(AssignedDevice *dev)
 {
     char name[PATH_MAX], dir[PATH_MAX], driver[PATH_MAX] = {}, *ns;

@@ -1123,6 +1136,27 @@ static int get_msix_entries_max_nr(AssignedDevice *adev)
     return entries_max_nr;
 }
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+static int assigned_dev_msix_entry_masked(AssignedDevice *adev, int entry)
+{
+    struct kvm_assigned_msix_entry msix_entry;
+    int r;
+
+    memset(&msix_entry, 0, sizeof msix_entry);
+    msix_entry.assigned_dev_id = calc_assigned_dev_id(adev->h_segnr,
+            adev->h_busnr, (uint8_t)adev->h_devfn);
+    msix_entry.entry = entry;
+    msix_entry.flags = KVM_MSIX_FLAG_QUERY_MASK;
+    r = kvm_assign_get_msix_entry(kvm_context, &msix_entry);
+    if (r) {
+        fprintf(stderr, "assigned_dev_msix_entry_masked: "
+			"Fail to get mask bit of entry %d\n", entry);
+        return 1;
+    }
+    return (msix_entry.flags & KVM_MSIX_FLAG_MASK);
+}
+#endif
+
 static int get_msix_valid_entries_nr(AssignedDevice *adev,
 				     uint16_t entries_max_nr)
 {

@@ -1136,7 +1170,11 @@ static int get_msix_valid_entries_nr(AssignedDevice *adev,
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
         /* Ignore unused entry even it's unmasked */
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else
         if (msg_data == 0)
+#endif
             continue;
         entries_nr ++;
     }

@@ -1165,6 +1203,8 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
     }
 
     free_dev_irq_entries(adev);
+    memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+                                        sizeof(*pci_dev->msix_entry_used));
     adev->irq_entries_nr = entries_nr;
     adev->entry = calloc(entries_nr, sizeof(struct kvm_irq_routing_entry));
     if (!adev->entry) {

@@ -1179,7 +1219,11 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
             break;
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else
         if (msg_data == 0)
+#endif
             continue;
 
         memcpy(&msg_addr, va + i * 16, 4);

@@ -1200,6 +1244,7 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
 
         msix_entry.gsi = adev->entry[entries_nr].gsi;
         msix_entry.entry = i;
+        pci_dev->msix_entry_used[i] = 1;
         r = kvm_assign_set_msix_entry(kvm_context, &msix_entry);
         if (r) {
             fprintf(stderr, "fail to set MSI-X entry! %s\n", strerror(-r));

@@ -1243,6 +1288,8 @@ static void assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix)
             perror("assigned_dev_update_msix: deassign irq");
 
         assigned_dev->irq_requested_type = 0;
+        memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+                                        sizeof(*pci_dev->msix_entry_used));
     }
 
     entries_max_nr = get_msix_entries_max_nr(assigned_dev);

@@ -1250,10 +1297,16 @@ static void assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix)
         fprintf(stderr, "assigned_dev_update_msix: MSI-X entries_max_nr == 0");
         return;
     }
+    /*
+     * Guest may try to enable MSI-X before setting MSI-X entry done, so
+     * let's wait until guest unmask the entries.
+     */
     entries_nr = get_msix_valid_entries_nr(assigned_dev, entries_max_nr);
     if (entries_nr == 0) {
+#ifndef KVM_CAP_DEVICE_MSIX_MASK
         if (enable_msix)
             fprintf(stderr, "MSI-X entry number is zero!\n");
+#endif
         return;
     }
     if (enable_msix) {

@@ -1297,7 +1350,8 @@ static void assigned_device_pci_cap_write_config(PCIDevice *pci_dev, uint32_t ad
         if (address <= ctrl_pos && address + len > ctrl_pos) {
             ctrl_pos--; /* control is word long */
             ctrl_word = (uint16_t *)(pci_dev->config + ctrl_pos);
-            assigned_dev_update_msix(pci_dev, (*ctrl_word & PCI_MSIX_ENABLE));
+            assigned_dev_update_msix(pci_dev,
+                    (*ctrl_word & PCI_MSIX_ENABLE) && !(*ctrl_word & PCI_MSIX_MASK));
 	}
         pos += PCI_CAPABILITY_CONFIG_MSIX_LENGTH;
     }

@@ -1395,10 +1449,101 @@ static void msix_mmio_writel(void *opaque,
     AssignedDevice *adev = opaque;
     unsigned int offset = addr & 0xfff;
     void *page = adev->msix_table_page;
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    int pos, ctrl_word, index;
+    struct kvm_irq_routing_entry new_entry = {};
+    int entry_idx, entries_max_nr, r = 0, i;
+    uint32_t msg_ctrl, msg_data, msg_upper_addr, msg_addr;
+#endif
 
     DEBUG("write to MSI-X entry table mmio offset 0x%lx, val 0x%x\n",
 		    addr, val);
     memcpy((void *)((char *)page + offset), &val, 4);
+
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    index = offset / 16;
+
+    /* Check if mask bit is being accessed */
+    memcpy(&msg_addr, (char *)page + index * 16, 4);
+    memcpy(&msg_upper_addr, (char *)page + index * 16 + 4, 4);
+    memcpy(&msg_data, (char *)page + index * 16 + 8, 4);
+    memcpy(&msg_ctrl, (char *)page + index * 16 + 12, 4);
+    DEBUG("MSI-X entries index %d: "
+            "msg_addr 0x%x, msg_upper_addr 0x%x, msg_data 0x%x, vec_ctl 0x%x\n",
+            index, msg_addr, msg_upper_addr, msg_data, msg_ctrl);
+
+    if (adev->cap.available & ASSIGNED_DEVICE_CAP_MSI)
+        pos = adev->dev.cap.start + PCI_CAPABILITY_CONFIG_MSI_LENGTH;
+    else
+        pos = adev->dev.cap.start;
+
+    ctrl_word = *(uint16_t *)(adev->dev.config + pos + 2);
+
+    if (!((ctrl_word & PCI_MSIX_ENABLE) && !(ctrl_word & PCI_MSIX_MASK)))
+        return;
+
+    if (!assigned_dev_msix_entry_masked(adev, index)) {
+        if (!adev->dev.msix_entry_used[index]) {
+            DEBUG("Try to modify unenabled MSI-X entry %d's mask. "
+                    "Reenable MSI-X.\n",
+                    index);
+            assigned_dev_update_msix(&adev->dev, 1);
+        }
+        return;
+    }
+
+    if (!adev->dev.msix_entry_used[index])
+        return;
+
+    /*
+     * We're here only because guest want to modify MSI data/addr, and
+     * kernel would filter those writing with mask bit unset.
+     */
+    entries_max_nr = get_msix_entries_max_nr(adev);
+
+    /*
+     * Find the index of routing entry, it can be different from 'index' if
+     * empty entry existed in between
+     */
+    entry_idx = -1;
+    for (i = 0; i <= index; i++) {
+        if (adev->dev.msix_entry_used[i])
+            entry_idx ++;
+    }
+    if (entry_idx >= entries_max_nr || entry_idx == -1) {
+        fprintf(stderr, "msix_mmio_writel: Entry idx %d exceed limit!\n",
+			entry_idx);
+        return;
+    }
+
+    if (!assigned_dev_msix_entry_masked(adev, index)) {
+        fprintf(stderr, "msix_mmio_writel: Trying write to unmasked entry!\n");
+        return;
+    }
+
+    new_entry.gsi = adev->entry[entry_idx].gsi;
+    new_entry.type = KVM_IRQ_ROUTING_MSI;
+    new_entry.flags = 0;
+    new_entry.u.msi.address_lo = msg_addr;
+    new_entry.u.msi.address_hi = msg_upper_addr;
+    new_entry.u.msi.data = msg_data;
+    if (memcmp(&adev->entry[entry_idx].u.msi, &new_entry.u.msi,
+                sizeof new_entry.u.msi)) {
+        r = kvm_update_routing_entry(&adev->entry[entry_idx], &new_entry);
+        if (r) {
+            perror("msix_mmio_writel: kvm_update_routing_entry failed\n");
+            return;
+        }
+        r = kvm_commit_irq_routes();
+        if (r) {
+            perror("msix_mmio_writel: kvm_commit_irq_routes failed\n");
+            return;
+        }
+    }
+    adev->entry[entry_idx].u.msi.address_lo = msg_addr;
+    adev->entry[entry_idx].u.msi.address_hi = msg_upper_addr;
+    adev->entry[entry_idx].u.msi.data = msg_data;
+#endif
 }
 
 static void msix_mmio_writew(void *opaque,

@@ -1436,6 +1581,8 @@ static int assigned_dev_register_msix_mmio(AssignedDevice *dev)
     memset(dev->msix_table_page, 0, 0x1000);
     dev->mmio_index = cpu_register_io_memory(
                         msix_mmio_read, msix_mmio_write, dev);
+    dev->dev.msix_entry_used = qemu_mallocz(KVM_MAX_MSIX_PER_DEV *
+                                        sizeof *dev->dev.msix_entry_used);
     return 0;
 }

@@ -1452,6 +1599,8 @@ static void assigned_dev_unregister_msix_mmio(AssignedDevice *dev)
                 strerror(errno));
     }
     dev->msix_table_page = NULL;
+    free(dev->dev.msix_entry_used);
+    dev->dev.msix_entry_used = NULL;
 }
 
 static int assigned_initfn(struct PCIDevice *pci_dev)

Signature · 1 line

1.7.0.1

Michael S. Tsirkin

2010-11-04 09:44 UTC #10 lore raw

Re: [PATCH 3/3] qemu-kvm: device assignment: emulate MSI-X mask bits

Quoted 72 lines · Sheng Yang

This patch emulated MSI-X per vector mask bit on assigned device.

Signed-off-by Sheng Yang <[email protected]>

---
 hw/device-assignment.c |  161 ++++++++++++++++++++++++++++++++++++++++++++++--
 1 files changed, 155 insertions(+), 6 deletions(-)

hw/device-assignment.c +24 -5

index 8a98876..639aa0b 100644
@@ -62,6 +62,11 @@ static void assigned_dev_load_option_rom(AssignedDevice *dev);
 
 static void assigned_dev_unregister_msix_mmio(AssignedDevice *dev);
 
+static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus, uint8_t devfn)
+{
+    return (uint32_t)seg << 16 | (uint32_t)bus << 8 | (uint32_t)devfn;
+}
+
 static uint32_t assigned_dev_ioport_rw(AssignedDevRegion *dev_region,
                                        uint32_t addr, int len, uint32_t *val)
 {

@@ -264,6 +269,9 @@ static void assigned_dev_iomem_map(PCIDevice *pci_dev, int region_num,
     AssignedDevRegion *region = &r_dev->v_addrs[region_num];
     PCIRegion *real_region = &r_dev->real_device.regions[region_num];
     int ret = 0;
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    struct kvm_assigned_msix_mmio msix_mmio;
+#endif
 
     DEBUG("e_phys=%08" FMT_PCIBUS " r_virt=%p type=%d len=%08" FMT_PCIBUS " region_num=%d \n",
           e_phys, region->u.r_virtbase, type, e_size, region_num);

@@ -282,6 +290,16 @@ static void assigned_dev_iomem_map(PCIDevice *pci_dev, int region_num,
 
             cpu_register_physical_memory(e_phys + offset,
                     TARGET_PAGE_SIZE, r_dev->mmio_index);
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+	    memset(&msix_mmio, 0, sizeof(struct kvm_assigned_msix_mmio));
+	    msix_mmio.assigned_dev_id = calc_assigned_dev_id(r_dev->h_segnr,
+			    r_dev->h_busnr, r_dev->h_devfn);
+	    msix_mmio.base_addr = e_phys + offset;
+            /* We required kernel MSI-X support */
+	    ret = kvm_assign_reg_msix_mmio(kvm_context, &msix_mmio);
+	    if (ret)
+                fprintf(stderr, "fail to register in-kernel msix_mmio!\n");
+#endif
         }
     }

@@ -824,11 +842,6 @@ static void free_assigned_device(AssignedDevice *dev)
     }
 }
 
-static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus, uint8_t devfn)
-{
-    return (uint32_t)seg << 16 | (uint32_t)bus << 8 | (uint32_t)devfn;
-}
-
 static void assign_failed_examine(AssignedDevice *dev)
 {
     char name[PATH_MAX], dir[PATH_MAX], driver[PATH_MAX] = {}, *ns;

@@ -1123,6 +1136,27 @@ static int get_msix_entries_max_nr(AssignedDevice *adev)
     return entries_max_nr;
 }
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+static int assigned_dev_msix_entry_masked(AssignedDevice *adev, int entry)
+{
+    struct kvm_assigned_msix_entry msix_entry;
+    int r;
+

hw/device-assignment.c static int get_msix_entries_max_nr(AssignedDevice *adev) · line 1154

Quoted 10 lines

hw/device-assignment.c +10 -0

+    memset(&msix_entry, 0, sizeof msix_entry);
+    msix_entry.assigned_dev_id = calc_assigned_dev_id(adev->h_segnr,
+            adev->h_busnr, (uint8_t)adev->h_devfn);
+    msix_entry.entry = entry;
+    msix_entry.flags = KVM_MSIX_FLAG_QUERY_MASK;
+    r = kvm_assign_get_msix_entry(kvm_context, &msix_entry);
+    if (r) {
+        fprintf(stderr, "assigned_dev_msix_entry_masked: "
+			"Fail to get mask bit of entry %d\n", entry);
+        return 1;

This error handling seems pretty useless. assert?

Quoted 9 lines

hw/device-assignment.c +5 -0

+    }
+    return (msix_entry.flags & KVM_MSIX_FLAG_MASK);
+}
+#endif
+
 static int get_msix_valid_entries_nr(AssignedDevice *adev,
 				     uint16_t entries_max_nr)
 {

hw/device-assignment.c static int get_msix_valid_entries_nr(AssignedDevice *adev, · line 1177

Quoted 8 lines

hw/device-assignment.c +4 -0

@@ -1136,7 +1170,11 @@ static int get_msix_valid_entries_nr(AssignedDevice *adev,
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
         /* Ignore unused entry even it's unmasked */
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else
         if (msg_data == 0)
+#endif

So, we are replacing msg_data == 0 check with masked check? If yes why not do this for non-KVM_CAP_DEVICE_MSIX_MASK too?

Quoted 13 lines

hw/device-assignment.c +2 -0

             continue;
         entries_nr ++;
     }

@@ -1165,6 +1203,8 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
     }
 
     free_dev_irq_entries(adev);
+    memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+                                        sizeof(*pci_dev->msix_entry_used));
     adev->irq_entries_nr = entries_nr;
     adev->entry = calloc(entries_nr, sizeof(struct kvm_irq_routing_entry));
     if (!adev->entry) {

hw/device-assignment.c static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev, · line 1226

Quoted 8 lines

hw/device-assignment.c +4 -0

@@ -1179,7 +1219,11 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
             break;
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else
         if (msg_data == 0)
+#endif

You can't use ifdef to check that kernel supports an ioctl. You must check this at runtime.

Quoted 21 lines

hw/device-assignment.c +3 -0

             continue;
 
         memcpy(&msg_addr, va + i * 16, 4);

@@ -1200,6 +1244,7 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
 
         msix_entry.gsi = adev->entry[entries_nr].gsi;
         msix_entry.entry = i;
+        pci_dev->msix_entry_used[i] = 1;
         r = kvm_assign_set_msix_entry(kvm_context, &msix_entry);
         if (r) {
             fprintf(stderr, "fail to set MSI-X entry! %s\n", strerror(-r));

@@ -1243,6 +1288,8 @@ static void assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix)
             perror("assigned_dev_update_msix: deassign irq");
 
         assigned_dev->irq_requested_type = 0;
+        memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+                                        sizeof(*pci_dev->msix_entry_used));
     }
 
     entries_max_nr = get_msix_entries_max_nr(assigned_dev);

hw/device-assignment.c static void assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix) · line 1303

Quoted 7 lines

hw/device-assignment.c +4 -0

@@ -1250,10 +1297,16 @@ static void assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix)
         fprintf(stderr, "assigned_dev_update_msix: MSI-X entries_max_nr == 0");
         return;
     }
+    /*
+     * Guest may try to enable MSI-X before setting MSI-X entry done, so
+     * let's wait until guest unmask the entries.
+     */

Well it can also set up any number of entries, enable msix then set up more entries. Now what?

hw/device-assignment.c static void assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix) · line 1308

Quoted 5 lines

hw/device-assignment.c +1 -0

     entries_nr = get_msix_valid_entries_nr(assigned_dev, entries_max_nr);
     if (entries_nr == 0) {
+#ifndef KVM_CAP_DEVICE_MSIX_MASK
         if (enable_msix)
             fprintf(stderr, "MSI-X entry number is zero!\n");

And what happens then?

Quoted 136 lines

hw/device-assignment.c +98 -1

+#endif
         return;
     }
     if (enable_msix) {

@@ -1297,7 +1350,8 @@ static void assigned_device_pci_cap_write_config(PCIDevice *pci_dev, uint32_t ad
         if (address <= ctrl_pos && address + len > ctrl_pos) {
             ctrl_pos--; /* control is word long */
             ctrl_word = (uint16_t *)(pci_dev->config + ctrl_pos);
-            assigned_dev_update_msix(pci_dev, (*ctrl_word & PCI_MSIX_ENABLE));
+            assigned_dev_update_msix(pci_dev,
+                    (*ctrl_word & PCI_MSIX_ENABLE) && !(*ctrl_word & PCI_MSIX_MASK));
 	}
         pos += PCI_CAPABILITY_CONFIG_MSIX_LENGTH;
     }

@@ -1395,10 +1449,101 @@ static void msix_mmio_writel(void *opaque,
     AssignedDevice *adev = opaque;
     unsigned int offset = addr & 0xfff;
     void *page = adev->msix_table_page;
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    int pos, ctrl_word, index;
+    struct kvm_irq_routing_entry new_entry = {};
+    int entry_idx, entries_max_nr, r = 0, i;
+    uint32_t msg_ctrl, msg_data, msg_upper_addr, msg_addr;
+#endif
 
     DEBUG("write to MSI-X entry table mmio offset 0x%lx, val 0x%x\n",
 		    addr, val);
     memcpy((void *)((char *)page + offset), &val, 4);
+
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    index = offset / 16;
+
+    /* Check if mask bit is being accessed */
+    memcpy(&msg_addr, (char *)page + index * 16, 4);
+    memcpy(&msg_upper_addr, (char *)page + index * 16 + 4, 4);
+    memcpy(&msg_data, (char *)page + index * 16 + 8, 4);
+    memcpy(&msg_ctrl, (char *)page + index * 16 + 12, 4);
+    DEBUG("MSI-X entries index %d: "
+            "msg_addr 0x%x, msg_upper_addr 0x%x, msg_data 0x%x, vec_ctl 0x%x\n",
+            index, msg_addr, msg_upper_addr, msg_data, msg_ctrl);
+
+    if (adev->cap.available & ASSIGNED_DEVICE_CAP_MSI)
+        pos = adev->dev.cap.start + PCI_CAPABILITY_CONFIG_MSI_LENGTH;
+    else
+        pos = adev->dev.cap.start;
+
+    ctrl_word = *(uint16_t *)(adev->dev.config + pos + 2);
+
+    if (!((ctrl_word & PCI_MSIX_ENABLE) && !(ctrl_word & PCI_MSIX_MASK)))
+        return;
+
+    if (!assigned_dev_msix_entry_masked(adev, index)) {
+        if (!adev->dev.msix_entry_used[index]) {
+            DEBUG("Try to modify unenabled MSI-X entry %d's mask. "
+                    "Reenable MSI-X.\n",
+                    index);
+            assigned_dev_update_msix(&adev->dev, 1);
+        }
+        return;
+    }
+
+    if (!adev->dev.msix_entry_used[index])
+        return;
+
+    /*
+     * We're here only because guest want to modify MSI data/addr, and
+     * kernel would filter those writing with mask bit unset.
+     */
+    entries_max_nr = get_msix_entries_max_nr(adev);
+
+    /*
+     * Find the index of routing entry, it can be different from 'index' if
+     * empty entry existed in between
+     */
+    entry_idx = -1;
+    for (i = 0; i <= index; i++) {
+        if (adev->dev.msix_entry_used[i])
+            entry_idx ++;
+    }
+    if (entry_idx >= entries_max_nr || entry_idx == -1) {
+        fprintf(stderr, "msix_mmio_writel: Entry idx %d exceed limit!\n",
+			entry_idx);
+        return;
+    }
+
+    if (!assigned_dev_msix_entry_masked(adev, index)) {
+        fprintf(stderr, "msix_mmio_writel: Trying write to unmasked entry!\n");
+        return;
+    }
+
+    new_entry.gsi = adev->entry[entry_idx].gsi;
+    new_entry.type = KVM_IRQ_ROUTING_MSI;
+    new_entry.flags = 0;
+    new_entry.u.msi.address_lo = msg_addr;
+    new_entry.u.msi.address_hi = msg_upper_addr;
+    new_entry.u.msi.data = msg_data;
+    if (memcmp(&adev->entry[entry_idx].u.msi, &new_entry.u.msi,
+                sizeof new_entry.u.msi)) {
+        r = kvm_update_routing_entry(&adev->entry[entry_idx], &new_entry);
+        if (r) {
+            perror("msix_mmio_writel: kvm_update_routing_entry failed\n");
+            return;
+        }
+        r = kvm_commit_irq_routes();
+        if (r) {
+            perror("msix_mmio_writel: kvm_commit_irq_routes failed\n");
+            return;
+        }
+    }
+    adev->entry[entry_idx].u.msi.address_lo = msg_addr;
+    adev->entry[entry_idx].u.msi.address_hi = msg_upper_addr;
+    adev->entry[entry_idx].u.msi.data = msg_data;
+#endif
 }
 
 static void msix_mmio_writew(void *opaque,

@@ -1436,6 +1581,8 @@ static int assigned_dev_register_msix_mmio(AssignedDevice *dev)
     memset(dev->msix_table_page, 0, 0x1000);
     dev->mmio_index = cpu_register_io_memory(
                         msix_mmio_read, msix_mmio_write, dev);
+    dev->dev.msix_entry_used = qemu_mallocz(KVM_MAX_MSIX_PER_DEV *
+                                        sizeof *dev->dev.msix_entry_used);
     return 0;
 }

@@ -1452,6 +1599,8 @@ static void assigned_dev_unregister_msix_mmio(AssignedDevice *dev)
                 strerror(errno));
     }
     dev->msix_table_page = NULL;
+    free(dev->dev.msix_entry_used);
+    dev->dev.msix_entry_used = NULL;
 }
 
 static int assigned_initfn(struct PCIDevice *pci_dev)

Signature · 1 line

1.7.0.1

Sheng Yang

author

2010-11-05 03:20 UTC #11 lore raw

Re: [PATCH 3/3] qemu-kvm: device assignment: emulate MSI-X mask bits

On Thursday 04 November 2010 17:44:27 Michael S. Tsirkin wrote:

Thanks very much for reviewing this! Seems nobody cares about userspace one before...

Quoted 8 lines · Sheng Yang

Quoted 7 lines
This patch emulated MSI-X per vector mask bit on assigned device.
Signed-off-by Sheng Yang <[email protected]>
---
 hw/device-assignment.c |  161
++++++++++++++++++++++++++++++++++++++++++++++-- 1 files

changed, 155

Quoted 97 lines

insertions(+), 6 deletions(-)

hw/device-assignment.c +7 -0

index 8a98876..639aa0b 100644
@@ -62,6 +62,11 @@ static void
 assigned_dev_load_option_rom(AssignedDevice *dev);
 
 static void assigned_dev_unregister_msix_mmio(AssignedDevice *dev);
 
+static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus, uint8_t
 devfn) +{
+    return (uint32_t)seg << 16 | (uint32_t)bus << 8 | (uint32_t)devfn;
+}
+
 
 static uint32_t assigned_dev_ioport_rw(AssignedDevRegion *dev_region,
 
                                        uint32_t addr, int len, uint32_t
                                        *val)
 
 {

@@ -264,6 +269,9 @@ static void assigned_dev_iomem_map(PCIDevice
 *pci_dev, int region_num,
 
     AssignedDevRegion *region = &r_dev->v_addrs[region_num];
     PCIRegion *real_region = &r_dev->real_device.regions[region_num];
     int ret = 0;
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    struct kvm_assigned_msix_mmio msix_mmio;
+#endif

     DEBUG("e_phys=%08" FMT_PCIBUS " r_virt=%p type=%d len=%08"
     FMT_PCIBUS " region_num=%d \n",
     
           e_phys, region->u.r_virtbase, type, e_size, region_num);

hw/device-assignment.c +9 -0

@@ -282,6 +290,16 @@ static void assigned_dev_iomem_map(PCIDevice
 *pci_dev, int region_num,
 
             cpu_register_physical_memory(e_phys + offset,
 
                     TARGET_PAGE_SIZE, r_dev->mmio_index);
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+	    memset(&msix_mmio, 0, sizeof(struct kvm_assigned_msix_mmio));
+	    msix_mmio.assigned_dev_id = calc_assigned_dev_id(r_dev->h_segnr,
+			    r_dev->h_busnr, r_dev->h_devfn);
+	    msix_mmio.base_addr = e_phys + offset;
+            /* We required kernel MSI-X support */
+	    ret = kvm_assign_reg_msix_mmio(kvm_context, &msix_mmio);
+	    if (ret)
+                fprintf(stderr, "fail to register in-kernel

msix_mmio!\n"); +#endif

         }
     
     }

hw/device-assignment.c +0 -1

@@ -824,11 +842,6 @@ static void free_assigned_device(AssignedDevice
 *dev)
 
     }
 
 }
 
-static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus, uint8_t

devfn) -{

hw/device-assignment.c +2 -3

-    return (uint32_t)seg << 16 | (uint32_t)bus << 8 | (uint32_t)devfn;
-}
-
 
 static void assign_failed_examine(AssignedDevice *dev)
 {
 
     char name[PATH_MAX], dir[PATH_MAX], driver[PATH_MAX] = {}, *ns;

@@ -1123,6 +1136,27 @@ static int get_msix_entries_max_nr(AssignedDevice
 *adev)
 
     return entries_max_nr;
 
 }
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+static int assigned_dev_msix_entry_masked(AssignedDevice *adev, int

entry) +{

hw/device-assignment.c +7 -0

+    struct kvm_assigned_msix_entry msix_entry;
+    int r;
+
+    memset(&msix_entry, 0, sizeof msix_entry);
+    msix_entry.assigned_dev_id = calc_assigned_dev_id(adev->h_segnr,
+            adev->h_busnr, (uint8_t)adev->h_devfn);
+    msix_entry.entry = entry;

Quoted 8 lines

hw/device-assignment.c static int get_msix_entries_max_nr(AssignedDevice *adev) · line 1154
Quoted 6 lines
hw/device-assignment.c +6 -0
+    msix_entry.flags = KVM_MSIX_FLAG_QUERY_MASK;
+    r = kvm_assign_get_msix_entry(kvm_context, &msix_entry);
+    if (r) {
+        fprintf(stderr, "assigned_dev_msix_entry_masked: "
+			"Fail to get mask bit of entry %d\n", entry);
+        return 1;
This error handling seems pretty useless. assert?

Maybe we can continue with it. Assert seems a little strict.

Quoted 24 lines

hw/device-assignment.c +8 -0

+    }
+    return (msix_entry.flags & KVM_MSIX_FLAG_MASK);
+}
+#endif
+
 
 static int get_msix_valid_entries_nr(AssignedDevice *adev,
 
 				     uint16_t entries_max_nr)
 
 {

@@ -1136,7 +1170,11 @@ static int
 get_msix_valid_entries_nr(AssignedDevice *adev,
 
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
         /* Ignore unused entry even it's unmasked */
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else

Quoted 6 lines

Quoted 3 lines
         if (msg_data == 0)
+#endif
So, we are replacing msg_data == 0 check with masked check? If yes why not do this for non-KVM_CAP_DEVICE_MSIX_MASK too?

Because the old one doesn't have idea about mask bit...

Quoted 34 lines

             continue;
         
         entries_nr ++;
     
     }

hw/device-assignment.c +2 -0

@@ -1165,6 +1203,8 @@ static int assigned_dev_update_msix_mmio(PCIDevice
 *pci_dev,
 
     }
 
     free_dev_irq_entries(adev);
 
+    memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+

sizeof(*pci_dev->msix_entry_used));

     adev->irq_entries_nr = entries_nr;
     adev->entry = calloc(entries_nr, sizeof(struct
     kvm_irq_routing_entry)); if (!adev->entry) {

hw/device-assignment.c +3 -0

@@ -1179,7 +1219,11 @@ static int assigned_dev_update_msix_mmio(PCIDevice
 *pci_dev,
 
             break;
 
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else

Quoted 6 lines

Quoted 3 lines
         if (msg_data == 0)
+#endif
You can't use ifdef to check that kernel supports an ioctl. You must check this at runtime.

Maybe we can convert them in bulk later.

Quoted 42 lines

             continue;
         
         memcpy(&msg_addr, va + i * 16, 4);

hw/device-assignment.c +1 -0

@@ -1200,6 +1244,7 @@ static int assigned_dev_update_msix_mmio(PCIDevice
 *pci_dev,
 
         msix_entry.gsi = adev->entry[entries_nr].gsi;
         msix_entry.entry = i;
 
+        pci_dev->msix_entry_used[i] = 1;

         r = kvm_assign_set_msix_entry(kvm_context, &msix_entry);
         if (r) {
         
             fprintf(stderr, "fail to set MSI-X entry! %s\n",
             strerror(-r));

hw/device-assignment.c +2 -0

@@ -1243,6 +1288,8 @@ static void assigned_dev_update_msix(PCIDevice
 *pci_dev, int enable_msix)
 
             perror("assigned_dev_update_msix: deassign irq");
 
         assigned_dev->irq_requested_type = 0;
 
+        memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+

sizeof(*pci_dev->msix_entry_used));

     }
     
     entries_max_nr = get_msix_entries_max_nr(assigned_dev);

hw/device-assignment.c +0 -0

@@ -1250,10 +1297,16 @@ static void assigned_dev_update_msix(PCIDevice
 *pci_dev, int enable_msix)
 
         fprintf(stderr, "assigned_dev_update_msix: MSI-X entries_max_nr
         == 0"); return;
 
     }

Quoted 7 lines

hw/device-assignment.c static void assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix) · line 1303
Quoted 4 lines
hw/device-assignment.c +4 -0
+    /*
+     * Guest may try to enable MSI-X before setting MSI-X entry done, so
+     * let's wait until guest unmask the entries.
+     */
Well it can also set up any number of entries, enable msix then set up more entries. Now what?

It's the same. If it want to set up more entries, it have to unmask them. Then we would get them.

Quoted 11 lines

hw/device-assignment.c static void assigned_dev_update_msix(PCIDevice

Quoted 9 lines

hw/device-assignment.c +1 -0

     entries_nr = get_msix_valid_entries_nr(assigned_dev,
     entries_max_nr); if (entries_nr == 0) {
 
+#ifndef KVM_CAP_DEVICE_MSIX_MASK
 
         if (enable_msix)
 
             fprintf(stderr, "MSI-X entry number is zero!\n");

And what happens then?

MSI-X can't work for old ones without MSIX mask support. Reload the guest module may help.

--
regards
Yang, Sheng

Quoted 166 lines

hw/device-assignment.c +1 -0

+#endif
 
         return;
 
     }
     if (enable_msix) {

@@ -1297,7 +1350,8 @@ static void

assigned_device_pci_cap_write_config(PCIDevice *pci_dev, uint32_t ad

         if (address <= ctrl_pos && address + len > ctrl_pos) {
         
             ctrl_pos--; /* control is word long */
             ctrl_word = (uint16_t *)(pci_dev->config + ctrl_pos);

- assigned_dev_update_msix(pci_dev, (*ctrl_word &
PCI_MSIX_ENABLE)); + assigned_dev_update_msix(pci_dev,
+ (*ctrl_word & PCI_MSIX_ENABLE) && !(*ctrl_word &
PCI_MSIX_MASK));

hw/device-assignment.c +17 -0

 	}
 
         pos += PCI_CAPABILITY_CONFIG_MSIX_LENGTH;
 
     }

@@ -1395,10 +1449,101 @@ static void msix_mmio_writel(void *opaque,
 
     AssignedDevice *adev = opaque;
     unsigned int offset = addr & 0xfff;
     void *page = adev->msix_table_page;
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    int pos, ctrl_word, index;
+    struct kvm_irq_routing_entry new_entry = {};
+    int entry_idx, entries_max_nr, r = 0, i;
+    uint32_t msg_ctrl, msg_data, msg_upper_addr, msg_addr;
+#endif
 
     DEBUG("write to MSI-X entry table mmio offset 0x%lx, val 0x%x\n",
 
 		    addr, val);
 
     memcpy((void *)((char *)page + offset), &val, 4);
 
+
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    index = offset / 16;
+
+    /* Check if mask bit is being accessed */
+    memcpy(&msg_addr, (char *)page + index * 16, 4);
+    memcpy(&msg_upper_addr, (char *)page + index * 16 + 4, 4);
+    memcpy(&msg_data, (char *)page + index * 16 + 8, 4);
+    memcpy(&msg_ctrl, (char *)page + index * 16 + 12, 4);
+    DEBUG("MSI-X entries index %d: "
+            "msg_addr 0x%x, msg_upper_addr 0x%x, msg_data 0x%x, vec_ctl

0x%x\n", + index, msg_addr, msg_upper_addr, msg_data,
msg_ctrl); +

hw/device-assignment.c +8 -0

+    if (adev->cap.available & ASSIGNED_DEVICE_CAP_MSI)
+        pos = adev->dev.cap.start + PCI_CAPABILITY_CONFIG_MSI_LENGTH;
+    else
+        pos = adev->dev.cap.start;
+
+    ctrl_word = *(uint16_t *)(adev->dev.config + pos + 2);
+
+    if (!((ctrl_word & PCI_MSIX_ENABLE) && !(ctrl_word &

PCI_MSIX_MASK))) + return;

hw/device-assignment.c +22 -0

+
+    if (!assigned_dev_msix_entry_masked(adev, index)) {
+        if (!adev->dev.msix_entry_used[index]) {
+            DEBUG("Try to modify unenabled MSI-X entry %d's mask. "
+                    "Reenable MSI-X.\n",
+                    index);
+            assigned_dev_update_msix(&adev->dev, 1);
+        }
+        return;
+    }
+
+    if (!adev->dev.msix_entry_used[index])
+        return;
+
+    /*
+     * We're here only because guest want to modify MSI data/addr, and
+     * kernel would filter those writing with mask bit unset.
+     */
+    entries_max_nr = get_msix_entries_max_nr(adev);
+
+    /*
+     * Find the index of routing entry, it can be different from 'index'

if + * empty entry existed in between

hw/device-assignment.c +8 -0

+     */
+    entry_idx = -1;
+    for (i = 0; i <= index; i++) {
+        if (adev->dev.msix_entry_used[i])
+            entry_idx ++;
+    }
+    if (entry_idx >= entries_max_nr || entry_idx == -1) {
+        fprintf(stderr, "msix_mmio_writel: Entry idx %d exceed

limit!\n", + entry_idx);

hw/device-assignment.c +5 -0

+        return;
+    }
+
+    if (!assigned_dev_msix_entry_masked(adev, index)) {
+        fprintf(stderr, "msix_mmio_writel: Trying write to unmasked

entry!\n"); + return;

hw/device-assignment.c +11 -0

+    }
+
+    new_entry.gsi = adev->entry[entry_idx].gsi;
+    new_entry.type = KVM_IRQ_ROUTING_MSI;
+    new_entry.flags = 0;
+    new_entry.u.msi.address_lo = msg_addr;
+    new_entry.u.msi.address_hi = msg_upper_addr;
+    new_entry.u.msi.data = msg_data;
+    if (memcmp(&adev->entry[entry_idx].u.msi, &new_entry.u.msi,
+                sizeof new_entry.u.msi)) {
+        r = kvm_update_routing_entry(&adev->entry[entry_idx],

&new_entry); + if (r) {
+ perror("msix_mmio_writel: kvm_update_routing_entry
failed\n"); + return;

hw/device-assignment.c +13 -0

+        }
+        r = kvm_commit_irq_routes();
+        if (r) {
+            perror("msix_mmio_writel: kvm_commit_irq_routes failed\n");
+            return;
+        }
+    }
+    adev->entry[entry_idx].u.msi.address_lo = msg_addr;
+    adev->entry[entry_idx].u.msi.address_hi = msg_upper_addr;
+    adev->entry[entry_idx].u.msi.data = msg_data;
+#endif
 
 }
 
 static void msix_mmio_writew(void *opaque,

@@ -1436,6 +1581,8 @@ static int
 assigned_dev_register_msix_mmio(AssignedDevice *dev)
 
     memset(dev->msix_table_page, 0, 0x1000);
     dev->mmio_index = cpu_register_io_memory(
 
                         msix_mmio_read, msix_mmio_write, dev);
 
+    dev->dev.msix_entry_used = qemu_mallocz(KVM_MAX_MSIX_PER_DEV *
+                                        sizeof

*dev->dev.msix_entry_used);

     return 0;

}

hw/device-assignment.c +2 -0

@@ -1452,6 +1599,8 @@ static void
 assigned_dev_unregister_msix_mmio(AssignedDevice *dev)
 
                 strerror(errno));
 
     }
     dev->msix_table_page = NULL;
 
+    free(dev->dev.msix_entry_used);
+    dev->dev.msix_entry_used = NULL;
 
 }
 
 static int assigned_initfn(struct PCIDevice *pci_dev)

Michael S. Tsirkin

2010-11-05 09:05 UTC #12 lore raw

Re: [PATCH 3/3] qemu-kvm: device assignment: emulate MSI-X mask bits

Quoted 120 lines · Sheng Yang

On Thursday 04 November 2010 17:44:27 Michael S. Tsirkin wrote:

Thanks very much for reviewing this! Seems nobody cares about userspace one before...

Quoted 8 lines · Sheng Yang

Quoted 7 lines
This patch emulated MSI-X per vector mask bit on assigned device.
Signed-off-by Sheng Yang <[email protected]>
---
 hw/device-assignment.c |  161
++++++++++++++++++++++++++++++++++++++++++++++-- 1 files

changed, 155

Quoted 97 lines

insertions(+), 6 deletions(-)

hw/device-assignment.c +7 -0

index 8a98876..639aa0b 100644
@@ -62,6 +62,11 @@ static void
 assigned_dev_load_option_rom(AssignedDevice *dev);
 
 static void assigned_dev_unregister_msix_mmio(AssignedDevice *dev);
 
+static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus, uint8_t
 devfn) +{
+    return (uint32_t)seg << 16 | (uint32_t)bus << 8 | (uint32_t)devfn;
+}
+
 
 static uint32_t assigned_dev_ioport_rw(AssignedDevRegion *dev_region,
 
                                        uint32_t addr, int len, uint32_t
                                        *val)
 
 {

@@ -264,6 +269,9 @@ static void assigned_dev_iomem_map(PCIDevice
 *pci_dev, int region_num,
 
     AssignedDevRegion *region = &r_dev->v_addrs[region_num];
     PCIRegion *real_region = &r_dev->real_device.regions[region_num];
     int ret = 0;
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    struct kvm_assigned_msix_mmio msix_mmio;
+#endif

     DEBUG("e_phys=%08" FMT_PCIBUS " r_virt=%p type=%d len=%08"
     FMT_PCIBUS " region_num=%d \n",
     
           e_phys, region->u.r_virtbase, type, e_size, region_num);

hw/device-assignment.c +9 -0

@@ -282,6 +290,16 @@ static void assigned_dev_iomem_map(PCIDevice
 *pci_dev, int region_num,
 
             cpu_register_physical_memory(e_phys + offset,
 
                     TARGET_PAGE_SIZE, r_dev->mmio_index);
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+	    memset(&msix_mmio, 0, sizeof(struct kvm_assigned_msix_mmio));
+	    msix_mmio.assigned_dev_id = calc_assigned_dev_id(r_dev->h_segnr,
+			    r_dev->h_busnr, r_dev->h_devfn);
+	    msix_mmio.base_addr = e_phys + offset;
+            /* We required kernel MSI-X support */
+	    ret = kvm_assign_reg_msix_mmio(kvm_context, &msix_mmio);
+	    if (ret)
+                fprintf(stderr, "fail to register in-kernel

msix_mmio!\n"); +#endif

         }
     
     }

hw/device-assignment.c +0 -1

@@ -824,11 +842,6 @@ static void free_assigned_device(AssignedDevice
 *dev)
 
     }
 
 }
 
-static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus, uint8_t

devfn) -{

hw/device-assignment.c +2 -3

-    return (uint32_t)seg << 16 | (uint32_t)bus << 8 | (uint32_t)devfn;
-}
-
 
 static void assign_failed_examine(AssignedDevice *dev)
 {
 
     char name[PATH_MAX], dir[PATH_MAX], driver[PATH_MAX] = {}, *ns;

@@ -1123,6 +1136,27 @@ static int get_msix_entries_max_nr(AssignedDevice
 *adev)
 
     return entries_max_nr;
 
 }
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+static int assigned_dev_msix_entry_masked(AssignedDevice *adev, int

entry) +{

hw/device-assignment.c +7 -0

+    struct kvm_assigned_msix_entry msix_entry;
+    int r;
+
+    memset(&msix_entry, 0, sizeof msix_entry);
+    msix_entry.assigned_dev_id = calc_assigned_dev_id(adev->h_segnr,
+            adev->h_busnr, (uint8_t)adev->h_devfn);
+    msix_entry.entry = entry;

Quoted 8 lines

hw/device-assignment.c static int get_msix_entries_max_nr(AssignedDevice *adev) · line 1154
Quoted 6 lines
hw/device-assignment.c +6 -0
+    msix_entry.flags = KVM_MSIX_FLAG_QUERY_MASK;
+    r = kvm_assign_get_msix_entry(kvm_context, &msix_entry);
+    if (r) {
+        fprintf(stderr, "assigned_dev_msix_entry_masked: "
+			"Fail to get mask bit of entry %d\n", entry);
+        return 1;
This error handling seems pretty useless. assert?

Quoted 1 line

Maybe we can continue with it. Assert seems a little strict.

Well need to consider whether to return 1 or 0 then.

Quoted 24 lines

hw/device-assignment.c +8 -0

+    }
+    return (msix_entry.flags & KVM_MSIX_FLAG_MASK);
+}
+#endif
+
 
 static int get_msix_valid_entries_nr(AssignedDevice *adev,
 
 				     uint16_t entries_max_nr)
 
 {

@@ -1136,7 +1170,11 @@ static int
 get_msix_valid_entries_nr(AssignedDevice *adev,
 
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
         /* Ignore unused entry even it's unmasked */
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else

Quoted 8 lines

Quoted 6 lines
Quoted 3 lines
         if (msg_data == 0)
+#endif
So, we are replacing msg_data == 0 check with masked check? If yes why not do this for non-KVM_CAP_DEVICE_MSIX_MASK too?
Because the old one doesn't have idea about mask bit...

So track mask bits in userspace.

Quoted 34 lines

             continue;
         
         entries_nr ++;
     
     }

hw/device-assignment.c +2 -0

@@ -1165,6 +1203,8 @@ static int assigned_dev_update_msix_mmio(PCIDevice
 *pci_dev,
 
     }
 
     free_dev_irq_entries(adev);
 
+    memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+

sizeof(*pci_dev->msix_entry_used));

     adev->irq_entries_nr = entries_nr;
     adev->entry = calloc(entries_nr, sizeof(struct
     kvm_irq_routing_entry)); if (!adev->entry) {

hw/device-assignment.c +3 -0

@@ -1179,7 +1219,11 @@ static int assigned_dev_update_msix_mmio(PCIDevice
 *pci_dev,
 
             break;
 
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else

Quoted 8 lines

Quoted 6 lines
Quoted 3 lines
         if (msg_data == 0)
+#endif
You can't use ifdef to check that kernel supports an ioctl. You must check this at runtime.
Maybe we can convert them in bulk later.

Well you are adding a bug, all writes on old kernels will spew out a ton of stderr.
We can't just assume new kernels.

Quoted 50 lines

Quoted 42 lines

             continue;
         
         memcpy(&msg_addr, va + i * 16, 4);

hw/device-assignment.c +1 -0

@@ -1200,6 +1244,7 @@ static int assigned_dev_update_msix_mmio(PCIDevice
 *pci_dev,
 
         msix_entry.gsi = adev->entry[entries_nr].gsi;
         msix_entry.entry = i;
 
+        pci_dev->msix_entry_used[i] = 1;

         r = kvm_assign_set_msix_entry(kvm_context, &msix_entry);
         if (r) {
         
             fprintf(stderr, "fail to set MSI-X entry! %s\n",
             strerror(-r));

hw/device-assignment.c +2 -0

@@ -1243,6 +1288,8 @@ static void assigned_dev_update_msix(PCIDevice
 *pci_dev, int enable_msix)
 
             perror("assigned_dev_update_msix: deassign irq");
 
         assigned_dev->irq_requested_type = 0;
 
+        memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+

sizeof(*pci_dev->msix_entry_used));

     }
     
     entries_max_nr = get_msix_entries_max_nr(assigned_dev);

hw/device-assignment.c +0 -0

@@ -1250,10 +1297,16 @@ static void assigned_dev_update_msix(PCIDevice
 *pci_dev, int enable_msix)
 
         fprintf(stderr, "assigned_dev_update_msix: MSI-X entries_max_nr
         == 0"); return;
 
     }

Quoted 7 lines

hw/device-assignment.c static void assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix) · line 1303
Quoted 4 lines
hw/device-assignment.c +4 -0
+    /*
+     * Guest may try to enable MSI-X before setting MSI-X entry done, so
+     * let's wait until guest unmask the entries.
+     */
Well it can also set up any number of entries, enable msix then set up more entries. Now what?

Quoted 2 lines

It's the same. If it want to set up more entries, it have to unmask them. Then we would get them.

Why can't we handle the first enable in the same way?

Quoted 6 lines

hw/device-assignment.c +1 -0

     entries_nr = get_msix_valid_entries_nr(assigned_dev,
     entries_max_nr); if (entries_nr == 0) {
 
+#ifndef KVM_CAP_DEVICE_MSIX_MASK

Quoted 7 lines

Quoted 5 lines
Quoted 3 lines
         if (enable_msix)
         
             fprintf(stderr, "MSI-X entry number is zero!\n");
And what happens then?
MSI-X can't work for old ones without MSIX mask support.

Old ones?

Quoted 2 lines

Reload the guest module
may help.

Guest might not have any concept of modules.

Quoted 3 lines

--
regards
Yang, Sheng

Sheng Yang

author

2010-11-05 11:01 UTC #13 lore raw

Re: [PATCH 3/3] qemu-kvm: device assignment: emulate MSI-X mask bits

Quoted 125 lines · On Friday 05 November 2010 17:05:32 Michael S. Tsirkin wrote:

Quoted 124 lines · Sheng Yang

On Thursday 04 November 2010 17:44:27 Michael S. Tsirkin wrote:

Thanks very much for reviewing this! Seems nobody cares about userspace one before...

Quoted 8 lines · Sheng Yang

Quoted 7 lines
This patch emulated MSI-X per vector mask bit on assigned device.
Signed-off-by Sheng Yang <[email protected]>
---
 hw/device-assignment.c |  161
++++++++++++++++++++++++++++++++++++++++++++++-- 1 files

changed, 155

Quoted 107 lines

Quoted 95 lines

insertions(+), 6 deletions(-)

hw/device-assignment.c +6 -0

index 8a98876..639aa0b 100644
@@ -62,6 +62,11 @@ static void
 assigned_dev_load_option_rom(AssignedDevice *dev);
 
 static void assigned_dev_unregister_msix_mmio(AssignedDevice *dev);
 
+static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus,
 uint8_t devfn) +{
+    return (uint32_t)seg << 16 | (uint32_t)bus << 8 |
 (uint32_t)devfn; +}
+
 
 static uint32_t assigned_dev_ioport_rw(AssignedDevRegion
 *dev_region,
 
                                        uint32_t addr, int len,
                                        uint32_t *val)
 
 {

@@ -264,6 +269,9 @@ static void assigned_dev_iomem_map(PCIDevice
 *pci_dev, int region_num,
 
     AssignedDevRegion *region = &r_dev->v_addrs[region_num];
     PCIRegion *real_region =
     &r_dev->real_device.regions[region_num]; int ret = 0;
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    struct kvm_assigned_msix_mmio msix_mmio;
+#endif

     DEBUG("e_phys=%08" FMT_PCIBUS " r_virt=%p type=%d len=%08"
     FMT_PCIBUS " region_num=%d \n",
     
           e_phys, region->u.r_virtbase, type, e_size, region_num);

hw/device-assignment.c +3 -0

@@ -282,6 +290,16 @@ static void assigned_dev_iomem_map(PCIDevice
 *pci_dev, int region_num,
 
             cpu_register_physical_memory(e_phys + offset,
 
                     TARGET_PAGE_SIZE, r_dev->mmio_index);
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+	    memset(&msix_mmio, 0, sizeof(struct kvm_assigned_msix_mmio));
+	    msix_mmio.assigned_dev_id =

calc_assigned_dev_id(r_dev->h_segnr, + r_dev->h_busnr,
r_dev->h_devfn);

hw/device-assignment.c +5 -0

+	    msix_mmio.base_addr = e_phys + offset;
+            /* We required kernel MSI-X support */
+	    ret = kvm_assign_reg_msix_mmio(kvm_context, &msix_mmio);
+	    if (ret)
+                fprintf(stderr, "fail to register in-kernel

msix_mmio!\n"); +#endif

         }
     
     }

hw/device-assignment.c +0 -1

@@ -824,11 +842,6 @@ static void free_assigned_device(AssignedDevice
 *dev)
 
     }
 
 }
 
-static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus,

uint8_t devfn) -{
- return (uint32_t)seg << 16 | (uint32_t)bus << 8 |
(uint32_t)devfn; -}

hw/device-assignment.c +2 -1

-
 
 static void assign_failed_examine(AssignedDevice *dev)
 {
 
     char name[PATH_MAX], dir[PATH_MAX], driver[PATH_MAX] = {}, *ns;

@@ -1123,6 +1136,27 @@ static int
 get_msix_entries_max_nr(AssignedDevice *adev)
 
     return entries_max_nr;
 
 }
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+static int assigned_dev_msix_entry_masked(AssignedDevice *adev, int

entry) +{

hw/device-assignment.c +3 -0

+    struct kvm_assigned_msix_entry msix_entry;
+    int r;
+

hw/device-assignment.c static int get_msix_entries_max_nr(AssignedDevice *adev) · line 1154

Quoted 10 lines

hw/device-assignment.c +10 -0

+    memset(&msix_entry, 0, sizeof msix_entry);
+    msix_entry.assigned_dev_id = calc_assigned_dev_id(adev->h_segnr,
+            adev->h_busnr, (uint8_t)adev->h_devfn);
+    msix_entry.entry = entry;
+    msix_entry.flags = KVM_MSIX_FLAG_QUERY_MASK;
+    r = kvm_assign_get_msix_entry(kvm_context, &msix_entry);
+    if (r) {
+        fprintf(stderr, "assigned_dev_msix_entry_masked: "
+			"Fail to get mask bit of entry %d\n", entry);
+        return 1;

This error handling seems pretty useless. assert?

Quoted 3 lines

Quoted 1 line
Maybe we can continue with it. Assert seems a little strict.
Well need to consider whether to return 1 or 0 then.

Tell it it's masked then routing change may be continue.

Quoted 26 lines

hw/device-assignment.c +8 -0

+    }
+    return (msix_entry.flags & KVM_MSIX_FLAG_MASK);
+}
+#endif
+
 
 static int get_msix_valid_entries_nr(AssignedDevice *adev,
 
 				     uint16_t entries_max_nr)
 
 {

@@ -1136,7 +1170,11 @@ static int
 get_msix_valid_entries_nr(AssignedDevice *adev,
 
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
         /* Ignore unused entry even it's unmasked */
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else
 
         if (msg_data == 0)

Quoted 8 lines

Quoted 6 lines
Quoted 4 lines
Quoted 1 line
+#endif
So, we are replacing msg_data == 0 check with masked check? If yes why not do this for non-KVM_CAP_DEVICE_MSIX_MASK too?
Because the old one doesn't have idea about mask bit...
So track mask bits in userspace.

That's not the target of this patch. And it still won't work.

Quoted 96 lines

Quoted 34 lines

             continue;
         
         entries_nr ++;
     
     }

hw/device-assignment.c +2 -0

@@ -1165,6 +1203,8 @@ static int
 assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
 
     }
 
     free_dev_irq_entries(adev);
 
+    memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+

sizeof(*pci_dev->msix_entry_used));

     adev->irq_entries_nr = entries_nr;
     adev->entry = calloc(entries_nr, sizeof(struct
     kvm_irq_routing_entry)); if (!adev->entry) {

hw/device-assignment.c +3 -0

@@ -1179,7 +1219,11 @@ static int
 assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
 
             break;
 
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else

Quoted 8 lines

Quoted 6 lines
Quoted 3 lines
         if (msg_data == 0)
+#endif
You can't use ifdef to check that kernel supports an ioctl. You must check this at runtime.
Maybe we can convert them in bulk later.

Well you are adding a bug, all writes on old kernels will spew out a ton of stderr.
We can't just assume new kernels.

Quoted 49 lines

Quoted 48 lines

Quoted 43 lines

             continue;
         
         memcpy(&msg_addr, va + i * 16, 4);

hw/device-assignment.c +1 -0

@@ -1200,6 +1244,7 @@ static int
 assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
 
         msix_entry.gsi = adev->entry[entries_nr].gsi;
         msix_entry.entry = i;
 
+        pci_dev->msix_entry_used[i] = 1;

         r = kvm_assign_set_msix_entry(kvm_context, &msix_entry);
         if (r) {
         
             fprintf(stderr, "fail to set MSI-X entry! %s\n",
             strerror(-r));

hw/device-assignment.c +2 -0

@@ -1243,6 +1288,8 @@ static void assigned_dev_update_msix(PCIDevice
 *pci_dev, int enable_msix)
 
             perror("assigned_dev_update_msix: deassign irq");
 
         assigned_dev->irq_requested_type = 0;
 
+        memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+

sizeof(*pci_dev->msix_entry_used));

     }
     
     entries_max_nr = get_msix_entries_max_nr(assigned_dev);

hw/device-assignment.c +2 -0

@@ -1250,10 +1297,16 @@ static void
 assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix)
 
         fprintf(stderr, "assigned_dev_update_msix: MSI-X
         entries_max_nr == 0"); return;
 
     }
 
+    /*
+     * Guest may try to enable MSI-X before setting MSI-X entry

Quoted 2 lines

done, so + * let's wait until guest unmask the entries.
+ */

Well it can also set up any number of entries, enable msix then set up more entries. Now what?

Quoted 4 lines

Quoted 2 lines
It's the same. If it want to set up more entries, it have to unmask them. Then we would get them.
Why can't we handle the first enable in the same way?

Don't understand the question, but I guess the answer is pci_enable_msix().

Quoted 8 lines

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
entries_nr = get_msix_valid_entries_nr(assigned_dev, entries_max_nr); if (entries_nr == 0) {
+#ifndef KVM_CAP_DEVICE_MSIX_MASK
if (enable_msix)

Quoted 7 lines

Quoted 5 lines
Quoted 3 lines
Quoted 1 line
             fprintf(stderr, "MSI-X entry number is zero!\n");
And what happens then?
MSI-X can't work for old ones without MSIX mask support.
Old ones?

Old userspace without KVM_CAP_DEVICE_MSIX_MASK

Quoted 5 lines

Quoted 3 lines
Reload the guest module
may help.
Guest might not have any concept of modules.

Just an workaround. Not a suggestion method.

--
regards
Yang, Sheng

Quoted 4 lines

Quoted 4 lines
--
regards
Yang, Sheng

Michael S. Tsirkin

2010-11-05 13:54 UTC #14 lore raw

Re: [PATCH 3/3] qemu-kvm: device assignment: emulate MSI-X mask bits

Quoted 126 lines · Sheng Yang

Quoted 125 lines · On Friday 05 November 2010 17:05:32 Michael S. Tsirkin wrote:

Quoted 124 lines · Sheng Yang

On Thursday 04 November 2010 17:44:27 Michael S. Tsirkin wrote:

Thanks very much for reviewing this! Seems nobody cares about userspace one before...

Quoted 8 lines · Sheng Yang

Quoted 7 lines
This patch emulated MSI-X per vector mask bit on assigned device.
Signed-off-by Sheng Yang <[email protected]>
---
 hw/device-assignment.c |  161
++++++++++++++++++++++++++++++++++++++++++++++-- 1 files

changed, 155

Quoted 107 lines

Quoted 95 lines

insertions(+), 6 deletions(-)

hw/device-assignment.c +6 -0

index 8a98876..639aa0b 100644
@@ -62,6 +62,11 @@ static void
 assigned_dev_load_option_rom(AssignedDevice *dev);
 
 static void assigned_dev_unregister_msix_mmio(AssignedDevice *dev);
 
+static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus,
 uint8_t devfn) +{
+    return (uint32_t)seg << 16 | (uint32_t)bus << 8 |
 (uint32_t)devfn; +}
+
 
 static uint32_t assigned_dev_ioport_rw(AssignedDevRegion
 *dev_region,
 
                                        uint32_t addr, int len,
                                        uint32_t *val)
 
 {

@@ -264,6 +269,9 @@ static void assigned_dev_iomem_map(PCIDevice
 *pci_dev, int region_num,
 
     AssignedDevRegion *region = &r_dev->v_addrs[region_num];
     PCIRegion *real_region =
     &r_dev->real_device.regions[region_num]; int ret = 0;
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    struct kvm_assigned_msix_mmio msix_mmio;
+#endif

     DEBUG("e_phys=%08" FMT_PCIBUS " r_virt=%p type=%d len=%08"
     FMT_PCIBUS " region_num=%d \n",
     
           e_phys, region->u.r_virtbase, type, e_size, region_num);

hw/device-assignment.c +3 -0

@@ -282,6 +290,16 @@ static void assigned_dev_iomem_map(PCIDevice
 *pci_dev, int region_num,
 
             cpu_register_physical_memory(e_phys + offset,
 
                     TARGET_PAGE_SIZE, r_dev->mmio_index);
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+	    memset(&msix_mmio, 0, sizeof(struct kvm_assigned_msix_mmio));
+	    msix_mmio.assigned_dev_id =

calc_assigned_dev_id(r_dev->h_segnr, + r_dev->h_busnr,
r_dev->h_devfn);

hw/device-assignment.c +5 -0

+	    msix_mmio.base_addr = e_phys + offset;
+            /* We required kernel MSI-X support */
+	    ret = kvm_assign_reg_msix_mmio(kvm_context, &msix_mmio);
+	    if (ret)
+                fprintf(stderr, "fail to register in-kernel

msix_mmio!\n"); +#endif

         }
     
     }

hw/device-assignment.c +0 -1

@@ -824,11 +842,6 @@ static void free_assigned_device(AssignedDevice
 *dev)
 
     }
 
 }
 
-static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus,

uint8_t devfn) -{
- return (uint32_t)seg << 16 | (uint32_t)bus << 8 |
(uint32_t)devfn; -}

hw/device-assignment.c +2 -1

-
 
 static void assign_failed_examine(AssignedDevice *dev)
 {
 
     char name[PATH_MAX], dir[PATH_MAX], driver[PATH_MAX] = {}, *ns;

@@ -1123,6 +1136,27 @@ static int
 get_msix_entries_max_nr(AssignedDevice *adev)
 
     return entries_max_nr;
 
 }
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+static int assigned_dev_msix_entry_masked(AssignedDevice *adev, int

entry) +{

hw/device-assignment.c +3 -0

+    struct kvm_assigned_msix_entry msix_entry;
+    int r;
+

hw/device-assignment.c static int get_msix_entries_max_nr(AssignedDevice *adev) · line 1154

Quoted 10 lines

hw/device-assignment.c +10 -0

+    memset(&msix_entry, 0, sizeof msix_entry);
+    msix_entry.assigned_dev_id = calc_assigned_dev_id(adev->h_segnr,
+            adev->h_busnr, (uint8_t)adev->h_devfn);
+    msix_entry.entry = entry;
+    msix_entry.flags = KVM_MSIX_FLAG_QUERY_MASK;
+    r = kvm_assign_get_msix_entry(kvm_context, &msix_entry);
+    if (r) {
+        fprintf(stderr, "assigned_dev_msix_entry_masked: "
+			"Fail to get mask bit of entry %d\n", entry);
+        return 1;

This error handling seems pretty useless. assert?

Quoted 5 lines

Quoted 3 lines
Quoted 1 line
Maybe we can continue with it. Assert seems a little strict.
Well need to consider whether to return 1 or 0 then.
Tell it it's masked then routing change may be continue.

OK, then
1. we probably want to print this only once, not on each write
2. only try doing this if capability is present.

Quoted 31 lines

Quoted 30 lines

Quoted 17 lines

hw/device-assignment.c +5 -0

+    }
+    return (msix_entry.flags & KVM_MSIX_FLAG_MASK);
+}
+#endif
+
 
 static int get_msix_valid_entries_nr(AssignedDevice *adev,
 
 				     uint16_t entries_max_nr)
 
 {

@@ -1136,7 +1170,11 @@ static int
 get_msix_valid_entries_nr(AssignedDevice *adev,
 
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);

hw/device-assignment.c static int · line 1177

Quoted 10 lines

hw/device-assignment.c +4 -0

         memcpy(&msg_data, va + i * 16 + 8, 4);
         /* Ignore unused entry even it's unmasked */
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else
 
         if (msg_data == 0)
 
+#endif

So, we are replacing msg_data == 0 check with masked check? If yes why not do this for non-KVM_CAP_DEVICE_MSIX_MASK too?

Quoted 5 lines

Quoted 3 lines
Quoted 1 line
Because the old one doesn't have idea about mask bit...
So track mask bits in userspace.
That's not the target of this patch. And it still won't work.

Sure it will, at least to figure out # of entries.

Quoted 96 lines

Quoted 34 lines

             continue;
         
         entries_nr ++;
     
     }

hw/device-assignment.c +2 -0

@@ -1165,6 +1203,8 @@ static int
 assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
 
     }
 
     free_dev_irq_entries(adev);
 
+    memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+

sizeof(*pci_dev->msix_entry_used));

     adev->irq_entries_nr = entries_nr;
     adev->entry = calloc(entries_nr, sizeof(struct
     kvm_irq_routing_entry)); if (!adev->entry) {

hw/device-assignment.c +3 -0

@@ -1179,7 +1219,11 @@ static int
 assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
 
             break;
 
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else

Quoted 8 lines

Quoted 6 lines
Quoted 3 lines
         if (msg_data == 0)
+#endif
You can't use ifdef to check that kernel supports an ioctl. You must check this at runtime.
Maybe we can convert them in bulk later.

Well you are adding a bug, all writes on old kernels will spew out a ton of stderr.
We can't just assume new kernels.

Quoted 49 lines

Quoted 48 lines

Quoted 43 lines

             continue;
         
         memcpy(&msg_addr, va + i * 16, 4);

hw/device-assignment.c +1 -0

@@ -1200,6 +1244,7 @@ static int
 assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
 
         msix_entry.gsi = adev->entry[entries_nr].gsi;
         msix_entry.entry = i;
 
+        pci_dev->msix_entry_used[i] = 1;

         r = kvm_assign_set_msix_entry(kvm_context, &msix_entry);
         if (r) {
         
             fprintf(stderr, "fail to set MSI-X entry! %s\n",
             strerror(-r));

hw/device-assignment.c +2 -0

@@ -1243,6 +1288,8 @@ static void assigned_dev_update_msix(PCIDevice
 *pci_dev, int enable_msix)
 
             perror("assigned_dev_update_msix: deassign irq");
 
         assigned_dev->irq_requested_type = 0;
 
+        memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+

sizeof(*pci_dev->msix_entry_used));

     }
     
     entries_max_nr = get_msix_entries_max_nr(assigned_dev);

hw/device-assignment.c +2 -0

@@ -1250,10 +1297,16 @@ static void
 assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix)
 
         fprintf(stderr, "assigned_dev_update_msix: MSI-X
         entries_max_nr == 0"); return;
 
     }
 
+    /*
+     * Guest may try to enable MSI-X before setting MSI-X entry

Quoted 2 lines

done, so + * let's wait until guest unmask the entries.
+ */

Well it can also set up any number of entries, enable msix then set up more entries. Now what?

Quoted 6 lines

Quoted 4 lines
Quoted 2 lines
It's the same. If it want to set up more entries, it have to unmask them. Then we would get them.
Why can't we handle the first enable in the same way?
Don't understand the question, but I guess the answer is pci_enable_msix().

pci_enable_msix is an internal kernel API to enable msix, isn't it? We are discussing qemu patch here.

Quoted 16 lines

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
entries_nr = get_msix_valid_entries_nr(assigned_dev, entries_max_nr); if (entries_nr == 0) {
+#ifndef KVM_CAP_DEVICE_MSIX_MASK
if (enable_msix)
Quoted 7 lines
Quoted 5 lines
Quoted 3 lines
Quoted 1 line
             fprintf(stderr, "MSI-X entry number is zero!\n");
And what happens then?
MSI-X can't work for old ones without MSIX mask support.
Old ones?

Quoted 8 lines

Old userspace without KVM_CAP_DEVICE_MSIX_MASK
Quoted 5 lines
Quoted 3 lines
Reload the guest module
may help.
Guest might not have any concept of modules.
Just an workaround. Not a suggestion method.

Look, if there's some failure mode we need a way to recover, or to make it very unlikely. If this is guest doing something illegal let's add a comment explaining what it is and why it's illegal. If this is legal, why the print out?

Quoted 8 lines

--
regards
Yang, Sheng
Quoted 4 lines
Quoted 4 lines
--
regards
Yang, Sheng

Sheng Yang

author

2010-11-08 05:17 UTC #15 lore raw

Re: [PATCH 3/3] qemu-kvm: device assignment: emulate MSI-X mask bits

Quoted 8 lines · On Friday 05 November 2010 21:54:16 Michael S. Tsirkin wrote:

Quoted 7 lines · Sheng Yang
Quoted 6 lines · On Friday 05 November 2010 17:05:32 Michael S. Tsirkin wrote:
Quoted 5 lines · Sheng Yang
On Thursday 04 November 2010 17:44:27 Michael S. Tsirkin wrote:
Thanks very much for reviewing this! Seems nobody cares about userspace one before...

Quoted 8 lines

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines · Sheng Yang
Quoted 7 lines
This patch emulated MSI-X per vector mask bit on assigned device.
Signed-off-by Sheng Yang <[email protected]>
---
 hw/device-assignment.c |  161
++++++++++++++++++++++++++++++++++++++++++++++-- 1

files

Quoted 156 lines

Quoted 49 lines

changed, 155

Quoted 46 lines

insertions(+), 6 deletions(-)

hw/device-assignment.c +2 -0

index 8a98876..639aa0b 100644
@@ -62,6 +62,11 @@ static void
 assigned_dev_load_option_rom(AssignedDevice *dev);
 
 static void assigned_dev_unregister_msix_mmio(AssignedDevice
 *dev);
 
+static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus,
 uint8_t devfn) +{
+    return (uint32_t)seg << 16 | (uint32_t)bus << 8 |

(uint32_t)devfn; +}

hw/device-assignment.c +4 -0

+
 
 static uint32_t assigned_dev_ioport_rw(AssignedDevRegion
 *dev_region,
 
                                        uint32_t addr, int len,
                                        uint32_t *val)
 
 {

@@ -264,6 +269,9 @@ static void assigned_dev_iomem_map(PCIDevice
 *pci_dev, int region_num,
 
     AssignedDevRegion *region = &r_dev->v_addrs[region_num];
     PCIRegion *real_region =
     &r_dev->real_device.regions[region_num]; int ret = 0;
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    struct kvm_assigned_msix_mmio msix_mmio;
+#endif

     DEBUG("e_phys=%08" FMT_PCIBUS " r_virt=%p type=%d len=%08"
     FMT_PCIBUS " region_num=%d \n",
     
           e_phys, region->u.r_virtbase, type, e_size,
           region_num);

hw/device-assignment.c +0 -0

@@ -282,6 +290,16 @@ static void assigned_dev_iomem_map(PCIDevice
 *pci_dev, int region_num,

Quoted 8 lines

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
hw/device-assignment.c +2 -0
             cpu_register_physical_memory(e_phys + offset,
 
                     TARGET_PAGE_SIZE, r_dev->mmio_index);
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+	    memset(&msix_mmio, 0, sizeof(struct
kvm_assigned_msix_mmio)); + msix_mmio.assigned_dev_id =
calc_assigned_dev_id(r_dev->h_segnr, + r_dev-

h_busnr,

Quoted 58 lines

Quoted 57 lines

Quoted 48 lines

r_dev->h_devfn);

hw/device-assignment.c +5 -0

+	    msix_mmio.base_addr = e_phys + offset;
+            /* We required kernel MSI-X support */
+	    ret = kvm_assign_reg_msix_mmio(kvm_context, &msix_mmio);
+	    if (ret)
+                fprintf(stderr, "fail to register in-kernel

msix_mmio!\n"); +#endif

         }
     
     }

hw/device-assignment.c +0 -1

@@ -824,11 +842,6 @@ static void
 free_assigned_device(AssignedDevice *dev)
 
     }
 
 }
 
-static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus,

uint8_t devfn) -{
- return (uint32_t)seg << 16 | (uint32_t)bus << 8 |
(uint32_t)devfn; -}

hw/device-assignment.c +2 -1

-
 
 static void assign_failed_examine(AssignedDevice *dev)
 {
 
     char name[PATH_MAX], dir[PATH_MAX], driver[PATH_MAX] = {},
     *ns;

@@ -1123,6 +1136,27 @@ static int
 get_msix_entries_max_nr(AssignedDevice *adev)
 
     return entries_max_nr;
 
 }
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+static int assigned_dev_msix_entry_masked(AssignedDevice *adev,

int entry) +{

hw/device-assignment.c +5 -0

+    struct kvm_assigned_msix_entry msix_entry;
+    int r;
+
+    memset(&msix_entry, 0, sizeof msix_entry);
+    msix_entry.assigned_dev_id =

calc_assigned_dev_id(adev->h_segnr, + adev->h_busnr,
(uint8_t)adev->h_devfn);

hw/device-assignment.c static int get_msix_entries_max_nr(AssignedDevice *adev) · line 1154

Quoted 7 lines

hw/device-assignment.c +7 -0

+    msix_entry.entry = entry;
+    msix_entry.flags = KVM_MSIX_FLAG_QUERY_MASK;
+    r = kvm_assign_get_msix_entry(kvm_context, &msix_entry);
+    if (r) {
+        fprintf(stderr, "assigned_dev_msix_entry_masked: "
+			"Fail to get mask bit of entry %d\n", entry);
+        return 1;

This error handling seems pretty useless. assert?

Quoted 5 lines

Quoted 3 lines
Quoted 1 line
Maybe we can continue with it. Assert seems a little strict.
Well need to consider whether to return 1 or 0 then.
Tell it it's masked then routing change may be continue.

OK, then
1. we probably want to print this only once, not on each write
2. only try doing this if capability is present.

Quoted 30 lines

Quoted 29 lines

Quoted 16 lines

hw/device-assignment.c +5 -0

+    }
+    return (msix_entry.flags & KVM_MSIX_FLAG_MASK);
+}
+#endif
+
 
 static int get_msix_valid_entries_nr(AssignedDevice *adev,
 
 				     uint16_t entries_max_nr)
 
 {

@@ -1136,7 +1170,11 @@ static int
 get_msix_valid_entries_nr(AssignedDevice *adev,
 
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);

hw/device-assignment.c static int · line 1177

Quoted 10 lines

hw/device-assignment.c +4 -0

         memcpy(&msg_data, va + i * 16 + 8, 4);
         /* Ignore unused entry even it's unmasked */
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else
 
         if (msg_data == 0)
 
+#endif

So, we are replacing msg_data == 0 check with masked check? If yes why not do this for non-KVM_CAP_DEVICE_MSIX_MASK too?

Quoted 7 lines

Quoted 5 lines
Quoted 3 lines
Quoted 1 line
Because the old one doesn't have idea about mask bit...
So track mask bits in userspace.
That's not the target of this patch. And it still won't work.
Sure it will, at least to figure out # of entries.

OK, we can only tracking them.

Quoted 98 lines

Quoted 33 lines

             continue;
         
         entries_nr ++;
     
     }

hw/device-assignment.c +2 -0

@@ -1165,6 +1203,8 @@ static int
 assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
 
     }
 
     free_dev_irq_entries(adev);
 
+    memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+

sizeof(*pci_dev->msix_entry_used));

     adev->irq_entries_nr = entries_nr;
     adev->entry = calloc(entries_nr, sizeof(struct
     kvm_irq_routing_entry)); if (!adev->entry) {

hw/device-assignment.c +3 -0

@@ -1179,7 +1219,11 @@ static int
 assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
 
             break;
 
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else

Quoted 8 lines

Quoted 6 lines
Quoted 3 lines
         if (msg_data == 0)
+#endif
You can't use ifdef to check that kernel supports an ioctl. You must check this at runtime.
Maybe we can convert them in bulk later.

Well you are adding a bug, all writes on old kernels will spew out a ton of stderr.
We can't just assume new kernels.

Quoted 51 lines

Quoted 41 lines

             continue;
         
         memcpy(&msg_addr, va + i * 16, 4);

hw/device-assignment.c +1 -0

@@ -1200,6 +1244,7 @@ static int
 assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
 
         msix_entry.gsi = adev->entry[entries_nr].gsi;
         msix_entry.entry = i;
 
+        pci_dev->msix_entry_used[i] = 1;

         r = kvm_assign_set_msix_entry(kvm_context, &msix_entry);
         if (r) {
         
             fprintf(stderr, "fail to set MSI-X entry! %s\n",
             strerror(-r));

hw/device-assignment.c +1 -0

@@ -1243,6 +1288,8 @@ static void
 assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix)
 
             perror("assigned_dev_update_msix: deassign irq");
 
         assigned_dev->irq_requested_type = 0;
 
+        memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV

* +
sizeof(*pci_dev->msix_entry_used));

     }
     
     entries_max_nr = get_msix_entries_max_nr(assigned_dev);

hw/device-assignment.c +0 -0

@@ -1250,10 +1297,16 @@ static void
 assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix)
 
         fprintf(stderr, "assigned_dev_update_msix: MSI-X
         entries_max_nr == 0"); return;
 
     }

Quoted 7 lines

Quoted 4 lines
hw/device-assignment.c +2 -0
+    /*
+     * Guest may try to enable MSI-X before setting MSI-X entry
done, so + * let's wait until guest unmask the entries.
+ */
Well it can also set up any number of entries, enable msix then set up more entries. Now what?

It's the same. If it want to set up more entries, it have to unmask them. Then we would get them.

Quoted 7 lines

Quoted 4 lines
Quoted 1 line
Why can't we handle the first enable in the same way?
Don't understand the question, but I guess the answer is pci_enable_msix().
pci_enable_msix is an internal kernel API to enable msix, isn't it? We are discussing qemu patch here.

OK, I understand the question now. The others entries would be enabled with unmask, but not very first ones. Guest can unmask the entries first, then enable MSI-X. We should scan the table when MSI-X enabled, that's why the first one is different from others.

Quoted 20 lines

Quoted 20 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
entries_nr = get_msix_valid_entries_nr(assigned_dev, entries_max_nr); if (entries_nr == 0) {
+#ifndef KVM_CAP_DEVICE_MSIX_MASK
if (enable_msix)
Quoted 7 lines
Quoted 5 lines
Quoted 3 lines
Quoted 1 line
             fprintf(stderr, "MSI-X entry number is zero!\n");
And what happens then?
MSI-X can't work for old ones without MSIX mask support.
Old ones?
Old userspace without KVM_CAP_DEVICE_MSIX_MASK
Quoted 3 lines
Quoted 2 lines
Reload the guest module
may help.

Quoted 8 lines

Quoted 3 lines
Quoted 1 line
Guest might not have any concept of modules.
Just an workaround. Not a suggestion method.
Look, if there's some failure mode we need a way to recover, or to make it very unlikely. If this is guest doing something illegal let's add a comment explaining what it is and why it's illegal. If this is legal, why the print out?

It's unsupported in the old QEmu. I can change the comment, but I do think it's good to let user know something important is wrong.

Maybe it's better to put it into DEBUG.

--
regards
Yang, Sheng

Quoted 8 lines

Quoted 8 lines
--
regards
Yang, Sheng
Quoted 3 lines
Quoted 3 lines
--
regards
Yang, Sheng

Michael S. Tsirkin

2010-11-08 06:27 UTC #16 lore raw

Re: [PATCH 3/3] qemu-kvm: device assignment: emulate MSI-X mask bits

Quoted 24 lines · Sheng Yang

Quoted 24 lines
Quoted 23 lines
Quoted 16 lines
Quoted 16 lines
Quoted 15 lines
Quoted 10 lines
(quoted diff) +2 -0
@@ -1250,10 +1297,16 @@ static void
 assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix)
 
         fprintf(stderr, "assigned_dev_update_msix: MSI-X
         entries_max_nr == 0"); return;
 
     }
 
+    /*
+     * Guest may try to enable MSI-X before setting MSI-X entry
Quoted 2 lines
done, so + * let's wait until guest unmask the entries.
+ */
Well it can also set up any number of entries, enable msix then set up more entries. Now what?
Quoted 4 lines
Quoted 2 lines
It's the same. If it want to set up more entries, it have to unmask them. Then we would get them.
Why can't we handle the first enable in the same way?
Don't understand the question, but I guess the answer is pci_enable_msix().

Quoted 7 lines

Quoted 2 lines
pci_enable_msix is an internal kernel API to enable msix, isn't it? We are discussing qemu patch here.
OK, I understand the question now. The others entries would be enabled with unmask, but not very first ones. Guest can unmask the entries first, then enable MSI-X. We should scan the table when MSI-X enabled, that's why the first one is different from others.

I see. I think the comment mislead me: you really mean 'Guest can unmask MSI-X entries when MSI-X is disabled,
don't do anything until MSI-X is enabled'.

Quoted 21 lines

Quoted 21 lines
Quoted 21 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
     entries_max_nr); if (entries_nr == 0) {
+#ifndef KVM_CAP_DEVICE_MSIX_MASK
if (enable_msix)
Quoted 7 lines
Quoted 5 lines
Quoted 3 lines
Quoted 1 line
             fprintf(stderr, "MSI-X entry number is zero!\n");
And what happens then?
MSI-X can't work for old ones without MSIX mask support.
Old ones?
Old userspace without KVM_CAP_DEVICE_MSIX_MASK
Quoted 4 lines
Quoted 2 lines
Reload the guest module
may help.
Guest might not have any concept of modules.

Quoted 8 lines

Quoted 6 lines
Quoted 1 line
Just an workaround. Not a suggestion method.
Look, if there's some failure mode we need a way to recover, or to make it very unlikely. If this is guest doing something illegal let's add a comment explaining what it is and why it's illegal. If this is legal, why the print out?
It's unsupported in the old QEmu.

I think I have it. The comment you want is:

/* Error: number of MSI-X entries is zero. Using this device
requires KVM_CAP_DEVICE_MSIX_MASK support in kvm which is missing.
*/

As I said prebiously, we must check runtime capabilities for this, ifdef is not enough to know kernel supports a feature because there's no dependency between kernel and qemu packages.

Quoted 2 lines

I can change the comment, but I do think it's good to let user know something important is wrong.

stderr is really there for developers.
You can't expect the user to understand this message.

Quoted 13 lines

Maybe it's better to put it into DEBUG.
--
regards
Yang, Sheng
Quoted 8 lines
Quoted 8 lines
--
regards
Yang, Sheng
Quoted 3 lines
Quoted 3 lines
--
regards
Yang, Sheng

Michael S. Tsirkin

2010-11-04 10:04 UTC #17 lore raw

Re: [PATCH 3/3] qemu-kvm: device assignment: emulate MSI-X mask bits

In reply to Sheng Yang

Quoted 3 lines · Sheng Yang

This patch emulated MSI-X per vector mask bit on assigned device.
Signed-off-by Sheng Yang <[email protected]>

Also pls update the in-tree header for the new ioctls.

Quoted 286 lines

---
 hw/device-assignment.c |  161 ++++++++++++++++++++++++++++++++++++++++++++++--
 1 files changed, 155 insertions(+), 6 deletions(-)

hw/device-assignment.c +155 -6

index 8a98876..639aa0b 100644
@@ -62,6 +62,11 @@ static void assigned_dev_load_option_rom(AssignedDevice *dev);
 
 static void assigned_dev_unregister_msix_mmio(AssignedDevice *dev);
 
+static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus, uint8_t devfn)
+{
+    return (uint32_t)seg << 16 | (uint32_t)bus << 8 | (uint32_t)devfn;
+}
+
 static uint32_t assigned_dev_ioport_rw(AssignedDevRegion *dev_region,
                                        uint32_t addr, int len, uint32_t *val)
 {

@@ -264,6 +269,9 @@ static void assigned_dev_iomem_map(PCIDevice *pci_dev, int region_num,
     AssignedDevRegion *region = &r_dev->v_addrs[region_num];
     PCIRegion *real_region = &r_dev->real_device.regions[region_num];
     int ret = 0;
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    struct kvm_assigned_msix_mmio msix_mmio;
+#endif
 
     DEBUG("e_phys=%08" FMT_PCIBUS " r_virt=%p type=%d len=%08" FMT_PCIBUS " region_num=%d \n",
           e_phys, region->u.r_virtbase, type, e_size, region_num);

@@ -282,6 +290,16 @@ static void assigned_dev_iomem_map(PCIDevice *pci_dev, int region_num,
 
             cpu_register_physical_memory(e_phys + offset,
                     TARGET_PAGE_SIZE, r_dev->mmio_index);
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+	    memset(&msix_mmio, 0, sizeof(struct kvm_assigned_msix_mmio));
+	    msix_mmio.assigned_dev_id = calc_assigned_dev_id(r_dev->h_segnr,
+			    r_dev->h_busnr, r_dev->h_devfn);
+	    msix_mmio.base_addr = e_phys + offset;
+            /* We required kernel MSI-X support */
+	    ret = kvm_assign_reg_msix_mmio(kvm_context, &msix_mmio);
+	    if (ret)
+                fprintf(stderr, "fail to register in-kernel msix_mmio!\n");
+#endif
         }
     }

@@ -824,11 +842,6 @@ static void free_assigned_device(AssignedDevice *dev)
     }
 }
 
-static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus, uint8_t devfn)
-{
-    return (uint32_t)seg << 16 | (uint32_t)bus << 8 | (uint32_t)devfn;
-}
-
 static void assign_failed_examine(AssignedDevice *dev)
 {
     char name[PATH_MAX], dir[PATH_MAX], driver[PATH_MAX] = {}, *ns;

@@ -1123,6 +1136,27 @@ static int get_msix_entries_max_nr(AssignedDevice *adev)
     return entries_max_nr;
 }
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+static int assigned_dev_msix_entry_masked(AssignedDevice *adev, int entry)
+{
+    struct kvm_assigned_msix_entry msix_entry;
+    int r;
+
+    memset(&msix_entry, 0, sizeof msix_entry);
+    msix_entry.assigned_dev_id = calc_assigned_dev_id(adev->h_segnr,
+            adev->h_busnr, (uint8_t)adev->h_devfn);
+    msix_entry.entry = entry;
+    msix_entry.flags = KVM_MSIX_FLAG_QUERY_MASK;
+    r = kvm_assign_get_msix_entry(kvm_context, &msix_entry);
+    if (r) {
+        fprintf(stderr, "assigned_dev_msix_entry_masked: "
+			"Fail to get mask bit of entry %d\n", entry);
+        return 1;
+    }
+    return (msix_entry.flags & KVM_MSIX_FLAG_MASK);
+}
+#endif
+
 static int get_msix_valid_entries_nr(AssignedDevice *adev,
 				     uint16_t entries_max_nr)
 {

@@ -1136,7 +1170,11 @@ static int get_msix_valid_entries_nr(AssignedDevice *adev,
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
         /* Ignore unused entry even it's unmasked */
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else
         if (msg_data == 0)
+#endif
             continue;
         entries_nr ++;
     }

@@ -1165,6 +1203,8 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
     }
 
     free_dev_irq_entries(adev);
+    memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+                                        sizeof(*pci_dev->msix_entry_used));
     adev->irq_entries_nr = entries_nr;
     adev->entry = calloc(entries_nr, sizeof(struct kvm_irq_routing_entry));
     if (!adev->entry) {

@@ -1179,7 +1219,11 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
             break;
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else
         if (msg_data == 0)
+#endif
             continue;
 
         memcpy(&msg_addr, va + i * 16, 4);

@@ -1200,6 +1244,7 @@ static int assigned_dev_update_msix_mmio(PCIDevice *pci_dev,
 
         msix_entry.gsi = adev->entry[entries_nr].gsi;
         msix_entry.entry = i;
+        pci_dev->msix_entry_used[i] = 1;
         r = kvm_assign_set_msix_entry(kvm_context, &msix_entry);
         if (r) {
             fprintf(stderr, "fail to set MSI-X entry! %s\n", strerror(-r));

@@ -1243,6 +1288,8 @@ static void assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix)
             perror("assigned_dev_update_msix: deassign irq");
 
         assigned_dev->irq_requested_type = 0;
+        memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+                                        sizeof(*pci_dev->msix_entry_used));
     }
 
     entries_max_nr = get_msix_entries_max_nr(assigned_dev);

@@ -1250,10 +1297,16 @@ static void assigned_dev_update_msix(PCIDevice *pci_dev, int enable_msix)
         fprintf(stderr, "assigned_dev_update_msix: MSI-X entries_max_nr == 0");
         return;
     }
+    /*
+     * Guest may try to enable MSI-X before setting MSI-X entry done, so
+     * let's wait until guest unmask the entries.
+     */
     entries_nr = get_msix_valid_entries_nr(assigned_dev, entries_max_nr);
     if (entries_nr == 0) {
+#ifndef KVM_CAP_DEVICE_MSIX_MASK
         if (enable_msix)
             fprintf(stderr, "MSI-X entry number is zero!\n");
+#endif
         return;
     }
     if (enable_msix) {

@@ -1297,7 +1350,8 @@ static void assigned_device_pci_cap_write_config(PCIDevice *pci_dev, uint32_t ad
         if (address <= ctrl_pos && address + len > ctrl_pos) {
             ctrl_pos--; /* control is word long */
             ctrl_word = (uint16_t *)(pci_dev->config + ctrl_pos);
-            assigned_dev_update_msix(pci_dev, (*ctrl_word & PCI_MSIX_ENABLE));
+            assigned_dev_update_msix(pci_dev,
+                    (*ctrl_word & PCI_MSIX_ENABLE) && !(*ctrl_word & PCI_MSIX_MASK));
 	}
         pos += PCI_CAPABILITY_CONFIG_MSIX_LENGTH;
     }

@@ -1395,10 +1449,101 @@ static void msix_mmio_writel(void *opaque,
     AssignedDevice *adev = opaque;
     unsigned int offset = addr & 0xfff;
     void *page = adev->msix_table_page;
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    int pos, ctrl_word, index;
+    struct kvm_irq_routing_entry new_entry = {};
+    int entry_idx, entries_max_nr, r = 0, i;
+    uint32_t msg_ctrl, msg_data, msg_upper_addr, msg_addr;
+#endif
 
     DEBUG("write to MSI-X entry table mmio offset 0x%lx, val 0x%x\n",
 		    addr, val);
     memcpy((void *)((char *)page + offset), &val, 4);
+
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    index = offset / 16;
+
+    /* Check if mask bit is being accessed */
+    memcpy(&msg_addr, (char *)page + index * 16, 4);
+    memcpy(&msg_upper_addr, (char *)page + index * 16 + 4, 4);
+    memcpy(&msg_data, (char *)page + index * 16 + 8, 4);
+    memcpy(&msg_ctrl, (char *)page + index * 16 + 12, 4);
+    DEBUG("MSI-X entries index %d: "
+            "msg_addr 0x%x, msg_upper_addr 0x%x, msg_data 0x%x, vec_ctl 0x%x\n",
+            index, msg_addr, msg_upper_addr, msg_data, msg_ctrl);
+
+    if (adev->cap.available & ASSIGNED_DEVICE_CAP_MSI)
+        pos = adev->dev.cap.start + PCI_CAPABILITY_CONFIG_MSI_LENGTH;
+    else
+        pos = adev->dev.cap.start;
+
+    ctrl_word = *(uint16_t *)(adev->dev.config + pos + 2);
+
+    if (!((ctrl_word & PCI_MSIX_ENABLE) && !(ctrl_word & PCI_MSIX_MASK)))
+        return;
+
+    if (!assigned_dev_msix_entry_masked(adev, index)) {
+        if (!adev->dev.msix_entry_used[index]) {
+            DEBUG("Try to modify unenabled MSI-X entry %d's mask. "
+                    "Reenable MSI-X.\n",
+                    index);
+            assigned_dev_update_msix(&adev->dev, 1);
+        }
+        return;
+    }
+
+    if (!adev->dev.msix_entry_used[index])
+        return;
+
+    /*
+     * We're here only because guest want to modify MSI data/addr, and
+     * kernel would filter those writing with mask bit unset.
+     */
+    entries_max_nr = get_msix_entries_max_nr(adev);
+
+    /*
+     * Find the index of routing entry, it can be different from 'index' if
+     * empty entry existed in between
+     */
+    entry_idx = -1;
+    for (i = 0; i <= index; i++) {
+        if (adev->dev.msix_entry_used[i])
+            entry_idx ++;
+    }
+    if (entry_idx >= entries_max_nr || entry_idx == -1) {
+        fprintf(stderr, "msix_mmio_writel: Entry idx %d exceed limit!\n",
+			entry_idx);
+        return;
+    }
+
+    if (!assigned_dev_msix_entry_masked(adev, index)) {
+        fprintf(stderr, "msix_mmio_writel: Trying write to unmasked entry!\n");
+        return;
+    }
+
+    new_entry.gsi = adev->entry[entry_idx].gsi;
+    new_entry.type = KVM_IRQ_ROUTING_MSI;
+    new_entry.flags = 0;
+    new_entry.u.msi.address_lo = msg_addr;
+    new_entry.u.msi.address_hi = msg_upper_addr;
+    new_entry.u.msi.data = msg_data;
+    if (memcmp(&adev->entry[entry_idx].u.msi, &new_entry.u.msi,
+                sizeof new_entry.u.msi)) {
+        r = kvm_update_routing_entry(&adev->entry[entry_idx], &new_entry);
+        if (r) {
+            perror("msix_mmio_writel: kvm_update_routing_entry failed\n");
+            return;
+        }
+        r = kvm_commit_irq_routes();
+        if (r) {
+            perror("msix_mmio_writel: kvm_commit_irq_routes failed\n");
+            return;
+        }
+    }
+    adev->entry[entry_idx].u.msi.address_lo = msg_addr;
+    adev->entry[entry_idx].u.msi.address_hi = msg_upper_addr;
+    adev->entry[entry_idx].u.msi.data = msg_data;
+#endif
 }
 
 static void msix_mmio_writew(void *opaque,

@@ -1436,6 +1581,8 @@ static int assigned_dev_register_msix_mmio(AssignedDevice *dev)
     memset(dev->msix_table_page, 0, 0x1000);
     dev->mmio_index = cpu_register_io_memory(
                         msix_mmio_read, msix_mmio_write, dev);
+    dev->dev.msix_entry_used = qemu_mallocz(KVM_MAX_MSIX_PER_DEV *
+                                        sizeof *dev->dev.msix_entry_used);
     return 0;
 }

@@ -1452,6 +1599,8 @@ static void assigned_dev_unregister_msix_mmio(AssignedDevice *dev)
                 strerror(errno));
     }
     dev->msix_table_page = NULL;
+    free(dev->dev.msix_entry_used);
+    dev->dev.msix_entry_used = NULL;
 }
 
 static int assigned_initfn(struct PCIDevice *pci_dev)

Signature · 1 line

1.7.0.1

Sheng Yang

author

2010-11-05 03:14 UTC #18 lore raw

Re: [PATCH 3/3] qemu-kvm: device assignment: emulate MSI-X mask bits

Quoted 6 lines · On Thursday 04 November 2010 18:04:08 Michael S. Tsirkin wrote:

Quoted 3 lines · Sheng Yang
This patch emulated MSI-X per vector mask bit on assigned device.
Signed-off-by Sheng Yang <[email protected]>
Also pls update the in-tree header for the new ioctls.

I thought this should be done by maintainer?

I just use one script to sync them everytime. I don't mind to send the diff out.

--
regards
Yang, Sheng

Quoted 5 lines

Quoted 5 lines
---
 hw/device-assignment.c |  161
++++++++++++++++++++++++++++++++++++++++++++++-- 1 files

changed, 155

Quoted 387 lines

insertions(+), 6 deletions(-)

hw/device-assignment.c +7 -0

index 8a98876..639aa0b 100644
@@ -62,6 +62,11 @@ static void
 assigned_dev_load_option_rom(AssignedDevice *dev);
 
 static void assigned_dev_unregister_msix_mmio(AssignedDevice *dev);
 
+static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus, uint8_t
 devfn) +{
+    return (uint32_t)seg << 16 | (uint32_t)bus << 8 | (uint32_t)devfn;
+}
+
 
 static uint32_t assigned_dev_ioport_rw(AssignedDevRegion *dev_region,
 
                                        uint32_t addr, int len, uint32_t
                                        *val)
 
 {

@@ -264,6 +269,9 @@ static void assigned_dev_iomem_map(PCIDevice
 *pci_dev, int region_num,
 
     AssignedDevRegion *region = &r_dev->v_addrs[region_num];
     PCIRegion *real_region = &r_dev->real_device.regions[region_num];
     int ret = 0;
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    struct kvm_assigned_msix_mmio msix_mmio;
+#endif

     DEBUG("e_phys=%08" FMT_PCIBUS " r_virt=%p type=%d len=%08"
     FMT_PCIBUS " region_num=%d \n",
     
           e_phys, region->u.r_virtbase, type, e_size, region_num);

hw/device-assignment.c +9 -0

@@ -282,6 +290,16 @@ static void assigned_dev_iomem_map(PCIDevice
 *pci_dev, int region_num,
 
             cpu_register_physical_memory(e_phys + offset,
 
                     TARGET_PAGE_SIZE, r_dev->mmio_index);
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+	    memset(&msix_mmio, 0, sizeof(struct kvm_assigned_msix_mmio));
+	    msix_mmio.assigned_dev_id = calc_assigned_dev_id(r_dev->h_segnr,
+			    r_dev->h_busnr, r_dev->h_devfn);
+	    msix_mmio.base_addr = e_phys + offset;
+            /* We required kernel MSI-X support */
+	    ret = kvm_assign_reg_msix_mmio(kvm_context, &msix_mmio);
+	    if (ret)
+                fprintf(stderr, "fail to register in-kernel

msix_mmio!\n"); +#endif

         }
     
     }

hw/device-assignment.c +0 -1

@@ -824,11 +842,6 @@ static void free_assigned_device(AssignedDevice
 *dev)
 
     }
 
 }
 
-static uint32_t calc_assigned_dev_id(uint16_t seg, uint8_t bus, uint8_t

devfn) -{

hw/device-assignment.c +2 -3

-    return (uint32_t)seg << 16 | (uint32_t)bus << 8 | (uint32_t)devfn;
-}
-
 
 static void assign_failed_examine(AssignedDevice *dev)
 {
 
     char name[PATH_MAX], dir[PATH_MAX], driver[PATH_MAX] = {}, *ns;

@@ -1123,6 +1136,27 @@ static int get_msix_entries_max_nr(AssignedDevice
 *adev)
 
     return entries_max_nr;
 
 }
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+static int assigned_dev_msix_entry_masked(AssignedDevice *adev, int

entry) +{

hw/device-assignment.c +24 -0

+    struct kvm_assigned_msix_entry msix_entry;
+    int r;
+
+    memset(&msix_entry, 0, sizeof msix_entry);
+    msix_entry.assigned_dev_id = calc_assigned_dev_id(adev->h_segnr,
+            adev->h_busnr, (uint8_t)adev->h_devfn);
+    msix_entry.entry = entry;
+    msix_entry.flags = KVM_MSIX_FLAG_QUERY_MASK;
+    r = kvm_assign_get_msix_entry(kvm_context, &msix_entry);
+    if (r) {
+        fprintf(stderr, "assigned_dev_msix_entry_masked: "
+			"Fail to get mask bit of entry %d\n", entry);
+        return 1;
+    }
+    return (msix_entry.flags & KVM_MSIX_FLAG_MASK);
+}
+#endif
+
 
 static int get_msix_valid_entries_nr(AssignedDevice *adev,
 
 				     uint16_t entries_max_nr)
 
 {

@@ -1136,7 +1170,11 @@ static int
 get_msix_valid_entries_nr(AssignedDevice *adev,
 
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
         /* Ignore unused entry even it's unmasked */
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else
 
         if (msg_data == 0)
 
+#endif
 
             continue;
 
         entries_nr ++;
 
     }

@@ -1165,6 +1203,8 @@ static int assigned_dev_update_msix_mmio(PCIDevice
 *pci_dev,
 
     }
 
     free_dev_irq_entries(adev);
 
+    memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+

sizeof(*pci_dev->msix_entry_used));

     adev->irq_entries_nr = entries_nr;
     adev->entry = calloc(entries_nr, sizeof(struct
     kvm_irq_routing_entry)); if (!adev->entry) {

hw/device-assignment.c +5 -0

@@ -1179,7 +1219,11 @@ static int assigned_dev_update_msix_mmio(PCIDevice
 *pci_dev,
 
             break;
 
         memcpy(&msg_ctrl, va + i * 16 + 12, 4);
         memcpy(&msg_data, va + i * 16 + 8, 4);
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+        if (assigned_dev_msix_entry_masked(adev, i))
+#else
 
         if (msg_data == 0)
 
+#endif
 
             continue;
 
         memcpy(&msg_addr, va + i * 16, 4);

@@ -1200,6 +1244,7 @@ static int assigned_dev_update_msix_mmio(PCIDevice
 *pci_dev,
 
         msix_entry.gsi = adev->entry[entries_nr].gsi;
         msix_entry.entry = i;
 
+        pci_dev->msix_entry_used[i] = 1;

         r = kvm_assign_set_msix_entry(kvm_context, &msix_entry);
         if (r) {
         
             fprintf(stderr, "fail to set MSI-X entry! %s\n",
             strerror(-r));

hw/device-assignment.c +2 -0

@@ -1243,6 +1288,8 @@ static void assigned_dev_update_msix(PCIDevice
 *pci_dev, int enable_msix)
 
             perror("assigned_dev_update_msix: deassign irq");
 
         assigned_dev->irq_requested_type = 0;
 
+        memset(pci_dev->msix_entry_used, 0, KVM_MAX_MSIX_PER_DEV *
+

sizeof(*pci_dev->msix_entry_used));

     }
     
     entries_max_nr = get_msix_entries_max_nr(assigned_dev);

hw/device-assignment.c +6 -0

@@ -1250,10 +1297,16 @@ static void assigned_dev_update_msix(PCIDevice
 *pci_dev, int enable_msix)
 
         fprintf(stderr, "assigned_dev_update_msix: MSI-X entries_max_nr
         == 0"); return;
 
     }
 
+    /*
+     * Guest may try to enable MSI-X before setting MSI-X entry done, so
+     * let's wait until guest unmask the entries.
+     */
 
     entries_nr = get_msix_valid_entries_nr(assigned_dev,
     entries_max_nr); if (entries_nr == 0) {
 
+#ifndef KVM_CAP_DEVICE_MSIX_MASK
 
         if (enable_msix)
 
             fprintf(stderr, "MSI-X entry number is zero!\n");
 
+#endif
 
         return;
 
     }
     if (enable_msix) {

@@ -1297,7 +1350,8 @@ static void

assigned_device_pci_cap_write_config(PCIDevice *pci_dev, uint32_t ad

         if (address <= ctrl_pos && address + len > ctrl_pos) {
         
             ctrl_pos--; /* control is word long */
             ctrl_word = (uint16_t *)(pci_dev->config + ctrl_pos);

- assigned_dev_update_msix(pci_dev, (*ctrl_word &
PCI_MSIX_ENABLE)); + assigned_dev_update_msix(pci_dev,
+ (*ctrl_word & PCI_MSIX_ENABLE) && !(*ctrl_word &
PCI_MSIX_MASK));

hw/device-assignment.c +17 -0

 	}
 
         pos += PCI_CAPABILITY_CONFIG_MSIX_LENGTH;
 
     }

@@ -1395,10 +1449,101 @@ static void msix_mmio_writel(void *opaque,
 
     AssignedDevice *adev = opaque;
     unsigned int offset = addr & 0xfff;
     void *page = adev->msix_table_page;
 
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    int pos, ctrl_word, index;
+    struct kvm_irq_routing_entry new_entry = {};
+    int entry_idx, entries_max_nr, r = 0, i;
+    uint32_t msg_ctrl, msg_data, msg_upper_addr, msg_addr;
+#endif
 
     DEBUG("write to MSI-X entry table mmio offset 0x%lx, val 0x%x\n",
 
 		    addr, val);
 
     memcpy((void *)((char *)page + offset), &val, 4);
 
+
+#ifdef KVM_CAP_DEVICE_MSIX_MASK
+    index = offset / 16;
+
+    /* Check if mask bit is being accessed */
+    memcpy(&msg_addr, (char *)page + index * 16, 4);
+    memcpy(&msg_upper_addr, (char *)page + index * 16 + 4, 4);
+    memcpy(&msg_data, (char *)page + index * 16 + 8, 4);
+    memcpy(&msg_ctrl, (char *)page + index * 16 + 12, 4);
+    DEBUG("MSI-X entries index %d: "
+            "msg_addr 0x%x, msg_upper_addr 0x%x, msg_data 0x%x, vec_ctl

0x%x\n", + index, msg_addr, msg_upper_addr, msg_data,
msg_ctrl); +

hw/device-assignment.c +8 -0

+    if (adev->cap.available & ASSIGNED_DEVICE_CAP_MSI)
+        pos = adev->dev.cap.start + PCI_CAPABILITY_CONFIG_MSI_LENGTH;
+    else
+        pos = adev->dev.cap.start;
+
+    ctrl_word = *(uint16_t *)(adev->dev.config + pos + 2);
+
+    if (!((ctrl_word & PCI_MSIX_ENABLE) && !(ctrl_word &

PCI_MSIX_MASK))) + return;

hw/device-assignment.c +22 -0

+
+    if (!assigned_dev_msix_entry_masked(adev, index)) {
+        if (!adev->dev.msix_entry_used[index]) {
+            DEBUG("Try to modify unenabled MSI-X entry %d's mask. "
+                    "Reenable MSI-X.\n",
+                    index);
+            assigned_dev_update_msix(&adev->dev, 1);
+        }
+        return;
+    }
+
+    if (!adev->dev.msix_entry_used[index])
+        return;
+
+    /*
+     * We're here only because guest want to modify MSI data/addr, and
+     * kernel would filter those writing with mask bit unset.
+     */
+    entries_max_nr = get_msix_entries_max_nr(adev);
+
+    /*
+     * Find the index of routing entry, it can be different from 'index'

if + * empty entry existed in between

hw/device-assignment.c +8 -0

+     */
+    entry_idx = -1;
+    for (i = 0; i <= index; i++) {
+        if (adev->dev.msix_entry_used[i])
+            entry_idx ++;
+    }
+    if (entry_idx >= entries_max_nr || entry_idx == -1) {
+        fprintf(stderr, "msix_mmio_writel: Entry idx %d exceed

limit!\n", + entry_idx);

hw/device-assignment.c +5 -0

+        return;
+    }
+
+    if (!assigned_dev_msix_entry_masked(adev, index)) {
+        fprintf(stderr, "msix_mmio_writel: Trying write to unmasked

entry!\n"); + return;

hw/device-assignment.c +11 -0

+    }
+
+    new_entry.gsi = adev->entry[entry_idx].gsi;
+    new_entry.type = KVM_IRQ_ROUTING_MSI;
+    new_entry.flags = 0;
+    new_entry.u.msi.address_lo = msg_addr;
+    new_entry.u.msi.address_hi = msg_upper_addr;
+    new_entry.u.msi.data = msg_data;
+    if (memcmp(&adev->entry[entry_idx].u.msi, &new_entry.u.msi,
+                sizeof new_entry.u.msi)) {
+        r = kvm_update_routing_entry(&adev->entry[entry_idx],

&new_entry); + if (r) {
+ perror("msix_mmio_writel: kvm_update_routing_entry
failed\n"); + return;

hw/device-assignment.c +13 -0

+        }
+        r = kvm_commit_irq_routes();
+        if (r) {
+            perror("msix_mmio_writel: kvm_commit_irq_routes failed\n");
+            return;
+        }
+    }
+    adev->entry[entry_idx].u.msi.address_lo = msg_addr;
+    adev->entry[entry_idx].u.msi.address_hi = msg_upper_addr;
+    adev->entry[entry_idx].u.msi.data = msg_data;
+#endif
 
 }
 
 static void msix_mmio_writew(void *opaque,

@@ -1436,6 +1581,8 @@ static int
 assigned_dev_register_msix_mmio(AssignedDevice *dev)
 
     memset(dev->msix_table_page, 0, 0x1000);
     dev->mmio_index = cpu_register_io_memory(
 
                         msix_mmio_read, msix_mmio_write, dev);
 
+    dev->dev.msix_entry_used = qemu_mallocz(KVM_MAX_MSIX_PER_DEV *
+                                        sizeof

*dev->dev.msix_entry_used);

     return 0;

}

hw/device-assignment.c +2 -0

@@ -1452,6 +1599,8 @@ static void
 assigned_dev_unregister_msix_mmio(AssignedDevice *dev)
 
                 strerror(errno));
 
     }
     dev->msix_table_page = NULL;
 
+    free(dev->dev.msix_entry_used);
+    dev->dev.msix_entry_used = NULL;
 
 }
 
 static int assigned_initfn(struct PCIDevice *pci_dev)

Michael S. Tsirkin

2010-11-05 08:59 UTC #19 lore raw

Re: [PATCH 3/3] qemu-kvm: device assignment: emulate MSI-X mask bits

Quoted 9 lines · Sheng Yang

Quoted 6 lines · On Thursday 04 November 2010 18:04:08 Michael S. Tsirkin wrote:
Quoted 3 lines · Sheng Yang
This patch emulated MSI-X per vector mask bit on assigned device.
Signed-off-by Sheng Yang <[email protected]>
Also pls update the in-tree header for the new ioctls.
I thought this should be done by maintainer?

No idea. I do this for my patches, no one complained.

Quoted 1 line

I just use one script to sync them everytime. I don't mind to send the diff out.

Just make sure there's no extra cruft which scripts sometimes produce ...