qemu: support event_idx parameter for virtio disk and net devices

In some versions of qemu, both virtio-blk-pci and virtio-net-pci devices can have an event_idx setting that determines some details of event processing. When it is enabled, it "reduces the number of interrupts and exits for the guest". qemu will automatically enable this feature when it is available, but there may be cases where this new feature could actually make performance worse (NB: no such case has been found so far). As a safety switch in case such a situation is encountered in the field, this patch adds a new attribute "event_idx" to the <driver> element of both disk and interface devices. event_idx can be set to "on" (to force event_idx on in case qemu has it disabled by default) or "off" (for force event_idx off). In the case that event_idx support isn't present in qemu, the attribute is ignored (this on the advice of the qemu developer). docs/formatdomain.html.in: document the new flag (marking it as "don't mess with this!" docs/schemas/domain.rng: add event_idx in appropriate places src/conf/domain_conf.[ch]: add event_idx to parser and formatter src/libvirt_private.syms: export virDomainVirtioEventIdx(From|To)String src/qemu/qemu_capabilities.[ch]: detect and report event_idx in disk/net src/qemu/qemu_command.c: add event_idx parameter to qemu commandline when appropriate. tests/qemuxml2argvdata/qemuxml2argv-event_idx.args, tests/qemuxml2argvdata/qemuxml2argv-event_idx.xml, tests/qemuxml2argvtest.c, tests/qemuxml2xmltest.c: test cases for event_idx.
2011-08-13 02:32:45 -04:00 · 2011-08-13 02:32:45 -04:00 · 73c0a148af
parent d17101e011
commit 73c0a148af
12 changed files with 196 additions and 4 deletions
--- a/docs/formatdomain.html.in
+++ b/docs/formatdomain.html.in
@ -904,7 +904,7 @@
    &lt;/disk&gt;
      ...
    &lt;disk type='network'&gt;
-      &lt;driver name="qemu" type="raw" io="threads" ioeventfd="on"/&gt;
+      &lt;driver name="qemu" type="raw" io="threads" ioeventfd="on" event_idx="off"/&gt;
      &lt;source protocol="sheepdog" name="image_name"&gt;
        &lt;host name="hostname" port="7000"/&gt;
      &lt;/source&gt;
@ -1004,6 +1004,19 @@
            <b>In general you should leave this option alone, unless you
            are very certain you know what you are doing.</b>
          </li>
+          <li>
+            The optional <code>event_idx</code> attribute controls
+            some aspects of device event processing. The value can be
+            either 'on' or 'off' - if it is on, it will reduce the
+            number of interupts and exits for the guest. The default
+            is determined by QEMU; usually if the feature is
+            supported, default is on. In case there is a situation
+            where this behavior is suboptimal, this attribute provides
+            a way to force the feature off.
+            <span class="since">Since 0.9.5 (QEMU and KVM only)</span>
+            <b>In general you should leave this option alone, unless you
+            are very certain you know what you are doing.</b>
+          </li>
        </ul>
      </dd>
      <dt><code>boot</code></dt>
@ -1850,7 +1863,7 @@ qemu-kvm -net nic,model=? /dev/null
      &lt;source network='default'/&gt;
      &lt;target dev='vnet1'/&gt;
      &lt;model type='virtio'/&gt;
-      <b>&lt;driver name='vhost' txmode='iothread' ioeventfd='on'/&gt;</b>
+      <b>&lt;driver name='vhost' txmode='iothread' ioeventfd='on' event_idx='off'/&gt;</b>
    &lt;/interface&gt;
  &lt;/devices&gt;
  ...</pre>
@ -1914,6 +1927,20 @@ qemu-kvm -net nic,model=? /dev/null
        on overloaded host it could increase guest I/O latency.
        <span class="since">Since 0.9.3 (QEMU and KVM only)</span><br/><br/>

+        <b>In general you should leave this option alone, unless you
+        are very certain you know what you are doing.</b>
+      </dd>
+      <dt><code>event_idx</code></dt>
+      <dd>
+        The <code>event_idx</code> attribute controls some aspects of
+        device event processing. The value can be either 'on' or 'off'
+        - if it is on, it will reduce the number of interupts and
+        exits for the guest. The default is determined by QEMU;
+        usually if the feature is supported, default is on. In case
+        there is a situation where this behavior is suboptimal, this
+        attribute provides a way to force the feature off.
+        <span class="since">Since 0.9.5 (QEMU and KVM only)</span><br/><br/>
+
        <b>In general you should leave this option alone, unless you
        are very certain you know what you are doing.</b>
      </dd>
--- a/docs/schemas/domain.rng
+++ b/docs/schemas/domain.rng
@ -807,6 +807,9 @@
      <optional>
        <ref name="ioeventfd"/>
      </optional>
+      <optional>
+        <ref name="event_idx"/>
+      </optional>
      <empty/>
    </element>
  </define>
@ -854,6 +857,14 @@
      </choice>
    </attribute>
  </define>
+  <define name="event_idx">
+    <attribute name="event_idx">
+      <choice>
+        <value>on</value>
+        <value>off</value>
+      </choice>
+    </attribute>
+  </define>
  <define name="controller">
    <element name="controller">
      <choice>
@ -1165,6 +1176,9 @@
          <optional>
            <ref name="ioeventfd"/>
          </optional>
+          <optional>
+            <ref name="event_idx"/>
+          </optional>
          <empty/>
        </element>
      </optional>
--- a/src/conf/domain_conf.c
+++ b/src/conf/domain_conf.c
@ -182,6 +182,10 @@ VIR_ENUM_IMPL(virDomainIoEventFd, VIR_DOMAIN_IO_EVENT_FD_LAST,
              "on",
              "off")

+VIR_ENUM_IMPL(virDomainVirtioEventIdx, VIR_DOMAIN_VIRTIO_EVENT_IDX_LAST,
+              "default",
+              "on",
+              "off")

 VIR_ENUM_IMPL(virDomainController, VIR_DOMAIN_CONTROLLER_TYPE_LAST,
              "ide",
@ -2081,6 +2085,7 @@ virDomainDiskDefParseXML(virCapsPtr caps,
    char *error_policy = NULL;
    char *iotag = NULL;
    char *ioeventfd = NULL;
+    char *event_idx = NULL;
    char *devaddr = NULL;
    virStorageEncryptionPtr encryption = NULL;
    char *serial = NULL;
@ -2197,6 +2202,7 @@ virDomainDiskDefParseXML(virCapsPtr caps,
                error_policy = virXMLPropString(cur, "error_policy");
                iotag = virXMLPropString(cur, "io");
                ioeventfd = virXMLPropString(cur, "ioeventfd");
+                event_idx = virXMLPropString(cur, "event_idx");
            } else if (xmlStrEqual(cur->name, BAD_CAST "readonly")) {
                def->readonly = 1;
            } else if (xmlStrEqual(cur->name, BAD_CAST "shareable")) {
@ -2351,6 +2357,24 @@ virDomainDiskDefParseXML(virCapsPtr caps,
        def->ioeventfd=i;
    }

+    if (event_idx) {
+        if (def->bus != VIR_DOMAIN_DISK_BUS_VIRTIO) {
+            virDomainReportError(VIR_ERR_CONFIG_UNSUPPORTED,
+                                 _("disk event_idx mode supported "
+                                   "only for virtio bus"));
+            goto error;
+        }
+
+        int idx;
+        if ((idx = virDomainVirtioEventIdxTypeFromString(event_idx)) <= 0) {
+            virDomainReportError(VIR_ERR_CONFIG_UNSUPPORTED,
+                                 _("unknown disk event_idx mode '%s'"),
+                                 event_idx);
+            goto error;
+        }
+        def->event_idx = idx;
+    }
+
    if (devaddr) {
        if (virDomainParseLegacyDeviceAddress(devaddr,
                                              &def->info.addr.pci) < 0) {
@ -2414,6 +2438,7 @@ cleanup:
    VIR_FREE(error_policy);
    VIR_FREE(iotag);
    VIR_FREE(ioeventfd);
+    VIR_FREE(event_idx);
    VIR_FREE(devaddr);
    VIR_FREE(serial);
    virStorageEncryptionFree(encryption);
@ -2744,6 +2769,7 @@ virDomainNetDefParseXML(virCapsPtr caps,
    char *backend = NULL;
    char *txmode = NULL;
    char *ioeventfd = NULL;
+    char *event_idx = NULL;
    char *filter = NULL;
    char *internal = NULL;
    char *devaddr = NULL;
@ -2835,6 +2861,7 @@ virDomainNetDefParseXML(virCapsPtr caps,
                backend = virXMLPropString(cur, "name");
                txmode = virXMLPropString(cur, "txmode");
                ioeventfd = virXMLPropString(cur, "ioeventfd");
+                event_idx = virXMLPropString(cur, "event_idx");
            } else if (xmlStrEqual (cur->name, BAD_CAST "filterref")) {
                filter = virXMLPropString(cur, "filter");
                VIR_FREE(filterparams);
@ -3075,6 +3102,16 @@ virDomainNetDefParseXML(virCapsPtr caps,
            }
            def->driver.virtio.ioeventfd = i;
        }
+        if (event_idx) {
+            int idx;
+            if ((idx = virDomainVirtioEventIdxTypeFromString(event_idx)) <= 0) {
+                virDomainReportError(VIR_ERR_CONFIG_UNSUPPORTED,
+                                     _("unknown interface event_idx mode '%s'"),
+                                     event_idx);
+                goto error;
+            }
+            def->driver.virtio.event_idx = idx;
+        }
    }

    if (filter != NULL) {
@ -3118,6 +3155,7 @@ cleanup:
    VIR_FREE(backend);
    VIR_FREE(txmode);
    VIR_FREE(ioeventfd);
+    VIR_FREE(event_idx);
    VIR_FREE(filter);
    VIR_FREE(type);
    VIR_FREE(internal);
@ -8576,6 +8614,7 @@ virDomainDiskDefFormat(virBufferPtr buf,
    const char *error_policy = virDomainDiskErrorPolicyTypeToString(def->error_policy);
    const char *iomode = virDomainDiskIoTypeToString(def->iomode);
    const char *ioeventfd = virDomainIoEventFdTypeToString(def->ioeventfd);
+    const char *event_idx = virDomainVirtioEventIdxTypeToString(def->event_idx);

    if (!type) {
        virDomainReportError(VIR_ERR_INTERNAL_ERROR,
@ -8608,7 +8647,7 @@ virDomainDiskDefFormat(virBufferPtr buf,
                      type, device);

    if (def->driverName || def->driverType || def->cachemode ||
-        def->ioeventfd) {
+        def->ioeventfd || def->event_idx) {
        virBufferAsprintf(buf, "      <driver");
        if (def->driverName)
            virBufferAsprintf(buf, " name='%s'", def->driverName);
@ -8622,6 +8661,8 @@ virDomainDiskDefFormat(virBufferPtr buf,
            virBufferAsprintf(buf, " io='%s'", iomode);
        if (def->ioeventfd)
            virBufferAsprintf(buf, " ioeventfd='%s'", ioeventfd);
+        if (def->event_idx)
+            virBufferAsprintf(buf, " event_idx='%s'", event_idx);
        virBufferAsprintf(buf, "/>\n");
    }

@ -8996,6 +9037,10 @@ virDomainNetDefFormat(virBufferPtr buf,
                virBufferAsprintf(buf, " ioeventfd='%s'",
                                  virDomainIoEventFdTypeToString(def->driver.virtio.ioeventfd));
            }
+            if (def->driver.virtio.event_idx) {
+                virBufferAsprintf(buf, " event_idx='%s'",
+                                  virDomainVirtioEventIdxTypeToString(def->driver.virtio.event_idx));
+            }
            virBufferAddLit(buf, "/>\n");
        }
    }
--- a/src/conf/domain_conf.h
+++ b/src/conf/domain_conf.h
@ -209,6 +209,14 @@ enum virDomainIoEventFd {
    VIR_DOMAIN_IO_EVENT_FD_LAST
 };

+enum virDomainVirtioEventIdx {
+    VIR_DOMAIN_VIRTIO_EVENT_IDX_DEFAULT = 0,
+    VIR_DOMAIN_VIRTIO_EVENT_IDX_ON,
+    VIR_DOMAIN_VIRTIO_EVENT_IDX_OFF,
+
+    VIR_DOMAIN_VIRTIO_EVENT_IDX_LAST
+};
+
 /* Stores the virtual disk configuration */
 typedef struct _virDomainDiskDef virDomainDiskDef;
 typedef virDomainDiskDef *virDomainDiskDefPtr;
@ -229,6 +237,7 @@ struct _virDomainDiskDef {
    int bootIndex;
    int iomode;
    int ioeventfd;
+    int event_idx;
    unsigned int readonly : 1;
    unsigned int shared : 1;
    virDomainDeviceInfo info;
@ -377,6 +386,7 @@ struct _virDomainNetDef {
            enum virDomainNetBackendType name; /* which driver backend to use */
            enum virDomainNetVirtioTxModeType txmode;
            enum virDomainIoEventFd ioeventfd;
+            enum virDomainVirtioEventIdx event_idx;
        } virtio;
    } driver;
    union {
@ -1661,6 +1671,7 @@ VIR_ENUM_DECL(virDomainDiskErrorPolicy)
 VIR_ENUM_DECL(virDomainDiskProtocol)
 VIR_ENUM_DECL(virDomainDiskIo)
 VIR_ENUM_DECL(virDomainIoEventFd)
+VIR_ENUM_DECL(virDomainVirtioEventIdx)
 VIR_ENUM_DECL(virDomainController)
 VIR_ENUM_DECL(virDomainControllerModel)
 VIR_ENUM_DECL(virDomainFS)
--- a/src/libvirt_private.syms
+++ b/src/libvirt_private.syms
@ -414,6 +414,8 @@ virDomainVideoDefaultRAM;
 virDomainVideoDefaultType;
 virDomainVideoTypeFromString;
 virDomainVideoTypeToString;
+virDomainVirtioEventIdxTypeFromString;
+virDomainVirtioEventIdxTypeToString;
 virDomainVirtTypeToString;
 virDomainWatchdogActionTypeFromString;
 virDomainWatchdogActionTypeToString;
--- a/src/qemu/qemu_capabilities.c
+++ b/src/qemu/qemu_capabilities.c
@ -123,6 +123,8 @@ VIR_ENUM_IMPL(qemuCaps, QEMU_CAPS_LAST,
              "pci-multifunction", /* 60 */
              "virtio-blk-pci.ioeventfd",
              "sga",
+              "virtio-blk-pci.event_idx",
+              "virtio-net-pci.event_idx",
    );

 struct qemu_feature_flags {
@ -1215,6 +1217,10 @@ qemuCapsParseDeviceStr(const char *str, virBitmapPtr flags)
        qemuCapsSet(flags, QEMU_CAPS_VIRTIO_IOEVENTFD);
    if (strstr(str, "name \"sga\""))
        qemuCapsSet(flags, QEMU_CAPS_SGA);
+    if (strstr(str, "virtio-blk-pci.event_idx"))
+        qemuCapsSet(flags, QEMU_CAPS_VIRTIO_BLK_EVENT_IDX);
+    if (strstr(str, "virtio-net-pci.event_idx"))
+        qemuCapsSet(flags, QEMU_CAPS_VIRTIO_NET_EVENT_IDX);

    return 0;
 }
--- a/src/qemu/qemu_capabilities.h
+++ b/src/qemu/qemu_capabilities.h
@ -98,6 +98,8 @@ enum qemuCapsFlags {
    QEMU_CAPS_PCI_MULTIFUNCTION = 60, /* -device multifunction=on|off */
    QEMU_CAPS_VIRTIO_IOEVENTFD  = 61, /* IOeventFD feature: virtio-{net|blk}-pci.ioeventfd=on/off */
    QEMU_CAPS_SGA               = 62, /* Serial Graphics Adapter */
+    QEMU_CAPS_VIRTIO_BLK_EVENT_IDX = 63, /* virtio-blk-pci.event_idx */
+    QEMU_CAPS_VIRTIO_NET_EVENT_IDX = 64, /* virtio-net-pci.event_idx */

    QEMU_CAPS_LAST,                   /* this must always be the last item */
 };
--- a/src/qemu/qemu_command.c
+++ b/src/qemu/qemu_command.c
@ -1585,6 +1585,11 @@ qemuBuildDriveDevStr(virDomainDiskDefPtr disk,
    case VIR_DOMAIN_DISK_BUS_VIRTIO:
        virBufferAddLit(&opt, "virtio-blk-pci");
        qemuBuildIoEventFdStr(&opt, disk->ioeventfd, qemuCaps);
+        if (disk->event_idx &&
+            qemuCapsGet(qemuCaps, QEMU_CAPS_VIRTIO_BLK_EVENT_IDX)) {
+            virBufferAsprintf(&opt, ",event_idx=%s",
+                              virDomainVirtioEventIdxTypeToString(disk->event_idx));
+        }
        qemuBuildDeviceAddressStr(&opt, &disk->info, qemuCaps);
        break;
    case VIR_DOMAIN_DISK_BUS_USB:
@ -1808,8 +1813,14 @@ qemuBuildNicDevStr(virDomainNetDefPtr net,
            goto error;
        }
    }
-    if (usingVirtio)
+    if (usingVirtio) {
        qemuBuildIoEventFdStr(&buf, net->driver.virtio.ioeventfd, qemuCaps);
+        if (net->driver.virtio.event_idx &&
+            qemuCapsGet(qemuCaps, QEMU_CAPS_VIRTIO_NET_EVENT_IDX)) {
+            virBufferAsprintf(&buf, ",event_idx=%s",
+                              virDomainVirtioEventIdxTypeToString(net->driver.virtio.event_idx));
+        }
+    }
    if (vlan == -1)
        virBufferAsprintf(&buf, ",netdev=host%s", net->info.alias);
    else
--- a/tests/qemuxml2argvdata/qemuxml2argv-event_idx.args
+++ b/tests/qemuxml2argvdata/qemuxml2argv-event_idx.args
@ -0,0 +1,11 @@
+LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test QEMU_AUDIO_DRV=none \
+/usr/bin/qemu -S -M pc-0.13 -m 1024 -smp 1 -nodefaults \
+-monitor unix:/tmp/test-monitor,server,nowait -no-acpi \
+-boot dc -device virtio-serial-pci,id=virtio-serial0,bus=pci.0,addr=0x6 \
+-drive file=/var/lib/libvirt/images/f14.img,if=none,id=drive-virtio-disk0 \
+-device virtio-blk-pci,event_idx=on,bus=pci.0,addr=0x4,drive=drive-virtio-disk0,id=virtio-disk0 \
+-drive file=/var/lib/libvirt/Fedora-14-x86_64-Live-KDE.iso,if=none,media=cdrom,id=drive-ide0-1-0 \
+-device ide-drive,bus=ide.1,unit=0,drive=drive-ide0-1-0,id=ide0-1-0 \
+-device virtio-net-pci,event_idx=off,vlan=0,id=net0,mac=52:54:00:e5:48:58,bus=pci.0,addr=0x3 \
+-net user,vlan=0,name=hostnet0 -serial pty -usb -vnc 127.0.0.1:-809 -std-vga \
+-device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x5
--- a/tests/qemuxml2argvdata/qemuxml2argv-event_idx.xml
+++ b/tests/qemuxml2argvdata/qemuxml2argv-event_idx.xml
@ -0,0 +1,57 @@
+<domain type='qemu'>
+  <name>test</name>
+  <uuid>bba65c0e-c049-934f-b6aa-4e2c0582acdf</uuid>
+  <memory>1048576</memory>
+  <currentMemory>1048576</currentMemory>
+  <vcpu>1</vcpu>
+  <os>
+    <type arch='x86_64' machine='pc-0.13'>hvm</type>
+    <boot dev='cdrom'/>
+    <boot dev='hd'/>
+    <bootmenu enable='yes'/>
+  </os>
+  <clock offset='utc'/>
+  <on_poweroff>destroy</on_poweroff>
+  <on_reboot>restart</on_reboot>
+  <on_crash>restart</on_crash>
+  <devices>
+    <emulator>/usr/bin/qemu</emulator>
+    <disk type='file' device='disk'>
+      <driver name='qemu' type='qcow2' event_idx='on'/>
+      <source file='/var/lib/libvirt/images/f14.img'/>
+      <target dev='vda' bus='virtio'/>
+      <address type='pci' domain='0x0000' bus='0x00' slot='0x04' function='0x0'/>
+    </disk>
+    <disk type='file' device='cdrom'>
+      <driver name='qemu' type='raw'/>
+      <source file='/var/lib/libvirt/Fedora-14-x86_64-Live-KDE.iso'/>
+      <target dev='hdc' bus='ide'/>
+      <readonly/>
+      <address type='drive' controller='0' bus='1' unit='0'/>
+    </disk>
+    <controller type='virtio-serial' index='0'>
+      <address type='pci' domain='0x0000' bus='0x00' slot='0x06' function='0x0'/>
+    </controller>
+    <controller type='ide' index='0'/>
+    <interface type='user'>
+      <mac address='52:54:00:e5:48:58'/>
+      <model type='virtio'/>
+      <driver name='vhost' event_idx='off'/>
+    </interface>
+    <serial type='pty'>
+      <target port='0'/>
+    </serial>
+    <console type='pty'>
+      <target type='serial' port='0'/>
+    </console>
+    <input type='mouse' bus='ps2'/>
+    <graphics type='vnc' port='5091' autoport='no' listen='127.0.0.1'>
+      <listen type='address' address='127.0.0.1'/>
+    </graphics>
+    <video>
+      <model type='vga' vram='9216' heads='1'/>
+      <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/>
+    </video>
+    <memballoon model='virtio'/>
+  </devices>
+</domain>
--- a/tests/qemuxml2argvtest.c
+++ b/tests/qemuxml2argvtest.c
@ -359,6 +359,11 @@ mymain(void)
    DO_TEST("disk-ioeventfd", false,
            QEMU_CAPS_DRIVE, QEMU_CAPS_VIRTIO_IOEVENTFD,
            QEMU_CAPS_VIRTIO_TX_ALG, QEMU_CAPS_DEVICE);
+    DO_TEST("event_idx", false,
+            QEMU_CAPS_DRIVE,
+            QEMU_CAPS_VIRTIO_BLK_EVENT_IDX,
+            QEMU_CAPS_VIRTIO_NET_EVENT_IDX,
+            QEMU_CAPS_DEVICE);

    DO_TEST("graphics-vnc", false, NONE);
    DO_TEST("graphics-vnc-socket", false, NONE);
--- a/tests/qemuxml2xmltest.c
+++ b/tests/qemuxml2xmltest.c
@ -187,6 +187,7 @@ mymain(void)

    DO_TEST("smp");
    DO_TEST("lease");
+    DO_TEST("event_idx");

    /* These tests generate different XML */
    DO_TEST_DIFFERENT("balloon-device-auto");