[libvirt] [PATCH 0/5] Add support for turning off offloading for virtio-net
Ján Tomko (5): conf: split out virtio net driver formatting Add virSwitch to data types conf: remove redundant local variable conf: add options for disabling segment offloading qemu: wire up virtio-net segment offloading options docs/formatdomain.html.in | 27 ++++ docs/schemas/basictypes.rng | 6 + docs/schemas/domaincommon.rng | 25 +++ src/conf/domain_conf.c | 176 ++++++++++++++++----- src/conf/domain_conf.h | 5 + src/qemu/qemu_command.c | 20 +++ .../qemuxml2argv-net-virtio-disable-offloads.args | 8 + .../qemuxml2argv-net-virtio-disable-offloads.xml | 32 ++++ tests/qemuxml2argvtest.c | 2 + tests/qemuxml2xmltest.c | 1 + 10 files changed, 262 insertions(+), 40 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.xml -- 1.8.5.5
Instead of checking upfront if the <driver> element will be needed in a big condition, just format all the attributes into a string and output the <driver> element if the string is not empty. --- src/conf/domain_conf.c | 68 ++++++++++++++++++++++++++++++++------------------ 1 file changed, 44 insertions(+), 24 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index a2a7d92..fcf7fb6 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -16401,6 +16401,41 @@ virDomainActualNetDefFormat(virBufferPtr buf, return 0; } + +static int +virDomainVirtioNetDriverFormat(char **outstr, + virDomainNetDefPtr def) +{ + virBuffer buf = VIR_BUFFER_INITIALIZER; + if (def->driver.virtio.name) { + virBufferAsprintf(&buf, "name='%s' ", + virDomainNetBackendTypeToString(def->driver.virtio.name)); + } + if (def->driver.virtio.txmode) { + virBufferAsprintf(&buf, "txmode='%s' ", + virDomainNetVirtioTxModeTypeToString(def->driver.virtio.txmode)); + } + if (def->driver.virtio.ioeventfd) { + virBufferAsprintf(&buf, "ioeventfd='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.ioeventfd)); + } + if (def->driver.virtio.event_idx) { + virBufferAsprintf(&buf, "event_idx='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.event_idx)); + } + if (def->driver.virtio.queues) + virBufferAsprintf(&buf, "queues='%u' ", def->driver.virtio.queues); + + virBufferTrim(&buf, " ", -1); + + if (virBufferCheckError(&buf) < 0) + return -1; + + *outstr = virBufferContentAndReset(&buf); + return 0; +} + + int virDomainNetDefFormat(virBufferPtr buf, virDomainNetDefPtr def, @@ -16576,30 +16611,15 @@ virDomainNetDefFormat(virBufferPtr buf, if (def->model) { virBufferEscapeString(buf, "<model type='%s'/>\n", def->model); - if (STREQ(def->model, "virtio") && - (def->driver.virtio.name || def->driver.virtio.txmode || - def->driver.virtio.ioeventfd || def->driver.virtio.event_idx || - def->driver.virtio.queues)) { - virBufferAddLit(buf, "<driver"); - if (def->driver.virtio.name) { - virBufferAsprintf(buf, " name='%s'", - virDomainNetBackendTypeToString(def->driver.virtio.name)); - } - if (def->driver.virtio.txmode) { - virBufferAsprintf(buf, " txmode='%s'", - virDomainNetVirtioTxModeTypeToString(def->driver.virtio.txmode)); - } - if (def->driver.virtio.ioeventfd) { - virBufferAsprintf(buf, " ioeventfd='%s'", - virTristateSwitchTypeToString(def->driver.virtio.ioeventfd)); - } - if (def->driver.virtio.event_idx) { - virBufferAsprintf(buf, " event_idx='%s'", - virTristateSwitchTypeToString(def->driver.virtio.event_idx)); - } - if (def->driver.virtio.queues) - virBufferAsprintf(buf, " queues='%u'", def->driver.virtio.queues); - virBufferAddLit(buf, "/>\n"); + if (STREQ(def->model, "virtio")) { + char *str; + + if (virDomainVirtioNetDriverFormat(&str, def) < 0) + return -1; + + if (str) + virBufferAsprintf(buf, "<driver %s/>\n", str); + VIR_FREE(str); } } if (def->filter) { -- 1.8.5.5
On 09/11/2014 07:43 AM, Ján Tomko wrote:
Instead of checking upfront if the <driver> element will be needed in a big condition, just format all the attributes into a string and output the <driver> element if the string is not empty. --- src/conf/domain_conf.c | 68 ++++++++++++++++++++++++++++++++------------------ 1 file changed, 44 insertions(+), 24 deletions(-)
ACK I left some thoughts below, but I think I convinced myself that this was the way to go - just figured I'd keep the thoughts there though. John
diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index a2a7d92..fcf7fb6 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -16401,6 +16401,41 @@ virDomainActualNetDefFormat(virBufferPtr buf, return 0; }
+ +static int +virDomainVirtioNetDriverFormat(char **outstr, + virDomainNetDefPtr def)
Could perhaps have been: static char * virDomainVirtioNetDriverFormat(virDomainNetDefPtr def)
+{ + virBuffer buf = VIR_BUFFER_INITIALIZER; + if (def->driver.virtio.name) { + virBufferAsprintf(&buf, "name='%s' ", + virDomainNetBackendTypeToString(def->driver.virtio.name)); + } + if (def->driver.virtio.txmode) { + virBufferAsprintf(&buf, "txmode='%s' ", + virDomainNetVirtioTxModeTypeToString(def->driver.virtio.txmode)); + } + if (def->driver.virtio.ioeventfd) { + virBufferAsprintf(&buf, "ioeventfd='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.ioeventfd)); + } + if (def->driver.virtio.event_idx) { + virBufferAsprintf(&buf, "event_idx='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.event_idx)); + } + if (def->driver.virtio.queues) + virBufferAsprintf(&buf, "queues='%u' ", def->driver.virtio.queues); + + virBufferTrim(&buf, " ", -1); + + if (virBufferCheckError(&buf) < 0) + return -1;
So an 'error' here is more than likely a memory one (similar to old code), but the only difference here is you're checking and failing because of that; whereas, the old code didn't fail right away... Or at least until the similar error checking was done on the buffer. Hmm. so I guess this is why you took the approach of passing a string address to return your string buffer into. This will just cause a bit faster failure (more than likely) from the
+ + *outstr = virBufferContentAndReset(&buf);
and return virBufferContentAndReset(&buf);
+ return 0; +} + + int virDomainNetDefFormat(virBufferPtr buf, virDomainNetDefPtr def, @@ -16576,30 +16611,15 @@ virDomainNetDefFormat(virBufferPtr buf, if (def->model) { virBufferEscapeString(buf, "<model type='%s'/>\n", def->model); - if (STREQ(def->model, "virtio") && - (def->driver.virtio.name || def->driver.virtio.txmode || - def->driver.virtio.ioeventfd || def->driver.virtio.event_idx || - def->driver.virtio.queues)) { - virBufferAddLit(buf, "<driver"); - if (def->driver.virtio.name) { - virBufferAsprintf(buf, " name='%s'", - virDomainNetBackendTypeToString(def->driver.virtio.name)); - } - if (def->driver.virtio.txmode) { - virBufferAsprintf(buf, " txmode='%s'", - virDomainNetVirtioTxModeTypeToString(def->driver.virtio.txmode)); - } - if (def->driver.virtio.ioeventfd) { - virBufferAsprintf(buf, " ioeventfd='%s'", - virTristateSwitchTypeToString(def->driver.virtio.ioeventfd)); - } - if (def->driver.virtio.event_idx) { - virBufferAsprintf(buf, " event_idx='%s'", - virTristateSwitchTypeToString(def->driver.virtio.event_idx)); - } - if (def->driver.virtio.queues) - virBufferAsprintf(buf, " queues='%u'", def->driver.virtio.queues); - virBufferAddLit(buf, "/>\n"); + if (STREQ(def->model, "virtio")) { + char *str; + + if (virDomainVirtioNetDriverFormat(&str, def) < 0) + return -1; + + if (str) + virBufferAsprintf(buf, "<driver %s/>\n", str);
Of course if you went with return char * option, it'd be if ((str = virDomainVirtioNetDriverFormat(def))) virBufferAsprintf(buf, "<driver %s/>\n", str);
+ VIR_FREE(str); } } if (def->filter) {
On 09/15/2014 11:55 PM, John Ferlan wrote:
On 09/11/2014 07:43 AM, Ján Tomko wrote:
Instead of checking upfront if the <driver> element will be needed in a big condition, just format all the attributes into a string and output the <driver> element if the string is not empty. --- src/conf/domain_conf.c | 68 ++++++++++++++++++++++++++++++++------------------ 1 file changed, 44 insertions(+), 24 deletions(-)
ACK
I left some thoughts below, but I think I convinced myself that this was the way to go - just figured I'd keep the thoughts there though.
John
diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index a2a7d92..fcf7fb6 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -16401,6 +16401,41 @@ virDomainActualNetDefFormat(virBufferPtr buf, return 0; }
+ +static int +virDomainVirtioNetDriverFormat(char **outstr, + virDomainNetDefPtr def)
Could perhaps have been:
static char * virDomainVirtioNetDriverFormat(virDomainNetDefPtr def)
+{ + virBuffer buf = VIR_BUFFER_INITIALIZER; + if (def->driver.virtio.name) { + virBufferAsprintf(&buf, "name='%s' ", + virDomainNetBackendTypeToString(def->driver.virtio.name)); + } + if (def->driver.virtio.txmode) { + virBufferAsprintf(&buf, "txmode='%s' ", + virDomainNetVirtioTxModeTypeToString(def->driver.virtio.txmode)); + } + if (def->driver.virtio.ioeventfd) { + virBufferAsprintf(&buf, "ioeventfd='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.ioeventfd)); + } + if (def->driver.virtio.event_idx) { + virBufferAsprintf(&buf, "event_idx='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.event_idx)); + } + if (def->driver.virtio.queues) + virBufferAsprintf(&buf, "queues='%u' ", def->driver.virtio.queues); + + virBufferTrim(&buf, " ", -1); + + if (virBufferCheckError(&buf) < 0) + return -1;
So an 'error' here is more than likely a memory one (similar to old code), but the only difference here is you're checking and failing because of that; whereas, the old code didn't fail right away... Or at least until the similar error checking was done on the buffer.
Hmm. so I guess this is why you took the approach of passing a string address to return your string buffer into. This will just cause a bit faster failure (more than likely) from the
I wanted to get rid of the giant condition, failing sooner is just a side effect.
+ + *outstr = virBufferContentAndReset(&buf);
and return virBufferContentAndReset(&buf);
+ return 0; +} + + int virDomainNetDefFormat(virBufferPtr buf, virDomainNetDefPtr def, @@ -16576,30 +16611,15 @@ virDomainNetDefFormat(virBufferPtr buf, if (def->model) { virBufferEscapeString(buf, "<model type='%s'/>\n", def->model); - if (STREQ(def->model, "virtio") && - (def->driver.virtio.name || def->driver.virtio.txmode || - def->driver.virtio.ioeventfd || def->driver.virtio.event_idx || - def->driver.virtio.queues)) { - virBufferAddLit(buf, "<driver"); - if (def->driver.virtio.name) { - virBufferAsprintf(buf, " name='%s'", - virDomainNetBackendTypeToString(def->driver.virtio.name)); - } - if (def->driver.virtio.txmode) { - virBufferAsprintf(buf, " txmode='%s'", - virDomainNetVirtioTxModeTypeToString(def->driver.virtio.txmode)); - } - if (def->driver.virtio.ioeventfd) { - virBufferAsprintf(buf, " ioeventfd='%s'", - virTristateSwitchTypeToString(def->driver.virtio.ioeventfd)); - } - if (def->driver.virtio.event_idx) { - virBufferAsprintf(buf, " event_idx='%s'", - virTristateSwitchTypeToString(def->driver.virtio.event_idx)); - } - if (def->driver.virtio.queues) - virBufferAsprintf(buf, " queues='%u'", def->driver.virtio.queues); - virBufferAddLit(buf, "/>\n"); + if (STREQ(def->model, "virtio")) { + char *str; + + if (virDomainVirtioNetDriverFormat(&str, def) < 0) + return -1; + + if (str) + virBufferAsprintf(buf, "<driver %s/>\n", str);
Of course if you went with return char * option, it'd be
if ((str = virDomainVirtioNetDriverFormat(def))) virBufferAsprintf(buf, "<driver %s/>\n", str);
We wouldn't know if str is NULL because the driver has no special options, or if it's because we were out of memory. In the unlikely event of virDomainVirtioNetDriverFormat getting OOM error and the other virBuffer* calls succeeding, the <driver> element would not get formatted, but we wouldn't report an error either. Jan
Just to make this series work until Martin pushes his more complete cleanup: https://www.redhat.com/archives/libvir-list/2014-September/msg00391.html --- docs/schemas/basictypes.rng | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/docs/schemas/basictypes.rng b/docs/schemas/basictypes.rng index 75d5238..3e90262 100644 --- a/docs/schemas/basictypes.rng +++ b/docs/schemas/basictypes.rng @@ -446,4 +446,10 @@ </optional> </define> + <define name="virSwitch"> + <choice> + <value>on</value> + <value>off</value> + </choice> + </define> </grammar> -- 1.8.5.5
On 09/11/2014 07:43 AM, Ján Tomko wrote:
Just to make this series work until Martin pushes his more complete cleanup: https://www.redhat.com/archives/libvir-list/2014-September/msg00391.html --- docs/schemas/basictypes.rng | 6 ++++++ 1 file changed, 6 insertions(+)
Looks like this goes away with Martin's virOnOff John
diff --git a/docs/schemas/basictypes.rng b/docs/schemas/basictypes.rng index 75d5238..3e90262 100644 --- a/docs/schemas/basictypes.rng +++ b/docs/schemas/basictypes.rng @@ -446,4 +446,10 @@ </optional> </define>
+ <define name="virSwitch"> + <choice> + <value>on</value> + <value>off</value> + </choice> + </define> </grammar>
Use just one int variable for all the FromString calls. --- src/conf/domain_conf.c | 27 +++++++++++---------------- 1 file changed, 11 insertions(+), 16 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index fcf7fb6..0fdfa6f 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -6904,7 +6904,7 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, virNWFilterHashTablePtr filterparams = NULL; virDomainActualNetDefPtr actual = NULL; xmlNodePtr oldnode = ctxt->node; - int ret; + int ret, val; if (VIR_ALLOC(def) < 0) return NULL; @@ -7248,13 +7248,12 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, } if (mode != NULL) { - int m; - if ((m = virNetDevMacVLanModeTypeFromString(mode)) < 0) { + if ((val = virNetDevMacVLanModeTypeFromString(mode)) < 0) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", _("Unknown mode has been specified")); goto error; } - def->data.direct.mode = m; + def->data.direct.mode = val; } else { def->data.direct.mode = VIR_NETDEV_MACVLAN_MODE_VEPA; } @@ -7329,31 +7328,28 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, if (def->type != VIR_DOMAIN_NET_TYPE_HOSTDEV && STREQ_NULLABLE(def->model, "virtio")) { if (backend != NULL) { - int name; - if ((name = virDomainNetBackendTypeFromString(backend)) < 0 || - name == VIR_DOMAIN_NET_BACKEND_TYPE_DEFAULT) { + if ((val = virDomainNetBackendTypeFromString(backend)) < 0 || + val == VIR_DOMAIN_NET_BACKEND_TYPE_DEFAULT) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, _("Unknown interface <driver name='%s'> " "has been specified"), backend); goto error; } - def->driver.virtio.name = name; + def->driver.virtio.name = val; } if (txmode != NULL) { - int m; - if ((m = virDomainNetVirtioTxModeTypeFromString(txmode)) < 0 || - m == VIR_DOMAIN_NET_VIRTIO_TX_MODE_DEFAULT) { + if ((val = virDomainNetVirtioTxModeTypeFromString(txmode)) < 0 || + val == VIR_DOMAIN_NET_VIRTIO_TX_MODE_DEFAULT) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, _("Unknown interface <driver txmode='%s'> " "has been specified"), txmode); goto error; } - def->driver.virtio.txmode = m; + def->driver.virtio.txmode = val; } if (ioeventfd) { - int val; if ((val = virTristateSwitchTypeFromString(ioeventfd)) <= 0) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, _("unknown interface ioeventfd mode '%s'"), @@ -7363,14 +7359,13 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, def->driver.virtio.ioeventfd = val; } if (event_idx) { - int idx; - if ((idx = virTristateSwitchTypeFromString(event_idx)) <= 0) { + if ((val = virTristateSwitchTypeFromString(event_idx)) <= 0) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, _("unknown interface event_idx mode '%s'"), event_idx); goto error; } - def->driver.virtio.event_idx = idx; + def->driver.virtio.event_idx = val; } if (queues) { unsigned int q; -- 1.8.5.5
On 09/11/2014 07:43 AM, Ján Tomko wrote:
Use just one int variable for all the FromString calls. --- src/conf/domain_conf.c | 27 +++++++++++---------------- 1 file changed, 11 insertions(+), 16 deletions(-)
ACK
Add the following attributes: csum, gso, guest_tso4, guest_tso6, guest_ecn to the <driver> element of network interface which control the virtio-net device properties of the same names. --- docs/formatdomain.html.in | 27 ++++++++ docs/schemas/domaincommon.rng | 25 +++++++ src/conf/domain_conf.c | 81 ++++++++++++++++++++++ src/conf/domain_conf.h | 5 ++ .../qemuxml2argv-net-virtio-disable-offloads.xml | 32 +++++++++ tests/qemuxml2xmltest.c | 1 + 6 files changed, 171 insertions(+) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.xml diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index a2ea758..5b2758a 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -3847,6 +3847,12 @@ qemu-kvm -net nic,model=? /dev/null <model type='virtio'/> <b><driver name='vhost' txmode='iothread' ioeventfd='on' event_idx='off' queues='5'/></b> </interface> + <interface type='network'> + <source network='default'/> + <target dev='vnet2'/> + <model type='virtio'/> + <b><driver csum='off' gso='off' guest_tso4='off' guest_tso6='off' guest_ecn='off'/></b> + </interface> </devices> ...</pre> @@ -3949,6 +3955,27 @@ qemu-kvm -net nic,model=? /dev/null processor, resulting in much higher throughput. <span class="since">Since 1.0.6 (QEMU and KVM only)</span> </dd> + <dt><code>csum</code></dt> + <dd> + The <code>csum</code> attribute with possible values <code>on</code> + and <code>off</code> controls host-side support for packets with + partial checksum values. + <span class="since">Since 1.2.9 (QEMU 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>segment offloading options</dt> + <dd> + The attributes <code>gso</code>, <code>guest_tso4</code>, + <code>guest_tso6</code>, <code>guest_ecn</code> with possible + values of <code>on</code> and <code>off</code> can be used + to tune segment offloading. + <span class="since">Since 1.2.9 (QEMU 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> </dl> <h5><a name="elementsNICSTargetOverride">Overriding the target element</a></h5> diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index 6ae940a..c5b092f 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -2373,6 +2373,31 @@ <optional> <ref name="event_idx"/> </optional> + <optional> + <attribute name="csum"> + <ref name="virSwitch"/> + </attribute> + </optional> + <optional> + <attribute name="gso"> + <ref name="virSwitch"/> + </attribute> + </optional> + <optional> + <attribute name="guest_tso4"> + <ref name="virSwitch"/> + </attribute> + </optional> + <optional> + <attribute name="guest_tso6"> + <ref name="virSwitch"/> + </attribute> + </optional> + <optional> + <attribute name="guest_ecn"> + <ref name="virSwitch"/> + </attribute> + </optional> </group> </choice> <empty/> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 0fdfa6f..cc67e35 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -6892,6 +6892,11 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, char *ioeventfd = NULL; char *event_idx = NULL; char *queues = NULL; + char *csum = NULL; + char *gso = NULL; + char *guest_tso4 = NULL; + char *guest_tso6 = NULL; + char *guest_ecn = NULL; char *filter = NULL; char *internal = NULL; char *devaddr = NULL; @@ -7015,6 +7020,11 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, ioeventfd = virXMLPropString(cur, "ioeventfd"); event_idx = virXMLPropString(cur, "event_idx"); queues = virXMLPropString(cur, "queues"); + csum = virXMLPropString(cur, "csum"); + gso = virXMLPropString(cur, "gso"); + guest_tso4 = virXMLPropString(cur, "guest_tso4"); + guest_tso6 = virXMLPropString(cur, "guest_tso6"); + guest_ecn = virXMLPropString(cur, "guest_ecn"); } else if (xmlStrEqual(cur->name, BAD_CAST "filterref")) { if (filter) { virReportError(VIR_ERR_XML_ERROR, "%s", @@ -7377,6 +7387,51 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, } def->driver.virtio.queues = q; } + if (csum) { + if ((val = virTristateSwitchTypeFromString(csum)) <= 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("unknown interface csum mode '%s'"), + csum); + goto error; + } + def->driver.virtio.csum = val; + } + if (gso) { + if ((val = virTristateSwitchTypeFromString(gso)) <= 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("unknown interface gso mode '%s'"), + gso); + goto error; + } + def->driver.virtio.gso = val; + } + if (guest_tso4) { + if ((val = virTristateSwitchTypeFromString(guest_tso4)) <= 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("unknown interface guest_tso4 mode '%s'"), + guest_tso4); + goto error; + } + def->driver.virtio.guest_tso4 = val; + } + if (guest_tso6) { + if ((val = virTristateSwitchTypeFromString(guest_tso6)) <= 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("unknown interface guest_tso6 mode '%s'"), + guest_tso6); + goto error; + } + def->driver.virtio.guest_tso6 = val; + } + if (guest_ecn) { + if ((val = virTristateSwitchTypeFromString(guest_ecn)) <= 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("unknown interface guest_ecn mode '%s'"), + guest_ecn); + goto error; + } + def->driver.virtio.guest_ecn = val; + } } def->linkstate = VIR_DOMAIN_NET_INTERFACE_LINK_STATE_DEFAULT; @@ -7434,6 +7489,11 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, VIR_FREE(ioeventfd); VIR_FREE(event_idx); VIR_FREE(queues); + VIR_FREE(csum); + VIR_FREE(gso); + VIR_FREE(guest_tso4); + VIR_FREE(guest_tso6); + VIR_FREE(guest_ecn); VIR_FREE(filter); VIR_FREE(type); VIR_FREE(internal); @@ -16421,6 +16481,27 @@ virDomainVirtioNetDriverFormat(char **outstr, if (def->driver.virtio.queues) virBufferAsprintf(&buf, "queues='%u' ", def->driver.virtio.queues); + if (def->driver.virtio.csum) { + virBufferAsprintf(&buf, "csum='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.csum)); + } + if (def->driver.virtio.gso) { + virBufferAsprintf(&buf, "gso='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.gso)); + } + if (def->driver.virtio.guest_tso4) { + virBufferAsprintf(&buf, "guest_tso4='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.guest_tso4)); + } + if (def->driver.virtio.guest_tso6) { + virBufferAsprintf(&buf, "guest_tso6='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.guest_tso6)); + } + if (def->driver.virtio.guest_ecn) { + virBufferAsprintf(&buf, "guest_ecn='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.guest_ecn)); + } + virBufferTrim(&buf, " ", -1); if (virBufferCheckError(&buf) < 0) diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index efae2f5..99a6aa9 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -895,6 +895,11 @@ struct _virDomainNetDef { virTristateSwitch ioeventfd; virTristateSwitch event_idx; unsigned int queues; /* Multiqueue virtio-net */ + virTristateSwitch csum; + virTristateSwitch gso; + virTristateSwitch guest_tso4; + virTristateSwitch guest_tso6; + virTristateSwitch guest_ecn; } virtio; } driver; union { diff --git a/tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.xml b/tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.xml new file mode 100644 index 0000000..5603604 --- /dev/null +++ b/tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.xml @@ -0,0 +1,32 @@ +<domain type='qemu'> + <name>QEMUGuest1</name> + <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid> + <memory unit='KiB'>219136</memory> + <currentMemory unit='KiB'>219136</currentMemory> + <vcpu placement='static'>1</vcpu> + <os> + <type arch='i686' machine='pc'>hvm</type> + <boot dev='hd'/> + </os> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <emulator>/usr/bin/qemu</emulator> + <disk type='block' device='disk'> + <source dev='/dev/HostVG/QEMUGuest7'/> + <target dev='hda' bus='ide'/> + <address type='drive' controller='0' bus='0' target='0' unit='0'/> + </disk> + <controller type='usb' index='0'/> + <controller type='ide' index='0'/> + <controller type='pci' index='0' model='pci-root'/> + <interface type='user'> + <mac address='00:22:44:66:88:aa'/> + <model type='virtio'/> + <driver csum='off' gso='off' guest_tso4='off' guest_tso6='off' guest_ecn='off'/> + </interface> + <memballoon model='virtio'/> + </devices> +</domain> diff --git a/tests/qemuxml2xmltest.c b/tests/qemuxml2xmltest.c index 34cdb97..183c13f 100644 --- a/tests/qemuxml2xmltest.c +++ b/tests/qemuxml2xmltest.c @@ -260,6 +260,7 @@ mymain(void) DO_TEST("net-user"); DO_TEST("net-virtio"); DO_TEST("net-virtio-device"); + DO_TEST("net-virtio-disable-offloads"); DO_TEST("net-eth"); DO_TEST("net-eth-ifname"); DO_TEST("net-virtio-network-portgroup"); -- 1.8.5.5
On 09/11/2014 07:43 AM, Ján Tomko wrote:
Add the following attributes: csum, gso, guest_tso4, guest_tso6, guest_ecn to the <driver> element of network interface which control the virtio-net device properties of the same names. --- docs/formatdomain.html.in | 27 ++++++++ docs/schemas/domaincommon.rng | 25 +++++++ src/conf/domain_conf.c | 81 ++++++++++++++++++++++ src/conf/domain_conf.h | 5 ++ .../qemuxml2argv-net-virtio-disable-offloads.xml | 32 +++++++++ tests/qemuxml2xmltest.c | 1 + 6 files changed, 171 insertions(+) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.xml
diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index a2ea758..5b2758a 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -3847,6 +3847,12 @@ qemu-kvm -net nic,model=? /dev/null <model type='virtio'/> <b><driver name='vhost' txmode='iothread' ioeventfd='on' event_idx='off' queues='5'/></b> </interface> + <interface type='network'> + <source network='default'/> + <target dev='vnet2'/> + <model type='virtio'/> + <b><driver csum='off' gso='off' guest_tso4='off' guest_tso6='off' guest_ecn='off'/></b> + </interface>
This doesn't require a driver name='' value?
</devices> ...</pre>
@@ -3949,6 +3955,27 @@ qemu-kvm -net nic,model=? /dev/null processor, resulting in much higher throughput. <span class="since">Since 1.0.6 (QEMU and KVM only)</span> </dd> + <dt><code>csum</code></dt> + <dd> + The <code>csum</code> attribute with possible values <code>on</code> + and <code>off</code> controls host-side support for packets with + partial checksum values. + <span class="since">Since 1.2.9 (QEMU 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>segment offloading options</dt> + <dd> + The attributes <code>gso</code>, <code>guest_tso4</code>, + <code>guest_tso6</code>, <code>guest_ecn</code> with possible + values of <code>on</code> and <code>off</code> can be used + to tune segment offloading. + <span class="since">Since 1.2.9 (QEMU 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>
s/this option/these options/ [oh - I'm having a flashback to something similar I had to do at my former employer with their virtualization switch software... these got enabled by some application and caused shall we say significant performance issues, especially for older drivers. That particular software was loaded/started after the virtualization network switch and thus reset what our code had done... Bugger to find as it embedded in some output from some network command that was executed rarely in our vswitch network daemon layer] Anyway, I understand it's desired to not say much about them, but is there any need to say what the defaults are? Furthermore, does one domain's setting affect other domains? I guess my curiosity is more is this a domain function or an interface (host) setting. We may want to indicate that here... Is the difference dependent upon the driver name? I know from my previous experience it had a wider affect, but that code had a very different implementation. The vswitch was separate from the guest as a host process to which guests could configure ports. The vswitch software had the configuration magic to the lower level network driver(s) which is where the tso/cko, etc. were managed in the kernel. The equivalent of 'em0', 'eth0', etc - the physical NIC. A vswitch was "tied" to a particular physical NIC. So any changes to the pNIC cast a wide net, which is why "other" software setting TSO/CKO options on the same pNIC our vswitch used after our code disabled it caused all sorts of issues. (Just so you understand why I'm asking the question).
+ </dd> </dl>
<h5><a name="elementsNICSTargetOverride">Overriding the target element</a></h5> diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index 6ae940a..c5b092f 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -2373,6 +2373,31 @@ <optional> <ref name="event_idx"/> </optional> + <optional> + <attribute name="csum"> + <ref name="virSwitch"/> + </attribute> + </optional> + <optional> + <attribute name="gso"> + <ref name="virSwitch"/> + </attribute> + </optional> + <optional> + <attribute name="guest_tso4"> + <ref name="virSwitch"/> + </attribute> + </optional> + <optional> + <attribute name="guest_tso6"> + <ref name="virSwitch"/> + </attribute> + </optional> + <optional> + <attribute name="guest_ecn"> + <ref name="virSwitch"/> + </attribute> + </optional>
s/virSwitch/virOnOff/g The rest looks fine... ACK with the nits fixed... John
</group> </choice> <empty/> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 0fdfa6f..cc67e35 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -6892,6 +6892,11 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, char *ioeventfd = NULL; char *event_idx = NULL; char *queues = NULL; + char *csum = NULL; + char *gso = NULL; + char *guest_tso4 = NULL; + char *guest_tso6 = NULL; + char *guest_ecn = NULL; char *filter = NULL; char *internal = NULL; char *devaddr = NULL; @@ -7015,6 +7020,11 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, ioeventfd = virXMLPropString(cur, "ioeventfd"); event_idx = virXMLPropString(cur, "event_idx"); queues = virXMLPropString(cur, "queues"); + csum = virXMLPropString(cur, "csum"); + gso = virXMLPropString(cur, "gso"); + guest_tso4 = virXMLPropString(cur, "guest_tso4"); + guest_tso6 = virXMLPropString(cur, "guest_tso6"); + guest_ecn = virXMLPropString(cur, "guest_ecn"); } else if (xmlStrEqual(cur->name, BAD_CAST "filterref")) { if (filter) { virReportError(VIR_ERR_XML_ERROR, "%s", @@ -7377,6 +7387,51 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, } def->driver.virtio.queues = q; } + if (csum) { + if ((val = virTristateSwitchTypeFromString(csum)) <= 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("unknown interface csum mode '%s'"), + csum); + goto error; + } + def->driver.virtio.csum = val; + } + if (gso) { + if ((val = virTristateSwitchTypeFromString(gso)) <= 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("unknown interface gso mode '%s'"), + gso); + goto error; + } + def->driver.virtio.gso = val; + } + if (guest_tso4) { + if ((val = virTristateSwitchTypeFromString(guest_tso4)) <= 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("unknown interface guest_tso4 mode '%s'"), + guest_tso4); + goto error; + } + def->driver.virtio.guest_tso4 = val; + } + if (guest_tso6) { + if ((val = virTristateSwitchTypeFromString(guest_tso6)) <= 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("unknown interface guest_tso6 mode '%s'"), + guest_tso6); + goto error; + } + def->driver.virtio.guest_tso6 = val; + } + if (guest_ecn) { + if ((val = virTristateSwitchTypeFromString(guest_ecn)) <= 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("unknown interface guest_ecn mode '%s'"), + guest_ecn); + goto error; + } + def->driver.virtio.guest_ecn = val; + } }
def->linkstate = VIR_DOMAIN_NET_INTERFACE_LINK_STATE_DEFAULT; @@ -7434,6 +7489,11 @@ virDomainNetDefParseXML(virDomainXMLOptionPtr xmlopt, VIR_FREE(ioeventfd); VIR_FREE(event_idx); VIR_FREE(queues); + VIR_FREE(csum); + VIR_FREE(gso); + VIR_FREE(guest_tso4); + VIR_FREE(guest_tso6); + VIR_FREE(guest_ecn); VIR_FREE(filter); VIR_FREE(type); VIR_FREE(internal); @@ -16421,6 +16481,27 @@ virDomainVirtioNetDriverFormat(char **outstr, if (def->driver.virtio.queues) virBufferAsprintf(&buf, "queues='%u' ", def->driver.virtio.queues);
+ if (def->driver.virtio.csum) { + virBufferAsprintf(&buf, "csum='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.csum)); + } + if (def->driver.virtio.gso) { + virBufferAsprintf(&buf, "gso='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.gso)); + } + if (def->driver.virtio.guest_tso4) { + virBufferAsprintf(&buf, "guest_tso4='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.guest_tso4)); + } + if (def->driver.virtio.guest_tso6) { + virBufferAsprintf(&buf, "guest_tso6='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.guest_tso6)); + } + if (def->driver.virtio.guest_ecn) { + virBufferAsprintf(&buf, "guest_ecn='%s' ", + virTristateSwitchTypeToString(def->driver.virtio.guest_ecn)); + } + virBufferTrim(&buf, " ", -1);
if (virBufferCheckError(&buf) < 0) diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index efae2f5..99a6aa9 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -895,6 +895,11 @@ struct _virDomainNetDef { virTristateSwitch ioeventfd; virTristateSwitch event_idx; unsigned int queues; /* Multiqueue virtio-net */ + virTristateSwitch csum; + virTristateSwitch gso; + virTristateSwitch guest_tso4; + virTristateSwitch guest_tso6; + virTristateSwitch guest_ecn; } virtio; } driver; union { diff --git a/tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.xml b/tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.xml new file mode 100644 index 0000000..5603604 --- /dev/null +++ b/tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.xml @@ -0,0 +1,32 @@ +<domain type='qemu'> + <name>QEMUGuest1</name> + <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid> + <memory unit='KiB'>219136</memory> + <currentMemory unit='KiB'>219136</currentMemory> + <vcpu placement='static'>1</vcpu> + <os> + <type arch='i686' machine='pc'>hvm</type> + <boot dev='hd'/> + </os> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <emulator>/usr/bin/qemu</emulator> + <disk type='block' device='disk'> + <source dev='/dev/HostVG/QEMUGuest7'/> + <target dev='hda' bus='ide'/> + <address type='drive' controller='0' bus='0' target='0' unit='0'/> + </disk> + <controller type='usb' index='0'/> + <controller type='ide' index='0'/> + <controller type='pci' index='0' model='pci-root'/> + <interface type='user'> + <mac address='00:22:44:66:88:aa'/> + <model type='virtio'/> + <driver csum='off' gso='off' guest_tso4='off' guest_tso6='off' guest_ecn='off'/> + </interface> + <memballoon model='virtio'/> + </devices> +</domain> diff --git a/tests/qemuxml2xmltest.c b/tests/qemuxml2xmltest.c index 34cdb97..183c13f 100644 --- a/tests/qemuxml2xmltest.c +++ b/tests/qemuxml2xmltest.c @@ -260,6 +260,7 @@ mymain(void) DO_TEST("net-user"); DO_TEST("net-virtio"); DO_TEST("net-virtio-device"); + DO_TEST("net-virtio-disable-offloads"); DO_TEST("net-eth"); DO_TEST("net-eth-ifname"); DO_TEST("net-virtio-network-portgroup");
On 09/16/2014 12:24 AM, John Ferlan wrote:
On 09/11/2014 07:43 AM, Ján Tomko wrote:
Add the following attributes: csum, gso, guest_tso4, guest_tso6, guest_ecn to the <driver> element of network interface which control the virtio-net device properties of the same names. --- docs/formatdomain.html.in | 27 ++++++++ docs/schemas/domaincommon.rng | 25 +++++++ src/conf/domain_conf.c | 81 ++++++++++++++++++++++ src/conf/domain_conf.h | 5 ++ .../qemuxml2argv-net-virtio-disable-offloads.xml | 32 +++++++++ tests/qemuxml2xmltest.c | 1 + 6 files changed, 171 insertions(+) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.xml
diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index a2ea758..5b2758a 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -3847,6 +3847,12 @@ qemu-kvm -net nic,model=? /dev/null <model type='virtio'/> <b><driver name='vhost' txmode='iothread' ioeventfd='on' event_idx='off' queues='5'/></b> </interface> + <interface type='network'> + <source network='default'/> + <target dev='vnet2'/> + <model type='virtio'/> + <b><driver csum='off' gso='off' guest_tso4='off' guest_tso6='off' guest_ecn='off'/></b> + </interface>
This doesn't require a driver name='' value?
AFAIK only queues require name='vhost'.
</devices> ...</pre>
@@ -3949,6 +3955,27 @@ qemu-kvm -net nic,model=? /dev/null processor, resulting in much higher throughput. <span class="since">Since 1.0.6 (QEMU and KVM only)</span> </dd> + <dt><code>csum</code></dt> + <dd> + The <code>csum</code> attribute with possible values <code>on</code> + and <code>off</code> controls host-side support for packets with + partial checksum values. + <span class="since">Since 1.2.9 (QEMU 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>segment offloading options</dt> + <dd> + The attributes <code>gso</code>, <code>guest_tso4</code>, + <code>guest_tso6</code>, <code>guest_ecn</code> with possible + values of <code>on</code> and <code>off</code> can be used + to tune segment offloading. + <span class="since">Since 1.2.9 (QEMU 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>
s/this option/these options/
[oh - I'm having a flashback to something similar I had to do at my former employer with their virtualization switch software... these got enabled by some application and caused shall we say significant performance issues, especially for older drivers. That particular software was loaded/started after the virtualization network switch and thus reset what our code had done... Bugger to find as it embedded in some output from some network command that was executed rarely in our vswitch network daemon layer]
Anyway, I understand it's desired to not say much about them, but is there any need to say what the defaults are? Furthermore, does one domain's setting affect other domains? I guess my curiosity is more is this a domain function or an interface (host) setting. We may want to indicate that here... Is the difference dependent upon the driver name?
This is a per (guest) interface setting, I'm not aware of it affecting other interfaces or domains. Perhaps the 'leave this option alone' was too harsh. I'll add the defaults before pushing/sending another version. Jan
I know from my previous experience it had a wider affect, but that code had a very different implementation. The vswitch was separate from the guest as a host process to which guests could configure ports. The vswitch software had the configuration magic to the lower level network driver(s) which is where the tso/cko, etc. were managed in the kernel. The equivalent of 'em0', 'eth0', etc - the physical NIC. A vswitch was "tied" to a particular physical NIC. So any changes to the pNIC cast a wide net, which is why "other" software setting TSO/CKO options on the same pNIC our vswitch used after our code disabled it caused all sorts of issues. (Just so you understand why I'm asking the question).
+ </dd> </dl>
<h5><a name="elementsNICSTargetOverride">Overriding the target element</a></h5>
On 09/11/2014 05:43 AM, Ján Tomko wrote:
Add the following attributes: csum, gso, guest_tso4, guest_tso6, guest_ecn to the <driver> element of network interface which control the virtio-net device properties of the same names. --- docs/formatdomain.html.in | 27 ++++++++ docs/schemas/domaincommon.rng | 25 +++++++ src/conf/domain_conf.c | 81 ++++++++++++++++++++++ src/conf/domain_conf.h | 5 ++ .../qemuxml2argv-net-virtio-disable-offloads.xml | 32 +++++++++ tests/qemuxml2xmltest.c | 1 + 6 files changed, 171 insertions(+) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.xml
diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index a2ea758..5b2758a 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -3847,6 +3847,12 @@ qemu-kvm -net nic,model=? /dev/null <model type='virtio'/> <b><driver name='vhost' txmode='iothread' ioeventfd='on' event_idx='off' queues='5'/></b> </interface> + <interface type='network'> + <source network='default'/> + <target dev='vnet2'/> + <model type='virtio'/> + <b><driver csum='off' gso='off' guest_tso4='off' guest_tso6='off' guest_ecn='off'/></b> + </interface>
Are we stuck with names with underscores in our XML? I'm still not sure if we've come up with the best naming for exposing all these knobs. -- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org
On 09/16/2014 12:30 AM, Eric Blake wrote:
On 09/11/2014 05:43 AM, Ján Tomko wrote:
Add the following attributes: csum, gso, guest_tso4, guest_tso6, guest_ecn to the <driver> element of network interface which control the virtio-net device properties of the same names. --- docs/formatdomain.html.in | 27 ++++++++ docs/schemas/domaincommon.rng | 25 +++++++ src/conf/domain_conf.c | 81 ++++++++++++++++++++++ src/conf/domain_conf.h | 5 ++ .../qemuxml2argv-net-virtio-disable-offloads.xml | 32 +++++++++ tests/qemuxml2xmltest.c | 1 + 6 files changed, 171 insertions(+) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.xml
diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index a2ea758..5b2758a 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -3847,6 +3847,12 @@ qemu-kvm -net nic,model=? /dev/null <model type='virtio'/> <b><driver name='vhost' txmode='iothread' ioeventfd='on' event_idx='off' queues='5'/></b> </interface> + <interface type='network'> + <source network='default'/> + <target dev='vnet2'/> + <model type='virtio'/> + <b><driver csum='off' gso='off' guest_tso4='off' guest_tso6='off' guest_ecn='off'/></b> + </interface>
Are we stuck with names with underscores in our XML? I'm still not sure if we've come up with the best naming for exposing all these knobs.
I'd rather not mix underscores (event_idx) and other word separators in the same element. Alternatively, we could do something like: <driver csum='off' gso='off'> <guest tso4='off' tso6='off' ecn='off'/> </driver> to get rid of the multi-word attributes, but I like the underscores better because they match QEMU arguments. Jan
On Mon, Sep 15, 2014 at 04:30:46PM -0600, Eric Blake wrote:
On 09/11/2014 05:43 AM, Ján Tomko wrote:
Add the following attributes: csum, gso, guest_tso4, guest_tso6, guest_ecn to the <driver> element of network interface which control the virtio-net device properties of the same names. --- docs/formatdomain.html.in | 27 ++++++++ docs/schemas/domaincommon.rng | 25 +++++++ src/conf/domain_conf.c | 81 ++++++++++++++++++++++ src/conf/domain_conf.h | 5 ++ .../qemuxml2argv-net-virtio-disable-offloads.xml | 32 +++++++++ tests/qemuxml2xmltest.c | 1 + 6 files changed, 171 insertions(+) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.xml
diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index a2ea758..5b2758a 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -3847,6 +3847,12 @@ qemu-kvm -net nic,model=? /dev/null <model type='virtio'/> <b><driver name='vhost' txmode='iothread' ioeventfd='on' event_idx='off' queues='5'/></b> </interface> + <interface type='network'> + <source network='default'/> + <target dev='vnet2'/> + <model type='virtio'/> + <b><driver csum='off' gso='off' guest_tso4='off' guest_tso6='off' guest_ecn='off'/></b> + </interface>
Are we stuck with names with underscores in our XML? I'm still not sure if we've come up with the best naming for exposing all these knobs.
I'm not really convinced having a 'guest_' prefix really buys us anything here, since there's no naming clash to avoid. Why don't we just kill the 'guest_' prefixes. NB, remember that precisely matching QEMU naming is a non-goal, we should be designing something that makes sense in general. Regards, Daniel -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
On 09/17/2014 08:57 AM, Daniel P. Berrange wrote:
+ <b><driver csum='off' gso='off' guest_tso4='off' guest_tso6='off' guest_ecn='off'/></b> + </interface>
Are we stuck with names with underscores in our XML? I'm still not sure if we've come up with the best naming for exposing all these knobs.
I'm not really convinced having a 'guest_' prefix really buys us anything here, since there's no naming clash to avoid. Why don't we just kill the 'guest_' prefixes.
NB, remember that precisely matching QEMU naming is a non-goal, we should be designing something that makes sense in general.
I agree; I'd be fine with: <driver csum='off' gso='off' tso4='off' tso6='off' ecn='off'/> with no need for a <guest> sub-structure. Yeah, we'll have to do some glue logic to translate to qemu names, but that's what libvirt is for. -- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org
On 09/17/2014 04:57 PM, Daniel P. Berrange wrote:
On Mon, Sep 15, 2014 at 04:30:46PM -0600, Eric Blake wrote:
On 09/11/2014 05:43 AM, Ján Tomko wrote:
Add the following attributes: csum, gso, guest_tso4, guest_tso6, guest_ecn to the <driver> element of network interface which control the virtio-net device properties of the same names. --- docs/formatdomain.html.in | 27 ++++++++ docs/schemas/domaincommon.rng | 25 +++++++ src/conf/domain_conf.c | 81 ++++++++++++++++++++++ src/conf/domain_conf.h | 5 ++ .../qemuxml2argv-net-virtio-disable-offloads.xml | 32 +++++++++ tests/qemuxml2xmltest.c | 1 + 6 files changed, 171 insertions(+) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.xml
diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index a2ea758..5b2758a 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -3847,6 +3847,12 @@ qemu-kvm -net nic,model=? /dev/null <model type='virtio'/> <b><driver name='vhost' txmode='iothread' ioeventfd='on' event_idx='off' queues='5'/></b> </interface> + <interface type='network'> + <source network='default'/> + <target dev='vnet2'/> + <model type='virtio'/> + <b><driver csum='off' gso='off' guest_tso4='off' guest_tso6='off' guest_ecn='off'/></b> + </interface>
Are we stuck with names with underscores in our XML? I'm still not sure if we've come up with the best naming for exposing all these knobs.
I'm not really convinced having a 'guest_' prefix really buys us anything here, since there's no naming clash to avoid. Why don't we just kill the 'guest_' prefixes.
The clash is in the options I didn't expose: http://git.qemu.org/?p=qemu.git;a=blob;f=include/hw/virtio/virtio-net.h;h=6c... because they weren't requested by the (private :() bug 1139364 Jan
On Wed, Sep 17, 2014 at 05:36:18PM +0200, Ján Tomko wrote:
On 09/17/2014 04:57 PM, Daniel P. Berrange wrote:
On Mon, Sep 15, 2014 at 04:30:46PM -0600, Eric Blake wrote:
On 09/11/2014 05:43 AM, Ján Tomko wrote:
Add the following attributes: csum, gso, guest_tso4, guest_tso6, guest_ecn to the <driver> element of network interface which control the virtio-net device properties of the same names. --- docs/formatdomain.html.in | 27 ++++++++ docs/schemas/domaincommon.rng | 25 +++++++ src/conf/domain_conf.c | 81 ++++++++++++++++++++++ src/conf/domain_conf.h | 5 ++ .../qemuxml2argv-net-virtio-disable-offloads.xml | 32 +++++++++ tests/qemuxml2xmltest.c | 1 + 6 files changed, 171 insertions(+) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.xml
diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index a2ea758..5b2758a 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -3847,6 +3847,12 @@ qemu-kvm -net nic,model=? /dev/null <model type='virtio'/> <b><driver name='vhost' txmode='iothread' ioeventfd='on' event_idx='off' queues='5'/></b> </interface> + <interface type='network'> + <source network='default'/> + <target dev='vnet2'/> + <model type='virtio'/> + <b><driver csum='off' gso='off' guest_tso4='off' guest_tso6='off' guest_ecn='off'/></b> + </interface>
Are we stuck with names with underscores in our XML? I'm still not sure if we've come up with the best naming for exposing all these knobs.
I'm not really convinced having a 'guest_' prefix really buys us anything here, since there's no naming clash to avoid. Why don't we just kill the 'guest_' prefixes.
The clash is in the options I didn't expose: http://git.qemu.org/?p=qemu.git;a=blob;f=include/hw/virtio/virtio-net.h;h=6c...
because they weren't requested by the (private :() bug 1139364
Ah, so this is why you shouldn't take the precise solution requested in a bug too literally, and instead look at the general picture :-) So QEMU exposes alot of stuff: $ qemu-kvm -device virtio-net,? virtio-net-pci.ioeventfd=on/off virtio-net-pci.vectors=uint32 virtio-net-pci.indirect_desc=on/off virtio-net-pci.event_idx=on/off virtio-net-pci.any_layout=on/off virtio-net-pci.csum=on/off virtio-net-pci.guest_csum=on/off virtio-net-pci.gso=on/off virtio-net-pci.guest_tso4=on/off virtio-net-pci.guest_tso6=on/off virtio-net-pci.guest_ecn=on/off virtio-net-pci.guest_ufo=on/off virtio-net-pci.host_tso4=on/off virtio-net-pci.host_tso6=on/off virtio-net-pci.host_ecn=on/off virtio-net-pci.host_ufo=on/off virtio-net-pci.mrg_rxbuf=on/off virtio-net-pci.status=on/off virtio-net-pci.ctrl_vq=on/off virtio-net-pci.ctrl_rx=on/off virtio-net-pci.ctrl_vlan=on/off virtio-net-pci.ctrl_rx_extra=on/off virtio-net-pci.ctrl_mac_addr=on/off virtio-net-pci.ctrl_guest_offloads=on/off virtio-net-pci.mq=on/off virtio-net-pci.mac=macaddr virtio-net-pci.vlan=vlan virtio-net-pci.netdev=netdev virtio-net-pci.bootindex=int32 virtio-net-pci.x-txtimer=uint32 virtio-net-pci.x-txburst=int32 virtio-net-pci.tx=str virtio-net-pci.addr=pci-devfn virtio-net-pci.romfile=str virtio-net-pci.rombar=uint32 virtio-net-pci.multifunction=on/off virtio-net-pci.command_serr_enable=on/off Which to me indicates that Eric's suggestion for sub-elements is a good idea. eg go for: <driver> <guest ..../> <host ..../> </driver> and support the host bits too Regards, Daniel -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
On 09/17/2014 09:52 AM, Daniel P. Berrange wrote:
Ah, so this is why you shouldn't take the precise solution requested in a bug too literally, and instead look at the general picture :-)
So QEMU exposes alot of stuff:
$ qemu-kvm -device virtio-net,?
virtio-net-pci.guest_tso4=on/off virtio-net-pci.guest_tso6=on/off virtio-net-pci.guest_ecn=on/off virtio-net-pci.guest_ufo=on/off virtio-net-pci.host_tso4=on/off virtio-net-pci.host_tso6=on/off virtio-net-pci.host_ecn=on/off virtio-net-pci.host_ufo=on/off
Which to me indicates that Eric's suggestion for sub-elements is a
Wasn't my idea, but Jan's.
good idea. eg go for:
<driver> <guest ..../> <host ..../> </driver>
and support the host bits too
but yes, I could live with that, now that I'm seeing how many more options there are to be exposed. -- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org
--- src/qemu/qemu_command.c | 20 ++++++++++++++++++++ .../qemuxml2argv-net-virtio-disable-offloads.args | 8 ++++++++ tests/qemuxml2argvtest.c | 2 ++ 3 files changed, 30 insertions(+) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.args diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 665a590..0b88a48 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -4430,6 +4430,26 @@ qemuBuildNicDevStr(virDomainDefPtr def, virBufferAsprintf(&buf, ",event_idx=%s", virTristateSwitchTypeToString(net->driver.virtio.event_idx)); } + if (net->driver.virtio.csum) { + virBufferAsprintf(&buf, ",csum=%s", + virTristateSwitchTypeToString(net->driver.virtio.csum)); + } + if (net->driver.virtio.gso) { + virBufferAsprintf(&buf, ",gso=%s", + virTristateSwitchTypeToString(net->driver.virtio.gso)); + } + if (net->driver.virtio.guest_tso4) { + virBufferAsprintf(&buf, ",guest_tso4=%s", + virTristateSwitchTypeToString(net->driver.virtio.guest_tso4)); + } + if (net->driver.virtio.guest_tso6) { + virBufferAsprintf(&buf, ",guest_tso6=%s", + virTristateSwitchTypeToString(net->driver.virtio.guest_tso6)); + } + if (net->driver.virtio.guest_ecn) { + virBufferAsprintf(&buf, ",guest_ecn=%s", + virTristateSwitchTypeToString(net->driver.virtio.guest_ecn)); + } } if (usingVirtio && vhostfdSize > 1) { /* As advised at http://www.linux-kvm.org/page/Multiqueue diff --git a/tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.args b/tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.args new file mode 100644 index 0000000..49e8bf7 --- /dev/null +++ b/tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.args @@ -0,0 +1,8 @@ +LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test QEMU_AUDIO_DRV=none \ +/usr/bin/qemu -S -M pc -m 214 -smp 1 -nographic -nodefconfig -nodefaults \ +-monitor unix:/tmp/test-monitor,server,nowait -no-acpi -boot c -usb \ +-hda /dev/HostVG/QEMUGuest7 \ +-device virtio-net-pci,csum=off,gso=off,guest_tso4=off,guest_tso6=off\ +,guest_ecn=off,vlan=0,id=net0,mac=00:22:44:66:88:aa,bus=pci.0,addr=0x3 \ +-net user,vlan=0,name=hostnet0 \ +-device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x4 diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index 5c28253..9da1214 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -959,6 +959,8 @@ mymain(void) DO_TEST("net-virtio", NONE); DO_TEST("net-virtio-device", QEMU_CAPS_DEVICE, QEMU_CAPS_NODEFCONFIG, QEMU_CAPS_VIRTIO_TX_ALG); + DO_TEST("net-virtio-disable-offloads", + QEMU_CAPS_DEVICE, QEMU_CAPS_NODEFCONFIG); DO_TEST("net-virtio-netdev", QEMU_CAPS_DEVICE, QEMU_CAPS_NETDEV, QEMU_CAPS_NODEFCONFIG); DO_TEST("net-virtio-s390", -- 1.8.5.5
On 09/11/2014 07:43 AM, Ján Tomko wrote:
--- src/qemu/qemu_command.c | 20 ++++++++++++++++++++ .../qemuxml2argv-net-virtio-disable-offloads.args | 8 ++++++++ tests/qemuxml2argvtest.c | 2 ++ 3 files changed, 30 insertions(+) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-disable-offloads.args
Kind of light on the commit message... ACK John
participants (4)
-
Daniel P. Berrange -
Eric Blake -
John Ferlan -
Ján Tomko