- Devel - libvirt List Archives

[libvirt] Coming release plans for libvirt-1.2.10
by Daniel Veillard 27 Oct '14

27 Oct '14

It's already the end of the month and I didn't pushed plans yet (forgot !) for the coming release. We have only until the end of the week to get to next month, o we should freeze this week. There isn't that many commits yet sine 1.2.9, less than 140 and a number of them are the large splitting changes for the hypervisor driver. So a slightly smaller freeze may be just fine, I'm tempted to push the rc1 for 1.2.10 freeze very early on Wednesday and targetting a release out next Monday. Hopefully that works for everyone, thanks, Daniel -- Daniel Veillard | Open Source and Standards, Red Hat veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

2 1

Re: [libvirt] Close the BlockDriverState when guest eject the media
by Weidong Huang 27 Oct '14

27 Oct '14

On 2014/10/20 19:39, Kevin Wolf wrote: > Am 20.10.2014 um 13:27 hat Weidong Huang geschrieben: >> On 2014/10/20 17:41, Kevin Wolf wrote: >> >>> Am 18.10.2014 um 12:02 hat Weidong Huang geschrieben: >>>> Hi ALL: >>>> >>>> There are two ways to eject the cdrom tray. One is by the eject's qmp commmand(eject_device). >>>> The another one is by the guest(bdrv_eject). They have different results. >>> >>> Yes, they are different things. >>> >>> If a guest opens the tray (using bdrv_eject) and then closes it again, >>> with no user interaction in between, the virtual media must still be in >>> the drive and the guest must be able to access the same image again. >>> Calling bdrv_close() in this case would be a bug. >>> >>> The goal of the monitor command "eject" on the other hand is to remove >>> the medium so that the drive is empty. That a device with a closed tray >>> has to be opened for this is only secondary. >> >> >> Thanks for your reply. >> >> There is a problem. >> >> 1. Qemu receive the "eject" command. >> 2. Runs "eject_request_cb" when an eject request is issued from the monitor, the tray >> is closed, and the medium is locked. But the drive is not closed. >> 3. Guest agree with opening tray and qemu will call bdrv_eject to complete. The drive is >> still not close. >> >> So the result of the monitor command "eject" is not to remove the medium in this situation. > > Now I understand, thanks for explaining. > > But I think libvirt can actually work correctly with what qemu offers > today. qemu returns an error if the medium cannot be removed with the > 'eject' command and it only sends an eject request to the guest. > > With this error, libvirt can know that the DEVICE_TRAY_MOVED event > doesn't mean that the medium has removed, but that it needs to issue > another 'eject' command. > > If this isn't implemented correctly in libvirt today, this needs a > libvirt fix rather than a qemu one. > hi all! How about fix it in libvirt? >>>> eject_device: close the BlockDriverState(bdrv_close(bs)) >>>> bdrv_eject: don't close the BlockDriverState, >>>> >>>> This is ambiguous. So libvirt can't handle some situations. >>>> >>>> libvirt send eject qmp command ---> qemu send eject request to guest ---> >>>> guest respond to qemu ---> qemu emit tray_open event to libvirt ---> >>>> libvirt will not send change qmp command if media source is null. So >>>> the media is not be replace to the null. >>> >>> What is the problem that libvirt has with the guest opening the tray? I >>> don't think libvirt should even care about that case. >> >> >> For example, using libvirt to change media by xml below(media source is null): >> <disk type='file' device='cdrom'> >> <driver name='qemu'/> >> <target dev='hdb' bus='ide'/> >> </disk> >> >> libivrt return ok. But media still is in the guest. >> This is confused. > > Kevin > > . >

4 3

[libvirt] [PATCHv4 0/3] Add non-FreeBSD guest support to Bhyve driver.
by Conrad Meyer 26 Oct '14

26 Oct '14

1 3

[libvirt] [PATCH] util: eliminate "use after free" in callers of virNetDevLinkDump
by Laine Stump 26 Oct '14

26 Oct '14

virNetDevLinkDump() gets a message from netlink into "resp", then calls nlmsg_parse() to fill the table "tb" with pointers into resp. It then returns tb to its caller, but not before freeing the buffer at resp. That means that all the callers of virNetDevLinkDump() are examining memory that has already been freed. This can be verified by filling the buffer at resp with garbage prior to freeing it (or, I suppose, just running libvirtd under valgrind) then performing some operation that calls virNetDevLinkDump(). The code has been like this ever since virNetDevLinkDump() was written - the original author didn't notice it, and neither did later additional users of the function. It has only been pure luck (or maybe a lack of heavy load, and/or maybe an allocation algorithm in malloc() that delays re-use of just-freed memory) that has kept this from causing errors, for example when configuring a PCI passthrough or macvtap passthrough network interface. The solution taken in this patch is the simplest - just return resp to the caller along with tb, then have the caller free it after they are finished using the data (pointers) in tb. I alternately could have made a cleaner interface by creating a new struct that put tb and resp together along with a vir*Free() function for it, but this function is only used in a couple places, and I'm not sure there will be additional new uses of virNetDevLinkDump(), so the value of adding a new type, extra APIs, etc. is dubious. --- src/util/virnetdev.c | 26 +++++++++++++++++--------- src/util/virnetdev.h | 2 +- src/util/virnetdevvportprofile.c | 17 ++++++++++++----- 3 files changed, 30 insertions(+), 15 deletions(-) diff --git a/src/util/virnetdev.c b/src/util/virnetdev.c index db5623a..c22a0e6 100644 --- a/src/util/virnetdev.c +++ b/src/util/virnetdev.c @@ -1353,23 +1353,25 @@ static struct nla_policy ifla_vf_policy[IFLA_VF_MAX+1] = { /** * virNetDevLinkDump: * - * @ifname: The name of the interface; only use if ifindex < 0 - * @ifindex: The interface index; may be < 0 if ifname is given - * @nlattr: pointer to a pointer of netlink attributes that will contain - * the results + * @ifname: The name of the interface; only use if ifindex <= 0 + * @ifindex: The interface index; may be <= 0 if ifname is given + * @data: Gets a pointer to the raw data from netlink. + MUST BE FREED BY CALLER! + * @nlattr: Pointer to a pointer of netlink attributes that will contain + * the results * @src_pid: pid used for nl_pid of the local end of the netlink message * (0 == "use getpid()") * @dst_pid: pid of destination nl_pid if the kernel * is not the target of the netlink message but it is to be * sent to another process (0 if sending to the kernel) * - * Get information about an interface given its name or index. + * Get information from netlink about an interface given its name or index. * * Returns 0 on success, -1 on fatal error. */ int virNetDevLinkDump(const char *ifname, int ifindex, - struct nlattr **tb, + void **nlData, struct nlattr **tb, uint32_t src_pid, uint32_t dst_pid) { int rc = -1; @@ -1451,7 +1453,9 @@ virNetDevLinkDump(const char *ifname, int ifindex, rc = 0; cleanup: nlmsg_free(nl_msg); - VIR_FREE(resp); + if (rc < 0) + VIR_FREE(resp); + *nlData = resp; return rc; malformed_resp: @@ -1647,15 +1651,18 @@ virNetDevGetVfConfig(const char *ifname, int vf, virMacAddrPtr mac, int *vlanid) { int rc = -1; + void *nlData = NULL; struct nlattr *tb[IFLA_MAX + 1] = {NULL, }; int ifindex = -1; - rc = virNetDevLinkDump(ifname, ifindex, tb, 0, 0); + rc = virNetDevLinkDump(ifname, ifindex, &nlData, tb, 0, 0); if (rc < 0) - return rc; + goto cleanup; rc = virNetDevParseVfConfig(tb, vf, mac, vlanid); + cleanup: + VIR_FREE(nlData); return rc; } @@ -1798,6 +1805,7 @@ virNetDevRestoreNetConfig(const char *linkdev, int vf, const char *stateDir) int virNetDevLinkDump(const char *ifname ATTRIBUTE_UNUSED, int ifindex ATTRIBUTE_UNUSED, + void **nlData ATTRIBUTE_UNUSED, struct nlattr **tb ATTRIBUTE_UNUSED, uint32_t src_pid ATTRIBUTE_UNUSED, uint32_t dst_pid ATTRIBUTE_UNUSED) diff --git a/src/util/virnetdev.h b/src/util/virnetdev.h index 2a6e67d..5c463ef 100644 --- a/src/util/virnetdev.h +++ b/src/util/virnetdev.h @@ -166,7 +166,7 @@ int virNetDevGetVirtualFunctions(const char *pfname, ATTRIBUTE_NONNULL(4) ATTRIBUTE_RETURN_CHECK; int virNetDevLinkDump(const char *ifname, int ifindex, - struct nlattr **tb, + void **nlData, struct nlattr **tb, uint32_t src_pid, uint32_t dst_pid) ATTRIBUTE_RETURN_CHECK; diff --git a/src/util/virnetdevvportprofile.c b/src/util/virnetdevvportprofile.c index 75679b0..6ee20d3 100644 --- a/src/util/virnetdevvportprofile.c +++ b/src/util/virnetdevvportprofile.c @@ -787,7 +787,8 @@ virNetDevVPortProfileGetNthParent(const char *ifname, int ifindex, unsigned int int *parent_ifindex, char *parent_ifname, unsigned int *nth) { - int rc; + int rc = -1; + void *nlData = NULL; struct nlattr *tb[IFLA_MAX + 1] = { NULL, }; bool end = false; size_t i = 0; @@ -798,7 +799,8 @@ virNetDevVPortProfileGetNthParent(const char *ifname, int ifindex, unsigned int return -1; while (!end && i <= nthParent) { - rc = virNetDevLinkDump(ifname, ifindex, tb, 0, 0); + VIR_FREE(nlData); + rc = virNetDevLinkDump(ifname, ifindex, &nlData, tb, 0, 0); if (rc < 0) break; @@ -807,7 +809,8 @@ virNetDevVPortProfileGetNthParent(const char *ifname, int ifindex, unsigned int IFNAMSIZ)) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("buffer for root interface name is too small")); - return -1; + rc = -1; + goto cleanup; } *parent_ifindex = ifindex; } @@ -824,6 +827,8 @@ virNetDevVPortProfileGetNthParent(const char *ifname, int ifindex, unsigned int *nth = i - 1; + cleanup: + VIR_FREE(nlData); return rc; } @@ -845,6 +850,7 @@ virNetDevVPortProfileOpCommon(const char *ifname, int ifindex, int rc; int src_pid = 0; uint32_t dst_pid = 0; + void *nlData = NULL; struct nlattr *tb[IFLA_MAX + 1] = { NULL, }; int repeats = STATUS_POLL_TIMEOUT_USEC / STATUS_POLL_INTERVL_USEC; uint16_t status = 0; @@ -877,7 +883,8 @@ virNetDevVPortProfileOpCommon(const char *ifname, int ifindex, } while (--repeats >= 0) { - rc = virNetDevLinkDump(NULL, ifindex, tb, src_pid, dst_pid); + VIR_FREE(nlData); + rc = virNetDevLinkDump(NULL, ifindex, &nlData, tb, src_pid, dst_pid); if (rc < 0) goto cleanup; @@ -909,7 +916,7 @@ virNetDevVPortProfileOpCommon(const char *ifname, int ifindex, } cleanup: - + VIR_FREE(nlData); return rc; } -- 1.9.3

2 2

[libvirt] [PATCH v2] tests: Adding SELINUX_LIBS to fix viridentitytest linker bug.
by Julio Faracco 25 Oct '14

25 Oct '14

In a clean build system (Ubuntu 14.04), the viridentitytest failed to compile. Even if all the SELINUX libraries and depedencies are installed. See the error message below: [...] CC viridentitytest.o CCLD viridentitytest /usr/bin/ld: viridentitytest.o: undefined reference to symbol 'security_disable' //lib/x86_64-linux-gnu/libselinux.so.1: error adding symbols: DSO missing from command line collect2: error: ld returned 1 exit status make: *** [viridentitytest] Error 1 Simply adding the variable SELINUX_LIBS in viridentitytest rules of Makefile.am to include SELINUX libraries into viridentitytest solved that compilation issue. A conditional was added to consider when libvirt is using or not SELINUX. Signed-off-by: Julio Faracco <jcfaracco(a)gmail.com> --- tests/Makefile.am | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/tests/Makefile.am b/tests/Makefile.am index 4c3d4ef..363c902 100644 --- a/tests/Makefile.am +++ b/tests/Makefile.am @@ -926,11 +926,13 @@ virstoragetest_LDADD = $(LDADDS) \ viridentitytest_SOURCES = \ viridentitytest.c testutils.h testutils.c -viridentitytest_LDADD = $(LDADDS) if WITH_SELINUX +viridentitytest_LDADD = $(LDADDS) $(SELINUX_LIBS) viridentitytest_DEPENDENCIES = libsecurityselinuxhelper.la \ ../src/libvirt.la -endif WITH_SELINUX +else ! WITH_SELINUX +viridentitytest_LDADD = $(LDADDS) +endif ! WITH_SELINUX viriscsitest_SOURCES = \ viriscsitest.c testutils.h testutils.c -- 1.9.1

2 3

[libvirt] [PATCH] qemu: add capability probing for block-stream
by Shanzhi Yu 24 Oct '14

24 Oct '14

Since block-stream is not supported on qemu-kvm, so libvirt should post more accurate error info when do blockpull with qemu-kvm but not "Command 'block-stream' is not found" Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1140981 Signed-off-by: Shanzhi Yu <shyu(a)redhat.com> --- src/qemu/qemu_capabilities.c | 2 ++ src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_driver.c | 13 +++++-------- 3 files changed, 8 insertions(+), 8 deletions(-) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 81ada48..5674f4f 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -269,6 +269,7 @@ VIR_ENUM_IMPL(virQEMUCaps, QEMU_CAPS_LAST, "splash-timeout", /* 175 */ "iothread", + "block-stream", ); @@ -1426,6 +1427,7 @@ struct virQEMUCapsStringFlags virQEMUCapsCommands[] = { { "query-spice", QEMU_CAPS_SPICE }, { "query-kvm", QEMU_CAPS_KVM }, { "block-commit", QEMU_CAPS_BLOCK_COMMIT }, + { "block-stream", QEMU_CAPS_BLOCK_STREAM }, { "query-vnc", QEMU_CAPS_VNC }, { "drive-mirror", QEMU_CAPS_DRIVE_MIRROR }, { "blockdev-snapshot-sync", QEMU_CAPS_DISK_SNAPSHOT }, diff --git a/src/qemu/qemu_capabilities.h b/src/qemu/qemu_capabilities.h index 0980c00..6701965 100644 --- a/src/qemu/qemu_capabilities.h +++ b/src/qemu/qemu_capabilities.h @@ -216,6 +216,7 @@ typedef enum { QEMU_CAPS_RTC_RESET_REINJECTION = 174, /* rtc-reset-reinjection monitor command */ QEMU_CAPS_SPLASH_TIMEOUT = 175, /* -boot splash-time */ QEMU_CAPS_OBJECT_IOTHREAD = 176, /* -object iothread */ + QEMU_CAPS_BLOCK_STREAM = 177, /* block-stream */ QEMU_CAPS_LAST, /* this must always be the last item */ } virQEMUCapsFlags; diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 73edda3..b38bc91 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -14980,15 +14980,12 @@ qemuDomainBlockJobImpl(virDomainObjPtr vm, virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", _("block jobs not supported with this QEMU binary")); goto cleanup; - } else if (base) { - virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", - _("partial block pull not supported with this " - "QEMU binary")); - goto cleanup; - } else if (mode == BLOCK_JOB_PULL && bandwidth) { + } + + if (mode == BLOCK_JOB_PULL && + !(virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_BLOCK_STREAM))){ virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", - _("setting bandwidth at start of block pull not " - "supported with this QEMU binary")); + _("block pull is not supported with this QEMU binary")); goto cleanup; } -- 1.8.3.1

2 3

[libvirt] [PATCH] vbox: network: make sure driver not be NULL in virRegisterNetworkDriver
by Shanzhi Yu 24 Oct '14

24 Oct '14

libvirtd will report below error if does not make sure driver not be NULL in virRegisterNetworkDriver $ libvirtd 2014-10-24 09:24:36.443+0000: 28876: info : libvirt version: 1.2.10 2014-10-24 09:24:36.443+0000: 28876: error : virRegisterNetworkDriver:549 : driver in virRegisterNetworkDriver must not be NULL 2014-10-24 09:24:36.443+0000: 28876: error : virDriverLoadModule:99 : Failed module registration vboxNetworkRegister Signed-off-by: Shanzhi Yu <shyu(a)redhat.com> --- src/vbox/vbox_driver.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/vbox/vbox_driver.c b/src/vbox/vbox_driver.c index 743a488..ff69069 100644 --- a/src/vbox/vbox_driver.c +++ b/src/vbox/vbox_driver.c @@ -152,7 +152,7 @@ int vboxNetworkRegister(void) if (VBoxCGlueInit(&uVersion) == 0) networkDriver = vboxGetNetworkDriver(uVersion); - if (virRegisterNetworkDriver(networkDriver) < 0) + if ((networkDriver != NULL) && (virRegisterNetworkDriver(networkDriver) < 0)) return -1; return 0; } -- 1.9.3

3 4

[libvirt] [PATCHv3] qemu: NUMA/network tuning shouldn't be supported in session mode
by Erik Skultety 24 Oct '14

24 Oct '14

Tuning NUMA or network interface parameters require root privileges to manage cgroups, thus an attempt to set some of these parameters in session mode on a running domain should be invalid followed by an error. As an example might be memory tuning which raises an error in such case. Following behavior in session mode will be present after applying this patch: Tuning | SET | GET | ----------|---------------|--------| NUMA | shut off only | always | Memory | never | never | Interface | never | always | Resolves https://bugzilla.redhat.com/show_bug.cgi?id=1126762 --- src/qemu/qemu_command.c | 13 ++++++++++++- src/qemu/qemu_driver.c | 35 +++++++++++++++++++++++++---------- 2 files changed, 37 insertions(+), 11 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index eb72451..4c335dc 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -7671,7 +7671,7 @@ qemuBuildCommandLine(virConnectPtr conn, emulator = def->emulator; if (!cfg->privileged) { - /* If we have no cgroups than we can have no tunings that + /* If we have no cgroups then we can have no tunings that * require them */ if (def->mem.hard_limit || def->mem.soft_limit || @@ -7694,6 +7694,17 @@ qemuBuildCommandLine(virConnectPtr conn, _("CPU tuning is not available in session mode")); goto error; } + + virDomainNetDefPtr *nets = def->nets; + virNetDevBandwidthPtr bandwidth = NULL; + size_t nnets = def->nnets; + for (i = 0; i < nnets; i++) { + if ((bandwidth = virDomainNetGetActualBandwidth(nets[i])) != NULL) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Network bandwidth tuning is not available in session mode")); + goto error; + } + } } for (i = 0; i < def->ngraphics; ++i) { diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 6606154..c64d272 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -8974,6 +8974,13 @@ qemuDomainSetNumaParameters(virDomainPtr dom, &persistentDef) < 0) goto cleanup; + if (!cfg->privileged && + flags & VIR_DOMAIN_AFFECT_LIVE) { + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, "%s", + _("NUMA tuning is not available in session mode")); + goto cleanup; + } + if (flags & VIR_DOMAIN_AFFECT_LIVE) { if (!virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_CPUSET)) { virReportError(VIR_ERR_OPERATION_INVALID, "%s", @@ -9058,6 +9065,7 @@ qemuDomainGetNumaParameters(virDomainPtr dom, size_t i; virDomainObjPtr vm = NULL; virDomainDefPtr persistentDef = NULL; + virQEMUDriverConfigPtr cfg = NULL; char *nodeset = NULL; int ret = -1; virCapsPtr caps = NULL; @@ -9076,6 +9084,7 @@ qemuDomainGetNumaParameters(virDomainPtr dom, return -1; priv = vm->privateData; + cfg = virQEMUDriverGetConfig(driver); if (virDomainGetNumaParametersEnsureACL(dom->conn, vm->def) < 0) goto cleanup; @@ -9093,14 +9102,6 @@ qemuDomainGetNumaParameters(virDomainPtr dom, goto cleanup; } - if (flags & VIR_DOMAIN_AFFECT_LIVE) { - if (!virCgroupHasController(priv->cgroup, VIR_CGROUP_CONTROLLER_MEMORY)) { - virReportError(VIR_ERR_OPERATION_INVALID, - "%s", _("cgroup memory controller is not mounted")); - goto cleanup; - } - } - for (i = 0; i < QEMU_NB_NUMA_PARAM && i < *nparams; i++) { virMemoryParameterPtr param = &params[i]; @@ -9123,9 +9124,16 @@ qemuDomainGetNumaParameters(virDomainPtr dom, if (!nodeset) goto cleanup; } else { - if (virCgroupGetCpusetMems(priv->cgroup, &nodeset) < 0) - goto cleanup; + if (!virCgroupHasController(priv->cgroup, + VIR_CGROUP_CONTROLLER_MEMORY) || + virCgroupGetCpusetMems(priv->cgroup, &nodeset) < 0) { + nodeset = virDomainNumatuneFormatNodeset(vm->def->numatune, + NULL, -1); + if (!nodeset) + goto cleanup; + } } + if (virTypedParameterAssign(param, VIR_DOMAIN_NUMA_NODESET, VIR_TYPED_PARAM_STRING, nodeset) < 0) goto cleanup; @@ -9150,6 +9158,7 @@ qemuDomainGetNumaParameters(virDomainPtr dom, if (vm) virObjectUnlock(vm); virObjectUnref(caps); + virObjectUnref(cfg); return ret; } @@ -10120,6 +10129,12 @@ qemuDomainSetInterfaceParameters(virDomainPtr dom, if (virDomainSetInterfaceParametersEnsureACL(dom->conn, vm->def, flags) < 0) goto cleanup; + if (!cfg->privileged) { + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, "%s", + _("Network bandwidth tuning is not available in session mode")); + goto cleanup; + } + if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto cleanup; -- 1.9.3

3 3

[libvirt] [PATCH 00/12] Split up the libvirt.h.in file
by Daniel P. Berrange 24 Oct '14

24 Oct '14

Similar to the previous series, this splits up the libvirt.h.in header file to many pieces. This is backwards compatible as applications will continue to use the main <libvirt/libvirt.h> header and are forbidden from including the individual pieces (to allow us to change them again if we want to in the future). Daniel P. Berrange (12): Move virDomainSnapshot related APIs out of libvirt.h.in Move virNetwork related APIs out of libvirt.h.in Move virInterface related APIs out of libvirt.h.in Move virNWFilter related APIs out of libvirt.h.in Move virNodeDevice related APIs out of libvirt.h.in Move virSecret related APIs out of libvirt.h.in Move virStream related APIs out of libvirt.h.in Move virStoragePool/Vol related APIs out of libvirt.h.in Move virEvent related APIs out of libvirt.h.in Move virDomain related APIs out of libvirt.h.in Move virConnect related APIs out of libvirt.h.in Fix preprocessor indentation in libvirt.h.in cfg.mk | 4 +- docs/apibuild.py | 11 + include/libvirt/Makefile.am | 11 + include/libvirt/libvirt-domain-snapshot.h | 212 ++ include/libvirt/libvirt-domain.h | 3496 +++++++++++++++++ include/libvirt/libvirt-event.h | 190 + include/libvirt/libvirt-host.h | 892 +++++ include/libvirt/libvirt-interface.h | 110 + include/libvirt/libvirt-network.h | 337 ++ include/libvirt/libvirt-nodedev.h | 125 + include/libvirt/libvirt-nwfilter.h | 95 + include/libvirt/libvirt-secret.h | 113 + include/libvirt/libvirt-storage.h | 359 ++ include/libvirt/libvirt-stream.h | 152 + include/libvirt/libvirt.h.in | 5837 +---------------------------- 15 files changed, 6160 insertions(+), 5784 deletions(-) create mode 100644 include/libvirt/libvirt-domain-snapshot.h create mode 100644 include/libvirt/libvirt-domain.h create mode 100644 include/libvirt/libvirt-event.h create mode 100644 include/libvirt/libvirt-host.h create mode 100644 include/libvirt/libvirt-interface.h create mode 100644 include/libvirt/libvirt-network.h create mode 100644 include/libvirt/libvirt-nodedev.h create mode 100644 include/libvirt/libvirt-nwfilter.h create mode 100644 include/libvirt/libvirt-secret.h create mode 100644 include/libvirt/libvirt-storage.h create mode 100644 include/libvirt/libvirt-stream.h -- 2.1.0

2 25

[libvirt] [RFC] Add support for NIC offload discovery
by Chapman, James P 24 Oct '14

24 Oct '14

Currently libvirt provides lots of useful information on the HW capabilities of the host and its attached devices. I'm interested in extending the HW capability discovery of libvirt, to a point where it is capable of discovering the HW offload capabilities of host PCI devices. Take a NIC for example, this feature would enable libvirt discover the supported NIC offload features, csum, tso, etc. So where might a feature like this be used. Consider an Openstack deployment, a VM may have a workload that depends on specific NIC HW offload capabilities in order to stay within certain throughput, CPU utilisation or latency thresholds. For this scenario, it's important that the VM is placed on a host with PCI devices that provide the required NIC HW offload capabilities. If libvirt could provide this functionality, this information could be used during the VM scheduling phase when all nodes are evaluated for their suitability. >From what I understand, libvirt currently uses ioctl's to get the NIC state, MTU, etc. Could the ethtool ioctl framework be used to discover the HW offload features provided by the hosts network controllers. E.g. * ETHTOOL_GRXCSUM * ETHTOOL_GTXCSUM * ETHTOOL_GTSO * ETHTOOL_GFLAGS So I'm looking for your opinions on this feature proposal, does it make sense, is it already being worked on. Thanks James

2 1