- Devel - libvirt List Archives

[libvirt] [PATCH] build: fix building error when building without libvirtd
by Wen Congyang 20 Jun '11

20 Jun '11

When building libvirt without libvirtd, we will receive the following error message: make[3]: Entering directory `/home/wency/rpmbuild/BUILD/libvirt-0.9.2/tools' CC virsh-virsh.o CC virsh-console.o GEN virt-xml-validate GEN virt-pki-validate CCLD virsh ../src/.libs/libvirt.so: undefined reference to `numa_available' ../src/.libs/libvirt.so: undefined reference to `numa_max_node' collect2: ld returned 1 exit status The reason is that: we check numactl only when building qemu driver, and qemu driver will not be built when bulding without libvirtd. So with_numactl's value is check and we will not link libnuma.so. In the other function, we call numa_available() and numa_max_node() only when HAVE_NUMACTL is 1. We should do the same check in the function nodeGetMemoryStats(). --- src/nodeinfo.c | 6 ++++++ 1 files changed, 6 insertions(+), 0 deletions(-) diff --git a/src/nodeinfo.c b/src/nodeinfo.c index 647cb1e..c6dbb84 100644 --- a/src/nodeinfo.c +++ b/src/nodeinfo.c @@ -685,17 +685,23 @@ int nodeGetMemoryStats(virConnectPtr conn ATTRIBUTE_UNUSED, return -1; } } else { +#if HAVE_NUMACTL if (numa_available() < 0) { +#endif nodeReportError(VIR_ERR_NO_SUPPORT, "%s", _("NUMA not supported on this host")); return -1; +#if HAVE_NUMACTL } +#endif +#if HAVE_NUMACTL if (cellNum > numa_max_node()) { nodeReportError(VIR_ERR_INVALID_ARG, "%s", _("Invalid cell number")); return -1; } +#endif if (virAsprintf(&meminfo_path, "%s/node%d/meminfo", NODE_SYS_PATH, cellNum) < 0) { -- 1.7.1

2 2

[libvirt] [PATCH v3 0/3] add support for changing blkio parameters for inactive domains
by Hu Tao 20 Jun '11

20 Jun '11

This series enables user to change blkio parameters for inactive domains from virsh command line. CHANGES: v2-v3: - based on new macros VIR_DOMAIN_AFFECT_XXX Hu Tao (3): Add new parameters for blkiotune update qemuDomainGetBlkioParameters to use flags Update qemuDomainSetBlkioParameters to use flags src/qemu/qemu_driver.c | 250 ++++++++++++++++++++++++++++++++++-------------- tools/virsh.c | 26 +++++- tools/virsh.pod | 7 ++ 3 files changed, 206 insertions(+), 77 deletions(-) -- 1.7.3.1

3 8

[libvirt] Schedule for 0.9.3 release
by Daniel Veillard 20 Jun '11

20 Jun '11

As stated before I really want the next release schedule to be aligned with month end, so I would like to get 0.9.3 out around Friday 1st July, which means that there is one week left before we enter the feature freeze. Make sure your patches get reviewed in time this week if they need to make 0.9.3 :-) Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

1 0

[libvirt] [PATCH] apparmor: implement AppArmorSetFDLabel()
by Jamie Strandboge 20 Jun '11

20 Jun '11

During a savevm operation, libvirt will now use fd migration if qemu supports it. When the AppArmor driver is enabled, AppArmorSetFDLabel() is used but since this function simply returns '0', the dynamic AppArmor profile is not updated and AppArmor blocks access to the save file. This patch implements AppArmorSetFDLabel() to get the pathname of the file by resolving the fd symlink in /proc, and then gives that pathname to reload_profile(), which fixes 'virsh save' when AppArmor is enabled. Passes 'check' and 'syntax-check' (though po_check failed for unrelated reasons). Reference: https://launchpad.net/bugs/795800 -- Jamie Strandboge | http://www.canonical.com

3 2

[libvirt] [PATCH v3]: set and restore MAC address of a NIC when using PASSTHROUGH mode
by Gerhard Stenzel 19 Jun '11

19 Jun '11

This is a another rework of the patch Dirk sent out last week taking into account most propsosed changes The following patch addresses the problem that when a PASSTHROUGH mode DIRECT NIC connection is made the MAC address of the NIC is not automatically set and reset to the configured VM MAC and back again. The attached patch fixes this problem by setting and resetting the MAC while remembering the previous setting while the VM is running. This also works if libvirtd is restarted while the VM is running. the patch passes make syntax-check Signed-off-by: Dirk Herrendoerfer <d.herrendoerfer at herrendoerfer.name> Signed-off-by: Gerhard Stenzel <gerhard.stenzel(a)de.ibm.com> --- Index: libvirt/src/libvirt_macvtap.syms =================================================================== --- libvirt.orig/src/libvirt_macvtap.syms +++ libvirt/src/libvirt_macvtap.syms @@ -5,6 +5,8 @@ # macvtap.h delMacvtap; +getMacaddr; openMacvtapTap; +setMacaddr; vpAssociatePortProfileId; vpDisassociatePortProfileId; Index: libvirt/src/qemu/qemu_command.c =================================================================== --- libvirt.orig/src/qemu/qemu_command.c +++ libvirt/src/qemu/qemu_command.c @@ -128,7 +128,7 @@ qemuPhysIfaceConnect(virDomainDefPtr def rc = openMacvtapTap(net->ifname, net->mac, net->data.direct.linkdev, net->data.direct.mode, vnet_hdr, def->uuid, &net->data.direct.virtPortProfile, &res_ifname, - vmop); + vmop, driver->stateDir); if (rc >= 0) { qemuAuditNetDevice(def, net, res_ifname, true); VIR_FREE(net->ifname); @@ -149,7 +149,7 @@ qemuPhysIfaceConnect(virDomainDefPtr def if (err) { VIR_FORCE_CLOSE(rc); delMacvtap(net->ifname, net->mac, net->data.direct.linkdev, - &net->data.direct.virtPortProfile); + net->data.direct.mode, &net->data.direct.virtPortProfile, driver->stateDir); VIR_FREE(net->ifname); } } Index: libvirt/src/qemu/qemu_process.c =================================================================== --- libvirt.orig/src/qemu/qemu_process.c +++ libvirt/src/qemu/qemu_process.c @@ -2707,7 +2707,8 @@ void qemuProcessStop(struct qemud_driver virDomainNetDefPtr net = def->nets[i]; if (net->type == VIR_DOMAIN_NET_TYPE_DIRECT) { delMacvtap(net->ifname, net->mac, net->data.direct.linkdev, - &net->data.direct.virtPortProfile); + net->data.direct.mode, + &net->data.direct.virtPortProfile, driver->stateDir); VIR_FREE(net->ifname); } } Index: libvirt/src/util/macvtap.c =================================================================== --- libvirt.orig/src/util/macvtap.c +++ libvirt/src/util/macvtap.c @@ -87,6 +87,7 @@ # define LLDPAD_PID_FILE "/var/run/lldpad.pid" +#define MACADDRSIZE 6 enum virVirtualPortOp { ASSOCIATE = 0x1, @@ -191,6 +192,149 @@ err_exit: # if WITH_MACVTAP +/** + * getMacaddr: + * Get the MAC address of a network device + * + * @macaddress: Pointer where the MAC address will be stored + * @srcdev: The interface name of the NIC to get the MAC from + * + * Returns zero in case of success, + * negative value otherwise with error reported. + * + */ +int +getMacaddr(const unsigned char *macaddress, const char *srcdev ) +{ + int sockfd; + int io; + struct ifreq ifr; + + strcpy(ifr.ifr_name, srcdev); + + sockfd = socket(AF_INET, SOCK_STREAM, 0); + if(sockfd < 0){ + return -1; + } + + io = ioctl(sockfd, SIOCGIFHWADDR, (char *)&ifr); + if(io < 0){ + return -1; + } + + memcpy(macaddress, ifr.ifr_ifru.ifru_hwaddr.sa_data, MACADDRSIZE); + + return 0; +} + +/** + * setMacaddr: + * Set the MAC address of a network device + * + * @macaddress: MAC address to assign to the NIC + * @srcdev: The interface name of the NIC + * + * Returns zero in case of success, + * negative value otherwise with error reported. + * + */ +int +setMacaddr(const unsigned char *macaddress, const char *srcdev ) +{ + int rc = 0; + struct nlmsghdr *resp; + struct nlmsgerr *err; + struct ifinfomsg ifinfo = { .ifi_family = AF_UNSPEC }; + int ifindex; + unsigned char *recvbuf = NULL; + unsigned int recvbuflen; + struct nl_msg *nl_msg; + + if (ifaceGetIndex(true, srcdev, &ifindex) != 0) + return -1; + + nl_msg = nlmsg_alloc_simple(RTM_SETLINK, NLM_F_REQUEST); + + if (!nl_msg) { + virReportOOMError(); + return -1; + } + + if (nlmsg_append(nl_msg, &ifinfo, sizeof(ifinfo), NLMSG_ALIGNTO) < 0) + goto buffer_too_small; + + if (nla_put_u32(nl_msg, IFLA_LINK, ifindex) < 0) + goto buffer_too_small; + + if (nla_put(nl_msg, IFLA_ADDRESS, MACADDRSIZE, macaddress) < 0) + goto buffer_too_small; + + if (srcdev && + nla_put(nl_msg, IFLA_IFNAME, strlen(srcdev)+1, srcdev) < 0) + goto buffer_too_small; + + if (nlComm(nl_msg, &recvbuf, &recvbuflen, 0) < 0) { + rc = -1; + goto err_exit; + } + + if (recvbuflen < NLMSG_LENGTH(0) || recvbuf == NULL) + goto malformed_resp; + + resp = (struct nlmsghdr *)recvbuf; + + switch (resp->nlmsg_type) { + case NLMSG_ERROR: + err = (struct nlmsgerr *)NLMSG_DATA(resp); + if (resp->nlmsg_len < NLMSG_LENGTH(sizeof(*err))) + goto malformed_resp; + + switch (err->error) { + + case 0: + break; + + case -EEXIST: + rc = -1; + break; + + default: + macvtapError(VIR_ERR_INTERNAL_ERROR, "%s", + _("error setting device mac address")); + rc = -1; + } + break; + + case NLMSG_DONE: + break; + + default: + goto malformed_resp; + } + +err_exit: + nlmsg_free(nl_msg); + + VIR_FREE(recvbuf); + + return rc; + +malformed_resp: + nlmsg_free(nl_msg); + + macvtapError(VIR_ERR_INTERNAL_ERROR, "%s", + _("malformed netlink response message")); + VIR_FREE(recvbuf); + return -1; + +buffer_too_small: + nlmsg_free(nl_msg); + + macvtapError(VIR_ERR_INTERNAL_ERROR, "%s", + _("allocated netlink buffer is too small")); + return -1; +} + static int link_add(const char *type, const unsigned char *macaddress, int macaddrsize, @@ -575,7 +719,8 @@ openMacvtapTap(const char *tgifname, const unsigned char *vmuuid, virVirtualPortProfileParamsPtr virtPortProfile, char **res_ifname, - enum virVMOperationType vmOp) + enum virVMOperationType vmOp, + char *stateDir) { const char *type = "macvtap"; int c, rc; @@ -589,6 +734,54 @@ openMacvtapTap(const char *tgifname, VIR_DEBUG("%s: VM OPERATION: %s", __FUNCTION__, virVMOperationTypeToString(vmOp)); + /** Note: When using PASSTHROUGH mode with MACVTAP devices the link + * device's MAC address must be set to the VMs MAC address. In + * order to not confuse the first switch or bridge in line this MAC + * address must be reset when the VM is shut down. + * This is especially important when using SRIOV capable cards that + * emulate their switch in firmware. + */ + if (mode == VIR_DOMAIN_NETDEV_MACVTAP_MODE_PASSTHRU) { + unsigned char oldmac[6]; + + rc = getMacaddr(&oldmac, linkdev); + if (rc) { + virReportSystemError(rc, + _("Getting MAC address from '%s' " + "to '%02x:%02x:%02x:%02x:%02x:%02x' failed."), + linkdev, + oldmac[0], oldmac[1], oldmac[2], + oldmac[3], oldmac[4], oldmac[5]); + } else { + char *path = NULL; + + char macstr[VIR_MAC_STRING_BUFLEN]; + if (virAsprintf(&path, "%s/%s", + stateDir, + linkdev) < 0) { + virReportOOMError(); + return errno; + } + virFormatMacAddr(oldmac, macstr); + if (virFileWriteStr(path, macstr, O_CREAT|O_TRUNC|O_WRONLY) < 0) { + virReportSystemError(errno, _("Unable to preserve mac for %s"), + linkdev); + return errno; + } + } + + rc = setMacaddr(macaddress, linkdev); + if (rc) { + virReportSystemError(errno, + _("Setting MAC address on '%s' to " + "'%02x:%02x:%02x:%02x:%02x:%02x' failed."), + linkdev, + macaddress[0], macaddress[1], macaddress[2], + macaddress[3], macaddress[4], macaddress[5]); + } + } + + if (tgifname) { if(ifaceGetIndex(false, tgifname, &ifindex) == 0) { if (STRPREFIX(tgifname, @@ -684,8 +886,43 @@ void delMacvtap(const char *ifname, const unsigned char *macaddr, const char *linkdev, - virVirtualPortProfileParamsPtr virtPortProfile) -{ + int mode, + virVirtualPortProfileParamsPtr virtPortProfile, + char *stateDir) +{ + if (mode == VIR_DOMAIN_NETDEV_MACVTAP_MODE_PASSTHRU) { + int ret; + char *oldmacname = NULL; + char *macstr = NULL; + char *path = NULL; + unsigned char oldmac[6]; + + if (virAsprintf(&path, "%s/%s", + stateDir, + linkdev) < 0) { + virReportOOMError(); + goto disassociate_exit; + } + + if (virFileReadAll(path, VIR_MAC_STRING_BUFLEN, &macstr) < 0) { + virReportSystemError(errno, _("Unable to preserve mac for %s"), + linkdev); + goto disassociate_exit; + } + + if (virParseMacAddr(macstr, &oldmac[0]) != 0) { + virReportSystemError(VIR_ERR_INTERNAL_ERROR, + _("Cannot parse MAC address from '%s'"), + oldmacname); + } + + /*reset mac and remove file-ignore results*/ + ret = setMacaddr(oldmac, linkdev); + ret = unlink(path); + VIR_FREE(macstr); + } + +disassociate_exit: if (ifname) { vpDisassociatePortProfileId(ifname, macaddr, linkdev, Index: libvirt/src/util/macvtap.h =================================================================== --- libvirt.orig/src/util/macvtap.h +++ libvirt/src/util/macvtap.h @@ -75,6 +75,10 @@ enum virVMOperationType { # include "internal.h" +int getMacaddr(const unsigned char *macaddress, const char *srcdev ); + +int setMacaddr(const unsigned char *macaddress, const char *srcdev ); + int openMacvtapTap(const char *ifname, const unsigned char *macaddress, const char *linkdev, @@ -83,12 +87,15 @@ int openMacvtapTap(const char *ifname, const unsigned char *vmuuid, virVirtualPortProfileParamsPtr virtPortProfile, char **res_ifname, - enum virVMOperationType vmop); + enum virVMOperationType vmop, + char *stateDir); void delMacvtap(const char *ifname, const unsigned char *macaddress, const char *linkdev, - virVirtualPortProfileParamsPtr virtPortProfile); + int mode, + virVirtualPortProfileParamsPtr virtPortProfile, + char *stateDir); int vpAssociatePortProfileId(const char *macvtap_ifname, const unsigned char *macvtap_macaddr, Index: libvirt/src/qemu/qemu_hotplug.c =================================================================== --- libvirt.orig/src/qemu/qemu_hotplug.c +++ libvirt/src/qemu/qemu_hotplug.c @@ -1610,7 +1610,7 @@ int qemuDomainDetachNetDevice(struct qem #if WITH_MACVTAP if (detach->type == VIR_DOMAIN_NET_TYPE_DIRECT) { delMacvtap(detach->ifname, detach->mac, detach->data.direct.linkdev, - &detach->data.direct.virtPortProfile); + detach->data.direct.mode, &detach->data.direct.virtPortProfile, driver->stateDir); VIR_FREE(detach->ifname); } #endif =================================================================== Best regards, Gerhard Stenzel ------------------------------------------------------------------------------------- IBM Deutschland Research & Development GmbH Vorsitzender des Aufsichtsrats: Martin Jetter Geschaeftsfuehrung: Dirk Wittkopp Sitz der Gesellschaft: Boeblingen Registergericht: Amtsgericht Stuttgart, HRB 243294

2 1

[libvirt] [Libvirt-test-API][PATCH] Add a pair of keywords in parser.py for supportting testcases loop
by Nan Zhang 19 Jun '11

19 Jun '11

For example: ----------------- domain:start guestname vm1 domain:save start_loop guestname vm1 domain:restore end_loop 10 guestname vm1 domain:destroy guestname vm1 ----------------- --- parser.py | 46 +++++++++++++++++++++++++++++++++++++++++++++- 1 files changed, 45 insertions(+), 1 deletions(-) diff --git a/parser.py b/parser.py index 5b3ce58..e9cce58 100644 --- a/parser.py +++ b/parser.py @@ -31,13 +31,18 @@ class CaseFileParser(object): """ Parser the case configuration file to generate a data list. """ def __init__(self, casefile=None, debug=False): - """ Initialize the list and optionally parse case file. """ self.list = [[]] self.variables = {} self.missing_variables = [] self.debug = debug self.casefile = casefile self.env = env_parser.Envparser("env.cfg") + self.loop_finish = False + self.loop_start = 0 + self.loop_end = 0 + self.loop_times = 0 + self.loop_list = [] + if casefile: self.parse_file(casefile) @@ -306,6 +311,7 @@ class CaseFileParser(object): def parse(self, fh, list): """ For the testcase name parsing. """ + while True: if self.debug: self.debug_print("the list is", list) @@ -330,6 +336,44 @@ class CaseFileParser(object): self.debug_print("we begin to handle the case", tripped_casename) + if self.loop_finish: + for i in range(len(list)): + self.loop_list.append([]) + + i = 0 + for caselist in list: + for j in range(self.loop_start, self.loop_end): + self.loop_list[i].append(caselist.pop()) + + self.loop_list[i].reverse() + self.debug_print("loop_list is", self.loop_list) + caselist.extend(self.loop_list[i] * self.loop_times) + i += 1 + + self.loop_finish = False + + if len(tripped_caselist) == 2 and \ + tripped_caselist[1] == "start_loop": + for caselist in list: + newdict = {} + newdict[tripped_casename] = {} + caselist.append(newdict) + self.loop_start = len(caselist) - 1 + continue + + if len(tripped_caselist) == 3 and \ + tripped_caselist[1] == "end_loop": + looptimes = tripped_caselist[2] + self.debug_print("looptimes is", looptimes) + self.loop_times = int(looptimes) + self.loop_finish = True + for caselist in list: + newdict = {} + newdict[tripped_casename] = {} + caselist.append(newdict) + self.loop_end = len(caselist) + continue + if len(tripped_caselist) == 3 and \ tripped_caselist[1] == "times": times = tripped_caselist[2] -- 1.7.4.4

3 6

[libvirt] [PATCH]: C# bindings. Added some virStream bindings (not all right now)
by Anthony Fox 18 Jun '11

18 Jun '11

2 2

[libvirt] [PATCH 0/6] Introduce virDomainGetControlInfo API
by Jiri Denemark 16 Jun '11

16 Jun '11

The API can be used to query current state of an interface to VMM used to control a domain. In QEMU world this translates into monitor connection and one can use the API, e.g., to check whether (and for how long) libvirtd is currently waiting for a reply from qemu process. Jiri Denemark (6): Introduce virTimeMs for getting current time in ms Use virTimeMs when appropriate Introduce virDomainGetControlInfo API Wire protocol and remote driver for virDomainGetControlInfo qemu: Implement virDomainGetControlInfo virsh: Add support for virDomainGetControlInfo daemon/remote_generator.pl | 20 +++++++--- include/libvirt/libvirt.h.in | 40 +++++++++++++++++++++ python/generator.py | 1 + python/libvirt-override-api.xml | 6 +++ src/driver.h | 5 +++ src/libvirt.c | 48 ++++++++++++++++++++++++++ src/libvirt_private.syms | 1 + src/libvirt_public.syms | 5 +++ src/qemu/qemu_domain.c | 28 ++++++--------- src/qemu/qemu_domain.h | 2 + src/qemu/qemu_driver.c | 72 ++++++++++++++++++++++++++++++++++----- src/qemu/qemu_migration.c | 31 +++++----------- src/qemu/qemu_process.c | 3 ++ src/remote/remote_driver.c | 1 + src/remote/remote_protocol.x | 14 +++++++- src/remote_protocol-structs | 9 +++++ src/util/event_poll.c | 45 +++++++++--------------- src/util/util.c | 24 +++++++++++++ src/util/util.h | 2 + tools/virsh.c | 66 +++++++++++++++++++++++++++++++++++ tools/virsh.pod | 7 ++++ 21 files changed, 348 insertions(+), 82 deletions(-) -- 1.7.5.3

4 21

[libvirt] [PATCH RFC 0/5] support cpu bandwidth in libvirt
by Wen Congyang 16 Jun '11

16 Jun '11

I need this feature immediately after CFS bandwidth patchset is merged into upstream kernel. So I prepare this patchset, and post it here for reviewing before CFS bandwidth is merged into upstream kernel. quota is an optional attribute specified in microseconds not a percentage of period. TODO: 1. quota should be in the range [1000, 18446744073709551(2^64/1000)] or less than 0. But I only limit it less or equal than 2^64/1000. Does anyone have a better way to limit quota? Wen Congyang (5): cgroup: Implement cpu.cfs_period_us and cpu.cfs_quota_us tuning API Update XML Schema for new entries qemu: Implement period and quota tunable XML configuration and parsing. qemu: Implement cfs_period and cfs_quota's modification doc: Add documentation for new cputune elements period and quota docs/formatdomain.html.in | 19 +++ docs/schemas/domain.rng | 25 ++++- src/conf/domain_conf.c | 20 +++- src/conf/domain_conf.h | 2 + src/libvirt_private.syms | 4 + src/qemu/qemu_cgroup.c | 43 +++++- src/qemu/qemu_driver.c | 162 +++++++++++++++++++---- src/util/cgroup.c | 81 +++++++++++- src/util/cgroup.h | 6 + tests/qemuxml2argvdata/qemuxml2argv-cputune.xml | 2 + 10 files changed, 328 insertions(+), 36 deletions(-)

2 9

[libvirt] [RFC PATCH] Set and reset MAC for PASSTHROUGH mode
by Dirk Herrendoerfer 16 Jun '11

16 Jun '11

there is a problem present in libvirt that when a PASSTHROUGH mode DIRECT NIC connection is made the MAC address of the NIC is not automatically set and reset to the configured VM MAC and back again. The attached patch fixes this problem by setting and resetting the MAC while remembering the previous setting while the VM is running. This also works if libvirtd is restarted while the VM is running. The storing of the previous MAC address is done in a link ( currently in /tmp, but should probably go into /var/run somewhere ). Best regards Dirk

3 5