- Devel - libvirt List Archives

[libvirt] virDomainMemoryStats help
by arnaud.champion＠devatom.fr 04 Nov '10

04 Nov '10

?Hi, I need a little help about "virDomainMemoryStats" method. I'm trying to bind it for csharp, and I don't understand the parameter "stats" of the method. Should this array be filled with desired memory statistics or is it fully returned by the method ? Can anyone make a little C sample code that I can understand ? Regards, Arnaud

1 1

Re: [libvirt] [virt-devel] Idea for a "net-info" command in virsh
by Osier Yang 04 Nov '10

04 Nov '10

于 2010年11月04日 16:19, Daniel Veillard 写道: > On Thu, Nov 04, 2010 at 04:04:31AM -0400, jclift(a)redhat.com wrote: >> Hey Osier, >> >> I'm thinking there would be good value for people, in us adding a "net-info" command to virsh, as it seems to be missing. >> >> Something that displays the details for a virtual network, like this: >> >> Name: default >> UUID: aaaaa-bbbbb-ccccc-ddddd >> State: started >> Name on host: virbr0 >> Persistent: yes >> Autostarted: yes >> >> Guest interfaces attached >> ==================== >> >> Domain Interface UUID >> ****** ************** >> guest1 efef-fgfg-efef-fgfg >> abab-cdcd-abab-cdcd >> guest2 zzyy-aabb-zzyy-aabb >> >> It's just a rough concept idea, but thought it worth asking if it's your kind of thing. Hi Justin, yeah, it's good thought from my point, would like implement it after finishing the patch for virsh output.. :-) >> >> What do you think? :) > > I think this kind of suggestions should go to the upstream list :-) > forward to libvir-list for discussing.. :/ - Osier > Daniel >

1 0

[libvirt] [PATCH] macvtap: convert send / recv function to use libnl
by Stefan Berger 04 Nov '10

04 Nov '10

In a second step I am converting the netlink send/receive function to use libnl. I tested this with 802.1Qbg profiles and my test server and did not see a regression. Caveat: The online documentation of libnl talks about nl_socket_alloc() but the header file provides nl_handle_alloc() -- this could be a hint to a possible problem between libnl versions... http://www.infradead.org/~tgr/libnl/doc/group__socket.html versus http://libnl.sourcearchive.com/documentation/1.1/group__socket_gf903c9ea089… Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com> --- src/util/macvtap.c | 89 ++++++++++++++++++----------------------------------- 1 file changed, 31 insertions(+), 58 deletions(-) Index: libvirt-acl/src/util/macvtap.c =================================================================== --- libvirt-acl.orig/src/util/macvtap.c +++ libvirt-acl/src/util/macvtap.c @@ -82,22 +82,6 @@ enum virVirtualPortOp { }; -static int nlOpen(void) -{ - int fd = socket(AF_NETLINK, SOCK_RAW, NETLINK_ROUTE); - if (fd < 0) - virReportSystemError(errno, - "%s",_("cannot open netlink socket")); - return fd; -} - - -static void nlClose(int fd) -{ - close(fd); -} - - /** * nlComm: * @nlmsg: pointer to netlink message @@ -111,8 +95,8 @@ static void nlClose(int fd) * buffer will be returned. */ static -int nlComm(struct nlmsghdr *nlmsg, - char **respbuf, unsigned int *respbuflen, +int nlComm(struct nl_msg *nl_msg, + unsigned char **respbuf, unsigned int *respbuflen, int nl_pid) { int rc = 0; @@ -121,24 +105,29 @@ int nlComm(struct nlmsghdr *nlmsg, .nl_pid = nl_pid, .nl_groups = 0, }; - int rcvChunkSize = 1024; // expecting less than that - int rcvoffset = 0; ssize_t nbytes; struct timeval tv = { .tv_sec = NETLINK_ACK_TIMEOUT_S, }; fd_set readfds; - int fd = nlOpen(); + int fd; int n; + struct nl_handle *nlhandle = nl_handle_alloc(); + struct nlmsghdr *nlmsg = nlmsg_hdr(nl_msg); - if (fd < 0) + if (!nlhandle) return -1; + if (nl_connect(nlhandle, NETLINK_ROUTE) < 0) { + rc = -1; + goto err_exit; + } + + nlmsg_set_dst(nl_msg, &nladdr); + nlmsg->nlmsg_pid = getpid(); - nlmsg->nlmsg_flags |= NLM_F_ACK; - nbytes = sendto(fd, (void *)nlmsg, nlmsg->nlmsg_len, 0, - (struct sockaddr *)&nladdr, sizeof(nladdr)); + nbytes = nl_send_auto_complete(nlhandle, nl_msg); if (nbytes < 0) { virReportSystemError(errno, "%s", _("cannot send to netlink socket")); @@ -146,6 +135,8 @@ int nlComm(struct nlmsghdr *nlmsg, goto err_exit; } + fd = nl_socket_get_fd(nlhandle); + FD_ZERO(&readfds); FD_SET(fd, &readfds); @@ -161,28 +152,9 @@ int nlComm(struct nlmsghdr *nlmsg, goto err_exit; } - while (1) { - if (VIR_REALLOC_N(*respbuf, rcvoffset+rcvChunkSize) < 0) { - virReportOOMError(); - rc = -1; - goto err_exit; - } - - socklen_t addrlen = sizeof(nladdr); - nbytes = recvfrom(fd, &((*respbuf)[rcvoffset]), rcvChunkSize, 0, - (struct sockaddr *)&nladdr, &addrlen); - if (nbytes < 0) { - if (errno == EAGAIN || errno == EINTR) - continue; - virReportSystemError(errno, "%s", - _("error receiving from netlink socket")); - rc = -1; - goto err_exit; - } - rcvoffset += nbytes; - break; - } - *respbuflen = rcvoffset; + *respbuflen = nl_recv(nlhandle, &nladdr, respbuf, NULL); + if (*respbuflen <= 0) + rc = -1; err_exit: if (rc == -1) { @@ -191,7 +163,8 @@ err_exit: *respbuflen = 0; } - nlClose(fd); + nl_handle_destroy(nlhandle); + return rc; } @@ -211,7 +184,7 @@ link_add(const char *type, struct nlmsgerr *err; struct ifinfomsg ifinfo = { .ifi_family = AF_UNSPEC }; int ifindex; - char *recvbuf = NULL; + unsigned char *recvbuf = NULL; unsigned int recvbuflen; struct nl_msg *nl_msg; struct nlattr *linkinfo, *info_data; @@ -260,7 +233,7 @@ link_add(const char *type, nla_nest_end(nl_msg, linkinfo); - if (nlComm(nlmsg_hdr(nl_msg), &recvbuf, &recvbuflen, 0) < 0) { + if (nlComm(nl_msg, &recvbuf, &recvbuflen, 0) < 0) { rc = -1; goto err_exit; } @@ -332,7 +305,7 @@ link_del(const char *ifname) struct nlmsghdr *resp; struct nlmsgerr *err; struct ifinfomsg ifinfo = { .ifi_family = AF_UNSPEC }; - char *recvbuf = NULL; + unsigned char *recvbuf = NULL; unsigned int recvbuflen; struct nl_msg *nl_msg; @@ -349,7 +322,7 @@ link_del(const char *ifname) if (nla_put(nl_msg, IFLA_IFNAME, strlen(ifname)+1, ifname) < 0) goto buffer_too_small; - if (nlComm(nlmsg_hdr(nl_msg), &recvbuf, &recvbuflen, 0) < 0) { + if (nlComm(nl_msg, &recvbuf, &recvbuflen, 0) < 0) { rc = -1; goto err_exit; } @@ -746,7 +719,7 @@ getLldpadPid(void) { static int link_dump(bool nltarget_kernel, const char *ifname, int ifindex, - struct nlattr **tb, char **recvbuf) + struct nlattr **tb, unsigned char **recvbuf) { int rc = 0; struct nlmsghdr *resp; @@ -783,7 +756,7 @@ link_dump(bool nltarget_kernel, const ch } } - if (nlComm(nlmsg_hdr(nl_msg), recvbuf, &recvbuflen, pid) < 0) { + if (nlComm(nl_msg, recvbuf, &recvbuflen, pid) < 0) { rc = -1; goto err_exit; } @@ -869,7 +842,7 @@ ifaceGetNthParent(int ifindex, const cha { int rc; struct nlattr *tb[IFLA_MAX + 1] = { NULL, }; - char *recvbuf = NULL; + unsigned char *recvbuf = NULL; bool end = false; unsigned int i = 0; @@ -1029,7 +1002,7 @@ doPortProfileOpSetLink(bool nltarget_ker .ifi_family = AF_UNSPEC, .ifi_index = ifindex, }; - char *recvbuf = NULL; + unsigned char *recvbuf = NULL; unsigned int recvbuflen = 0; uint32_t pid = 0; struct nl_msg *nl_msg; @@ -1140,7 +1113,7 @@ doPortProfileOpSetLink(bool nltarget_ker } } - if (nlComm(nlmsg_hdr(nl_msg), &recvbuf, &recvbuflen, pid) < 0) { + if (nlComm(nl_msg, &recvbuf, &recvbuflen, pid) < 0) { rc = -1; goto err_exit; } @@ -1208,7 +1181,7 @@ doPortProfileOpCommon(bool nltarget_kern uint8_t op) { int rc; - char *recvbuf = NULL; + unsigned char *recvbuf = NULL; struct nlattr *tb[IFLA_MAX + 1] = { NULL , }; int repeats = STATUS_POLL_TIMEOUT_USEC / STATUS_POLL_INTERVL_USEC; uint16_t status = 0;

1 0

[libvirt] [PATCH] rpm: fix /var/lib/libvirt permissions
by Eric Blake 04 Nov '10

04 Nov '10

https://bugzilla.redhat.com/show_bug.cgi?id=649511 Regression of forcing 0700 permissions (which breaks guest startup because the qemu user can't see /var/lib/libvirt/*.monitor) was introduced in commit 66823690e, as part of libvirt 0.8.2. * libvirt.spec.in (%{_localstatedir}/lib/libvirt): Guarantee 755 permissions. --- Stumbled on this one today when upgrading from F13 to F14, and my VMs no longer started. Turns out that the upgrade process reinitializes permissions according to the spec file, and the spec file has been wrong for a couple of releases now. libvirt.spec.in | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/libvirt.spec.in b/libvirt.spec.in index 869b5cc..827b8c5 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -801,7 +801,7 @@ fi %dir %{_localstatedir}/run/libvirt/ -%dir %{_localstatedir}/lib/libvirt/ +%dir %attr(0755, root, root) %{_localstatedir}/lib/libvirt/ %dir %attr(0711, root, root) %{_localstatedir}/lib/libvirt/images/ %dir %attr(0711, root, root) %{_localstatedir}/lib/libvirt/boot/ %dir %attr(0700, root, root) %{_localstatedir}/cache/libvirt/ @@ -895,7 +895,7 @@ fi %{_sysconfdir}/rc.d/init.d/libvirt-guests %config(noreplace) %{_sysconfdir}/sysconfig/libvirt-guests -%dir %attr(0700, root, root) %{_localstatedir}/lib/libvirt +%dir %attr(0755, root, root) %{_localstatedir}/lib/libvirt/ %if %{with_sasl} %config(noreplace) %{_sysconfdir}/sasl2/libvirt.conf -- 1.7.3.2

2 1

[libvirt] virsh reboot not working with QEMU driver?
by Jon Stanley 04 Nov '10

04 Nov '10

I've noticed when using the QEMU driver that virsh reboot is not supported. Can this be supported, or is there some functionality that the underlying QEMU is missing?

3 4

[libvirt] [PATCH] qemu: record timestamps in qemu guest log
by Osier Yang 04 Nov '10

04 Nov '10

Currently only record timestamps for domain start and shutdown, for domain start, record timestamps before qemu command line, for domain shutdown, just says it's shutting down. * src/qemu/qemu_driver.c (qemudStartVMDaemon, qemudShutdownVMDaemon) --- src/qemu/qemu_driver.c | 62 ++++++++++++++++++++++++++++++++++++++++++++++- 1 files changed, 60 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 1eea3a9..89b4d11 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -3828,6 +3828,10 @@ static int qemudStartVMDaemon(virConnectPtr conn, char ebuf[1024]; char *pidfile = NULL; int logfile = -1; + struct timeval cur_time; + struct tm time_info; + char timestr[100]; + char *timestamp; qemuDomainObjPrivatePtr priv = vm->privateData; struct qemudHookData hookData; @@ -4015,7 +4019,27 @@ static int qemudStartVMDaemon(virConnectPtr conn, goto cleanup; } + gettimeofday(&cur_time, NULL); + localtime_r(&cur_time.tv_sec, &time_info); + + strftime(timestr, sizeof(timestr), "%Y-%m-%d %H:%M:%S", &time_info); + + if (virAsprintf(&timestamp, "%s.%3d: ", + timestr, (int) cur_time.tv_usec / 1000) < 0) { + VIR_FREE(timestamp); + virReportOOMError(); + goto cleanup; + } + + if (safewrite(logfile, timestamp, strlen(timestamp)) < 0) { + VIR_WARN("Unable to write timestamp to logfile: %s", + virStrerror(errno, ebuf, sizeof ebuf)); + } + + VIR_FREE(timestamp); + tmp = progenv; + while (*tmp) { if (safewrite(logfile, *tmp, strlen(*tmp)) < 0) VIR_WARN("Unable to write envv to logfile: %s", @@ -4166,7 +4190,6 @@ cleanup: return -1; } - static void qemudShutdownVMDaemon(struct qemud_driver *driver, virDomainObjPtr vm, int migrated) { @@ -4176,6 +4199,42 @@ static void qemudShutdownVMDaemon(struct qemud_driver *driver, virErrorPtr orig_err; virDomainDefPtr def; int i; + int logfile = -1; + char timestr[100]; + char *timestamp; + char ebuf[1024]; + struct timeval cur_time; + struct tm time_info; + + VIR_DEBUG0("Creating domain log file"); + if ((logfile = qemudLogFD(driver, vm->def->name)) < 0) { + /* To not break the normal domain shutdown process, skip the + * timestamp log writing if failed on opening log file. */ + VIR_WARN("Unable to open logfile: %s", + virStrerror(errno, ebuf, sizeof ebuf)); + } else { + gettimeofday(&cur_time, NULL); + localtime_r(&cur_time.tv_sec, &time_info); + + strftime(timestr, sizeof(timestr), "%Y-%m-%d %H:%M:%S", &time_info); + + if (virAsprintf(&timestamp, "%s.%3d: shutting down\n", + timestr, (int) cur_time.tv_usec / 1000) < 0) { + VIR_FREE(timestamp); + virReportOOMError(); + } + + if (close(logfile) < 0) + VIR_WARN("Unable to close logfile: %s", + virStrerror(errno, ebuf, sizeof ebuf)); + } + + if (safewrite(logfile, timestamp, strlen(timestamp)) < 0) { + VIR_WARN("Unable to write timestamp to logfile: %s", + virStrerror(errno, ebuf, sizeof ebuf)); + } + + VIR_FREE(timestamp); VIR_DEBUG("Shutting down VM '%s' pid=%d migrated=%d", vm->def->name, vm->pid, migrated); @@ -4315,7 +4374,6 @@ retry: } } - static virDrvOpenStatus qemudOpen(virConnectPtr conn, virConnectAuthPtr auth ATTRIBUTE_UNUSED, int flags ATTRIBUTE_UNUSED) { -- 1.7.2.3

4 12

[libvirt] [PATCH] macvtap: convert nl msg construction to use libnl
by Stefan Berger 03 Nov '10

03 Nov '10

In a first step I am converting the netlink message construction in macvtap code to use libnl. It's pretty much a 1:1 conversion except that now the message needs to be allocated and deallocated. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com> --- src/util/macvtap.c | 145 +++++++++++++++++------------------------------------ 1 file changed, 49 insertions(+), 96 deletions(-) Index: libvirt-acl/src/util/macvtap.c =================================================================== --- libvirt-acl.orig/src/util/macvtap.c +++ libvirt-acl/src/util/macvtap.c @@ -196,51 +196,6 @@ err_exit: } -static struct rtattr * -rtattrCreate(char *buffer, int bufsize, int type, - const void *data, int datalen) -{ - struct rtattr *r = (struct rtattr *)buffer; - r->rta_type = type; - r->rta_len = RTA_LENGTH(datalen); - if (r->rta_len > bufsize) - return NULL; - memcpy(RTA_DATA(r), data, datalen); - return r; -} - - -static void -nlInit(struct nlmsghdr *nlm, int flags, int type) -{ - nlm->nlmsg_len = NLMSG_LENGTH(0); - nlm->nlmsg_flags = flags; - nlm->nlmsg_type = type; -} - - -static void -nlAlign(struct nlmsghdr *nlm) -{ - nlm->nlmsg_len = NLMSG_ALIGN(nlm->nlmsg_len); -} - - -static void * -nlAppend(struct nlmsghdr *nlm, int totlen, const void *data, int datalen) -{ - char *pos; - nlAlign(nlm); - if (nlm->nlmsg_len + NLMSG_ALIGN(datalen) > totlen) - return NULL; - pos = (char *)nlm + nlm->nlmsg_len; - memcpy(pos, data, datalen); - nlm->nlmsg_len += datalen; - nlAlign(nlm); - return pos; -} - - # if WITH_MACVTAP static int @@ -252,73 +207,60 @@ link_add(const char *type, int *retry) { int rc = 0; - char nlmsgbuf[NLMSGBUF_SIZE]; - struct nlmsghdr *nlm = (struct nlmsghdr *)nlmsgbuf, *resp; + struct nlmsghdr *resp; struct nlmsgerr *err; - char rtattbuf[RATTBUF_SIZE]; - struct rtattr *rta, *rta1, *li; struct ifinfomsg ifinfo = { .ifi_family = AF_UNSPEC }; int ifindex; char *recvbuf = NULL; unsigned int recvbuflen; + struct nl_msg *nl_msg; + struct nlattr *linkinfo, *info_data; if (ifaceGetIndex(true, srcdev, &ifindex) != 0) return -1; - *retry = 0; - - memset(&nlmsgbuf, 0, sizeof(nlmsgbuf)); + nl_msg = nlmsg_alloc_simple(RTM_NEWLINK, + NLM_F_REQUEST | NLM_F_CREATE | NLM_F_EXCL); + if (!nl_msg) { + virReportOOMError(); + return -1; + } - nlInit(nlm, NLM_F_REQUEST | NLM_F_CREATE | NLM_F_EXCL, RTM_NEWLINK); + *retry = 0; - if (!nlAppend(nlm, sizeof(nlmsgbuf), &ifinfo, sizeof(ifinfo))) + if (nlmsg_append(nl_msg, &ifinfo, sizeof(ifinfo), NLMSG_ALIGNTO) < 0) goto buffer_too_small; - rta = rtattrCreate(rtattbuf, sizeof(rtattbuf), IFLA_LINK, - &ifindex, sizeof(ifindex)); - if (!rta || !nlAppend(nlm, sizeof(nlmsgbuf), rtattbuf, rta->rta_len)) + if (nla_put_u32(nl_msg, IFLA_LINK, ifindex) < 0) goto buffer_too_small; - rta = rtattrCreate(rtattbuf, sizeof(rtattbuf), IFLA_ADDRESS, - macaddress, macaddrsize); - if (!rta || !nlAppend(nlm, sizeof(nlmsgbuf), rtattbuf, rta->rta_len)) + if (nla_put(nl_msg, IFLA_ADDRESS, macaddrsize, macaddress) < 0) goto buffer_too_small; - if (ifname) { - rta = rtattrCreate(rtattbuf, sizeof(rtattbuf), IFLA_IFNAME, - ifname, strlen(ifname) + 1); - if (!rta || !nlAppend(nlm, sizeof(nlmsgbuf), rtattbuf, rta->rta_len)) - goto buffer_too_small; - } + if (ifname && + nla_put(nl_msg, IFLA_IFNAME, strlen(ifname)+1, ifname) < 0) + goto buffer_too_small; - rta = rtattrCreate(rtattbuf, sizeof(rtattbuf), IFLA_LINKINFO, NULL, 0); - if (!rta || - !(li = nlAppend(nlm, sizeof(nlmsgbuf), rtattbuf, rta->rta_len))) + if (!(linkinfo = nla_nest_start(nl_msg, IFLA_LINKINFO))) goto buffer_too_small; - rta = rtattrCreate(rtattbuf, sizeof(rtattbuf), IFLA_INFO_KIND, - type, strlen(type)); - if (!rta || !nlAppend(nlm, sizeof(nlmsgbuf), rtattbuf, rta->rta_len)) + if (nla_put(nl_msg, IFLA_INFO_KIND, strlen(type), type) < 0) goto buffer_too_small; if (macvlan_mode > 0) { - rta = rtattrCreate(rtattbuf, sizeof(rtattbuf), IFLA_INFO_DATA, - NULL, 0); - if (!rta || - !(rta1 = nlAppend(nlm, sizeof(nlmsgbuf), rtattbuf, rta->rta_len))) + if (!(info_data = nla_nest_start(nl_msg, IFLA_INFO_DATA))) goto buffer_too_small; - rta = rtattrCreate(rtattbuf, sizeof(rtattbuf), IFLA_MACVLAN_MODE, - &macvlan_mode, sizeof(macvlan_mode)); - if (!rta || !nlAppend(nlm, sizeof(nlmsgbuf), rtattbuf, rta->rta_len)) + if (nla_put(nl_msg, IFLA_MACVLAN_MODE, sizeof(macvlan_mode), + &macvlan_mode) < 0) goto buffer_too_small; - rta1->rta_len = (char *)nlm + nlm->nlmsg_len - (char *)rta1; + nla_nest_end(nl_msg, info_data); } - li->rta_len = (char *)nlm + nlm->nlmsg_len - (char *)li; + nla_nest_end(nl_msg, linkinfo); - if (nlComm(nlm, &recvbuf, &recvbuflen, 0) < 0) + if (nlComm(nlmsg_hdr(nl_msg), &recvbuf, &recvbuflen, 0) < 0) return -1; if (recvbuflen < NLMSG_LENGTH(0) || recvbuf == NULL) @@ -359,17 +301,23 @@ link_add(const char *type, VIR_FREE(recvbuf); + nlmsg_free(nl_msg); + 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", - _("internal buffer is too small")); + _("allocated netlink buffer is too small")); return -1; } @@ -378,28 +326,27 @@ static int link_del(const char *name) { int rc = 0; - char nlmsgbuf[NLMSGBUF_SIZE]; - struct nlmsghdr *nlm = (struct nlmsghdr *)nlmsgbuf, *resp; + struct nlmsghdr *resp; struct nlmsgerr *err; - char rtattbuf[RATTBUF_SIZE]; - struct rtattr *rta; struct ifinfomsg ifinfo = { .ifi_family = AF_UNSPEC }; char *recvbuf = NULL; unsigned int recvbuflen; + struct nl_msg *nl_msg; - memset(&nlmsgbuf, 0, sizeof(nlmsgbuf)); - - nlInit(nlm, NLM_F_REQUEST | NLM_F_CREATE | NLM_F_EXCL, RTM_DELLINK); + nl_msg = nlmsg_alloc_simple(RTM_DELLINK, + NLM_F_REQUEST | NLM_F_CREATE | NLM_F_EXCL); + if (!nl_msg) { + virReportOOMError(); + return -1; + } - if (!nlAppend(nlm, sizeof(nlmsgbuf), &ifinfo, sizeof(ifinfo))) + if (nlmsg_append(nl_msg, &ifinfo, sizeof(ifinfo), NLMSG_ALIGNTO) < 0) goto buffer_too_small; - rta = rtattrCreate(rtattbuf, sizeof(rtattbuf), IFLA_IFNAME, - name, strlen(name)+1); - if (!rta || !nlAppend(nlm, sizeof(nlmsgbuf), rtattbuf, rta->rta_len)) + if (nla_put(nl_msg, IFLA_IFNAME, strlen(name)+1, name) < 0) goto buffer_too_small; - if (nlComm(nlm, &recvbuf, &recvbuflen, 0) < 0) + if (nlComm(nlmsg_hdr(nl_msg), &recvbuf, &recvbuflen, 0) < 0) return -1; if (recvbuflen < NLMSG_LENGTH(0) || recvbuf == NULL) @@ -430,17 +377,23 @@ link_del(const char *name) VIR_FREE(recvbuf); + nlmsg_free(nl_msg); + 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", - _("internal buffer is too small")); + _("allocated netlink buffer is too small")); return -1; }

1 0

[libvirt] [PATCH] doc: fix typo in virsh.pod
by Osier Yang 03 Nov '10

03 Nov '10

--- tools/virsh.pod | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/virsh.pod b/tools/virsh.pod index 5932aaa..f677383 100644 --- a/tools/virsh.pod +++ b/tools/virsh.pod @@ -301,7 +301,7 @@ L<http://libvirt.org/formatdomain.html#elementsCPU> The following commands manipulate domains directly, as stated previously most commands take domain-id as the first parameter. The -I<domain-id> can be specified as an short integer, a name or a full UUID. +I<domain-id> can be specified as a short integer, a name or a full UUID. =over 4 @@ -326,7 +326,7 @@ If I<--console> is requested, attach to the console after creation. B<Example> virsh dumpxml <domain-id> > domain.xml - edit domain.xml + virsh edit domain.xml virsh create < domain.xml =item B<define> I<FILE> -- 1.7.2.3

5 9

[libvirt] [PATCH 6/7] Add sysinfo/smbios support to the QEmu driver
by Daniel Veillard 03 Nov '10

03 Nov '10

The patch is based on the possiblity in the QEmu command line to add -smbios options allowing to override the default values picked by QEmu. We need to detect this first from QEmu help output. If the domain is defined with smbios to be inherited from host then we pass the values coming from the Host own SMBIOS, but if the domain is defined with smbios to come from sysinfo, we use the ones coming from the domain definition. * src/qemu/qemu_conf.h: add the QEMUD_CMD_FLAG_SMBIOS_TYPE enum value * src/qemu/qemu_conf.c: scan the help output for the smbios support, and if available add support based on the domain definitions, and host data * tests/qemuhelptest.c: add the new enum in the outputs Signed-off-by: Daniel Veillard <veillard(a)redhat.com> --- src/qemu/qemu_conf.c | 117 ++++++++++++++++++++++++++++++++++++++++++++++++++ src/qemu/qemu_conf.h | 3 +- tests/qemuhelptest.c | 15 ++++-- 3 files changed, 129 insertions(+), 6 deletions(-)

2 2

[libvirt] [PATCH 4/7] Add a sysinfo util module and read host info API
by Daniel Veillard 03 Nov '10

03 Nov '10

Move existing routines about virSysinfoDef to an util module, add a new entry point virSysinfoRead() to read the host values with dmidecode * src/conf/domain_conf.c src/conf/domain_conf.h src/util/sysinfo.c src/util/sysinfo.h: move to a new module, add virSysinfoRead() * src/Makefile.am: handle the new module build * src/libvirt_private.syms: new internal symbols * include/libvirt/virterror.h src/util/virterror.c: defined a new error code for that module Signed-off-by: Daniel Veillard <veillard(a)redhat.com> --- include/libvirt/virterror.h | 3 +- po/POTFILES.in | 1 + src/Makefile.am | 1 + src/conf/domain_conf.c | 19 ---- src/conf/domain_conf.h | 25 +---- src/libvirt_private.syms | 5 + src/util/sysinfo.c | 250 +++++++++++++++++++++++++++++++++++++++++++ src/util/sysinfo.h | 58 ++++++++++ src/util/virterror.c | 3 + 9 files changed, 321 insertions(+), 44 deletions(-) create mode 100644 src/util/sysinfo.c create mode 100644 src/util/sysinfo.h

2 2