- Devel - libvirt List Archives

[libvirt] Repost [PATCH] Fix whitespace in domain.rng
by Matthew Booth 17 Feb '10

17 Feb '10

Rebased so it applies.

1 1

[libvirt] [PATCH] Fix whitespace in domain.rng
by Matthew Booth 17 Feb '10

17 Feb '10

* src/schemas/domain.rng: Replace tabs with 8 spaces --- docs/schemas/domain.rng | 52 +++++++++++++++++++++++----------------------- 1 files changed, 26 insertions(+), 26 deletions(-) diff --git a/docs/schemas/domain.rng b/docs/schemas/domain.rng index a834705..1ac5943 100644 --- a/docs/schemas/domain.rng +++ b/docs/schemas/domain.rng @@ -538,7 +538,7 @@ <ref name="unsignedInt"/> </attribute> <optional> - <ref name="address"/> + <ref name="address"/> </optional> </element> </define> @@ -603,7 +603,7 @@ </group> </choice> <optional> - <ref name="address"/> + <ref name="address"/> </optional> </element> </define> @@ -762,7 +762,7 @@ </element> </optional> <optional> - <ref name="address"/> + <ref name="address"/> </optional> </interleave> </define> @@ -1015,7 +1015,7 @@ </element> </optional> <optional> - <ref name="address"/> + <ref name="address"/> </optional> </interleave> </define> @@ -1159,9 +1159,9 @@ <ref name="guestfwdTarget"/> <ref name="virtioTarget"/> </choice> - <optional> - <ref name="address"/> - </optional> + <optional> + <ref name="address"/> + </optional> </interleave> </element> </define> @@ -1183,7 +1183,7 @@ </attribute> </optional> <optional> - <ref name="address"/> + <ref name="address"/> </optional> </element> </define> @@ -1212,12 +1212,12 @@ <group> <element name="source"> <choice> - <group> + <group> <ref name="usbproduct"/> - <optional> - <ref name="usbaddress"/> - </optional> - </group> + <optional> + <ref name="usbaddress"/> + </optional> + </group> <ref name="usbaddress"/> <element name="address"> <ref name="pciaddress"/> @@ -1276,7 +1276,7 @@ </optional> <optional> <attribute name="bus"> - <ref name="driveBus"/> + <ref name="driveBus"/> </attribute> </optional> <attribute name="unit"> @@ -1415,18 +1415,18 @@ <define name="address"> <element name="address"> <choice> - <group> - <attribute name="type"> - <value>pci</value> - </attribute> - <ref name="pciaddress"/> - </group> - <group> - <attribute name="type"> - <value>drive</value> - </attribute> - <ref name="driveaddress"/> - </group> + <group> + <attribute name="type"> + <value>pci</value> + </attribute> + <ref name="pciaddress"/> + </group> + <group> + <attribute name="type"> + <value>drive</value> + </attribute> + <ref name="driveaddress"/> + </group> </choice> </element> </define> -- 1.6.6

2 1

[libvirt] [PATCH] build: inform libtool of m4 directory
by Eric Blake 17 Feb '10

17 Feb '10

Silence a libtoolize warning. * configure.ac (AC_CONFIG_MACRO_DIR): Add. --- Libtool spits out this warning during ./autogen.sh: | libtoolize: Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and | libtoolize: rerunning libtoolize, to keep the correct libtool macros in-tree. I've checked that there is no significant difference in the generated files when applying this patch to silence the warning, and that autoconf 2.59 supplied this macro. configure.ac | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/configure.ac b/configure.ac index 1c7c24c..743a357 100644 --- a/configure.ac +++ b/configure.ac @@ -4,6 +4,7 @@ AC_INIT([libvirt], [0.7.6]) AC_CONFIG_SRCDIR([src/libvirt.c]) AC_CONFIG_AUX_DIR([build-aux]) AC_CONFIG_HEADERS([config.h]) +AC_CONFIG_MACRO_DIR([m4]) dnl Make automake keep quiet about wildcards & other GNUmake-isms AM_INIT_AUTOMAKE([-Wno-portability]) -- 1.6.6

2 1

[libvirt] [PATCH] Fix macvtap device tear down problem on virsh destroy
by Stefan Berger 17 Feb '10

17 Feb '10

This patch fixes the problem with the tear down of the macvtap device when issuing a 'virsh destroy' by moving the tear down block past the point of killing the Qemu process. It seems necessary to loop at least once in the case of 'virsh destory' since the device seems to be busy for a while after the Qemu process has been killed. This also still properly tears down the macvtap device when the VM is 'virsh shutdown'ed or halted from inside. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com>

2 5

[libvirt] [PATCH v2 0/9] New public API for computing baseline CPU
by Jiri Denemark 17 Feb '10

17 Feb '10

Version 2 changes: - rebased (mainly for conn argument removal) - swapped array and number of elements in array parameters - fixed absolute paths in generated RPC code - added python bindings for virConnectBaselineCPU - fixed uninitialized variable in cpuDecode, which wasn't revealed by gcc :-( This series introduces a new virConnectBaselineCPU() API call for computing baseline CPU model from a set of host CPU definitions. The call accepts a list of <cpu>...</cpu> xml definitions taken from host capabilities and returns an xml description of a guest CPU which is supported by all hosts. virsh is little bit less picky and accepts a file containing host CPU definitions. It doesn't care what is outside of the <cpu/> tag, which allows for concatenating outputs of virsh capabilities from several hosts and feeding the result to virsh cpu-baseline. This series can also be found on cpu-baseline branch of git://gitorious.org/~jirka/libvirt/jirka-staging.git In practice it works like this: # virsh cpu-baseline /dev/stdin >guest-cpu <<EOF <capabilities> <host> <cpu> <arch>x86_64</arch> <model>core2duo</model> <topology sockets='1' cores='2' threads='1'/> <feature name='lahf_lm'/> <feature name='xtpr'/> <feature name='cx16'/> <feature name='tm2'/> <feature name='est'/> <feature name='vmx'/> </cpu> </host> </capabilities> <cpu> <arch>i686</arch> <model>pentiumpro</model> <topology sockets='1' cores='2' threads='1'/> <feature name='vme'/> <feature name='pse36'/> <feature name='pni'/> <feature name='monitor'/> <feature name='ssse3'/> </cpu> <cpu> <arch>x86_64</arch> <model>phenom</model> </cpu> EOF # cat guest-cpu <cpu match='exact'> <model>qemu32</model> <feature policy='require' name='monitor'/> <feature policy='require' name='pse36'/> </cpu> Jiri Denemark (9): Functions for computing baseline CPU from a set of host CPUs Implement cpuArchBaseline in generic CPU driver Implement cpuArchBaseline in x86 CPU driver virConnectBaselineCPU public API Internal driver API for virConnectBaselineCPU virConnectBaselineCPU public API implementation Wire protocol format and dispatcher for virConnectBaselineCPU Implement cpuBaseline in remote and qemu drivers cpu-baseline command for virsh daemon/remote.c | 25 +++++++ daemon/remote_dispatch_args.h | 1 + daemon/remote_dispatch_prototypes.h | 8 ++ daemon/remote_dispatch_ret.h | 1 + daemon/remote_dispatch_table.h | 5 ++ include/libvirt/libvirt.h.in | 18 +++++ python/generator.py | 1 + python/libvirt-override-api.xml | 7 ++ python/libvirt-override.c | 52 +++++++++++++++ src/cpu/cpu.c | 123 ++++++++++++++++++++++++++++++++++- src/cpu/cpu.h | 21 ++++++- src/cpu/cpu_generic.c | 104 +++++++++++++++++++++++++++++- src/cpu/cpu_x86.c | 88 ++++++++++++++++++++++++- src/driver.h | 6 ++ src/esx/esx_driver.c | 1 + src/libvirt.c | 57 ++++++++++++++++ src/libvirt_private.syms | 2 + src/libvirt_public.syms | 1 + src/lxc/lxc_driver.c | 1 + src/opennebula/one_driver.c | 1 + src/openvz/openvz_driver.c | 1 + src/phyp/phyp_driver.c | 1 + src/qemu/qemu_driver.c | 14 ++++ src/remote/remote_driver.c | 32 +++++++++ src/remote/remote_protocol.c | 22 ++++++ src/remote/remote_protocol.h | 20 ++++++ src/remote/remote_protocol.x | 18 +++++- src/test/test_driver.c | 1 + src/uml/uml_driver.c | 1 + src/vbox/vbox_tmpl.c | 1 + src/xen/xen_driver.c | 1 + tools/virsh.c | 72 ++++++++++++++++++++ tools/virsh.pod | 8 ++ 33 files changed, 709 insertions(+), 6 deletions(-)

4 32

[libvirt] [PATCH] vbox_tmpl.c: avoid leak on OOM error path
by Jim Meyering 17 Feb '10

17 Feb '10

We have to free def here, rather than after "cleanup:", because "cleanup:" is on the successful-return path. >From 2c7800c3560f03fd77e4a458cb0e557165faed3f Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 15 Feb 2010 17:54:15 +0100 Subject: [PATCH] vbox_tmpl.c: avoid leak on OOM error path * src/vbox/vbox_tmpl.c (vboxDomainDumpXML): Free def upon virAsprintf failure. --- src/vbox/vbox_tmpl.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/src/vbox/vbox_tmpl.c b/src/vbox/vbox_tmpl.c index 68dffd2..1fb8d56 100644 --- a/src/vbox/vbox_tmpl.c +++ b/src/vbox/vbox_tmpl.c @@ -6059,14 +6059,15 @@ static char *vboxNetworkDumpXML(virNetworkPtr network, int flags ATTRIBUTE_UNUSE if (VIR_ALLOC(def) < 0) { virReportOOMError(); goto cleanup; } if (virAsprintf(&networkNameUtf8, "HostInterfaceNetworking-%s", network->name) < 0) { virReportOOMError(); + VIR_FREE(def); goto cleanup; } PRUnichar *networkInterfaceNameUtf16 = NULL; IHostNetworkInterface *networkInterface = NULL; VBOX_UTF8_TO_UTF16(network->name, &networkInterfaceNameUtf16); -- 1.7.0.181.g41533

2 6

Re: [libvirt] Cannot enumerate physical devices with libvirt 0.7.6
by Frédéric Grelot 16 Feb '10

16 Feb '10

Yep, here is what I get (see at the end) : 14:06:22.071: debug : do_open:1110 : name "qemu:///system" to URI components: scheme qemu opaque (null) authority (null) server (null) user (null) port 0 path /system 14:06:22.071: debug : do_open:1120 : trying driver 0 (Test) ... 14:06:22.071: debug : do_open:1126 : driver 0 Test returned DECLINED 14:06:22.071: debug : do_open:1120 : trying driver 1 (OPENVZ) ... 14:06:22.071: debug : do_open:1126 : driver 1 OPENVZ returned DECLINED 14:06:22.071: debug : do_open:1120 : trying driver 2 (VBOX) ... 14:06:22.071: debug : do_open:1126 : driver 2 VBOX returned DECLINED 14:06:22.071: debug : do_open:1120 : trying driver 3 (remote) ... 14:06:22.071: debug : do_open:1126 : driver 3 remote returned DECLINED 14:06:22.071: debug : do_open:1120 : trying driver 4 (QEMU) ... 14:06:22.071: debug : do_open:1126 : driver 4 QEMU returned SUCCESS 14:06:22.071: debug : do_open:1146 : network driver 0 Test returned DECLINED 14:06:22.071: debug : do_open:1146 : network driver 1 VBOX returned DECLINED 14:06:22.071: debug : do_open:1146 : network driver 2 remote returned DECLINED 14:06:22.071: debug : do_open:1146 : network driver 3 Network returned SUCCESS 14:06:22.071: debug : do_open:1165 : interface driver 0 Test returned DECLINED 14:06:22.071: debug : do_open:1165 : interface driver 1 remote returned DECLINED 14:06:22.072: debug : do_open:1185 : storage driver 0 Test returned DECLINED 14:06:22.072: debug : do_open:1185 : storage driver 1 VBOX returned DECLINED 14:06:22.072: debug : do_open:1185 : storage driver 2 remote returned DECLINED 14:06:22.072: debug : do_open:1185 : storage driver 3 storage returned SUCCESS 14:06:22.072: debug : do_open:1205 : node driver 0 Test returned DECLINED 14:06:22.072: debug : do_open:1205 : node driver 1 remote returned DECLINED 14:06:22.072: debug : do_open:1232 : secret driver 0 Test returned DECLINED 14:06:22.072: debug : do_open:1232 : secret driver 1 remote returned DECLINED 14:06:22.072: debug : do_open:1232 : secret driver 2 secret returned SUCCESS 14:06:22.072: debug : virConnectClose:1381 : conn=0x1eaf0b0 I think the times when I get those lines matches with my "virsh nodedev-list" requests (or something like this : virt-manager tries to create a vm, add hardware...). Does this help? Frederic. ----- "Dave Allan" <dallan(a)redhat.com> a écrit : > On 02/16/2010 03:00 AM, Frédéric Grelot wrote: > >> Daniel was referring to the libvirt udev driver--the libvirt code > that > >> > >> interacts with udev--not to udev itself, which is what you're > testing > >> > >> with udevadm. > > > > Ok, sorry... At least I ensured that udev itself is on... > > > >> In any of your logs when you have libvirt debug logging > >> turned on, do you see the message: > >> > >> Registering udev node device backend > > > > Nope, I checked in /var/log/* and libvirt.log in debug mode (which, > after 1 1/2 day of "debug" activity enabled, is already 2Gb!) > > There is no mention of udev. I think it means that the udev driver > is not loaded in libvirt? How could enable it then? (or produce some > message telling me why it is not) ? > > > > > >> These messages are the result of not being able to find the > device, > >> which is consistent with the other things you're seeing. These > are > >> all symptoms of the node device code not running. > > > > OK, I'll wait for your suggestions... > > Do you see messages like: > > > 14:35:01.849: debug : do_open:1205 : node driver 0 Test returned > DECLINED > 14:35:01.849: debug : do_open:1205 : node driver 1 ESX returned > DECLINED > 14:35:01.849: debug : do_open:1205 : node driver 2 remote returned > DECLINED > 14:35:01.849: debug : do_open:1205 : node driver 3 udevDeviceMonitor > returned SUCCESS > > in your debug log? > > > > Frederic. > > > > > >> > >>> Have you any other Idea? > >>> Frederic. > >>> > >>> > >>> ----- "Daniel P. Berrange"<berrange(a)redhat.com> a écrit : > >>> > >>>> On Fri, Feb 12, 2010 at 08:05:21PM +0100, Fr?d?ric Grelot wrote: > >>>>> Hi all, > >>>>> > >>>>> I often update my F12 box with the rawvirt repository, which > >>>>> just > >>>> installed libvirt 0.7.6. Since that, I cannot enumerate physical > >>>> devices (and thus get a beautiful error when I start one of my > vms > >>>> that should use a USB device). > >>>>> > >>>>> I don't think I changed anything else on that box, and, of > >>>>> course, I > >>>> already restarted the libvirt daemon. > >>>>> I haven't restarted the server itself, because it is currently > >>>>> used, > >>>> but I could do it if necessary... > >>>>> > >>>>> I make good use of the qemu hypervisor, and I'm looking forward > >>>>> to > >>>> solve this issue! > >>>>> > >>>>> Frederic. > >>>>> > >>>>> # libvirtd --version libvirtd (libvirt) 0.7.6 > >>>>> > >>>>> # virsh --version 0.7.6 > >>>>> > >>>>> # virsh nodedev-list error :Failed to count node devices error > >>>>> :this function is not supported by the hypervisor: > >>>> virNodeNumOfDevices > >>>> > >>>> It would appear that the udev driver failed to start - there are > >>>> probably messages in syslog about the problem > >>>> > >>>> > >>>> Regards, Daniel -- |: Red Hat, Engineering, London -o- > >>>> http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- > >>>> http://virt-manager.org -o- http://ovirt.org :| |: > >>>> http://autobuild.org -o- http://search.cpan.org/~danberr/ > :| > >>>> |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 > >>>> 7D3B 9505 :| > >>> > >>> -- libvir-list mailing list libvir-list(a)redhat.com > >>> https://www.redhat.com/mailman/listinfo/libvir-list > > > > -- > > libvir-list mailing list > > libvir-list(a)redhat.com > > https://www.redhat.com/mailman/listinfo/libvir-list

2 1

Re: [libvirt] Cannot enumerate physical devices with libvirt 0.7.6
by Frédéric Grelot 16 Feb '10

16 Feb '10

The udev driver is actually fully loaded, I even entered "udevadm monitor" in a console, and it reacted as expected when I plugged/unplugged an usb mouse... There is nothing about it in dmesg/boot/message.log, nor libvirt.log (I set up libvirt daemon to log at debug level in /var/log/libvirt/libvirt.log). By the way, I noticed that libvirt generates tons of "error : qemuMonitorTextGetBlockStatsInfo:698 : invalid argument in no stats found for device virtio-diskX" while virt-manager is running (it stops as soon as I stop virt-manager). Is it linked with the bug I saw on the list about libvirt taking over memory after time isn't it? Have you any other Idea? Frederic. ----- "Daniel P. Berrange" <berrange(a)redhat.com> a écrit : > On Fri, Feb 12, 2010 at 08:05:21PM +0100, Fr?d?ric Grelot wrote: > > Hi all, > > > > I often update my F12 box with the rawvirt repository, which just > installed libvirt 0.7.6. Since that, I cannot enumerate physical > devices (and thus get a beautiful error when I start one of my vms > that should use a USB device). > > > > I don't think I changed anything else on that box, and, of course, I > already restarted the libvirt daemon. > > I haven't restarted the server itself, because it is currently used, > but I could do it if necessary... > > > > I make good use of the qemu hypervisor, and I'm looking forward to > solve this issue! > > > > Frederic. > > > > # libvirtd --version > > libvirtd (libvirt) 0.7.6 > > > > # virsh --version > > 0.7.6 > > > > # virsh nodedev-list > > error :Failed to count node devices > > error :this function is not supported by the hypervisor: > virNodeNumOfDevices > > It would appear that the udev driver failed to start - there are > probably > messages in syslog about the problem > > > Regards, > Daniel > -- > |: Red Hat, Engineering, London -o- > http://people.redhat.com/berrange/ :| > |: http://libvirt.org -o- http://virt-manager.org -o- > http://ovirt.org :| > |: http://autobuild.org -o- > http://search.cpan.org/~danberr/ :| > |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B > 9505 :|

2 3

[libvirt] [PATCH] vshCommandParse: placate coverity
by Jim Meyering 16 Feb '10

16 Feb '10

Here are two patches to avoid semblance of a leak. One adds a useless use of VIR_FREE. The other adds an assertion that the pointer to would-be-leaked storage is NULL. There's already an assertion in this file, so no need to add an "#include ...". Take your pick. >From 2070262e7f1a4803e06a6f75bd96091c5754164d Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Tue, 16 Feb 2010 14:34:48 +0100 Subject: [PATCH] vshCommandParse: placate coverity There is no real leak here, but Coverity-Prevent thinks there is. It does not see that while there are four ways to return from vshCommandGetToken with VSH_TK_END, none of them results in allocation of a result. * tools/virsh.c (vshCommandParse): Add a (currently) useless VIR_FREE, to ensure that we never leak when vshCommandGetToken returns VSH_TK_END. --- tools/virsh.c | 4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 7db48d9..4ffa154 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -8222,8 +8222,10 @@ vshCommandParse(vshControl *ctl, char *cmdstr) str = end; - if (tk == VSH_TK_END) + if (tk == VSH_TK_END) { + VIR_FREE(tkdata); break; + } if (tk == VSH_TK_ERROR) goto syntaxError; -- 1.7.0.181.g41533 >From 98b0c148d7fd7e11eed7786189db60b485612b51 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Tue, 16 Feb 2010 14:34:48 +0100 Subject: [PATCH] vshCommandParse: placate coverity * tools/virsh.c (vshCommandParse): Inform static analyzers that there is no risk of leak when vshCommandGetToken returns VSH_TK_END. --- tools/virsh.c | 4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 7db48d9..5c45d1c 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -8222,8 +8222,10 @@ vshCommandParse(vshControl *ctl, char *cmdstr) str = end; - if (tk == VSH_TK_END) + if (tk == VSH_TK_END) { + assert (tkdata == NULL); break; + } if (tk == VSH_TK_ERROR) goto syntaxError; -- 1.7.0.181.g41533

2 2

[libvirt] [PATCH] qemuMonitorTextAddUSBDisk: avoid unconditional leak
by Jim Meyering 16 Feb '10

16 Feb '10

First the patch, then the (fixed) function: >From d3f0ee1ee11ca5f3709168fdc4e6eb1c56fdfbc0 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Tue, 16 Feb 2010 13:51:38 +0100 Subject: [PATCH] qemuMonitorTextAddUSBDisk: avoid unconditional leak * src/qemu/qemu_monitor_text.c (qemuMonitorTextAddUSBDisk): Free command output buffer. --- src/qemu/qemu_monitor_text.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/src/qemu/qemu_monitor_text.c b/src/qemu/qemu_monitor_text.c index c71bf1d..62ffcc6 100644 --- a/src/qemu/qemu_monitor_text.c +++ b/src/qemu/qemu_monitor_text.c @@ -1229,6 +1229,7 @@ int qemuMonitorTextAddUSBDisk(qemuMonitorPtr mon, cleanup: VIR_FREE(cmd); VIR_FREE(safepath); + VIR_FREE(info); return ret; } -- 1.7.0.181.g41533 int qemuMonitorTextAddUSBDisk(qemuMonitorPtr mon, const char *path) { char *cmd = NULL; char *safepath; int ret = -1; char *info = NULL; safepath = qemuMonitorEscapeArg(path); if (!safepath) { virReportOOMError(); return -1; } if (virAsprintf(&cmd, "usb_add disk:%s", safepath) < 0) { virReportOOMError(); goto cleanup; } if (qemuMonitorCommand(mon, cmd, &info) < 0) { qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("cannot run monitor command to add usb disk")); goto cleanup; } /* If the command failed qemu prints: * Could not add ... */ if (strstr(info, "Could not add ")) { qemuReportError(VIR_ERR_OPERATION_FAILED, _("unable to add USB disk %s: %s"), path, info); goto cleanup; } ret = 0; cleanup: VIR_FREE(cmd); VIR_FREE(safepath); VIR_FREE(info); return ret; }

2 1

[libvirt] [PATCH] tests: avoid NULL deref upon OOM failure
by Jim Meyering 16 Feb '10

16 Feb '10

Pretty obvious: >From 3e79d8839c5051461197a7c484adb32f7f04377a Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Tue, 16 Feb 2010 09:36:57 +0100 Subject: [PATCH] tests: avoid NULL deref upon OOM failure * tests/qemuxml2argvtest.c (testCompareXMLToArgvFiles): Handle malloc failure. --- tests/qemuxml2argvtest.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index 3ee71a3..aa42f99 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -98,7 +98,8 @@ static int testCompareXMLToArgvFiles(const char *xml, len += strlen(*tmp) + 1; tmp++; } - actualargv = malloc(sizeof(*actualargv)*len); + if ((actualargv = malloc(sizeof(*actualargv)*len)) == NULL) + goto fail; actualargv[0] = '\0'; tmp = qenv; while (*tmp) { -- 1.7.0.181.g41533

2 1

[libvirt] [PATCH] qemuInitPasswords: avoid unconditional leak
by Jim Meyering 16 Feb '10

16 Feb '10

A successful getVolumeQcowPassphrase allocates "secret". Nothing freed it. >From 2aa6d3c4254f587133e473f4d193ccb023507665 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Tue, 16 Feb 2010 09:22:02 +0100 Subject: [PATCH] qemuInitPasswords: avoid unconditional leak * src/qemu/qemu_driver.c (qemuInitPasswords): Free pass-phrase buffer. --- src/qemu/qemu_driver.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 03feaf4..e8b7cf6 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -1733,34 +1733,35 @@ qemuInitPasswords(virConnectPtr conn, for (i = 0 ; i < vm->def->ndisks ; i++) { char *secret; size_t secretLen; if (!vm->def->disks[i]->encryption || !vm->def->disks[i]->src) continue; if (getVolumeQcowPassphrase(conn, vm->def->disks[i], &secret, &secretLen) < 0) goto cleanup; qemuDomainObjEnterMonitorWithDriver(driver, vm); ret = qemuMonitorSetDrivePassphrase(priv->mon, vm->def->disks[i]->info.alias, secret); + VIR_FREE(secret); qemuDomainObjExitMonitorWithDriver(driver, vm); if (ret < 0) goto cleanup; } } cleanup: return ret; } #define QEMU_PCI_VENDOR_INTEL 0x8086 #define QEMU_PCI_VENDOR_LSI_LOGIC 0x1000 #define QEMU_PCI_VENDOR_REDHAT 0x1af4 #define QEMU_PCI_VENDOR_CIRRUS 0x1013 #define QEMU_PCI_VENDOR_REALTEK 0x10ec #define QEMU_PCI_VENDOR_AMD 0x1022 -- 1.7.0.181.g41533

2 1

[libvirt] [PATCH] qemuMonitorTextAddDevice: avoid unconditional leak
by Jim Meyering 16 Feb '10

16 Feb '10

Here's the patch (entire function below): >From 1ee0edab5356ca8cd84a88e4e4a2e4f3cd916f57 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Tue, 16 Feb 2010 08:22:34 +0100 Subject: [PATCH] qemuMonitorTextAddDevice: avoid unconditional leak * src/qemu/qemu_monitor_text.c (qemuMonitorTextAddDevice): Free the device name buffer. --- src/qemu/qemu_monitor_text.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/src/qemu/qemu_monitor_text.c b/src/qemu/qemu_monitor_text.c index e993699..c71bf1d 100644 --- a/src/qemu/qemu_monitor_text.c +++ b/src/qemu/qemu_monitor_text.c @@ -2087,6 +2087,7 @@ int qemuMonitorTextAddDevice(qemuMonitorPtr mon, cleanup: VIR_FREE(cmd); VIR_FREE(reply); + VIR_FREE(safedev); return ret; } -- 1.7.0.181.g41533 For reference, here's the entire function: int qemuMonitorTextAddDevice(qemuMonitorPtr mon, const char *devicestr) { char *cmd = NULL; char *reply = NULL; char *safedev; int ret = -1; if (!(safedev = qemuMonitorEscapeArg(devicestr))) { virReportOOMError(); goto cleanup; } if (virAsprintf(&cmd, "device_add %s", safedev) < 0) { virReportOOMError(); goto cleanup; } if (qemuMonitorCommand(mon, cmd, &reply) < 0) { qemuReportError(VIR_ERR_OPERATION_FAILED, _("cannot attach %s device"), devicestr); goto cleanup; } if (STRNEQ(reply, "")) { qemuReportError(VIR_ERR_OPERATION_FAILED, _("adding %s device failed: %s"), devicestr, reply); goto cleanup; } ret = 0; cleanup: VIR_FREE(cmd); VIR_FREE(reply); VIR_FREE(safedev); return ret; }

2 1

[libvirt] PATCH [0/3] Volume zeroing
by David Allan 16 Feb '10

16 Feb '10

The following patches implement overwriting a volume with zeros when the volume is deleted. The zeroing happens before the delete, so it works for storage backends that don't support actually deleting volumes as well as the ones that do. The intent is that any future VM assigned that volume will not be able to recover any data belonging to the previous VM. It is not intended to prevent attackers with physical access to the medium from recovering data--it simply writes a single pass of zeros over the medium. If the filesystem containing the volume supports the fiemap ioctl and the volume is a sparse file, the volume zeroing code attempts to use fiemap to locate the mapped extents. It does not attempt to zero a sparse file if it cannot use fiemap. Such an operation could take an essentially unbounded amount of time. Since the volume is being deleted, zeroing has less value in the context of backends that support delete, but does provide value with storage backends that do not zero volumes if they are deleted and recreated. Dave

6 15

[libvirt] [PATCH] Support VCPU hotplug in QEMU guests
by Daniel P. Berrange 16 Feb '10

16 Feb '10

QEMU has a monitor command 'set_cpu' which allows a specific CPU to be toggled between online& offline state. libvirt CPU hotplug does not work in terms of individual indexes CPUs. Thus to support this, we iteratively toggle the online state when the total number of vCPUs is adjusted via libvirt NB, currently untested since QEMU segvs when running this! * src/qemu/qemu_driver.c: Toggle online state for CPUs when doing hotplug * src/qemu/qemu_monitor.c, src/qemu/qemu_monitor.h, src/qemu/qemu_monitor_json.c, src/qemu/qemu_monitor_json.h, src/qemu/qemu_monitor_text.c, src/qemu/qemu_monitor_text.h: Add monitor API for toggling a CPU's online status via 'set_cpu --- src/qemu/qemu_driver.c | 59 ++++++++++++++++++++++++++++++++++++----- src/qemu/qemu_monitor.c | 14 ++++++++++ src/qemu/qemu_monitor.h | 2 + src/qemu/qemu_monitor_json.c | 44 +++++++++++++++++++++++++++++++ src/qemu/qemu_monitor_json.h | 1 + src/qemu/qemu_monitor_text.c | 38 +++++++++++++++++++++++++++ src/qemu/qemu_monitor_text.h | 1 + 7 files changed, 151 insertions(+), 8 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 03feaf4..6ef784f 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -4206,6 +4206,51 @@ cleanup: } +static int qemudDomainHotplugVcpus(virDomainObjPtr vm, unsigned int nvcpus) +{ + qemuDomainObjPrivatePtr priv = vm->privateData; + int i, rc; + int ret = -1; + + /* We need different branches here, because we want to offline + * in reverse order to onlining, so any partial fail leaves us in a + * reasonably sensible state */ + if (nvcpus > vm->def->vcpus) { + for (i = vm->def->vcpus ; i < nvcpus ; i++) { + /* Online new CPU */ + rc = qemuMonitorSetCPU(priv->mon, i, 1); + if (rc == 0) + goto unsupported; + if (rc < 0) + goto cleanup; + + vm->def->vcpus++; + } + } else { + for (i = vm->def->vcpus - 1 ; i >= nvcpus ; i--) { + /* Offline old CPU */ + rc = qemuMonitorSetCPU(priv->mon, i, 0); + if (rc == 0) + goto unsupported; + if (rc < 0) + goto cleanup; + + vm->def->vcpus--; + } + } + + ret = 0; + +cleanup: + return ret; + +unsupported: + qemudReportError(NULL, NULL, NULL, VIR_ERR_OPERATION_INVALID, "%s", + _("cannot change vcpu count of an active domain")); + goto cleanup; +} + + static int qemudDomainSetVcpus(virDomainPtr dom, unsigned int nvcpus) { struct qemud_driver *driver = dom->conn->privateData; virDomainObjPtr vm; @@ -4228,12 +4273,6 @@ static int qemudDomainSetVcpus(virDomainPtr dom, unsigned int nvcpus) { if (qemuDomainObjBeginJob(vm) < 0) goto cleanup; - if (virDomainObjIsActive(vm)) { - qemuReportError(VIR_ERR_OPERATION_INVALID, "%s", - _("cannot change vcpu count of an active domain")); - goto endjob; - } - if (!(type = virDomainVirtTypeToString(vm->def->virtType))) { qemuReportError(VIR_ERR_INTERNAL_ERROR, _("unknown virt type in domain definition '%d'"), @@ -4254,8 +4293,12 @@ static int qemudDomainSetVcpus(virDomainPtr dom, unsigned int nvcpus) { goto endjob; } - vm->def->vcpus = nvcpus; - ret = 0; + if (virDomainObjIsActive(vm)) { + ret = qemudDomainHotplugVcpus(vm, nvcpus); + } else { + vm->def->vcpus = nvcpus; + ret = 0; + } endjob: if (qemuDomainObjEndJob(vm) == 0) diff --git a/src/qemu/qemu_monitor.c b/src/qemu/qemu_monitor.c index c1d369b..541d7a1 100644 --- a/src/qemu/qemu_monitor.c +++ b/src/qemu/qemu_monitor.c @@ -912,6 +912,20 @@ int qemuMonitorSetBalloon(qemuMonitorPtr mon, return ret; } + +int qemuMonitorSetCPU(qemuMonitorPtr mon, int cpu, int online) +{ + int ret; + DEBUG("mon=%p, fd=%d cpu=%d online=%d", mon, mon->fd, cpu, online); + + if (mon->json) + ret = qemuMonitorJSONSetCPU(mon, cpu, online); + else + ret = qemuMonitorTextSetCPU(mon, cpu, online); + return ret; +} + + int qemuMonitorEjectMedia(qemuMonitorPtr mon, const char *devname) { diff --git a/src/qemu/qemu_monitor.h b/src/qemu/qemu_monitor.h index 786ad7a..d560c63 100644 --- a/src/qemu/qemu_monitor.h +++ b/src/qemu/qemu_monitor.h @@ -144,6 +144,8 @@ int qemuMonitorSetVNCPassword(qemuMonitorPtr mon, const char *password); int qemuMonitorSetBalloon(qemuMonitorPtr mon, unsigned long newmem); +int qemuMonitorSetCPU(qemuMonitorPtr mon, int cpu, int online); + /* XXX should we pass the virDomainDiskDefPtr instead * and hide devname details inside monitor. Reconsider diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index c9b8d60..35e529e 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -873,6 +873,50 @@ cleanup: } +/* + * Returns: 0 if CPU hotplug not supported, +1 if CPU hotplug worked + * or -1 on failure + */ +int qemuMonitorJSONSetCPU(qemuMonitorPtr mon, + int cpu, int online) +{ + int ret; + virJSONValuePtr cmd = qemuMonitorJSONMakeCommand("balloon", + "U:cpu", (unsigned long long)cpu, + "s:state", online ? "online" : "offline", + NULL); + virJSONValuePtr reply = NULL; + if (!cmd) + return -1; + + ret = qemuMonitorJSONCommand(mon, cmd, &reply); + + if (ret == 0) { + /* XXX See if CPU soft-failed due to lack of ACPI */ +#if 0 + if (qemuMonitorJSONHasError(reply, "DeviceNotActive") || + qemuMonitorJSONHasError(reply, "KVMMissingCap")) + goto cleanup; +#endif + + /* See if any other fatal error occurred */ + ret = qemuMonitorJSONCheckError(cmd, reply); + + /* Real success */ + if (ret == 0) + ret = 1; + } + +#if 0 +cleanup: +#endif + + virJSONValueFree(cmd); + virJSONValueFree(reply); + return ret; +} + + int qemuMonitorJSONEjectMedia(qemuMonitorPtr mon, const char *devname) { diff --git a/src/qemu/qemu_monitor_json.h b/src/qemu/qemu_monitor_json.h index 65a70e3..88dfd3c 100644 --- a/src/qemu/qemu_monitor_json.h +++ b/src/qemu/qemu_monitor_json.h @@ -57,6 +57,7 @@ int qemuMonitorJSONSetVNCPassword(qemuMonitorPtr mon, const char *password); int qemuMonitorJSONSetBalloon(qemuMonitorPtr mon, unsigned long newmem); +int qemuMonitorJSONSetCPU(qemuMonitorPtr mon, int cpu, int online); int qemuMonitorJSONEjectMedia(qemuMonitorPtr mon, const char *devname); diff --git a/src/qemu/qemu_monitor_text.c b/src/qemu/qemu_monitor_text.c index e993699..1745b20 100644 --- a/src/qemu/qemu_monitor_text.c +++ b/src/qemu/qemu_monitor_text.c @@ -797,6 +797,44 @@ int qemuMonitorTextSetBalloon(qemuMonitorPtr mon, return ret; } + +/* + * Returns: 0 if balloon not supported, +1 if balloon adjust worked + * or -1 on failure + */ +int qemuMonitorTextSetCPU(qemuMonitorPtr mon, int cpu, int online) +{ + char *cmd; + char *reply = NULL; + int ret = -1; + + if (virAsprintf(&cmd, "set_cpu %d %s", cpu, online ? "online" : "offline") < 0) { + virReportOOMError(NULL); + return -1; + } + + if (qemuMonitorCommand(mon, cmd, &reply) < 0) { + qemudReportError(NULL, NULL, NULL, VIR_ERR_OPERATION_FAILED, + "%s", _("could nt change CPU online status")); + VIR_FREE(cmd); + return -1; + } + VIR_FREE(cmd); + + /* If the command failed qemu prints: 'unknown command' + * No message is printed on success it seems */ + if (strstr(reply, "\nunknown command:")) { + /* Don't set error - it is expected CPU onlining fails on many qemu - caller will handle */ + ret = 0; + } else { + ret = 1; + } + + VIR_FREE(reply); + return ret; +} + + int qemuMonitorTextEjectMedia(qemuMonitorPtr mon, const char *devname) { diff --git a/src/qemu/qemu_monitor_text.h b/src/qemu/qemu_monitor_text.h index 1937e99..77b8b0f 100644 --- a/src/qemu/qemu_monitor_text.h +++ b/src/qemu/qemu_monitor_text.h @@ -61,6 +61,7 @@ int qemuMonitorTextSetVNCPassword(qemuMonitorPtr mon, const char *password); int qemuMonitorTextSetBalloon(qemuMonitorPtr mon, unsigned long newmem); +int qemuMonitorTextSetCPU(qemuMonitorPtr mon, int cpu, int online); int qemuMonitorTextEjectMedia(qemuMonitorPtr mon, const char *devname); -- 1.6.6

3 4

Re: [libvirt] Live Migration with non-shared storage for kvm
by Kenneth Nagin 16 Feb '10

16 Feb '10

>"Daniel P. Berrange" <berrange(a)redhat.com> wrote on 26/01/2010 16:20:22: > "Daniel P. Berrange" <berrange(a)redhat.com> > 26/01/2010 16:20 > > Please respond to > "Daniel P. Berrange" <berrange(a)redhat.com> > > To > > Kenneth Nagin/Haifa/IBM@IBMIL > > cc > > Subject > > Re: [libvirt] Live Migration with non-shared storage for kvm > > On Tue, Jan 26, 2010 at 04:09:58PM +0200, Kenneth Nagin wrote: > > "Daniel P. Berrange" <berrange(a)redhat.com> wrote on 26/01/2010 12:14:47: > > Since, this my first time submitting something to libvirt I wanted to make > > the initially submission simple, i.e. just adding the flags. > > That sounds fine > > > > > > > > I suggest adding these flags to virDomainMigrate. > > > > > > That sounds reasonable. > > > > > > > If I'm not mistaken qemuMonitorTextMigrate is the function that > > actually > > > > invokes the kvm monitor. > > > > Thus, it would be necessary to pass the flags to > > qemuMonitorTextMigrate.. > > > > But qemuMonitorTextMigrate does not have a flag input parameter. I > > think > > > > the least disruptive way to support the new flags is use the existing > > > > "background" parameter > > > > to pass the flags. Of course this would require some changes to the > > > > upstream functions that are invoked for migration. > > > > > > That is an internal function, so adding extra parameters to it is no > > > problem at all - no need to reuse/abuse existing parameters. > > > > > Are you recommending adding another "flag" parameter or exploiting the > > existing "background" parameter? > > Probably define some flags for the internal API. eg in src/qemu/qemu_monitor.h > add a small enum > > enum { > QEMU_MONITOR_MIGRATE_BACKGROUND = 1 << 0, > QEMU_MONITOR_MIGRATE_STORAGE = 1 << 1 > ...etc > }; > > and then just replace the existing background parameter with a flags > parameter > > > Daniel > -- > |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| > |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| > |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| > |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :| Sorry about the delayed response. I had some pressing work items and family issues that took precedence. In order to support kvm's live migration with non-shared storage I added two migration flags that user can invoke: VIR_MIGRATE_NON_SHARED_DISK = (1 << 6), /* migration with non-shared storage with full disk copy */ VIR_MIGRATE_NON_SHARED_INC = (1 << 7), /* migration with non-shared storage with incremental copy */ /* (same base image shared between source and destination) */ Likewise I add complementary flags to virsh migrate: non_shared_disk and non_shared_inc As you suggested I also added internal flags to be passed in the "background" parameter to the qemu monitor: typedef enum { QEMU_MONITOR_MIGRATE_BACKGROUND = 1 << 0, QEMU_MONITOR_MIGRATE_NON_SHARED_DISK = 1 << 1, /* migration with non-shared storage with full disk copy */ QEMU_MONITOR_MIGRATE_NON_SHARED_INC = 1 << 2, /* migration with non-shared storage with incremental copy */ QEMU_MONITOR_MIGRATION_FLAGS_LAST }; I updated qemu_driver.c's doNativeMigrate and doTunnelMigrate to map the external flags to the internal flags. I updated qemu_monitor_text's qemuMonitorTextMigrate to map the internal flags to the qemu monitor's command line flags. I tested the doNativeMigrate usage of the two flags. However, I could not test the doTunnelMigrate usage because I wasn't sure how it works. Also, qemudDomainSave ends up calling qemuMonitorTextMigrate, but it didn't support the external flags so I assumed that it was not relevant. Here is the diff patch file from the current git: (See attached file: migration_with_non_shared.patch) - Kenneth Nagin

1 0

[libvirt] [PATCH 0/5] macvtap support for Qemu/KVM VMs via libvirt
by Stefan Berger 16 Feb '10

16 Feb '10

Hello! This is a re-post of previously posted patches following Daniel Berrange's request for changes along with other fixes PLUS a rebase to the latest code where the conn parameter is missing along with changes requested by Daniel Veillard. The following patches provide support for making the macvtap networking device type available to Qemu/KVM VMs. The patches rely on the macvtap driver that just became available through the Linux net-next tree (fixes still may be necessary) and make the device available to Qemu/KVM via a tap file descriptor similar to a 'regular' tap device. Following up on previous discussions, the libvirt patches allow using the following XML in the domain description to enable qemu network connectivity via this type of device: <interface type='direct'> <source dev='eth1' mode='vepa'/> <model type='virtio'/> </interface> The above XML indicates that eth1 is the Ethernet interface to link the macvtap device to and communicate to the network. As a consequence, libvirt will create an instance of a macvtap device, assign it the same MAC address as the VM's interface has and open a file descriptor of the associated character device /dev/tap%d and pass it via command line to Qemu/kvm. In the above XML the mode can be chosen as 'vepa', 'private' or 'bridge' and is by default set to 'vepa'(by the driver) if omitted. Attachment and detachment of macvtap to/from a running VM also works. Regards, Stefan

3 4

[libvirt] [PATCH] libvirt-override.c: avoid a leak upon call with invalid argument
by Jim Meyering 16 Feb '10

16 Feb '10

>From 49d7a268574772d362dc374b531ff80ec87f5062 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Tue, 16 Feb 2010 08:07:38 +0100 Subject: [PATCH] libvirt-override.c: avoid a leak upon call with invalid argument * python/libvirt-override.c (libvirt_virConnectBaselineCPU): Don't leak the xmlcpus buffer upon encountering a non-string list element. --- python/libvirt-override.c | 6 ++++-- 1 files changed, 4 insertions(+), 2 deletions(-) diff --git a/python/libvirt-override.c b/python/libvirt-override.c index a71766a..2447ad7 100644 --- a/python/libvirt-override.c +++ b/python/libvirt-override.c @@ -1,16 +1,16 @@ /* * libvir.c: this modules implements the main part of the glue of the * libvir library and the Python interpreter. It provides the * entry points where an automatically generated stub is * unpractical * - * Copyright (C) 2005, 2007-2009 Red Hat, Inc. + * Copyright (C) 2005, 2007-2010 Red Hat, Inc. * * Daniel Veillard <veillard(a)redhat.com> */ #include <config.h> /* Horrible kludge to work around even more horrible name-space pollution via Python.h. That file includes /usr/include/python2.5/pyconfig*.h, which has over 180 autoconf-style HAVE_* definitions. Shame on them. */ @@ -2040,20 +2040,22 @@ libvirt_virConnectBaselineCPU(PyObject *self ATTRIBUTE_UNUSED, if (PyList_Check(list)) { int i; ncpus = PyList_Size(list); if ((xmlcpus = malloc(ncpus * sizeof(*xmlcpus))) == NULL) return VIR_PY_INT_FAIL; for (i = 0; i < ncpus; i++) { xmlcpus[i] = PyString_AsString(PyList_GetItem(list, i)); - if (xmlcpus[i] == NULL) + if (xmlcpus[i] == NULL) { + free(xmlcpus); return VIR_PY_INT_FAIL; + } } } LIBVIRT_BEGIN_ALLOW_THREADS; base_cpu = virConnectBaselineCPU(conn, xmlcpus, ncpus, flags); LIBVIRT_END_ALLOW_THREADS; free(xmlcpus); -- 1.7.0.181.g41533

2 1

[libvirt] [PATCH 4/4] remote: Detect 'nc' version incompatibilities
by Cole Robinson 16 Feb '10

16 Feb '10

This ugly thing is a shell script to detect availability of the -q option for 'nc': debian and suse based distros need this flag to ensure the remote nc will exit on EOF, so it will go away when we close the tunnel. If it doesn't go away, a useless 'nc' process is left sitting on the remote host. Fedora's 'nc' doesn't have this option, so we can't blindly pass -q. More info here: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=564034 We also detect the -U option, since some nc versions don't support it, which has caused quite some user confusion. Connecting to a box and using the incorrect nc: virsh --connect qemu+ssh://root@debhost/system?netcat=/bin/nc.traditional error: server closed connection: /bin/nc.traditional: invalid option -- U nc -h for help openbsd nc is required error: failed to connect to the hypervisor Test with Fedora 12, RHEL 5.4, and Debian Lenny --- src/remote/remote_driver.c | 52 ++++++++++++++++++++++++++++++++++++++------ 1 files changed, 45 insertions(+), 7 deletions(-) diff --git a/src/remote/remote_driver.c b/src/remote/remote_driver.c index 7f92fd0..c3258d3 100644 --- a/src/remote/remote_driver.c +++ b/src/remote/remote_driver.c @@ -732,8 +732,49 @@ doRemoteOpen (virConnectPtr conn, } case trans_ssh: { - int j, nr_args = 6; + int j, nr_args = 0; + char *nc_command; + + const char *nc_bin = netcat ? netcat : "nc"; + const char *sock_path = (sockname ? sockname : + (flags & VIR_CONNECT_RO + ? LIBVIRTD_PRIV_UNIX_SOCKET_RO + : LIBVIRTD_PRIV_UNIX_SOCKET)); + + /* + * Build 'nc' command run on the remote host + * + * This ugly thing is a shell script to detect availability of + * the -q option for 'nc': debian and suse based distros need this + * flag to ensure the remote nc will exit on EOF, so it will go away + * when we close the tunnel. If it doesn't go away, a useless 'nc' + * process is left sitting on the remote host. + * + * Fedora's 'nc' doesn't have this option, and apparently defaults + * to the desired behavior. + */ + const char *nc_detect_template = ( + "NCOUT=$(%s -U 2>&1);" + "echo \"$NCOUT\" | grep -q 'invalid option';" + "if [ $? -eq 0 ] ; then" + " echo \"$NCOUT\" >&2;" + " echo openbsd 'nc' is required >&2;" + " exit 1;" + "fi;" + "" + "%s -q 2>&1 | grep -q 'requires an argument';" + "if [ $? -eq 0 ] ; then" + " CMD='-q 0';" + "else" + " CMD='';" + "fi;" + "%s $CMD -U %s;"); + + if (virAsprintf(&nc_command, nc_detect_template, + nc_bin, nc_bin, nc_bin, sock_path) < 0) + goto out_of_memory; + nr_args += 4; /* ssh $hostname $netcat_command NULL*/ if (username) nr_args += 2; /* For -l username */ if (no_tty) nr_args += 5; /* For -T -o BatchMode=yes -e none */ if (port) nr_args += 2; /* For -p port */ @@ -765,12 +806,9 @@ doRemoteOpen (virConnectPtr conn, cmd_argv[j++] = strdup ("none"); } cmd_argv[j++] = strdup (priv->hostname); - cmd_argv[j++] = strdup (netcat ? netcat : "nc"); - cmd_argv[j++] = strdup ("-U"); - cmd_argv[j++] = strdup (sockname ? sockname : - (flags & VIR_CONNECT_RO - ? LIBVIRTD_PRIV_UNIX_SOCKET_RO - : LIBVIRTD_PRIV_UNIX_SOCKET)); + + cmd_argv[j++] = nc_command; + cmd_argv[j++] = 0; assert (j == nr_args); for (j = 0; j < (nr_args-1); j++) -- 1.6.5.2

3 7

[libvirt] [PATCH] vboxDomainDumpXML: avoid a leak on OOM error path
by Jim Meyering 15 Feb '10

15 Feb '10

Yet another... >From 05e87caf9da26f9038935d7e54b6038b9362875b Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 15 Feb 2010 20:09:55 +0100 Subject: [PATCH] vboxDomainDumpXML: avoid a leak on OOM error path * src/vbox/vbox_tmpl.c (vboxDomainDumpXML): Free vboxCallback buffer upon OOM. --- src/vbox/vbox_tmpl.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/src/vbox/vbox_tmpl.c b/src/vbox/vbox_tmpl.c index d1a701e..8a9af52 100644 --- a/src/vbox/vbox_tmpl.c +++ b/src/vbox/vbox_tmpl.c @@ -5276,10 +5276,11 @@ static nsresult vboxCallbackQueryInterface(nsISupports *pThis, const nsID *iid, static IVirtualBoxCallback *vboxAllocCallbackObj(void) { IVirtualBoxCallback *vboxCallback = NULL; - /* Allocate, Initialize and return a validi + /* Allocate, Initialize and return a valid * IVirtualBoxCallback object here */ if ((VIR_ALLOC(vboxCallback) < 0) || (VIR_ALLOC(vboxCallback->vtbl) < 0)) { + VIR_FREE(vboxCallback); virReportOOMError(); return NULL; } -- 1.7.0.181.g41533

2 1

[libvirt] [PATCH] virNodeDevCapScsiHostParseXML: avoid an unconditional leak
by Jim Meyering 15 Feb '10

15 Feb '10

It looks like there may still be error-path leaks (the buffers pointed to by nodes[i]), but that can wait for another day. >From c293a97ed55c99c87bc658aaffc10831f8ba8c68 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 15 Feb 2010 19:54:45 +0100 Subject: [PATCH] virNodeDevCapScsiHostParseXML: avoid an unconditional leak * src/conf/node_device_conf.c (virNodeDevCapScsiHostParseXML): Free the "nodes" buffer allocated by virXPathNodeSet. --- src/conf/node_device_conf.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/src/conf/node_device_conf.c b/src/conf/node_device_conf.c index 4b65953..09c0f41 100644 --- a/src/conf/node_device_conf.c +++ b/src/conf/node_device_conf.c @@ -797,6 +797,7 @@ virNodeDevCapScsiHostParseXML(xmlXPathContextPtr ctxt, out: VIR_FREE(type); ctxt->node = orignode; + VIR_FREE(nodes); return ret; } -- 1.7.0.181.g41533

2 1

[libvirt] [PATCH] uml_driver.c: avoid leak upon failure
by Jim Meyering 15 Feb '10

15 Feb '10

Without this, it looked like every caller (there's only one; the other use is if-0'd out) could leak the result buffer. >From 8b8874b956c822562151cb90b1a8950d17b565ac Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 15 Feb 2010 19:22:38 +0100 Subject: [PATCH] uml_driver.c: avoid leak upon failure * src/uml/uml_driver.c (umlMonitorCommand): This function would sometimes return -1, yet fail to free the "reply" it had allocated. Hence, no caller would know to free the corresponding argument. When returning -1, be sure to free all allocated resources. --- src/uml/uml_driver.c | 7 +++++-- 1 files changed, 5 insertions(+), 2 deletions(-) diff --git a/src/uml/uml_driver.c b/src/uml/uml_driver.c index 5049c92..10268db 100644 --- a/src/uml/uml_driver.c +++ b/src/uml/uml_driver.c @@ -1,7 +1,7 @@ /* * uml_driver.c: core driver methods for managing UML guests * - * Copyright (C) 2006, 2007, 2008, 2009 Red Hat, Inc. + * Copyright (C) 2006-2010 Red Hat, Inc. * Copyright (C) 2006-2008 Daniel P. Berrange * * This library is free software; you can redistribute it and/or @@ -757,7 +757,10 @@ static int umlMonitorCommand(virConnectPtr conn, VIR_DEBUG("Command reply is '%s'", NULLSTR(retdata)); - *reply = retdata; + if (ret < 0) + VIR_FREE(retdata); + else + *reply = retdata; return ret; -- 1.7.0.181.g41533

2 1

[libvirt] [PATCH 2/4] qemu: Make SetVcpu command hotplug only
by Cole Robinson 15 Feb '10

15 Feb '10

Similar to the Set*Mem commands, this implementation was bogus and misleading. Make it clear this is a hotplug only operation, and that the hotplug piece isn't even implemented. Also drop the overkill maxvcpus validation: we don't perform this check at XML define time so clearly no one is missing it, and there is always the risk that our info will be out of date, possibly preventing legitimate CPU values. --- src/qemu/qemu_driver.c | 42 +++++++----------------------------------- 1 files changed, 7 insertions(+), 35 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 56a450c..ef1f638 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -4100,12 +4100,11 @@ cleanup: } -static int qemudDomainSetVcpus(virDomainPtr dom, unsigned int nvcpus) { +static int qemudDomainSetVcpus(virDomainPtr dom, + ATTRIBUTE_UNUSED unsigned int nvcpus) { struct qemud_driver *driver = dom->conn->privateData; virDomainObjPtr vm; - int max; int ret = -1; - const char *type; qemuDriverLock(driver); vm = virDomainFindByUUID(&driver->domains, dom->uuid); @@ -4119,41 +4118,14 @@ static int qemudDomainSetVcpus(virDomainPtr dom, unsigned int nvcpus) { goto cleanup; } - if (qemuDomainObjBeginJob(vm) < 0) + if (!virDomainObjIsActive(vm)) { + qemuReportError(VIR_ERR_OPERATION_INVALID, + "%s", _("domain is not running")); goto cleanup; - - if (virDomainObjIsActive(vm)) { - qemuReportError(VIR_ERR_OPERATION_INVALID, "%s", - _("cannot change vcpu count of an active domain")); - goto endjob; - } - - if (!(type = virDomainVirtTypeToString(vm->def->virtType))) { - qemuReportError(VIR_ERR_INTERNAL_ERROR, - _("unknown virt type in domain definition '%d'"), - vm->def->virtType); - goto endjob; } - if ((max = qemudGetMaxVCPUs(NULL, type)) < 0) { - qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s", - _("could not determine max vcpus for the domain")); - goto endjob; - } - - if (nvcpus > max) { - qemuReportError(VIR_ERR_INVALID_ARG, - _("requested vcpus is greater than max allowable" - " vcpus for the domain: %d > %d"), nvcpus, max); - goto endjob; - } - - vm->def->vcpus = nvcpus; - ret = 0; - -endjob: - if (qemuDomainObjEndJob(vm) == 0) - vm = NULL; + qemuReportError(VIR_ERR_NO_SUPPORT, + "%s", _("cpu hotplug not yet supported")); cleanup: if (vm) -- 1.6.5.2

2 2

[libvirt] [PATCH] openvz_driver.c: avoid leak on OOM error path
by Jim Meyering 15 Feb '10

15 Feb '10

It looks like openvzFreeDriver cannot call VIR_FREE on its buffer argument, because of the use in the function that just happens to be a few lines below. These are the only two uses of openvzFreeDriver. >From 0a616cd03cca9c9d89f23dadd52cd23c30789aca Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 15 Feb 2010 18:33:38 +0100 Subject: [PATCH] openvz_driver.c: avoid leak on OOM error path * src/openvz/openvz_driver.c (openvzOpen): Free "driver" buffer upon failure. --- src/openvz/openvz_driver.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/src/openvz/openvz_driver.c b/src/openvz/openvz_driver.c index 68d0398..3a8a82f 100644 --- a/src/openvz/openvz_driver.c +++ b/src/openvz/openvz_driver.c @@ -1245,22 +1245,23 @@ static virDrvOpenStatus openvzOpen(virConnectPtr conn, goto cleanup; if (openvzExtractVersion(conn, driver) < 0) goto cleanup; conn->privateData = driver; return VIR_DRV_OPEN_SUCCESS; cleanup: openvzFreeDriver(driver); + VIR_FREE(driver); return VIR_DRV_OPEN_ERROR; }; static int openvzClose(virConnectPtr conn) { struct openvz_driver *driver = conn->privateData; openvzFreeDriver(driver); conn->privateData = NULL; return 0; } -- 1.7.0.181.g41533

2 3

[libvirt] [PATCH 3/5] [FIX] macvtap support for libvirt -- qemu support
by Stefan Berger 15 Feb '10

15 Feb '10

I needed to fix the case where macvtap support is not to be compiled in. This part adds support for qemu making a macvtap tap device available via file descriptor passed to qemu command line. This also attempts to tear down the macvtap device when a VM terminates. This includes support for attachment and detachment to/from running VM. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com>

3 3

[libvirt] [PATCH] Treat missing QEMU 'thread_id' as non-fatal in JSON monitor
by Daniel P. Berrange 15 Feb '10

15 Feb '10

The plain QEMU tree does not include 'thread_id' in the JSON output. Thus we need to treat it as non-fatal if missing. * src/qemu/qemu_monitor_json.c: Treat missing thread_id as non-fatal --- src/qemu/qemu_monitor_json.c | 5 +++-- 1 files changed, 3 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index 9875f38..49e0370 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -630,8 +630,9 @@ qemuMonitorJSONExtractCPUInfo(virJSONValuePtr reply, } if (virJSONValueObjectGetNumberInt(entry, "thread_id", &thread) < 0) { - qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s", - _("cpu information was missing thread ID")); + /* Only qemu-kvm tree includs thread_id, so treat this as + non-fatal, simply returning no data */ + ret = 0; goto cleanup; } -- 1.6.6

2 1

[libvirt] [PATCH] Make error reporting for QEMU JSON mode more friendly
by Daniel P. Berrange 15 Feb '10

15 Feb '10

Current error reporting for JSON mode returns the full JSON command string and full JSON error string. This is not very user friendly, so this change makes the error report only contain the basic command name, and friendly error message description string. The full JSON data is logged instead. * src/qemu/qemu_monitor_json.c: Always return the 'desc' field from the JSON error message to users. --- src/qemu/qemu_monitor_json.c | 58 ++++++++++++++++++++++++++++------------- 1 files changed, 39 insertions(+), 19 deletions(-) diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index f0dcf81..9875f38 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -246,15 +246,33 @@ qemuMonitorJSONCommand(qemuMonitorPtr mon, * * XXX see qerror.h for different klasses & fill out useful params */ -static char *qemuMonitorJSONStringifyError(virJSONValuePtr error) +static const char * +qemuMonitorJSONStringifyError(virJSONValuePtr error) { const char *klass = virJSONValueObjectGetString(error, "class"); + const char *detail = NULL; - if (klass) { - return strdup(klass); - } else { - return strdup(_("Missing QEMU error klass")); - } + /* The QMP 'desc' field is usually sufficient for our generic + * error reporting needs. + */ + if (klass) + detail = virJSONValueObjectGetString(error, "desc"); + + + if (!detail) + detail = "unknown QEMU command error"; + + return detail; +} + +static const char * +qemuMonitorJSONCommandName(virJSONValuePtr cmd) +{ + const char *name = virJSONValueObjectGetString(cmd, "execute"); + if (name) + return name; + else + return "<unknown>"; } static int @@ -266,20 +284,21 @@ qemuMonitorJSONCheckError(virJSONValuePtr cmd, char *cmdstr = virJSONValueToString(cmd); char *replystr = virJSONValueToString(reply); - if (!error) { - VIR_DEBUG("Saw a JSON error, but value is null for %s: %s", - cmdstr, replystr); + /* Log the full JSON formatted command & error */ + VIR_DEBUG("unable to execute QEMU command %s: %s", + cmdstr, replystr); + + /* Only send the user the command name + friendly error */ + if (!error) qemuReportError(VIR_ERR_INTERNAL_ERROR, - _("error running QEMU command '%s': '%s'"), - cmdstr, replystr); - } else { - VIR_DEBUG("Got a JSON error set for %s", cmdstr); - char *detail = qemuMonitorJSONStringifyError(error); + _("unable to execute QEMU command '%s'"), + qemuMonitorJSONCommandName(cmd)); + else qemuReportError(VIR_ERR_INTERNAL_ERROR, - _("error running QEMU command '%s': %s ('%s')"), - cmdstr, detail, replystr); - VIR_FREE(detail); - } + _("unable to execute QEMU command '%s': %s"), + qemuMonitorJSONCommandName(cmd), + qemuMonitorJSONStringifyError(error)); + VIR_FREE(cmdstr); VIR_FREE(replystr); return -1; @@ -290,7 +309,8 @@ qemuMonitorJSONCheckError(virJSONValuePtr cmd, VIR_DEBUG("Neither 'return' nor 'error' is set in the JSON reply %s: %s", cmdstr, replystr); qemuReportError(VIR_ERR_INTERNAL_ERROR, - _("error running QEMU command '%s': '%s'"), cmdstr, replystr); + _("unable to execute QEMU command '%s'"), + qemuMonitorJSONCommandName(cmd)); VIR_FREE(cmdstr); VIR_FREE(replystr); return -1; -- 1.6.6

2 1

[libvirt] [PATCH] Fix check for primary IDE controller in QEMU PCI slot assignment
by Daniel P. Berrange 15 Feb '10

15 Feb '10

A typo in the check for the primary IDE controller could cause a crash on restore depending on the exact guest config. * src/qemu/qemu_conf.c: Fix s/video/controller/ typo & slot number typo --- src/qemu/qemu_conf.c | 8 ++++---- 1 files changed, 4 insertions(+), 4 deletions(-) diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c index 00f761d..dd5efd6 100644 --- a/src/qemu/qemu_conf.c +++ b/src/qemu/qemu_conf.c @@ -2085,10 +2085,10 @@ qemuAssignDevicePCISlots(virDomainDefPtr def, qemuDomainPCIAddressSetPtr addrs) if (def->controllers[i]->type == VIR_DOMAIN_CONTROLLER_TYPE_IDE && def->controllers[i]->idx == 0) { if (def->controllers[i]->info.type == VIR_DOMAIN_DEVICE_ADDRESS_TYPE_PCI) { - if (def->videos[i]->info.addr.pci.domain != 0 || - def->videos[i]->info.addr.pci.bus != 0 || - def->videos[i]->info.addr.pci.slot != 2 || - def->videos[i]->info.addr.pci.function != 0) { + if (def->controllers[i]->info.addr.pci.domain != 0 || + def->controllers[i]->info.addr.pci.bus != 0 || + def->controllers[i]->info.addr.pci.slot != 1 || + def->controllers[i]->info.addr.pci.function != 1) { qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Primary IDE controller must have PCI address 0:0:1.1")); goto error; -- 1.6.6

2 1

[libvirt] [PATCH] interface: Use proper return codes in the open function
by Matthias Bolte 15 Feb '10

15 Feb '10

The open function returned -1 in case of an error, but -1 maps to VIR_DRV_OPEN_DECLINED instead of VIR_DRV_OPEN_ERROR. --- src/interface/netcf_driver.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/interface/netcf_driver.c b/src/interface/netcf_driver.c index 2753049..7f4d43d 100644 --- a/src/interface/netcf_driver.c +++ b/src/interface/netcf_driver.c @@ -139,7 +139,7 @@ static virDrvOpenStatus interfaceOpenInterface(virConnectPtr conn, } conn->interfacePrivateData = driverState; - return 0; + return VIR_DRV_OPEN_SUCCESS; netcf_error: if (driverState->netcf) @@ -150,7 +150,7 @@ netcf_error: mutex_error: VIR_FREE(driverState); alloc_error: - return -1; + return VIR_DRV_OPEN_ERROR; } static int interfaceCloseInterface(virConnectPtr conn) -- 1.6.3.3

2 2

Re: [libvirt] Cannot enumerate physical devices with libvirt 0.7.6
by Frédéric Grelot 15 Feb '10

15 Feb '10

> > # libvirtd --version > > libvirtd (libvirt) 0.7.6 > > > > # virsh --version > > 0.7.6 > > > > # virsh nodedev-list > > error :Failed to count node devices > > error :this function is not supported by the hypervisor: > virNodeNumOfDevices I restarted the server, with no change. I uninstalled and reinstalled libvirt, same result. I had a look at libvirt logs, there is nothing special. I tried several other things, with no luck. I check that "lshal" gives result : it is actually the case... > It would appear that the udev driver failed to start - there are > probably > messages in syslog about the problem I'll check that. However I don't know what happened with this last update, physical devices worked great before it... If what you say is right, it may be linked with some other update that occured at the same time... Thanks for you help! > > Regards, > Daniel > -- > |: Red Hat, Engineering, London -o- > http://people.redhat.com/berrange/ :| > |: http://libvirt.org -o- http://virt-manager.org -o- > http://ovirt.org :| > |: http://autobuild.org -o- > http://search.cpan.org/~danberr/ :| > |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B > 9505 :|

1 0

[libvirt] Cannot enumerate physical devices with libvirt 0.7.6
by Frédéric Grelot 15 Feb '10

15 Feb '10

Hi all, I often update my F12 box with the rawvirt repository, which just installed libvirt 0.7.6. Since that, I cannot enumerate physical devices (and thus get a beautiful error when I start one of my vms that should use a USB device). I don't think I changed anything else on that box, and, of course, I already restarted the libvirt daemon. I haven't restarted the server itself, because it is currently used, but I could do it if necessary... I make good use of the qemu hypervisor, and I'm looking forward to solve this issue! Frederic. # libvirtd --version libvirtd (libvirt) 0.7.6 # virsh --version 0.7.6 # virsh nodedev-list error :Failed to count node devices error :this function is not supported by the hypervisor: virNodeNumOfDevices

2 1

[libvirt] [PATCH] Convert virSecurityReportError into a macro
by Matthias Bolte 15 Feb '10

15 Feb '10

--- src/security/security_driver.c | 18 ------------------ src/security/security_driver.h | 6 +++--- 2 files changed, 3 insertions(+), 21 deletions(-) diff --git a/src/security/security_driver.c b/src/security/security_driver.c index 27945a6..4c98190 100644 --- a/src/security/security_driver.c +++ b/src/security/security_driver.c @@ -90,24 +90,6 @@ virSecurityDriverStartup(virSecurityDriverPtr *drv, return -2; } -void -virSecurityReportError(int code, const char *fmt, ...) -{ - va_list args; - char errorMessage[1024]; - - if (fmt) { - va_start(args, fmt); - vsnprintf(errorMessage, sizeof(errorMessage) - 1, fmt, args); - va_end(args); - } else - errorMessage[0] = '\0'; - - virRaiseError(NULL, NULL, NULL, VIR_FROM_SECURITY, code, - VIR_ERR_ERROR, NULL, NULL, NULL, -1, -1, "%s", - errorMessage); -} - /* * Helpers */ diff --git a/src/security/security_driver.h b/src/security/security_driver.h index 8860d81..15671b3 100644 --- a/src/security/security_driver.h +++ b/src/security/security_driver.h @@ -88,9 +88,9 @@ int virSecurityDriverStartup(virSecurityDriverPtr *drv, int virSecurityDriverVerify(virDomainDefPtr def); -void -virSecurityReportError(int code, const char *fmt, ...) - ATTRIBUTE_FMT_PRINTF(2, 3); +#define virSecurityReportError(code, fmt...) \ + virReportErrorHelper(NULL, VIR_FROM_SECURITY, code, __FILE__, \ + __FUNCTION__, __LINE__, fmt) /* Helpers */ void virSecurityDriverInit(virSecurityDriverPtr drv); -- 1.6.3.3

2 2

[libvirt] How to shut down every VM when libvirt is stopped
by Laurent Léonard 13 Feb '10

13 Feb '10

Hi, With libvirt 0.4.6 for example, the behaviour when we stopped libvirt was to shut down every VM. That was a problem in many cases (upgrades, etc.) but that behaviour was very useful with high avalability systems like Heartbeat. With an Hearbeat + DRBD system, if the libvirt service is stopped by Heartbeat, every VM should be stopped too. Otherwise we will experience problems and disk corruption. Is there any way to configure libvirt to have that old behaviour ? Or do I have to shut down/kill every VM in the stop target in the init script ? Thank you, -- Laurent Léonard

1 0

[libvirt] [PATCH] Add persistence of PCI addresses to QEMU
by Daniel P. Berrange 12 Feb '10

12 Feb '10

Current PCI addresses are allocated at time of VM startup. To make them truely persistent, it is neccessary to do this at time of virDomainDefine/virDomainCreate. The code in qemuStartVMDaemon still remains in order to cope with upgrades from older libvirt releases * src/qemu/qemu_driver.c: Rename existing qemuAssignPCIAddresses to qemuDetectPCIAddresses. Add new qemuAssignPCIAddresses which does auto-allocation upfront. Call qemuAssignPCIAddresses from qemuDomainDefine and qemuDomainCreate to assign PCI addresses that can then be persisted. Don't clear PCI addresses at shutdown if they are intended to be persistent --- src/qemu/qemu_driver.c | 84 +++++++++++++++++++++++++++++++++++++++++++++--- 1 files changed, 79 insertions(+), 5 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 03d0f5f..69187fc 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -95,6 +95,7 @@ struct _qemuDomainObjPrivate { int *vcpupids; qemuDomainPCIAddressSetPtr pciaddrs; + int persistentAddrs; }; static int qemudShutdown(void); @@ -857,6 +858,7 @@ qemuReconnectDomain(void *payload, const char *name ATTRIBUTE_UNUSED, void *opaq virDomainObjPtr obj = payload; struct qemud_driver *driver = opaque; qemuDomainObjPrivatePtr priv; + unsigned long long qemuCmdFlags; virDomainObjLock(obj); @@ -872,6 +874,15 @@ qemuReconnectDomain(void *payload, const char *name ATTRIBUTE_UNUSED, void *opaq goto error; } + /* XXX we should be persisting the original flags in the XML + * not re-detecting them, since the binary may have changed + * since launch time */ + if (qemudExtractVersionInfo(obj->def->emulator, + NULL, + &qemuCmdFlags) >= 0 && + (qemuCmdFlags & QEMUD_CMD_FLAG_DEVICE)) + priv->persistentAddrs = 1; + if (!(priv->pciaddrs = qemuDomainPCIAddressSetCreate(obj->def))) goto error; @@ -1976,7 +1987,7 @@ qemuGetPCIWatchdogVendorProduct(virDomainWatchdogDefPtr def, * some static addrs on CLI. Have to check that... */ static int -qemuAssignPCIAddresses(virDomainObjPtr vm, +qemuDetectPCIAddresses(virDomainObjPtr vm, qemuMonitorPCIAddress *addrs, int naddrs) { @@ -2098,7 +2109,7 @@ qemuInitPCIAddresses(struct qemud_driver *driver, &addrs); qemuDomainObjExitMonitorWithDriver(driver, vm); - ret = qemuAssignPCIAddresses(vm, addrs, naddrs); + ret = qemuDetectPCIAddresses(vm, addrs, naddrs); VIR_FREE(addrs); @@ -2141,6 +2152,44 @@ static int qemudNextFreeVNCPort(struct qemud_driver *driver ATTRIBUTE_UNUSED) { return -1; } + +static int +qemuAssignPCIAddresses(virDomainDefPtr def) +{ + int ret = -1; + unsigned long long qemuCmdFlags = 0; + qemuDomainPCIAddressSetPtr addrs = NULL; + struct stat sb; + + if (stat(def->emulator, &sb) < 0) { + virReportSystemError(errno, + _("Cannot find QEMU binary %s"), + def->emulator); + goto cleanup; + } + + if (qemudExtractVersionInfo(def->emulator, + NULL, + &qemuCmdFlags) < 0) + goto cleanup; + + if (qemuCmdFlags & QEMUD_CMD_FLAG_DEVICE) { + if (!(addrs = qemuDomainPCIAddressSetCreate(def))) + goto cleanup; + + if (qemuAssignDevicePCISlots(def, addrs) < 0) + goto cleanup; + } + + ret = 0; + +cleanup: + qemuDomainPCIAddressSetFree(addrs); + + return ret; +} + + static pciDeviceList * qemuGetPciHostDeviceList(virDomainDefPtr def) { @@ -2662,7 +2711,15 @@ static int qemudStartVMDaemon(virConnectPtr conn, goto cleanup; } + /* + * Normally PCI addresses are assigned inhe virDomainCreate + * or virDomainDefine methods. We might still need to assign + * some here to cope with the question of upgrades. Regardless + * we also need to populate the PCi address set cache for later + * use in hotplug + */ if (qemuCmdFlags & QEMUD_CMD_FLAG_DEVICE) { + /* Populate cache with current addresses */ if (priv->pciaddrs) { qemuDomainPCIAddressSetFree(priv->pciaddrs); priv->pciaddrs = NULL; @@ -2670,8 +2727,14 @@ static int qemudStartVMDaemon(virConnectPtr conn, if (!(priv->pciaddrs = qemuDomainPCIAddressSetCreate(vm->def))) goto cleanup; + + /* Assign any remaining addresses */ if (qemuAssignDevicePCISlots(vm->def, priv->pciaddrs) < 0) goto cleanup; + + priv->persistentAddrs = 1; + } else { + priv->persistentAddrs = 0; } vm->def->id = driver->nextvmid++; @@ -2903,10 +2966,12 @@ static void qemudShutdownVMDaemon(struct qemud_driver *driver, VIR_FREE(vm->def->seclabel.imagelabel); } - virDomainDefClearPCIAddresses(vm->def); virDomainDefClearDeviceAliases(vm->def); - qemuDomainPCIAddressSetFree(priv->pciaddrs); - priv->pciaddrs = NULL; + if (!priv->persistentAddrs) { + virDomainDefClearPCIAddresses(vm->def); + qemuDomainPCIAddressSetFree(priv->pciaddrs); + priv->pciaddrs = NULL; + } qemuDomainReAttachHostDevices(driver, vm->def); @@ -3352,6 +3417,12 @@ static virDomainPtr qemudDomainCreate(virConnectPtr conn, const char *xml, if (virDomainObjIsDuplicate(&driver->domains, def, 1) < 0) goto cleanup; + if (qemudCanonicalizeMachine(driver, def) < 0) + goto cleanup; + + if (qemuAssignPCIAddresses(def) < 0) + goto cleanup; + if (!(vm = virDomainAssignDef(driver->caps, &driver->domains, def))) @@ -5049,6 +5120,9 @@ static virDomainPtr qemudDomainDefine(virConnectPtr conn, const char *xml) { if (qemudCanonicalizeMachine(driver, def) < 0) goto cleanup; + if (qemuAssignPCIAddresses(def) < 0) + goto cleanup; + if (!(vm = virDomainAssignDef(driver->caps, &driver->domains, def))) { -- 1.6.5.2

2 3

[libvirt] [PATCH] Fix compliation of AppArmor related code
by Matthias Bolte 12 Feb '10

12 Feb '10

Broken by the latest commits to remove the virConnectPtr parameter from internal functions. I just pushed this one, as it is an obvious fix. --- src/security/security_apparmor.c | 4 ++-- src/security/virt-aa-helper.c | 8 +++----- 2 files changed, 5 insertions(+), 7 deletions(-) diff --git a/src/security/security_apparmor.c b/src/security/security_apparmor.c index 23f40f8..db04d5c 100644 --- a/src/security/security_apparmor.c +++ b/src/security/security_apparmor.c @@ -262,7 +262,7 @@ use_apparmor(void) char *libvirt_daemon = NULL; if (virFileResolveLink("/proc/self/exe", &libvirt_daemon) < 0) { - virSecurityReportError(NULL, VIR_ERR_INTERNAL_ERROR, + virSecurityReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("could not find libvirtd")); return rc; } @@ -295,7 +295,7 @@ AppArmorSecurityDriverProbe(void) } if (!virFileExists(template)) { - virSecurityReportError(NULL, VIR_ERR_INTERNAL_ERROR, + virSecurityReportError(VIR_ERR_INTERNAL_ERROR, _("template \'%s\' does not exist"), template); goto clean; } diff --git a/src/security/virt-aa-helper.c b/src/security/virt-aa-helper.c index 619c8c3..066e18b 100644 --- a/src/security/virt-aa-helper.c +++ b/src/security/virt-aa-helper.c @@ -690,7 +690,7 @@ get_definition(vahControl * ctl, const char *xmlStr) goto exit; } - ctl->def = virDomainDefParseString(NULL, ctl->caps, xmlStr, 0); + ctl->def = virDomainDefParseString(ctl->caps, xmlStr, 0); if (ctl->def == NULL) { vah_error(ctl, 0, "could not parse XML"); goto exit; @@ -767,8 +767,7 @@ vah_add_file(virBufferPtr buf, const char *path, const char *perms) } static int -file_iterate_cb(virConnectPtr conn ATTRIBUTE_UNUSED, - usbDevice *dev ATTRIBUTE_UNUSED, +file_iterate_cb(usbDevice *dev ATTRIBUTE_UNUSED, const char *file, void *opaque) { virBufferPtr buf = opaque; @@ -844,8 +843,7 @@ get_files(vahControl * ctl) if (usb == NULL) continue; - rc = usbDeviceFileIterate(NULL, usb, - file_iterate_cb, &buf); + rc = usbDeviceFileIterate(usb, file_iterate_cb, &buf); usbFreeDevice(usb); if (rc != 0) goto clean; -- 1.6.3.3

2 1

[libvirt] dname parameter ignored in qemudDomainMigratePrepareTunnel{, 2}
by Jim Meyering 11 Feb '10

11 Feb '10

clang warned about a dead-store in src/qemu/qemu_driver.c, since dname is never used thereafter: http://libvirt.org/git/?p=libvirt.git;a=blob;f=src/qemu/qemu_driver.c;h=0d7… /* Target domain name, maybe renamed. */ dname = dname ? dname : def->name; That stmt was initially added with a following use, if (!vm) vm = virDomainFindByName(&driver->domains, dname); But the use was removed by this commit, rendering the store "dead", and the dname parameter effectively unused: http://libvirt.org/git/?p=libvirt.git;a=commitdiff;h=c26cb9234f4b9fa46d7caa… The same thing happened in both qemudDomainMigratePrepareTunnel and qemudDomainMigratePrepareTunnel2 Shouldn't they be doing something like this instead? if (dname) def->name = dname; But def->name is already allocated. Better free it first. And is dname "known" to be allocated from the heap? That's not clear from the little documentation I've read so far, so I've strdup'd it below: Here's a possible patch: >From b51a39aed62ee5c8db733cecfe6eef0c34a7f989 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Thu, 11 Feb 2010 16:46:21 +0100 Subject: [PATCH] qemu_driver.c: honor dname parameter once again Since c26cb9234f4b9fa46d7caa3385ae36704167c53f, the dname parameter has been ignored by these two functions. Use it. * src/qemu/qemu_driver.c (qemudDomainMigratePrepareTunnel): Honor dname parameter once again. (qemudDomainMigratePrepare2): Likewise. --- src/qemu/qemu_driver.c | 14 ++++++++++++-- 1 files changed, 12 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 0d77d57..9ff712c 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -7434,7 +7434,12 @@ qemudDomainMigratePrepareTunnel(virConnectPtr dconn, } /* Target domain name, maybe renamed. */ - dname = dname ? dname : def->name; + if (dname) { + VIR_FREE(def->name); + def->name = strdup(dname); + if (def->name == NULL) + goto cleanup; + } if (virDomainObjIsDuplicate(&driver->domains, def, 1) < 0) goto cleanup; @@ -7660,7 +7665,12 @@ qemudDomainMigratePrepare2 (virConnectPtr dconn, } /* Target domain name, maybe renamed. */ - dname = dname ? dname : def->name; + if (dname) { + VIR_FREE(def->name); + def->name = strdup(dname); + if (def->name == NULL) + goto cleanup; + } if (virDomainObjIsDuplicate(&driver->domains, def, 1) < 0) goto cleanup; -- 1.7.0.rc2.170.gbc565

2 2

[libvirt] [PATCH] Mark all error messages for translation
by Jiri Denemark 11 Feb '10

11 Feb '10

Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/conf/storage_conf.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/src/conf/storage_conf.c b/src/conf/storage_conf.c index eb1c9d9..2352637 100644 --- a/src/conf/storage_conf.c +++ b/src/conf/storage_conf.c @@ -1464,7 +1464,7 @@ virStoragePoolObjLoad(virStoragePoolObjListPtr pools, if (!virFileMatchesNameSuffix(file, def->name, ".xml")) { virStorageReportError(VIR_ERR_INVALID_STORAGE_POOL, - "Storage pool config filename '%s' does not match pool name '%s'", + _("Storage pool config filename '%s' does not match pool name '%s'"), path, def->name); virStoragePoolDefFree(def); return NULL; @@ -1524,7 +1524,7 @@ virStoragePoolLoadAllConfigs(virStoragePoolObjListPtr pools, if (virFileBuildPath(configDir, entry->d_name, NULL, path, PATH_MAX) < 0) { virStorageReportError(VIR_ERR_INTERNAL_ERROR, - "Config filename '%s/%s' is too long", + _("Config filename '%s/%s' is too long"), configDir, entry->d_name); continue; } @@ -1532,7 +1532,7 @@ virStoragePoolLoadAllConfigs(virStoragePoolObjListPtr pools, if (virFileBuildPath(autostartDir, entry->d_name, NULL, autostartLink, PATH_MAX) < 0) { virStorageReportError(VIR_ERR_INTERNAL_ERROR, - "Autostart link path '%s/%s' is too long", + _("Autostart link path '%s/%s' is too long"), autostartDir, entry->d_name); continue; } -- 1.6.6.1

2 2

[libvirt] [PATCH] remote_driver.c: avoid leak on OOM error path
by Jim Meyering 11 Feb '10

11 Feb '10

Any "goto out_of_memory" taken between the vars = qparam_query_parse (query) allocation and the free_qparam_set call would result in a leak. Here's the fix: >From a3a21142e40962c960eae882ac9edcc3bd17cb8b Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Thu, 11 Feb 2010 12:51:31 +0100 Subject: [PATCH] remote_driver.c: avoid leak on OOM error path * src/remote/remote_driver.c (doRemoteOpen): Don't leak a qparam_set buffer upon OOM error. --- src/remote/remote_driver.c | 7 +++++-- 1 files changed, 5 insertions(+), 2 deletions(-) diff --git a/src/remote/remote_driver.c b/src/remote/remote_driver.c index e17db3d..cc543fc 100644 --- a/src/remote/remote_driver.c +++ b/src/remote/remote_driver.c @@ -2,7 +2,7 @@ * remote_internal.c: driver to provide access to libvirtd running * on a remote machine * - * Copyright (C) 2007-2009 Red Hat, Inc. + * Copyright (C) 2007-2010 Red Hat, Inc. * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public @@ -370,6 +370,7 @@ doRemoteOpen (virConnectPtr conn, virConnectAuthPtr auth ATTRIBUTE_UNUSED, int flags) { + struct qparam_set *vars = NULL; int wakeupFD[2] = { -1, -1 }; char *transport_str = NULL; enum { @@ -458,7 +459,6 @@ doRemoteOpen (virConnectPtr conn, * feasibly it might contain variables needed by the real driver, * although that won't be the case for now). */ - struct qparam_set *vars; struct qparam *var; int i; char *query; @@ -551,6 +551,7 @@ doRemoteOpen (virConnectPtr conn, } free_qparam_set (vars); + vars = NULL; } else { /* Probe URI server side */ name = strdup(""); @@ -933,6 +934,8 @@ doRemoteOpen (virConnectPtr conn, out_of_memory: virReportOOMError(); + if (vars) + free_qparam_set (vars); failed: /* Close the socket if we failed. */ -- 1.7.0.rc2.170.gbc565

2 2

[libvirt] [PATCH] Swap position of nmodels and models parameters in cpuDecode()
by Jiri Denemark 11 Feb '10

11 Feb '10

All other libvirt functions use array first and then number of elements in that array. Let's make cpuDecode follow this rule. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/cpu/cpu.c | 6 +++--- src/cpu/cpu.h | 8 ++++---- src/cpu/cpu_x86.c | 4 ++-- src/qemu/qemu_conf.c | 4 ++-- 4 files changed, 11 insertions(+), 11 deletions(-) diff --git a/src/cpu/cpu.c b/src/cpu/cpu.c index ae8d37d..e9ecc98 100644 --- a/src/cpu/cpu.c +++ b/src/cpu/cpu.c @@ -119,8 +119,8 @@ cpuCompare(virCPUDefPtr host, int cpuDecode(virCPUDefPtr cpu, const union cpuData *data, - unsigned int nmodels, - const char **models) + const char **models, + unsigned int nmodels) { struct cpuArchDriver *driver; @@ -146,7 +146,7 @@ cpuDecode(virCPUDefPtr cpu, return -1; } - return driver->decode(cpu, data, nmodels, models); + return driver->decode(cpu, data, models, nmodels); } diff --git a/src/cpu/cpu.h b/src/cpu/cpu.h index 1e04993..7ee0ce6 100644 --- a/src/cpu/cpu.h +++ b/src/cpu/cpu.h @@ -48,8 +48,8 @@ typedef virCPUCompareResult typedef int (*cpuArchDecode) (virCPUDefPtr cpu, const union cpuData *data, - unsigned int nmodels, - const char **models); + const char **models, + unsigned int nmodels); typedef int (*cpuArchEncode) (const virCPUDefPtr cpu, @@ -95,8 +95,8 @@ cpuCompare (virCPUDefPtr host, extern int cpuDecode (virCPUDefPtr cpu, const union cpuData *data, - unsigned int nmodels, - const char **models); + const char **models, + unsigned int nmodels); extern int cpuEncode (const char *arch, diff --git a/src/cpu/cpu_x86.c b/src/cpu/cpu_x86.c index 7ae7fbf..c6ed078 100644 --- a/src/cpu/cpu_x86.c +++ b/src/cpu/cpu_x86.c @@ -940,8 +940,8 @@ x86GuestData(virCPUDefPtr host, static int x86Decode(virCPUDefPtr cpu, const union cpuData *data, - unsigned int nmodels, - const char **models) + const char **models, + unsigned int nmodels) { int ret = -1; struct x86_map *map; diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c index 3988582..ca32dc0 100644 --- a/src/qemu/qemu_conf.c +++ b/src/qemu/qemu_conf.c @@ -1016,7 +1016,7 @@ qemudCapsInitCPU(virCapsPtr caps, cpu->threads = nodeinfo.threads; if (!(data = cpuNodeData(arch)) - || cpuDecode(cpu, data, 0, NULL) < 0) + || cpuDecode(cpu, data, NULL, 0) < 0) goto error; caps->host.cpu = cpu; @@ -2959,7 +2959,7 @@ qemuBuildCpuArgStr(const struct qemud_driver *driver, if (VIR_ALLOC(guest) < 0 || !(guest->arch = strdup(ut->machine))) goto no_memory; - if (cpuDecode(guest, data, ncpus, cpus) < 0) + if (cpuDecode(guest, data, cpus, ncpus) < 0) goto cleanup; virBufferVSprintf(&buf, "%s", guest->model); -- 1.6.6.1

2 2

[libvirt] [PATCH] Fix <cpu> element in domain XML schema
by Jiri Denemark 11 Feb '10

11 Feb '10

The current schema is more permissive than the XML parsing code in libvirt. Precisely, 'match' attribute is optional in schema while in reality its use is bound to <model> element: - <cpu> element without 'match' attribute is allowed only if <topology> element is the only child element of <cpu> - <cpu> element with 'match' attribute requires <model> element to be present; <topology> and <feature> elements are optional Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- docs/schemas/domain.rng | 109 +++++++++++++++++++++++++++-------------------- 1 files changed, 63 insertions(+), 46 deletions(-) diff --git a/docs/schemas/domain.rng b/docs/schemas/domain.rng index bb6d00d..c913a50 100644 --- a/docs/schemas/domain.rng +++ b/docs/schemas/domain.rng @@ -1313,52 +1313,69 @@ --> <define name="cpu"> <element name="cpu"> - <optional> - <attribute name="match"> - <choice> - <value>minimum</value> - <value>exact</value> - <value>strict</value> - </choice> - </attribute> - </optional> - <interleave> - <optional> - <element name="model"> - <text/> - </element> - </optional> - <optional> - <element name="topology"> - <attribute name="sockets"> - <ref name="positiveInteger"/> - </attribute> - <attribute name="cores"> - <ref name="positiveInteger"/> - </attribute> - <attribute name="threads"> - <ref name="positiveInteger"/> - </attribute> - </element> - </optional> - <zeroOrMore> - <element name="feature"> - <attribute name="policy"> - <choice> - <value>force</value> - <value>require</value> - <value>optional</value> - <value>disable</value> - <value>forbid</value> - </choice> - </attribute> - <attribute name="name"> - <ref name="featureName"/> - </attribute> - <empty/> - </element> - </zeroOrMore> - </interleave> + <choice> + <ref name="cpuTopology"/> + <group> + <ref name="cpuMatch"/> + <interleave> + <ref name="cpuModel"/> + <optional> + <ref name="cpuTopology"/> + </optional> + <zeroOrMore> + <ref name="cpuFeature"/> + </zeroOrMore> + </interleave> + </group> + </choice> + </element> + </define> + + <define name="cpuMatch"> + <attribute name="match"> + <choice> + <value>minimum</value> + <value>exact</value> + <value>strict</value> + </choice> + </attribute> + </define> + + <define name="cpuModel"> + <element name="model"> + <text/> + </element> + </define> + + <define name="cpuFeature"> + <element name="feature"> + <attribute name="policy"> + <choice> + <value>force</value> + <value>require</value> + <value>optional</value> + <value>disable</value> + <value>forbid</value> + </choice> + </attribute> + <attribute name="name"> + <ref name="featureName"/> + </attribute> + <empty/> + </element> + </define> + + <define name="cpuTopology"> + <element name="topology"> + <attribute name="sockets"> + <ref name="positiveInteger"/> + </attribute> + <attribute name="cores"> + <ref name="positiveInteger"/> + </attribute> + <attribute name="threads"> + <ref name="positiveInteger"/> + </attribute> </element> </define> -- 1.6.6.1

2 2

[libvirt] [PATCH 0/9] New public API for computing baseline CPU
by Jiri Denemark 11 Feb '10

11 Feb '10

Hi, This series introduces a new virConnectBaselineCPU() API call for computing baseline CPU model from a set of host CPU definitions. The call accepts a list of <cpu>...</cpu> xml definitions taken from host capabilities and returns an xml description of a guest CPU which is supported by all hosts. virsh is little bit less picky and accepts a file containing host CPU definitions. It doesn't care what is outside of the <cpu/> tag, which allows for concatenating outputs of virsh capabilities from several hosts and feeding the result to virsh cpu-baseline. This series can also be found on cpu-baseline branch of git://gitorious.org/~jirka/libvirt/jirka-staging.git In practice it works like this: # virsh cpu-baseline /dev/stdin >guest-cpu <<EOF <capabilities> <host> <cpu> <arch>x86_64</arch> <model>core2duo</model> <topology sockets='1' cores='2' threads='1'/> <feature name='lahf_lm'/> <feature name='xtpr'/> <feature name='cx16'/> <feature name='tm2'/> <feature name='est'/> <feature name='vmx'/> </cpu> </host> </capabilities> <cpu> <arch>i686</arch> <model>pentiumpro</model> <topology sockets='1' cores='2' threads='1'/> <feature name='vme'/> <feature name='pse36'/> <feature name='pni'/> <feature name='monitor'/> <feature name='ssse3'/> </cpu> <cpu> <arch>x86_64</arch> <model>phenom</model> </cpu> EOF # cat guest-cpu <cpu match='exact'> <model>qemu32</model> <feature policy='require' name='monitor'/> <feature policy='require' name='pse36'/> </cpu> Jiri Denemark (9): Functions for computing baseline CPU from a set of host CPUs Implement cpuArchBaseline in generic CPU driver Implement cpuArchBaseline in x86 CPU driver virConnectBaselineCPU public API Internal driver API for virConnectBaselineCPU virConnectBaselineCPU public API implementation Wire protocol format and dispatcher for virConnectBaselineCPU Implement cpuBaseline in remote and qemu drivers cpu-baseline command for virsh daemon/remote.c | 25 +++++++ daemon/remote_dispatch_args.h | 1 + daemon/remote_dispatch_prototypes.h | 8 ++ daemon/remote_dispatch_ret.h | 1 + daemon/remote_dispatch_table.h | 5 ++ include/libvirt/libvirt.h.in | 18 +++++ src/cpu/cpu.c | 125 ++++++++++++++++++++++++++++++++++- src/cpu/cpu.h | 23 ++++++- src/cpu/cpu_generic.c | 104 ++++++++++++++++++++++++++++- src/cpu/cpu_x86.c | 88 ++++++++++++++++++++++++- src/driver.h | 6 ++ src/esx/esx_driver.c | 1 + src/libvirt.c | 57 ++++++++++++++++ src/libvirt_private.syms | 2 + src/libvirt_public.syms | 5 ++ src/lxc/lxc_driver.c | 1 + src/opennebula/one_driver.c | 1 + src/openvz/openvz_driver.c | 1 + src/phyp/phyp_driver.c | 1 + src/qemu/qemu_driver.c | 14 ++++ src/remote/remote_driver.c | 32 +++++++++ src/remote/remote_protocol.c | 24 +++++++- src/remote/remote_protocol.h | 20 ++++++ src/remote/remote_protocol.x | 18 +++++- src/test/test_driver.c | 1 + src/uml/uml_driver.c | 1 + src/vbox/vbox_tmpl.c | 1 + src/xen/xen_driver.c | 1 + tools/virsh.c | 72 ++++++++++++++++++++ tools/virsh.pod | 8 ++ 30 files changed, 658 insertions(+), 7 deletions(-)

3 15

[libvirt] [PATCH] Add cpu_generic.c to the list of translated files
by Jiri Denemark 11 Feb '10

11 Feb '10

Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- po/POTFILES.in | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/po/POTFILES.in b/po/POTFILES.in index 3e6440e..6655005 100644 --- a/po/POTFILES.in +++ b/po/POTFILES.in @@ -12,6 +12,7 @@ src/conf/secret_conf.c src/conf/storage_conf.c src/conf/storage_encryption_conf.c src/cpu/cpu.c +src/cpu/cpu_generic.c src/cpu/cpu_map.c src/cpu/cpu_x86.c src/datatypes.c -- 1.6.6.1

2 1

[libvirt] [PATCH] Fix crash in LXC driver open method when URI has no path
by Daniel P. Berrange 11 Feb '10

11 Feb '10

If giving a lxc:// URI instead of lxc:/// the open method would crash ona NULL pointer * src/lxc/lxc_driver.c: Cope with a NULL URI path --- src/lxc/lxc_driver.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index 1a73ca6..6d568a3 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -126,7 +126,8 @@ static virDrvOpenStatus lxcOpen(virConnectPtr conn, return VIR_DRV_OPEN_DECLINED; /* If path isn't '/' then they typoed, tell them correct path */ - if (STRNEQ(conn->uri->path, "/")) { + if (conn->uri->path != NULL && + STRNEQ(conn->uri->path, "/")) { lxcError(VIR_ERR_INTERNAL_ERROR, _("Unexpected LXC URI path '%s', try lxc:///"), conn->uri->path); -- 1.6.5.2

2 1

[libvirt] [PATCH] Fix USB device path formatting mixup
by Daniel P. Berrange 11 Feb '10

11 Feb '10

* src/util/hostusb.c: The device path for a USB device wants the bus/device IDs in decimal not octal --- src/util/hostusb.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/util/hostusb.c b/src/util/hostusb.c index 3cce66b..bf96539 100644 --- a/src/util/hostusb.c +++ b/src/util/hostusb.c @@ -184,7 +184,7 @@ usbGetDevice(unsigned bus, snprintf(dev->name, sizeof(dev->name), "%.3o:%.3o", dev->bus, dev->dev); snprintf(dev->path, sizeof(dev->path), - USB_DEVFS "%03o/%03o", dev->bus, dev->dev); + USB_DEVFS "%03d/%03d", dev->bus, dev->dev); /* XXX fixme. this should be product/vendor */ snprintf(dev->id, sizeof(dev->id), "%d %d", dev->bus, dev->dev); -- 1.6.5.2

2 1

[libvirt] [PATCH] Support 'block_passwd' command for QEMU disk encryption
by Daniel P. Berrange 11 Feb '10

11 Feb '10

The old text mode monitor prompts for a password when disks are encrypted. This interactive approach doesn't work for JSON mode monitor. Thus there is a new 'block_passwd' command that can be used. * src/qemu/qemu_driver.c: Split out code for looking up a disk secret from findVolumeQcowPassphrase, into a new method getVolumeQcowPassphrase. Enhance qemuInitPasswords() to also set the disk encryption password via the monitor * src/qemu/qemu_monitor.c, src/qemu/qemu_monitor.h, src/qemu/qemu_monitor_json.c, src/qemu/qemu_monitor_json.h, src/qemu/qemu_monitor_text.c, src/qemu/qemu_monitor_text.h: Add support for the 'block_passwd' monitor command. --- src/qemu/qemu_driver.c | 117 +++++++++++++++++++++++++++++------------- src/qemu/qemu_monitor.c | 15 +++++ src/qemu/qemu_monitor.h | 4 ++ src/qemu/qemu_monitor_json.c | 33 ++++++++++++ src/qemu/qemu_monitor_json.h | 4 ++ src/qemu/qemu_monitor_text.c | 47 +++++++++++++++++ src/qemu/qemu_monitor_text.h | 4 ++ 7 files changed, 188 insertions(+), 36 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 0d77d57..03d0f5f 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -694,51 +694,46 @@ qemuHandleMonitorEOF(qemuMonitorPtr mon ATTRIBUTE_UNUSED, } -static virStorageEncryptionPtr -findDomainDiskEncryption(virDomainObjPtr vm, - const char *path) +static virDomainDiskDefPtr +findDomainDiskByPath(virDomainObjPtr vm, + const char *path) { - bool seen_volume; int i; - seen_volume = false; for (i = 0; i < vm->def->ndisks; i++) { virDomainDiskDefPtr disk; disk = vm->def->disks[i]; - if (disk->src != NULL && STREQ(disk->src, path)) { - seen_volume = true; - if (disk->encryption != NULL) - return disk->encryption; - } + if (disk->src != NULL && STREQ(disk->src, path)) + return disk; } - if (seen_volume) - qemuReportError(VIR_ERR_INVALID_DOMAIN, - _("missing <encryption> for volume %s"), path); - else - qemuReportError(VIR_ERR_INTERNAL_ERROR, - _("unexpected passphrase request for volume %s"), - path); + + qemuReportError(VIR_ERR_INTERNAL_ERROR, + _("no disk found with path %s"), + path); return NULL; } - static int -findVolumeQcowPassphrase(qemuMonitorPtr mon ATTRIBUTE_UNUSED, - virConnectPtr conn, - virDomainObjPtr vm, - const char *path, - char **secretRet, - size_t *secretLen) +getVolumeQcowPassphrase(virConnectPtr conn, + virDomainDiskDefPtr disk, + char **secretRet, + size_t *secretLen) { - virStorageEncryptionPtr enc; virSecretPtr secret; char *passphrase; unsigned char *data; size_t size; int ret = -1; + virStorageEncryptionPtr enc; - virDomainObjLock(vm); + if (!disk->encryption) { + qemuReportError(VIR_ERR_INTERNAL_ERROR, + _("disk %s does not have any encryption information"), + disk->src); + return -1; + } + enc = disk->encryption; if (!conn) { qemuReportError(VIR_ERR_NO_SUPPORT, @@ -754,16 +749,12 @@ findVolumeQcowPassphrase(qemuMonitorPtr mon ATTRIBUTE_UNUSED, goto cleanup; } - enc = findDomainDiskEncryption(vm, path); - if (enc == NULL) - return -1; - if (enc->format != VIR_STORAGE_ENCRYPTION_FORMAT_QCOW || enc->nsecrets != 1 || enc->secrets[0]->type != VIR_STORAGE_ENCRYPTION_SECRET_TYPE_PASSPHRASE) { qemuReportError(VIR_ERR_INVALID_DOMAIN, - _("invalid <encryption> for volume %s"), path); + _("invalid <encryption> for volume %s"), disk->src); goto cleanup; } @@ -782,7 +773,7 @@ findVolumeQcowPassphrase(qemuMonitorPtr mon ATTRIBUTE_UNUSED, VIR_FREE(data); qemuReportError(VIR_ERR_INVALID_SECRET, _("format='qcow' passphrase for %s must not contain a " - "'\\0'"), path); + "'\\0'"), disk->src); goto cleanup; } @@ -804,8 +795,30 @@ findVolumeQcowPassphrase(qemuMonitorPtr mon ATTRIBUTE_UNUSED, ret = 0; cleanup: - virDomainObjUnlock(vm); + return ret; +} +static int +findVolumeQcowPassphrase(qemuMonitorPtr mon ATTRIBUTE_UNUSED, + virConnectPtr conn, + virDomainObjPtr vm, + const char *path, + char **secretRet, + size_t *secretLen) +{ + virDomainDiskDefPtr disk; + int ret = -1; + + virDomainObjLock(vm); + disk = findDomainDiskByPath(vm, path); + + if (!disk) + goto cleanup; + + ret = getVolumeQcowPassphrase(conn, disk, secretRet, secretLen); + +cleanup: + virDomainObjUnlock(vm); return ret; } @@ -1681,8 +1694,10 @@ qemudInitCpuAffinity(virDomainObjPtr vm) static int -qemuInitPasswords(struct qemud_driver *driver, - virDomainObjPtr vm) { +qemuInitPasswords(virConnectPtr conn, + struct qemud_driver *driver, + virDomainObjPtr vm, + unsigned long long qemuCmdFlags) { int ret = 0; qemuDomainObjPrivatePtr priv = vm->privateData; @@ -1698,6 +1713,36 @@ qemuInitPasswords(struct qemud_driver *driver, qemuDomainObjExitMonitorWithDriver(driver, vm); } + if (ret < 0) + goto cleanup; + + if (qemuCmdFlags & QEMUD_CMD_FLAG_DEVICE) { + int i; + + for (i = 0 ; i < vm->def->ndisks ; i++) { + char *secret; + size_t secretLen; + + if (!vm->def->disks[i]->encryption || + !vm->def->disks[i]->src) + continue; + + if (getVolumeQcowPassphrase(conn, + vm->def->disks[i], + &secret, &secretLen) < 0) + goto cleanup; + + qemuDomainObjEnterMonitorWithDriver(driver, vm); + ret = qemuMonitorSetDrivePassphrase(priv->mon, + vm->def->disks[i]->info.alias, + secret); + qemuDomainObjExitMonitorWithDriver(driver, vm); + if (ret < 0) + goto cleanup; + } + } + +cleanup: return ret; } @@ -2721,7 +2766,7 @@ static int qemudStartVMDaemon(virConnectPtr conn, if (qemudInitCpuAffinity(vm) < 0) goto abort; - if (qemuInitPasswords(driver, vm) < 0) + if (qemuInitPasswords(conn, driver, vm, qemuCmdFlags) < 0) goto abort; /* If we have -device, then addresses are assigned explicitly. diff --git a/src/qemu/qemu_monitor.c b/src/qemu/qemu_monitor.c index 64c6cba..c1d369b 100644 --- a/src/qemu/qemu_monitor.c +++ b/src/qemu/qemu_monitor.c @@ -1331,3 +1331,18 @@ int qemuMonitorAddDrive(qemuMonitorPtr mon, ret = qemuMonitorTextAddDrive(mon, drivestr); return ret; } + + +int qemuMonitorSetDrivePassphrase(qemuMonitorPtr mon, + const char *alias, + const char *passphrase) +{ + DEBUG("mon=%p, fd=%d alias=%s passphrase=%p(value hidden)", mon, mon->fd, alias, passphrase); + int ret; + + if (mon->json) + ret = qemuMonitorJSONSetDrivePassphrase(mon, alias, passphrase); + else + ret = qemuMonitorTextSetDrivePassphrase(mon, alias, passphrase); + return ret; +} diff --git a/src/qemu/qemu_monitor.h b/src/qemu/qemu_monitor.h index a330eff..786ad7a 100644 --- a/src/qemu/qemu_monitor.h +++ b/src/qemu/qemu_monitor.h @@ -291,4 +291,8 @@ int qemuMonitorAddDevice(qemuMonitorPtr mon, int qemuMonitorAddDrive(qemuMonitorPtr mon, const char *drivestr); +int qemuMonitorSetDrivePassphrase(qemuMonitorPtr mon, + const char *alias, + const char *passphrase); + #endif /* QEMU_MONITOR_H */ diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index 032afef..c9b8d60 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -1841,3 +1841,36 @@ int qemuMonitorJSONAddDrive(qemuMonitorPtr mon, virJSONValueFree(reply); return ret; } + + +int qemuMonitorJSONSetDrivePassphrase(qemuMonitorPtr mon, + const char *alias, + const char *passphrase) +{ + int ret; + virJSONValuePtr cmd; + virJSONValuePtr reply = NULL; + char *drive; + + if (virAsprintf(&drive, "%s%s", QEMU_DRIVE_HOST_PREFIX, alias) < 0) { + virReportOOMError(); + return -1; + } + + cmd = qemuMonitorJSONMakeCommand("block_passwd", + "s:device", drive, + "s:password", passphrase, + NULL); + VIR_FREE(drive); + if (!cmd) + return -1; + + ret = qemuMonitorJSONCommand(mon, cmd, &reply); + + if (ret == 0) + ret = qemuMonitorJSONCheckError(cmd, reply); + + virJSONValueFree(cmd); + virJSONValueFree(reply); + return ret; +} diff --git a/src/qemu/qemu_monitor_json.h b/src/qemu/qemu_monitor_json.h index ac6458c..65a70e3 100644 --- a/src/qemu/qemu_monitor_json.h +++ b/src/qemu/qemu_monitor_json.h @@ -162,4 +162,8 @@ int qemuMonitorJSONAddDevice(qemuMonitorPtr mon, int qemuMonitorJSONAddDrive(qemuMonitorPtr mon, const char *drivestr); +int qemuMonitorJSONSetDrivePassphrase(qemuMonitorPtr mon, + const char *alias, + const char *passphrase); + #endif /* QEMU_MONITOR_JSON_H */ diff --git a/src/qemu/qemu_monitor_text.c b/src/qemu/qemu_monitor_text.c index a6a4598..e993699 100644 --- a/src/qemu/qemu_monitor_text.c +++ b/src/qemu/qemu_monitor_text.c @@ -2127,3 +2127,50 @@ cleanup: VIR_FREE(safe_str); return ret; } + + +int qemuMonitorTextSetDrivePassphrase(qemuMonitorPtr mon, + const char *alias, + const char *passphrase) +{ + char *cmd = NULL; + char *reply = NULL; + int ret = -1; + char *safe_str; + + safe_str = qemuMonitorEscapeArg(passphrase); + if (!safe_str) { + virReportOOMError(); + return -1; + } + + ret = virAsprintf(&cmd, "block_passwd %s%s \"%s\"", QEMU_DRIVE_HOST_PREFIX, alias, safe_str); + if (ret == -1) { + virReportOOMError(); + goto cleanup; + } + + if (qemuMonitorCommand(mon, cmd, &reply) < 0) { + qemuReportError(VIR_ERR_OPERATION_FAILED, + _("failed to close fd in qemu with '%s'"), cmd); + goto cleanup; + } + + if (strstr(reply, "\nunknown command:")) { + qemuReportError(VIR_ERR_OPERATION_FAILED, "%s", + _("setting disk password is not supported")); + goto cleanup; + } else if (strstr(reply, "The entered password is invalid")) { + qemuReportError(VIR_ERR_OPERATION_FAILED, "%s", + _("the disk password is incorrect")); + goto cleanup; + } + + ret = 0; + +cleanup: + VIR_FREE(cmd); + VIR_FREE(reply); + VIR_FREE(safe_str); + return ret; +} diff --git a/src/qemu/qemu_monitor_text.h b/src/qemu/qemu_monitor_text.h index 12d75f5..1937e99 100644 --- a/src/qemu/qemu_monitor_text.h +++ b/src/qemu/qemu_monitor_text.h @@ -166,4 +166,8 @@ int qemuMonitorTextAddDevice(qemuMonitorPtr mon, int qemuMonitorTextAddDrive(qemuMonitorPtr mon, const char *drivestr); +int qemuMonitorTextSetDrivePassphrase(qemuMonitorPtr mon, + const char *alias, + const char *passphrase); + #endif /* QEMU_MONITOR_TEXT_H */ -- 1.6.5.2

2 2

[libvirt] [PATCH 5/5] macvtap support for libvirt -- schema extensions
by Stefan Berger 11 Feb '10

11 Feb '10

This part extends the domain xml schema to support the new interface type 'direct'. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com>

1 0

[libvirt] [PATCH 4/5] macvtap support for libvirt -- helper code
by Stefan Berger 11 Feb '10

11 Feb '10

This part adds the helper code to setup and tear down macvtap devices using direct communication with the device driver via netlink sockets. The rather short messages received from the netlink layer are now written into a dynamically allocated buffer. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com>

1 0

[libvirt] [PATCH 3/5] macvtap support for libvirt -- qemu support
by Stefan Berger 11 Feb '10

11 Feb '10

This part adds support for qemu making a macvtap tap device available via file descriptor passed to qemu command line. This also attempts to tear down the macvtap device when a VM terminates. This includes support for attachment and detachment to/from running VM. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com>

1 0

[libvirt] [PATCH 2/5] macvtap support for libvirt -- parse new interface XML
by Stefan Berger 11 Feb '10

11 Feb '10

This part adds support to domain_conf.{c|h} for parsing the new interface XML of type 'direct'. The parsed mode is now stored as an int. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com>

1 0

[libvirt] [PATCH 1/5] macvtap support for libvirt -- build support
by Stefan Berger 11 Feb '10

11 Feb '10

This patch adds build support for libvirt checking for certain contents of /usr/include/linux/if_link.h to see whether macvtap support is compilable on that system. One can disable macvtap support in libvirt via --without-macvtap passed to configure. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com>

1 0

[libvirt] [PATCH] virStorageBackendIsMultipath: avoid dead store
by Jim Meyering 11 Feb '10

11 Feb '10

This suppresses a warning reported by clang. Since "target_type" is initialized, there's no point in storing dm_get_next_target's return value. >From d0cd9b4233194b27fb168e835223dd51e77fa6f2 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Thu, 11 Feb 2010 15:26:37 +0100 Subject: [PATCH] virStorageBackendIsMultipath: avoid dead store * src/storage/storage_backend_mpath.c (virStorageBackendIsMultipath): The result of dm_get_next_target was never used (and isn't needed), so don't store it. --- src/storage/storage_backend_mpath.c | 6 ++---- 1 files changed, 2 insertions(+), 4 deletions(-) diff --git a/src/storage/storage_backend_mpath.c b/src/storage/storage_backend_mpath.c index dbcd6d8..1e7504b 100644 --- a/src/storage/storage_backend_mpath.c +++ b/src/storage/storage_backend_mpath.c @@ -1,7 +1,7 @@ /* * storage_backend_mpath.c: storage backend for multipath handling * - * Copyright (C) 2009-2009 Red Hat, Inc. + * Copyright (C) 2009-2010 Red Hat, Inc. * Copyright (C) 2009-2008 Dave Allan * * This library is free software; you can redistribute it and/or @@ -150,7 +150,6 @@ virStorageBackendIsMultipath(const char *devname) { int ret = 0; struct dm_task *dmt = NULL; - void *next = NULL; uint64_t start, length; char *target_type = NULL; char *params = NULL; @@ -173,8 +172,7 @@ virStorageBackendIsMultipath(const char *devname) goto out; } - next = dm_get_next_target(dmt, next, &start, &length, - &target_type, &params); + dm_get_next_target(dmt, next, &start, &length, &target_type, &params); if (target_type == NULL) { ret = -1; -- 1.7.0.rc2.170.gbc565

2 1

[libvirt] [PATCH] Add virConnectGetVersion Python API
by Taizo ITO 11 Feb '10

11 Feb '10

Hi all, The attached patch adds a new python API call for retrieving the running hypervisor version used by a connection: virConnectGetVersion. Regards, Taizo -- Taizo ITO

2 1

[libvirt] [PATCH 4/5] macvtap support for libvirt -- helper code
by Stefan Berger 11 Feb '10

11 Feb '10

This part adds the helper code to setup and tear down macvtap devices using direct communication with the device driver via netlink sockets. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com>

2 3

[libvirt] [PATCH 3/5] macvtap support for libvirt -- qemu support
by Stefan Berger 11 Feb '10

11 Feb '10

This part adds support for qemu making a macvtap tap device available via file descriptor passed to qemu command line. This also attempts to tear down the macvtap device when a VM terminates. This includes support for attachment and detachment to/from running VM. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com>

3 4

[libvirt] [PATCH 2/5] macvtap support for libvirt -- parse new interface XML
by Stefan Berger 11 Feb '10

11 Feb '10

This part adds support to domain_conf.{c|h} for parsing the new interface XML of type 'direct'. The parsed mode is now stored as an int. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com>

3 3

[libvirt] [PATCH 0/5] macvtap support for Qemu/KVM VMs via libvirt
by Stefan Berger 11 Feb '10

11 Feb '10

Hello! This is a re-post of previously posted patches following Daniel Berrange's request for changes along with other fixes. The following patches provide support for making the macvtap networking device type available to Qemu/KVM VMs. The patches rely on the macvtap driver that just became available through the Linux net-next tree (fixes still may be necessary) and make the device available to Qemu/KVM via a tap file descriptor similar to a 'regular' tap device. Following up on previous discussions, the libvirt patches allow using the following XML in the domain description to enable qemu network connectivity via this type of device: <interface type='direct'> <source dev='eth1' mode='vepa'/> <model type='virtio'/> </interface> The above XML indicates that eth1 is the Ethernet interface to link the macvtap device to and communicate to the network. As a consequence, libvirt will create an instance of a macvtap device, assign it the same MAC address as the VM's interface has and open a file descriptor of the associated character device /dev/tap%d and pass it via command line to Qemu/kvm. In the above XML the mode can be chosen as 'vepa', 'private' or 'bridge' and is by default set to 'vepa'(by the driver) if omitted. Attachment and detachment of macvtap to/from a running VM also works. Regards, Stefan

3 3

[libvirt] [PATCH 5/5] macvtap support for libvirt -- schema extensions
by Stefan Berger 11 Feb '10

11 Feb '10

This part extends the domain xml schema to support the new interface type 'direct'. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com>

2 1

[libvirt] virStoragePoolSourceFree(S) does not free S
by Jim Meyering 11 Feb '10

11 Feb '10

I was surprised to see that virStoragePoolSourceFree(S) does not free S. The other three vir*Free functions in storage_conf *do* free S. There are at least three places where this causes leaks. Here's one: virStoragePoolSourcePtr virStoragePoolDefParseSourceString(virConnectPtr conn, const char *srcSpec, int pool_type) { xmlDocPtr doc = NULL; xmlNodePtr node = NULL; xmlXPathContextPtr xpath_ctxt = NULL; virStoragePoolSourcePtr def = NULL, ret = NULL; doc = xmlReadDoc((const xmlChar *)srcSpec, "srcSpec.xml", NULL, XML_PARSE_NOENT | XML_PARSE_NONET | XML_PARSE_NOERROR | XML_PARSE_NOWARNING); if (doc == NULL) { virStorageReportError(conn, VIR_ERR_XML_ERROR, "%s", _("bad <source> spec")); goto cleanup; } xpath_ctxt = xmlXPathNewContext(doc); if (xpath_ctxt == NULL) { virReportOOMError(conn); goto cleanup; } if (VIR_ALLOC(def) < 0) { <<<====== def is alloc'd virReportOOMError(conn); goto cleanup; } node = virXPathNode(conn, "/source", xpath_ctxt); if (!node) { virStorageReportError(conn, VIR_ERR_XML_ERROR, "%s", _("root element was not source")); goto cleanup; } if (virStoragePoolDefParseSource(conn, xpath_ctxt, def, pool_type, node) < 0) goto cleanup; ret = def; def = NULL; cleanup: if (def) virStoragePoolSourceFree(def); <<<<======== yet not freed xmlFreeDoc(doc); xmlXPathFreeContext(xpath_ctxt); return ret; } One fix might be to call VIR_FREE(def) in the "if (def)..." block, but that seems short-sighted, since the name virStoragePoolSourceFree makes me think *it* should be doing the whole job. However, if I make the logical change to virStoragePoolSourceFree, diff --git a/src/conf/storage_conf.c b/src/conf/storage_conf.c index 62b8394..ffe38cc 100644 --- a/src/conf/storage_conf.c +++ b/src/conf/storage_conf.c @@ -291,6 +291,7 @@ virStoragePoolSourceFree(virStoragePoolSourcePtr source) { VIR_FREE(source->auth.chap.login); VIR_FREE(source->auth.chap.passwd); } + VIR_FREE(source); } that causes "make check" to fail miserably due to heap corruption, as reported by valgrind: ==30311== Invalid free() / delete / delete[] ==30311== at 0x4A04D72: free (vg_replace_malloc.c:325) ==30311== by 0x42950C: virFree (memory.c:177) ==30311== by 0x4A2687: virStoragePoolSourceFree (storage_conf.c:294) ==30311== by 0x4A26BD: virStoragePoolDefFree (storage_conf.c:304) ==30311== by 0x4A2723: virStoragePoolObjFree (storage_conf.c:319) ==30311== by 0x4A27A2: virStoragePoolObjListFree (storage_conf.c:334) ==30311== by 0x4757AD: storageDriverShutdown (storage_driver.c:246) ==30311== by 0x4C1E3F: virStateCleanup (libvirt.c:909) ==30311== by 0x4103C5: qemudCleanup (libvirtd.c:2402) ==30311== by 0x41273B: main (libvirtd.c:3196) ==30311== Address 0x5c48c38 is 56 bytes inside a block of size 184 alloc'd ==30311== at 0x4A04481: calloc (vg_replace_malloc.c:418) ==30311== by 0x4293F5: virAlloc (memory.c:100) ==30311== by 0x4A330E: virStoragePoolDefParseXML (storage_conf.c:615) ==30311== by 0x4A3A21: virStoragePoolDefParseNode (storage_conf.c:762) ==30311== by 0x4A3BE6: virStoragePoolDefParse (storage_conf.c:810) ==30311== by 0x4A3C7F: virStoragePoolDefParseFile (storage_conf.c:834) ==30311== by 0x4A5816: virStoragePoolObjLoad (storage_conf.c:1495) ==30311== by 0x4A5BDB: virStoragePoolLoadAllConfigs (storage_conf.c:1575) ==30311== by 0x475598: storageDriverStartup (storage_driver.c:164) ==30311== by 0x4C1D9D: virStateInitialize (libvirt.c:888) ==30311== by 0x4125E9: main (libvirtd.c:3153) I tracked the problem down to this definition in src/conf/storage_conf.h: typedef struct _virStoragePoolDef virStoragePoolDef; typedef virStoragePoolDef *virStoragePoolDefPtr; struct _virStoragePoolDef { /* General metadata */ char *name; unsigned char uuid[VIR_UUID_BUFLEN]; int type; /* virStoragePoolType */ unsigned long long allocation; unsigned long long capacity; unsigned long long available; virStoragePoolSource source; <== this is a *STRUCT*, not a ptr virStoragePoolTarget target; <== Likewise }; Hence, when calling virStoragePoolDefFree (defined like this) void virStoragePoolDefFree(virStoragePoolDefPtr def) { if (!def) return; VIR_FREE(def->name); virStoragePoolSourceFree(&def->source); That function must *not* call VIR_FREE on &def->source, because as valgrind tells us, above, that is in the middle of a virStoragePoolDef buffer. Thus, the naive patch seems to be the way to go, unless you want to change the struct members to be pointers (and adjust all uses -- much more invasive). There are other leaks associated with similar uses of virStoragePoolSourceFree just like this one. Here's a patch that fixes them all at once. If you agree in principle, I'll write a longer commit log than the one below: [note that there's no need to guard uses of virStoragePoolSourceFree against NULL args, so I removed those. ] >From a53efcee7ec74948a2cf4c1c02902419856b0154 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Fri, 5 Feb 2010 17:09:43 +0100 Subject: [PATCH] plug four virStoragePoolSourceFree-related leaks * src/conf/storage_conf.c (virStoragePoolDefParseSourceString): * src/storage/storage_backend_fs.c: (virStorageBackendFileSystemNetFindPoolSourcesFunc): (virStorageBackendFileSystemNetFindPoolSources): * src/test/test_driver.c (testStorageFindPoolSources): --- src/conf/storage_conf.c | 4 ++-- src/storage/storage_backend_fs.c | 8 ++++---- src/test/test_driver.c | 3 ++- 3 files changed, 8 insertions(+), 7 deletions(-) diff --git a/src/conf/storage_conf.c b/src/conf/storage_conf.c index 62b8394..b98b8e9 100644 --- a/src/conf/storage_conf.c +++ b/src/conf/storage_conf.c @@ -524,8 +524,8 @@ virStoragePoolDefParseSourceString(virConnectPtr conn, ret = def; def = NULL; cleanup: - if (def) - virStoragePoolSourceFree(def); + virStoragePoolSourceFree(def); + VIR_FREE(def); xmlFreeDoc(doc); xmlXPathFreeContext(xpath_ctxt); diff --git a/src/storage/storage_backend_fs.c b/src/storage/storage_backend_fs.c index 0d1c7a7..70e91a0 100644 --- a/src/storage/storage_backend_fs.c +++ b/src/storage/storage_backend_fs.c @@ -174,8 +174,8 @@ virStorageBackendFileSystemNetFindPoolSourcesFunc(virConnectPtr conn, src = NULL; ret = 0; cleanup: - if (src) - virStoragePoolSourceFree(src); + virStoragePoolSourceFree(src); + VIR_FREE(src); return ret; } @@ -236,8 +236,8 @@ virStorageBackendFileSystemNetFindPoolSources(virConnectPtr conn, for (i = 0; i < state.list.nsources; i++) virStoragePoolSourceFree(&state.list.sources[i]); - if (source) - virStoragePoolSourceFree(source); + virStoragePoolSourceFree(source); + VIR_FREE(source); VIR_FREE(state.list.sources); diff --git a/src/test/test_driver.c b/src/test/test_driver.c index 2122a1b..8e5ecf2 100644 --- a/src/test/test_driver.c +++ b/src/test/test_driver.c @@ -1,7 +1,7 @@ /* * test.c: A "mock" hypervisor for use by application unit tests * - * Copyright (C) 2006-2009 Red Hat, Inc. + * Copyright (C) 2006-2010 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange * * This library is free software; you can redistribute it and/or @@ -3763,6 +3763,7 @@ testStorageFindPoolSources(virConnectPtr conn, cleanup: virStoragePoolSourceFree(source); + VIR_FREE(source); return ret; } -- 1.7.0.rc1.204.gb96e

2 2

[libvirt] [PATCH 1/5] macvtap support for libvirt -- build support
by Stefan Berger 11 Feb '10

11 Feb '10

This patch adds build support for libvirt checking for certain contents of /usr/include/linux/if_link.h to see whether macvtap support is compilable on that system. One can disable macvtap support in libvirt via --without-macvtap passed to configure. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com>

3 2

[libvirt] Node devices?
by Ed Swierk 11 Feb '10

11 Feb '10

After staring at the code implementing node devices in libvirt, I have decided that they're really neat, but I have no idea what they are for or how to use them. Reading the API docs and searching libvir-list didn't help much, and the user documentation (http://libvirt.org/archnode.html and http://libvirt.org/formatnode.html) is a bit, err, high-level. Can someone enlighten me? --Ed

5 5

[libvirt] [PATCH] Fix disk stats retrieval with QEMU >= 0.12
by Daniel P. Berrange 10 Feb '10

10 Feb '10

With QEMU >= 0.12 the host and guest side of disks no longer have the same naming convention. Specifically the host side will now get a 'drive-' prefix added to its name. The 'info blockstats' monitor command returns the host side name, so it is neccessary to strip this off when looking up stats since libvirt stores the guest side name ! * src/qemu/qemu_conf.c, src/qemu/qemu_conf.h: Move 'drive-' prefix string to a defined constant * src/qemu/qemu_monitor_json.c, src/qemu/qemu_monitor_text.c: Strip off 'drive-' prefix (if found) when looking up disk stats --- src/qemu/qemu_conf.c | 4 ++-- src/qemu/qemu_conf.h | 2 ++ src/qemu/qemu_monitor_json.c | 7 +++++++ src/qemu/qemu_monitor_text.c | 7 +++++++ 4 files changed, 18 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c index 3988582..c1d03cd 100644 --- a/src/qemu/qemu_conf.c +++ b/src/qemu/qemu_conf.c @@ -2301,7 +2301,7 @@ qemuBuildDriveStr(virDomainDiskDefPtr disk, virBufferAddLit(&opt, ",media=cdrom"); if (qemuCmdFlags & QEMUD_CMD_FLAG_DEVICE) { - virBufferVSprintf(&opt, ",id=drive-%s", disk->info.alias); + virBufferVSprintf(&opt, ",id=%s%s", QEMU_DRIVE_HOST_PREFIX, disk->info.alias); } else { if (busid == -1 && unitid == -1) { if (idx != -1) @@ -2390,7 +2390,7 @@ qemuBuildDriveDevStr(virDomainDiskDefPtr disk) _("unsupported disk bus '%s' with device setup"), bus); goto error; } - virBufferVSprintf(&opt, ",drive=drive-%s", disk->info.alias); + virBufferVSprintf(&opt, ",drive=%s%s", QEMU_DRIVE_HOST_PREFIX, disk->info.alias); virBufferVSprintf(&opt, ",id=%s", disk->info.alias); if (virBufferError(&opt)) { diff --git a/src/qemu/qemu_conf.h b/src/qemu/qemu_conf.h index d26bb90..498f1d1 100644 --- a/src/qemu/qemu_conf.h +++ b/src/qemu/qemu_conf.h @@ -157,6 +157,8 @@ typedef qemuDomainPCIAddressSet *qemuDomainPCIAddressSetPtr; /* Config type for XML import/export conversions */ #define QEMU_CONFIG_FORMAT_ARGV "qemu-argv" +#define QEMU_DRIVE_HOST_PREFIX "drive-" + #define qemuReportError(code, fmt...) \ virReportErrorHelper(NULL, VIR_FROM_QEMU, code, __FILE__, \ __FUNCTION__, __LINE__, fmt) diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index e5288ce..032afef 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -754,6 +754,13 @@ int qemuMonitorJSONGetBlockStatsInfo(qemuMonitorPtr mon, goto cleanup; } + /* New QEMU has separate names for host & guest side of the disk + * and libvirt gives the host side a 'drive-' prefix. The passed + * in devname is the guest side though + */ + if (STRPREFIX(thisdev, QEMU_DRIVE_HOST_PREFIX)) + thisdev += strlen(QEMU_DRIVE_HOST_PREFIX); + if (STRNEQ(thisdev, devname)) continue; diff --git a/src/qemu/qemu_monitor_text.c b/src/qemu/qemu_monitor_text.c index 23fc4cf..a6a4598 100644 --- a/src/qemu/qemu_monitor_text.c +++ b/src/qemu/qemu_monitor_text.c @@ -649,6 +649,13 @@ int qemuMonitorTextGetBlockStatsInfo(qemuMonitorPtr mon, p = info; while (*p) { + /* New QEMU has separate names for host & guest side of the disk + * and libvirt gives the host side a 'drive-' prefix. The passed + * in devname is the guest side though + */ + if (STRPREFIX(p, QEMU_DRIVE_HOST_PREFIX)) + p += strlen(QEMU_DRIVE_HOST_PREFIX); + if (STREQLEN (p, devname, devnamelen) && p[devnamelen] == ':' && p[devnamelen+1] == ' ') { -- 1.6.5.2

3 2

[libvirt] domblkstat not working for qemu-kvm > 0.11
by Thomas Treutner 10 Feb '10

10 Feb '10

Hi, with all qemu-kvm > 0.11, virsh domblkstat is not working: virsh -c qemu://node03/system domblkstat wp01 vdb : invalid argument in no stats found for device virtio-disk1 (vda, vdb etc. works with 0.11) "virtio-disk1" seems to be a red herring: virsh -c qemu://node03/system domblkstat wp01 virtio-disk1 : invalid argument in invalid path: virtio-disk1 Most probably there have been made some changes with 0.12 libvirt does not resemble yet. If you need more debugging output, just tell. kr,t

2 2

[libvirt] [PATCH] qemu: Properly report a startup timeout error
by Cole Robinson 10 Feb '10

10 Feb '10

The timeout errors were unconditionally being overwritten by the less helpful 'unable to start guest' error. Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- src/qemu/qemu_driver.c | 17 +++++++---------- 1 files changed, 7 insertions(+), 10 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 4374291..2172c99 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -1358,15 +1358,15 @@ qemudReadLogOutput(virConnectPtr conn, buf[got] = '\0'; if (got == buflen-1) { qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, - _("Out of space while reading %s log output"), - what); + _("Out of space while reading %s log output: %s"), + what, buf); return -1; } if (isdead) { qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, - _("Process exited while reading %s log output"), - what); + _("Process exited while reading %s log output: %s"), + what, buf); return -1; } @@ -1378,7 +1378,8 @@ qemudReadLogOutput(virConnectPtr conn, } qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, - _("Timed out while reading %s log output"), what); + _("Timed out while reading %s log output: %s"), + what, buf); return -1; } @@ -1557,12 +1558,8 @@ qemudWaitForMonitor(virConnectPtr conn, virStrerror(errno, ebuf, sizeof ebuf)); } - if (ret < 0) { - /* Unexpected end of file - inform user of QEMU log data */ - qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, - _("unable to start guest: %s"), buf); + if (ret < 0) return -1; - } VIR_DEBUG("Connect monitor to %p '%s'", vm, vm->def->name); if (qemuConnectMonitor(vm) < 0) -- 1.6.5.2

3 6

[libvirt] [PATCH] test: Fake security driver support in capabilities
by Cole Robinson 10 Feb '10

10 Feb '10

Having some value in capabilities helps testing this stuff in virt-manager. Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- src/test/test_driver.c | 8 ++++++++ 1 files changed, 8 insertions(+), 0 deletions(-) diff --git a/src/test/test_driver.c b/src/test/test_driver.c index 411c5cd..3f95a22 100644 --- a/src/test/test_driver.c +++ b/src/test/test_driver.c @@ -205,6 +205,14 @@ testBuildCapabilities(virConnectPtr conn) { caps->privateDataAllocFunc = testDomainObjPrivateAlloc; caps->privateDataFreeFunc = testDomainObjPrivateFree; + caps->host.secModel.model = strdup("testSecurity"); + if (!caps->host.secModel.model) + goto no_memory; + + caps->host.secModel.doi = strdup(""); + if (!caps->host.secModel.doi) + goto no_memory; + return caps; no_memory: -- 1.6.5.2

2 2

[libvirt] [PATCH] (absolutePathFromBaseFile): fix up preceding commit
by Jim Meyering 10 Feb '10

10 Feb '10

To my chagrin, I saw that my most recent commit introduced compilation errors. Sorry about that. Here's how I propose to fix it. >From 2d948a373ecebec6c06274f61b31d1ae9c40ae41 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Fri, 5 Feb 2010 14:57:35 +0100 Subject: [PATCH] (absolutePathFromBaseFile): fix up preceding commit * src/util/storage_file.c: Include <assert.h>. (absolutePathFromBaseFile): Assert that converting size_t to int is valid. Reverse length/string args to match "%.*s". Explicitly ignore the return value of virAsprintf. --- src/util/storage_file.c | 7 ++++++- 1 files changed, 6 insertions(+), 1 deletions(-) diff --git a/src/util/storage_file.c b/src/util/storage_file.c index 2c79fa9..135acec 100644 --- a/src/util/storage_file.c +++ b/src/util/storage_file.c @@ -26,7 +26,9 @@ #include <unistd.h> #include <fcntl.h> +#include <assert.h> #include "dirname.h" +#include "ignore-value.h" #include "memory.h" #include "virterror_internal.h" @@ -255,7 +257,10 @@ absolutePathFromBaseFile(const char *base_file, const char *path) if (*path == '/' || d_len == 0) return strdup(path); - virAsprintf(&res, "%.*s/%s", base_file, d_len, path); + /* Ensure that the following cast-to-int is valid. */ + assert (d_len <= INT_MAX); + + ignore_value(virAsprintf(&res, "%.*s/%s", (int) d_len, base_file, path)); return res; } -- 1.7.0.rc1.204.gb96e

3 4

[libvirt] [PATCH 0/8] Remove virConnectPtr from yet more places
by Daniel P. Berrange 10 Feb '10

10 Feb '10

Following on from previous two postings http://www.redhat.com/archives/libvir-list/2010-February/msg00198.html http://www.redhat.com/archives/libvir-list/2010-February/msg00279.html With this series, none of the shared driver code uses virConnectPtr anymore, with the exception of src/conf/domain_events.h which has a genuine need for it - events are dispatched back to a specifici connection handle.

2 17

[libvirt] [PATCH 0/4] Remove use of virConnectPtr from more internal APIs
by Daniel P. Berrange 10 Feb '10

10 Feb '10

Following on from Matthias' work http://www.redhat.com/archives/libvir-list/2010-February/msg00198.html This series kills the virConnectPtr from QEMU, and LXC drivers, the security driver support code and the domain XML handling code.

3 11

[libvirt] [PATCH] Convert qemu command line flags to 64-bit int
by Daniel P. Berrange 10 Feb '10

10 Feb '10

The QEMU flags are commonly stored as a signed or unsigned int, allowing only 31 flags. This limit is rather close, so to aid future patches, change it to a 64-bit int * src/qemu/qemu_conf.c, src/qemu/qemu_conf.h, src/qemu/qemu_driver.c, tests/qemuargv2xmltest.c, tests/qemuhelptest.c, tests/qemuxml2argvtest.c: Use 'unsigned long long' for QEMU flags --- src/qemu/qemu_conf.c | 28 ++++++++++++++-------------- src/qemu/qemu_conf.h | 14 +++++++------- src/qemu/qemu_driver.c | 12 ++++++------ tests/qemuargv2xmltest.c | 2 +- tests/qemuhelptest.c | 13 +++++++------ tests/qemuxml2argvtest.c | 7 ++++--- 6 files changed, 39 insertions(+), 37 deletions(-) diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c index 257b4a4..e3769c8 100644 --- a/src/qemu/qemu_conf.c +++ b/src/qemu/qemu_conf.c @@ -1108,12 +1108,12 @@ virCapsPtr qemudCapsInit(virCapsPtr old_caps) { return NULL; } -static unsigned int qemudComputeCmdFlags(const char *help, - unsigned int version, - unsigned int is_kvm, - unsigned int kvm_version) +static unsigned long long qemudComputeCmdFlags(const char *help, + unsigned int version, + unsigned int is_kvm, + unsigned int kvm_version) { - unsigned int flags = 0; + unsigned long long flags = 0; if (strstr(help, "-no-kqemu")) flags |= QEMUD_CMD_FLAG_KQEMU; @@ -1236,7 +1236,7 @@ static unsigned int qemudComputeCmdFlags(const char *help, #define SKIP_BLANKS(p) do { while ((*(p) == ' ') || (*(p) == '\t')) (p)++; } while (0) int qemudParseHelpStr(const char *help, - unsigned int *flags, + unsigned long long *flags, unsigned int *version, unsigned int *is_kvm, unsigned int *kvm_version) @@ -1316,14 +1316,14 @@ fail: int qemudExtractVersionInfo(const char *qemu, unsigned int *retversion, - unsigned int *retflags) { + unsigned long long *retflags) { const char *const qemuarg[] = { qemu, "-help", NULL }; const char *const qemuenv[] = { "LC_ALL=C", NULL }; pid_t child; int newstdout = -1; int ret = -1, status; unsigned int version, is_kvm, kvm_version; - unsigned int flags = 0; + unsigned long long flags = 0; if (retflags) *retflags = 0; @@ -1425,7 +1425,7 @@ int qemudNetworkIfaceConnect(virConnectPtr conn, struct qemud_driver *driver, virDomainNetDefPtr net, - int qemuCmdFlags) + unsigned long long qemuCmdFlags) { char *brname = NULL; int err; @@ -1637,7 +1637,7 @@ no_memory: int -qemuAssignDeviceDiskAlias(virDomainDiskDefPtr def, int qemuCmdFlags) +qemuAssignDeviceDiskAlias(virDomainDiskDefPtr def, unsigned long long qemuCmdFlags) { if (qemuCmdFlags & QEMUD_CMD_FLAG_DRIVE) { if (qemuCmdFlags & QEMUD_CMD_FLAG_DEVICE) @@ -1720,7 +1720,7 @@ qemuAssignDeviceControllerAlias(virDomainControllerDefPtr controller) static int -qemuAssignDeviceAliases(virDomainDefPtr def, int qemuCmdFlags) +qemuAssignDeviceAliases(virDomainDefPtr def, unsigned long long qemuCmdFlags) { int i; @@ -2189,7 +2189,7 @@ qemuSafeSerialParamValue(virConnectPtr conn, char * qemuBuildDriveStr(virDomainDiskDefPtr disk, int bootable, - int qemuCmdFlags) + unsigned long long qemuCmdFlags) { virBuffer opt = VIR_BUFFER_INITIALIZER; const char *bus = virDomainDiskQEMUBusTypeToString(disk->bus); @@ -3017,7 +3017,7 @@ no_memory: static char * qemuBuildSmpArgStr(const virDomainDefPtr def, - int qemuCmdFlags) + unsigned long long qemuCmdFlags) { virBuffer buf = VIR_BUFFER_INITIALIZER; @@ -3057,7 +3057,7 @@ int qemudBuildCommandLine(virConnectPtr conn, virDomainDefPtr def, virDomainChrDefPtr monitor_chr, int monitor_json, - unsigned int qemuCmdFlags, + unsigned long long qemuCmdFlags, const char ***retargv, const char ***retenv, int **tapfds, diff --git a/src/qemu/qemu_conf.h b/src/qemu/qemu_conf.h index 5525b6c..f6397b7 100644 --- a/src/qemu/qemu_conf.h +++ b/src/qemu/qemu_conf.h @@ -170,10 +170,10 @@ virCapsPtr qemudCapsInit (virCapsPtr old_caps); int qemudExtractVersion (struct qemud_driver *driver); int qemudExtractVersionInfo (const char *qemu, unsigned int *version, - unsigned int *flags); + unsigned long long *qemuCmdFlags); int qemudParseHelpStr (const char *str, - unsigned int *flags, + unsigned long long *qemuCmdFlags, unsigned int *version, unsigned int *is_kvm, unsigned int *kvm_version); @@ -183,7 +183,7 @@ int qemudBuildCommandLine (virConnectPtr conn, virDomainDefPtr def, virDomainChrDefPtr monitor_chr, int monitor_json, - unsigned int qemuCmdFlags, + unsigned long long qemuCmdFlags, const char ***retargv, const char ***retenv, int **tapfds, @@ -205,12 +205,12 @@ char * qemuBuildNicStr(virConnectPtr conn, /* Current, best practice */ char * qemuBuildNicDevStr(virDomainNetDefPtr net, - int qemuCmdFlags); + int vlan); /* Both legacy & current support */ char *qemuBuildDriveStr(virDomainDiskDefPtr disk, int bootable, - int qemuCmdFlags); + unsigned long long qemuCmdFlags); /* Current, best practice */ char * qemuBuildDriveDevStr(virConnectPtr conn, @@ -244,7 +244,7 @@ char * qemuBuildUSBHostdevDevStr(virDomainHostdevDefPtr dev); int qemudNetworkIfaceConnect (virConnectPtr conn, struct qemud_driver *driver, virDomainNetDefPtr net, - int qemuCmdFlags); + unsigned long long qemuCmdFlags); int qemudProbeMachineTypes (const char *binary, virCapsGuestMachinePtr **machines, @@ -283,7 +283,7 @@ int qemuAssignDevicePCISlots(virDomainDefPtr def, qemuDomainPCIAddressSetPtr ad int qemuDomainNetVLAN(virDomainNetDefPtr def); int qemuAssignDeviceNetAlias(virDomainDefPtr def, virDomainNetDefPtr net, int idx); -int qemuAssignDeviceDiskAlias(virDomainDiskDefPtr def, int qemuCmdFlags); +int qemuAssignDeviceDiskAlias(virDomainDiskDefPtr def, unsigned long long qemuCmdFlags); int qemuAssignDeviceHostdevAlias(virDomainDefPtr def, virDomainHostdevDefPtr net, int idx); int qemuAssignDeviceControllerAlias(virDomainControllerDefPtr controller); diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 4374291..54d33cd 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -2520,7 +2520,7 @@ static int qemudStartVMDaemon(virConnectPtr conn, struct stat sb; int *tapfds = NULL; int ntapfds = 0; - unsigned int qemuCmdFlags; + unsigned long long qemuCmdFlags; fd_set keepfd; const char *emulator; pid_t child; @@ -4707,7 +4707,7 @@ static char *qemuDomainXMLToNative(virConnectPtr conn, virDomainDefPtr def = NULL; virDomainChrDef monConfig; const char *emulator; - unsigned int qemuCmdFlags; + unsigned long long qemuCmdFlags; struct stat sb; const char **retargv = NULL; const char **retenv = NULL; @@ -5586,7 +5586,7 @@ static int qemudDomainAttachNetDevice(virConnectPtr conn, struct qemud_driver *driver, virDomainObjPtr vm, virDomainNetDefPtr net, - unsigned int qemuCmdFlags) + unsigned long long qemuCmdFlags) { qemuDomainObjPrivatePtr priv = vm->privateData; char *tapfd_name = NULL; @@ -5901,7 +5901,7 @@ static int qemudDomainAttachDevice(virDomainPtr dom, struct qemud_driver *driver = dom->conn->privateData; virDomainObjPtr vm; virDomainDeviceDefPtr dev = NULL; - unsigned int qemuCmdFlags; + unsigned long long qemuCmdFlags; virCgroupPtr cgroup = NULL; int ret = -1; @@ -7437,7 +7437,7 @@ qemudDomainMigratePrepareTunnel(virConnectPtr dconn, int ret = -1; int internalret; char *unixfile = NULL; - unsigned int qemuCmdFlags; + unsigned long long qemuCmdFlags; struct qemuStreamMigFile *qemust = NULL; qemuDriverLock(driver); @@ -7884,7 +7884,7 @@ static int doTunnelMigrate(virDomainPtr dom, virStreamPtr st = NULL; char *unixfile = NULL; int internalret; - unsigned int qemuCmdFlags; + unsigned long long qemuCmdFlags; int status; unsigned long long transferred, remaining, total; diff --git a/tests/qemuargv2xmltest.c b/tests/qemuargv2xmltest.c index 1f1914b..c125a34 100644 --- a/tests/qemuargv2xmltest.c +++ b/tests/qemuargv2xmltest.c @@ -75,7 +75,7 @@ static int testCompareXMLToArgvFiles(const char *xml, struct testInfo { const char *name; - int extraFlags; + unsigned long long extraFlags; const char *migrateFrom; }; diff --git a/tests/qemuhelptest.c b/tests/qemuhelptest.c index 34a6048..0f2b509 100644 --- a/tests/qemuhelptest.c +++ b/tests/qemuhelptest.c @@ -12,10 +12,10 @@ struct testInfo { const char *name; - unsigned flags; - unsigned version; - unsigned is_kvm; - unsigned kvm_version; + unsigned long long flags; + unsigned int version; + unsigned int is_kvm; + unsigned int kvm_version; }; static char *progname; @@ -41,7 +41,8 @@ static int testHelpStrParsing(const void *data) char path[PATH_MAX]; char helpStr[MAX_HELP_OUTPUT_SIZE]; char *help = &(helpStr[0]); - unsigned flags, version, is_kvm, kvm_version; + unsigned int version, is_kvm, kvm_version; + unsigned long long flags; snprintf(path, PATH_MAX, "%s/qemuhelpdata/%s", abs_srcdir, info->name); @@ -52,7 +53,7 @@ static int testHelpStrParsing(const void *data) return -1; if (flags != info->flags) { - fprintf(stderr, "Computed flags do not match: got 0x%x, expected 0x%x\n", + fprintf(stderr, "Computed flags do not match: got 0x%llx, expected 0x%llx\n", flags, info->flags); if (getenv("VIR_TEST_DEBUG")) diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index 67dc47e..f4908b8 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -24,7 +24,7 @@ static struct qemud_driver driver; static int testCompareXMLToArgvFiles(const char *xml, const char *cmd, - int extraFlags, + unsigned long long extraFlags, const char *migrateFrom) { char argvData[MAX_FILE]; char *expectargv = &(argvData[0]); @@ -32,7 +32,8 @@ static int testCompareXMLToArgvFiles(const char *xml, const char **argv = NULL; const char **qenv = NULL; const char **tmp = NULL; - int ret = -1, len, flags; + int ret = -1, len; + unsigned long long flags; virDomainDefPtr vmdef = NULL; virDomainChrDef monitor_chr; @@ -142,7 +143,7 @@ static int testCompareXMLToArgvFiles(const char *xml, struct testInfo { const char *name; - int extraFlags; + unsigned long long extraFlags; const char *migrateFrom; }; -- 1.6.6

3 6

[libvirt] [PATCH] Fix security driver configuration
by Daniel P. Berrange 10 Feb '10

10 Feb '10

The security driver was mistakenly initialized before the QEMU config file was loaded. This prevents it being turned off again. The capabilities XML was also getting the wrong security driver name, due to the stacked driver arrangement. * src/qemu/qemu_driver.c: Fix initialization order and capabilities model name --- src/qemu/qemu_driver.c | 20 ++++++++++---------- 1 files changed, 10 insertions(+), 10 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 54d33cd..cac5b8a 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -933,7 +933,7 @@ qemudSecurityInit(struct qemud_driver *qemud_drv) static virCapsPtr qemuCreateCapabilities(virCapsPtr oldcaps, - virSecurityDriverPtr secDriver) + struct qemud_driver *driver) { virCapsPtr caps; @@ -951,11 +951,11 @@ qemuCreateCapabilities(virCapsPtr oldcaps, /* Security driver data */ - if (secDriver) { + if (driver->securityPrimaryDriver) { const char *doi, *model; - doi = virSecurityDriverGetDOI(secDriver); - model = virSecurityDriverGetModel(secDriver); + doi = virSecurityDriverGetDOI(driver->securityPrimaryDriver); + model = virSecurityDriverGetModel(driver->securityPrimaryDriver); if (!(caps->host.secModel.model = strdup(model))) goto no_memory; @@ -1098,20 +1098,20 @@ qemudStartup(int privileged) { virStrerror(-rc, buf, sizeof(buf))); } + if (qemudLoadDriverConfig(qemu_driver, driverConf) < 0) { + goto error; + } + if (qemudSecurityInit(qemu_driver) < 0) goto error; if ((qemu_driver->caps = qemuCreateCapabilities(NULL, - qemu_driver->securityDriver)) == NULL) + qemu_driver)) == NULL) goto error; if ((qemu_driver->activePciHostdevs = pciDeviceListNew()) == NULL) goto error; - if (qemudLoadDriverConfig(qemu_driver, driverConf) < 0) { - goto error; - } - if (privileged) { if (chown(qemu_driver->libDir, qemu_driver->user, qemu_driver->group) < 0) { virReportSystemError(errno, @@ -3065,7 +3065,7 @@ static char *qemudGetCapabilities(virConnectPtr conn) { qemuDriverLock(driver); if ((caps = qemuCreateCapabilities(qemu_driver->caps, - qemu_driver->securityDriver)) == NULL) { + qemu_driver)) == NULL) { virCapabilitiesFree(caps); goto cleanup; } -- 1.6.6

2 1

[libvirt] PATCH: Depend on libcgroup & turn on cgconfig initscript
by Daniel P. Berrange 10 Feb '10

10 Feb '10

Since libvirt uses cgroups for various aspects of guest tuning, we need to make sure that the host is configured with them enabled out of the box. The libcgroup maintainer has now provided a suitable default config that mounts them all in /cgroup. So libvirt simply needs to depend on libcgroup RPM, and turn on the 'cgconfig' initscript at install time if QEMU or LXC is enabled. diff -u -r1.81 libvirt.spec --- libvirt.spec 3 Feb 2010 17:54:54 -0000 1.81 +++ libvirt.spec 9 Feb 2010 12:49:25 -0000 @@ -61,6 +61,7 @@ %define with_udev 0%{!?_without_udev:0} %define with_hal 0%{!?_without_hal:0} %define with_yajl 0%{!?_without_yajl:0} +%define with_cgconfig 0%{!?_without_cgconfig:0} # Non-server/HV driver defaults which are always enabled %define with_python 0%{!?_without_python:1} @@ -147,6 +148,13 @@ %define with_yajl 0%{!?_without_yajl:%{server_drivers}} %endif +# Pull in cgroups config system +%if 0%{?fedora} >= 12 || 0%{?rhel} >= 6 +%if %{with_qemu} || %{with_lxc} +%define with_cgconfig 0%{!?_without_cgconfig:1} +%endif +%endif + # Force QEMU to run as non-root %if 0%{?fedora} >= 12 || 0%{?rhel} >= 6 %define qemu_user qemu @@ -244,6 +252,9 @@ # For multipath support Requires: device-mapper %endif +%if %{with_cgconfig} +Requires: libcgroup +%endif %if %{with_xen} BuildRequires: xen-devel %endif @@ -634,6 +645,12 @@ fi %endif +%if %{with_cgconfig} +if [ "$1" = "1" ]; then +/sbin/chkconfig cgconfig on +fi +%endif + /sbin/chkconfig --add libvirtd if [ "$1" -ge "1" ]; then /sbin/service libvirtd condrestart > /dev/null 2>&1 -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 1

[libvirt] [PATCH] escape strings serialized in XML
by Daniel Veillard 10 Feb '10

10 Feb '10

A large part of the serialization code in src/conf/node_device_conf.c forgot that rule, this led to https://bugzilla.redhat.com/show_bug.cgi?id=547045 Escape strings serialized in XML * src/conf/node_device_conf.c: string saved into XML should be escaped with the use of virBufferEscapeString() diff --git a/src/conf/node_device_conf.c b/src/conf/node_device_conf.c index 4e8a492..69f42a9 100644 --- a/src/conf/node_device_conf.c +++ b/src/conf/node_device_conf.c @@ -372,19 +372,21 @@ char *virNodeDeviceDefFormat(virConnectPtr conn, virBufferVSprintf(&buf, " <protocol>%d</protocol>\n", data->usb_if.protocol); if (data->usb_if.description) - virBufferVSprintf(&buf, " <description>%s</description>\n", + virBufferEscapeString(&buf, + " <description>%s</description>\n", data->usb_if.description); break; case VIR_NODE_DEV_CAP_NET: - virBufferVSprintf(&buf, " <interface>%s</interface>\n", + virBufferEscapeString(&buf, " <interface>%s</interface>\n", data->net.ifname); if (data->net.address) - virBufferVSprintf(&buf, " <address>%s</address>\n", + virBufferEscapeString(&buf, " <address>%s</address>\n", data->net.address); if (data->net.subtype != VIR_NODE_DEV_CAP_NET_LAST) { const char *subtyp = virNodeDevNetCapTypeToString(data->net.subtype); - virBufferVSprintf(&buf, " <capability type='%s'/>\n", subtyp); + virBufferEscapeString(&buf, " <capability type='%s'/>\n", + subtyp); } break; case VIR_NODE_DEV_CAP_SCSI_HOST: @@ -392,10 +394,10 @@ char *virNodeDeviceDefFormat(virConnectPtr conn, data->scsi_host.host); if (data->scsi_host.flags & VIR_NODE_DEV_CAP_FLAG_HBA_FC_HOST) { virBufferAddLit(&buf, " <capability type='fc_host'>\n"); - virBufferVSprintf(&buf, - " <wwnn>%s</wwnn>\n", data->scsi_host.wwnn); - virBufferVSprintf(&buf, - " <wwpn>%s</wwpn>\n", data->scsi_host.wwpn); + virBufferEscapeString(&buf, " <wwnn>%s</wwnn>\n", + data->scsi_host.wwnn); + virBufferEscapeString(&buf, " <wwpn>%s</wwpn>\n", + data->scsi_host.wwpn); virBufferAddLit(&buf, " </capability>\n"); } if (data->scsi_host.flags & VIR_NODE_DEV_CAP_FLAG_HBA_VPORT_OPS) { @@ -405,8 +407,8 @@ char *virNodeDeviceDefFormat(virConnectPtr conn, break; case VIR_NODE_DEV_CAP_SCSI_TARGET: - virBufferVSprintf(&buf, " <target>%s</target>\n", - data->scsi_target.name); + virBufferEscapeString(&buf, " <target>%s</target>\n", + data->scsi_target.name); break; case VIR_NODE_DEV_CAP_SCSI: @@ -416,23 +418,23 @@ char *virNodeDeviceDefFormat(virConnectPtr conn, data->scsi.target); virBufferVSprintf(&buf, " <lun>%d</lun>\n", data->scsi.lun); if (data->scsi.type) - virBufferVSprintf(&buf, " <type>%s</type>\n", - data->scsi.type); + virBufferEscapeString(&buf, " <type>%s</type>\n", + data->scsi.type); break; case VIR_NODE_DEV_CAP_STORAGE: - virBufferVSprintf(&buf, " <block>%s</block>\n", + virBufferEscapeString(&buf, " <block>%s</block>\n", data->storage.block); if (data->storage.bus) - virBufferVSprintf(&buf, " <bus>%s</bus>\n", + virBufferEscapeString(&buf, " <bus>%s</bus>\n", data->storage.bus); if (data->storage.drive_type) - virBufferVSprintf(&buf, " <drive_type>%s</drive_type>\n", + virBufferEscapeString(&buf, " <drive_type>%s</drive_type>\n", data->storage.drive_type); if (data->storage.model) - virBufferVSprintf(&buf, " <model>%s</model>\n", + virBufferEscapeString(&buf, " <model>%s</model>\n", data->storage.model); if (data->storage.vendor) - virBufferVSprintf(&buf, " <vendor>%s</vendor>\n", + virBufferEscapeString(&buf, " <vendor>%s</vendor>\n", data->storage.vendor); if (data->storage.serial) virBufferVSprintf(&buf, " <serial>%s</serial>\n", -- 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/

2 2

[libvirt] [0/3] Rebased patches for SPICE graphics support
by Daniel Veillard 10 Feb '10

10 Feb '10

This is a rebase of the 4 RHEL specific patches posted last sptember by Dan Berrange (but with patch 3 and 4 merged). It's actually a rebase from the gitorious spice set from Dan http://www.redhat.com/archives/libvir-list/2009-September/msg00865.html -- 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/

2 7

[libvirt] [PATCH] fix AppArmor security driver when libvirt is compiled with libcap-ng
by Jamie Strandboge 10 Feb '10

10 Feb '10

The calls to virExec() in security_apparmor.c when invoking virt-aa-helper use VIR_EXEC_CLEAR_CAPS. When compiled without libcap-ng, this is not a problem (it's effectively a no-op) but with libcap-ng this causes MAC_ADMIN to be cleared. MAC_ADMIN is needed by virt-aa-helper to manipulate apparmor profiles and without it VMs will not start[1]. This patch calls virExec with the default VIR_EXEC_NONE instead. [1] https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/517714 -- Jamie Strandboge | http://www.canonical.com

2 1

[libvirt] [PATCH 0/9] virDomain{Attach,Detach}DeviceFlags patches
by Jim Fehlig 09 Feb '10

09 Feb '10

This set implements two new APIs, virDomainAttachDeviceFlags and virDomainDetachDeviceFlags as discussed here https://www.redhat.com/archives/libvir-list/2009-December/msg00124.html Introduce two new APIs virDomainAttachDeviceFlags(virDomainPtr dom, const char *xml, unsigned flags) virDomainDetachDeviceFlags(virDomainPtr dom, const char *xml, unsigned flags) with flags being one or more from VIR_DOMAIN_DEVICE_MODIFY_CURRENT, VIR_DOMAIN_DEVICE_MODIFY_LIVE, VIR_DOMAIN_DEVICE_MODIFY_CONFIG. If caller specifies CURRENT (default), add or remove the device depending on the current state of domain. E.g. if domain is active add or remove the device to/from live domain, if it is inactive change the persistent config. If caller specifies LIVE, only change the active domain. If caller specifies CONFIG, only change persistent config - even if the domain is active. If caller specifies both LIVE and CONFIG, then change both. If a driver can not satisfy the exact requested flags it must return an error. E.g if user specified LIVE but the driver can only change live and persisted config, the driver must fail the request. The existing virDomain{Attach,Detach}Device is now explicitly restricted to active domains and is equivalent to virDomain{Attach,Detach}DeviceFlags(LIVE). Finally, virsh {attach,detach}-{disk,interface,device} has been modified to add a --persistent flag in order to set the appropriate flags when calling the new APIs. Jim Fehlig (9): Restrict virDomain{Attach,Detach}Device to active domains Public API Internal API Public API Implementation Wire protocol format Remote driver Server side dispatcher domain{Attach,Detach}DeviceFlags handler for drivers Modify virsh commands daemon/remote.c | 53 ++++++++++++++++++ include/libvirt/libvirt.h.in | 13 +++++ src/driver.h | 10 ++++ src/esx/esx_driver.c | 2 + src/libvirt.c | 120 ++++++++++++++++++++++++++++++++++++++---- src/libvirt_public.syms | 6 ++ src/lxc/lxc_driver.c | 2 + src/opennebula/one_driver.c | 2 + src/openvz/openvz_driver.c | 2 + src/phyp/phyp_driver.c | 2 + src/qemu/qemu_driver.c | 26 +++++++++ src/remote/remote_driver.c | 54 +++++++++++++++++++ src/remote/remote_protocol.x | 17 ++++++- src/test/test_driver.c | 2 + src/uml/uml_driver.c | 2 + src/vbox/vbox_tmpl.c | 24 ++++++++ src/xen/proxy_internal.c | 4 +- src/xen/xen_driver.c | 42 +++++++++++++-- src/xen/xen_driver.h | 4 +- src/xen/xen_hypervisor.c | 4 +- src/xen/xen_inotify.c | 4 +- src/xen/xend_internal.c | 98 ++++++++++++++++++++++++++++------ src/xen/xm_internal.c | 30 +++++++---- src/xen/xs_internal.c | 4 +- tools/virsh.c | 55 +++++++++++++++++-- 25 files changed, 523 insertions(+), 59 deletions(-)

3 33

[libvirt] [PATCH] Create raw storage files with O_DSYNC
by Jiri Denemark 09 Feb '10

09 Feb '10

When creating preallocated large raw files opening them with O_DSYNC prevents long delays in reading because cache pages can be immediately reused without writing them on a disk first. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/storage/storage_backend.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/storage/storage_backend.c b/src/storage/storage_backend.c index e79294a..8aab9ab 100644 --- a/src/storage/storage_backend.c +++ b/src/storage/storage_backend.c @@ -301,7 +301,7 @@ virStorageBackendCreateRaw(virConnectPtr conn, goto cleanup; } - if ((fd = open(vol->target.path, O_RDWR | O_EXCL)) < 0) { + if ((fd = open(vol->target.path, O_RDWR | O_EXCL | O_DSYNC)) < 0) { virReportSystemError(conn, errno, _("cannot open new path '%s'"), vol->target.path); -- 1.6.6.1

2 2

[libvirt] ANNOUNCE: New releases virt-manager 0.8.3 and virtinst 0.500.2
by Cole Robinson 09 Feb '10

09 Feb '10

I'm happy to announce two new releases: virt-manager 0.8.3: virt-manager is a desktop application for managing KVM and Xen virtual machines via libvirt. virtinst 0.500.2: virtinst is a collection of command line tools for provisioning libvirt virtual machines, including virt-install and virt-clone. The releases can be downloaded from: http://virt-manager.org/download.html The direct download links are: http://virt-manager.org/download/sources/virt-manager/virt-manager-0.8.3.ta… http://virt-manager.org/download/sources/virtinst/virtinst-0.500.2.tar.gz The virt-manager release includes: - New ability to manage network interfaces: start, stop, and view existing interfaces. Provision new bridge, bond, and vlan devices. - New option to 'customize VM before install', which allows adjusting most VM options from the install wizard. The virtinst release includes: - virt-install: --autostart option for setting domain autostart flag - virt-install: Allow specifying --host-device with lsusb and lspci format (0x1234:0x5678 for USB, etc) - virtinst: New 'Interface' module for building libvirt interface XML Thanks to everyone who has contributed to this release through testing, bug reporting, submitting patches, and otherwise sending in feedback! Thanks, Cole

1 0

Re: [libvirt] Live migration woes
by Lance Albertson 09 Feb '10

09 Feb '10

> On Fri, Feb 05, 2010 at 03:21:32PM -0500, Thomas Sjolshagen wrote: >>> is anyone using libvirt-git and qemu-kvm post-0.11 and can live >>> migrate VMs successfully for- and backwards? qemu-kvm-0.11 seems >>> to be the latest release were live migration works, and I'm >>> trying to figure out if its a kvm or libvirt problem, since there >>> were other things (balloon parameter f.e.) that were changed with >>> qemu-kvm-0.12 and caused troubles when using libvirt. >> >> Yes. >> >> I'm using the virt-preview repo (rawhide) and I just noticed, >> today, after re-upgrading to 0.12.2-5 (since -4 has a rather nasty >> IO problem for virtio devices against raw image files) that my >> live-migrations fail with the following error message on the >> destination system (i.e. the system the I'm trying to migrate to): > >> /var/log/libvirt/qemu/<guest>.log: Unknown savevm section >> -2038417646 load of migration failed > > This looks like a new QEMU bug to me I can confirm this bug is happening on Ganeti as well [1]. It appears that if you sync the time on the guest VM before migrating back to the other node it will "succeed" most of the time. Has anyone opened a bug for this on the qemu-kvm bug tracker yet? I was about to do that myself. Thanks- [1] http://groups.google.com/group/ganeti/browse_thread/thread/9b2c556557e749cf -- Lance Albertson lance <at> osuosl.org Systems Administrator / Architect Open Source Lab Network Services Oregon State University Work: 541-737-9975 PGP Key: 0x27F4B742 GPG Fingerprint 0423 92F3 544A 1282 5AB1 4D07 416F A15D 27F4 B742

2 1

[libvirt] Inbound NAT and iptables rules
by Karl Vogel 09 Feb '10

09 Feb '10

This issue has been brought up a few times, but I haven't found any real solution yet. The problem is with the way libvirt adds iptables rules. I have a KVM Host with 1 public IP address, on which a few virtual machines run. I want to be able to direct certain ports on the public IP to certain virtual machine's internal IP. For example: allow conneciton to 'public-ip' tcp port: 80 to 'vm1' Which is possible using following rules: $ iptables -I FORWARD 1 -d vm1 -p tcp --dport 80 -j ACCEPT $ iptables -t nat -A PREROUTING -p tcp -d public-ip --dport 80 -j DNAT --to-destination vm1 However the problem is that libvirt _always_ adds it's rules to the top of the FORWARD rules, hence it blocks any 'custom' rules to the VM machines. ie. Chain FORWARD (policy ACCEPT 0 packets, 0 bytes) pkts bytes target prot opt in out source destination 39 2964 ACCEPT all -- * virbr0 0.0.0.0/0 192.168.122.0/24 state RELATED,ESTABLISHED 39 2964 ACCEPT all -- virbr0 * 192.168.122.0/24 0.0.0.0/0 2 656 ACCEPT all -- virbr0 virbr0 0.0.0.0/0 0.0.0.0/0 2 120 REJECT all -- * virbr0 0.0.0.0/0 0.0.0.0/0 reject-with icmp-port-unreachable 0 0 REJECT all -- virbr0 * 0.0.0.0/0 0.0.0.0/0 reject-with icmp-port-unreachable Only established connections are allowed with the libvirt rules, hence the forwarded connection to port 80 will be rejected by the 4th libvirt rule. The only way to make it work is to insert the rule _after_ libvirt has started the network.. and if the libvirt network ever gets restarted, the rule will be disabled since libvirt inserts it's rules at the top of the chain. Is there a proper way to achieve this with libvirt ?!

1 1

[libvirt] qemu can't parse floppy cmdline
by Cole Robinson 09 Feb '10

09 Feb '10

Hi Dan, With qemu-0.12.2-5.fc13.x86_64 in fedora, libvirt is generating a cmdline which qemu-kvm can't parse. XML is: <domain type='kvm'> <name>winxp_32</name> <uuid>3485e26d-c370-c0b0-32e2-a041a12f5cfc</uuid> <memory>1048576</memory> <currentMemory>1048576</currentMemory> <vcpu>1</vcpu> <os> <type arch='x86_64' machine='pc-0.11'>hvm</type> <boot dev='hd'/> </os> <features> <acpi/> <apic/> <pae/> </features> <clock offset='localtime'/> <on_poweroff>destroy</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>restart</on_crash> <devices> <emulator>/usr/bin/qemu-kvm</emulator> <disk type='file' device='disk'> <driver name='qemu'/> <source file='/mnt/devel/vms/disks/winxp_32.qcow2.img'/> <target dev='hda' bus='ide'/> <address type='drive' controller='0' bus='0' unit='0'/> </disk> <disk type='file' device='cdrom'> <driver name='qemu'/> <source file='/mnt/devel/media/win_2003_64.iso'/> <target dev='hdc' bus='ide'/> <readonly/> <address type='drive' controller='0' bus='1' unit='0'/> </disk> <disk type='file' device='cdrom'> <target dev='hdb' bus='ide'/> <readonly/> <address type='drive' controller='0' bus='0' unit='1'/> </disk> <disk type='file' device='floppy'> <driver name='qemu' type='raw'/> <source file='/var/lib/libvirt/images/1.img'/> <target dev='fda' bus='fdc'/> <address type='drive' controller='0' bus='0' unit='0'/> </disk> <controller type='ide' index='0'/> <controller type='fdc' index='0'/> <interface type='network'> <mac address='52:54:00:5d:46:22'/> <source network='default'/> <target dev='vnet0'/> </interface> <serial type='pty'> <target port='0'/> </serial> <console type='pty'> <target port='0'/> </console> <input type='tablet' bus='usb'/> <input type='mouse' bus='ps2'/> <graphics type='vnc' port='-1' autoport='yes'/> <sound model='es1370'/> <video> <model type='vga' vram='9216' heads='1'/> </video> </devices> </domain> The generated cmdline is: LC_ALL=C PATH=/sbin:/usr/sbin:/bin:/usr/bin QEMU_AUDIO_DRV=none /usr/bin/qemu-kvm -S -M pc-0.11 -enable-kvm -m 1024 -smp 1,sockets=1,cores=1,threads=1 -name winxp_32 -uuid 3485e26d-c370-c0b0-32e2-a041a12f5cfc -nodefaults -chardev socket,id=monitor,path=/var/lib/libvirt/qemu/winxp_32.monitor,server,nowait -mon chardev=monitor,mode=readline -localtime -boot c -drive file=/mnt/devel/vms/disks/winxp_32.qcow2.img,if=none,id=drive-ide0-0-0,boot=on -device ide-drive,bus=ide.0,unit=0,drive=drive-ide0-0-0,id=ide0-0-0 -drive file=/mnt/devel/media/win_2003_64.iso,if=none,media=cdrom,id=drive-ide0-1-0 -device ide-drive,bus=ide.1,unit=0,drive=drive-ide0-1-0,id=ide0-1-0 -drive if=none,media=cdrom,id=drive-ide0-0-1 -device ide-drive,bus=ide.0,unit=1,drive=drive-ide0-0-1,id=ide0-0-1 -drive file=/var/lib/libvirt/images/1.img,if=none,id=drive-fdc0-0-0,format=raw -global isa-fdc,driveA=drive-fdc0-0-0 -device rtl8139,vlan=0,id=net0,mac=52:54:00:5d:46:22,bus=pci.0,addr=0x4 -net tap,fd=38,vlan=0,name=hostnet0 -chardev pty,id=serial0 -device isa-serial,chardev=serial0 -usb -device usb-tablet,id=input0 -vnc 127.0.0.1:0 -vga std -device ES1370,id=sound0,bus=pci.0,addr=0x5 -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x3 can't parse: "isa-fdc,driveA=drive-fdc0-0-0" If I drop the floppy dev and controller, the guest boots, but a few warnings are printed. Not sure if any of this reflects qemu bugs or libvirt bugs: LC_ALL=C PATH=/sbin:/usr/sbin:/bin:/usr/bin QEMU_AUDIO_DRV=none /usr/bin/qemu-kvm -S -M pc-0.11 -enable-kvm -m 1024 -smp 1,sockets=1,cores=1,threads=1 -name winxp_32 -uuid 3485e26d-c370-c0b0-32e2-a041a12f5cfc -nodefaults -chardev socket,id=monitor,path=/var/lib/libvirt/qemu/winxp_32.monitor,server,nowait -mon chardev=monitor,mode=readline -localtime -boot c -drive file=/mnt/devel/vms/disks/winxp_32.qcow2.img,if=none,id=drive-ide0-0-0,boot=on -device ide-drive,bus=ide.0,unit=0,drive=drive-ide0-0-0,id=ide0-0-0 -drive file=/mnt/devel/media/win_2003_64.iso,if=none,media=cdrom,id=drive-ide0-1-0 -device ide-drive,bus=ide.1,unit=0,drive=drive-ide0-1-0,id=ide0-1-0 -drive if=none,media=cdrom,id=drive-ide0-0-1 -device ide-drive,bus=ide.0,unit=1,drive=drive-ide0-0-1,id=ide0-0-1 -device rtl8139,vlan=0,id=net0,mac=52:54:00:5d:46:22,bus=pci.0,addr=0x4 -net tap,fd=37,vlan=0,name=hostnet0 -chardev pty,id=serial0 -device isa-serial,chardev=serial0 -usb -device usb-tablet,id=input0 -vnc 127.0.0.1:0 -vga std -device ES1370,id=sound0,bus=pci.0,addr=0x5 -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x3 char device redirected to /dev/pts/2 Warning: vlan 0 with no nics Option 'ipv4': Use 'on' or 'off' Failed to parse "yes" for "dummy.ipv4" Thanks, Cole

2 1

[libvirt] [PATCH 0/9] Remove conn parameter from several functions
by Matthias Bolte 09 Feb '10

09 Feb '10

This set of patches removes the conn parameter from several internal functions where it was used of error reporting only. Since errors are reported to an error object in thread local storage the conn parameter is not necessary for this anymore. Matthias

2 11

[libvirt] [PATCH 2/5] macvtap support for libvirt -- parse new interface XML
by Stefan Berger 08 Feb '10

08 Feb '10

This part adds support to domain_conf.{c|h} for parsing the new interface XML of type 'direct'. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com>

2 3

[libvirt] [PATCH 4/5] macvtap support for libvirt -- helper code
by Stefan Berger 08 Feb '10

08 Feb '10

This part adds the helper code to setup and tear down macvtap devices using direct communication with the device driver via netlink sockets. Signed-off-by: Stefan Berger <stefanb(a)us.ibm.com>

2 2

[libvirt] RFC: solving problem with qemu domain save on root-squashing NFS
by Laine Stump 08 Feb '10

08 Feb '10

(I had sent this message last week as a response to a message in a patch thread, but thought today that possibly some mail readers would bury it, so I'm sending it again disconnected from any existing thread. I can send the patches I mention below to anyone interested in playing around with them, but they're not useful since they still fail at the final step...) On 01/24/2010 11:57 AM, Dan Kenigsberg wrote: > On Wed, Jan 20, 2010 at 03:14:57PM +0000, Daniel P. Berrange wrote: >> This patch series does some work on te security drivers, and the QEMU >> code >> for managing DAC permissions on files. >> >> The core goal is to turn the QEMU driver DAC file management code into a >> security driver. Instead of QEMU calling into the SELinux/AppArmour >> drivers >> directly, a stacked driver module is introduced. This delegates all >> operations >> to first the QEMU DAC driver, and then the main SELinux/AppArmour >> driver. >> The end result is that all the permissions management code is removed >> from >> the QEMU driver, and we're left with just simple security driver calls. >> >> In the process of this a number of flaws in the current hotplug code >> were >> found, and code was generally tidied up with a view to making it >> easier to >> manage. >> >> Finally, we add the ability to turn off the QEMU DAC file managment >> code, >> and also deal gracefully with failures to change ownership (eg on NFS >> with >> root squash, or readonly FS). > Thanks for this series. However, it seems that we still have a problem > when trying to save domain to a root-squashing nfs export. > When using qemu directly, as a user with write permissions to that > export, there is no problem. When using libvirt, libvirt tries to write > its own state to the target file. I would not want to pre-create the > target file as world redable. > > How about performing open(path, O_CREAT|O_TRUNC|O_WRONLY, > S_IRUSR|S_IWUSR)) with the euid of the qemu process? I came up with a patch that creates the domain state file by first forking, followed by the child process doing setuid(qemu) and all the write-to-the-file stuff that creates the file (similar to the recent virFileCreate patch). This works all the way up through qemu's appending its state onto the file. The trouble comes when qemudDomainSave calls securityDriver->domainRestoreSavedStateLabel() (in this case, qemuSecurityStackedRestoreSavedStateLabel()). That function first calls the primary security driver's qemuSecurityDACRestoreSavedStateLabel(), which will fail because it attempts to chown the file to 0,0 (to prevent other qemu instances from messing with it). That can be disabled by setting "dynamic_ownership=0" in qemu.conf (needed in several other places when stuff is stored on root-squashing NFS anyway :-/). Then the stacked security driver calls the secondary security driver's SELinuxRestoreSavedStateLabel(), which also fails because of an SELinux function (matchpathcon) returning failure. We could ignore this failure, but that would create an unnecessary security hole for all those installations that *aren't* storing their domains on root-squashing NFS. So I *think* what I'm looking for a good way to avoid the SELinux stuff when the destination file is located on a root-squashing NFS share, but not ignore it otherwise (I believe the consensus is that we *don't* want to require another configuration knob if at all possible, since that would not only lead to extra complexity of config, but also to uninformed users turning it off when not necessary). Does anyone have any good ideas for doing this (or for a different approach)? (I suppose I could always attempt the entire operation assuming not, then redo with fork/setuid if the first attempt failed, but that seems less than elegant).

1 0

[libvirt] Live migration woes
by Thomas Treutner 08 Feb '10

08 Feb '10

Hi, is anyone using libvirt-git and qemu-kvm post-0.11 and can live migrate VMs successfully for- and backwards? qemu-kvm-0.11 seems to be the latest release were live migration works, and I'm trying to figure out if its a kvm or libvirt problem, since there were other things (balloon parameter f.e.) that were changed with qemu-kvm-0.12 and caused troubles when using libvirt. thx,t

3 5

[libvirt] Regarding lxc driver for libvirt.
by Kumar L Srikanth-B22348 08 Feb '10

08 Feb '10

Hi, I am new to libvirt. I want to create a Domain using libvirt XML. In order to mount the host's '/home/srikanth' directory to the new container's '/' directory, my XML format is shown below: <domain type='lxc' id='1'> <name>container1_vm</name> <memory>500000</memory> <os> <type>exe</type> <init>/bin/sh</init> </os> <vcpu>1</vcpu> <clock offset='utc'/> <on_poweroff>destroy</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>destroy</on_crash> <devices> <emulator>/usr/libexec/libvirt_lxc</emulator> <filesystem type='mount'> <source dir='/home/srikanth'/> <target dir='/'/> </filesystem> <console type='pty' /> </devices> </domain> With the above libvirt XML, Domain is defining, but not starting. When I issue the start command it's saying "Domain started", but showing "shut off" status. If I changed the target directory(<traget dir='/'/>) from '/' to '/home/container1'(<traget dir='/home/container1'/>), the domain is starting normally and I am able to see the contents in the target directory. Can you please let me know, how can I set the target directory to '/'? By the way, I am using libvirt version o.7.6. Regards, Srikanth.

2 1

[libvirt] [PATCH UPDATE] Add support for virtio channel
by Matthew Booth 08 Feb '10

08 Feb '10

Amit pointed out that byte_limit, guest_byte_limit and cache_buffers are now gone. They are removed in these updated patches.

2 3

[libvirt] FYI, [PATCH] build: update gnulib submodule to latest
by Jim Meyering 05 Feb '10

05 Feb '10

FYI, I've just pushed this: >From 26bd7bef58296300cd81756297de514b9c2073df Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Fri, 5 Feb 2010 17:57:39 +0100 Subject: [PATCH] build: update gnulib submodule to latest * .gnulib: This fixes a warning in test-gettimeofday.c, seen via "make check". Reported by Daniel Veillard. --- .gnulib | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/.gnulib b/.gnulib index 9d0ad65..11fbc57 160000 --- a/.gnulib +++ b/.gnulib @@ -1 +1 @@ -Subproject commit 9d0ad652de159d08e5f679842f8a2a5658196361 +Subproject commit 11fbc57405a118e6ec9a3ebc19bbf5ececdae4d6 -- 1.7.0.rc1.204.gb96e

1 0

[libvirt] FW: Regarding lxc driver for libvirt.
by Kumar L Srikanth-B22348 05 Feb '10

05 Feb '10

Can anyone help in the below issue? Regards, Srikanth. -----Original Message----- From: Kumar L Srikanth-B22348 Sent: Friday, February 05, 2010 4:02 PM To: 'libvir-list(a)redhat.com' Subject: Regarding lxc driver for libvirt. Hi, I am new to libvirt. I want to create a Domain using libvirt XML. In order to mount the host's '/home/srikanth' directory to the new container's '/' directory, my XML format is shown below: <domain type='lxc' id='1'> <name>container1_vm</name> <memory>500000</memory> <os> <type>exe</type> <init>/bin/sh</init> </os> <vcpu>1</vcpu> <clock offset='utc'/> <on_poweroff>destroy</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>destroy</on_crash> <devices> <emulator>/usr/libexec/libvirt_lxc</emulator> <filesystem type='mount'> <source dir='/home/srikanth'/> <target dir='/'/> </filesystem> <console type='pty' /> </devices> </domain> With the above libvirt XML, Domain is defining, but not starting. When I issue the start command it's saying "Domain started", but showing "shut off" status. If I changed the target directory(<traget dir='/'/>) from '/' to '/home/container1'(<traget dir='/home/container1'/>), the domain is starting normally and I am able to see the contents in the target directory. Can you please let me know, how can I set the target directory to '/'? By the way, I am using libvirt version o.7.6. Regards, Srikanth.

1 0

[libvirt] [PATCH] absolutePathFromBaseFile: don't leak when first arg contains no "/"
by Jim Meyering 05 Feb '10

05 Feb '10

Not only did this function leak(p), but it would also over-allocate (by the length of basename(base_file)), and then later, re-alloc to compensate, so I rewrote it: >From 1dc52930daa000b407d8a8f18588a19cf4e8b7f5 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Thu, 4 Feb 2010 16:55:57 +0100 Subject: [PATCH] absolutePathFromBaseFile: don't leak when first arg contains no "/" * src/util/storage_file.c: Include "dirname.h". (absolutePathFromBaseFile): Rewrite not to leak, and to require fewer allocations. * bootstrap (modules): Add dirname-lgpl and stpncpy. * .gnulib: Update submodule to the latest. --- .gnulib | 2 +- bootstrap | 2 ++ src/util/storage_file.c | 27 ++++++++++----------------- 3 files changed, 13 insertions(+), 18 deletions(-) diff --git a/.gnulib b/.gnulib index 146d914..9d0ad65 160000 --- a/.gnulib +++ b/.gnulib @@ -1 +1 @@ -Subproject commit 146d9145073e62a2096a2d6b33f75e93908fedf3 +Subproject commit 9d0ad652de159d08e5f679842f8a2a5658196361 diff --git a/bootstrap b/bootstrap index cc3c6ef..5cc43c5 100755 --- a/bootstrap +++ b/bootstrap @@ -71,6 +71,7 @@ c-ctype canonicalize-lgpl close connect +dirname-lgpl getaddrinfo gethostname getpass @@ -93,6 +94,7 @@ send setsockopt socket stpcpy +stpncpy strchrnul strndup strerror diff --git a/src/util/storage_file.c b/src/util/storage_file.c index 44057d2..8c53fba 100644 --- a/src/util/storage_file.c +++ b/src/util/storage_file.c @@ -1,7 +1,7 @@ /* * storage_file.c: file utility functions for FS storage backend * - * Copyright (C) 2007-2009 Red Hat, Inc. + * Copyright (C) 2007-2010 Red Hat, Inc. * Copyright (C) 2007-2008 Daniel P. Berrange * * This library is free software; you can redistribute it and/or @@ -26,6 +26,7 @@ #include <unistd.h> #include <fcntl.h> +#include "dirname.h" #include "memory.h" #include "virterror_internal.h" @@ -246,26 +247,18 @@ vmdk4GetBackingStore(virConnectPtr conn, static char * absolutePathFromBaseFile(const char *base_file, const char *path) { - size_t base_size, path_size; - char *res, *p; + char *res; + size_t d_len = dir_len (base_file); - if (*path == '/') + /* If path is already absolute, or if dirname(base_file) is ".", + just return a copy of path. */ + if (*path == '/' || d_len == 0) return strdup(path); - base_size = strlen(base_file) + 1; - path_size = strlen(path) + 1; - if (VIR_ALLOC_N(res, base_size - 1 + path_size) < 0) + if (VIR_ALLOC_N(res, d_len + 1 + strlen(path) + 1) < 0) return NULL; - memcpy(res, base_file, base_size); - p = strrchr(res, '/'); - if (p != NULL) - p++; - else - p = res; - memcpy(p, path, path_size); - if (VIR_REALLOC_N(res, (p + path_size) - res) < 0) { - /* Ignore failure */ - } + + stpcpy(stpcpy(stpncpy(res, base_file, d_len), "/"), path); return res; } -- 1.7.0.rc1.199.g9253a

3 6

[libvirt] [PATCH 0/12] Improve security driver handling & QEMU DAC management
by Daniel P. Berrange 05 Feb '10

05 Feb '10

This patch series does some work on te security drivers, and the QEMU code for managing DAC permissions on files. The core goal is to turn the QEMU driver DAC file management code into a security driver. Instead of QEMU calling into the SELinux/AppArmour drivers directly, a stacked driver module is introduced. This delegates all operations to first the QEMU DAC driver, and then the main SELinux/AppArmour driver. The end result is that all the permissions management code is removed from the QEMU driver, and we're left with just simple security driver calls. In the process of this a number of flaws in the current hotplug code were found, and code was generally tidied up with a view to making it easier to manage. Finally, we add the ability to turn off the QEMU DAC file managment code, and also deal gracefully with failures to change ownership (eg on NFS with root squash, or readonly FS).

5 37

[libvirt] [PATCH] storage: Replace storageLog with VIR_ERROR
by Matthias Bolte 04 Feb '10

04 Feb '10

--- src/storage/storage_driver.c | 23 ++++++++++------------- 1 files changed, 10 insertions(+), 13 deletions(-) diff --git a/src/storage/storage_driver.c b/src/storage/storage_driver.c index 50fcbe2..37be77d 100644 --- a/src/storage/storage_driver.c +++ b/src/storage/storage_driver.c @@ -40,11 +40,10 @@ #include "storage_conf.h" #include "memory.h" #include "storage_backend.h" +#include "logging.h" #define VIR_FROM_THIS VIR_FROM_STORAGE -#define storageLog(msg...) fprintf(stderr, msg) - static virStorageDriverStatePtr driverState; static int storageDriverShutdown(void); @@ -70,8 +69,7 @@ storageDriverAutostart(virStorageDriverStatePtr driver) { !virStoragePoolObjIsActive(pool)) { virStorageBackendPtr backend; if ((backend = virStorageBackendForType(pool->def->type)) == NULL) { - storageLog("Missing backend %d", - pool->def->type); + VIR_ERROR("Missing backend %d", pool->def->type); virStoragePoolObjUnlock(pool); continue; } @@ -79,9 +77,9 @@ storageDriverAutostart(virStorageDriverStatePtr driver) { if (backend->startPool && backend->startPool(NULL, pool) < 0) { virErrorPtr err = virGetLastError(); - storageLog("Failed to autostart storage pool '%s': %s", - pool->def->name, err ? err->message : - "no error message found"); + VIR_ERROR("Failed to autostart storage pool '%s': %s", + pool->def->name, err ? err->message : + "no error message found"); virStoragePoolObjUnlock(pool); continue; } @@ -90,9 +88,9 @@ storageDriverAutostart(virStorageDriverStatePtr driver) { virErrorPtr err = virGetLastError(); if (backend->stopPool) backend->stopPool(NULL, pool); - storageLog("Failed to autostart storage pool '%s': %s", - pool->def->name, err ? err->message : - "no error message found"); + VIR_ERROR("Failed to autostart storage pool '%s': %s", + pool->def->name, err ? err->message : + "no error message found"); virStoragePoolObjUnlock(pool); continue; } @@ -132,7 +130,6 @@ storageDriverStartup(int privileged) { goto error; if (virAsprintf(&base, "%s/.libvirt", userdir) == -1) { - storageLog("out of memory in virAsprintf"); VIR_FREE(userdir); goto out_of_memory; } @@ -175,7 +172,7 @@ storageDriverStartup(int privileged) { return 0; out_of_memory: - storageLog("virStorageStartup: out of memory"); + virReportOOMError(NULL); error: VIR_FREE(base); storageDriverUnlock(driverState); @@ -635,7 +632,7 @@ storagePoolUndefine(virStoragePoolPtr obj) { if (unlink(pool->autostartLink) < 0 && errno != ENOENT && errno != ENOTDIR) { char ebuf[1024]; - storageLog("Failed to delete autostart link '%s': %s", + VIR_ERROR("Failed to delete autostart link '%s': %s", pool->autostartLink, virStrerror(errno, ebuf, sizeof ebuf)); } -- 1.6.3.3

3 4

[libvirt] [PATCH] website: Add a 1em right margin
by Matthias Bolte 04 Feb '10

04 Feb '10

This stops the text and pre-boxes from touching the right border. --- docs/libvirt.css | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/docs/libvirt.css b/docs/libvirt.css index a4d1ddd..dfc93c6 100644 --- a/docs/libvirt.css +++ b/docs/libvirt.css @@ -30,6 +30,7 @@ h2, h3, h4, h5, h6 { #content { margin-left: 230px; + margin-right: 1em; padding: 0px; padding-bottom: 1em; } -- 1.6.3.3

2 2

[libvirt] [PATCH] opennebula: Remove unnecessary casts
by Matthias Bolte 04 Feb '10

04 Feb '10

--- src/opennebula/one_driver.c | 36 ++++++++++++++++++------------------ 1 files changed, 18 insertions(+), 18 deletions(-) diff --git a/src/opennebula/one_driver.c b/src/opennebula/one_driver.c index ad7faca..c96d24d 100644 --- a/src/opennebula/one_driver.c +++ b/src/opennebula/one_driver.c @@ -108,7 +108,7 @@ static int oneIsSecure(virConnectPtr conn ATTRIBUTE_UNUSED) static virDomainPtr oneDomainLookupByID(virConnectPtr conn, int id) { - one_driver_t *driver = (one_driver_t *)conn->privateData; + one_driver_t *driver = conn->privateData; virDomainPtr dom = NULL; virDomainObjPtr vm = NULL; @@ -137,7 +137,7 @@ return_point: static virDomainPtr oneDomainLookupByUUID(virConnectPtr conn, const unsigned char *uuid) { - one_driver_t *driver = (one_driver_t *)conn->privateData; + one_driver_t *driver = conn->privateData; virDomainPtr dom = NULL; virDomainObjPtr vm = NULL; @@ -165,7 +165,7 @@ return_point: static virDomainPtr oneDomainLookupByName(virConnectPtr conn, const char *name) { - one_driver_t *driver = (one_driver_t *)conn->privateData; + one_driver_t *driver = conn->privateData; virDomainObjPtr vm = NULL; virDomainPtr dom=NULL; @@ -192,7 +192,7 @@ return_point: static int oneListDomains(virConnectPtr conn, int *ids, int nids) { - one_driver_t *driver = (one_driver_t *)conn->privateData; + one_driver_t *driver = conn->privateData; int n; oneDriverLock(driver); @@ -204,7 +204,7 @@ static int oneListDomains(virConnectPtr conn, int *ids, int nids) static int oneNumDomains(virConnectPtr conn) { - one_driver_t *driver = (one_driver_t *)conn->privateData; + one_driver_t *driver = conn->privateData; int n; oneDriverLock(driver); @@ -216,7 +216,7 @@ static int oneNumDomains(virConnectPtr conn) static int oneListDefinedDomains(virConnectPtr conn, char **const names, int nnames) { - one_driver_t *driver = (one_driver_t *)conn->privateData; + one_driver_t *driver = conn->privateData; int n; oneDriverLock(driver); @@ -228,7 +228,7 @@ static int oneListDefinedDomains(virConnectPtr conn, static int oneNumDefinedDomains(virConnectPtr conn) { - one_driver_t *driver = (one_driver_t *)conn->privateData; + one_driver_t *driver = conn->privateData; int n; oneDriverLock(driver); @@ -240,7 +240,7 @@ static int oneNumDefinedDomains(virConnectPtr conn) static virDomainPtr oneDomainDefine(virConnectPtr conn, const char *xml) { - one_driver_t *driver = (one_driver_t *)conn->privateData; + one_driver_t *driver = conn->privateData; virDomainDefPtr def; virDomainObjPtr vm; virDomainPtr dom=NULL; @@ -272,7 +272,7 @@ return_point: static int oneDomainUndefine(virDomainPtr dom) { - one_driver_t *driver = (one_driver_t *)dom->conn->privateData; + one_driver_t *driver = dom->conn->privateData; virDomainObjPtr vm = NULL; int ret=-1; @@ -302,7 +302,7 @@ return_point: static int oneDomainGetInfo(virDomainPtr dom, virDomainInfoPtr info) { - one_driver_t *driver = (one_driver_t *)dom->conn->privateData; + one_driver_t *driver = dom->conn->privateData; struct timeval tv; virDomainObjPtr vm; oneDriverLock(driver); @@ -377,7 +377,7 @@ static int oneDomainGetInfo(virDomainPtr dom, static char *oneGetOSType(virDomainPtr dom) { - one_driver_t *driver = (one_driver_t *)dom->conn->privateData; + one_driver_t *driver = dom->conn->privateData; virDomainObjPtr vm = NULL; char *ret = NULL; @@ -403,7 +403,7 @@ cleanup: static int oneDomainStart(virDomainPtr dom) { virConnectPtr conn = dom->conn; - one_driver_t *driver = (one_driver_t *)(conn->privateData); + one_driver_t *driver = conn->privateData; virDomainObjPtr vm; int ret = -1; int oneid; @@ -436,7 +436,7 @@ static virDomainPtr oneDomainCreateAndStart(virConnectPtr conn, const char *xml, unsigned int flags ATTRIBUTE_UNUSED) { - one_driver_t *driver = (one_driver_t *)conn->privateData; + one_driver_t *driver = conn->privateData; virDomainObjPtr vm = NULL; virDomainDefPtr def; virDomainPtr dom = NULL; @@ -486,7 +486,7 @@ return_point: static int oneDomainShutdown(virDomainPtr dom) { - one_driver_t *driver = (one_driver_t*)dom->conn->privateData; + one_driver_t *driver = dom->conn->privateData; virDomainObjPtr vm; int ret=-1; @@ -520,7 +520,7 @@ return_point: static int oneDomainDestroy(virDomainPtr dom) { - one_driver_t *driver = (one_driver_t*)dom->conn->privateData; + one_driver_t *driver = dom->conn->privateData; virDomainObjPtr vm; int ret=-1; @@ -556,7 +556,7 @@ return_point: static int oneDomainSuspend(virDomainPtr dom) { - one_driver_t* driver=dom->conn->privateData; + one_driver_t* driver = dom->conn->privateData; virDomainObjPtr vm; int ret=-1; @@ -590,7 +590,7 @@ return_point: static int oneDomainResume(virDomainPtr dom) { - one_driver_t* driver=dom->conn->privateData; + one_driver_t* driver = dom->conn->privateData; virDomainObjPtr vm; int ret=-1; @@ -698,7 +698,7 @@ static int oneGetAutostart(virDomainPtr domain ATTRIBUTE_UNUSED, int *autostart) } static char* oneGetCapabilities(virConnectPtr conn){ - one_driver_t* privconn=conn->privateData; + one_driver_t* privconn = conn->privateData; char *xml; oneDriverLock(privconn); if ((xml = virCapabilitiesFormatXML(privconn->caps)) == NULL) -- 1.6.3.3

2 2

[libvirt] [PATCH] cpu conf: Use virBufferFreeAndReset instead of virBufferContentAndReset and VIR_FREE
by Matthias Bolte 04 Feb '10

04 Feb '10

--- src/conf/cpu_conf.c | 4 +--- 1 files changed, 1 insertions(+), 3 deletions(-) diff --git a/src/conf/cpu_conf.c b/src/conf/cpu_conf.c index e7924c1..244d670 100644 --- a/src/conf/cpu_conf.c +++ b/src/conf/cpu_conf.c @@ -249,7 +249,6 @@ virCPUDefFormat(virConnectPtr conn, int flags) { virBuffer buf = VIR_BUFFER_INITIALIZER; - char *tmp; if (virCPUDefFormatBuf(conn, &buf, def, indent, flags) < 0) goto cleanup; @@ -262,8 +261,7 @@ virCPUDefFormat(virConnectPtr conn, no_memory: virReportOOMError(conn); cleanup: - tmp = virBufferContentAndReset(&buf); - VIR_FREE(tmp); + virBufferFreeAndReset(&buf); return NULL; } -- 1.6.3.3

2 2

[libvirt] [PATCH] docs: Refer to virReportOOMError in the HACKING file
by Matthias Bolte 04 Feb '10

04 Feb '10

Instead of refering to __virRaiseError(VIR_ERROR_NO_MEMORY). --- HACKING | 22 +++++++++++----------- docs/hacking.html.in | 20 ++++++++++---------- 2 files changed, 21 insertions(+), 21 deletions(-) diff --git a/HACKING b/HACKING index 3844e50..4bf216c 100644 --- a/HACKING +++ b/HACKING @@ -166,13 +166,13 @@ codebase, because they encourage a number of serious coding bugs and do not enable compile time verification of checks for NULL. Instead of these routines, use the macros from memory.h - - eg to allocate a single object: + - eg to allocate a single object: virDomainPtr domain; if (VIR_ALLOC(domain) < 0) { - __virRaiseError(VIR_ERROR_NO_MEMORY) - return NULL; + virReportOOMError(NULL); + return NULL; } @@ -182,8 +182,8 @@ routines, use the macros from memory.h int ndomains = 10; if (VIR_ALLOC_N(domains, ndomains) < 0) { - __virRaiseError(VIR_ERROR_NO_MEMORY) - return NULL; + virReportOOMError(NULL); + return NULL; } - eg to allocate an array of object pointers @@ -192,8 +192,8 @@ routines, use the macros from memory.h int ndomains = 10; if (VIR_ALLOC_N(domains, ndomains) < 0) { - __virRaiseError(VIR_ERROR_NO_MEMORY) - return NULL; + virReportOOMError(NULL); + return NULL; } - eg to re-allocate the array of domains to be longer @@ -201,8 +201,8 @@ routines, use the macros from memory.h ndomains = 20 if (VIR_REALLOC_N(domains, ndomains) < 0) { - __virRaiseError(VIR_ERROR_NO_MEMORY) - return NULL; + virReportOOMError(NULL); + return NULL; } - eg to free the domain @@ -297,7 +297,7 @@ eg typical usage is as follows: if (virBufferError(&buf)) { virBufferFreeAndReset(&buf); - virReportOOMError(...); + virReportOOMError(NULL); return NULL; } @@ -385,7 +385,7 @@ to build: - if a recently commited patch breaks compilation on a platform or for a given driver then it's fine to commit a minimal fix directly without getting the review feedback first - - similary if make check or make syntax-chek breaks, if there is + - similary if make check or make syntax-check breaks, if there is an obvious fix, it's fine to commit immediately The patch should still be sent to the list (or tell what the fix was if trivial) and 'make check syntax-check' should pass too before commiting diff --git a/docs/hacking.html.in b/docs/hacking.html.in index 96f6657..71b4f47 100644 --- a/docs/hacking.html.in +++ b/docs/hacking.html.in @@ -192,8 +192,8 @@ virDomainPtr domain; if (VIR_ALLOC(domain) < 0) { - __virRaiseError(VIR_ERROR_NO_MEMORY) - return NULL; + virReportOOMError(NULL); + return NULL; } </pre></li> @@ -204,8 +204,8 @@ int ndomains = 10; if (VIR_ALLOC_N(domains, ndomains) < 0) { - __virRaiseError(VIR_ERROR_NO_MEMORY) - return NULL; + virReportOOMError(NULL); + return NULL; } </pre></li> @@ -216,8 +216,8 @@ int ndomains = 10; if (VIR_ALLOC_N(domains, ndomains) < 0) { - __virRaiseError(VIR_ERROR_NO_MEMORY) - return NULL; + virReportOOMError(NULL); + return NULL; } </pre></li> @@ -227,8 +227,8 @@ ndomains = 20 if (VIR_REALLOC_N(domains, ndomains) < 0) { - __virRaiseError(VIR_ERROR_NO_MEMORY) - return NULL; + virReportOOMError(NULL); + return NULL; } </pre></li> @@ -314,7 +314,7 @@ if (virBufferError(&buf)) { virBufferFreeAndReset(&buf); - virReportOOMError(...); + virReportOOMError(NULL); return NULL; } @@ -419,7 +419,7 @@ <li>if a recently commited patch breaks compilation on a platform or for a given driver then it's fine to commit a minimal fix directly without getting the review feedback first</li> - <li>if make check or make syntax-chek breaks, if there is + <li>if make check or make syntax-check breaks, if there is an obvious fix, it's fine to commit immediately. The patch should still be sent to the list (or tell what the fix was if trivial) and 'make check syntax-check' should pass too before commiting -- 1.6.3.3

2 2

[libvirt] [PATCH] esx: Cleanup preprocessing structure in esxVI_EnsureSession
by Matthias Bolte 04 Feb '10

04 Feb '10

--- src/esx/esx_vi.c | 14 ++++++-------- 1 files changed, 6 insertions(+), 8 deletions(-) diff --git a/src/esx/esx_vi.c b/src/esx/esx_vi.c index bc94326..19c85c5 100644 --- a/src/esx/esx_vi.c +++ b/src/esx/esx_vi.c @@ -1317,16 +1317,18 @@ esxVI_BuildFullTraversalSpecList(esxVI_SelectionSpec **fullTraversalSpecList) * you try to call it. Query the session manager for the current session of * this connection instead and re-login if there is no current session for this * connection. + * + * Update: 'ESX 4.0.0 build-171294' doesn't implement this method. */ #define ESX_VI_USE_SESSION_IS_ACTIVE 0 int esxVI_EnsureSession(esxVI_Context *ctx) { - int result = 0; #if ESX_VI_USE_SESSION_IS_ACTIVE esxVI_Boolean active = esxVI_Boolean_Undefined; #else + int result = 0; esxVI_String *propertyNameList = NULL; esxVI_ObjectContent *sessionManager = NULL; esxVI_DynamicProperty *dynamicProperty = NULL; @@ -1352,6 +1354,8 @@ esxVI_EnsureSession(esxVI_Context *ctx) return -1; } } + + return 0; #else if (esxVI_String_AppendValueToList(&propertyNameList, "currentSession") < 0 || @@ -1388,16 +1392,11 @@ esxVI_EnsureSession(esxVI_Context *ctx) "last login"); goto failure; } -#endif cleanup: -#if ESX_VI_USE_SESSION_IS_ACTIVE - /* nothing */ -#else esxVI_String_Free(&propertyNameList); esxVI_ObjectContent_Free(&sessionManager); esxVI_UserSession_Free(&currentSession); -#endif return result; @@ -1405,8 +1404,7 @@ esxVI_EnsureSession(esxVI_Context *ctx) result = -1; goto cleanup; - - return 0; +#endif } -- 1.6.3.3

2 2

[libvirt] [PATCH] docs: Emphasize that devices have to be inside the <devices> element
by Matthias Bolte 04 Feb '10

04 Feb '10

Also cleanup indentation of domain XML examples. --- docs/drvxen.html.in | 12 +- docs/formatdomain.html.in | 776 +++++++++++++++++++++++++-------------------- 2 files changed, 430 insertions(+), 358 deletions(-) diff --git a/docs/drvxen.html.in b/docs/drvxen.html.in index a6f0e57..4e35afa 100644 --- a/docs/drvxen.html.in +++ b/docs/drvxen.html.in @@ -234,9 +234,9 @@ vif = [ "mac=00:16:3e:60:36:ba,bridge=virbr0,script=vif-bridge,vifname=vif5.0" ] <on_reboot>restart</on_reboot> <on_crash>restart</on_crash> <features> - <pae/> - <acpi/> - <apic/> + <pae/> + <acpi/> + <apic/> </features> <clock sync="localtime"/> <devices> @@ -288,9 +288,9 @@ vif = [ "mac=00:16:3e:60:36:ba,bridge=virbr0,script=vif-bridge,vifname=vif5.0" ] <on_reboot>restart</on_reboot> <on_crash>restart</on_crash> <features> - <pae/> - <acpi/> - <apic/> + <pae/> + <acpi/> + <apic/> </features> <clock sync="localtime"/> <devices> diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index ce49f7d..083a80a 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -28,11 +28,11 @@ <h3><a name="elementsMetadata">General metadata</a></h3> - <pre> - <domain type='xen' id='3'> - <name>fv0</name> - <uuid>4dea22b31d52d8f32516782e98ab3fa0</uuid> - ...</pre> +<pre> +<domain type='xen' id='3'> + <name>fv0</name> + <uuid>4dea22b31d52d8f32516782e98ab3fa0</uuid> + ...</pre> <dl> <dt><code>name</code></dt> @@ -66,14 +66,14 @@ to obtain/find the boot image. </p> - <pre> - ... - <os> - <type>hvm</type> - <loader>/usr/lib/xen/boot/hvmloader</loader> - <boot dev='hd'/> - </os> - ...</pre> +<pre> + ... + <os> + <type>hvm</type> + <loader>/usr/lib/xen/boot/hvmloader</loader> + <boot dev='hd'/> + </os> + ...</pre> <dl> <dt><code>type</code></dt> @@ -109,11 +109,11 @@ An example is <code>pygrub</code> with Xen. </p> - <pre> - ... - <bootloader>/usr/bin/pygrub</bootloader> - <bootloader_args>--append single</bootloader_args> - ...</pre> +<pre> + ... + <bootloader>/usr/bin/pygrub</bootloader> + <bootloader_args>--append single</bootloader_args> + ...</pre> <dl> <dt><code>bootloader</code></dt> @@ -139,16 +139,16 @@ is usually available for both para and full virtualized guests. </p> - <pre> - ... - <os> - <type>hvm</type> - <loader>/usr/lib/xen/boot/hvmloader</loader> - <kernel>/root/f8-i386-vmlinuz</kernel> - <initrd>/root/f8-i386-initrd</initrd> - <cmdline>console=ttyS0 ks=http://example.com/f8-i386/os/</cmdline> - </os> - ...</pre> +<pre> + ... + <os> + <type>hvm</type> + <loader>/usr/lib/xen/boot/hvmloader</loader> + <kernel>/root/f8-i386-vmlinuz</kernel> + <initrd>/root/f8-i386-initrd</initrd> + <cmdline>console=ttyS0 ks=http://example.com/f8-i386/os/</cmdline> + </os> + ...</pre> <dl> <dt><code>type</code></dt> @@ -172,15 +172,15 @@ <h3><a name="elementsResources">Basic resources</a></h3> - <pre> - ... - <memory>524288</memory> - <currentMemory>524288</currentMemory> - <memoryBacking> - <hugepages/> - </memoryBacking> - <vcpu>1</vcpu> - ...</pre> +<pre> + ... + <memory>524288</memory> + <currentMemory>524288</currentMemory> + <memoryBacking> + <hugepages/> + </memoryBacking> + <vcpu>1</vcpu> + ...</pre> <dl> <dt><code>memory</code></dt> @@ -209,14 +209,14 @@ <span class="since">Since 0.7.5</span> </p> - <pre> - ... - <cpu match='exact'> - <model>core2duo</model> - <topology sockets='1' cores='2' threads='1'/> - <feature policy='disable' name='lahf_lm'/> - </cpu> - ...</pre> +<pre> + ... + <cpu match='exact'> + <model>core2duo</model> + <topology sockets='1' cores='2' threads='1'/> + <feature policy='disable' name='lahf_lm'/> + </cpu> + ...</pre> <p> In case no restrictions need to be put on CPU model and its features, a @@ -224,12 +224,12 @@ <span class="since">Since 0.7.6</span> </p> - <pre> - ... - <cpu> - <topology sockets='1' cores='2' threads='1'/> - </cpu> - ...</pre> +<pre> + ... + <cpu> + <topology sockets='1' cores='2' threads='1'/> + </cpu> + ...</pre> <dl> <dt><code>cpu</code></dt> @@ -305,12 +305,12 @@ re-configured for the first post-install bootup. </p> - <pre> - ... - <on_poweroff>destroy</on_poweroff> - <on_reboot>restart</on_reboot> - <on_crash>restart</on_crash> - ...</pre> +<pre> + ... + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>restart</on_crash> + ...</pre> <dl> <dt><code>on_poweroff</code></dt> @@ -350,14 +350,14 @@ toggled on/off. </p> - <pre> - ... - <features> - <pae/> - <acpi/> - <apic/> - </features> - ...</pre> +<pre> + ... + <features> + <pae/> + <acpi/> + <apic/> + </features> + ...</pre> <p> All features are listed within the <code>features</code> @@ -386,10 +386,10 @@ it to be in so called 'localtime'. </p> - <pre> - ... - <clock offset="localtime"/> - ...</pre> +<pre> + ... + <clock offset="localtime"/> + ...</pre> <dl> <dt><code>clock</code></dt> @@ -408,11 +408,12 @@ <span class="since">Since 0.1.3</span> </p> - <pre> - ... - <devices> - <emulator>/usr/lib/xen/bin/qemu-dm</emulator> - ...</pre> +<pre> + ... + <devices> + <emulator>/usr/lib/xen/bin/qemu-dm</emulator> + </devices> + ...</pre> <dl> <dt><code>emulator</code></dt> @@ -433,18 +434,20 @@ element. </p> - <pre> - ... - <disk type='file'> - <driver name="tap" type="aio" cache="default"> - <source file='/var/lib/xen/images/fv0'/> - <target dev='hda' bus='ide'/> - <encryption type='...'> - ... - </encryption> - <shareable/> - </disk> - ...</pre> +<pre> + ... + <devices> + <disk type='file'> + <driver name="tap" type="aio" cache="default"> + <source file='/var/lib/xen/images/fv0'/> + <target dev='hda' bus='ide'/> + <encryption type='...'> + ... + </encryption> + <shareable/> + </disk> + </devices> + ...</pre> <dl> <dt><code>disk</code></dt> @@ -500,24 +503,30 @@ 0.4.4 for USB and 0.6.0 for PCI (KVM only)</span>: </p> - <pre> - ... - <hostdev mode='subsystem' type='usb'> - <source> - <vendor id='0x1234'/> - <product id='0xbeef'/> - </source> - </hostdev> - ...</pre> +<pre> + ... + <devices> + <hostdev mode='subsystem' type='usb'> + <source> + <vendor id='0x1234'/> + <product id='0xbeef'/> + </source> + </hostdev> + </devices> + ...</pre> + <p>or:</p> - <pre> - ... - <hostdev mode='subsystem' type='pci'> - <source> - <address bus='0x06' slot='0x02' function='0x0'/> - </source> - </hostdev> - ...</pre> + +<pre> + ... + <devices> + <hostdev mode='subsystem' type='pci'> + <source> + <address bus='0x06' slot='0x02' function='0x0'/> + </source> + </hostdev> + </devices> + ...</pre> <dl> <dt><code>hostdev</code></dt> @@ -556,14 +565,16 @@ <h4><a name="elementsNICS">Network interfaces</a></h4> - <pre> - ... - <interface type='bridge'> - <source bridge='xenbr0'/> - <mac address='00:16:3e:5d:c7:9e'/> - <script path='vif-bridge'/> - </interface> - ...</pre> +<pre> + ... + <devices> + <interface type='bridge'> + <source bridge='xenbr0'/> + <mac address='00:16:3e:5d:c7:9e'/> + <script path='vif-bridge'/> + </interface> + </devices> + ...</pre> <h5><a name="elementsNICSVirtual">Virtual network</a></h5> @@ -589,18 +600,20 @@ <a href="#elementsNICSTargetOverride">overriding the target element</a>). </p> - <pre> - ... - <interface type='network'> - <source network='default'/> - </interface> - ... - <interface type='network'> - <source network='default'/> - <target dev='vnet7'/> - <mac address="11:22:33:44:55:66"/> - </interface> - ...</pre> +<pre> + ... + <devices> + <interface type='network'> + <source network='default'/> + </interface> + ... + <interface type='network'> + <source network='default'/> + <target dev='vnet7'/> + <mac address="11:22:33:44:55:66"/> + </interface> + </devices> + ...</pre> <h5><a name="elementsNICSBridge">Bridge to LAN</a></h5> @@ -623,18 +636,20 @@ full incoming & outgoing net access just like a physical machine. </p> - <pre> - ... - <interface type='bridge'> - <source bridge='br0'/> - </interface> - - <interface type='bridge'> - <source bridge='br0'/> - <target dev='vnet7'/> - <mac address="11:22:33:44:55:66"/> - </interface> - ...</pre> +<pre> + ... + <devices> + <interface type='bridge'> + <source bridge='br0'/> + </interface> + ... + <interface type='bridge'> + <source bridge='br0'/> + <target dev='vnet7'/> + <mac address="11:22:33:44:55:66"/> + </interface> + </devices> + ...</pre> <h5><a name="elementsNICSSlirp">Userspace SLIRP stack</a></h5> @@ -647,14 +662,16 @@ VMs to have outgoing access. </p> - <pre> - ... - <interface type='user'/> - ... - <interface type='user'> - <mac address="11:22:33:44:55:66"/> - </interface> - ...</pre> +<pre> + ... + <devices> + <interface type='user'/> + ... + <interface type='user'> + <mac address="11:22:33:44:55:66"/> + </interface> + </devices> + ...</pre> <h5><a name="elementsNICSEthernet">Generic ethernet connection</a></h5> @@ -669,15 +686,17 @@ overridden. </p> - <pre> - ... - <interface type='ethernet'/> - ... - <interface type='ethernet'> - <target dev='vnet7'/> - <script path='/etc/qemu-ifup-mynet'/> - </interface> - ...</pre> +<pre> + ... + <devices> + <interface type='ethernet'/> + ... + <interface type='ethernet'> + <target dev='vnet7'/> + <script path='/etc/qemu-ifup-mynet'/> + </interface> + </devices> + ...</pre> <h5><a name="elementsNICSMulticast">Multicast tunnel</a></h5> @@ -693,12 +712,14 @@ multicast address block. </p> - <pre> - ... - <interface type='mcast'> - <source address='230.0.0.1' port='5558'/> - </interface> - ...</pre> +<pre> + ... + <devices> + <interface type='mcast'> + <source address='230.0.0.1' port='5558'/> + </interface> + </devices> + ...</pre> <h5><a name="elementsNICSTCP">TCP tunnel</a></h5> @@ -711,27 +732,31 @@ network access, one of the VMs should have a 2nd NIC which is connected to one of the first 4 network types and do the appropriate routing.</p> - <pre> - ... - <interface type='server'> - <source address='192.168.0.1' port='5558'/> - </interface> - ... - <interface type='client'> +<pre> + ... + <devices> + <interface type='server'> <source address='192.168.0.1' port='5558'/> - </interface> - ...</pre> + </interface> + ... + <interface type='client'> + <source address='192.168.0.1' port='5558'/> + </interface> + </devices> + ...</pre> <h5><a name="elementsNICSModel">Setting the NIC model</a></h5> - <pre> - ... - <interface type='network'> - <source network='default'/> - <target dev='vnet1'/> - <b><model type='ne2k_pci'/></b> - </interface> - ...</pre> +<pre> + ... + <devices> + <interface type='network'> + <source network='default'/> + <target dev='vnet1'/> + <b><model type='ne2k_pci'/></b> + </interface> + </devices> + ...</pre> <p> For hypervisors which support this, you can set the model of @@ -757,13 +782,15 @@ qemu-kvm -net nic,model=? /dev/null <h5><a name="elementsNICSTargetOverride">Overriding the target element</a></h5> - <pre> - ... - <interface type='network'> - <source network='default'/> - <b><target dev='vnet1'/></b> - </interface> - ...</pre> +<pre> + ... + <devices> + <interface type='network'> + <source network='default'/> + <b><target dev='vnet1'/></b> + </interface> + </devices> + ...</pre> <p> If no target is specified, certain hypervisors will automatically @@ -783,10 +810,12 @@ qemu-kvm -net nic,model=? /dev/null to provide a graphics tablet for absolute cursor movement. </p> - <pre> - ... - <input type='mouse' bus='usb'/> - ...</pre> +<pre> + ... + <devices> + <input type='mouse' bus='usb'/> + </devices> + ...</pre> <dl> <dt><code>input</code></dt> @@ -807,13 +836,15 @@ qemu-kvm -net nic,model=? /dev/null admin. </p> - <pre> - ... - <graphics type='sdl' display=':0.0'/> - <graphics type='vnc' port='5904'/> - <graphics type='rdp' autoport='yes' multiUser='yes' /> - <graphics type='desktop' fullscreen='yes'/> - ...</pre> +<pre> + ... + <devices> + <graphics type='sdl' display=':0.0'/> + <graphics type='vnc' port='5904'/> + <graphics type='rdp' autoport='yes' multiUser='yes' /> + <graphics type='desktop' fullscreen='yes'/> + </devices> + ...</pre> <dl> <dt><code>graphics</code></dt> @@ -867,14 +898,16 @@ qemu-kvm -net nic,model=? /dev/null A video device. </p> - <pre> - ... - <video> - <model type='vga' vram='8192' heads='1'> - <acceleration accel3d='yes' accel3d='yes'/> - </model> - </video> - ...</pre> +<pre> + ... + <devices> + <video> + <model type='vga' vram='8192' heads='1'> + <acceleration accel3d='yes' accel3d='yes'/> + </model> + </video> + </devices> + ...</pre> <dl> <dt><code>video</code></dt> @@ -907,26 +940,27 @@ qemu-kvm -net nic,model=? /dev/null all classed as character devices and so represented using the same syntax. </p> - <pre> - ... - <parallel type='pty'> - <source path='/dev/pts/2'/> - <target port='0'/> - </parallel> - <serial type='pty'> - <source path='/dev/pts/3'/> - <target port='0'/> - </serial> - <console type='pty'> - <source path='/dev/pts/4'/> - <target port='0'/> - </console> - <channel type='unix'> - <source mode='bind' path='/tmp/guestfwd'/> - <target type='guestfwd' address='10.0.2.1' port='4600'/> - </channel> - </devices> - </domain></pre> +<pre> + ... + <devices> + <parallel type='pty'> + <source path='/dev/pts/2'/> + <target port='0'/> + </parallel> + <serial type='pty'> + <source path='/dev/pts/3'/> + <target port='0'/> + </serial> + <console type='pty'> + <source path='/dev/pts/4'/> + <target port='0'/> + </console> + <channel type='unix'> + <source mode='bind' path='/tmp/guestfwd'/> + <target type='guestfwd' address='10.0.2.1' port='4600'/> + </channel> + </devices> + ...</pre> <p> In each of these directives, the top-level element name (parallel, serial, @@ -949,13 +983,15 @@ qemu-kvm -net nic,model=? /dev/null <h6><a name="elementCharParallel">Parallel port</a></h6> - <pre> - ... - <parallel type='pty'> - <source path='/dev/pts/2'/> - <target port='0'/> - </parallel> - ...</pre> +<pre> + ... + <devices> + <parallel type='pty'> + <source path='/dev/pts/2'/> + <target port='0'/> + </parallel> + </devices> + ...</pre> <p> <code>target</code> can have a <code>port</code> attribute, which @@ -965,13 +1001,15 @@ qemu-kvm -net nic,model=? /dev/null <h6><a name="elementCharSerial">Serial port</a></h6> - <pre> - ... - <serial type='pty'> - <source path='/dev/pts/3'/> - <target port='0'/> - </serial> - ...</pre> +<pre> + ... + <devices> + <serial type='pty'> + <source path='/dev/pts/3'/> + <target port='0'/> + </serial> + </devices> + ...</pre> <p> <code>target</code> can have a <code>port</code> attribute, which @@ -987,13 +1025,15 @@ qemu-kvm -net nic,model=? /dev/null virtualized guests without a paravirtualized console. </p> - <pre> - ... - <console type='pty'> - <source path='/dev/pts/4'/> - <target port='0'/> - </console> - ...</pre> +<pre> + ... + <devices> + <console type='pty'> + <source path='/dev/pts/4'/> + <target port='0'/> + </console> + </devices> + ...</pre> <p> If the console is presented as a serial port, the <code>target</code> @@ -1008,13 +1048,15 @@ qemu-kvm -net nic,model=? /dev/null guest. </p> - <pre> - ... - <channel type='unix'> - <source mode='bind' path='/tmp/guestfwd'/> - <target type='guestfwd' address='10.0.2.1' port='4600'/> - </channel> - ...</pre> +<pre> + ... + <devices> + <channel type='unix'> + <source mode='bind' path='/tmp/guestfwd'/> + <target type='guestfwd' address='10.0.2.1' port='4600'/> + </channel> + </devices> + ...</pre> <p> This can be implemented in a variety of ways. The specific type of @@ -1045,12 +1087,14 @@ qemu-kvm -net nic,model=? /dev/null into the virtual machine's logfile </p> - <pre> - ... - <console type='stdio'> - <target port='1'> - </console> - ...</pre> +<pre> + ... + <devices> + <console type='stdio'> + <target port='1'> + </console> + </devices> + ...</pre> <h6><a name="elementsCharFle">Device logfile</a></h6> @@ -1060,13 +1104,15 @@ qemu-kvm -net nic,model=? /dev/null device is written to the file. </p> - <pre> - ... - <serial type="file"> - <source path="/var/log/vm/vm-serial.log"/> - <target port="1"/> - </serial> - ...</pre> +<pre> + ... + <devices> + <serial type="file"> + <source path="/var/log/vm/vm-serial.log"/> + <target port="1"/> + </serial> + </devices> + ...</pre> <h6><a name="elementsCharVC">Virtual console</a></h6> @@ -1076,12 +1122,14 @@ qemu-kvm -net nic,model=? /dev/null hotkey sequence such as "ctrl+alt+3" </p> - <pre> - ... - <serial type='vc'> - <target port="1"/> - </serial> - ...</pre> +<pre> + ... + <devices> + <serial type='vc'> + <target port="1"/> + </serial> + </devices> + ...</pre> <h6><a name="elementsCharNull">Null device</a></h6> @@ -1090,12 +1138,14 @@ qemu-kvm -net nic,model=? /dev/null provided to the input. All data written is discarded. </p> - <pre> - ... - <serial type='null'> - <target port="1"/> - </serial> - ...</pre> +<pre> + ... + <devices> + <serial type='null'> + <target port="1"/> + </serial> + </devices> + ...</pre> <h6><a name="elementsCharPTY">Pseudo TTY</a></h6> @@ -1105,13 +1155,15 @@ qemu-kvm -net nic,model=? /dev/null serial port locally. </p> - <pre> - ... - <serial type="pty"> - <source path="/dev/pts/3"/> - <target port="1"/> - </serial> - ...</pre> +<pre> + ... + <devices> + <serial type="pty"> + <source path="/dev/pts/3"/> + <target port="1"/> + </serial> + </devices> + ...</pre> <p> NB special case if <console type='pty'>, then the TTY @@ -1130,13 +1182,15 @@ qemu-kvm -net nic,model=? /dev/null port. </p> - <pre> - ... - <serial type="dev"> - <source path="/dev/ttyS0"/> - <target port="1"/> - </serial> - ...</pre> +<pre> + ... + <devices> + <serial type="dev"> + <source path="/dev/ttyS0"/> + <target port="1"/> + </serial> + </devices> + ...</pre> <h6><a name="elementsCharPipe">Named pipe</a></h6> @@ -1145,13 +1199,15 @@ qemu-kvm -net nic,model=? /dev/null more info. </p> - <pre> - ... - <serial type="pipe"> - <source path="/tmp/mypipe"/> - <target port="1"/> - </serial> - ...</pre> +<pre> + ... + <devices> + <serial type="pipe"> + <source path="/tmp/mypipe"/> + <target port="1"/> + </serial> + </devices> + ...</pre> <h6><a name="elementsCharTCP">TCP client/server</a></h6> @@ -1160,46 +1216,52 @@ qemu-kvm -net nic,model=? /dev/null remote server. </p> - <pre> - ... - <serial type="tcp"> - <source mode="connect" host="0.0.0.0" service="2445"/> - <protocol type="raw"/> - <target port="1"/> - </serial> - ...</pre> +<pre> + ... + <devices> + <serial type="tcp"> + <source mode="connect" host="0.0.0.0" service="2445"/> + <protocol type="raw"/> + <target port="1"/> + </serial> + </devices> + ...</pre> <p> Or as a TCP server waiting for a client connection. </p> - <pre> - ... - <serial type="tcp"> - <source mode="bind" host="127.0.0.1" service="2445"/> - <protocol type="raw"/> - <target port="1"/> - </serial> - ...</pre> +<pre> + ... + <devices> + <serial type="tcp"> + <source mode="bind" host="127.0.0.1" service="2445"/> + <protocol type="raw"/> + <target port="1"/> + </serial> + </devices> + ...</pre> <p> Alternatively you can use telnet instead of raw TCP. <p> - <pre> - ... - <serial type="tcp"> - <source mode="connect" host="0.0.0.0" service="2445"/> - <protocol type="telnet"/> - <target port="1"/> - </serial> - ... - <serial type="tcp"> - <source mode="bind" host="127.0.0.1" service="2445"/> - <protocol type="telnet"/> - <target port="1"/> - </serial> - ...</pre> +<pre> + ... + <devices> + <serial type="tcp"> + <source mode="connect" host="0.0.0.0" service="2445"/> + <protocol type="telnet"/> + <target port="1"/> + </serial> + ... + <serial type="tcp"> + <source mode="bind" host="127.0.0.1" service="2445"/> + <protocol type="telnet"/> + <target port="1"/> + </serial> + </devices> + ...</pre> <h6><a name="elementsCharUDP">UDP network console</a></h6> @@ -1208,14 +1270,16 @@ qemu-kvm -net nic,model=? /dev/null sending and receiving packets. This is a lossy service. </p> - <pre> - ... - <serial type="udp"> - <source mode="bind" host="0.0.0.0" service="2445"/> - <source mode="connect" host="0.0.0.0" service="2445"/> - <target port="1"/> - </serial> - ...</pre> +<pre> + ... + <devices> + <serial type="udp"> + <source mode="bind" host="0.0.0.0" service="2445"/> + <source mode="connect" host="0.0.0.0" service="2445"/> + <target port="1"/> + </serial> + </devices> + ...</pre> <h6><a name="elementsCharUNIX">UNIX domain socket client/server</a></h6> @@ -1224,13 +1288,15 @@ qemu-kvm -net nic,model=? /dev/null accepting connections from local clients. </p> - <pre> - ... - <serial type="unix"> - <source mode="bind" path="/tmp/foo"/> - <target port="1"/> - </serial> - ...</pre> +<pre> + ... + <devices> + <serial type="unix"> + <source mode="bind" path="/tmp/foo"/> + <target port="1"/> + </serial> + </devices> + ...</pre> <h4><a name="elementsSound">Sound devices</a></h4> @@ -1240,10 +1306,12 @@ qemu-kvm -net nic,model=? /dev/null <code>sound</code> element. <span class="since">Since 0.4.3</span> </p> - <pre> - ... - <sound model='es1370'/> - ...</pre> +<pre> + ... + <devices> + <sound model='es1370'/> + </devices> + ...</pre> <dl> <dt><code>sound</code></dt> @@ -1276,15 +1344,19 @@ qemu-kvm -net nic,model=? /dev/null libvirt. </p> - <pre> - ... - <watchdog model='i6300esb'/> - ...</pre> +<pre> + ... + <devices> + <watchdog model='i6300esb'/> + </devices> + ...</pre> - <pre> - ... - <watchdog model='i6300esb' action='poweroff'/> - ...</pre> +<pre> + ... + <devices> + <watchdog model='i6300esb' action='poweroff'/> + </devices> +</domain></pre> <dl> <dt><code>model</code></dt> -- 1.6.3.3

2 2

[libvirt] [PATCH 0/9] Progress monitoring & cancellation of jobs
by Daniel P. Berrange 04 Feb '10

04 Feb '10

This series introduces 2 new APIs to allow long running jobs to be monitored and cancelled. This applies to virDomainMigrate, virDomainSave, virDomainRestore and virDomainCoreDump at least. The implementation is provided for QEMU only, any others that are able to implementation can do so as desired. Usefulness of these new APIs requires use of multiple threads on a single connection, or multiple connections, since the main API calls are still all fully blocking. Instead a parallel thread or process can monitor. This can be seen with virsh In termainl 1 $ virsh save demo demo.img In terminal 2 $ ./virsh domjobinfo demo Job type: Unbounded Time elapsed: 795 ms Data processed: 22446080 bytes Data remaining: 219119616 bytes Data total: 241565696 bytes Memory processed: 22446080 bytes Memory remaining: 219119616 bytes Memory total: 241565696 bytes $ ./virsh domjobinfo demo Job type: Unbounded Time elapsed: 1669 ms Data processed: 139784192 bytes Data remaining: 101781504 bytes Data total: 241565696 bytes Memory processed: 139784192 bytes Memory remaining: 101781504 bytes Memory total: 241565696 bytes $ ./virsh domjobabort demo Back in terminal 1 error: Failed to save domain demo to demo.img error: operation failed: Migration was cancelled by client

1 9

[libvirt] [PATCH] sexpr_string: avoid leak on OOM error path
by Jim Meyering 04 Feb '10

04 Feb '10

>From cc06142a81dc1891dc75306b986ea331f5fd5368 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Thu, 4 Feb 2010 17:55:15 +0100 Subject: [PATCH] sexpr_string: avoid leak on OOM error path * src/xen/sexpr.c (sexpr_string): Free sexpr buffer upon allocation failure. --- src/xen/sexpr.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/src/xen/sexpr.c b/src/xen/sexpr.c index 085500d..669f8a2 100644 --- a/src/xen/sexpr.c +++ b/src/xen/sexpr.c @@ -107,30 +107,31 @@ sexpr_nil(void) struct sexpr * sexpr_string(const char *str, ssize_t len) { struct sexpr *ret = sexpr_new(); if (ret == NULL) return ret; ret->kind = SEXPR_VALUE; if (len > 0) { ret->u.value = strndup(str, len); } else { ret->u.value = strdup(str); } if (ret->u.value == NULL) { + VIR_FREE(ret); return NULL; } return ret; } /** * sexpr_cons: * @car: the left operand * @cdr: the right operand * * Implement the CONS operation assembling 2 existing S-Expressions. * Note that in case of error the input data are not freed. * * Returns the resulting S-Expression pointer or NULL in case of error. -- 1.7.0.rc1.199.g9253a

2 1

[libvirt] Suggested schedule for next release
by Daniel Veillard 04 Feb '10

04 Feb '10

Considering the amount of changes accumulating in a month, I suggest to continue with the end-of-month release cycle, which means basically that we would push features to head until Friday 19th and push the next release on the 26. This is a short cycle since Feb is a short month and we skipped a bit on 0.7.6 release, but I have no doubt we will have enough substance :-) There is a number of patches alredy lining up, the new APIs from Jim Fehlig, Dan Berrange clock sync patches, the macvlan patches, and the proposed snapshotting APIs (if we actually end up adding snapshotting I think next release would deserve being named 0.8.0 as this would be a major enhancement, otherwise 0.7.7), etc... Happy hacking ! 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