- Devel - libvirt List Archives

[libvirt] [PATCH] qemu_driver.c: avoid double free on error path
by Jim Meyering 15 Dec '09

15 Dec '09

Note that there's nearly identical code in the same file where there is already an "event = NULL;" statement right after two identical lines. >From 69f3c74b6cdc36fb4958f4c9cc2d4497a03d40c1 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 14 Dec 2009 15:55:19 +0100 Subject: [PATCH] qemu_driver.c: avoid double free on error path * src/qemu/qemu_driver.c (qemudDomainMigrateFinish2): Set "event" to NULL after qemuDomainEventQueue frees it, so a subsequent free (after endjob label) upon qemuMonitorStartCPUs failure does not cause a double free. --- 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 34fa582..43f7927 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -7668,6 +7668,7 @@ qemudDomainMigrateFinish2 (virConnectPtr dconn, VIR_DOMAIN_EVENT_DEFINED_UPDATED); if (event) qemuDomainEventQueue(driver, event); + event = NULL; } qemuDomainObjPrivatePtr priv = vm->privateData; -- 1.6.6.rc2.275.g51e2d

3 2

[libvirt] [PATCH] virsh: avoid double-free
by Jim Meyering 15 Dec '09

15 Dec '09

I didn't try to trigger the double free, but it looks feasible. >From 864cc161ffaf6c08cb980e66a673b53ac5ea52ab Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 14 Dec 2009 14:34:29 +0100 Subject: [PATCH] virsh: avoid double-free * tools/virsh.c (vshCommandParse): Avoid double-free of "tkdata". Set it to NULL immediately after free in the (cmd == NULL) case, just as in the other case, in case the final free(tkdata) is triggered by a syntax error. --- tools/virsh.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 62c1270..0312945 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -8044,6 +8044,7 @@ vshCommandParse(vshControl *ctl, char *cmdstr) goto syntaxError; /* ... or ignore this command only? */ } free(tkdata); + tkdata = NULL; } else if (tk == VSH_TK_OPTION) { if (!(opt = vshCmddefGetOption(cmd, tkdata))) { vshError(ctl, -- 1.6.6.rc2.275.g51e2d

3 2

[libvirt] [PATCH] remote_driver.c: also zero out ->saslDecodedOffset member
by Jim Meyering 15 Dec '09

15 Dec '09

Technically, the initialization of ->saslDecodedOffset is not required, but... What could go wrong if saslDecodedOffset is left with some positive value in the if-block below? Well, in that same if-block priv->saslDecoded is set to NULL, and in that case, the code above *usually* sets priv->saslDecodedOffset to 0, but not if we hit one of the intervening returns. In that case, you have to wonder if priv->saslDecoded can be set to non-NULL, yet with priv->saslDecodedOffset still left with the now-bogus positive value. In that case, with an initial saslDecodedLength of 0, we'd compute this: wantData = (priv->saslDecodedLength - priv->saslDecodedOffset and that would be negative. The very next statement would cause trouble, as the negative value treated-as-size_t would wreak havoc: memcpy(priv->buffer + priv->bufferOffset, priv->saslDecoded + priv->saslDecodedOffset, wantData); Luckily, the semantics of sasl_decode ensure that does not happen. Hence, an alternate (though riskier) patch would be to skip that of priv->saslDecodedOffset altogether. >From a3578fc03714a210678e6bcf62d162d92c9d07c0 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 14 Dec 2009 13:00:01 +0100 Subject: [PATCH] remote_driver.c: also zero out ->saslDecodedOffset member * src/remote/remote_driver.c (remoteIOReadMessage): ...rather than zeroing out priv->saslDecodedLength twice. --- src/remote/remote_driver.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/remote/remote_driver.c b/src/remote/remote_driver.c index affeb0f..77962fe 100644 --- a/src/remote/remote_driver.c +++ b/src/remote/remote_driver.c @@ -7782,7 +7782,7 @@ remoteIOReadMessage(virConnectPtr conn, struct private_data *priv, priv->saslDecodedOffset += wantData; priv->bufferOffset += wantData; if (priv->saslDecodedOffset == priv->saslDecodedLength) { - priv->saslDecodedLength = priv->saslDecodedLength = 0; + priv->saslDecodedOffset = priv->saslDecodedLength = 0; priv->saslDecoded = NULL; } -- 1.6.6.rc2.275.g51e2d

2 1

[libvirt] problem connecting to vbox
by anuj rampal 15 Dec '09

15 Dec '09

Hi, I have installed VBOX 3.0.12 on fc12 which is running libvirt-0.7.1. Now when i try to connect using virsh using uri "virsh -c vbox+tcp:///system" this is the error i get: error: internal error unable to initialize VirtualBox driver API error: failed to connect to the hypervisor But it connects successfully with this ur: "virsh -c vbox:///system" i have also qemu installed on the same machine and connects successfully with uri "virsh -c qemu+tcp:///system" Libvirt is listening on tcp port:16509 any idea what am i doing wrong? Here is a detailed message: LIBVIRT_DEBUG=1 virsh -c vbox+tcp:///system 11:10:45.469: debug : virInitialize:279 : register drivers 11:10:45.495: debug : virRegisterDriver:776 : registering Test as driver 0 11:10:45.495: debug : virRegisterNetworkDriver:614 : registering Test as network driver 0 11:10:45.495: debug : virRegisterInterfaceDriver:645 : registering Test as interface driver 0 11:10:45.495: debug : virRegisterStorageDriver:676 : registering Test as storage driver 0 11:10:45.495: debug : virRegisterDeviceMonitor:707 : registering Test as device driver 0 11:10:45.495: debug : virRegisterSecretDriver:738 : registering Test as secret driver 0 11:10:45.496: debug : virRegisterDriver:776 : registering Xen as driver 1 11:10:45.496: debug : virRegisterDriver:776 : registering OPENVZ as driver 2 11:10:45.496: debug : virRegisterDriver:776 : registering PHYP as driver 3 11:10:45.506: debug : vboxRegister:80 : VBoxCGlueInit found API version: 3.0.12 (3000012) 11:10:45.506: debug : vboxRegister:92 : VirtualBox API version: 3.0 11:10:45.506: debug : virRegisterDriver:776 : registering VBOX as driver 4 11:10:45.506: debug : virRegisterNetworkDriver:614 : registering VBOX as network driver 1 11:10:45.506: debug : virRegisterStorageDriver:676 : registering VBOX as storage driver 1 11:10:45.506: debug : virRegisterDriver:776 : registering ESX as driver 5 11:10:45.506: debug : virRegisterDriver:776 : registering remote as driver 6 11:10:45.506: debug : virRegisterNetworkDriver:614 : registering remote as network driver 2 11:10:45.506: debug : virRegisterInterfaceDriver:645 : registering remote as interface driver 1 11:10:45.506: debug : virRegisterStorageDriver:676 : registering remote as storage driver 2 11:10:45.506: debug : virRegisterDeviceMonitor:707 : registering remote as device driver 1 11:10:45.506: debug : virRegisterSecretDriver:738 : registering remote as secret driver 1 11:10:45.506: debug : virConnectOpenAuth:1273 : name=vbox+tcp:///system, auth=0x43d940, flags=0 11:10:45.506: debug : do_open:1042 : name "vbox+tcp:///system" to URI components: scheme vbox+tcp opaque (null) authority (null) server (null) user (null) port 0 path /system 11:10:45.506: debug : do_open:1052 : trying driver 0 (Test) ... 11:10:45.506: debug : do_open:1058 : driver 0 Test returned DECLINED 11:10:45.506: debug : do_open:1052 : trying driver 1 (Xen) ... 11:10:45.506: debug : do_open:1058 : driver 1 Xen returned DECLINED 11:10:45.506: debug : do_open:1052 : trying driver 2 (OPENVZ) ... 11:10:45.506: debug : do_open:1058 : driver 2 OPENVZ returned DECLINED 11:10:45.506: debug : do_open:1052 : trying driver 3 (PHYP) ... 11:10:45.506: debug : do_open:1058 : driver 3 PHYP returned DECLINED 11:10:45.506: debug : do_open:1052 : trying driver 4 (VBOX) ... 11:10:45.506: debug : do_open:1058 : driver 4 VBOX returned DECLINED 11:10:45.506: debug : do_open:1052 : trying driver 5 (ESX) ... 11:10:45.506: debug : do_open:1058 : driver 5 ESX returned DECLINED 11:10:45.506: debug : do_open:1052 : trying driver 6 (remote) ... 11:10:45.506: debug : doRemoteOpen:535 : proceeding with name = vbox:///system 11:10:45.508: debug : remoteIO:7421 : Do proc=66 serial=0 length=28 wait=(nil) 11:10:45.508: debug : remoteIO:7483 : We have the buck 66 0xb75df008 0xb75df008 11:10:45.509: debug : remoteIODecodeMessageLength:7032 : Got length, now need 64 total (60 more) 11:10:45.509: debug : remoteIOEventLoop:7347 : Giving up the buck 66 0xb75df008 (nil) 11:10:45.509: debug : remoteIO:7514 : All done with our call 66 (nil) 0xb75df008 11:10:45.509: debug : remoteIO:7421 : Do proc=1 serial=1 length=56 wait=(nil) 11:10:45.509: debug : remoteIO:7483 : We have the buck 1 0x9fb2ea8 0x9fb2ea8 11:10:45.510: debug : remoteIODecodeMessageLength:7032 : Got length, now need 208 total (204 more) 11:10:45.510: debug : remoteIOEventLoop:7347 : Giving up the buck 1 0x9fb2ea8 (nil) 11:10:45.510: debug : remoteIO:7514 : All done with our call 1 (nil) 0x9fb2ea8 11:10:45.510: debug : do_open:1058 : driver 6 remote returned ERROR 11:10:45.510: debug : virUnrefConnect:257 : unref connection 0x9fb0b30 1 11:10:45.510: debug : virReleaseConnect:214 : release connection 0x9fb0b30 error: internal error unable to initialize VirtualBox driver API error: failed to connect to the hypervisor Regards Anuj

1 1

Re: [libvirt] getSchedulerParameters() always returning nullpointerexception in JAVA
by Marc Gonzalez Mateo 15 Dec '09

15 Dec '09

Hi Bryan, I've applied and tested the PATCH and still does not work. Here is the console output: -> [chipiron04, Domain-0, XenTest, XenTest2, chipiron03] Dec 15, 2009 9:49:14 AM net.emotivecloud.virtmonitor.VirtMonitor getCPUPriority SEVERE: Error: getting CPU priority of "XenTest".class java.lang.NullPointerException java.lang.NullPointerException at org.libvirt.SchedParameter.create(Unknown Source) at org.libvirt.Domain.getSchedulerParameters(Unknown Source) at net.emotivecloud.virtmonitor.VirtMonitor.getCPUPriority(VirtMonitor.java:495) at net.emotivecloud.virtmonitor.VirtMonitor.main(VirtMonitor.java:775) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:616) at org.codehaus.mojo.exec.ExecJavaMojo$1.run(ExecJavaMojo.java:283) at java.lang.Thread.run(Thread.java:636) -1 Dec 15, 2009 9:49:14 AM net.emotivecloud.virtmonitor.VirtMonitor getCPUCapacity SEVERE: Error: getting CPU capacity of "XenTest". java.lang.NullPointerException at org.libvirt.SchedParameter.create(Unknown Source) at org.libvirt.Domain.getSchedulerParameters(Unknown Source) at net.emotivecloud.virtmonitor.VirtMonitor.getCPUCapacity(VirtMonitor.java:470) at net.emotivecloud.virtmonitor.VirtMonitor.main(VirtMonitor.java:776) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:616) at org.codehaus.mojo.exec.ExecJavaMojo$1.run(ExecJavaMojo.java:283) at java.lang.Thread.run(Thread.java:636) -1 Both GetCPUCapacity and GetCPUPriority use the schedparameter type. Here is the code and the main. getDomain method just returns the domain object itself and I've also tested it's not null value at this point of the code: public int getCPUCapacity(String name) { int ret = -1; try { Domain d = getDomain(name); SchedParameter[] pars= d.getSchedulerParameters(); for (SchedParameter pri : pars) { if (pri.field=="cap") ret=Integer.parseInt(pri.getValueAsString()); } } catch (Exception e) { log.error(" Error: getting CPU capacity of \""+name+"\"."); e.printStackTrace(); } return ret; } /** * Returns CPU Priority assigned to VM name * @param name * @return */ public int getCPUPriority(String name) { int res=-1; try { Domain d = getDomain(name); SchedParameter[] pars = d.getSchedulerParameters(); for (SchedParameter pri : pars) { if (pri.field=="weight") res=Integer.parseInt(pri.getValueAsString()); } } catch (Exception e) { log.error(" Error: getting CPU priority of \""+name+"\"." + e.getClass()); e.printStackTrace(); } return res; } public static void main(String args[]) { VirtMonitor virt = new VirtMonitor(); System.out.println(" -> " + virt.getDomains()); System.out.println(virt.getCPUPriority("XenTest")); System.out.println(virt.getCPUCapacity("XenTest")); } Thanks in advance, Marc Gonzalez Mateo marcg(a)ac.upc.edu Universitat Politècnica de Catalunya El 14/12/2009, a las 16:41, Bryan Kearney escribió: Can you send me more info on the use case.. and I will try and recreate it? In the mean time.. tell me if the patched jar file at: http://bkearney.fedorapeople.org/libvirt-0.4.0-PATCH.jar works. You will need to download this and copy it into /usr/share/java/libvirt-0.4.0.jar -- bk On 12/14/2009 04:22 AM, Marc Gonzalez Mateo wrote: OK, let's see if Bryan has an idea/solution about this. Thanks guys! MARC El 11/12/2009, a las 14:29, Daniel Veillard escribió: On Fri, Dec 11, 2009 at 02:01:36PM +0100, Marc Gonzalez Mateo wrote: Hi everyvody, I'm developing a new API based on libvirt. I'm currently stucked using getSchedulerParameters, always is returning a nullpointerexception, no matter which Xen Domain I'm passing to the function. Enclosing both the code and the error console: public int getCPUPriority(String name) { int res=-1; try { Domain d = getDomain(name); SchedParameter[] pars = d.getSchedulerParameters(); Hum, it seems Domain.getSchedulerParameters() does SchedParameter[] returnValue = new SchedParameter[0]; and public static SchedParameter create(virSchedParameter vParam) { SchedParameter returnValue = null; switch (vParam.type) { case (1): and the create method getting there gets a null pointer as the initialization argument, which it first dereference ... so not surprizing looking at the code, maybe Bryan has an idea of what is going on there, I'm a bit lost in this initialization process ... Daniel for (SchedParameter pri : pars) { if (pri.field=="weight") res=Integer.parseInt(pri.getValueAsString()); } } catch (LibvirtException e) { log.error(" Error: getting CPU priority of \""+name+"\"." + e.getClass()); e.printStackTrace(); } return res; } Dec 11, 2009 1:04:18 PM net.emotivecloud.virtmonitor.VirtMonitor getCPUCapacity SEVERE: Error: getting CPU capacity of "XenTest". java.lang.NullPointerException at org.libvirt.SchedParameter.create(Unknown Source) at org.libvirt.Domain.getSchedulerParameters(Unknown Source) at net.emotivecloud.virtmonitor.VirtMonitor.getCPUCapacity(VirtMonitor.java:462) at net.emotivecloud.virtmonitor.VirtMonitor.main(VirtMonitor.java:763) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:616) at org.codehaus.mojo.exec.ExecJavaMojo$1.run(ExecJavaMojo.java:283) at java.lang.Thread.run(Thread.java:636) Any ideas? Thanks in advance, Marc Gonzalez Mateo -- Libvir-list mailing list Libvir-list(a)redhat.com https://www.redhat.com/mailman/listinfo/libvir-list -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

1 0

[libvirt] [PATCH v2 00/13] CPU selection patches
by Jiri Denemark 15 Dec '09

15 Dec '09

Hi, This is a second version of CPU selection patchset. Changes in version 2: - virConnectGetHostCPU() API call was completely removed - 'CPU driver implementation' (11/14) patch was dropped - virConnectCompareCPU() API call is implemented directly by hypervisor drivers - new cpuCompareXML() internal function to make virConnectCompareCPU() simpler Jirka Jiri Denemark (13): XML schema for CPU flags XML parsing/formating code for CPU flags Public API Internal driver API Public API implementation Wire protocol format Remote driver Server side dispatcher CPU selection infrastructure CPU map for models and features Set cpuCompare handler in drivers Implement CPU selection in QEMU driver Implement virsh commands configure.in | 20 + daemon/remote.c | 22 + docs/schemas/capability.rng | 46 ++- docs/schemas/domain.rng | 62 +++ include/libvirt/libvirt.h.in | 13 + include/libvirt/virterror.h | 1 + src/Makefile.am | 23 +- src/conf/capabilities.c | 31 ++- src/conf/capabilities.h | 6 + src/conf/cpu_conf.c | 366 ++++++++++++++ src/conf/cpu_conf.h | 116 +++++ src/conf/domain_conf.c | 15 + src/conf/domain_conf.h | 2 + src/cpu/cpu.c | 242 +++++++++ src/cpu/cpu.h | 129 +++++ src/cpu/cpu_generic.c | 123 +++++ src/cpu/cpu_generic.h | 32 ++ src/cpu/cpu_map.c | 130 +++++ src/cpu/cpu_map.h | 41 ++ src/cpu/cpu_map.xml | 304 +++++++++++ src/cpu/cpu_x86.c | 1142 ++++++++++++++++++++++++++++++++++++++++++ src/cpu/cpu_x86.h | 31 ++ src/cpu/cpu_x86_data.h | 45 ++ src/driver.h | 5 + src/esx/esx_driver.c | 1 + src/libvirt.c | 41 ++ src/libvirt_private.syms | 19 + src/libvirt_public.syms | 5 + src/lxc/lxc_driver.c | 1 + src/openvz/openvz_driver.c | 1 + src/qemu/qemu_conf.c | 401 ++++++++++++++- src/qemu/qemu_conf.h | 7 + src/qemu/qemu_driver.c | 43 ++- src/remote/remote_driver.c | 27 + src/remote/remote_protocol.x | 12 +- src/test/test_driver.c | 1 + src/uml/uml_driver.c | 1 + src/util/virterror.c | 3 + src/vbox/vbox_tmpl.c | 1 + tools/virsh.c | 65 +++ 40 files changed, 3541 insertions(+), 35 deletions(-) create mode 100644 src/conf/cpu_conf.c create mode 100644 src/conf/cpu_conf.h create mode 100644 src/cpu/cpu.c create mode 100644 src/cpu/cpu.h create mode 100644 src/cpu/cpu_generic.c create mode 100644 src/cpu/cpu_generic.h create mode 100644 src/cpu/cpu_map.c create mode 100644 src/cpu/cpu_map.h create mode 100644 src/cpu/cpu_map.xml create mode 100644 src/cpu/cpu_x86.c create mode 100644 src/cpu/cpu_x86.h create mode 100644 src/cpu/cpu_x86_data.h

1 13

[libvirt] getSchedulerParameters() always returning nullpointerexception in JAVA
by Marc Gonzalez Mateo 15 Dec '09

15 Dec '09

Hi everyvody, I'm developing a new API based on libvirt. I'm currently stucked using getSchedulerParameters, always is returning a nullpointerexception, no matter which Xen Domain I'm passing to the function. Enclosing both the code and the error console: public int getCPUPriority(String name) { int res=-1; try { Domain d = getDomain(name); SchedParameter[] pars = d.getSchedulerParameters(); for (SchedParameter pri : pars) { if (pri.field=="weight") res=Integer.parseInt(pri.getValueAsString()); } } catch (LibvirtException e) { log.error(" Error: getting CPU priority of \""+name+"\"." + e.getClass()); e.printStackTrace(); } return res; } Dec 11, 2009 1:04:18 PM net.emotivecloud.virtmonitor.VirtMonitor getCPUCapacity SEVERE: Error: getting CPU capacity of "XenTest". java.lang.NullPointerException at org.libvirt.SchedParameter.create(Unknown Source) at org.libvirt.Domain.getSchedulerParameters(Unknown Source) at net.emotivecloud.virtmonitor.VirtMonitor.getCPUCapacity(VirtMonitor.java:462) at net.emotivecloud.virtmonitor.VirtMonitor.main(VirtMonitor.java:763) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:616) at org.codehaus.mojo.exec.ExecJavaMojo$1.run(ExecJavaMojo.java:283) at java.lang.Thread.run(Thread.java:636) Any ideas? Thanks in advance, Marc Gonzalez Mateo

4 4

[libvirt] [PATCH 00/14] CPU selection patches
by Jiri Denemark 15 Dec '09

15 Dec '09

Hi, Here it is, finally. Jirka Jiri Denemark (14): XML schema for CPU flags XML parsing/formating code for CPU flags Public API Internal driver API Public API implementation Wire protocol format Remote driver Server side dispatcher CPU selection infrastructure CPU map for models and features CPU driver implementation Set getHostCPU handler in drivers Implement CPU selection in QEMU driver Implement virsh commands configure.in | 20 + daemon/Makefile.am | 2 + daemon/libvirtd.c | 3 + daemon/remote.c | 42 ++ daemon/remote_dispatch_prototypes.h | 16 + docs/schemas/capability.rng | 46 ++- docs/schemas/domain.rng | 62 ++ include/libvirt/libvirt.h.in | 15 + include/libvirt/virterror.h | 2 + src/Makefile.am | 43 ++- src/conf/capabilities.c | 31 +- src/conf/capabilities.h | 6 + src/conf/cpu_conf.c | 366 +++++++++++ src/conf/cpu_conf.h | 116 ++++ src/conf/domain_conf.c | 15 + src/conf/domain_conf.h | 2 + src/cpu/cpu.c | 206 +++++++ src/cpu/cpu.h | 124 ++++ src/cpu/cpu_driver.c | 190 ++++++ src/cpu/cpu_driver.h | 40 ++ src/cpu/cpu_generic.c | 123 ++++ src/cpu/cpu_generic.h | 32 + src/cpu/cpu_map.c | 130 ++++ src/cpu/cpu_map.h | 41 ++ src/cpu/cpu_map.xml | 304 ++++++++++ src/cpu/cpu_x86.c | 1142 +++++++++++++++++++++++++++++++++++ src/cpu/cpu_x86.h | 31 + src/cpu/cpu_x86_data.h | 45 ++ src/datatypes.h | 2 + src/driver.h | 20 + src/esx/esx_driver.c | 1 + src/libvirt.c | 129 ++++ src/libvirt_private.syms | 19 + src/libvirt_public.syms | 6 + src/lxc/lxc_driver.c | 1 + src/openvz/openvz_driver.c | 1 + src/qemu/qemu_conf.c | 401 ++++++++++++- src/qemu/qemu_conf.h | 7 + src/qemu/qemu_driver.c | 37 +- src/remote/remote_driver.c | 125 ++++ src/remote/remote_protocol.c | 27 + src/remote/remote_protocol.x | 18 +- src/test/test_driver.c | 1 + src/uml/uml_driver.c | 1 + src/util/virterror.c | 8 + src/vbox/vbox_tmpl.c | 1 + tools/virsh.c | 65 ++ 47 files changed, 4029 insertions(+), 36 deletions(-) create mode 100644 src/conf/cpu_conf.c create mode 100644 src/conf/cpu_conf.h create mode 100644 src/cpu/cpu.c create mode 100644 src/cpu/cpu.h create mode 100644 src/cpu/cpu_driver.c create mode 100644 src/cpu/cpu_driver.h create mode 100644 src/cpu/cpu_generic.c create mode 100644 src/cpu/cpu_generic.h create mode 100644 src/cpu/cpu_map.c create mode 100644 src/cpu/cpu_map.h create mode 100644 src/cpu/cpu_map.xml create mode 100644 src/cpu/cpu_x86.c create mode 100644 src/cpu/cpu_x86.h create mode 100644 src/cpu/cpu_x86_data.h

1 15

[libvirt] [PATCH] libvirtd: avoid a NULL dereference on error path
by Jim Meyering 15 Dec '09

15 Dec '09

>From f124b132913884ced6e9fbae8233056e0e97cd0a Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 14 Dec 2009 16:45:11 +0100 Subject: [PATCH] libvirtd: avoid a NULL dereference on error path * daemon/libvirtd.c (qemudDispatchServer): Since "client" may be NULL in the "cleanup:" block, free client->rx only when it's not. --- daemon/libvirtd.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/daemon/libvirtd.c b/daemon/libvirtd.c index 6b7e33d..de6fc27 100644 --- a/daemon/libvirtd.c +++ b/daemon/libvirtd.c @@ -1422,7 +1422,8 @@ static int qemudDispatchServer(struct qemud_server *server, struct qemud_socket if (client && client->tlssession) gnutls_deinit (client->tlssession); close (fd); - VIR_FREE(client->rx); + if (client) + VIR_FREE(client->rx); VIR_FREE(client); return -1; } -- 1.6.6.rc2.275.g51e2d

2 2

[libvirt] libvirt -0.7.4 on fc8 gives "Starting libvirtd daemon: error: Unable to initialize network sockets"
by Nimal Prabudoss I 15 Dec '09

15 Dec '09

Hi, I'm trying to use the libvirt 0.7.4 in fc8. I have configured and done with make, make install. Virsh command also displays the version correctly as 0.7.4 But when I try to start the libvirtd service, I get the following error. "Starting libvirtd daemon: error: Unable to initialize network sockets [Failed]" My machine details is given below "2.6.21-2950.fc8xen #1 SMP Tue Oct 23 12:24:34 EDT 2007 i686 i686 i386 GNU/Linux" Have anybody encountered the same kind of error? Please help me on this. Thanks, Nimal.I

1 0

[libvirt] [PATCH] Fix and improve domain xml video element description
by Matthias Bolte 14 Dec '09

14 Dec '09

The description mismatched the actual structure since the video element was introduced. The nvram attribute is actually called vram. Specify the unit of the vram attribute. --- docs/formatdomain.html.in | 30 ++++++++++++++++++++++-------- 1 files changed, 22 insertions(+), 8 deletions(-) diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index 5b7aa10..3d418b9 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -768,20 +768,34 @@ qemu-kvm -net nic,model=? /dev/null <pre> ... - <video type='vga' nvram='8192' heads='1'> - <acceleration accel3d='yes' accel3d='yes' /> + <video> + <model type='vga' vram='8192' heads='1'> + <acceleration accel3d='yes' accel3d='yes'/> + </model> </video> ...</pre> <dl> <dt><code>video</code></dt> - <dd>The <code>video</code> element has a mandatory <code>type</code> - attribute which takes the value "vga", "cirrus", "vmvga", "xen" or "vbox". - You can also provide the amount of video memory using <code>nvram</code>, - the number of screen with <code>heads</code>, and whether acceleration - should be enabled (if supported) using the <code>accel3d</code> and - <code>accel2d</code> attributes in the <code>acceleration</code> element. + <dd> + The <code>video</code> element is the a container for describing + video devices. + </dd> + + <dt><code>model</code></dt> + <dd> + The <code>model</code> element has a mandatory <code>type</code> + attribute which takes the value "vga", "cirrus", "vmvga", "xen" or "vbox". + You can also provide the amount of video memory in kilobytes using + <code>vram</code> and the number of screen with <code>heads</code>. + </dd> + <dt><code>acceleration</code></dt> + <dd> + If acceleration should be enabled (if supported) using the + <code>accel3d</code> and <code>accel2d</code> attributes in the + <code>acceleration</code> element. + </dd> </dl> <h4><a name="elementsConsole">Consoles, serial, parallel & channel devices</a></h4> -- 1.6.0.4

2 2

[libvirt] Python script and extension module directory handling
by Matthias Bolte 14 Dec '09

14 Dec '09

Commit 66137344feb488ea87b0d92f3c03844d9a7a7786 changed the Python detection mechanism in configure. This results in a changed install location for the Python bindings at least on Fedora 12 64bit systems. For the explanation below I assume a Fedora 12 64bit systems, libvirt is configured with prefix /usr and Python 2.6 is installed. Before this commit libvirt.py and libvirtmod.so were installed to /usr/lib64/python2.6/site-packages. After this commit both are installed to /usr/lib/python2.6/site-packages. This is at least wrong for libvirtmod.so, as it should be installed to /usr/lib64/python2.6/site-packages on a 64bit system. Before this commit pythondir was set manually to /usr/lib64/python2.6/site-packages and python/Makefile.am contains 'python_LTLIBRARIES = libvirtmod.la' that results in installing libvirtmod.so to pythondir. AM_PATH_PYTHON defines two variables: pythondir and pyexecdir. But according to the AM_PATH_PYTHON documentation [1] pythondir is the directory where Python scripts should be installed. /usr/lib/python2.6/site-packages in this case, because Python scripts are not architecture dependent in general. pyexecdir is the directory where Python extension modules (shared libraries) should be installed. AM_PATH_PYTHON detects pyexecdir correctly as /usr/lib64/python2.6/site-packages, but python/Makefile.am contains 'python_LTLIBRARIES = libvirtmod.la' that results in installing libvirtmod.so to pythondir. So python/Makefile.am must be changed: diff --git a/python/Makefile.am b/python/Makefile.am index 736631e..04342b7 100644 --- a/python/Makefile.am +++ b/python/Makefile.am @@ -32,7 +32,7 @@ mylibs = $(top_builddir)/src/libvirt.la all-local: libvirt.py -python_LTLIBRARIES = libvirtmod.la +pyexec_LTLIBRARIES = libvirtmod.la libvirtmod_la_SOURCES = libvirt-override.c typewrappers.c libvirt.c libvirt.h # Python <= 2.4 header files contain a redundant decl, hence we Now libvirtmod.so is installed to the correct directory: /usr/lib64/python2.6/site-packages But libvirt.py gets installed to /usr/lib/python2.6/site-packages, this is different from the installation directory before the commit. A default Fedora 12 64bit installation has *.py files installed to /usr/lib/python2.6/site-packages and /usr/lib64/python2.6/site-packages. By looking at both directories the policy it pretty obvious: Packages that are pure Python go to /usr/lib/python2.6/site-packages and packages that are mixed (Python and shared libraries) go completely to /usr/lib64/python2.6/site-packages. So we restore the behavior from before the commit completely by applying this additional patch: diff --git a/python/Makefile.am b/python/Makefile.am index 736631e..04342b7 100644 --- a/python/Makefile.am +++ b/python/Makefile.am @@ -60,14 +60,14 @@ $(GENERATED): generated.stamp $(libvirtmod_la_OBJECTS): $(GENERATED) install-data-local: - $(mkinstalldirs) $(DESTDIR)$(pythondir) - @INSTALL@ -m 0644 libvirt.py $(DESTDIR)$(pythondir) + $(mkinstalldirs) $(DESTDIR)$(pyexecdir) + @INSTALL@ -m 0644 libvirt.py $(DESTDIR)$(pyexecdir) $(mkinstalldirs) $(DESTDIR)$(DOCS_DIR) @(for doc in $(DOCS) ; \ do @INSTALL@ -m 0644 $$doc $(DESTDIR)$(DOCS_DIR) ; done) uninstall-local: - rm -f $(DESTDIR)$(pythondir)/libvirt.py + rm -f $(DESTDIR)$(pyexecdir)/libvirt.py CLEANFILES= $(GENERATED) generated.stamp I discussed this with Daniel Veillard on IRC and there was some confusion what the actual problem was and how to solve it, so I wanted to recap it here. Also thanks to Daniel Hokka Zakrisson, who gave the important hint about pythondir vs pyexecdir on IRC. [1] http://www.gnu.org/software/hello/manual/automake/Python.html Matthias

3 3

[libvirt] [PATCH] node_device_driver.c: don't write beyond EOB for 4K-byte symlink
by Jim Meyering 14 Dec '09

14 Dec '09

Without this patch, a symlink pointing to a 4096-byte name could make this code write NUL into the byte beyond end of buffer: if ((n = readlink(driver_link, devpath, sizeof devpath - 1)) < 0) { virReportSystemError(conn, errno, _("cannot resolve driver link %s"), driver_link); goto cleanup; } devpath[n] = '\0'; >From a075e207bc8fb279c43c9f4f43a960ffbd9a8a70 Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Mon, 14 Dec 2009 12:05:38 +0100 Subject: [PATCH] node_device_driver.c: don't write beyond EOB for 4K-byte symlink * src/node_device/node_device_driver.c (update_driver_name): Leave one byte for the trailing NUL we'll append. --- src/node_device/node_device_driver.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/node_device/node_device_driver.c b/src/node_device/node_device_driver.c index f083f16..eda5d5e 100644 --- a/src/node_device/node_device_driver.c +++ b/src/node_device/node_device_driver.c @@ -97,7 +97,7 @@ static int update_driver_name(virConnectPtr conn, goto cleanup; } - if ((n = readlink(driver_link, devpath, sizeof devpath)) < 0) { + if ((n = readlink(driver_link, devpath, sizeof devpath - 1)) < 0) { virReportSystemError(conn, errno, _("cannot resolve driver link %s"), driver_link); goto cleanup; -- 1.6.6.rc2.275.g51e2d

3 6

[libvirt] [PATCH] Allow versioned pc machines for QEmu
by Diego Elio Pettenò 14 Dec '09

14 Dec '09

* docs/schemas/domain.rng: add pattern for pc-0.10, pc-0.11 and similar machine types --- docs/schemas/domain.rng | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/docs/schemas/domain.rng b/docs/schemas/domain.rng index d1d3efb..dd3f732 100644 --- a/docs/schemas/domain.rng +++ b/docs/schemas/domain.rng @@ -171,6 +171,9 @@ <value>xenfv</value> <value>pc</value> <value>isapc</value> + <data type="string"> + <param name="pattern">pc-[0-9\.]+</param> + </data> </choice> </attribute> </optional> -- 1.6.5.6

4 3

[libvirt] [PATCH] Fix owner and group in example volume XML
by Matthew Booth 14 Dec '09

14 Dec '09

The owner and group in the documentation examples were confusingly given as '0744'. They should be numeric uid and gid. Changed the examples to use the default uid and gid assigned to qemu in F12. * docs/formatstorage.html.in: Change example owner and group in volume XML --- docs/formatstorage.html.in | 16 ++++++++-------- 1 files changed, 8 insertions(+), 8 deletions(-) diff --git a/docs/formatstorage.html.in b/docs/formatstorage.html.in index ccf5a91..5c41c39 100644 --- a/docs/formatstorage.html.in +++ b/docs/formatstorage.html.in @@ -119,8 +119,8 @@ <target> <path>/dev/disk/by-path</path> <permissions> - <owner>0744</owner> - <group>0744</group> + <owner>107</owner> + <group>107</group> <mode>0744</mode> <label>virt_image_t</label> </permissions> @@ -244,8 +244,8 @@ <path>/var/lib/virt/images/sparse.img</path> <format type='qcow2'/> <permissions> - <owner>0744</owner> - <group>0744</group> + <owner>107</owner> + <group>107</group> <mode>0744</mode> <label>virt_image_t</label> </permissions> @@ -294,8 +294,8 @@ <path>/var/lib/virt/images/master.img</path> <format>raw</format> <permissions> - <owner>0744</owner> - <group>0744</group> + <owner>107</owner> + <group>107</group> <mode>0744</mode> <label>virt_image_t</label> </permissions> @@ -368,8 +368,8 @@ <target> <path>/var/lib/virt/images/sparse.img</path> <permissions> - <owner>0744</owner> - <group>0744</group> + <owner>107</owner> + <group>107</group> <mode>0744</mode> <label>virt_image_t</label> </permissions> -- 1.6.5.2

3 2

[libvirt] [PATCH] add missing doc for device <shareable/> option
by Daniel Veillard 14 Dec '09

14 Dec '09

This has been there forever, but we never documented it in the domain format page. The description is not fantastic, so improvements welcome :-) 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/

2 2

[libvirt] [PATCH] Eliminate failure to delete empty storage pools
by Laine Stump 14 Dec '09

14 Dec '09

virStorageBackendFileSystemDelete was incorrectly calling unlink() in an attempt to remove a directory. It should be calling rmdir() instead. (remove() would also work, but could potentially succeed on a non-empty pool if the pool was a symlink to somewhere else). --- src/storage/storage_backend_fs.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/storage/storage_backend_fs.c b/src/storage/storage_backend_fs.c index 16e4bd9..b7d4bd6 100644 --- a/src/storage/storage_backend_fs.c +++ b/src/storage/storage_backend_fs.c @@ -693,9 +693,9 @@ virStorageBackendFileSystemDelete(virConnectPtr conn, { /* XXX delete all vols first ? */ - if (unlink(pool->def->target.path) < 0) { + if (rmdir(pool->def->target.path) < 0) { virReportSystemError(conn, errno, - _("cannot unlink path '%s'"), + _("failed to remove pool '%s'"), pool->def->target.path); return -1; } -- 1.6.6.rc2.5.g49666

3 2

[libvirt] [PATCH] Fix use of virEventAddHandleImpl()
by Jiri Denemark 14 Dec '09

14 Dec '09

Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/node_device/node_device_udev.c | 10 +++++----- 1 files changed, 5 insertions(+), 5 deletions(-) diff --git a/src/node_device/node_device_udev.c b/src/node_device/node_device_udev.c index cf23be1..6b6e244 100644 --- a/src/node_device/node_device_udev.c +++ b/src/node_device/node_device_udev.c @@ -37,7 +37,7 @@ #include "uuid.h" #include "util.h" #include "buf.h" -#include "daemon/event.h" +#include "event.h" #define VIR_FROM_THIS VIR_FROM_NODEDEV @@ -1526,10 +1526,10 @@ static int udevDeviceMonitorStartup(int privileged ATTRIBUTE_UNUSED) * enumeration. The alternative is to register the callback after * we enumerate, in which case we will fail to create any devices * that appear while the enumeration is taking place. */ - if (virEventAddHandleImpl(udev_monitor_get_fd(udev_monitor), - VIR_EVENT_HANDLE_READABLE, - udevEventHandleCallback, - NULL, NULL) == -1) { + if (virEventAddHandle(udev_monitor_get_fd(udev_monitor), + VIR_EVENT_HANDLE_READABLE, + udevEventHandleCallback, + NULL, NULL) == -1) { ret = -1; goto out; } -- 1.6.5.6

2 1

[libvirt] [PATCH] Cleanup temporary #define after use
by Matthew Booth 14 Dec '09

14 Dec '09

* src/qemu/qemu_driver.c: #undef LOOKUP_PTYS after use --- 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 7e55c23..4567a66 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -1436,6 +1436,7 @@ qemudFindCharDevicePTYsMonitor(virConnectPtr conn, LOOKUP_PTYS(vm->def->serials, vm->def->nserials, "serial"); LOOKUP_PTYS(vm->def->parallels, vm->def->nparallels, "parallel"); LOOKUP_PTYS(vm->def->channels, vm->def->nchannels, "channel"); +#undef LOOKUP_PTYS return 0; } -- 1.6.5.2

2 1

[libvirt] [PATCH] implement --pool option for virsh vol-path
by Dave Allan 14 Dec '09

14 Dec '09

This patch fixes the problem reported in: https://bugzilla.redhat.com/show_bug.cgi?id=509306 The bug reporter says that vol-delete does not support the --pool option, but that's not the case in the current head. This patch makes vol-path behave the same way as vol-delete. Dave

2 1

[libvirt] [PATCH] nodedev: Add removable storage 'media_label' prop
by Cole Robinson 14 Dec '09

14 Dec '09

Provides the CDROM label for current media. Only implemented for the udev backend. Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- docs/schemas/nodedev.rng | 5 +++++ src/conf/node_device_conf.c | 8 ++++++++ src/conf/node_device_conf.h | 1 + src/node_device/node_device_udev.c | 5 +++++ tests/nodedevschemadata/DVD_with_media.xml | 16 ++++++++++++++++ tests/nodedevxml2xmltest.c | 1 + 6 files changed, 36 insertions(+), 0 deletions(-) create mode 100644 tests/nodedevschemadata/DVD_with_media.xml diff --git a/docs/schemas/nodedev.rng b/docs/schemas/nodedev.rng index 7060274..797b1af 100644 --- a/docs/schemas/nodedev.rng +++ b/docs/schemas/nodedev.rng @@ -314,6 +314,11 @@ <element name='media_size'> <ref name='uint'/> </element> + <optional> + <element name='media_label'> + <text/> + </element> + </optional> </element> </define> diff --git a/src/conf/node_device_conf.c b/src/conf/node_device_conf.c index 6003ab1..a0d256c 100644 --- a/src/conf/node_device_conf.c +++ b/src/conf/node_device_conf.c @@ -422,6 +422,11 @@ char *virNodeDeviceDefFormat(virConnectPtr conn, "</media_available>\n", avl ? 1 : 0); virBufferVSprintf(&buf, " <media_size>%llu</media_size>\n", data->storage.removable_media_size); + if (data->storage.media_label) + virBufferVSprintf(&buf, + " <media_label>%s</media_label>\n", + data->storage.media_label); + if (data->storage.logical_block_size > 0) virBufferVSprintf(&buf, " <logical_block_size>%llu" "</logical_block_size>\n", @@ -575,6 +580,8 @@ virNodeDevCapStorageParseXML(virConnectPtr conn, if (virXPathBoolean(conn, "count(./media_available[. = '1']) > 0", ctxt)) data->storage.flags |= VIR_NODE_DEV_CAP_STORAGE_REMOVABLE_MEDIA_AVAILABLE; + data->storage.media_label = virXPathString(conn, "string(./media_label[1])", ctxt); + val = 0; if (virNodeDevCapsDefParseULongLong(conn, "number(./media_size[1])", ctxt, &val, def, _("no removable media size supplied for '%s'"), @@ -1431,6 +1438,7 @@ void virNodeDevCapsDefFree(virNodeDevCapsDefPtr caps) VIR_FREE(data->storage.model); VIR_FREE(data->storage.vendor); VIR_FREE(data->storage.serial); + VIR_FREE(data->storage.media_label); break; case VIR_NODE_DEV_CAP_LAST: /* This case is here to shutup the compiler */ diff --git a/src/conf/node_device_conf.h b/src/conf/node_device_conf.h index 7a20bd6..e409a76 100644 --- a/src/conf/node_device_conf.h +++ b/src/conf/node_device_conf.h @@ -154,6 +154,7 @@ struct _virNodeDevCapsDef { char *model; char *vendor; char *serial; + char *media_label; unsigned flags; /* virNodeDevStorageCapFlags bits */ } storage; } data; diff --git a/src/node_device/node_device_udev.c b/src/node_device/node_device_udev.c index 9b48052..c78643e 100644 --- a/src/node_device/node_device_udev.c +++ b/src/node_device/node_device_udev.c @@ -867,6 +867,11 @@ static int udevProcessCDROM(struct udev_device *device, def->caps->data.storage.flags |= VIR_NODE_DEV_CAP_STORAGE_REMOVABLE_MEDIA_AVAILABLE; + if (udevGetStringProperty(device, "ID_FS_LABEL", + &data->storage.media_label) == PROPERTY_ERROR) { + goto out; + } + if (udevGetUint64SysfsAttr(device, "size", &data->storage.num_blocks) == PROPERTY_ERROR) { diff --git a/tests/nodedevschemadata/DVD_with_media.xml b/tests/nodedevschemadata/DVD_with_media.xml new file mode 100644 index 0000000..673e88f --- /dev/null +++ b/tests/nodedevschemadata/DVD_with_media.xml @@ -0,0 +1,16 @@ +<device> + <name>DVD_GCC_4247N</name> + <parent>pci_8086_27df_scsi_host_scsi_device_lun0</parent> + <capability type='storage'> + <block>/dev/sr0</block> + <bus>scsi</bus> + <drive_type>cdrom</drive_type> + <model>RW/DVD GCC-4247N</model> + <vendor>HL-DT-ST</vendor> + <capability type='removable'> + <media_available>1</media_available> + <media_size>12345678</media_size> + <media_label>Windows_XP_Label</media_label> + </capability> + </capability> +</device> diff --git a/tests/nodedevxml2xmltest.c b/tests/nodedevxml2xmltest.c index 7621212..bf8be7e 100644 --- a/tests/nodedevxml2xmltest.c +++ b/tests/nodedevxml2xmltest.c @@ -80,6 +80,7 @@ mymain(int argc, char **argv) DO_TEST("computer"); DO_TEST("DVD_GCC_4247N"); + DO_TEST("DVD_with_media"); DO_TEST("net_00_13_02_b9_f9_d3"); DO_TEST("net_00_15_58_2f_e9_55"); DO_TEST("pci_1002_71c4"); -- 1.6.5.2

3 2

[libvirt] [PATCH] expose SR IOV physical/virtual function relationships
by Dave Allan 14 Dec '09

14 Dec '09

Attached is a patch that exposes the relationships between physical and virtual functions on SR IOV capable devices. Dave >From cc5b72f99cd472aa0c07d8115e0abc970feab704 Mon Sep 17 00:00:00 2001 From: David Allan <dallan(a)redhat.com> Date: Mon, 30 Nov 2009 15:58:47 -0500 Subject: [PATCH 1/2] Add SR IOV physical and virtual function relationships --- src/conf/node_device_conf.c | 16 ++++ src/conf/node_device_conf.h | 3 + src/node_device/node_device_driver.h | 10 +++ src/node_device/node_device_hal.c | 6 ++ src/node_device/node_device_linux_sysfs.c | 122 ++++++++++++++++++++++++++++- 5 files changed, 156 insertions(+), 1 deletions(-) diff --git a/src/conf/node_device_conf.c b/src/conf/node_device_conf.c index 6003ab1..4b5d17c 100644 --- a/src/conf/node_device_conf.c +++ b/src/conf/node_device_conf.c @@ -246,6 +246,7 @@ char *virNodeDeviceDefFormat(virConnectPtr conn, { virBuffer buf = VIR_BUFFER_INITIALIZER; virNodeDevCapsDefPtr caps; + unsigned int i = 0; char *tmp; virBufferAddLit(&buf, "<device>\n"); @@ -318,6 +319,16 @@ char *virNodeDeviceDefFormat(virConnectPtr conn, data->pci_dev.vendor_name); else virBufferAddLit(&buf, " />\n"); + if (data->pci_dev.physical_function) { + virBufferEscapeString(&buf, " <physical_function>%s</physical_function>\n", + data->pci_dev.physical_function); + } + if (data->pci_dev.num_virtual_functions > 0) { + for (i = 0 ; i < data->pci_dev.num_virtual_functions ; i++) { + virBufferEscapeString(&buf, " <virtual_function>%s</virtual_function>\n", + data->pci_dev.virtual_functions[i]); + } + } break; case VIR_NODE_DEV_CAP_USB_DEV: virBufferVSprintf(&buf, " <bus>%d</bus>\n", data->usb_dev.bus); @@ -1387,6 +1398,7 @@ out: void virNodeDevCapsDefFree(virNodeDevCapsDefPtr caps) { + int i = 0; union _virNodeDevCapData *data = &caps->data; switch (caps->type) { @@ -1402,6 +1414,10 @@ void virNodeDevCapsDefFree(virNodeDevCapsDefPtr caps) case VIR_NODE_DEV_CAP_PCI_DEV: VIR_FREE(data->pci_dev.product_name); VIR_FREE(data->pci_dev.vendor_name); + VIR_FREE(data->pci_dev.physical_function); + for (i = 0 ; i < data->pci_dev.num_virtual_functions ; i++) { + VIR_FREE(data->pci_dev.virtual_functions[i]); + } break; case VIR_NODE_DEV_CAP_USB_DEV: VIR_FREE(data->usb_dev.product_name); diff --git a/src/conf/node_device_conf.h b/src/conf/node_device_conf.h index 7a20bd6..11b7539 100644 --- a/src/conf/node_device_conf.h +++ b/src/conf/node_device_conf.h @@ -105,6 +105,9 @@ struct _virNodeDevCapsDef { unsigned class; char *product_name; char *vendor_name; + char *physical_function; + char **virtual_functions; + unsigned num_virtual_functions; } pci_dev; struct { unsigned bus; diff --git a/src/node_device/node_device_driver.h b/src/node_device/node_device_driver.h index 4f0822c..d358276 100644 --- a/src/node_device/node_device_driver.h +++ b/src/node_device/node_device_driver.h @@ -61,6 +61,14 @@ int check_fc_host_linux(union _virNodeDevCapData *d); #define check_vport_capable(d) check_vport_capable_linux(d) int check_vport_capable_linux(union _virNodeDevCapData *d); +#define get_physical_function(s,d) get_physical_function_linux(s,d) +int get_physical_function_linux(const char *sysfs_path, + union _virNodeDevCapData *d); + +#define get_virtual_functions(s,d) get_virtual_functions_linux(s,d) +int get_virtual_functions_linux(const char *sysfs_path, + union _virNodeDevCapData *d); + #define read_wwn(host, file, wwn) read_wwn_linux(host, file, wwn) int read_wwn_linux(int host, const char *file, char **wwn); @@ -68,6 +76,8 @@ int read_wwn_linux(int host, const char *file, char **wwn); #define check_fc_host(d) #define check_vport_capable(d) +#define get_physical_function(sysfs_path, d) +#define get_virtual_functions(sysfs_path, d) #define read_wwn(host, file, wwn) #endif /* __linux__ */ diff --git a/src/node_device/node_device_hal.c b/src/node_device/node_device_hal.c index 31c1764..7cb931e 100644 --- a/src/node_device/node_device_hal.c +++ b/src/node_device/node_device_hal.c @@ -145,14 +145,20 @@ static int gather_pci_cap(LibHalContext *ctx, const char *udi, (void)virStrToLong_ui(p+1, &p, 16, &d->pci_dev.slot); (void)virStrToLong_ui(p+1, &p, 16, &d->pci_dev.function); } + + get_physical_function(sysfs_path, d); + get_virtual_functions(sysfs_path, d); + VIR_FREE(sysfs_path); } + (void)get_int_prop(ctx, udi, "pci.vendor_id", (int *)&d->pci_dev.vendor); if (get_str_prop(ctx, udi, "pci.vendor", &d->pci_dev.vendor_name) != 0) (void)get_str_prop(ctx, udi, "info.vendor", &d->pci_dev.vendor_name); (void)get_int_prop(ctx, udi, "pci.product_id", (int *)&d->pci_dev.product); if (get_str_prop(ctx, udi, "pci.product", &d->pci_dev.product_name) != 0) (void)get_str_prop(ctx, udi, "info.product", &d->pci_dev.product_name); + return 0; } diff --git a/src/node_device/node_device_linux_sysfs.c b/src/node_device/node_device_linux_sysfs.c index b7cf782..6b9738c 100644 --- a/src/node_device/node_device_linux_sysfs.c +++ b/src/node_device/node_device_linux_sysfs.c @@ -29,6 +29,7 @@ #include "virterror_internal.h" #include "memory.h" #include "logging.h" +#include <dirent.h> #define VIR_FROM_THIS VIR_FROM_NODEDEV @@ -70,7 +71,7 @@ int read_wwn_linux(int host, const char *file, char **wwn) char buf[64]; if (open_wwn_file(LINUX_SYSFS_FC_HOST_PREFIX, host, file, &fd) < 0) { - goto out; + goto out; } memset(buf, 0, sizeof(buf)); @@ -184,4 +185,123 @@ out: return retval; } + +static int get_sriov_function(const char *device_link, + char **pci_device) +{ + char *device_path = NULL, *device_name = NULL; + char errbuf[64]; + int ret = -1; + + VIR_DEBUG("Attempting to resolve device path from device link '%s'\n", + device_link); + + if (!virFileExists(device_link)) { + + VIR_DEBUG("SR IOV function link '%s' does not exist\n", device_link); + /* Not an SR IOV device, not an error, either. */ + ret = 0; + goto out; + + } + + device_path = realpath(device_link, device_path); + if (device_path == NULL) { + memset(errbuf, '\0', sizeof(errbuf)); + VIR_ERROR("Failed to resolve device link '%s': '%s'\n", device_link, + strerror_r(errno, errbuf, sizeof(errbuf))); + goto out; + } + + VIR_DEBUG("SR IOV device path is '%s'\n", device_path); + device_name = basename(device_path); + *pci_device = strdup(device_name); + if (*pci_device == NULL) { + VIR_ERROR0("Failed to allocate memory for PCI device name\n"); + goto out; + } + + VIR_DEBUG("SR IOV function is '%s'\n", *pci_device); + ret = 0; + +out: + VIR_FREE(device_path); + return ret; +} + + +int get_physical_function_linux(const char *sysfs_path, + union _virNodeDevCapData *d ATTRIBUTE_UNUSED) +{ + int ret = -1; + char *device_link = NULL; + + VIR_DEBUG("Attempting to get SR IOV physical function for device " + "with sysfs path '%s'\n", sysfs_path); + + if (virBuildPath(&device_link, sysfs_path, "physfn") == -1) { + virReportOOMError(NULL); + } else { + ret = get_sriov_function(device_link, &d->pci_dev.physical_function); + } + + VIR_FREE(device_link); + return ret; +} + + +int get_virtual_functions_linux(const char *sysfs_path, + union _virNodeDevCapData *d) +{ + int ret = -1; + DIR *dir = NULL; + struct dirent *entry = NULL; + char *device_link = NULL; + + VIR_DEBUG("Attempting to get SR IOV virtual functions for device" + "with sysfs path '%s'\n", sysfs_path); + + dir = opendir(sysfs_path); + if (dir == NULL) { + goto out; + } + + while ((entry = readdir(dir))) { + if (STRPREFIX(entry->d_name, "virtfn")) { + /* This local is just to avoid lines of code much > 80 col. */ + unsigned int *num_funcs = &d->pci_dev.num_virtual_functions; + + if (virBuildPath(&device_link, sysfs_path, entry->d_name) == -1) { + virReportOOMError(NULL); + goto out; + } + + VIR_DEBUG("Number of virtual functions: %d\n", *num_funcs); + if (VIR_REALLOC_N(d->pci_dev.virtual_functions, (*num_funcs) + 1) != 0) { + virReportOOMError(NULL); + goto out; + } + + if (get_sriov_function(device_link, + &d->pci_dev.virtual_functions[*num_funcs]) != 0) { + VIR_ERROR("Failed to get SR IOV function from device link '%s'\n", + device_link); + goto out; + } else { + (*num_funcs)++; + } + + VIR_FREE(device_link); + } + } + + closedir(dir); + + ret = 0; + +out: + VIR_FREE(device_link); + return 0; +} + #endif /* __linux__ */ -- 1.6.5.2

4 10

[libvirt] [PATCH 1/2] add --crash support to "virsh dump"
by Paolo Bonzini 14 Dec '09

14 Dec '09

This patch adds the --crash option (already present in "xm dump-core") to "virsh dump". virDomainCoreDump already has a flags argument, so the API/ABI is untouched. * include/libvirt/libvirt.h.in (virDomainCoreDumpFlags): New. * src/test/test_driver.c (testDomainCoreDump): Do not crash after dump unless VIR_DUMP_CRASH is given. * src/qemu/qemu_driver.c (qemudDomainCoreDump): Shutdown the domain instead of restarting it if --crash is passed. * src/xen/xend_internal.c (xenDaemonDomainCoreDump): Support --crash. * tools/virsh.c (opts_dump): Add --crash. (cmdDump): Map it to flags for virDomainCoreDump and pass them. --- include/libvirt/libvirt.h.in | 5 +++++ src/qemu/qemu_driver.c | 17 ++++++++++++++++- src/test/test_driver.c | 19 ++++++++++--------- src/xen/xend_internal.c | 6 ++++-- tools/virsh.c | 7 ++++++- 5 files changed, 41 insertions(+), 13 deletions(-) diff --git a/include/libvirt/libvirt.h.in b/include/libvirt/libvirt.h.in index 5bc7694..c04b552 100644 --- a/include/libvirt/libvirt.h.in +++ b/include/libvirt/libvirt.h.in @@ -334,6 +334,11 @@ struct _virDomainInterfaceStats { typedef virDomainInterfaceStatsStruct *virDomainInterfaceStatsPtr; +/* Domain core dump flags. */ +typedef enum { + VIR_DUMP_CRASH = (1 << 0), /* crash after dump */ +} virDomainCoreDumpFlags; + /* Domain migration flags. */ typedef enum { VIR_MIGRATE_LIVE = (1 << 0), /* live migration */ diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 92d4629..8e80144 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -3543,6 +3543,7 @@ static int qemudDomainCoreDump(virDomainPtr dom, virDomainObjPtr vm; int resume = 0, paused = 0; int ret = -1, fd = -1; + virDomainEventPtr event = NULL; const char *args[] = { "cat", NULL, @@ -3633,10 +3634,17 @@ static int qemudDomainCoreDump(virDomainPtr dom, goto endjob; endjob: + if ((ret == 0) && (flags & VIR_DUMP_CRASH)) { + qemudShutdownVMDaemon(dom->conn, driver, vm); + event = virDomainEventNewFromObj(vm, + VIR_DOMAIN_EVENT_STOPPED, + VIR_DOMAIN_EVENT_STOPPED_CRASHED); + } + /* Since the monitor is always attached to a pty for libvirt, it will support synchronous operations so we always get here after the migration is complete. */ - if (resume && paused) { + else if (resume && paused) { qemuDomainObjEnterMonitor(vm); if (qemuMonitorStartCPUs(priv->mon, dom->conn) < 0) { if (virGetLastError() == NULL) @@ -3647,12 +3655,19 @@ endjob: } qemuDomainObjEndJob(vm); + if ((ret == 0) && (flags & VIR_DUMP_CRASH) && !vm->persistent) { + virDomainRemoveInactive(&driver->domains, + vm); + vm = NULL; + } cleanup: if (ret != 0) unlink(path); if (vm) virDomainObjUnlock(vm); + if (event) + qemuDomainEventQueue(driver, event); return ret; } diff --git a/src/test/test_driver.c b/src/test/test_driver.c index 35f7571..7db9a4c 100644 --- a/src/test/test_driver.c +++ b/src/test/test_driver.c @@ -1911,15 +1911,16 @@ static int testDomainCoreDump(virDomainPtr domain, goto cleanup; } - testDomainShutdownState(domain, privdom); - event = virDomainEventNewFromObj(privdom, - VIR_DOMAIN_EVENT_STOPPED, - VIR_DOMAIN_EVENT_STOPPED_CRASHED); - - if (!privdom->persistent) { - virDomainRemoveInactive(&privconn->domains, - privdom); - privdom = NULL; + if (flags & VIR_DUMP_CRASH) { + testDomainShutdownState(domain, privdom); + event = virDomainEventNewFromObj(privdom, + VIR_DOMAIN_EVENT_STOPPED, + VIR_DOMAIN_EVENT_STOPPED_CRASHED); + if (!privdom->persistent) { + virDomainRemoveInactive(&privconn->domains, + privdom); + privdom = NULL; + } } ret = 0; diff --git a/src/xen/xend_internal.c b/src/xen/xend_internal.c index 66d2e7f..360390d 100644 --- a/src/xen/xend_internal.c +++ b/src/xen/xend_internal.c @@ -3243,8 +3243,10 @@ xenDaemonDomainCoreDump(virDomainPtr domain, const char *filename, return(-1); } - return xend_op(domain->conn, domain->name, "op", "dump", "file", filename, - "live", "0", "crash", "0", NULL); + return xend_op(domain->conn, domain->name, + "op", "dump", "file", filename, "live", "0", + "crash", (flags & VIR_DUMP_CRASH ? "1" : "0"), + NULL); } /** diff --git a/tools/virsh.c b/tools/virsh.c index 9faac35..65eaa3b 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -1431,6 +1431,7 @@ static const vshCmdInfo info_dump[] = { }; static const vshCmdOptDef opts_dump[] = { + {"crash", VSH_OT_BOOL, 0, gettext_noop("crash the domain after core dump")}, {"domain", VSH_OT_DATA, VSH_OFLAG_REQ, gettext_noop("domain name, id or uuid")}, {"file", VSH_OT_DATA, VSH_OFLAG_REQ, gettext_noop("where to dump the core")}, {NULL, 0, 0, NULL} @@ -1443,6 +1444,7 @@ cmdDump(vshControl *ctl, const vshCmd *cmd) char *name; char *to; int ret = TRUE; + int flags = 0; if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) return FALSE; @@ -1453,7 +1455,10 @@ cmdDump(vshControl *ctl, const vshCmd *cmd) if (!(dom = vshCommandOptDomain(ctl, cmd, &name))) return FALSE; - if (virDomainCoreDump(dom, to, 0) == 0) { + if (vshCommandOptBool (cmd, "crash")) + flags |= VIR_DUMP_CRASH; + + if (virDomainCoreDump(dom, to, flags) == 0) { vshPrint(ctl, _("Domain %s dumped to %s\n"), name, to); } else { vshError(ctl, _("Failed to core dump domain %s to %s"), name, to); -- 1.6.5.2

4 15

[libvirt] [PATCH] add AppArmor test and examples to dist
by Jamie Strandboge 14 Dec '09

14 Dec '09

tests/virt-aa-helper-test and examples/apparmor are not included in official tarballs, but should be. Attached is a patch to fix that which works when apparmor is and is not available. Thanks to Daniel P. Berrange for pointing me in the right direction. Jamie -- Jamie Strandboge | http://www.canonical.com

3 2

[libvirt] [PATCH] fix various breakages in qemu "virsh dump"
by Paolo Bonzini 14 Dec '09

14 Dec '09

1) qemuMigrateToCommand uses ">>" so we have to truncate the file before starting the migration; 2) the command wasn't updated to chown the driver and set/restore the security lavels; 3) the VM does not have to be resumed if migration fails; 4) the file is not removed when migration fails. * src/qemu/qemu_driver.c (qemuDomainCoreDump): Truncate file before dumping, set/restore ownership and security labels for the file. --- src/qemu/qemu_driver.c | 48 +++++++++++++++++++++++++++++++++++++++++++++--- 1 files changed, 45 insertions(+), 3 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index c9b5ac2..92d4629 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -3542,7 +3542,7 @@ static int qemudDomainCoreDump(virDomainPtr dom, struct qemud_driver *driver = dom->conn->privateData; virDomainObjPtr vm; int resume = 0, paused = 0; - int ret = -1; + int ret = -1, fd = -1; const char *args[] = { "cat", NULL, @@ -3569,6 +3569,33 @@ static int qemudDomainCoreDump(virDomainPtr dom, goto endjob; } + /* Create an empty file with appropriate ownership. */ + if ((fd = open(path, O_CREAT|O_TRUNC|O_WRONLY, S_IRUSR|S_IWUSR)) < 0) { + qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_FAILED, + _("failed to create '%s'"), path); + goto endjob; + } + + if (close(fd) < 0) { + virReportSystemError(dom->conn, errno, + _("unable to save file %s"), + path); + goto endjob; + } + + if (driver->privileged && + chown(path, driver->user, driver->group) < 0) { + virReportSystemError(NULL, errno, + _("unable to set ownership of '%s' to user %d:%d"), + path, driver->user, driver->group); + goto endjob; + } + + if (driver->securityDriver && + driver->securityDriver->domainSetSavedStateLabel && + driver->securityDriver->domainSetSavedStateLabel(dom->conn, vm, path) == -1) + goto endjob; + /* Migrate will always stop the VM, so once we support live dumping the resume condition will stay the same, independent of whether the stop command is issued. */ @@ -3590,8 +3617,22 @@ static int qemudDomainCoreDump(virDomainPtr dom, qemuDomainObjEnterMonitor(vm); ret = qemuMonitorMigrateToCommand(priv->mon, 0, args, path); qemuDomainObjExitMonitor(vm); - paused = 1; + paused |= (ret == 0); + + if (driver->privileged && + chown(path, 0, 0) < 0) { + virReportSystemError(NULL, errno, + _("unable to set ownership of '%s' to user %d:%d"), + path, 0, 0); + goto endjob; + } + if (driver->securityDriver && + driver->securityDriver->domainRestoreSavedStateLabel && + driver->securityDriver->domainRestoreSavedStateLabel(dom->conn, path) == -1) + goto endjob; + +endjob: /* Since the monitor is always attached to a pty for libvirt, it will support synchronous operations so we always get here after the migration is complete. */ @@ -3605,10 +3646,11 @@ static int qemudDomainCoreDump(virDomainPtr dom, qemuDomainObjExitMonitor(vm); } -endjob: qemuDomainObjEndJob(vm); cleanup: + if (ret != 0) + unlink(path); if (vm) virDomainObjUnlock(vm); return ret; -- 1.6.5.2

3 2