- Devel - libvirt List Archives

[libvirt] qmp monitor is broken
by Gerd Hoffmann 19 Mar '13

19 Mar '13

Hi, $subject says it. Bisecting points to: commit 23673ca740e0eda66901ca801a5a901df378b063 Author: Anthony Liguori <aliguori(a)us.ibm.com> Date: Tue Mar 5 23:21:23 2013 +0530 qemu-char: add watch support This allows a front-end to request for a callback when the backend is writable again. Signed-off-by: Anthony Liguori <aliguori(a)us.ibm.com> Signed-off-by: Amit Shah <amit.shah(a)redhat.com> Symtoms: "virsh start" hangs (not always, but most of the time). When killing qemu it prints an error message: error: Failed to start domain fedora-org-virtio error: internal error cannot parse json {"return": [{"name": "chardev-remove"}, {"name": "chardev-add"}, {"name": "query-target"}, {"name": "query-cpu-definitions"}, {"name": "query-machines"}, {"name": "device-list-properties"}, {"name": "qom-list-types"}, {"name": "change-vnc-password"}, {"name": "nbd-server-stop"}, {"name": "nbd-server-add"}, {"name": "nbd-server-start"}, {"name": "qom-get"}, {"name": "qom-set"}, {"name": "qom-list"}, {"name": "query-block-jobs"}, {"name": "query-balloon"}, {"name": "query-migrate-capabilities"}, {"name": "migrate-set-capabilities"}, {"name": "query-migrate"}, {"name": "query-uuid"}, {"name": "query-name"}, {"name": "query-spice"}, {"name": "query-vnc"}, {"name": "query-mice"}, {"name": "query-status"}, {"name": "query-kvm"}, {"name": "query-pci"}, {"name": "query-cpus"}, {"name": "query-blockstats"}, {"name": "query-block"}, {"name": "query-chardev"}, {"name": "query-events"}, {"name": "query-commands"}, {"name": "query-version"}, {"name": "human-monitor-command"}, {"name": "qmp_capabilities So it looks like a bulky qmp reply is written incomplete (or corrupted) and libvirt waits forever for the missing bits ... cheers, Gerd

1 0

[libvirt] [ANNOUNCE] libvirt-glib 0.1.6 release
by Daniel P. Berrange 19 Mar '13

19 Mar '13

I am pleased to announce that a new release of the libvirt-glib package, version 0.1.6, is now available from ftp://libvirt.org/libvirt/glib/ The packages are GPG signed with Key fingerprint: DAF3 A6FD B26B 6291 2D0E 8E3F BE86 EBB4 1510 4FDF (4096R) New in this release: - Prefer g_strlcpy over strncpy - Fix potential crashes in some error handling cases - Fix typo in gvir_config_init docs which broke vala binding - Use 'transfer none' for argv parameter in all init functions - Ensure stream data parameters are annotated to be guint8 arrays, not gchar, since they're not NULL terminated strings. - Replace FSF address with URL in copyright headers - Add API for changing Spice compression parameters. libvirt-glib comprises three distinct libraries: - libvirt-glib - Integrate with the GLib event loop and error handling - libvirt-gconfig - Representation of libvirt XML documents as GObjects - libvirt-gobject - Mapping of libvirt APIs into the GObject type system NB: While libvirt aims to be API/ABI stable forever, with libvirt-glib we are not yet guaranteeing that libvirt-glib libraries are API/ABI permanently stable. As of the 0.0.8 release, we have tentatively frozen the API/ABI with the intent of being longterm stable hereafter, but there is still a small chance we might find flaws requiring an API/ABI change. The likelihood of this is low, however, and we will strive to avoid it. Follow up comments about libvirt-glib should be directed to the regular libvir-list(a)redhat.com development list. Thanks to all the people involved in contributing to this release. Regards, Daniel -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|

1 0

[libvirt] [PATCH v8 0/3] DEVICE_DELETED event
by Michael S. Tsirkin 19 Mar '13

19 Mar '13

libvirt has a long-standing bug: when removing the device, it can request removal but does not know when the removal completes. Add an event so we can fix this in a robust way. First patch only adds the event with ID, second patch adds a path field. Split this way for ease of backport (stable downstreams without QOM would want to only take the first patch). Event without fields is still useful as management can use it to poll device list to figure out which device was removed. Signed-off-by: Michael S. Tsirkin <mst(a)redhat.com> If there are no more comments I'll stick this on my pci branch. Changes from v7: - none, v7 was malformed series sent by mistake Changes from v6: - make empty event use data: {}, Markus prefers this Changes from v5: - Emit an empty event on unnamed devices in patch 1/3, as suggested by Markus Changes from v4: - Add extra triggers and extra fields as requested by Markus Changes from v3: - Document that we only emit events for devices with and ID, as suggested by Markus Changes from v2: - move event toward the end of device_unparent, so that parents are reported after their children, as suggested by Paolo Changes from v1: - move to device_unparent - address comments by Andreas and Eric -- Anthony Liguori Michael S. Tsirkin (3): qdev: DEVICE_DELETED event qom: pass original path to unparent method qmp: add path to device_deleted event QMP/qmp-events.txt | 18 ++++++++++++++++++ hw/qdev.c | 15 +++++++++++++-- include/monitor/monitor.h | 1 + include/qom/object.h | 3 ++- monitor.c | 1 + qapi-schema.json | 4 +++- qom/object.c | 4 +++- 7 files changed, 41 insertions(+), 5 deletions(-) -- MST

5 10

[libvirt] [libvirt-sandbox][PATCH] Docs: fix a typo in help documents
by Alex Jia 18 Mar '13

18 Mar '13

Signed-off-by: Alex Jia <ajia(a)redhat.com> --- bin/virt-sandbox.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/bin/virt-sandbox.c b/bin/virt-sandbox.c index bd7ac49..654a205 100644 --- a/bin/virt-sandbox.c +++ b/bin/virt-sandbox.c @@ -87,10 +87,10 @@ int main(int argc, char **argv) { { "name", 'n', 0, G_OPTION_ARG_STRING, &name, N_("name of the sandbox"), "NAME" }, { "mount", 'm', 0, G_OPTION_ARG_STRING_ARRAY, &mounts, - N_("mount a filesystem int he guest"), "TYPE:TARGET=SOURCE" }, + N_("mount a filesystem in the guest"), "TYPE:TARGET=SOURCE" }, { "include", 'i', 0, G_OPTION_ARG_STRING_ARRAY, &includes, N_("file to copy into custom dir"), "GUEST-PATH=HOST-PATH", }, - { "includefile", 'I', 0, G_OPTION_ARG_STRING, &includefile, + { "interfile", 'I', 0, G_OPTION_ARG_STRING, &includefile, N_("file contain list of files to include"), "FILE" }, { "network", 'N', 0, G_OPTION_ARG_STRING_ARRAY, &networks, N_("setup network interface properties"), "PATH", }, -- 1.7.1

3 3

[libvirt] [PATCH] qemu: Fix startupPolicy regression
by Jiri Denemark 18 Mar '13

18 Mar '13

Commit 82d5fe543720da6d83c1d6bfa1c347d7d9fda278 qemu: check backing chains even when cgroup is omitted added backing file checks just before the code that removes optional disks if they are not present. However, the backing chain code fails in case the disk file does not exist, which makes qemuProcessStart fail regardless on configured startupPolicy. Note that startupPolicy implementation is still wrong after this patch since it only check the first file in a possible chain. It should rather check the complete backing chain. But this is an existing limitation that can be solved later. After all, startupPolicy is most useful for CDROM images and they won't make use of backing files in most cases. --- src/qemu/qemu_process.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 1941d4e..3d2b7d6 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -3621,14 +3621,15 @@ int qemuProcessStart(virConnectPtr conn, goto cleanup; VIR_DEBUG("Checking for CDROM and floppy presence"); + if (qemuDomainCheckDiskPresence(driver, vm, + flags & VIR_QEMU_PROCESS_START_COLD) < 0) + goto cleanup; + for (i = 0; i < vm->def->ndisks ; i++) { if (qemuDomainDetermineDiskChain(driver, vm->def->disks[i], false) < 0) goto cleanup; } - if (qemuDomainCheckDiskPresence(driver, vm, - flags & VIR_QEMU_PROCESS_START_COLD) < 0) - goto cleanup; /* Get the advisory nodeset from numad if 'placement' of * either <vcpu> or <numatune> is 'auto'. -- 1.8.1.5

3 3

[libvirt] [libvirt-sandbox][PATCH] Add missing break statements
by Alex Jia 18 Mar '13

18 Mar '13

Signed-off-by: Alex Jia <ajia(a)redhat.com> --- libvirt-sandbox/libvirt-sandbox-init-qemu.c | 4 ++++ 1 files changed, 4 insertions(+), 0 deletions(-) diff --git a/libvirt-sandbox/libvirt-sandbox-init-qemu.c b/libvirt-sandbox/libvirt-sandbox-init-qemu.c index 38ed284..dfa1374 100644 --- a/libvirt-sandbox/libvirt-sandbox-init-qemu.c +++ b/libvirt-sandbox/libvirt-sandbox-init-qemu.c @@ -496,12 +496,16 @@ insmod(const char *filename) switch (errno) { case ENOEXEC: msg = "Invalid module format"; + break; case ENOENT: msg = "Unknown symbol in module"; + break; case ESRCH: msg = "Module has wrong symbol version"; + break; case EINVAL: msg = "Invalid parameters"; + break; default: msg = strerror(errno); } -- 1.7.1

2 2

[libvirt] [libvirt-sandbox][PATCHv2] Docs: fix a typo in help documents
by Alex Jia 18 Mar '13

18 Mar '13

Signed-off-by: Alex Jia <ajia(a)redhat.com> --- bin/virt-sandbox.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/bin/virt-sandbox.c b/bin/virt-sandbox.c index bd7ac49..257d6c2 100644 --- a/bin/virt-sandbox.c +++ b/bin/virt-sandbox.c @@ -87,7 +87,7 @@ int main(int argc, char **argv) { { "name", 'n', 0, G_OPTION_ARG_STRING, &name, N_("name of the sandbox"), "NAME" }, { "mount", 'm', 0, G_OPTION_ARG_STRING_ARRAY, &mounts, - N_("mount a filesystem int he guest"), "TYPE:TARGET=SOURCE" }, + N_("mount a filesystem in the guest"), "TYPE:TARGET=SOURCE" }, { "include", 'i', 0, G_OPTION_ARG_STRING_ARRAY, &includes, N_("file to copy into custom dir"), "GUEST-PATH=HOST-PATH", }, { "includefile", 'I', 0, G_OPTION_ARG_STRING, &includefile, -- 1.7.1

2 2

[libvirt] [PATCH 0/4] Add cpuset cgroup support for LXC
by Gao feng 18 Mar '13

18 Mar '13

This patchset intend to add cpuset cgroup support for LXC. in order to don't create too many redundant codes, this patchset also rename some functions and structure. Gao feng (4): rename qemuGetNumadAdvice to virGetNumadAdvice LXC: allow uses advisory nodeset from querying numad remove the redundant codes LXC: add cpuset cgroup support for lxc po/POTFILES.in | 1 + src/Makefile.am | 1 + src/conf/domain_conf.h | 23 +------ src/libvirt_private.syms | 4 ++ src/lxc/lxc_cgroup.c | 57 +++++++++++++++- src/lxc/lxc_cgroup.h | 2 +- src/lxc/lxc_controller.c | 156 +++++++++++++++--------------------------- src/qemu/qemu_process.c | 154 +---------------------------------------- src/util/virnuma.c | 174 +++++++++++++++++++++++++++++++++++++++++++++++ src/util/virnuma.h | 52 ++++++++++++++ 10 files changed, 348 insertions(+), 276 deletions(-) create mode 100644 src/util/virnuma.c create mode 100644 src/util/virnuma.h -- 1.7.11.7

3 17

[libvirt] [libvirt-sandbox][PATCH v2] Docs: Fix security options wrongs in man page
by Alex Jia 18 Mar '13

18 Mar '13

Signed-off-by: Alex Jia <ajia(a)redhat.com> --- bin/virt-sandbox.c | 17 +++++++++++++---- 1 files changed, 13 insertions(+), 4 deletions(-) diff --git a/bin/virt-sandbox.c b/bin/virt-sandbox.c index 69d1e62..bd7ac49 100644 --- a/bin/virt-sandbox.c +++ b/bin/virt-sandbox.c @@ -372,13 +372,22 @@ separated by commas. The following options are valid for SELinux =over 4 -=item type=TYPE +=item dynamic -The SELinux security type, defaults to sandbox_t +Dynamically allocate an SELinux label, using the default base context. +The default base context is system_u:system_r:svirt_lxc_net_t:s0 for LXC, +system_u:system_r:svirt_t:s0 for KVM, system_u:system_r:svirt_tcg_t:s0 +for QEMU. -=item level=LEVEL +=item dynamic,label=USER:ROLE:TYPE:LEVEL -The SELinux MCS level, defaults to a randomly allocated level +Dynamically allocate an SELinux label, using the base context +USER:ROLE:TYPE:LEVEL, instead of the default base context. + +=item static,label=USER:ROLE:TYPE:LEVEL + +To set a completely static label. For example, +static,label=system_u:system_r:svirt_t:s0:c412,c355 =back -- 1.7.1

2 2

[libvirt] Tails bounty for supporting the 'removable' flag for USB disks
by intrigeri 17 Mar '13

17 Mar '13

Hello, I'm part of the people developing the live system called Tails [1]. [1] https://tails.boum.org/ We are now preparing a bounties program to help developers work on upstream software and fix bugs or implement features needed for Tails. Daniel P. Berrange suggested we write to this list about it. More specifically, for our libvirt-based test suite, we need a way to emulate both removable and non-removable USB 2.0+ devices. This distinction is supported in QEMU since --device can receive an optional removable= property. So, what we lack is a way to tell libvirt to pass this or that value to QEMU for the removable= property of a given emulated USB 2.0+ device. For instance, on a Linux guest it controls the value of /sys/block/$dev/removable. We propose a USD $3000 bounty to implement this feature. We are already in contact with someone who is very interested in doing this work. Ticket: https://bugzilla.redhat.com/show_bug.cgi?id=922495 Thanks for your work on libvirt! Cheers, -- intrigeri | GnuPG key @ https://gaffer.ptitcanardnoir.org/intrigeri/intrigeri.asc | OTR fingerprint @ https://gaffer.ptitcanardnoir.org/intrigeri/otr.asc

1 0

[libvirt] [PATCH v2 00/11] Add support for guests with TPM passthrough device
by Stefan Berger 16 Mar '13

16 Mar '13

Hello! The following set of patches adds support to libvirt for adding a TPM passthrough device to a QEMU guest. Support for this was recently accepted into QEMU. This set of patches borrows a lot from the recently added support for rng's. Regards, Stefan --- v1->v2: - Followed Daniel Berrange's comments (except for the one on de-consolidating the JSON monitor code)

1 11

[libvirt] [PATCH] util: portably check for unchanged uid
by Eric Blake 16 Mar '13

16 Mar '13

We've already scrubbed for comparisons of 'uid_t == -1' (which fail on platforms where uid_t is a u16), but another one snuck in. * src/util/virutil.c (virSetUIDGIDWithCaps): Correct uid comparison. * cfg.mk (sc_prohibit_risky_id_promotion): New rule. --- cfg.mk | 6 ++++++ src/util/virutil.c | 2 +- 2 files changed, 7 insertions(+), 1 deletion(-) diff --git a/cfg.mk b/cfg.mk index b95a90b..394521e 100644 --- a/cfg.mk +++ b/cfg.mk @@ -389,6 +389,12 @@ sc_prohibit_setuid: halt='use virSetUIDGID, not raw set*id' \ $(_sc_search_regexp) +# Don't compare *id_t against raw -1. +sc_prohibit_risky_id_promotion: + @prohibit='\b(user|group|[ug]id) *[=!]= *-' \ + halt='cast -1 to ([ug]id_t) before comparing against id' \ + $(_sc_search_regexp) + # Use snprintf rather than s'printf, even if buffer is provably large enough, # since gnulib has more guarantees for snprintf portability sc_prohibit_sprintf: diff --git a/src/util/virutil.c b/src/util/virutil.c index a0d1530..42b4295 100644 --- a/src/util/virutil.c +++ b/src/util/virutil.c @@ -3011,7 +3011,7 @@ virSetUIDGIDWithCaps(uid_t uid, gid_t gid, unsigned long long capBits, * change the capabilities bounding set. */ - if (clearExistingCaps || (uid != -1 && uid != 0)) + if (clearExistingCaps || (uid != (uid_t)-1 && uid != 0)) capng_clear(CAPNG_SELECT_BOTH); for (ii = 0; ii <= CAP_LAST_CAP; ii++) { -- 1.8.1.4

2 2

[libvirt] can this vm-xml-interface works,thanks
by yue 16 Mar '13

16 Mar '13

hi,all,please look at interface . actually ,virbr0 is NAT, my program product a vm-xml which is not standard. it use the way to product bridge interface to product a NAT interface. but i want to know if this can work rightly. <interface type='bridge'> # "type='network'" <mac address='00:1a:4a:a8:7a:09'/> <source bridge='virbr0'/># network='virbr0' <target dev='vnet0'/> <model type='virtio'/> <boot order='3'/> <alias name='net0'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </interface> thanks

2 2

[libvirt] [PATCHv3] audit: Audit resources used by VirtIO RNG
by Peter Krempa 16 Mar '13

16 Mar '13

This patch adds auditing of resources used by Virtio RNG devices. Only resources on the local filesystems are audited. The audit logs look like: For the 'random' backend: type=VIRT_RESOURCE msg=audit(1363099126.643:31): pid=995252 uid=0 auid=4294967295 ses=4294967295 msg='virt=kvm resrc=rng reason=start vm="qcow-test" uuid=118733ed-b658-3e22-a2cb-4fe5cb3ddf79 old-rng="?" new-rng="/dev/random": exe="/home/pipo/libvirt/daemon/.libs/libvirtd" hostname=? addr=? terminal=pts/0 res=success' For local character device source: type=VIRT_RESOURCE msg=audit(1363100164.240:96): pid=995252 uid=0 auid=4294967295 ses=4294967295 msg='virt=kvm resrc=rng reason=start vm="qcow-test" uuid=118733ed-b658-3e22-a2cb-4fe5cb3ddf79 old-rng="?" new-rng="/tmp/unix.sock": exe="/home/pipo/libvirt/daemon/.libs/libvirtd" hostname=? addr=? terminal=pts/0 res=success' --- Notes: Version 3: - don't log non-local resources for EGD backend - change order of blocks of code to optimize Version 2: - log also EGD backends - add example of audit message to commit message src/conf/domain_audit.c | 120 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 120 insertions(+) diff --git a/src/conf/domain_audit.c b/src/conf/domain_audit.c index 8cd522a..a776058 100644 --- a/src/conf/domain_audit.c +++ b/src/conf/domain_audit.c @@ -57,6 +57,37 @@ virDomainAuditGetRdev(const char *path ATTRIBUTE_UNUSED) } #endif + +static const char * +virDomainAuditChardevPath(virDomainChrSourceDefPtr chr) +{ + if (!chr) + return NULL; + + switch ((enum virDomainChrType) chr->type) { + case VIR_DOMAIN_CHR_TYPE_PTY: + case VIR_DOMAIN_CHR_TYPE_DEV: + case VIR_DOMAIN_CHR_TYPE_FILE: + case VIR_DOMAIN_CHR_TYPE_PIPE: + return chr->data.file.path; + + case VIR_DOMAIN_CHR_TYPE_UNIX: + return chr->data.nix.path; + + case VIR_DOMAIN_CHR_TYPE_TCP: + case VIR_DOMAIN_CHR_TYPE_UDP: + case VIR_DOMAIN_CHR_TYPE_NULL: + case VIR_DOMAIN_CHR_TYPE_VC: + case VIR_DOMAIN_CHR_TYPE_STDIO: + case VIR_DOMAIN_CHR_TYPE_SPICEVMC: + case VIR_DOMAIN_CHR_TYPE_LAST: + return NULL; + } + + return NULL; +} + + void virDomainAuditDisk(virDomainObjPtr vm, const char *oldDef, const char *newDef, @@ -100,6 +131,92 @@ cleanup: } +static void +virDomainAuditRNG(virDomainObjPtr vm, + virDomainRNGDefPtr newDef, virDomainRNGDefPtr oldDef, + const char *reason, bool success) +{ + char uuidstr[VIR_UUID_STRING_BUFLEN]; + char *vmname; + const char *newsrcpath = NULL; + const char *oldsrcpath = NULL; + char *oldsrc = NULL; + char *newsrc = NULL; + const char *virt; + + if (newDef) { + switch ((enum virDomainRNGBackend) newDef->backend) { + case VIR_DOMAIN_RNG_BACKEND_RANDOM: + if (newDef->source.file) + newsrcpath = newDef->source.file; + else + newsrcpath = "/dev/random"; + break; + + case VIR_DOMAIN_RNG_BACKEND_EGD: + newsrcpath = virDomainAuditChardevPath(newDef->source.chardev); + break; + + case VIR_DOMAIN_RNG_BACKEND_LAST: + break; + } + } + + if (oldDef) { + switch ((enum virDomainRNGBackend) oldDef->backend) { + case VIR_DOMAIN_RNG_BACKEND_RANDOM: + if (oldDef->source.file) + oldsrcpath = oldDef->source.file; + else + oldsrcpath = "/dev/random"; + break; + + case VIR_DOMAIN_RNG_BACKEND_EGD: + oldsrcpath = virDomainAuditChardevPath(oldDef->source.chardev); + break; + + case VIR_DOMAIN_RNG_BACKEND_LAST: + break; + } + } + + /* don't audit the RNG device if it doesn't use local resources */ + if (!oldsrcpath && !newsrcpath) + return; + + virUUIDFormat(vm->def->uuid, uuidstr); + if (!(vmname = virAuditEncode("vm", vm->def->name))) + goto no_memory; + + if (!(virt = virDomainVirtTypeToString(vm->def->virtType))) { + VIR_WARN("Unexpected virt type %d while encoding audit message", + vm->def->virtType); + virt = "?"; + } + + if (!(newsrc = virAuditEncode("new-rng", VIR_AUDIT_STR(newsrcpath)))) + goto no_memory; + + if (!(oldsrc = virAuditEncode("old-rng", VIR_AUDIT_STR(oldsrcpath)))) + goto no_memory; + + VIR_AUDIT(VIR_AUDIT_RECORD_RESOURCE, success, + "virt=%s resrc=rng reason=%s %s uuid=%s %s %s", + virt, reason, vmname, uuidstr, + oldsrc, newsrc); + +cleanup: + VIR_FREE(vmname); + VIR_FREE(oldsrc); + VIR_FREE(newsrc); + return; + +no_memory: + VIR_WARN("OOM while encoding audit message"); + goto cleanup; +} + + void virDomainAuditFS(virDomainObjPtr vm, virDomainFSDefPtr oldDef, virDomainFSDefPtr newDef, @@ -641,6 +758,9 @@ virDomainAuditStart(virDomainObjPtr vm, const char *reason, bool success) virDomainAuditRedirdev(vm, redirdev, "start", true); } + if (vm->def->rng) + virDomainAuditRNG(vm, vm->def->rng, NULL, "start", true); + virDomainAuditMemory(vm, 0, vm->def->mem.cur_balloon, "start", true); virDomainAuditVcpu(vm, 0, vm->def->vcpus, "start", true); -- 1.8.1.5

3 4

[libvirt] [PATCH 0/3] Fix virsh race and coredump
by Viktor Mihajlovski 15 Mar '13

15 Mar '13

I hope that this is an acceptable solution. The race is caused by referencing a disposed connection object in a callback. In the first patch we make sure that the object reference count is reflecting the callback registration. The second patch prevents the invocation of a NULL callback. The last one tries to ensure that we don't leak connection references. Viktor Mihajlovski (3): libvirt: Increase connection reference count for callbacks remote: Don't call NULL closeFreeCallback virsh: Unregister the connection close notifier upon termination src/libvirt.c | 5 +++++ src/remote/remote_driver.c | 3 ++- tools/virsh.c | 23 +++++++++++++++++------ 3 files changed, 24 insertions(+), 7 deletions(-) -- 1.7.9.5

1 3

[libvirt] [PATCH V1 0/6] Add support for guests with TPM passthrough device
by Stefan Berger 15 Mar '13

15 Mar '13

Hello! The following set of patches adds support to libvirt for adding a TPM passthrough device to a QEMU guest. Support for this was recently accepted into QEMU. This set of patches borrows a lot from the recently added support for rng's. This patch series builds on top of the patches for adding support for file descriptor sets. Regards, Stefan

2 21

[libvirt] [PATCH v4 0/3] Keep original file label
by Michal Privoznik 15 Mar '13

15 Mar '13

Yet another rework of $subj. I am still not solving atomicity problem for now. See diff to the patches if you want to know what's changed. Patch 1/3 has been already ACKed, however, I've changed it slightly. Michal Privoznik (3): virFile: Add APIs for extended attributes handling virfile: Introduce internal API for managing ACL security_dac: Favour ACLs over chown() configure.ac | 2 + libvirt.spec.in | 1 + m4/virt-acl.m4 | 9 ++ src/Makefile.am | 4 +- src/libvirt_private.syms | 6 + src/security/security_dac.c | 302 +++++++++++++++++++++++++++++++++++++++----- src/util/virfile.c | 301 +++++++++++++++++++++++++++++++++++++++++++ src/util/virfile.h | 28 ++++ 8 files changed, 622 insertions(+), 31 deletions(-) create mode 100644 m4/virt-acl.m4 -- 1.8.1.5

2 6

[libvirt] [libvirt-sandbox][PATCH] Docs: Fix security options wrongs in man page
by Alex Jia 15 Mar '13

15 Mar '13

Signed-off-by: Alex Jia <ajia(a)redhat.com> --- bin/virt-sandbox.c | 14 ++++++++++---- 1 files changed, 10 insertions(+), 4 deletions(-) diff --git a/bin/virt-sandbox.c b/bin/virt-sandbox.c index 69d1e62..66676c3 100644 --- a/bin/virt-sandbox.c +++ b/bin/virt-sandbox.c @@ -372,13 +372,19 @@ separated by commas. The following options are valid for SELinux =over 4 -=item type=TYPE +=item dynamic -The SELinux security type, defaults to sandbox_t +The SELinux security dynamic for the sandbox, defaults to svirt_sandbox_t -=item level=LEVEL +=item dynamic,label=USER:ROLE:TYPE:LEVEL -The SELinux MCS level, defaults to a randomly allocated level +To use dynamic and override the base label, defaults to +system_u:system_r:svirt_t:s0 + +=item static,label=USER:ROLE:TYPE:LEVEL + +To set a completely static label. For example, +static,label=system_u:system_r:svirt_t:s0:c412,c355 =back -- 1.7.1

2 2

[libvirt] [PATCH] remote: Prevent race when closing a connection
by Viktor Mihajlovski 15 Mar '13

15 Mar '13

A race condition can occur when virConnectClose is called parallel to the execution of the connection close callback in remoteClientCloseFunc. The race happens if the connection object is destroyed (including the mutex) while remoteClientCloseFunc is waiting for the connection mutex. After the destruction of the (non error checking) mutex remoteClientCloseFunc starts to process the callbacks. However the operations can occur against a freed (or even worse, reallocated) object. Another issue is that the closeFreeCallback is invoked even if it's NULL (this is the case for virsh). The solution is to clean out the callback pointers in virConnectDispose before destroying the mutex. This way remoteClientCloseFunc will return immediately after passing virMutexLock, thus avoiding potential data corruption. There's still the slight chance that the concluding virMutexUnlock could do harm on the freed connection object. This could be fixed using an error checking mutex which however has a much broader scope and impact. Signed-off-by: Viktor Mihajlovski <mihajlov(a)linux.vnet.ibm.com> --- src/datatypes.c | 8 +++++++- src/remote/remote_driver.c | 3 ++- 2 files changed, 9 insertions(+), 2 deletions(-) diff --git a/src/datatypes.c b/src/datatypes.c index b04e100..2358bdf 100644 --- a/src/datatypes.c +++ b/src/datatypes.c @@ -146,8 +146,14 @@ virConnectDispose(void *obj) virMutexLock(&conn->lock); - if (conn->closeFreeCallback) + if (conn->closeCallback) + conn->closeCallback = NULL; + + if (conn->closeFreeCallback) { conn->closeFreeCallback(conn->closeOpaque); + conn->closeFreeCallback = NULL; + conn->closeOpaque = NULL; + } virResetError(&conn->err); diff --git a/src/remote/remote_driver.c b/src/remote/remote_driver.c index 3721af9..885120e 100644 --- a/src/remote/remote_driver.c +++ b/src/remote/remote_driver.c @@ -358,7 +358,8 @@ static void remoteClientCloseFunc(virNetClientPtr client ATTRIBUTE_UNUSED, closeCallback(conn, reason, closeOpaque); virMutexLock(&conn->lock); conn->closeDispatch = false; - if (conn->closeUnregisterCount != closeUnregisterCount) + if (conn->closeUnregisterCount != closeUnregisterCount && + closeFreeCallback) closeFreeCallback(closeOpaque); } virMutexUnlock(&conn->lock); -- 1.7.9.5

3 5

[libvirt] [PATCHv2 1/1] Remove contiguous CPU indexes assumption
by Li Zhang 15 Mar '13

15 Mar '13

From: Li Zhang <zhlcindy(a)linux.vnet.ibm.com> When getting CPUs' information, it assumes that CPU indexes are not contiguous. But for ppc64 platform, CPU indexes are not contiguous because SMT is needed to be disabled, so CPU information is not right on ppc64 and vpuinfo, vcpupin can't work corretly. This patch is to remove the assumption to be compatible with ppc64. Test: 4 vcpus are assigned to one VM and execute vcpuinfo command. Without patch: There is only one vcpu informaion can be listed. With patch: All vcpus' information can be listed correctly. Signed-off-by: Li Zhang <zhlcindy(a)linux.vnet.ibm.com> --- v2 -> v1: * Remove unused cpu variable suggested by Daniel P.Berrange * Remove contigous assumption in qemu_monitor_text.c suggested by Daniel P.Berrange src/qemu/qemu_monitor_json.c | 14 -------------- src/qemu/qemu_monitor_text.c | 13 ++----------- 2 files changed, 2 insertions(+), 25 deletions(-) diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index 9991a0a..1bf8baf 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -1209,7 +1209,6 @@ qemuMonitorJSONExtractCPUInfo(virJSONValuePtr reply, for (i = 0 ; i < ncpus ; i++) { virJSONValuePtr entry = virJSONValueArrayGet(data, i); - int cpu; int thread; if (!entry) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", @@ -1217,12 +1216,6 @@ qemuMonitorJSONExtractCPUInfo(virJSONValuePtr reply, goto cleanup; } - if (virJSONValueObjectGetNumberInt(entry, "CPU", &cpu) < 0) { - virReportError(VIR_ERR_INTERNAL_ERROR, "%s", - _("cpu information was missing cpu number")); - goto cleanup; - } - if (virJSONValueObjectGetNumberInt(entry, "thread_id", &thread) < 0) { /* Only qemu-kvm tree includs thread_id, so treat this as non-fatal, simply returning no data */ @@ -1230,13 +1223,6 @@ qemuMonitorJSONExtractCPUInfo(virJSONValuePtr reply, goto cleanup; } - if (cpu != i) { - virReportError(VIR_ERR_INTERNAL_ERROR, - _("unexpected cpu index %d expecting %d"), - i, cpu); - goto cleanup; - } - threads[i] = thread; } diff --git a/src/qemu/qemu_monitor_text.c b/src/qemu/qemu_monitor_text.c index 58f6323..1b6efba 100644 --- a/src/qemu/qemu_monitor_text.c +++ b/src/qemu/qemu_monitor_text.c @@ -510,7 +510,6 @@ int qemuMonitorTextGetCPUInfo(qemuMonitorPtr mon, { char *qemucpus = NULL; char *line; - int lastVcpu = -1; pid_t *cpupids = NULL; size_t ncpupids = 0; @@ -530,16 +529,12 @@ int qemuMonitorTextGetCPUInfo(qemuMonitorPtr mon, do { char *offset = strchr(line, '#'); char *end = NULL; - int vcpu = 0, tid = 0; + int tid = 0; /* See if we're all done */ if (offset == NULL) break; - /* Extract VCPU number */ - if (virStrToLong_i(offset + 1, &end, 10, &vcpu) < 0) - goto error; - if (end == NULL || *end != ':') goto error; @@ -552,15 +547,11 @@ int qemuMonitorTextGetCPUInfo(qemuMonitorPtr mon, if (end == NULL || !c_isspace(*end)) goto error; - if (vcpu != (lastVcpu + 1)) - goto error; - if (VIR_REALLOC_N(cpupids, ncpupids+1) < 0) goto error; - VIR_DEBUG("vcpu=%d pid=%d", vcpu, tid); + VIR_DEBUG("pid=%d", tid); cpupids[ncpupids++] = tid; - lastVcpu = vcpu; /* Skip to next data line */ line = strchr(offset, '\r'); -- 1.7.10.1

3 2

[libvirt] [libvirt-sandbox][PATCH v2] Avoid infinite loop in gvir_sandbox_config_set_security_opts()
by Alex Jia 15 Mar '13

15 Mar '13

Signed-off-by: Alex Jia <ajia(a)redhat.com> --- libvirt-sandbox/libvirt-sandbox-config.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/libvirt-sandbox/libvirt-sandbox-config.c b/libvirt-sandbox/libvirt-sandbox-config.c index 90217f1..08d21b9 100644 --- a/libvirt-sandbox/libvirt-sandbox-config.c +++ b/libvirt-sandbox/libvirt-sandbox-config.c @@ -1365,6 +1365,7 @@ gboolean gvir_sandbox_config_set_security_opts(GVirSandboxConfig *config, "Unknown security option '%s'", name); return FALSE; } + i++; } return TRUE; } -- 1.7.1

2 2

[libvirt] [libvirt-sandbox][PATCH] Avoid infinite loop in gvir_sandbox_config_set_security_opts()
by Alex Jia 15 Mar '13

15 Mar '13

Signed-off-by: Alex Jia <ajia(a)redhat.com> --- libvirt-sandbox/libvirt-sandbox-config.c | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/libvirt-sandbox/libvirt-sandbox-config.c b/libvirt-sandbox/libvirt-sandbox-config.c index 90217f1..f96d417 100644 --- a/libvirt-sandbox/libvirt-sandbox-config.c +++ b/libvirt-sandbox/libvirt-sandbox-config.c @@ -1356,10 +1356,13 @@ gboolean gvir_sandbox_config_set_security_opts(GVirSandboxConfig *config, if (strncmp(name, "label=", 5) == 0) { gvir_sandbox_config_set_security_label(config, value); + break; } else if (g_str_equal(name, "dynamic")) { gvir_sandbox_config_set_security_dynamic(config, TRUE); + break; } else if (g_str_equal(name, "static")) { gvir_sandbox_config_set_security_dynamic(config, FALSE); + break; } else { g_set_error(error, GVIR_SANDBOX_CONFIG_ERROR, 0, "Unknown security option '%s'", name); -- 1.7.1

2 2

[libvirt] [libvirt-sandbox][PATCH] Docs: fix typos in IPv6 address
by Alex Jia 15 Mar '13

15 Mar '13

Signed-off-by: Alex Jia <ajia(a)redhat.com> --- bin/virt-sandbox.c | 2 +- libvirt-sandbox/libvirt-sandbox-config.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/bin/virt-sandbox.c b/bin/virt-sandbox.c index 9becd07..69d1e62 100644 --- a/bin/virt-sandbox.c +++ b/bin/virt-sandbox.c @@ -352,7 +352,7 @@ specifies the broadcast address. Some examples address=192.168.122.1/24 address=192.168.122.1/24%192.168.122.255 - address=2001:212::204.2/64 + address=2001:212::204:2/64 =item route=IP-NETWORK/PREFIX%GATEWAY diff --git a/libvirt-sandbox/libvirt-sandbox-config.c b/libvirt-sandbox/libvirt-sandbox-config.c index 3dc453a..90217f1 100644 --- a/libvirt-sandbox/libvirt-sandbox-config.c +++ b/libvirt-sandbox/libvirt-sandbox-config.c @@ -825,7 +825,7 @@ GList *gvir_sandbox_config_get_networks(GVirSandboxConfig *config) * * dhcp * address=192.168.122.1/24%192.168.122.255; - * address=192.168.122.1/24%192.168.122.255;address=2001:212::204.2/64 + * address=192.168.122.1/24%192.168.122.255;address=2001:212::204:2/64 * route=192.168.122.255/24%192.168.1.1 */ gboolean gvir_sandbox_config_add_network_strv(GVirSandboxConfig *config, -- 1.7.1

2 2

[libvirt] [[PATCHv2]] Add support for qxl.revision in domain XML
by Christophe Fergeau 15 Mar '13

15 Mar '13

QXL devices have an associated 'revision' which is raised when new features have been introduced which would break migration to older versions. This commit makes it possible to set this revision as QEMU sometimes support newer QXL revisions than what it defaults to. --- docs/formatdomain.html.in | 4 +++- docs/schemas/domaincommon.rng | 5 +++++ src/conf/domain_conf.c | 20 ++++++++++++++++++++ src/conf/domain_conf.h | 1 + src/qemu/qemu_command.c | 8 ++++++++ .../qemuxml2argv-graphics-spice-compression.args | 3 ++- .../qemuxml2argv-graphics-spice-compression.xml | 4 ++-- .../qemuxml2argv-graphics-spice-qxl-vga.args | 3 ++- .../qemuxml2argv-graphics-spice-qxl-vga.xml | 4 ++-- .../qemuxml2argv-graphics-spice.args | 3 ++- .../qemuxml2argvdata/qemuxml2argv-graphics-spice.xml | 4 ++-- 11 files changed, 49 insertions(+), 10 deletions(-) diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index a9003d7..bcdd90f 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -3584,7 +3584,9 @@ qemu-kvm -net nic,model=? /dev/null 1.0.2) is allowed for "qxl" type only and specifies the size of the primary bar, while <code>vram</code> specifies the secondary bar size. If "ram" or "vram" are not supplied a default - value is used. + value is used. The optional attribute <code>revision</code> (since 1.0.3) specifies the revision of + the QXL device, newer revisions provide more functionality. </dd> <dt><code>model</code></dt> diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index 63be4aa..57128ab 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -2283,6 +2283,11 @@ <ref name="unsignedInt"/> </attribute> </optional> + <optional> + <attribute name="revision"> + <ref name="unsignedInt"/> + </attribute> + </optional> </group> </choice> <optional> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 0c75838..f4f273c 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -7646,6 +7646,7 @@ virDomainVideoDefParseXML(const xmlNodePtr node, char *vram = NULL; char *ram = NULL; char *primary = NULL; + char *revision = NULL; if (VIR_ALLOC(def) < 0) { virReportOOMError(); @@ -7661,6 +7662,7 @@ virDomainVideoDefParseXML(const xmlNodePtr node, ram = virXMLPropString(cur, "ram"); vram = virXMLPropString(cur, "vram"); heads = virXMLPropString(cur, "heads"); + revision = virXMLPropString(cur, "revision"); if ((primary = virXMLPropString(cur, "primary")) != NULL) { if (STREQ(primary, "yes")) @@ -7713,6 +7715,19 @@ virDomainVideoDefParseXML(const xmlNodePtr node, def->vram = virDomainVideoDefaultRAM(dom, def->type); } + if (revision) { + if (def->type != VIR_DOMAIN_VIDEO_TYPE_QXL) { + virReportError(VIR_ERR_XML_ERROR, "%s", + _("revision attribute only supported for type of qxl")); + goto error; + } + if (virStrToLong_ui(revision, NULL, 10, &def->revision) < 0) { + virReportError(VIR_ERR_XML_ERROR, + _("cannot parse video revision '%s'"), revision); + goto error; + } + } + if (heads) { if (virStrToLong_ui(heads, NULL, 10, &def->heads) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, @@ -7730,6 +7745,7 @@ virDomainVideoDefParseXML(const xmlNodePtr node, VIR_FREE(ram); VIR_FREE(vram); VIR_FREE(heads); + VIR_FREE(revision); return def; @@ -7739,6 +7755,8 @@ error: VIR_FREE(ram); VIR_FREE(vram); VIR_FREE(heads); + VIR_FREE(revision); + return NULL; } @@ -13632,6 +13650,8 @@ virDomainVideoDefFormat(virBufferPtr buf, virBufferAsprintf(buf, " heads='%u'", def->heads); if (def->primary) virBufferAddLit(buf, " primary='yes'"); + if (def->revision) + virBufferAsprintf(buf, " revision='%u'", def->revision); if (def->accel) { virBufferAddLit(buf, ">\n"); virDomainVideoAccelDefFormat(buf, def->accel); diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index 4ffa4aa..1d7951b 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -1174,6 +1174,7 @@ struct _virDomainVideoDef { unsigned int ram; /* kibibytes (multiples of 1024) */ unsigned int vram; /* kibibytes (multiples of 1024) */ unsigned int heads; + unsigned int revision; bool primary; virDomainVideoAccelDefPtr accel; virDomainDeviceInfo info; diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index dee493f..5164bd5 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -3567,6 +3567,9 @@ qemuBuildDeviceVideoStr(virDomainVideoDefPtr video, /* QEMU accepts bytes for vram_size. */ virBufferAsprintf(&buf, ",vram_size=%u", video->vram * 1024); + + if (video->revision != 0) + virBufferAsprintf(&buf, ",revision=%u", video->revision); } if (qemuBuildDeviceAddressStr(&buf, &video->info, qemuCaps) < 0) @@ -6612,6 +6615,11 @@ qemuBuildCommandLine(virConnectPtr conn, virCommandAddArgFormat(cmd, "%s.vram_size=%u", dev, vram * 1024); } + if (def->videos[0]->revision) { + virCommandAddArg(cmd, "-global"); + virCommandAddArgFormat(cmd, "%s.revision=%u", + dev, def->videos[0]->revision); + } } } diff --git a/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-compression.args b/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-compression.args index 59f064b..05f5579 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-compression.args +++ b/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-compression.args @@ -7,5 +7,6 @@ image-compression=auto_glz,jpeg-wan-compression=auto,\ zlib-glz-wan-compression=auto,\ playback-compression=on,streaming-video=filter -vga \ qxl -global qxl.ram_size=67108864 -global qxl.vram_size=18874368 \ --device qxl,id=video1,ram_size=67108864,vram_size=33554432,bus=pci.0,addr=0x4 \ +-global qxl.revision=4 \ +-device qxl,id=video1,ram_size=67108864,vram_size=33554432,revision=4,bus=pci.0,addr=0x4 \ -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x3 diff --git a/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-compression.xml b/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-compression.xml index a8c4ad8..2dc5776 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-compression.xml +++ b/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-compression.xml @@ -31,10 +31,10 @@ <streaming mode='filter'/> </graphics> <video> - <model type='qxl' ram='65536' vram='18432' heads='1'/> + <model type='qxl' ram='65536' vram='18432' heads='1' revision='4'/> </video> <video> - <model type='qxl' ram='65536' vram='32768' heads='1'/> + <model type='qxl' ram='65536' vram='32768' heads='1' revision='4'/> </video> <memballoon model='virtio'/> </devices> diff --git a/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-qxl-vga.args b/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-qxl-vga.args index ef499e6..0b08038 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-qxl-vga.args +++ b/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-qxl-vga.args @@ -4,5 +4,6 @@ unix:/tmp/test-monitor,server,nowait -no-acpi -boot c -usb -hda \ /dev/HostVG/QEMUGuest1 -spice port=5903,tls-port=5904,addr=127.0.0.1,\ x509-dir=/etc/pki/libvirt-spice,tls-channel=main,plaintext-channel=inputs -vga \ qxl -global qxl-vga.ram_size=67108864 -global qxl-vga.vram_size=33554432 \ --device qxl,id=video1,ram_size=67108864,vram_size=67108864,bus=pci.0,addr=0x4 \ +-global qxl-vga.revision=4 \ +-device qxl,id=video1,ram_size=67108864,vram_size=67108864,revision=4,bus=pci.0,addr=0x4 \ -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x3 diff --git a/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-qxl-vga.xml b/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-qxl-vga.xml index 563d371..3cd0c42 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-qxl-vga.xml +++ b/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-qxl-vga.xml @@ -28,10 +28,10 @@ <channel name='inputs' mode='insecure'/> </graphics> <video> - <model type='qxl' ram='65536' vram='32768' heads='1'/> + <model type='qxl' ram='65536' vram='32768' heads='1' revision='4'/> </video> <video> - <model type='qxl' ram='65536' vram='65536' heads='1'/> + <model type='qxl' ram='65536' vram='65536' heads='1' revision='4'/> </video> <memballoon model='virtio'/> </devices> diff --git a/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice.args b/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice.args index d7cfae0..082eaf7 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice.args +++ b/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice.args @@ -6,5 +6,6 @@ x509-dir=/etc/pki/libvirt-spice,tls-channel=default,tls-channel=main,plaintext-c image-compression=auto_glz,jpeg-wan-compression=auto,zlib-glz-wan-compression=auto,\ playback-compression=on,streaming-video=filter,disable-copy-paste -vga \ qxl -global qxl.ram_size=67108864 -global qxl.vram_size=18874368 \ --device qxl,id=video1,ram_size=67108864,vram_size=33554432,bus=pci.0,addr=0x4 \ +-global qxl.revision=4 \ +-device qxl,id=video1,ram_size=67108864,vram_size=33554432,revision=4,bus=pci.0,addr=0x4 \ -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x3 diff --git a/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice.xml b/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice.xml index 9a36660..e99dbc8 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice.xml +++ b/tests/qemuxml2argvdata/qemuxml2argv-graphics-spice.xml @@ -34,10 +34,10 @@ <clipboard copypaste='no'/> </graphics> <video> - <model type='qxl' ram='65536' vram='18432' heads='1'/> + <model type='qxl' ram='65536' vram='18432' heads='1' revision='4'/> </video> <video> - <model type='qxl' ram='65536' vram='32768' heads='1'/> + <model type='qxl' ram='65536' vram='32768' heads='1' revision='4'/> </video> <memballoon model='virtio'/> </devices> -- 1.8.1.2

1 1

[libvirt] [PATCHv3 0/3] Enable virtio-scsi and virtio-rng for s390
by Viktor Mihajlovski 15 Mar '13

15 Mar '13

Early QEMU versions did not support virtio-scsi or virtio-rng for s390 machines. This series enables libvirt to exploit the capabilities provided by newer QEMUs. Note: I have to apologize for the superfluous iterations which could have been avoided by reading the comment in front of the enum... V2 Changes: - Dropped 1/3, the rename of virtio-scsi-pci capability, we will keep it for compatibility reasons. - Add a new generic virtio-scsi capability for non-pci busses V3 Changes - Reverting to three patches with rename of enum symbol only Viktor Mihajlovski (3): qemu: Rename virtio-scsi capability S390: Enable virtio-scsi and virtio-rng S390: Testcases for virtio-scsi and virtio-rng src/qemu/qemu_capabilities.c | 6 +++- src/qemu/qemu_capabilities.h | 2 +- src/qemu/qemu_command.c | 31 ++++++++++++++++---- tests/qemuhelptest.c | 6 ++-- .../qemuxml2argv-disk-virtio-scsi-ccw.args | 9 ++++++ .../qemuxml2argv-disk-virtio-scsi-ccw.xml | 31 ++++++++++++++++++++ .../qemuxml2argv-virtio-rng-ccw.args | 12 ++++++++ .../qemuxml2argv-virtio-rng-ccw.xml | 30 +++++++++++++++++++ tests/qemuxml2argvtest.c | 19 ++++++++---- 9 files changed, 129 insertions(+), 17 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-scsi-ccw.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-scsi-ccw.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-ccw.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-ccw.xml -- 1.7.9.5

2 7

[libvirt] [PATCH v2 1/2] json: add stream parser
by Dmitry Guryanov 15 Mar '13

15 Mar '13

Add function virJSONValueFromStream, which reads data from a stream and passes it to json parser. When end of the object is reached, it returns this object. To avoid reading from the stream by single bytes it reads to a buffer (in a structure virJSONStreamParserState), which should be passed to a consequent call of this function. So if the end of one object and the beginning of the next object have been read by a single system call - virJSONValueFromStream handle it correctly. example of usage: virJSONValuePtr v; virJSONStreamParserState state; memset(&state, 0, sizeof(state)); while (1) { v = virJSONValueFromStream(mon->fd, &state); if (v == (void *)-1) /* error */ break; if (v == NULL) /* file descriptor has been closed */ break; /* handle object 'v' */ } I need such function for the parallels driver. It caches info about domains and needs some mechanism to update this cache. There is a "prlsrvctl monitor" command which waits for events forever and prints info about events to stdout in json format. So parallels driver could start separate thread which will read from prlsrvctl's stdout and update cache accordingly. There is the same task in qemu_monitor_json, but each json object is printed in a separate line there. It's not possible in my case, because some fields could have line endings. Signed-off-by: Dmitry Guryanov <dguryanov(a)parallels.com> --- Changes: * add function to virJSONValueFromStream * fix bug with returning object which ends just before end of file * fix bug with handling objects without any characters between then, when each read returns data for one entire object. src/libvirt_private.syms | 1 + src/util/virjson.c | 110 ++++++++++++++++++++++++++++++++++++++++++++++ src/util/virjson.h | 8 +++ 3 files changed, 119 insertions(+), 0 deletions(-) diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index 5cad990..567055c 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -1320,6 +1320,7 @@ virJSONValueArrayGet; virJSONValueArraySize; virJSONValueFree; virJSONValueFromString; +virJSONValueFromStream; virJSONValueGetBoolean; virJSONValueGetNumberDouble; virJSONValueGetNumberInt; diff --git a/src/util/virjson.c b/src/util/virjson.c index e6a3b1b..731293f 100644 --- a/src/util/virjson.c +++ b/src/util/virjson.c @@ -990,6 +990,116 @@ cleanup: return ret; } +/* + * Read single JSON object from the stream. Store data, which + * have already been read from the stream, but belongs to the + * next object to the virJSONStreamParserState structure. So that + * consequent call to this function will parse it and return that + * object. + * + * @fd: file descriptor, opened for reading + * @state: pointer to the structure with buffer for data, read from fd. + * + * Return (void *)-1 in case of error, NULL when eof reached, + * pointer to the virJSONValuePtr in case of success. + */ + +virJSONValuePtr virJSONValueFromStream(int fd, virJSONStreamParserStatePtr state) +{ + yajl_handle hand; + virJSONParser parser = { NULL, NULL, 0 }; + virJSONValuePtr value = (void *)-1; +# ifndef WITH_YAJL2 + yajl_parser_config cfg = { 1, 1 }; +# endif + ssize_t len; + int ret = 0; + bool done = false; + +# ifdef WITH_YAJL2 + hand = yajl_alloc(&parserCallbacks, NULL, &parser); + if (hand) { + yajl_config(hand, yajl_allow_comments, 1); + yajl_config(hand, yajl_dont_validate_strings, 0); + yajl_config(hand, yajl_allow_trailing_garbage, 1); + } +# else + hand = yajl_alloc(&parserCallbacks, &cfg, NULL, &parser); +# endif + if (!hand) { + virReportError(VIR_ERR_INTERNAL_ERROR, "%s", + _("Unable to create JSON parser")); + goto cleanup; + } + + do { + if (strlen(&state->buf[state->pos]) == 0) { + state->pos = 0; + memset(state->buf, 0, sizeof(state->buf)); + len = read(fd, state->buf, sizeof(state->buf) - 1); + + if (len < 0) { + virReportSystemError(errno, _("cannot read from fd '%d'"), fd); + virJSONValueFree(parser.head); + goto cleanup; + } + + if (len == 0) { + value = parser.head; + goto cleanup; + } + } + + for (;state->pos < strlen(state->buf); state->pos++) { + unsigned char *buf = (unsigned char *)&state->buf[state->pos]; + /* + * New yaml library has useful function yajl_get_bytes_consumed + * which allows parsing by larger chunks. But rhel-6 has 1.0.7 + * version, which doesn't have it. + */ + ret = yajl_parse(hand, buf, 1); +# ifdef WITH_YAJL2 + if (ret == 0 && yajl_get_bytes_consumed(hand) == 0) { + done = true; + /* state->pos points to the first symbol after current + * object */ + break; + } else if (ret != 0) { +# else + if (ret == 0) { + done = true; + /* state->pos points to the last symbol of the + * current object */ + state->pos++; + break; + } else if (ret != yajl_status_insufficient_data) { +# endif + unsigned char *errstr = yajl_get_error(hand, 1, buf, 1); + virReportError(VIR_ERR_INTERNAL_ERROR, + _("cannot parse json: %s"), + (const char*) errstr); + VIR_FREE(errstr); + virJSONValueFree(parser.head); + goto cleanup; + } + } + } while (!done); + + value = parser.head; + +cleanup: + yajl_free(hand); + + if (parser.nstate) { + int i; + for (i = 0 ; i < parser.nstate ; i++) { + VIR_FREE(parser.state[i].key); + } + } + + VIR_DEBUG("result=%p", parser.head); + return value; +} static int virJSONValueToStringOne(virJSONValuePtr object, yajl_gen g) diff --git a/src/util/virjson.h b/src/util/virjson.h index 67f4398..aeb25ef 100644 --- a/src/util/virjson.h +++ b/src/util/virjson.h @@ -48,6 +48,8 @@ typedef virJSONObjectPair *virJSONObjectPairPtr; typedef struct _virJSONArray virJSONArray; typedef virJSONArray *virJSONArrayPtr; +typedef struct _virJSONStreamParserState virJSONStreamParserState; +typedef virJSONStreamParserState *virJSONStreamParserStatePtr; struct _virJSONObjectPair { char *key; @@ -77,6 +79,11 @@ struct _virJSONValue { } data; }; +struct _virJSONStreamParserState { + char buf[1024]; + size_t pos; +}; + void virJSONValueFree(virJSONValuePtr value); virJSONValuePtr virJSONValueNewString(const char *data); @@ -134,5 +141,6 @@ int virJSONValueObjectAppendNull(virJSONValuePtr object, const char *key); virJSONValuePtr virJSONValueFromString(const char *jsonstring); char *virJSONValueToString(virJSONValuePtr object, bool pretty); +virJSONValuePtr virJSONValueFromStream(int fd, virJSONStreamParserStatePtr state); #endif /* __VIR_JSON_H_ */ -- 1.7.1

1 1

[libvirt] [PATCH] Fix a copy&paste error for virsh dump help
by Yanbing Du 15 Mar '13

15 Mar '13

Signed-off-by: Yanbing Du <ydu(a)redhat.com> --- tools/virsh-domain.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index ab90f58..aa45e72 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -4358,7 +4358,7 @@ static const vshCmdOptDef opts_dump[] = { {.name = "bypass-cache", .type = VSH_OT_BOOL, .flags = 0, - .help = N_("avoid file system cache when saving") + .help = N_("avoid file system cache when dumping") }, {.name = "reset", .type = VSH_OT_BOOL, -- 1.7.1

2 1

[libvirt] [libvirt-sandbox][PATCH] Avoid segfault if broadcast address is NULL
by Alex Jia 15 Mar '13

15 Mar '13

It seems the broadcast address parameter is not optional now, if users haven't specified the broadcast address to the network interface then a segfault will be raised. This patch is just a temporary solution to avoid segfault, and with the patch, users will get a expected error "Unable to parse networks: Unable to parse broadcast address (null)", of course, a better patch should automatically generate a broadcast address according to actual ip address. But AFAIK, the IPv6 hasn't defined broadcast address, so it needs to do address protocol judgement firstly, I will try to implement it later. * How to reproduce? # virt-sandbox -c lxc:/// --network address=192.168.122.1 /bin/sh or # virt-sandbox -c lxc:/// --network address=192.168.122.1/24 /bin/sh Signed-off-by: Alex Jia <ajia(a)redhat.com> --- libvirt-sandbox/libvirt-sandbox-config.c | 8 ++++++++ 1 files changed, 8 insertions(+), 0 deletions(-) diff --git a/libvirt-sandbox/libvirt-sandbox-config.c b/libvirt-sandbox/libvirt-sandbox-config.c index 3dc453a..f392f4b 100644 --- a/libvirt-sandbox/libvirt-sandbox-config.c +++ b/libvirt-sandbox/libvirt-sandbox-config.c @@ -873,6 +873,14 @@ gboolean gvir_sandbox_config_add_network_strv(GVirSandboxConfig *config, goto cleanup; } + if (!bcast) { + g_set_error(error, GVIR_SANDBOX_CONFIG_ERROR, 0, + "Unable to parse broadcast address %s", bcast); + g_free(primary); + goto cleanup; + + } + if (!(bcastaddr = g_inet_address_new_from_string(bcast))) { g_set_error(error, GVIR_SANDBOX_CONFIG_ERROR, 0, "Unable to parse address %s", bcast); -- 1.7.1

2 2

[libvirt] [PATCH v6 0/3] DEVICE_DELETED event
by Michael S. Tsirkin 15 Mar '13

15 Mar '13

libvirt has a long-standing bug: when removing the device, it can request removal but does not know when the removal completes. Add an event so we can fix this in a robust way. First patch only adds the event with ID, second patch adds a path field. Split this way for ease of backport (stable downstreams without QOM would want to only take the first patch). Event without fields is still useful as management can use it to poll device list to figure out which device was removed. Signed-off-by: Michael S. Tsirkin <mst(a)redhat.com> Changes from v5: - Emit an empty event on unnamed devices in patch 1/3, as suggested by Markus Changes from v4: - Add extra triggers and extra fields as requested by Markus Changes from v3: - Document that we only emit events for devices with and ID, as suggested by Markus Changes from v2: - move event toward the end of device_unparent, so that parents are reported after their children, as suggested by Paolo Changes from v1: - move to device_unparent - address comments by Andreas and Eric -- Anthony Liguori Michael S. Tsirkin (3): qdev: DEVICE_DELETED event qom: pass original path to unparent method qmp: add path to device_deleted event QMP/qmp-events.txt | 18 ++++++++++++++++++ hw/qdev.c | 15 +++++++++++++-- include/monitor/monitor.h | 1 + include/qom/object.h | 3 ++- monitor.c | 1 + qapi-schema.json | 4 +++- qom/object.c | 4 +++- 7 files changed, 41 insertions(+), 5 deletions(-) -- MST

3 10

[libvirt] [PATCH] util: fix clear_emulator_capabilities=0
by Laine Stump 15 Mar '13

15 Mar '13

My commit 7a2e845a865dc7fa82d2393ea2a770cfc8cf00b4 (and its prerequisites) managed to effectively ignore the clear_emulator_capabilities setting in qemu.conf (visible in the code as the VIR_EXEC_CLEAR_CAPS flag when qemu is being exec'ed), with the result that the capabilities are always cleared regardless of the qemu.conf setting. This patch fixes it by passing the flag through to virSetUIDGIDWithCaps(), which uses it to decide whether or not to clear existing capabilities before adding in those that were requested. Note that the existing capabilities are *always* cleared if the new process is going to run as non-root, since the whole point of running non-root is to have the capabilities removed (it's still possible to add back individual capabilities as needed though). --- This will need to be backported to v1.0.3-maint. src/util/vircommand.c | 4 +++- src/util/virutil.c | 16 ++++++++++------ src/util/virutil.h | 3 ++- 3 files changed, 15 insertions(+), 8 deletions(-) diff --git a/src/util/vircommand.c b/src/util/vircommand.c index f672101..08d3d29 100644 --- a/src/util/vircommand.c +++ b/src/util/vircommand.c @@ -639,8 +639,10 @@ virExec(virCommandPtr cmd) cmd->capabilities || (cmd->flags & VIR_EXEC_CLEAR_CAPS)) { VIR_DEBUG("Setting child uid:gid to %d:%d with caps %llx", (int)cmd->uid, (int)cmd->gid, cmd->capabilities); - if (virSetUIDGIDWithCaps(cmd->uid, cmd->gid, cmd->capabilities) < 0) + if (virSetUIDGIDWithCaps(cmd->uid, cmd->gid, cmd->capabilities, + (cmd->flags & VIR_EXEC_CLEAR_CAPS)) < 0) { goto fork_error; + } } if (cmd->pwd) { diff --git a/src/util/virutil.c b/src/util/virutil.c index 4605c78..634b7e3 100644 --- a/src/util/virutil.c +++ b/src/util/virutil.c @@ -2998,18 +2998,21 @@ virGetGroupName(gid_t gid ATTRIBUTE_UNUSED) * errno). */ int -virSetUIDGIDWithCaps(uid_t uid, gid_t gid, unsigned long long capBits) +virSetUIDGIDWithCaps(uid_t uid, gid_t gid, unsigned long long capBits, + bool clearExistingCaps) { int ii, capng_ret, ret = -1; bool need_setgid = false, need_setuid = false; bool need_prctl = false, need_setpcap = false; - /* First drop all caps except those in capBits + the extra ones we - * need to change uid/gid and change the capabilities bounding - * set. + /* First drop all caps (unless the requested uid is "unchanged" or + * root and clearExistingCaps wasn't requested), then add back + * those in capBits + the extra ones we need to change uid/gid and + * change the capabilities bounding set. */ - capng_clear(CAPNG_SELECT_BOTH); + if (clearExistingCaps || (uid != 1 && uid != 0)) + capng_clear(CAPNG_SELECT_BOTH); for (ii = 0; ii <= CAP_LAST_CAP; ii++) { if (capBits & (1ULL << ii)) { @@ -3095,7 +3098,8 @@ cleanup: int virSetUIDGIDWithCaps(uid_t uid, gid_t gid, - unsigned long long capBits ATTRIBUTE_UNUSED) + unsigned long long capBits ATTRIBUTE_UNUSED, + bool clearExistingCaps ATTRIBUTE_UNUSED) { return virSetUIDGID(uid, gid); } diff --git a/src/util/virutil.h b/src/util/virutil.h index 2dc3403..665ad78 100644 --- a/src/util/virutil.h +++ b/src/util/virutil.h @@ -54,7 +54,8 @@ int virPipeReadUntilEOF(int outfd, int errfd, char **outbuf, char **errbuf); int virSetUIDGID(uid_t uid, gid_t gid); -int virSetUIDGIDWithCaps(uid_t uid, gid_t gid, unsigned long long capBits); +int virSetUIDGIDWithCaps(uid_t uid, gid_t gid, unsigned long long capBits, + bool clearExistingCaps); int virFileReadLimFD(int fd, int maxlen, char **buf) ATTRIBUTE_RETURN_CHECK; -- 1.8.1.4

3 5

[libvirt] [PATCHv2 0/2] Enable virtio-scsi and virtio-rng for s390
by Viktor Mihajlovski 15 Mar '13

15 Mar '13

Early QEMU versions did not support virtio-scsi or virtio-rng for s390 machines. This series enables libvirt to exploit the capabilities provided by newer QEMUs. V2 Changes: - Dropped 1/3, the rename of virtio-scsi-pci capability, we will keep it for compatibility reasons. - Add a new generic virtio-scsi capability for non-pci busses Viktor Mihajlovski (2): S390: Enable virtio-scsi and virtio-rng S390: Testcases for virtio-scsi and virtio-rng src/qemu/qemu_capabilities.c | 7 ++++- src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 32 ++++++++++++++++---- .../qemuxml2argv-disk-virtio-scsi-ccw.args | 9 ++++++ .../qemuxml2argv-disk-virtio-scsi-ccw.xml | 31 +++++++++++++++++++ .../qemuxml2argv-virtio-rng-ccw.args | 12 ++++++++ .../qemuxml2argv-virtio-rng-ccw.xml | 30 ++++++++++++++++++ tests/qemuxml2argvtest.c | 7 +++++ 8 files changed, 122 insertions(+), 7 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-scsi-ccw.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-scsi-ccw.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-ccw.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-ccw.xml -- 1.7.9.5

3 6

[libvirt] [PATCH] v1:Support for adding a static route to a bridge
by Gene Czarcinski 15 Mar '13

15 Mar '13

This patch adds support for adding a static route for a network. The "via" specifies the gateway's IP address. Both IPv4 and IPv6 static routes are supported although it is expected that this functionality will have more use with IPv6. Extensive tests are done to validate that the input definitions are correct. For example, for a static route ip definition, the address must be for a network and not a host. Signed-off-by: Gene Czarcinski <gene(a)czarc.net> --- docs/formatnetwork.html.in | 32 ++++++- docs/schemas/network.rng | 3 + src/conf/network_conf.c | 104 ++++++++++++++++++++- src/conf/network_conf.h | 2 + src/libvirt_private.syms | 1 + src/network/bridge_driver.c | 38 ++++++++ src/util/virnetdev.c | 47 ++++++++++ src/util/virnetdev.h | 5 + .../networkxml2xmlin/dhcp6host-routed-network.xml | 4 + .../networkxml2xmlout/dhcp6host-routed-network.xml | 4 + 10 files changed, 236 insertions(+), 4 deletions(-) diff --git a/docs/formatnetwork.html.in b/docs/formatnetwork.html.in index 4dd0415..f0cadf0 100644 --- a/docs/formatnetwork.html.in +++ b/docs/formatnetwork.html.in @@ -537,7 +537,9 @@ associated with the virtual network, and optionally enable DHCP services. These elements are only valid for isolated networks (no <code>forward</code> element specified), and for those with - a forward mode of 'route' or 'nat'. + a forward mode of 'route' or 'nat'. Another optional addressing + element <code>via</code> can be used to establish a static + route for IPv4 or IPv6 networks. <pre> @@ -560,6 +562,7 @@ </dhcp> </ip> <ip family="ipv6" address="2001:db8:ca2:2::1" prefix="64" /> + <ip family="ipv6" address="2001:db9:ca1:1::" prefix="64" via="2001:db8:ca2:2::2" /> </network></pre> <dl> @@ -633,7 +636,10 @@ IPv6, multiple addresses on a single network, <code>family</code>, and <code>prefix</code> are support Since 0.8.7. Similar to IPv4, one IPv6 address per network can also have - a <code>dhcp</code> definition. Since 1.0.1 + a <code>dhcp</code> definition. Since 1.0.1 The <code>via</code> + can be used to establish a static route for IPv4 or IPv6 networks. It is an error + to specify <code>via</code> and <code>dhcp</code> for the same IP address. + Since 1.0.4 <dl> <dt><code>tftp</code></dt> @@ -809,6 +815,28 @@ </ip> </network></pre> + + Below is yet another IPv6 variation. This variation has only IPv6 + defined with DHCPv6 on the primary IPv6 network. A second IPv6 + network has a static link to a host on the first (virtual) IPv6 + network. Since 1.0.4 + + + <pre> + <network> + <name>net7</name> + <bridge name="virbr7" /> + <forward mode="route"/> + <ip family="ipv6" address="2001:db8:ca2:7::1" prefix="64" > + <dhcp> + <range start="2001:db8:ca2:7::100" end="2001:db8:ca2::1ff" /> + <host id="0:4:7e:7d:f0:7d:a8:bc:c5:d2:13:32:11:ed:16:ea:84:63" name="lucas" ip="2001:db8:ca2:2:3::4" /> + </dhcp> + </ip> + <ip family="ipv6" address="2001:db8:ca2:8::" prefix="64" via="2001:db8:ca2::4" > + </ip> + </network></pre> + <h3><a name="examplesPrivate">Isolated network config</a></h3> diff --git a/docs/schemas/network.rng b/docs/schemas/network.rng index 6c3fae2..c1cca23 100644 --- a/docs/schemas/network.rng +++ b/docs/schemas/network.rng @@ -262,6 +262,9 @@ <attribute name="family"><ref name="addr-family"/></attribute> </optional> <optional> + <attribute name="via"><ref name="ipAddr"/></attribute> + </optional> + <optional> <element name="tftp"> <attribute name="root"><text/></attribute> </element> diff --git a/src/conf/network_conf.c b/src/conf/network_conf.c index c022fe4..a5f7be2 100644 --- a/src/conf/network_conf.c +++ b/src/conf/network_conf.c @@ -1149,8 +1149,10 @@ virNetworkIPDefParseXML(const char *networkName, xmlNodePtr cur, save; char *address = NULL, *netmask = NULL; - unsigned long prefix; + char *gateway = NULL; + unsigned long prefix = 0; int result = -1; + virSocketAddr testAddr; save = ctxt->node; ctxt->node = node; @@ -1162,6 +1164,7 @@ virNetworkIPDefParseXML(const char *networkName, def->prefix = 0; else def->prefix = prefix; + gateway = virXPathString("string(./@via)", ctxt); netmask = virXPathString("string(./@netmask)", ctxt); @@ -1175,7 +1178,17 @@ virNetworkIPDefParseXML(const char *networkName, } - /* validate family vs. address */ + if (gateway) { + if (virSocketAddrParse(&def->gateway, gateway, AF_UNSPEC) < 0) { + virReportError(VIR_ERR_XML_ERROR, + _("Bad gateway address '%s' in definition of network '%s'"), + gateway, networkName); + goto cleanup; + } + + } + + /* validate family vs. address (and gateway address too) */ if (def->family == NULL) { if (!(VIR_SOCKET_ADDR_IS_FAMILY(&def->address, AF_INET) || VIR_SOCKET_ADDR_IS_FAMILY(&def->address, AF_UNSPEC))) { @@ -1184,6 +1197,14 @@ virNetworkIPDefParseXML(const char *networkName, address, networkName); goto cleanup; } + if (gateway && + (!(VIR_SOCKET_ADDR_IS_FAMILY(&def->address, AF_INET) || + VIR_SOCKET_ADDR_IS_FAMILY(&def->address, AF_UNSPEC)))) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("no family specified for non-IPv4 gateway address '%s' in network '%s'"), + address, networkName); + goto cleanup; + } } else if (STREQ(def->family, "ipv4")) { if (!VIR_SOCKET_ADDR_IS_FAMILY(&def->address, AF_INET)) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, @@ -1191,6 +1212,12 @@ virNetworkIPDefParseXML(const char *networkName, address, networkName); goto cleanup; } + if (gateway && (!VIR_SOCKET_ADDR_IS_FAMILY(&def->address, AF_INET))) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("family 'ipv4' specified for non-IPv4 gateway address '%s' in network '%s'"), + address, networkName); + goto cleanup; + } } else if (STREQ(def->family, "ipv6")) { if (!VIR_SOCKET_ADDR_IS_FAMILY(&def->address, AF_INET6)) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, @@ -1198,6 +1225,24 @@ virNetworkIPDefParseXML(const char *networkName, address, networkName); goto cleanup; } + if (def->prefix == 0 || def->prefix > 64) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("invalid prefix=%u specified for IPv6 address '%s' in network '%s'"), + def->prefix, address, networkName); + goto cleanup; + } + if (netmask) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("specifying netmask invalid for IPv6 address '%s' in network '%s'"), + address, networkName); + goto cleanup; + } + if (gateway && (!VIR_SOCKET_ADDR_IS_FAMILY(&def->gateway, AF_INET6))) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("family 'ipv6' specified for non-IPv6 gateway address '%s' in network '%s'"), + gateway, networkName); + goto cleanup; + } } else { virReportError(VIR_ERR_XML_ERROR, _("Unrecognized family '%s' in definition of network '%s'"), @@ -1241,16 +1286,55 @@ virNetworkIPDefParseXML(const char *networkName, } } + /* if static route gateway specified, make sure the address is a network address */ + if (gateway) { + if (def->prefix > 0) { + if (virSocketAddrMaskByPrefix(&def->address, def->prefix, &testAddr) < 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("error converting address '%s' with prefix=%u to network-address in network '%s'"), + address, def->prefix, networkName); + goto cleanup; + } + } + if (netmask) { + if (virSocketAddrMask(&def->address, &def->netmask, &testAddr) < 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("error converting address '%s' with netmask '%s' to network-address in network '%s'"), + address, netmask, networkName); + goto cleanup; + } + } + if (!virSocketAddrEqual(&def->address, &testAddr)) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("address '%s' in network '%s' is not a network-address"), + address, networkName); + goto cleanup; + } + } + cur = node->children; while (cur != NULL) { if (cur->type == XML_ELEMENT_NODE && xmlStrEqual(cur->name, BAD_CAST "dhcp")) { + if (gateway) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("<dhcp> element unsupported when static route (via) previously defined in network '%s'"), + networkName); + goto cleanup; + } if (virNetworkDHCPDefParseXML(networkName, cur, def) < 0) goto cleanup; } else if (cur->type == XML_ELEMENT_NODE && xmlStrEqual(cur->name, BAD_CAST "tftp")) { char *root; + if (gateway) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("<tftp> element unsupported when static route (via) previously defined in network '%s'"), + networkName); + goto cleanup; + } + if (!VIR_SOCKET_ADDR_IS_FAMILY(&def->address, AF_INET)) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, _("Unsupported <tftp> element in an IPv6 element in network '%s'"), @@ -1274,6 +1358,7 @@ cleanup: } VIR_FREE(address); VIR_FREE(netmask); + VIR_FREE(gateway); ctxt->node = save; return result; @@ -2108,6 +2193,7 @@ virNetworkIpDefFormat(virBufferPtr buf, const virNetworkIpDefPtr def) { int result = -1; + bool routeFlg = false; virBufferAddLit(buf, "<ip"); @@ -2131,15 +2217,29 @@ virNetworkIpDefFormat(virBufferPtr buf, if (def->prefix > 0) { virBufferAsprintf(buf," prefix='%u'", def->prefix); } + if (VIR_SOCKET_ADDR_VALID(&def->gateway)) { + char *addr = virSocketAddrFormat(&def->gateway); + if (!addr) + goto error; + routeFlg = true; + virBufferAsprintf(buf, " via='%s'", addr); + VIR_FREE(addr); + } virBufferAddLit(buf, ">\n"); virBufferAdjustIndent(buf, 2); if (def->tftproot) { + if (routeFlg) /* this should not occur */ + goto error; /* but in case it does, better to error out */ virBufferEscapeString(buf, "<tftp root='%s' />\n", def->tftproot); } if ((def->nranges || def->nhosts)) { int ii; + + if (routeFlg) /* this should not occur */ + goto error; /* but in case it does, better to error out */ + virBufferAddLit(buf, "<dhcp>\n"); virBufferAdjustIndent(buf, 2); diff --git a/src/conf/network_conf.h b/src/conf/network_conf.h index 6ce9a00..91aff5f 100644 --- a/src/conf/network_conf.h +++ b/src/conf/network_conf.h @@ -115,6 +115,8 @@ struct _virNetworkIpDef { char *family; /* ipv4 or ipv6 - default is ipv4 */ virSocketAddr address; /* Bridge IP address */ + virSocketAddr gateway; /* gateway IP address for ip-route */ + /* One or the other of the following two will be used for a given * IP address, but never both. The parser guarantees this. * Use virNetworkIpDefPrefix/virNetworkIpDefNetmask rather diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index ed46479..988b97a 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -1470,6 +1470,7 @@ virNetDevReplaceMacAddress; virNetDevReplaceNetConfig; virNetDevRestoreMacAddress; virNetDevRestoreNetConfig; +virNetDevSetGateway; virNetDevSetIPv4Address; virNetDevSetMAC; virNetDevSetMTU; diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c index 31c8585..646a480 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.c @@ -2360,6 +2360,7 @@ out: return ret; } +/* add an IP address to a bridge */ static int networkAddAddrToBridge(virNetworkObjPtr network, virNetworkIpDefPtr ipdef) @@ -2380,6 +2381,28 @@ networkAddAddrToBridge(virNetworkObjPtr network, return 0; } +/* add an IP route to a bridge */ +static int +networkAddRouteToBridge(virNetworkObjPtr network, + virNetworkIpDefPtr ipdef) +{ + int prefix = virNetworkIpDefPrefix(ipdef); + + if (prefix < 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("bridge '%s' has an invalid netmask or IP address"), + network->def->bridge); + return -1; + } + + if (virNetDevSetGateway(network->def->bridge, + &ipdef->address, + prefix, + &ipdef->gateway) < 0) + return -1; + return 0; +} + static int networkStartNetworkVirtual(struct network_driver *driver, virNetworkObjPtr network) @@ -2446,9 +2469,12 @@ networkStartNetworkVirtual(struct network_driver *driver, if (networkAddIptablesRules(driver, network) < 0) goto err1; + /* first, do all of the addresses */ for (ii = 0; (ipdef = virNetworkDefGetIpByIndex(network->def, AF_UNSPEC, ii)); ii++) { + if (VIR_SOCKET_ADDR_VALID(&ipdef->gateway)) + continue; /* this is a static route specification */ if (VIR_SOCKET_ADDR_IS_FAMILY(&ipdef->address, AF_INET)) v4present = true; if (VIR_SOCKET_ADDR_IS_FAMILY(&ipdef->address, AF_INET6)) @@ -2464,6 +2490,18 @@ networkStartNetworkVirtual(struct network_driver *driver, if (virNetDevSetOnline(network->def->bridge, 1) < 0) goto err2; + /* then do all of the static routes */ + for (ii = 0; + (ipdef = virNetworkDefGetIpByIndex(network->def, AF_UNSPEC, ii)); + ii++) { + + /* Add the IP route to the bridge */ + if (VIR_SOCKET_ADDR_VALID(&ipdef->gateway)) { + if (networkAddRouteToBridge(network, ipdef) < 0) + goto err2; + } + } + /* If forward.type != NONE, turn on global IP forwarding */ if (network->def->forward.type != VIR_NETWORK_FORWARD_NONE && networkEnableIpForwarding(v4present, v6present) < 0) { diff --git a/src/util/virnetdev.c b/src/util/virnetdev.c index 296871c..c90b3d2 100644 --- a/src/util/virnetdev.c +++ b/src/util/virnetdev.c @@ -729,6 +729,7 @@ int virNetDevGetVLanID(const char *ifname ATTRIBUTE_UNUSED, * Add an IP address to an interface. This function *does not* remove * any previously added IP addresses - that must be done separately with * brDelInetAddress. + * TODO: what is "brDelInetAddress"? * * Returns 0 in case of success or -1 in case of error. */ @@ -769,6 +770,52 @@ cleanup: } /** + * virNetDevSetGateway: + * @ifname: the interface name + * @addr: the IP network address (IPv4 or IPv6) + * @prefix: number of 1 bits in the netmask + * @gateway: via address for route (same as @addr) + * + * Add a route for a network IP address to an interface. This function + * *does not* remove any previously added IP static routes. + * + * Returns 0 in case of success or -1 in case of error. + */ + +int virNetDevSetGateway(const char *ifname, + virSocketAddr *addr, + unsigned int prefix, + virSocketAddr *gateway) +{ + virCommandPtr cmd = NULL; + char *addrstr = NULL, *gatewaystr = NULL; + int ret = -1; + + if (!(addrstr = virSocketAddrFormat(addr))) + goto cleanup; + if (!(gatewaystr = virSocketAddrFormat(gateway))) + goto cleanup; + cmd = virCommandNew(IP_PATH); + virCommandAddArgList(cmd, "route", "add", NULL); + virCommandAddArgFormat(cmd, "%s/%u", addrstr, prefix); + virCommandAddArgList(cmd, "via", NULL); + virCommandAddArgFormat(cmd, "%s", gatewaystr); + virCommandAddArgList(cmd, "dev", ifname, NULL); + virCommandAddArgList(cmd, "proto", "static", "metric", NULL); + virCommandAddArgFormat(cmd, "%u", 1); + + if (virCommandRun(cmd, NULL) < 0) + goto cleanup; + + ret = 0; +cleanup: + VIR_FREE(addrstr); + VIR_FREE(gatewaystr); + virCommandFree(cmd); + return ret; +} + +/** * virNetDevClearIPv4Address: * @ifname: the interface name * @addr: the IP address (IPv4 or IPv6) diff --git a/src/util/virnetdev.h b/src/util/virnetdev.h index 06d0650..8b94ea8 100644 --- a/src/util/virnetdev.h +++ b/src/util/virnetdev.h @@ -42,6 +42,11 @@ int virNetDevSetIPv4Address(const char *ifname, virSocketAddr *addr, unsigned int prefix) ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(2) ATTRIBUTE_RETURN_CHECK; +int virNetDevSetGateway(const char *ifname, + virSocketAddr *addr, + unsigned int prefix, + virSocketAddr *gateway) + ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(2) ATTRIBUTE_RETURN_CHECK; int virNetDevClearIPv4Address(const char *ifname, virSocketAddr *addr, unsigned int prefix) diff --git a/tests/networkxml2xmlin/dhcp6host-routed-network.xml b/tests/networkxml2xmlin/dhcp6host-routed-network.xml index 2693d87..dcad62d 100644 --- a/tests/networkxml2xmlin/dhcp6host-routed-network.xml +++ b/tests/networkxml2xmlin/dhcp6host-routed-network.xml @@ -19,4 +19,8 @@ <host id='0:1:0:1:18:aa:62:fe:0:16:3e:44:55:66' name='badbob' ip='2001:db8:ac10:fd01::1:24' /> </dhcp> </ip> + <ip address="192.168.222.0" netmask="255.255.255.0" via="192.168.122.10"> + </ip> + <ip family="ipv6" address="2001:db8:ac10:fc00::" prefix="64" via="2001:db8:ac10:fd01::1:24"> + </ip> </network> diff --git a/tests/networkxml2xmlout/dhcp6host-routed-network.xml b/tests/networkxml2xmlout/dhcp6host-routed-network.xml index 7305043..880c2dd 100644 --- a/tests/networkxml2xmlout/dhcp6host-routed-network.xml +++ b/tests/networkxml2xmlout/dhcp6host-routed-network.xml @@ -21,4 +21,8 @@ <host id='0:1:0:1:18:aa:62:fe:0:16:3e:44:55:66' name='badbob' ip='2001:db8:ac10:fd01::1:24' /> </dhcp> </ip> + <ip address='192.168.222.0' netmask='255.255.255.0' via='192.168.122.10'> + </ip> + <ip family='ipv6' address='2001:db8:ac10:fc00::' prefix='64' via='2001:db8:ac10:fd01::1:24'> + </ip> </network> -- 1.8.1.4

1 2

[libvirt] when define NAT network, assign what kind of forward dev is better
by yue 15 Mar '13

15 Mar '13

hi all: in my case, my physical machine has one ethernet, i want to use both bridge and nat network. i create a bridge outbr0(addif eth0), when creating NAT network, there are 3 options for forward-dev 1.<forward mode="nat" /> 2.<forward mode="nat" dev="outbr0"/> 3.<forward mode="nat" dev="eth0"/> i do not know which one is the best? thanks

2 1

[libvirt] Adding and Clearing bridge addresses
by Gene Czarcinski 15 Mar '13

15 Mar '13

I have been working on this patch to have libvirt optionally set static routes. So I found the function that adds both IPv4 and IPv6 addresses to the bridge in virnetdev.c. I found that besides the virNetDevAddIPv4Address() there is also virNetDevCleanIPv4Address(). I patterned my virNetDevAddGateway() after the virNetDevAddIPv4Address() function. What I am puzzled about is that it appears that nobody calls the Clear function. What don't I understand? It appears to me that the addresses and routes assigned to a bridge are all removed when the network is destroyed but otherwise remain. Gene

3 4

[libvirt] [PATCH 0/3] Enable virtio-scsi and virtio-rng for s390
by Viktor Mihajlovski 15 Mar '13

15 Mar '13

Early QEMU versions did not support virtio-scsi or virtio-rng for s390 machines. This series enables libvirt to exploit the capabilities provided by newer QEMUs. Viktor Mihajlovski (3): qemu: rename virtio-scsi capability S390: Enable virtio-scsi and virtio-rng S390: Testcases for virtio-scsi and virtio-rng src/qemu/qemu_capabilities.c | 8 +++-- src/qemu/qemu_capabilities.h | 2 +- src/qemu/qemu_command.c | 31 ++++++++++++++++---- tests/qemuhelptest.c | 6 ++-- .../qemuxml2argv-disk-virtio-scsi-ccw.args | 9 ++++++ .../qemuxml2argv-disk-virtio-scsi-ccw.xml | 31 ++++++++++++++++++++ .../qemuxml2argv-virtio-rng-ccw.args | 12 ++++++++ .../qemuxml2argv-virtio-rng-ccw.xml | 30 +++++++++++++++++++ tests/qemuxml2argvtest.c | 19 ++++++++---- 9 files changed, 130 insertions(+), 18 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-scsi-ccw.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-scsi-ccw.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-ccw.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-ccw.xml -- 1.7.9.5

2 4

[libvirt] [PATCH] qemu: detect multi-head qxl via more than version check
by Eric Blake 14 Mar '13

14 Mar '13

Multi-head QXL support is so useful that distros have started to backport it to qemu earlier than 1.2. After discussion with Alon Levy, we determined that the existence of the qxl-vga.surfaces property is a reliable indicator of whether '-device qxl-vga' works, or whether we have to stick to the older '-vga qxl'. I'm leaving in the existing check for QEMU_CAPS_DEVICE_VIDEO_PRIMARY tied to qemu 1.2 and newer (in case qemu is built without qxl support), but for those distros that backport qxl, this additional capability check will allow the correct command line for both RHEL 6.3 (which lacks the feature) and RHEL 6.4 (where qemu still claims to be version 0.12.2.x, but has backported multi-head qxl). * src/qemu/qemu_capabilities.c (virQEMUCapsObjectPropsQxlVga): New property test. (virQEMUCapsExtractDeviceStr): Probe for backport of new capability to qemu earlier than 1.2. * tests/qemuhelpdata/qemu-kvm-1.2.0-device: Update test. * tests/qemuhelpdata/qemu-1.2.0-device: Likewise. * tests/qemuhelpdata/qemu-kvm-0.12.1.2-rhel62-beta-device: Likewise. --- src/qemu/qemu_capabilities.c | 7 +++++++ tests/qemuhelpdata/qemu-1.2.0-device | 16 ++++++++++++++++ tests/qemuhelpdata/qemu-kvm-0.12.1.2-rhel62-beta-device | 10 ++++++++++ tests/qemuhelpdata/qemu-kvm-1.2.0-device | 16 ++++++++++++++++ 4 files changed, 49 insertions(+) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 79cfdb3..17747c1 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -1383,6 +1383,10 @@ static struct virQEMUCapsStringFlags virQEMUCapsObjectPropsUsbHost[] = { { "bootindex", QEMU_CAPS_USB_HOST_BOOTINDEX }, }; +static struct virQEMUCapsStringFlags virQEMUCapsObjectPropsQxlVga[] = { + { "surfaces", QEMU_CAPS_DEVICE_VIDEO_PRIMARY }, +}; + struct virQEMUCapsObjectTypeProps { const char *type; struct virQEMUCapsStringFlags *props; @@ -1416,6 +1420,8 @@ static struct virQEMUCapsObjectTypeProps virQEMUCapsObjectProps[] = { ARRAY_CARDINALITY(virQEMUCapsObjectPropsUsbRedir) }, { "usb-host", virQEMUCapsObjectPropsUsbHost, ARRAY_CARDINALITY(virQEMUCapsObjectPropsUsbHost) }, + { "qxl-vga", virQEMUCapsObjectPropsQxlVga, + ARRAY_CARDINALITY(virQEMUCapsObjectPropsQxlVga) }, }; @@ -1613,6 +1619,7 @@ virQEMUCapsExtractDeviceStr(const char *qemu, "-device", "usb-redir,?", "-device", "ide-drive,?", "-device", "usb-host,?", + "-device", "qxl-vga,?", NULL); /* qemu -help goes to stdout, but qemu -device ? goes to stderr. */ virCommandSetErrorBuffer(cmd, &output); diff --git a/tests/qemuhelpdata/qemu-1.2.0-device b/tests/qemuhelpdata/qemu-1.2.0-device index 5613e00..027d99a 100644 --- a/tests/qemuhelpdata/qemu-1.2.0-device +++ b/tests/qemuhelpdata/qemu-1.2.0-device @@ -208,3 +208,19 @@ usb-host.bootindex=int32 usb-host.pipeline=on/off usb-host.port=string usb-host.full-path=on/off +qxl-vga.ram_size=uint32 +qxl-vga.vram_size=uint32 +qxl-vga.revision=uint32 +qxl-vga.debug=uint32 +qxl-vga.guestdebug=uint32 +qxl-vga.cmdlog=uint32 +qxl-vga.ram_size_mb=uint32 +qxl-vga.vram_size_mb=uint32 +qxl-vga.vram64_size_mb=uint32 +qxl-vga.vgamem_mb=uint32 +qxl-vga.surfaces=int32 +qxl-vga.addr=pci-devfn +qxl-vga.romfile=string +qxl-vga.rombar=uint32 +qxl-vga.multifunction=on/off +qxl-vga.command_serr_enable=on/off diff --git a/tests/qemuhelpdata/qemu-kvm-0.12.1.2-rhel62-beta-device b/tests/qemuhelpdata/qemu-kvm-0.12.1.2-rhel62-beta-device index ee0fd78..5eab539 100644 --- a/tests/qemuhelpdata/qemu-kvm-0.12.1.2-rhel62-beta-device +++ b/tests/qemuhelpdata/qemu-kvm-0.12.1.2-rhel62-beta-device @@ -118,3 +118,13 @@ virtio-net-pci.addr=pci-devfn virtio-net-pci.romfile=string virtio-net-pci.rombar=uint32 virtio-net-pci.multifunction=on/off +qxl-vga.ram_size=uint32 +qxl-vga.vram_size=uint32 +qxl-vga.revision=uint32 +qxl-vga.debug=uint32 +qxl-vga.guestdebug=uint32 +qxl-vga.cmdlog=uint32 +qxl-vga.addr=pci-devfn +qxl-vga.romfile=string +qxl-vga.rombar=uint32 +qxl-vga.multifunction=on/off diff --git a/tests/qemuhelpdata/qemu-kvm-1.2.0-device b/tests/qemuhelpdata/qemu-kvm-1.2.0-device index 879a049..ebc27f0 100644 --- a/tests/qemuhelpdata/qemu-kvm-1.2.0-device +++ b/tests/qemuhelpdata/qemu-kvm-1.2.0-device @@ -220,3 +220,19 @@ usb-host.bootindex=int32 usb-host.pipeline=on/off usb-host.port=string usb-host.full-path=on/off +qxl-vga.ram_size=uint32 +qxl-vga.vram_size=uint32 +qxl-vga.revision=uint32 +qxl-vga.debug=uint32 +qxl-vga.guestdebug=uint32 +qxl-vga.cmdlog=uint32 +qxl-vga.ram_size_mb=uint32 +qxl-vga.vram_size_mb=uint32 +qxl-vga.vram64_size_mb=uint32 +qxl-vga.vgamem_mb=uint32 +qxl-vga.surfaces=int32 +qxl-vga.addr=pci-devfn +qxl-vga.romfile=string +qxl-vga.rombar=uint32 +qxl-vga.multifunction=on/off +qxl-vga.command_serr_enable=on/off -- 1.8.1.4

2 2

[libvirt] [PATCH libvirt-glib v2 1/2] Add gvir_config_object_replace_child_with_attribute_enum()
by Zeeshan Ali (Khattak) 14 Mar '13

14 Mar '13

From: "Zeeshan Ali (Khattak)" <zeeshanak(a)gnome.org> Internal libvirt-gconfig helper to replace enum XML attributes. --- libvirt-gconfig/libvirt-gconfig-object-private.h | 5 +++++ libvirt-gconfig/libvirt-gconfig-object.c | 14 ++++++++++++++ 2 files changed, 19 insertions(+) diff --git a/libvirt-gconfig/libvirt-gconfig-object-private.h b/libvirt-gconfig/libvirt-gconfig-object-private.h index 830517c..91355fb 100644 --- a/libvirt-gconfig/libvirt-gconfig-object-private.h +++ b/libvirt-gconfig/libvirt-gconfig-object-private.h @@ -74,6 +74,11 @@ void gvir_config_object_replace_child_with_attribute(GVirConfigObject *object, const char *child_name, const char *attr_name, const char *attr_value); +void gvir_config_object_replace_child_with_attribute_enum(GVirConfigObject *object, + const char *child_name, + const char *attr_name, + GType attr_type, + unsigned int attr_value); void gvir_config_object_delete_child(GVirConfigObject *object, const char *child_name, const char *ns_href); diff --git a/libvirt-gconfig/libvirt-gconfig-object.c b/libvirt-gconfig/libvirt-gconfig-object.c index ac0545c..c4e4271 100644 --- a/libvirt-gconfig/libvirt-gconfig-object.c +++ b/libvirt-gconfig/libvirt-gconfig-object.c @@ -459,6 +459,20 @@ gvir_config_object_replace_child_with_attribute(GVirConfigObject *object, g_object_unref(G_OBJECT(child)); } +G_GNUC_INTERNAL void +gvir_config_object_replace_child_with_attribute_enum(GVirConfigObject *object, + const char *child_name, + const char *attr_name, + GType attr_type, + unsigned int attr_value) +{ + GVirConfigObject *child; + + child = gvir_config_object_replace_child(object, child_name); + gvir_config_object_set_attribute_with_type(child, attr_name, attr_type, attr_value, NULL); + g_object_unref(G_OBJECT(child)); +} + struct NodeMatch { const char *name; const char *ns; -- 1.8.1.4

2 3

[libvirt] [PATCH] Fix deps for generating RPC dispatch code
by Daniel P. Berrange 14 Mar '13

14 Mar '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> The src/lxc/lxc_*_dispatch.h files only had deps on the RPC generator script & the XDR definition file. So when the Makefile.am args passed to the generator were change, the disaptch code was not re-generated. This caused a build failure CC libvirt_lxc-lxc_controller.o lxc/lxc_controller.c: In function 'virLXCControllerSetupServer': lxc/lxc_controller.c:718:47: error: 'virLXCMonitorProcs' undeclared (first use in this function) lxc/lxc_controller.c:718:47: note: each undeclared identifier is reported only once for each function it appears in lxc/lxc_controller.c:719:47: error: 'virLXCMonitorNProcs' undeclared (first use in this function) make[3]: *** [libvirt_lxc-lxc_controller.o] Error 1 For added fun, the generated files were not listed in CLEANFILES, so only a 'git clean -f' would fix the build Pushed as a build break fix Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- src/Makefile.am | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/src/Makefile.am b/src/Makefile.am index 0c0dfb3..8b591d2 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -295,17 +295,17 @@ QEMU_PROTOCOL = $(srcdir)/remote/qemu_protocol.x REMOTE_DRIVER_PROTOCOL = $(REMOTE_PROTOCOL) $(QEMU_PROTOCOL) $(LXC_PROTOCOL) $(srcdir)/remote/remote_client_bodies.h: $(srcdir)/rpc/gendispatch.pl \ - $(REMOTE_PROTOCOL) + $(REMOTE_PROTOCOL) Makefile.am $(AM_V_GEN)$(PERL) -w $(srcdir)/rpc/gendispatch.pl \ -k remote REMOTE $(REMOTE_PROTOCOL) > $@ $(srcdir)/remote/lxc_client_bodies.h: $(srcdir)/rpc/gendispatch.pl \ - $(LXC_PROTOCOL) + $(LXC_PROTOCOL) Makefile.am $(AM_V_GEN)$(PERL) -w $(srcdir)/rpc/gendispatch.pl \ -k lxc LXC $(LXC_PROTOCOL) > $@ $(srcdir)/remote/qemu_client_bodies.h: $(srcdir)/rpc/gendispatch.pl \ - $(QEMU_PROTOCOL) + $(QEMU_PROTOCOL) Makefile.am $(AM_V_GEN)$(PERL) -w $(srcdir)/rpc/gendispatch.pl \ -k qemu QEMU $(QEMU_PROTOCOL) > $@ @@ -458,12 +458,12 @@ LXC_GENERATED = \ LXC_MONITOR_PROTOCOL = $(srcdir)/lxc/lxc_monitor_protocol.x $(srcdir)/lxc/lxc_monitor_dispatch.h: $(srcdir)/rpc/gendispatch.pl \ - $(LXC_MONITOR_PROTOCOL) + $(LXC_MONITOR_PROTOCOL) Makefile.am $(AM_V_GEN)$(PERL) -w $(srcdir)/rpc/gendispatch.pl \ -k virLXCMonitor VIR_LXC_MONITOR $(LXC_MONITOR_PROTOCOL) > $@ $(srcdir)/lxc/lxc_controller_dispatch.h: $(srcdir)/rpc/gendispatch.pl \ - $(REMOTE_PROTOCOL) + $(REMOTE_PROTOCOL) Makefile.am $(AM_V_GEN)$(PERL) -w $(srcdir)/rpc/gendispatch.pl \ -b virLXCMonitor VIR_LXC_MONITOR $(LXC_MONITOR_PROTOCOL) > $@ @@ -474,6 +474,8 @@ EXTRA_DIST += \ BUILT_SOURCES += $(LXC_GENERATED) +CLEANFILES += $(LXC_GENERATED) + LXC_DRIVER_SOURCES = \ $(LXC_MONITOR_PROTOCOL_GENERATED) \ $(LXC_MONITOR_GENERATED) \ -- 1.8.1.4

2 1

[libvirt] [PATCH V9 0/3] Add support for QEMU file descriptor sets
by Stefan Berger 14 Mar '13

14 Mar '13

The following patch series adds initial support for QEMU file descriptor sets implementing support for creating the proper command line. Some devices are using the sets now. Regards, Stefan

3 11

[libvirt] [PATCH] Fix size probing for VDI images
by Ján Tomko 14 Mar '13

14 Mar '13

Commit 027bf2ea used the wrong offset: the text field at the start of the header has 64 bytes, not 68. [1] Bug: https://bugzilla.redhat.com/show_bug.cgi?id=921452 [1] https://forums.virtualbox.org/viewtopic.php?p=29267#p29267 --- Relevant for v1.0.3-maint. src/util/virstoragefile.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/util/virstoragefile.c b/src/util/virstoragefile.c index cbcab5c..aabb5c8 100644 --- a/src/util/virstoragefile.c +++ b/src/util/virstoragefile.c @@ -199,7 +199,7 @@ static struct FileTypeInfo const fileTypeInfo[] = { [VIR_STORAGE_FILE_VDI] = { 64, "\x7f\x10\xda\xbe", ".vdi", LV_LITTLE_ENDIAN, 68, 0x00010001, - 68 + 5 * 4 + 256 + 7 * 4, 8, 1, -1, NULL}, + 64 + 5 * 4 + 256 + 7 * 4, 8, 1, -1, NULL}, /* Not direct file formats, but used for various drivers */ [VIR_STORAGE_FILE_FAT] = { 0, NULL, NULL, LV_LITTLE_ENDIAN, -- 1.8.1.5

2 2

[libvirt] Libvirt + LXC terminals
by Purcareata Bogdan-B43198 14 Mar '13

14 Mar '13

Hello all, I am looking for a way to start a container with libvirt and bind multiple consoles to it. My container is running busybox, the filesystem created using the LXC package (0.8.0). My libvirt version is 0.10.1. Configuration of the filesystem: - in the container rootfs, /dev contains 2 devices - /dev/console and /dev/tty1 - the inittab file looks like this: ::sysinit:/etc/init.d/rcS tty1::respawn:/bin/getty -L tty1 115200 vt100 console::askfirst:/bin/sh - it will link the /bin/sh process to /dev/console and the /bin/getty process to /dev/tty - when starting the container with lxc-start, each of these processes gets his own local /dev/pts device: 25609 ? Ss 0:00 lxc-start -n foo -d 25612 ? Ss 0:00 \_ init 25684 ? Ss 0:00 \_ /bin/syslogd 25686 pts/3 Ss+ 0:00 \_ /bin/getty -L tty1 115200 vt100 25687 pts/4 Ss+ 0:00 \_ init Here is the container.xml file I use to run with libvirt (you may ignore the console devices, since I don't know how much sense they make right now): <domain type='lxc'> <name>foo</name> <memory>500000</memory> <os> <type>exe</type> <init>/bin/init</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="/usr/var/lib/lxc/foo/rootfs" /> <target dir="/" /> </filesystem> <interface type="bridge"> <source bridge="br0" /> <name>eth0</name> <ip address="10.171.73.128" netmask="255.255.255.0" /> </interface> <console type='pty' tty='/dev/pts/3'> <source path='/dev/pts/3' /> <target path='/dev/console' /> </console> <serial type='pty' tty='/dev/pts/4'> <source path='/dev/pts/4' /> <target path='/dev/tty1' /> </serial> </devices> </domain> When starting the container with virsh start foo, the firt init process gets a local /dev/pts device, and it is inherited by its children: [root@everest][~]# ps axf 25858 ? Ss 0:00 /usr/libexec/libvirt_lxc --name foo --console 20 --security=selinux --handshake 23 --background --veth veth1 25859 pts/0 Ss+ 0:00 \_ init 25870 ? Ss 0:00 \_ /bin/syslogd 25872 ? Ss 0:00 \_ /bin/getty -L tty1 115200 vt100 25873 ? Ss 0:00 \_ init [root@everest][~]# ls -l /proc/25872/fd total 0 lrwx------. 1 root root 64 Mar 14 07:09 0 -> /dev/pts/0 lrwx------. 1 root root 64 Mar 14 07:09 1 -> /dev/pts/0 lrwx------. 1 root root 64 Mar 14 07:09 2 -> /dev/pts/0 [root@everest][~]# ls -l /proc/25873/fd total 0 lrwx------. 1 root root 64 Mar 14 07:09 0 -> /dev/pts/0 lrwx------. 1 root root 64 Mar 14 07:09 1 -> /dev/pts/0 lrwx------. 1 root root 64 Mar 14 07:09 2 -> /dev/pts/0 What I tried to do in the .xml, and I would need, is a way to bind a local /dev/pts devices to child processes in the container. Explicitly: - bind /dev/pts/3 to <container_rootfs>/dev/console - bind /dev/pts/4 to <container_rootfs>/dev/tty1 I've found some documentation at [1], hence my trials in the presented .xml. Is my scenario possible / supported? Thank you very much, Bogdan P. [1] http://libvirt.org/formatdomain.html#elementsConsole

2 2

[libvirt] [PATCHv6] virtio-rng: Add rate limiting options for virtio-RNG
by Peter Krempa 14 Mar '13

14 Mar '13

Qemu's implementation of virtio RNG supports rate limiting of the entropy used. This patch exposes the option to tune this functionality. This patch is based on qemu commit 904d6f588063fb5ad2b61998acdf1e73fb4 The rate limiting is exported in the XML as: <devices> ... <rng model='virtio'> <rate period='1234'>4321</rate> <backend model='random'/> </rng> ... --- Notes: Version 6: - use just attributes in XML instead of mixed attribute/contents - switch rate unit to bytes - tweak the rest to comply with the changes Version 5: - adjust after forbiding arbitrary name passthrough - rebase to master Version 4: - Reword docs - state it is available since 1.0.4 as the tree is frozen and this was actually never acked before Version 3: - State the time unit in docs Version 2: - qemu uses bytes/period, adapt the value according to that docs/formatdomain.html.in | 13 +++++++++++++ docs/schemas/domaincommon.rng | 21 ++++++++++++++++++++- src/conf/domain_conf.c | 19 +++++++++++++++++++ src/conf/domain_conf.h | 2 ++ src/qemu/qemu_command.c | 8 ++++++++ .../qemuxml2argv-virtio-rng-random.args | 2 +- .../qemuxml2argv-virtio-rng-random.xml | 1 + 7 files changed, 64 insertions(+), 2 deletions(-) diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index 35b47f2..1ef80b0 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -4294,6 +4294,7 @@ qemu-kvm -net nic,model=? /dev/null ... <devices> <rng model='virtio'> + <rate period="2000" bytes="1234"/> <backend model='random'>/dev/random</backend>  <backend model='egd' type='udp'> @@ -4316,6 +4317,18 @@ qemu-kvm -net nic,model=? /dev/null <li>'virtio' — supported by qemu and virtio-rng kernel module</li> </ul> </dd> + <dt><code>rate</code></dt> + <dd> + + The optional <code>rate</code> element allows limiting the rate at + which entropy can be consumed from the source. The mandatory + attribute <code>bytes</code> specifies how many bytes are permitted + to be consumed per period. An optional <code>period</code> attribute + specifies the duration of a period in milliseconds; if omitted, the + period is taken as 1000 milliseconds (1 second). + Since 1.0.4 + + </dd> <dt><code>backend</code></dt> <dd> diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index c40263c..9792065 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -3522,7 +3522,12 @@ <value>virtio</value> </choice> </attribute> - <ref name="rng-backend"/> + <interleave> + <ref name="rng-backend"/> + <optional> + <ref name="rng-rate"/> + </optional> + </interleave> </element> </define> @@ -3550,6 +3555,20 @@ </element> </define> + <define name="rng-rate"> + <element name="rate"> + <attribute name="bytes"> + <ref name="positiveInteger"/> + </attribute> + <optional> + <attribute name="period"> + <ref name="positiveInteger"/> + </attribute> + </optional> + <empty/> + </element> + </define> + <define name="usbmaster"> <element name="master"> <attribute name="startport"> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 95d2ff2..3278e9c 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -7570,6 +7570,19 @@ virDomainRNGDefParseXML(const xmlNodePtr node, ctxt->node = node; + if (virXPathUInt("string(./rate/@bytes)", ctxt, &def->rate) < -1) { + virReportError(VIR_ERR_XML_ERROR, "%s", + _("invalid RNG rate bytes value")); + goto error; + } + + if (def->rate > 0 && + virXPathUInt("string(./rate/@period)", ctxt, &def->period) < -1) { + virReportError(VIR_ERR_XML_ERROR, "%s", + _("invalid RNG rate period value")); + goto error; + } + if ((nbackends = virXPathNodeSet("./backend", ctxt, &backends)) < 0) goto error; @@ -13897,6 +13910,12 @@ virDomainRNGDefFormat(virBufferPtr buf, const char *backend = virDomainRNGBackendTypeToString(def->backend); virBufferAsprintf(buf, " <rng model='%s'>\n", model); + if (def->rate) { + virBufferAsprintf(buf, " <rate bytes='%u'", def->rate); + if (def->period) + virBufferAsprintf(buf, " period='%u'", def->period); + virBufferAddLit(buf, "/>\n"); + } virBufferAsprintf(buf, " <backend model='%s'", backend); switch ((enum virDomainRNGBackend) def->backend) { diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index 0fe43c5..96f11ba 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -1736,6 +1736,8 @@ enum virDomainRNGBackend { struct _virDomainRNGDef { int model; int backend; + unsigned int rate; /* bytes per period */ + unsigned int period; /* milliseconds */ union { char *file; /* file name for 'random' source */ diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index e7f2325..78c5cfc 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -4492,6 +4492,14 @@ qemuBuildRNGDeviceArgs(virCommandPtr cmd, virBufferAsprintf(&buf, "virtio-rng-pci,rng=%s", dev->info.alias); + if (dev->rate > 0) { + virBufferAsprintf(&buf, ",max-bytes=%u", dev->rate); + if (dev->period) + virBufferAsprintf(&buf, ",period=%u", dev->period); + else + virBufferAddLit(&buf, ",period=1000"); + } + if (qemuBuildDeviceAddressStr(&buf, &dev->info, qemuCaps) < 0) goto cleanup; diff --git a/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.args b/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.args index 7ab9dbc..5d296e8 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.args +++ b/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.args @@ -3,4 +3,4 @@ LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test /usr/bin/qemu \ -monitor unix:/tmp/test-monitor,server,nowait -no-acpi -boot c -usb \ -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x3 \ -object rng-random,id=rng0,filename=/dev/hwrng \ --device virtio-rng-pci,rng=rng0,bus=pci.0,addr=0x4 +-device virtio-rng-pci,rng=rng0,max-bytes=123,period=1234,bus=pci.0,addr=0x4 diff --git a/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.xml b/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.xml index 1e2c4be..354ae42 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.xml +++ b/tests/qemuxml2argvdata/qemuxml2argv-virtio-rng-random.xml @@ -17,6 +17,7 @@ <controller type='usb' index='0'/> <memballoon model='virtio'/> <rng model='virtio'> + <rate bytes='123' period='1234'/> <backend model='random'>/dev/hwrng</backend> </rng> </devices> -- 1.8.1.5

2 2

[libvirt] [PATCH v7 0/3] DEVICE_DELETED event
by Michael S. Tsirkin 14 Mar '13

14 Mar '13

libvirt has a long-standing bug: when removing the device, it can request removal but does not know when the removal completes. Add an event so we can fix this in a robust way. First patch only adds the event with ID, second patch adds a path field. Split this way for ease of backport (stable downstreams without QOM would want to only take the first patch). Event without fields is still useful as management can use it to poll device list to figure out which device was removed. Signed-off-by: Michael S. Tsirkin <mst(a)redhat.com> Changes from v6: - make empty event use data: {}, Markus prefers this Changes from v5: - Emit an empty event on unnamed devices in patch 1/3, as suggested by Markus Changes from v4: - Add extra triggers and extra fields as requested by Markus Changes from v3: - Document that we only emit events for devices with and ID, as suggested by Markus Changes from v2: - move event toward the end of device_unparent, so that parents are reported after their children, as suggested by Paolo Changes from v1: - move to device_unparent - address comments by Andreas and Eric -- Anthony Liguori Michael S. Tsirkin (2): qdev: DEVICE_DELETED event qom: pass original path to unparent method Stefan Hajnoczi (1): exec: assert that RAMBlock size is non-zero QMP/qmp-events.txt | 18 ++++++++++++++++++ exec.c | 2 ++ hw/qdev.c | 15 +++++++++++++-- include/monitor/monitor.h | 1 + include/qom/object.h | 3 ++- monitor.c | 1 + qapi-schema.json | 4 +++- qom/object.c | 4 +++- 8 files changed, 43 insertions(+), 5 deletions(-) -- MST

1 1

[libvirt] [PATCH] json: add stream parser
by Dmitry Guryanov 14 Mar '13

14 Mar '13

Add function virJSONValueFromStream, which reads data from a stream and passes it to json parser. When end of the object is reached, it returns this object. To avoid reading from the stream by single bytes it reads to a buffer (in a structure virJSONStreamParserState), which should be passed to a consequent call of this function. So if the end of one object and the beginning of the next object have been read by a single system call - virJSONValueFromStream handle it correctly. example of usage: virJSONValuePtr v; virJSONStreamParserState state; memset(&state, 0, sizeof(state)); while (1) { v = virJSONValueFromStream(mon->fd, &state); if (v == (void *)-1) /* error */ break; if (v == NULL) /* file descriptor has been closed */ break; /* handle object 'v' */ } I need such function for the parallels driver. It caches info about domains and needs some mechanism to update this cache. There is a "prlsrvctl monitor" command which waits for events forever and prints info about events to stdout in json format. So parallels driver could start separate thread which will read from prlsrvctl's stdout and update cache accordingly. There is the same task in qemu_monitor_json, but each json object is printed in a separate line there. It's not possible in my case, because some fields could have line endings. Signed-off-by: Dmitry Guryanov <dguryanov(a)parallels.com> --- src/util/virjson.c | 110 ++++++++++++++++++++++++++++++++++++++++++++++++++++ src/util/virjson.h | 8 ++++ 2 files changed, 118 insertions(+), 0 deletions(-) diff --git a/src/util/virjson.c b/src/util/virjson.c index e6a3b1b..d5c5a8c 100644 --- a/src/util/virjson.c +++ b/src/util/virjson.c @@ -990,6 +990,116 @@ cleanup: return ret; } +/* + * Read single JSON object from the stream. Store data, which + * have already been read from the stream, but belongs to the + * next object to the virJSONStreamParserState structure. So that + * consequent call to this function will parse it and return that + * object. + * + * @fd: file descriptor, opened for reading + * @state: pointer to the structure with buffer for data, read from fd. + * + * Return (void *)-1 in case of error, NULL when eof reached, + * pointer to the virJSONValuePtr in case of success. + */ + +virJSONValuePtr virJSONValueFromStream(int fd, virJSONStreamParserStatePtr state) +{ + yajl_handle hand; + virJSONParser parser = { NULL, NULL, 0 }; + virJSONValuePtr value = (void *)-1; +# ifndef WITH_YAJL2 + yajl_parser_config cfg = { 1, 1 }; +# endif + ssize_t len; + int ret = 0; + bool done = false; + +# ifdef WITH_YAJL2 + hand = yajl_alloc(&parserCallbacks, NULL, &parser); + if (hand) { + yajl_config(hand, yajl_allow_comments, 1); + yajl_config(hand, yajl_dont_validate_strings, 0); + yajl_config(hand, yajl_allow_trailing_garbage, 1); + } +# else + hand = yajl_alloc(&parserCallbacks, &cfg, NULL, &parser); +# endif + if (!hand) { + virReportError(VIR_ERR_INTERNAL_ERROR, "%s", + _("Unable to create JSON parser")); + goto cleanup; + } + + do { + if (state->pos == 0 || strlen(&state->buf[state->pos]) == 0) { + state->pos = 0; + memset(state->buf, 0, sizeof(state->buf)); + len = read(fd, state->buf, sizeof(state->buf) - 1); + + if (len < 0) { + virReportSystemError(errno, _("cannot read from fd '%d'"), fd); + virJSONValueFree(parser.head); + goto cleanup; + } + + if (len == 0) { + value = NULL; + goto cleanup; + } + } + + for (;state->pos < strlen(state->buf); state->pos++) { + unsigned char *buf = (unsigned char *)&state->buf[state->pos]; + /* + * New yaml library has useful function yajl_get_bytes_consumed + * which allows parsing by larger chunks. But rhel-6 has 1.0.7 + * version, which doesn't have it. + */ + ret = yajl_parse(hand, buf, 1); +# ifdef WITH_YAJL2 + if (ret == 0 && yajl_get_bytes_consumed(hand) == 0) { + done = true; + /* state->pos points to the first symbol after current + * object */ + break; + } else if (ret != 0) { +# else + if (ret == 0) { + done = true; + /* state->pos points to the last symbol of the + * current object */ + state->pos++; + break; + } else if (ret != yajl_status_insufficient_data) { +# endif + unsigned char *errstr = yajl_get_error(hand, 1, buf, 1); + virReportError(VIR_ERR_INTERNAL_ERROR, + _("cannot parse json: %s"), + (const char*) errstr); + VIR_FREE(errstr); + virJSONValueFree(parser.head); + goto cleanup; + } + } + } while (!done); + + value = parser.head; + +cleanup: + yajl_free(hand); + + if (parser.nstate) { + int i; + for (i = 0 ; i < parser.nstate ; i++) { + VIR_FREE(parser.state[i].key); + } + } + + VIR_DEBUG("result=%p", parser.head); + return value; +} static int virJSONValueToStringOne(virJSONValuePtr object, yajl_gen g) diff --git a/src/util/virjson.h b/src/util/virjson.h index 67f4398..aeb25ef 100644 --- a/src/util/virjson.h +++ b/src/util/virjson.h @@ -48,6 +48,8 @@ typedef virJSONObjectPair *virJSONObjectPairPtr; typedef struct _virJSONArray virJSONArray; typedef virJSONArray *virJSONArrayPtr; +typedef struct _virJSONStreamParserState virJSONStreamParserState; +typedef virJSONStreamParserState *virJSONStreamParserStatePtr; struct _virJSONObjectPair { char *key; @@ -77,6 +79,11 @@ struct _virJSONValue { } data; }; +struct _virJSONStreamParserState { + char buf[1024]; + size_t pos; +}; + void virJSONValueFree(virJSONValuePtr value); virJSONValuePtr virJSONValueNewString(const char *data); @@ -134,5 +141,6 @@ int virJSONValueObjectAppendNull(virJSONValuePtr object, const char *key); virJSONValuePtr virJSONValueFromString(const char *jsonstring); char *virJSONValueToString(virJSONValuePtr object, bool pretty); +virJSONValuePtr virJSONValueFromStream(int fd, virJSONStreamParserStatePtr state); #endif /* __VIR_JSON_H_ */ -- 1.7.1

2 2

[libvirt] [PATCH] Prevent streams from becoming libvirtd controlling TTY
by Daniel P. Berrange 14 Mar '13

14 Mar '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> When opening a stream to a device which is a TTY, that device may become the controlling TTY of libvirtd, if libvirtd was daemonized. This in turn means when the other end of the stream closes, libvirtd gets SIGHUP, causing it to reload its config. Prevent this by forcing O_NOCTTY on all streams that are opened Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- src/fdstream.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/fdstream.c b/src/fdstream.c index cc2dfe9..9a6f042 100644 --- a/src/fdstream.c +++ b/src/fdstream.c @@ -582,6 +582,8 @@ virFDStreamOpenFileInternal(virStreamPtr st, VIR_DEBUG("st=%p path=%s oflags=%x offset=%llu length=%llu mode=%o", st, path, oflags, offset, length, mode); + oflags |= O_NOCTTY; + if (oflags & O_CREAT) fd = open(path, oflags, mode); else -- 1.8.1.4

2 2

[libvirt] [libvirt-sandbox][PATCH] Fix copy&paste error in autogen.sh
by Alex Jia 14 Mar '13

14 Mar '13

Signed-off-by: Alex Jia <ajia(a)redhat.com> --- autogen.sh | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/autogen.sh b/autogen.sh index 86cd496..4b760fa 100755 --- a/autogen.sh +++ b/autogen.sh @@ -12,7 +12,7 @@ DIE=0 (autoconf --version) < /dev/null > /dev/null 2>&1 || { echo - echo "You must have autoconf installed to compile virt-viewer." + echo "You must have autoconf installed to compile libvirt-sandbox." echo "Download the appropriate package for your distribution," echo "or see http://www.gnu.org/software/autoconf" DIE=1 @@ -21,7 +21,7 @@ DIE=0 (automake --version) < /dev/null > /dev/null 2>&1 || { echo DIE=1 - echo "You must have automake installed to compile virt-viewer." + echo "You must have automake installed to compile libvirt-sandbox." echo "Download the appropriate package for your distribution," echo "or see http://www.gnu.org/software/automake" } -- 1.7.1

2 1

[libvirt] [libvirt-glib v2] gconfig: API for SPICE image compression
by Zeeshan Ali (Khattak) 14 Mar '13

14 Mar '13

From: "Zeeshan Ali (Khattak)" <zeeshanak(a)gnome.org> This patch adds API to set/get image compression configuration on domain/graphics[@type='spice'] nodes. Also included are simple tests for this API. --- V2: * No separate class for 'image' node. * Use trigraph extensions of glib-mkenums to assign desired nicks to enums rather than replacing '-' with '-' and viceversa at runtime. .../libvirt-gconfig-domain-graphics-spice.c | 39 +++++++++++++++++++++- .../libvirt-gconfig-domain-graphics-spice.h | 19 ++++++++++- libvirt-gconfig/libvirt-gconfig.sym | 8 +++++ libvirt-gconfig/tests/test-domain-create.c | 7 ++++ 4 files changed, 71 insertions(+), 2 deletions(-) diff --git a/libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.c b/libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.c index d090a3a..dc21b98 100644 --- a/libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.c +++ b/libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.c @@ -17,10 +17,12 @@ * License along with this library; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA * - * Author: Christophe Fergeau <cfergeau(a)gmail.com> + * Authors: Christophe Fergeau <cfergeau(a)gmail.com> + * Zeeshan Ali (Khattak) <zeeshanak(a)gnome.org> */ #include <config.h> +#include <string.h> #include "libvirt-gconfig/libvirt-gconfig.h" #include "libvirt-gconfig/libvirt-gconfig-private.h" @@ -122,3 +124,38 @@ void gvir_config_domain_graphics_spice_set_tls_port(GVirConfigDomainGraphicsSpic "tlsPort", G_TYPE_INT, port, NULL); } + +/** + * gvir_config_domain_graphics_spice_get_image_compression: + * @graphics: a #GVirConfigDomainGraphicsSpice + * + * Returns: (type GVirConfigDomainGraphicsSpiceImageCompression): image + * compression configuration of @graphics + */ +int +gvir_config_domain_graphics_spice_get_image_compression(GVirConfigDomainGraphicsSpice *graphics) +{ + g_return_val_if_fail(GVIR_CONFIG_IS_DOMAIN_GRAPHICS_SPICE(graphics), + GVIR_CONFIG_DOMAIN_GRAPHICS_SPICE_IMAGE_COMPRESSION_OFF); + + return gvir_config_object_get_attribute_genum + (GVIR_CONFIG_OBJECT(graphics), + "image", + "compression", + GVIR_CONFIG_TYPE_DOMAIN_GRAPHICS_SPICE_IMAGE_COMPRESSION, + GVIR_CONFIG_DOMAIN_GRAPHICS_SPICE_IMAGE_COMPRESSION_GLZ); +} + +void gvir_config_domain_graphics_spice_set_image_compression + (GVirConfigDomainGraphicsSpice *graphics, + GVirConfigDomainGraphicsSpiceImageCompression compression) +{ + g_return_if_fail(GVIR_CONFIG_IS_DOMAIN_GRAPHICS_SPICE(graphics)); + + gvir_config_object_add_child_with_attribute_enum + (GVIR_CONFIG_OBJECT(graphics), + "image", + "compression", + GVIR_CONFIG_TYPE_DOMAIN_GRAPHICS_SPICE_IMAGE_COMPRESSION, + compression); +} diff --git a/libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.h b/libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.h index c82615b..7e70cd4 100644 --- a/libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.h +++ b/libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.h @@ -17,7 +17,8 @@ * License along with this library; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA * - * Author: Christophe Fergeau <cfergeau(a)gmail.com> + * Authors: Christophe Fergeau <cfergeau(a)gmail.com> + * Zeeshan Ali (Khattak) <zeeshanak(a)gnome.org> */ #if !defined(__LIBVIRT_GCONFIG_H__) && !defined(LIBVIRT_GCONFIG_BUILD) @@ -56,6 +57,15 @@ struct _GVirConfigDomainGraphicsSpiceClass gpointer padding[20]; }; +typedef enum { + GVIR_CONFIG_DOMAIN_GRAPHICS_SPICE_IMAGE_COMPRESSION_AUTO_GLZ, /*< nick=auto_glz >*/ + GVIR_CONFIG_DOMAIN_GRAPHICS_SPICE_IMAGE_COMPRESSION_AUTO_LZ, /*< nick=auto_lz >*/ + GVIR_CONFIG_DOMAIN_GRAPHICS_SPICE_IMAGE_COMPRESSION_QUIC, + GVIR_CONFIG_DOMAIN_GRAPHICS_SPICE_IMAGE_COMPRESSION_GLZ, + GVIR_CONFIG_DOMAIN_GRAPHICS_SPICE_IMAGE_COMPRESSION_LZ, + GVIR_CONFIG_DOMAIN_GRAPHICS_SPICE_IMAGE_COMPRESSION_OFF +} GVirConfigDomainGraphicsSpiceImageCompression; + GType gvir_config_domain_graphics_spice_get_type(void); GVirConfigDomainGraphicsSpice *gvir_config_domain_graphics_spice_new(void); @@ -75,6 +85,13 @@ void gvir_config_domain_graphics_spice_set_port(GVirConfigDomainGraphicsSpice *g void gvir_config_domain_graphics_spice_set_tls_port(GVirConfigDomainGraphicsSpice *graphics, int port); +void gvir_config_domain_graphics_spice_set_image_compression + (GVirConfigDomainGraphicsSpice *graphics, + GVirConfigDomainGraphicsSpiceImageCompression compression); +int +gvir_config_domain_graphics_spice_get_image_compression + (GVirConfigDomainGraphicsSpice *graphics); + G_END_DECLS #endif /* __LIBVIRT_GCONFIG_DOMAIN_GRAPHICS_SPICE_H__ */ diff --git a/libvirt-gconfig/libvirt-gconfig.sym b/libvirt-gconfig/libvirt-gconfig.sym index d9cff90..a1b2cc1 100644 --- a/libvirt-gconfig/libvirt-gconfig.sym +++ b/libvirt-gconfig/libvirt-gconfig.sym @@ -504,4 +504,12 @@ LIBVIRT_GCONFIG_0.1.5 { gvir_config_domain_smartcard_passthrough_set_source; } LIBVIRT_GCONFIG_0.1.4; +LIBVIRT_GCONFIG_0.1.6 { + global: + gvir_config_domain_graphics_spice_get_image_compression; + gvir_config_domain_graphics_spice_set_image_compression; + + gvir_config_domain_graphics_spice_image_compression_get_type; +} LIBVIRT_GCONFIG_0.1.5; + # .... define new API here using predicted next version number .... diff --git a/libvirt-gconfig/tests/test-domain-create.c b/libvirt-gconfig/tests/test-domain-create.c index 4c94b2a..20fedc5 100644 --- a/libvirt-gconfig/tests/test-domain-create.c +++ b/libvirt-gconfig/tests/test-domain-create.c @@ -288,6 +288,13 @@ int main(int argc, char **argv) graphics = gvir_config_domain_graphics_spice_new(); gvir_config_domain_graphics_spice_set_port(graphics, 1234); g_assert(gvir_config_domain_graphics_spice_get_port(graphics) == 1234); + + /* SPICE image compression configuration */ + gvir_config_domain_graphics_spice_set_image_compression + (graphics, GVIR_CONFIG_DOMAIN_GRAPHICS_SPICE_IMAGE_COMPRESSION_AUTO_LZ); + g_assert(gvir_config_domain_graphics_spice_get_image_compression(graphics) == + GVIR_CONFIG_DOMAIN_GRAPHICS_SPICE_IMAGE_COMPRESSION_AUTO_LZ); + devices = g_list_append(devices, GVIR_CONFIG_DOMAIN_DEVICE(graphics)); /* video node */ -- 1.8.1.4

2 1

[libvirt] Problems with live migration
by Toni F. [ackstorm] 14 Mar '13

14 Mar '13

Hi all, I have working with Openstack Folsom and with Glusterfs as shared storage for /instances. All working fine, but when i'm trying to use "nova live-migration": 2013-03-07 18:33:42 3140 ERROR nova.virt.libvirt.driver [-] [instance: 773164b5-5e5c-4328-a762-d91f50f2ac33] Live Migration failure: Timed out during operation: cannot acquire state change lock I know that the real command that openstack executes virsh migrate --live instance-00000036 qemu+ssh://nova-compute-lnx001/system When i execute that command the same error is showed error: Timed out during operation: cannot acquire state change lock Versions: ii libvirt-bin 0.9.13-0ubuntu12.2~cloud0 programs for the libvirt library ii libvirt0 0.9.13-0ubuntu12.2~cloud0 library for interfacing with different virtualization systems ii python-libvirt 0.9.13-0ubuntu12.2~cloud0 libvirt Python bindings ii qemu-kvm 1.0+noroms-0ubuntu14.7 Full virtualization on i386 and amd64 hardware Any idea about this error? Thanks for advance Regards

1 0

[libvirt] Problems with live migration
by Toni F. [ackstorm] 14 Mar '13

14 Mar '13

Hi all, I have working with Openstack Folsom and with Glusterfs as shared storage for /instances. All working fine, but when i'm trying to use "nova live-migration": 2013-03-07 18:33:42 3140 ERROR nova.virt.libvirt.driver [-] [instance: 773164b5-5e5c-4328-a762-d91f50f2ac33] Live Migration failure: Timed out during operation: cannot acquire state change lock I know that the real command that openstack executes virsh migrate --live instance-00000036 qemu+ssh://nova-compute-lnx001/system When i execute that command the same error is showed error: Timed out during operation: cannot acquire state change lock Versions: ii libvirt-bin 0.9.13-0ubuntu12.2~cloud0 programs for the libvirt library ii libvirt0 0.9.13-0ubuntu12.2~cloud0 library for interfacing with different virtualization systems ii python-libvirt 0.9.13-0ubuntu12.2~cloud0 libvirt Python bindings ii qemu-kvm 1.0+noroms-0ubuntu14.7 Full virtualization on i386 and amd64 hardware Any idea about this error? Thanks for advance Regards

1 0

[libvirt] [PATCHv4 0/5] S390: Support for native CCW bus
by Viktor Mihajlovski 14 Mar '13

14 Mar '13

Originally, QEMU did not implement a native I/O bus for s390. The initial implementation had a machine type 's390-virtio' featuring a fully paravirtualized I/O system with an artificial bus type 'virtio-s390'. This bus had a number of short-comings, like the need for a non-standard device discovery mechanism, a limited number of devices, limited hotplugging capabilites and the lack of persistent device addresses. To resolve these issues a new machine type 's390-ccw-virtio' has been recently added to QEMU which implementa the native s390 CCW I/O bus. Guests with arch='s390x' and machine='s390-ccw-virtio' can now be defined with up to 262144 virtio devices. This series adds the support for the s390-ccw-virtio machine type and the CCW bus to libvirt. As usual we start with the documentation/schema, the generic configuration stuff, continue with the QEMU driver including device hotplug and finally add qemu2xml testcases. V2 Changes - use an attribute triple cssid, ssid, schid instead of devno to better represent the libvirt CCW address structure - rebase to current upstream, mainly address qemuCapsXXX rename V3 Changes - skip already acked patches 1/5 and 2/5 - revert machine-based capabilities - use machine definition from domain object V4 Changes - schid was technically wrong, the attribute name has to be devno: changed all occurrences accordingly - re-introduce reverted patches 1/5 and 2/5 with schid/devno correction - rebase to current upstream J.B. Joret (1): S390: Add hotplug support for s390 virtio devices Viktor Mihajlovski (4): S390: Documentation for CCW address type S390: domain_conf support for CCW S390: QEMU driver support for CCW addresses S390: Testcases for virtio-ccw machines docs/formatdomain.html.in | 14 + docs/schemas/domaincommon.rng | 52 ++++ src/conf/domain_conf.c | 107 +++++++- src/conf/domain_conf.h | 16 ++ src/libvirt_private.syms | 1 + src/qemu/qemu_capabilities.c | 7 +- src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 279 ++++++++++++++++++-- src/qemu/qemu_command.h | 6 + src/qemu/qemu_domain.c | 1 + src/qemu/qemu_domain.h | 3 + src/qemu/qemu_driver.c | 6 +- src/qemu/qemu_hotplug.c | 152 +++++++---- src/qemu/qemu_hotplug.h | 14 +- src/qemu/qemu_process.c | 3 + .../qemuxml2argv-console-virtio-ccw.args | 10 + .../qemuxml2argv-console-virtio-ccw.xml | 27 ++ .../qemuxml2argv-console-virtio-s390.args | 4 +- .../qemuxml2argv-console-virtio-s390.xml | 2 +- .../qemuxml2argv-disk-virtio-ccw-many.args | 12 + .../qemuxml2argv-disk-virtio-ccw-many.xml | 40 +++ .../qemuxml2argv-disk-virtio-ccw.args | 8 + .../qemuxml2argv-disk-virtio-ccw.xml | 30 +++ .../qemuxml2argv-net-virtio-ccw.args | 8 + .../qemuxml2argv-net-virtio-ccw.xml | 30 +++ tests/qemuxml2argvtest.c | 12 +- tests/testutilsqemu.c | 3 +- 27 files changed, 766 insertions(+), 82 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-console-virtio-ccw.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-console-virtio-ccw.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-ccw-many.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-ccw-many.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-ccw.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-ccw.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-ccw.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-ccw.xml -- 1.7.9.5

4 14