- Devel - libvirt List Archives

[libvirt PATCH 0/4] Code style documentation
by Tim Wiederhake 15 Jan '22

15 Jan '22

Some additions and clarifications to libvirt's code style documentation, based on points of feedback that are given regularly on the mailing list. Tim Wiederhake (4): docs: coding-style: Clarify on virXXXPtr types docs: coding-style: Rewrite section on shortening comparisons docs: coding-style: Remove "no_memory" as acceptable goto target docs: coding-style: One variable declaration per line docs/coding-style.rst | 79 ++++++++++++++++++++++++++++++------------- 1 file changed, 56 insertions(+), 23 deletions(-) -- 2.31.1

6 15

[libvirt PATCH 0/3] Bump some min versions
by Daniel P. Berrangé 14 Jan '22

14 Jan '22

Given previous platforms we've dropped, we can bump the min compilers. Daniel P. Berrangé (3): configure: bump min required GCC to 7.4.0 configure: bump min required CLang to 6.0 / XCode 10.0 examples: drop some conditionals checks from macros config.h | 14 +++++++------- examples/c/misc/event-test.c | 14 ++------------ 2 files changed, 9 insertions(+), 19 deletions(-) -- 2.33.1

2 4

[RFC 0/1] qemu: update index for serial device using taget.port
by divya 14 Jan '22

14 Jan '22

Issue ----- The port being provided in the xml file of the domain is not being used for the creation of qemu command. On adding the serial device : <serial> <target type='serial' port='3'/> </serial> Generated qemu command will look like : /usr/libexec/qemu-kvm ...\ -device isa-serial,chardev=charserial0,id=serial0 Actually it should be : /usr/libexec/qemu-kvm ...\ -device isa-serial,chardev=charserial0,id=serial0,index=3 Patch ----- Out already for the correction : https://listman.redhat.com/archives/libvir-list/2018-April/msg02302.html This patch was not followed up. According to me there are multiple reasons Reasons for not following up ---------------------------- Index : specifies the index number of a connector port. If not specified, the index is automatically incremented. This logic exists both on qemu as well as libvirt. https://github.com/qemu/qemu/blob/master/hw/char/serial-isa.c#L62 Issue 1: If we want two isa-serial devices and for the first one is we mention the port to be 3, then for the next device it not automatically assign the port number 4 which will throw the following error : error: internal error: process exited while connecting to monitor: 2021-11-12T11:05:31.169987Z qemu-kvm: -device isa-serial,chardev=charserial2,id=serial2,index=5: Max. supported number of ISA serial ports is 4. But we are left with 3 ports (0,1,2) which are unused. So ideally we should have used them. Issue 2: It is possible that two devices get the same port address which might lead to a lot of ambiguity. Example: we want two devices and for the second one we provide the index 0. Then from default logic the first device will be allotted port 0 and the second device will overwrite it and get port 0. Solution : ---------- Port allocation logic 1. Precedence should be given to serial devices as we only have the first 4 ports for them. 1.1. Check the command line/xml file, scan for all the devices mentioned and then start with the isa-serial devices for port allocation. 2.Maintain a buffer(bitmap) for marking the allocated ports. 3.While assigning a port to the device 3.1. If no port is provided by the user : provide the next available port. 3.2. Else check: 3.2.1. If the port is already allocated : throw the error. 3.2.2. Else allocate the port. 3.3. If out of ports : throw error -> qemu throws the error. Libvirt manages the port numbers with the similar logic(auto increment) along with the above mentioned bug. Hence need to add the above patch along with the Port allocation logic. root (1): update index for serial device using taget.port src/conf/domain_conf.c | 70 +++++++++++++++++-- src/qemu/qemu_command.c | 20 ++++-- tests/qemuhotplugtest.c | 1 - ...g-console-compat-2-live+console-virtio.xml | 4 +- .../qemuhotplug-console-compat-2-live.xml | 4 +- tests/qemuxml2argvdata/bios.args | 2 +- .../qemuxml2argvdata/console-compat-auto.args | 2 +- .../console-compat-chardev.args | 2 +- tests/qemuxml2argvdata/console-compat.args | 2 +- .../qemuxml2argvdata/console-virtio-many.args | 2 +- tests/qemuxml2argvdata/controller-order.args | 2 +- .../name-escape.x86_64-2.11.0.args | 4 +- tests/qemuxml2argvdata/name-escape.xml | 1 + .../q35-virt-manager-basic.args | 2 +- .../serial-dev-chardev-iobase.args | 2 +- .../qemuxml2argvdata/serial-dev-chardev.args | 2 +- .../qemuxml2argvdata/serial-file-chardev.args | 2 +- tests/qemuxml2argvdata/serial-file-log.args | 2 +- .../qemuxml2argvdata/serial-many-chardev.args | 4 +- .../qemuxml2argvdata/serial-pty-chardev.args | 2 +- tests/qemuxml2argvdata/serial-spiceport.args | 2 +- .../qemuxml2argvdata/serial-tcp-chardev.args | 2 +- .../serial-tcp-telnet-chardev.args | 2 +- .../serial-tcp-tlsx509-chardev-notls.args | 8 ++- .../serial-tcp-tlsx509-chardev-notls.xml | 18 ++++- .../serial-tcp-tlsx509-chardev-verify.args | 4 +- .../serial-tcp-tlsx509-chardev-verify.xml | 2 +- .../serial-tcp-tlsx509-chardev.args | 4 +- .../serial-tcp-tlsx509-chardev.xml | 2 +- .../serial-tcp-tlsx509-secret-chardev.args | 4 +- .../serial-tcp-tlsx509-secret-chardev.xml | 2 +- .../qemuxml2argvdata/serial-udp-chardev.args | 4 +- .../qemuxml2argvdata/serial-unix-chardev.args | 4 +- .../serial-unix-chardev.x86_64-latest.args | 4 +- tests/qemuxml2argvdata/serial-vc-chardev.args | 2 +- tests/qemuxml2argvdata/user-aliases.args | 4 +- .../virtio-9p-createmode.x86_64-latest.args | 2 +- .../virtio-9p-multidevs.x86_64-latest.args | 2 +- .../x86_64-pc-graphics.x86_64-latest.args | 2 +- .../x86_64-pc-headless.x86_64-latest.args | 2 +- .../x86_64-q35-graphics.x86_64-latest.args | 2 +- .../x86_64-q35-headless.x86_64-latest.args | 2 +- .../serial-tcp-tlsx509-chardev.xml | 2 +- 43 files changed, 149 insertions(+), 65 deletions(-) -- 2.27.0

9 53

[PATCH] tests: Report expected monitor command for simulated commands
by Michal Privoznik 14 Jan '22

14 Jan '22

There are two tests currently that simulate QMP talk: qemucapabilitiestest and qemuhotplugtest. In both cases they check whether currently executed command is the one for which reply was provided. If not an error message is reported. However, the error message contains only the actual command and not the expected one. This makes it harder to navigate through .replies files. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- tests/qemumonitortestutils.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/qemumonitortestutils.c b/tests/qemumonitortestutils.c index f7a0a37685..073ae5d73a 100644 --- a/tests/qemumonitortestutils.c +++ b/tests/qemumonitortestutils.c @@ -648,7 +648,8 @@ qemuMonitorTestProcessCommandVerbatim(qemuMonitorTest *test, ret = qemuMonitorTestAddResponse(test, data->response); } else { if (data->cmderr) { - qemuMonitorTestError("%s: %s", data->cmderr, cmdstr); + qemuMonitorTestError("%s: %s expected %s", + data->cmderr, cmdstr, data->command_name); } else { qemuMonitorTestErrorInvalidCommand(data->command_name, cmdstr); } -- 2.34.1

2 1

[PATCH partially-for-8.0 00/17] qemu: Fix use-after free when redefining snapshots and cleanup the code
by Peter Krempa 14 Jan '22

14 Jan '22

Patches 1 and 2 should be pushed for 8.0 as the bug was introduced in this dev cycle and the patches are specifically kept very simple. The rest of the series refactors the snapshot validation and helper code to have less weird semantics which lead to this bug. Peter Krempa (17): qemuSnapshotRedefine: Rename 'def' to 'snapdef' qemuSnapshotRedefine: Fix use of snapshot definition after free virDomainMomentAssignDef: Simplify error handling virDomainSnapshotRedefineValidate: Fix validation of VIR_DOMAIN_SNAPSHOT_CREATE_DISK_ONLY flag virDomainSnapshotAlignDisks: Improve function comment virDomainSnapshotAlignDisks: Convert @default_snapshot to virDomainSnapshotLocation virDomainSnapshotAlignDisks: Move 'require_match' selection logic inside virDomainSnapshotAlignDisks: Allow alternate domain definition when redefining virDomainSnapshotRedefineValidate: Unexport virDomainSnapshotRedefinePrep: Use 'snapdef' for snapshot definition object virDomainSnapshotRedefineValidate: Don't modify the snapshot definition testDomainSnapshotCreateXML: Extract snapshot redefinition code qemuSnapshotCreate: Use 'snapdef' instead of 'def' qemuSnapshotCreate: Standardize handling of the reference on @snapdef qemuDomainSnapshotLoad: Refactor handling of snapshot definition object virDomainSnapshotAssignDef: Clear second argument when it is consumed virDomainSnapshotRedefinePrep: Don't do partial redefine src/conf/snapshot_conf.c | 120 +++++++++++++++------------- src/conf/snapshot_conf.h | 13 +-- src/conf/virdomainmomentobjlist.c | 9 +-- src/conf/virdomainsnapshotobjlist.c | 29 ++++++- src/conf/virdomainsnapshotobjlist.h | 5 +- src/libvirt_private.syms | 1 + src/qemu/qemu_driver.c | 18 ++--- src/qemu/qemu_snapshot.c | 35 ++++---- src/test/test_driver.c | 89 ++++++++++++--------- src/vz/vz_sdk.c | 3 +- 10 files changed, 180 insertions(+), 142 deletions(-) -- 2.31.1

2 21

[PATCH 00/11] Misc cleanups
by Michal Privoznik 14 Jan '22

14 Jan '22

Couple of cleanups I've done whilst looking around our code base. Michal Prívozník (11): storage_file: Declare virStorageSourceParseRBDColonString only in one header virconf: Report an error in when virConfSetValue() fails xen_xl: Check for virConfSetValue() retval src: Declare and use g_autoptr(virConfValue) virconf: Make virConfSetValue() clear consumed pointer libxl: Don't use a static buffer in xenParseXLVnuma() libxl: Allocate @libxldisk in xenParseXLDisk() on stack xen_xl.c: Use g_autofree more xen_xl.c: Use g_autoptr() for virCPUDef libxl: Remove needless labels virsh: Remove needless labels src/libxl/xen_common.c | 62 ++--- src/libxl/xen_xl.c | 259 ++++++------------ src/libxl/xen_xm.c | 18 +- src/storage_file/storage_source.h | 5 - .../storage_source_backingstore.h | 3 +- src/util/virconf.c | 19 +- src/util/virconf.h | 3 +- tools/virsh-host.c | 20 +- tools/vsh.c | 11 +- 9 files changed, 135 insertions(+), 265 deletions(-) -- 2.34.1

2 14

Release of libvirt-8.0.0
by Jiri Denemark 14 Jan '22

14 Jan '22

The 8.0.0 release of both libvirt and libvirt-python is tagged and signed tarballs and source RPMs are available at https://libvirt.org/sources/ https://libvirt.org/sources/python/ Thanks everybody who helped with this release by sending patches, reviewing, testing, or providing feedback. Your work is greatly appreciated. * Security * libxl: Fix potential deadlock and crash (CVE-2021-4147) A rogue guest could continuously reboot itself and cause libvirtd on the host to deadlock or crash, resulting in a denial of service condition. * Removed features * qemu: Explicitly forbid live changing nodeset for strict numatune For ``strict`` mode of <numatune/> it can't be guaranteed that memory is moved completely onto new set of nodes (e.g. QEMU might have locked pieces of its memory) thus breaking the strict promise. If live migration of QEMU memory between NUMA nodes is desired, users are advised to use ``restrictive`` mode instead. * New features * qemu: Synchronous write mode for disk copy operations The ``blockdev-mirror`` block job supports a mode where writes from the VM are synchronously propagated to the destination of the copy. This ensures that the job will converge under heavy I/O. Implement the mode for the copy blockjob as ``VIR_DOMAIN_BLOCK_COPY_SYNCHRONOUS_WRITES`` flag exposed via ``virsh blockcopy --synchronous-writes`` and for non-shared storage migration as ``VIR_MIGRATE_NON_SHARED_SYNCHRONOUS_WRITES`` exposed via ``virsh migrate --copy-storage-synchronous-writes``. * Introduce TCG domain features Libvirt is now able to set the size of translation block cache size (tb-size) for TCG domains. * qemu: Add new API to inject a launch secret in a domain New API ``virDomainSetLaunchSecurityState()`` and virsh command ``domsetlaunchsecstate`` are added to support injecting a launch secret in a domain's memory. * Improvements * libxl: Implement the virDomainGetMessages API * qemu: Preserve qcow2 sub-cluster allocation state after external snapshots and block-copy The new image which is installed as an overlay on top of the current chain when taking an external snapshot, or the target of a block copy operation now enables sub-cluster allocation (``extended_l2``) if the original image has the option enabled. * Bug fixes * qemu: Fix device hot-unplug with ``libvirt-7.9`` or ``libvirt-7.10`` used with ``qemu-6.2`` An internal change to the configuration format used by the above libvirt versions triggers a bug in ``qemu-6.2`` where qemu no longer emits the event notifying that the device was unplugged successfully and thus libvirt never removes the device from the definition. This impacts only devices which were present at startup of the VM, hotplugged devices behave correctly. This is fixed in ``libvirt-8.0`` by reverting to the old configuration approach until qemu is fixed. As a workaround for ``libvirt-7.9`` and ``libvirt-7.10`` the old configuration approach can be forced by: Option 1, global ``qemu.conf``:: capability_filters = [ "device.json" ] Option 2, per VM XML override:: <domain type='kvm' xmlns:qemu='http://libvirt.org/schemas/domain/qemu/1.0'> [...] <qemu:capabilities> <qemu:del capability='device.json'/> </qemu:capabilities> </domain> * Fix sparse streams with split daemon In split daemon scenario, a client connected to a hypervisor driver and using sparse streams (e.g. ``virsh vol-download --sparse``) would make the hypervisor daemon enter an infinite loop without any data transfer. This is now fixed. * Build no longer requires RPC library Code and its cross dependencies were fixed so that build without remote driver and thus an RPC library (like ``tirpc``) fails no more. * virnetdevopenvswitch: Fix 'burst' value passed to ovs-vsctl When a ``<bandwidth/>`` was defined for a TAP device that's plugged into an OvS bridge values passed to the OvS were incorrectly recalculated resulting in slightly different limits being applied. Enjoy. Jirka

1 0

[libvirt PATCH 00/14] Random fixes and improvements
by Tim Wiederhake 13 Jan '22

13 Jan '22

A collection of drive-by fixes and improvements that quickly derailed into the g_auto-ification of src/conf/interface_conf.c. Tim Wiederhake (14): docs: coding-style: Clarify on virXXXPtr types virQEMUCapsSEVInfoCopy: Remove superfluous g_auto usage virInterfaceDefDevFormat: Add missing error handling conf: interface: Preparation for g_auto conf: interface: Use g_auto conf: interface: Remove ret and goto virInterfaceDefParseXML: Inline trivial virInterfaceDefParseName virInterfaceDefParseIP: Simplify and cleanup virInterfaceDefParseDhcp: Simplify and cleanup virInterfaceDefParseProtoIPv4: Simplify and cleanup virInterfaceDefParseProtoIPv6: Simplify and cleanup virInterfaceDefParseIfAdressing: Simplify and cleanup virInterfaceDefParseXML: Simplify and cleanup virInterfaceDefParse: Simplify and cleanup docs/coding-style.rst | 5 + src/conf/interface_conf.c | 357 ++++++++++++----------------------- src/qemu/qemu_capabilities.c | 4 +- 3 files changed, 128 insertions(+), 238 deletions(-) -- 2.31.1

2 19

[libvirt PATCH for-8.0.0] virt-ssh-helper: Move manual page to section 8
by Andrea Bolognani 13 Jan '22

13 Jan '22

While sections are somewhat loosely defined and thus the choice is not quite a clear-cut one, section 8 might be a slightly better fit in this case. Suggested-by: Olaf Hering <olaf(a)aepfle.de> Signed-off-by: Andrea Bolognani <abologna(a)redhat.com> --- We could also merge this after 8.0.0, but I think it would be a little bit nicer if the manual page didn't move around between releases. docs/manpages/meson.build | 2 +- docs/manpages/virt-ssh-helper.rst | 2 +- libvirt.spec.in | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/docs/manpages/meson.build b/docs/manpages/meson.build index 6763d19af8..150f45d296 100644 --- a/docs/manpages/meson.build +++ b/docs/manpages/meson.build @@ -19,11 +19,11 @@ docs_man_files = [ { 'name': 'virt-pki-query-dn', 'section': '1', 'install': true }, { 'name': 'virt-pki-validate', 'section': '1', 'install': true }, { 'name': 'virt-qemu-run', 'section': '1', 'install': conf.has('WITH_QEMU') }, - { 'name': 'virt-ssh-helper', 'section': '1', 'install': conf.has('WITH_LIBVIRTD') }, { 'name': 'virt-xml-validate', 'section': '1', 'install': true }, { 'name': 'libvirtd', 'section': '8', 'install': conf.has('WITH_LIBVIRTD') }, { 'name': 'virt-sanlock-cleanup', 'section': '8', 'install': conf.has('WITH_SANLOCK') }, + { 'name': 'virt-ssh-helper', 'section': '8', 'install': conf.has('WITH_LIBVIRTD') }, { 'name': 'virtbhyved', 'section': '8', 'install': conf.has('WITH_BHYVE') }, { 'name': 'virtinterfaced', 'section': '8', 'install': conf.has('WITH_INTERFACE') }, { 'name': 'virtlockd', 'section': '8', 'install': conf.has('WITH_LIBVIRTD') }, diff --git a/docs/manpages/virt-ssh-helper.rst b/docs/manpages/virt-ssh-helper.rst index 175ddacaec..9d0cf10bc3 100644 --- a/docs/manpages/virt-ssh-helper.rst +++ b/docs/manpages/virt-ssh-helper.rst @@ -6,7 +6,7 @@ virt-ssh-helper libvirt socket proxy -------------------- -:Manual section: 1 +:Manual section: 8 :Manual group: Virtualization Support .. contents:: diff --git a/libvirt.spec.in b/libvirt.spec.in index 7a90a1a9e6..db9d622699 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -1693,7 +1693,7 @@ exit 0 %{_mandir}/man1/virt-admin.1* %{_mandir}/man1/virt-host-validate.1* -%{_mandir}/man1/virt-ssh-helper.1* +%{_mandir}/man8/virt-ssh-helper.8* %{_mandir}/man8/libvirtd.8* %{_mandir}/man8/virtlogd.8* %{_mandir}/man8/virtlockd.8* -- 2.31.1

2 2

[PATCH] Print errors on daemon startup if logging setup fails
by Martin Kletzander 13 Jan '22

13 Jan '22

https://bugzilla.redhat.com/show_bug.cgi?id=2039652 Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- src/locking/lock_daemon.c | 4 +++- src/logging/log_daemon.c | 4 +++- src/remote/remote_daemon.c | 4 +++- 3 files changed, 9 insertions(+), 3 deletions(-) diff --git a/src/locking/lock_daemon.c b/src/locking/lock_daemon.c index ea81940a4325..75ebd059d79a 100644 --- a/src/locking/lock_daemon.c +++ b/src/locking/lock_daemon.c @@ -919,8 +919,10 @@ int main(int argc, char **argv) { config->log_outputs, privileged, verbose, - godaemon) < 0) + godaemon) < 0) { + virDispatchError(NULL); exit(EXIT_FAILURE); + } if (!pid_file && virPidFileConstructPath(privileged, diff --git a/src/logging/log_daemon.c b/src/logging/log_daemon.c index fe7fa8534aec..8c3eb66895dc 100644 --- a/src/logging/log_daemon.c +++ b/src/logging/log_daemon.c @@ -725,8 +725,10 @@ int main(int argc, char **argv) { config->log_outputs, privileged, verbose, - godaemon) < 0) + godaemon) < 0) { + virDispatchError(NULL); exit(EXIT_FAILURE); + } if (!pid_file && virPidFileConstructPath(privileged, diff --git a/src/remote/remote_daemon.c b/src/remote/remote_daemon.c index 84157e6cc19a..1b8e982a2f01 100644 --- a/src/remote/remote_daemon.c +++ b/src/remote/remote_daemon.c @@ -940,8 +940,10 @@ int main(int argc, char **argv) { config->log_outputs, privileged, verbose, - godaemon) < 0) + godaemon) < 0) { + virDispatchError(NULL); exit(EXIT_FAILURE); + } /* Let's try to initialize global variable that holds the host's boot time. */ if (virHostBootTimeInit() < 0) { -- 2.34.1

2 1