September 2018 - Devel - libvirt List Archives

[libvirt] [PATCH 00/13] Implement vsh-table API to virsh and virt-admin
by Simon Kobyda 21 Sep '18

21 Sep '18

API is implemented in virsh and virt-admin. It fixes problems with table-alignment, makes tables more readable and deals with unicode. Table however could not be implemented in places, which do use simple lists, or table implementation would change content of output, which users might rely on. If you know about other place where it could be implemented, let me know. https://bugzilla.redhat.com/show_bug.cgi?id=1574624 https://bugzilla.redhat.com/show_bug.cgi?id=1584630 Simon Kobyda (13): virsh: Implement vsh-table to iface-list virsh: Implement vshTable API to net-list and net-dhcp-leases virsh: Implement vshTable API to secret-list virsh: Implement vshTable API to nwfilter-list and nwfilterbinding-list virsh: Implement vshTable API to snapshot-list. virsh: Set up cmdDomblkinfo() and cmdDomblkinfoPrint() for vshTable API implementation virsh: Implement vshTable API to domblkinfo virsh: Implement vshTable API to domblklist virsh: Implement vshTable API to domiflist virsh: Implement vshTable API to vcpupin, iothreadinfo, domfsinfo virsh: Implement vshTable API to pool-list virsh: Implement vshTable API to vol-list virt-admin: Implement vshTable API to server-list and client-list tools/virsh-domain-monitor.c | 131 +++++++++++++++------------- tools/virsh-domain.c | 95 +++++++++++++++------ tools/virsh-interface.c | 27 ++++-- tools/virsh-network.c | 55 ++++++++---- tools/virsh-nwfilter.c | 47 +++++++--- tools/virsh-pool.c | 160 +++++++---------------------------- tools/virsh-secret.c | 30 +++++-- tools/virsh-snapshot.c | 33 +++++--- tools/virsh-volume.c | 127 ++++++--------------------- tools/virt-admin.c | 46 +++++++--- 10 files changed, 367 insertions(+), 384 deletions(-) -- 2.17.1

3 19

[libvirt] [PATCH 0/2] qemu: Drop QEMU_CAPS_DEVICE_SCSI_GENERIC
by Andrea Bolognani 21 Sep '18

21 Sep '18

2 3

[libvirt] [PATCH v2] virsh: Don't break loop of domblkinfo for disks
by Han Han 21 Sep '18

21 Sep '18

https://bugzilla.redhat.com/show_bug.cgi?id=1619625 --all option is added to cmdDomblkinfo since commit 62c39193 allowing to show all block devices info. Reset error when empty source in case error breaks the loop of domblkinfo for disks. Signed-off-by: Han Han <hhan(a)redhat.com> --- tools/virsh-domain-monitor.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/tools/virsh-domain-monitor.c b/tools/virsh-domain-monitor.c index b9b4f9739b..576610f005 100644 --- a/tools/virsh-domain-monitor.c +++ b/tools/virsh-domain-monitor.c @@ -475,6 +475,7 @@ cmdDomblkinfo(vshControl *ctl, const vshCmd *cmd) int ndisks; size_t i; xmlNodePtr *disks = NULL; + char *source = NULL; char *target = NULL; char *protocol = NULL; @@ -505,16 +506,18 @@ cmdDomblkinfo(vshControl *ctl, const vshCmd *cmd) for (i = 0; i < ndisks; i++) { ctxt->node = disks[i]; + source = virXPathString("string(./source)", ctxt); protocol = virXPathString("string(./source/@protocol)", ctxt); target = virXPathString("string(./target/@dev)", ctxt); rc = virDomainGetBlockInfo(dom, target, &info, 0); if (rc < 0) { - /* If protocol is present that's an indication of a networked - * storage device which cannot provide statistics, so generate - * 0 based data and get the next disk. */ - if (protocol && !active && + /* For the case of empty cdrom, networked disk which cannot + * provide statistics, generate 0 based data and get the next + * disk. + */ + if (!source && protocol && !active && virGetLastErrorCode() == VIR_ERR_INTERNAL_ERROR && virGetLastErrorDomain() == VIR_FROM_STORAGE) { memset(&info, 0, sizeof(info)); @@ -526,6 +529,7 @@ cmdDomblkinfo(vshControl *ctl, const vshCmd *cmd) cmdDomblkinfoPrint(ctl, &info, target, human, false); + VIR_FREE(source); VIR_FREE(target); VIR_FREE(protocol); } @@ -540,6 +544,7 @@ cmdDomblkinfo(vshControl *ctl, const vshCmd *cmd) cleanup: virshDomainFree(dom); + VIR_FREE(source); VIR_FREE(target); VIR_FREE(protocol); VIR_FREE(disks); -- 2.18.0

3 2

[libvirt] [PATCH 0/2] Couple of vhostuser fixes
by Michal Privoznik 21 Sep '18

21 Sep '18

Almost trivial ones. Michal Prívozník (2): qemu_hotplug: Fetch vhostuser ifname on hotplug virsh: Display vhostuser socket path in domiflist src/qemu/qemu_hotplug.c | 5 +++++ tools/virsh-domain-monitor.c | 3 ++- 2 files changed, 7 insertions(+), 1 deletion(-) -- 2.16.4

2 4

[libvirt] [QEMU PATCH v2 0/4] Deprecation patches net and enforce-config-section
by Thomas Huth 21 Sep '18

21 Sep '18

Deprecate the "name" parameter for -net, and "enforce-config-section" of -machine. v2: - Split the single net patch into multiple smaller patches - Add patch to deprecate enforce-config-section Thomas Huth (4): Makefile: Add missing dependency for qemu-deprecated.texi net: Deprecate the "name" parameter of -net net/slirp: Deprecate the [hub_id name] parameter tuple hw/core/machine: Officially deprecate the enforce-config-section parameter Makefile | 2 +- hw/core/machine.c | 3 +++ net/net.c | 4 ++++ net/slirp.c | 2 ++ qemu-deprecated.texi | 17 +++++++++++++++++ 5 files changed, 27 insertions(+), 1 deletion(-) -- 1.8.3.1

4 13

[libvirt] [QEMU PATCH] net: Deprecate the old way of using a legacy net via "name" instead of "id"
by Thomas Huth 21 Sep '18

21 Sep '18

In early times, network backends were specified by a "vlan" and "name" tuple. With the introduction of netdevs, the "name" was replaced by an "id" (which is supposed to be unique), but the "name" parameter stayed as an alias which could be used instead of "id". Unfortunately, we miss the duplication check for "name": $ qemu-system-x86_64 -net user,name=n1 -net user,name=n1 ... starts without an error, while "id" correctly complains: $ qemu-system-x86_64 -net user,id=n1 -net user,id=n1 qemu-system-x86_64: -net user,id=n1: Duplicate ID 'n1' for net Instead of trying to fix the code for the legacy "name" parameter, let's rather get rid of this old interface and deprecate the "name" parameter now - this will also be less confusing for the users in the long run. While we're at it, also deprecate the old syntax for the hostfwd_add and hostfwd_remove commands that still work with this legacy "hub" plus "name" tuple. It is enough to specify the netdev id there instead. Also add a missing dependency to the Makefile to make sure that the docs get correctly regenerated when qemu-deprecated.texi is changed. Signed-off-by: Thomas Huth <thuth(a)redhat.com> --- Makefile | 2 +- net/net.c | 4 ++++ net/slirp.c | 2 ++ qemu-deprecated.texi | 13 +++++++++++++ 4 files changed, 20 insertions(+), 1 deletion(-) diff --git a/Makefile b/Makefile index fe623e4..f42e176 100644 --- a/Makefile +++ b/Makefile @@ -978,7 +978,7 @@ txt: qemu-doc.txt docs/interop/qemu-qmp-ref.txt docs/interop/qemu-ga-ref.txt qemu-doc.html qemu-doc.info qemu-doc.pdf qemu-doc.txt: \ qemu-img.texi qemu-nbd.texi qemu-options.texi qemu-option-trace.texi \ - qemu-monitor.texi qemu-img-cmds.texi qemu-ga.texi \ + qemu-deprecated.texi qemu-monitor.texi qemu-img-cmds.texi qemu-ga.texi \ qemu-monitor-info.texi docs/qemu-block-drivers.texi \ docs/qemu-cpu-models.texi diff --git a/net/net.c b/net/net.c index 2a31339..cdcd5cf 100644 --- a/net/net.c +++ b/net/net.c @@ -984,6 +984,10 @@ static int net_client_init1(const void *object, bool is_netdev, Error **errp) /* missing optional values have been initialized to "all bits zero" */ name = net->has_id ? net->id : net->name; + if (net->has_name) { + warn_report("The 'name' parameter is deprecated, use 'id' instead"); + } + /* Map the old options to the new flat type */ switch (opts->type) { case NET_LEGACY_OPTIONS_TYPE_NONE: diff --git a/net/slirp.c b/net/slirp.c index c18060f..073bc69 100644 --- a/net/slirp.c +++ b/net/slirp.c @@ -404,6 +404,8 @@ static SlirpState *slirp_lookup(Monitor *mon, const char *hub_id, monitor_printf(mon, "unrecognized (hub-id, stackname) pair\n"); return NULL; } + warn_report("Using 'hub-id' is deprecated. Specify the netdev id " + "directly instead"); } else { nc = qemu_find_netdev(name); if (!nc) { diff --git a/qemu-deprecated.texi b/qemu-deprecated.texi index 8a2e399..eed7132 100644 --- a/qemu-deprecated.texi +++ b/qemu-deprecated.texi @@ -83,6 +83,11 @@ The 'file' driver for drives is no longer appropriate for character or host devices and will only accept regular files (S_IFREG). The correct driver for these file types is 'host_cdrom' or 'host_device' as appropriate. +@subsection -net ...,name=@var{name} (since 3.1) + +The @option{name} parameter of the @option{-net} option is an old synonym +for the @option{id} parameter which should now be used instead. + @section QEMU Machine Protocol (QMP) commands @subsection block-dirty-bitmap-add "autoload" parameter (since 2.12.0) @@ -99,6 +104,14 @@ The ``query-cpus'' command is replaced by the ``query-cpus-fast'' command. The ``arch'' output member of the ``query-cpus-fast'' command is replaced by the ``target'' output member. +@section System emulator human monitor commands + +@subsection The hub_id parameter of 'hostfwd_add' / 'hostfwd_remove' (since 3.1) + +The @option{hub_id} parameter of the 'hostfwd_add' and 'hostfwd_remove' HMP +commands is redundant. It is enough to specify the netdev ID of the backend +that should be changed. + @section System emulator devices @subsection ivshmem (since 2.6.0) -- 1.8.3.1

4 5

[libvirt] [libvirt PATCH v7 0/6] Finish the conversion to virConfGetValue* functions
by Fabiano Fidêncio 20 Sep '18

20 Sep '18

This patchset finishes the conversion to virConfGetValue* functions, started by Daniel Berrange a few months ago. Please, mind that although we could make virConfGetValue* functions more generic in order to support numbers and booleans as strings, that doesn't seem the safest path to take. The side-effect of this is that we will have to live with some specific code doing that as part of vmx and xen_common. Once this patchset gets merged, https://wiki.libvirt.org/page/BiteSizedTasks#Finish_conversion_to_virConfGe… can be removed. - Changes since v1: All the "values" from virConfGetValueString() are freed - Changes since v2: All comments from Ján Tomko have been addressed; A few leaks were (possibly) found and they're addressed in the last patch of the series; - Changes since v3: All comments from Ján Tomko have been addressed in order to lower the non-whitespace changes in the first patch; - Changes since v4: All comments from Ján Tomko have been addressed (hopefully, they actually were this time :-)); - Changes since v5: All comments from John Ferlan have been addressed, including splitting the patches per "function" touched. Some new patches were also needed due to a behaviour change introduced by the "xen_common: Change xenParsePCIList to using virConfGetValueStringList" patch. - Changes since v6: Actually properly building the code /o\ (John, --with-libxl is what you need!) Squashed all the patches fixing the leak with the patch that introduces the leak Followed John's suggestions and comments in order to try to keep the new checks as close as possible to the old ones. Fabiano Fidêncio (6): xen_common: Change xenConfigCopyStringInternal to using virConfGetValueString xen_common: Change xenConfigGetUUID to using virConfGetValueString xen_common: Change xenConfigGetString to using virConfGetValueString xen_common: Change xenParsePCIList to using virConfGetValueStringList xen_common: Change xenParseVfbs to using virConfGetValueStringList xen_common: Change xenParseCharDev to using virConfGetValueStringList src/xenconfig/xen_common.c | 202 ++++++++++++++++++++----------------- src/xenconfig/xen_common.h | 2 +- src/xenconfig/xen_xl.c | 10 +- src/xenconfig/xen_xm.c | 7 +- 4 files changed, 120 insertions(+), 101 deletions(-) -- 2.17.1

2 14

[libvirt] [jenkins-ci PATCH] guests: Hash test user password dynamically
by Andrea Bolognani 20 Sep '18

20 Sep '18

Current versions of Ansible support the password_hash() filter, so we can avoid hardcoding a pre-computed hash and make what's happening a bit clearer. Signed-off-by: Andrea Bolognani <abologna(a)redhat.com> --- guests/playbooks/update/tasks/users.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/guests/playbooks/update/tasks/users.yml b/guests/playbooks/update/tasks/users.yml index 0a930d6..abaeaa4 100644 --- a/guests/playbooks/update/tasks/users.yml +++ b/guests/playbooks/update/tasks/users.yml @@ -28,7 +28,7 @@ - name: '{{ flavor }}: Set password' user: name: '{{ flavor }}' - password: '$6$xSlfvkcsDgPmRAMX$mFh9qRmFFW9cyW1n5/jeHvq4OmJA8WzSD70Mfis3VHc3Z5imZeiQAg9VNL4sFEtmDy/siU3nJL.QeAapCgfL20' + password: '{{ "test"|password_hash("sha512") }}' when: - flavor == 'test' -- 2.17.1

2 2

[libvirt] [PATCH] util: Fix travis build error
by John Ferlan 20 Sep '18

20 Sep '18

Commit 12093f1f used %ld instead of %zd for a size_t. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- Sigh, always something simple. Pushed under the build breaker rule. src/util/virresctrl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/util/virresctrl.c b/src/util/virresctrl.c index eddcc41b67..1019bb52cc 100644 --- a/src/util/virresctrl.c +++ b/src/util/virresctrl.c @@ -676,7 +676,7 @@ virResctrlGetMonitorInfo(virResctrlInfoPtr resctrl) } features = virStringSplitCount(featurestr, "\n", 0, &nfeatures); - VIR_DEBUG("Resctrl supported %ld monitoring features", nfeatures); + VIR_DEBUG("Resctrl supported %zd monitoring features", nfeatures); info_monitor->nfeatures = nfeatures; VIR_STEAL_PTR(info_monitor->features, features); -- 2.17.1

1 0

[libvirt] [PATCH v3] qemu: fix deadlock if create qemuProcessReconnect thread failed
by Wang Yechao 20 Sep '18

20 Sep '18

qemuProcessReconnectHelper has hold lock on doms, if create qemuProcessReconnect thread failed, calls qemuDomainRemoveInactiveJob will lead to deadlock. Add a qemuDomainRemoveInactiveJobLocked, modify qemuProcessReconnectHelper to call it. Signed-off-by: Wang Yechao <wang.yechao255(a)zte.com.cn> --- v2 patch: https://www.redhat.com/archives/libvir-list/2018-September/msg00635.html Changes in v3: - drop v2 patch, cause it is not good. - split qemuDomainRemoveInactive into qemuDomainRemoveInactiveCommon and virDomainObjListRemove. - create a qemuDomainRemoveInactiveLocked, consists of qemuDomainRemoveInactiveCommon and virDomainObjListRemoveLocked. - create a qemuDomainRemoveInactiveJobLocked, which copies qemuDomainRemoveInactiveJob except calling qemuDomainRemoveInactiveLocked - Modify qemuProcessReconnectHelper to call qemuDomainRemoveInactiveJobLocked --- src/qemu/qemu_domain.c | 75 ++++++++++++++++++++++++++++++++++++++++--------- src/qemu/qemu_domain.h | 9 ++++++ src/qemu/qemu_process.c | 2 +- 3 files changed, 71 insertions(+), 15 deletions(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index 2fd8a2a..ef0d91d 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -8343,24 +8343,13 @@ qemuDomainSnapshotDiscardAllMetadata(virQEMUDriverPtr driver, return rem.err; } - -/** - * qemuDomainRemoveInactive: - * - * The caller must hold a lock to the vm. - */ void -qemuDomainRemoveInactive(virQEMUDriverPtr driver, - virDomainObjPtr vm) +qemuDomainRemoveInactiveCommon(virQEMUDriverPtr driver, + virDomainObjPtr vm) { char *snapDir; virQEMUDriverConfigPtr cfg; - if (vm->persistent) { - /* Short-circuit, we don't want to remove a persistent domain */ - return; - } - cfg = virQEMUDriverGetConfig(driver); /* Remove any snapshot metadata prior to removing the domain */ @@ -8379,9 +8368,47 @@ qemuDomainRemoveInactive(virQEMUDriverPtr driver, } qemuExtDevicesCleanupHost(driver, vm->def); + virObjectUnref(cfg); +} + +/** + * qemuDomainRemoveInactive: + * + * The caller must hold a lock to the vm. + */ +void +qemuDomainRemoveInactive(virQEMUDriverPtr driver, + virDomainObjPtr vm) +{ + if (vm->persistent) { + /* Short-circuit, we don't want to remove a persistent domain */ + return; + } + + qemuDomainRemoveInactiveCommon(driver, vm); virDomainObjListRemove(driver->domains, vm); - virObjectUnref(cfg); +} + +/** + * qemuDomainRemoveInactiveLocked: + * + * The caller must hold a lock to the vm , + * and hold a lock to the doms. + */ + +void +qemuDomainRemoveInactiveLocked(virQEMUDriverPtr driver, + virDomainObjPtr vm) +{ + if (vm->persistent) { + /* Short-circuit, we don't want to remove a persistent domain */ + return; + } + + qemuDomainRemoveInactiveCommon(driver, vm); + virDomainObjListRemoveLocked(driver->domains, vm); + } @@ -8407,6 +8434,26 @@ qemuDomainRemoveInactiveJob(virQEMUDriverPtr driver, qemuDomainObjEndJob(driver, vm); } +/** + * qemuDomainRemoveInactiveJobLocked: + * + * Just like qemuDomainRemoveInactiveJob but it hold lock + * on 'doms'. + */ + +void +qemuDomainRemoveInactiveJobLocked(virQEMUDriverPtr driver, + virDomainObjPtr vm) +{ + bool haveJob; + + haveJob = qemuDomainObjBeginJob(driver, vm, QEMU_JOB_MODIFY) >= 0; + + qemuDomainRemoveInactiveLocked(driver, vm); + + if (haveJob) + qemuDomainObjEndJob(driver, vm); +} void qemuDomainSetFakeReboot(virQEMUDriverPtr driver, diff --git a/src/qemu/qemu_domain.h b/src/qemu/qemu_domain.h index 914c9a6..1d80621 100644 --- a/src/qemu/qemu_domain.h +++ b/src/qemu/qemu_domain.h @@ -698,12 +698,21 @@ int qemuDomainSnapshotDiscardAll(void *payload, int qemuDomainSnapshotDiscardAllMetadata(virQEMUDriverPtr driver, virDomainObjPtr vm); +void qemuDomainRemoveInactiveCommon(virQEMUDriverPtr driver, + virDomainObjPtr vm); + void qemuDomainRemoveInactive(virQEMUDriverPtr driver, virDomainObjPtr vm); +void qemuDomainRemoveInactiveLocked(virQEMUDriverPtr driver, + virDomainObjPtr vm); + void qemuDomainRemoveInactiveJob(virQEMUDriverPtr driver, virDomainObjPtr vm); +void qemuDomainRemoveInactiveJobLocked(virQEMUDriverPtr driver, + virDomainObjPtr vm); + void qemuDomainSetFakeReboot(virQEMUDriverPtr driver, virDomainObjPtr vm, bool value); diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 72a59de..ed24447 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -8025,7 +8025,7 @@ qemuProcessReconnectHelper(virDomainObjPtr obj, */ qemuProcessStop(src->driver, obj, VIR_DOMAIN_SHUTOFF_FAILED, QEMU_ASYNC_JOB_NONE, 0); - qemuDomainRemoveInactiveJob(src->driver, obj); + qemuDomainRemoveInactiveJobLocked(src->driver, obj); virDomainObjEndAPI(&obj); virNWFilterUnlockFilterUpdates(); -- 1.8.3.1

2 1