September 2023 - Devel - libvirt List Archives

[PATCH] virnetdevopenvswitch: Propagate OVS error messages
by Michal Privoznik 08 Sep '23

08 Sep '23

When configuring OVS interfaces/bridges we spawn 'ovs-vsctl' with appropriate arguments and if it exited with a non-zero status we report a generic error message, like "Unable to add port vnet0 to OVS bridge ovsbr0". This is all cool, but the real reason why operation failed is hidden in (debug) logs because that's where virCommandRun() reports it unless caller requested otherwise. This is a bit clumsy because then we have to ask users to turn on debug logs and reproduce the problem again, e.g. [1]. Therefore, in cases where an error is reported to the user - just read ovs-vsctl's stderr and include it in the error message. For other cases (like VIR_DEBUG/VIR_WARN) - well they are meant to end up in (debug) logs anyway. 1: https://mail.openvswitch.org/pipermail/ovs-discuss/2023-September/052640.ht… Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/util/virnetdevopenvswitch.c | 93 ++++++++++++++++++++------------- 1 file changed, 58 insertions(+), 35 deletions(-) diff --git a/src/util/virnetdevopenvswitch.c b/src/util/virnetdevopenvswitch.c index 8dad6ed2bd..d836d05845 100644 --- a/src/util/virnetdevopenvswitch.c +++ b/src/util/virnetdevopenvswitch.c @@ -54,10 +54,14 @@ virNetDevOpenvswitchSetTimeout(unsigned int timeout) } static virCommand * -virNetDevOpenvswitchCreateCmd(void) +virNetDevOpenvswitchCreateCmd(char **errbuf) { virCommand *cmd = virCommandNew(OVS_VSCTL); + virCommandAddArgFormat(cmd, "--timeout=%u", virNetDevOpenvswitchTimeout); + if (errbuf) + virCommandSetErrorBuffer(cmd, errbuf); + return cmd; } @@ -137,6 +141,7 @@ int virNetDevOpenvswitchAddPort(const char *brname, const char *ifname, char ifuuidstr[VIR_UUID_STRING_BUFLEN]; char vmuuidstr[VIR_UUID_STRING_BUFLEN]; g_autoptr(virCommand) cmd = NULL; + g_autofree char *errbuf = NULL; g_autofree char *attachedmac_ex_id = NULL; g_autofree char *ifaceid_ex_id = NULL; g_autofree char *profile_ex_id = NULL; @@ -157,7 +162,7 @@ int virNetDevOpenvswitchAddPort(const char *brname, const char *ifname, ovsport->profileID); } - cmd = virNetDevOpenvswitchCreateCmd(); + cmd = virNetDevOpenvswitchCreateCmd(&errbuf); virCommandAddArgList(cmd, "--", "--may-exist", "add-port", brname, ifname, NULL); @@ -185,8 +190,8 @@ int virNetDevOpenvswitchAddPort(const char *brname, const char *ifname, if (virCommandRun(cmd, NULL) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, - _("Unable to add port %1$s to OVS bridge %2$s"), - ifname, brname); + _("Unable to add port %1$s to OVS bridge %2$s: %3$s"), + ifname, brname, NULLSTR(errbuf)); return -1; } @@ -203,13 +208,15 @@ int virNetDevOpenvswitchAddPort(const char *brname, const char *ifname, */ int virNetDevOpenvswitchRemovePort(const char *brname G_GNUC_UNUSED, const char *ifname) { - g_autoptr(virCommand) cmd = virNetDevOpenvswitchCreateCmd(); + g_autofree char *errbuf = NULL; + g_autoptr(virCommand) cmd = virNetDevOpenvswitchCreateCmd(&errbuf); virCommandAddArgList(cmd, "--", "--if-exists", "del-port", ifname, NULL); if (virCommandRun(cmd, NULL) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, - _("Unable to delete port %1$s from OVS"), ifname); + _("Unable to delete port %1$s from OVS: %2$s"), + ifname, NULLSTR(errbuf)); return -1; } @@ -228,7 +235,8 @@ int virNetDevOpenvswitchRemovePort(const char *brname G_GNUC_UNUSED, const char int virNetDevOpenvswitchGetMigrateData(char **migrate, const char *ifname) { size_t len; - g_autoptr(virCommand) cmd = virNetDevOpenvswitchCreateCmd(); + g_autofree char *errbuf = NULL; + g_autoptr(virCommand) cmd = virNetDevOpenvswitchCreateCmd(&errbuf); virCommandAddArgList(cmd, "--if-exists", "get", "Interface", ifname, "external_ids:PortData", NULL); @@ -238,8 +246,8 @@ int virNetDevOpenvswitchGetMigrateData(char **migrate, const char *ifname) /* Run the command */ if (virCommandRun(cmd, NULL) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, - _("Unable to run command to get OVS port data for interface %1$s"), - ifname); + _("Unable to run command to get OVS port data for interface %1$s: %2$s"), + ifname, NULLSTR(errbuf)); return -1; } @@ -263,21 +271,22 @@ int virNetDevOpenvswitchGetMigrateData(char **migrate, const char *ifname) int virNetDevOpenvswitchSetMigrateData(char *migrate, const char *ifname) { g_autoptr(virCommand) cmd = NULL; + g_autofree char *errbuf = NULL; if (!migrate) { VIR_DEBUG("No OVS port data for interface %s", ifname); return 0; } - cmd = virNetDevOpenvswitchCreateCmd(); + cmd = virNetDevOpenvswitchCreateCmd(&errbuf); virCommandAddArgList(cmd, "set", "Interface", ifname, NULL); virCommandAddArgFormat(cmd, "external_ids:PortData=%s", migrate); /* Run the command */ if (virCommandRun(cmd, NULL) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, - _("Unable to run command to set OVS port data for interface %1$s"), - ifname); + _("Unable to run command to set OVS port data for interface %1$s: %2$s"), + ifname, NULLSTR(errbuf)); return -1; } @@ -371,7 +380,8 @@ int virNetDevOpenvswitchInterfaceStats(const char *ifname, virDomainInterfaceStatsPtr stats) { - g_autoptr(virCommand) cmd = virNetDevOpenvswitchCreateCmd(); + g_autofree char *errbuf = NULL; + g_autoptr(virCommand) cmd = virNetDevOpenvswitchCreateCmd(&errbuf); g_autofree char *output = NULL; virCommandAddArgList(cmd, "--if-exists", "--format=list", "--data=json", @@ -390,8 +400,9 @@ virNetDevOpenvswitchInterfaceStats(const char *ifname, if (virCommandRun(cmd, NULL) < 0 || STREQ_NULLABLE(output, "")) { /* no ovs-vsctl or interface 'ifname' doesn't exists in ovs */ - virReportError(VIR_ERR_INTERNAL_ERROR, "%s", - _("Interface not found")); + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Interface not found: %1$s"), + NULLSTR(errbuf)); return -1; } @@ -433,7 +444,8 @@ virNetDevOpenvswitchInterfaceStats(const char *ifname, int virNetDevOpenvswitchInterfaceGetMaster(const char *ifname, char **master) { - g_autoptr(virCommand) cmd = virNetDevOpenvswitchCreateCmd(); + g_autofree char *errbuf = NULL; + g_autoptr(virCommand) cmd = virNetDevOpenvswitchCreateCmd(&errbuf); int exitstatus; *master = NULL; @@ -443,8 +455,8 @@ virNetDevOpenvswitchInterfaceGetMaster(const char *ifname, char **master) if (virCommandRun(cmd, &exitstatus) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, - _("Unable to run command to get OVS master for interface %1$s"), - ifname); + _("Unable to run command to get OVS master for interface %1$s: %2$s"), + ifname, NULLSTR(errbuf)); return -1; } @@ -546,7 +558,7 @@ virNetDevOpenvswitchGetVhostuserIfname(const char *path, return 0; } - cmd = virNetDevOpenvswitchCreateCmd(); + cmd = virNetDevOpenvswitchCreateCmd(NULL); if (server) { virCommandAddArgList(cmd, "--no-headings", "--columns=name", "find", @@ -600,7 +612,8 @@ virNetDevOpenvswitchGetVhostuserIfname(const char *path, int virNetDevOpenvswitchUpdateVlan(const char *ifname, const virNetDevVlan *virtVlan) { - g_autoptr(virCommand) cmd = virNetDevOpenvswitchCreateCmd(); + g_autofree char *errbuf = NULL; + g_autoptr(virCommand) cmd = virNetDevOpenvswitchCreateCmd(&errbuf); virCommandAddArgList(cmd, "--", "--if-exists", "clear", "Port", ifname, "tag", @@ -614,7 +627,8 @@ int virNetDevOpenvswitchUpdateVlan(const char *ifname, if (virCommandRun(cmd, NULL) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, - _("Unable to set vlan configuration on port %1$s"), ifname); + _("Unable to set vlan configuration on port %1$s: %2$s"), + ifname, NULLSTR(errbuf)); return -1; } @@ -629,7 +643,7 @@ virNetDevOpenvswitchFindUUID(const char *table, g_autoptr(virCommand) cmd = NULL; char *uuid = NULL; - cmd = virNetDevOpenvswitchCreateCmd(); + cmd = virNetDevOpenvswitchCreateCmd(NULL); virCommandAddArgList(cmd, "--no-heading", "--columns=_uuid", "find", table, vmid_ex_id, ifname_ex_id, NULL); virCommandSetOutputBuffer(cmd, &uuid); @@ -673,7 +687,7 @@ virNetDevOpenvswitchInterfaceClearTxQos(const char *ifname, if (!*line) { continue; } - listcmd = virNetDevOpenvswitchCreateCmd(); + listcmd = virNetDevOpenvswitchCreateCmd(NULL); virCommandAddArgList(listcmd, "--no-heading", "--columns=_uuid", "--if-exists", "list", "port", ifname, "qos", NULL); virCommandSetOutputBuffer(listcmd, &port_qos); @@ -681,13 +695,13 @@ virNetDevOpenvswitchInterfaceClearTxQos(const char *ifname, VIR_WARN("Unable to remove port qos on port %s", ifname); } if (port_qos && *port_qos) { - g_autoptr(virCommand) cmd = virNetDevOpenvswitchCreateCmd(); + g_autoptr(virCommand) cmd = virNetDevOpenvswitchCreateCmd(NULL); virCommandAddArgList(cmd, "remove", "port", ifname, "qos", line, NULL); if (virCommandRun(cmd, NULL) < 0) { VIR_WARN("Unable to remove port qos on port %s", ifname); } } - destroycmd = virNetDevOpenvswitchCreateCmd(); + destroycmd = virNetDevOpenvswitchCreateCmd(NULL); virCommandAddArgList(destroycmd, "destroy", "qos", line, NULL); if (virCommandRun(destroycmd, NULL) < 0) { VIR_WARN("Unable to destroy qos on port %s", ifname); @@ -706,7 +720,7 @@ virNetDevOpenvswitchInterfaceClearTxQos(const char *ifname, continue; } - cmd = virNetDevOpenvswitchCreateCmd(); + cmd = virNetDevOpenvswitchCreateCmd(NULL); virCommandAddArgList(cmd, "destroy", "queue", line, NULL); if (virCommandRun(cmd, NULL) < 0) { VIR_WARN("Unable to destroy queue on port %s", ifname); @@ -722,15 +736,17 @@ static int virNetDevOpenvswitchInterfaceClearRxQos(const char *ifname) { g_autoptr(virCommand) cmd = NULL; + g_autofree char *errbuf = NULL; - cmd = virNetDevOpenvswitchCreateCmd(); + cmd = virNetDevOpenvswitchCreateCmd(&errbuf); virCommandAddArgList(cmd, "set", "Interface", ifname, NULL); virCommandAddArgFormat(cmd, "ingress_policing_rate=%llu", 0llu); virCommandAddArgFormat(cmd, "ingress_policing_burst=%llu", 0llu); if (virCommandRun(cmd, NULL) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, - _("Unable to reset ingress on port %1$s"), ifname); + _("Unable to reset ingress on port %1$s: %2$s"), + ifname, NULLSTR(errbuf)); return -1; } @@ -754,6 +770,7 @@ virNetDevOpenvswitchInterfaceSetTxQos(const char *ifname, { char vmuuidstr[VIR_UUID_STRING_BUFLEN]; g_autoptr(virCommand) cmd = NULL; + g_autofree char *errbuf = NULL; g_autofree char *vmid_ex_id = NULL; g_autofree char *ifname_ex_id = NULL; g_autofree char *average = NULL; @@ -777,7 +794,7 @@ virNetDevOpenvswitchInterfaceSetTxQos(const char *ifname, qos_uuid = virNetDevOpenvswitchFindUUID("qos", vmid_ex_id, ifname_ex_id); /* create qos and set */ - cmd = virNetDevOpenvswitchCreateCmd(); + cmd = virNetDevOpenvswitchCreateCmd(&errbuf); if (queue_uuid && *queue_uuid) { g_auto(GStrv) lines = g_strsplit(queue_uuid, "\n", 0); virCommandAddArgList(cmd, "set", "queue", lines[0], NULL); @@ -806,17 +823,20 @@ virNetDevOpenvswitchInterfaceSetTxQos(const char *ifname, if (virCommandRun(cmd, NULL) < 0) { if (queue_uuid && *queue_uuid) { virReportError(VIR_ERR_INTERNAL_ERROR, - _("Unable to set queue configuration on port %1$s"), ifname); + _("Unable to set queue configuration on port %1$s: %2$s"), + ifname, NULLSTR(errbuf)); } else { virReportError(VIR_ERR_INTERNAL_ERROR, - _("Unable to create and set qos configuration on port %1$s"), ifname); + _("Unable to create and set qos configuration on port %1$s: %2$s"), + ifname, NULLSTR(errbuf)); } return -1; } if (qos_uuid && *qos_uuid) { g_auto(GStrv) lines = g_strsplit(qos_uuid, "\n", 0); - g_autoptr(virCommand) qoscmd = virNetDevOpenvswitchCreateCmd(); + g_autofree char *qoserrbuf = NULL; + g_autoptr(virCommand) qoscmd = virNetDevOpenvswitchCreateCmd(&qoserrbuf); virCommandAddArgList(qoscmd, "set", "qos", lines[0], NULL); virCommandAddArgFormat(qoscmd, "other_config:min-rate=%s", average); @@ -829,7 +849,8 @@ virNetDevOpenvswitchInterfaceSetTxQos(const char *ifname, virCommandAddArgList(qoscmd, vmid_ex_id, ifname_ex_id, NULL); if (virCommandRun(qoscmd, NULL) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, - _("Unable to set qos configuration on port %1$s"), ifname); + _("Unable to set qos configuration on port %1$s: %2$s"), + ifname, NULLSTR(qoserrbuf)); return -1; } } @@ -842,8 +863,9 @@ virNetDevOpenvswitchInterfaceSetRxQos(const char *ifname, const virNetDevBandwidthRate *rx) { g_autoptr(virCommand) cmd = NULL; + g_autofree char *errbuf = NULL; - cmd = virNetDevOpenvswitchCreateCmd(); + cmd = virNetDevOpenvswitchCreateCmd(&errbuf); virCommandAddArgList(cmd, "set", "Interface", ifname, NULL); virCommandAddArgFormat(cmd, "ingress_policing_rate=%llu", rx->average * VIR_NETDEV_RX_TO_OVS); @@ -853,7 +875,8 @@ virNetDevOpenvswitchInterfaceSetRxQos(const char *ifname, if (virCommandRun(cmd, NULL) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, - _("Unable to set vlan configuration on port %1$s"), ifname); + _("Unable to set vlan configuration on port %1$s: %2$s"), + ifname, NULLSTR(errbuf)); return -1; } -- 2.41.0

2 1

Error : virHostCPUGetKVMMaxVCPUs:1228 : KVM is not supported on this platform: Function not implemented.
by Mario Marietto 08 Sep '23

08 Sep '23

Hello. I'm running Debian bookworm on my ARM Chromebook,model "xe303c12" and I've recompiled the kernel (5.4) to enable KVM,so now my system sounds like this : $ lsb_release -a No LSB modules are available. Distributor ID: Debian Description: Debian GNU/Linux 12 (bookworm) Release: 12 Codename: bookworm $ uname -a Linux chromarietto 5.4.244-stb-cbe #8 SMP PREEMPT Sat Aug 19 22:19:32 UTC 2023 armv7l GNU/Linux $ uname -r 5.4.244-stb-cbe $ kvm-ok INFO: /dev/kvm exists KVM acceleration can be used $ qemu-system-arm --version QEMU emulator version 5.1.0 (v5.1.0-dirty) Copyright (c) 2003-2020 Fabrice Bellard and the QEMU Project developers $ python3 --version Python 3.11.2 I have installed libvirt 9.7.0,qemu 5.1 and virt-manager from source code with the final goal to be able to connect qemu,kvm and libvirt together to virtualize FreeBSD 13.2 for arm 32 bit. Some useful informations about my platform : root@chromarietto:/home/marietto/Desktop# virt-host-validate QEMU: Checking if device /dev/kvm exists : PASS QEMU: Checking if device /dev/kvm is accessible : PASS QEMU: Checking if device /dev/vhost-net exists : PASS QEMU: Checking if device /dev/net/tun exists : PASS QEMU: Checking for cgroup 'cpu' controller support : PASS QEMU: Checking for cgroup 'cpuacct' controller support : PASS QEMU: Checking for cgroup 'cpuset' controller support : PASS QEMU: Checking for cgroup 'memory' controller support : PASS QEMU: Checking for cgroup 'devices' controller support : PASS QEMU: Checking for cgroup 'blkio' controller support : PASS QEMU: Checking for device assignment IOMMU support : WARN (No ACPI IORT table found, IOMMU not supported by this hardware platform) QEMU: Checking for secure guest support : WARN (Unknown if this platform has Secure Guest support) LXC: Checking for Linux >= 2.6.26 : PASS LXC: Checking for namespace ipc : PASS LXC: Checking for namespace mnt : PASS LXC: Checking for namespace pid : PASS LXC: Checking for namespace uts : PASS LXC: Checking for namespace net : PASS LXC: Checking for namespace user : PASS LXC: Checking for cgroup 'cpu' controller support : PASS LXC: Checking for cgroup 'cpuacct' controller support : PASS LXC: Checking for cgroup 'cpuset' controller support : PASS LXC: Checking for cgroup 'memory' controller support : PASS LXC: Checking for cgroup 'devices' controller support : PASS LXC: Checking for cgroup 'freezer' controller support : FAIL (Enable 'freezer' in kernel Kconfig file or mount/enable cgroup controller in your system) LXC: Checking for cgroup 'blkio' controller support : PASS LXC: Checking if device /sys/fs/fuse/connections exists : PASS # lsmod | grep kvm no errors (I have embedded the options needed to enable KVM inside the kernel) # virsh --connect qemu:///system capabilities | grep baselabel <baselabel type='kvm'>+1002:+1002</baselabel> <baselabel type='qemu'>+1002:+1002</baselabel> The error that I'm not able to fix is the following one : root@chromarietto:~# virsh domcapabilities --machine virt --emulatorbin /usr/local/bin/qemu-system-arm 2023-08-29 10:17:59.110+0000: 1763: error : virHostCPUGetKVMMaxVCPUs:1228 : KVM is not supported on this platform: Function not implemented ; error: failed to get emulator capabilities error: KVM is not supported on this platform: Function not implemented and this is the log that I've got when I ran libvirtd with the debug option enabled root@chromarietto:~# libvirtd --debug [Tue, 29 Aug 2023 10:10:11 virt-manager 2141] DEBUG (createvm:494) UEFI found, setting it as default. [Tue, 29 Aug 2023 10:10:11 virt-manager 2141] DEBUG (createvm:728) Guest type set to os_type=hvm, arch=armv7l, dom_type=kvm [Tue, 29 Aug 2023 10:10:11 virt-manager 2141] DEBUG (guest:546) Prefer EFI => True 2023-08-29 10:10:12.972+0000: 1765: error : virHostCPUGetKVMMaxVCPUs:1228 : KVM is not supported on this platform: Function not implemented [Tue, 29 Aug 2023 10:10:12 virt-manager 2141] DEBUG (domcapabilities:250) Error fetching domcapabilities XML Traceback (most recent call last): File "/usr/local/share/virt-manager/virtinst/domcapabilities.py", line 245, in build_from_params xml = conn.getDomainCapabilities(emulator, arch, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/lib/python3/dist-packages/libvirt.py", line 4612, in getDomainCapabilities raise libvirtError('virConnectGetDomainCapabilities() failed') libvirt.libvirtError: KVM is not supported on this platform: Function not implemented *2023-08-29 10:10:14.157+0000: 1762: error : virHostCPUGetKVMMaxVCPUs:1228 : KVM is not supported on this platform: Function not implemented* [Tue, 29 Aug 2023 10:10:14 virt-manager 2141] DEBUG (domcapabilities:250) Error fetching domcapabilities XML Traceback (most recent call last): File "/usr/local/share/virt-manager/virtinst/domcapabilities.py", line 245, in build_from_params xml = conn.getDomainCapabilities(emulator, arch, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/lib/python3/dist-packages/libvirt.py", line 4612, in getDomainCapabilities raise libvirtError('virConnectGetDomainCapabilities() failed') libvirt.libvirtError: KVM is not supported on this platform: Function not implemented [Tue, 29 Aug 2023 10:10:14 virt-manager 2141] DEBUG (createvm:497) Error checking for UEFI default Traceback (most recent call last): File "/usr/local/share/virt-manager/virtManager/createvm.py", line 491, in _set_caps_state guest.enable_uefi() File "/usr/local/share/virt-manager/virtinst/guest.py", line 589, in enable_uefi path = self._lookup_default_uefi_path() ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/share/virt-manager/virtinst/guest.py", line 848, in _lookup_default_uefi_path raise RuntimeError(_("Libvirt version does not support UEFI.")) RuntimeError: Libvirt version does not support UEFI Does anyone know how to fix that error ? -- Mario.

3 6

[libvirt PATCH v3] ch: Fix cloud-hypervisor version processing
by Praveen K Paladugu 08 Sep '23

08 Sep '23

Refactor the version processing logic in ch driver to support versions from non-release cloud-hypervisor binaries. This version also supports versions with branch prefixes in them. Signed-off-by: Praveen K Paladugu <prapal(a)linux.microsoft.com> --- src/ch/ch_conf.c | 42 ++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 40 insertions(+), 2 deletions(-) diff --git a/src/ch/ch_conf.c b/src/ch/ch_conf.c index a8565d9537..5573119b23 100644 --- a/src/ch/ch_conf.c +++ b/src/ch/ch_conf.c @@ -172,6 +172,37 @@ virCHDriverConfigDispose(void *obj) #define MIN_VERSION ((15 * 1000000) + (0 * 1000) + (0)) +/** + * chPreProcessVersionString: + * + * Returns: a pointer to numerical version without branch/commit info + */ +static char * +chPreProcessVersionString(char *version) +{ + char *tmp_version = version; + g_autofree char *ret_version = NULL; + char *sub_string; + + if ((sub_string = strrchr(version, '/')) != NULL) { + tmp_version = sub_string + 1; + } + + if (tmp_version[0] == 'v') { + tmp_version = tmp_version + 1; + } + + // Duplicate the string in both cases. This would allow the calling method + // free the returned string in a consistent manner. + if ((sub_string = strchr(tmp_version, '-')) != NULL) { + ret_version = g_strndup(tmp_version, sub_string - tmp_version); + } else{ + ret_version = g_strdup(tmp_version); + } + + return g_steal_pointer(&ret_version); + +} int chExtractVersion(virCHDriver *driver) { @@ -193,13 +224,20 @@ chExtractVersion(virCHDriver *driver) tmp = help; - /* expected format: cloud-hypervisor v<major>.<minor>.<micro> */ - if ((tmp = STRSKIP(tmp, "cloud-hypervisor v")) == NULL) { + /* Below are some example version formats and expected outputs: + * cloud-hypervisor v32.0.0 (expected: 32.0.0) + * cloud-hypervisor v33.0-104-ge0e3779e-dirty (expected: 33.0) + * cloud-hypervisor testing/v32.0.131-1-ga5d6db5c-dirty (expected: 32.0.131) + */ + if ((tmp = STRSKIP(tmp, "cloud-hypervisor ")) == NULL) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Unexpected output of cloud-hypervisor binary")); return -1; } + tmp = chPreProcessVersionString(tmp); + VIR_DEBUG("Cloud-Hypervisor version detected: %s", tmp); + if (virStringParseVersion(&version, tmp, true) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, _("Unable to parse cloud-hypervisor version: %1$s"), tmp); -- 2.41.0

3 2

Re: [libvirt PATCH v3] ch: Fix cloud-hypervisor version processing
by Praveen Paladugu 08 Sep '23

08 Sep '23

On Thu, Sep 07, 2023 at 09:02:44AM +0200, Jiri Denemark wrote: > On Wed, Sep 06, 2023 at 10:50:30 -0500, Praveen K Paladugu wrote: > > Refactor the version processing logic in ch driver to support versions > > from non-release cloud-hypervisor binaries. This version also supports > > versions with branch prefixes in them. > > > > Signed-off-by: Praveen K Paladugu <prapal(a)linux.microsoft.com> > > --- > > src/ch/ch_conf.c | 42 ++++++++++++++++++++++++++++++++++++++++-- > > 1 file changed, 40 insertions(+), 2 deletions(-) > > > > diff --git a/src/ch/ch_conf.c b/src/ch/ch_conf.c > > index a8565d9537..5573119b23 100644 > > --- a/src/ch/ch_conf.c > > +++ b/src/ch/ch_conf.c > > @@ -172,6 +172,37 @@ virCHDriverConfigDispose(void *obj) > > > > #define MIN_VERSION ((15 * 1000000) + (0 * 1000) + (0)) > > > > +/** > > + * chPreProcessVersionString: > > + * > > + * Returns: a pointer to numerical version without branch/commit info > > + */ > > +static char * > > +chPreProcessVersionString(char *version) > > +{ > > + char *tmp_version = version; > > + g_autofree char *ret_version = NULL; > > As Daniel pointed out in previous version this g_autofree and > g_steal_pointer below are useless as there is no path that actually free > the string here. > Thanks Jirka, I expected "autofree" attribute to be carried to "tmp" in the calling method, which would be freed when tmp goes out of scope. So, I implemented it this way. With some experimentation, I confirmed the correct behavior. I will fix this in my next version. > > + char *sub_string; > > + > > + if ((sub_string = strrchr(version, '/')) != NULL) { > > + tmp_version = sub_string + 1; > > + } > > + > > + if (tmp_version[0] == 'v') { > > + tmp_version = tmp_version + 1; > > + } > > + > > + // Duplicate the string in both cases. This would allow the calling method > > + // free the returned string in a consistent manner. > > Our coding style document says we prefer /* ... */ comments. > > > + if ((sub_string = strchr(tmp_version, '-')) != NULL) { > > + ret_version = g_strndup(tmp_version, sub_string - tmp_version); > > + } else{ > > + ret_version = g_strdup(tmp_version); > > + } > > + > > + return g_steal_pointer(&ret_version); > > + > > +} > > int > > chExtractVersion(virCHDriver *driver) > > { > > @@ -193,13 +224,20 @@ chExtractVersion(virCHDriver *driver) > > > > tmp = help; > > > > - /* expected format: cloud-hypervisor v<major>.<minor>.<micro> */ > > - if ((tmp = STRSKIP(tmp, "cloud-hypervisor v")) == NULL) { > > + /* Below are some example version formats and expected outputs: > > + * cloud-hypervisor v32.0.0 (expected: 32.0.0) > > + * cloud-hypervisor v33.0-104-ge0e3779e-dirty (expected: 33.0) > > + * cloud-hypervisor testing/v32.0.131-1-ga5d6db5c-dirty (expected: 32.0.131) > > + */ > > + if ((tmp = STRSKIP(tmp, "cloud-hypervisor ")) == NULL) { > > virReportError(VIR_ERR_INTERNAL_ERROR, "%s", > > _("Unexpected output of cloud-hypervisor binary")); > > return -1; > > } > > > > + tmp = chPreProcessVersionString(tmp); > > So chPreProcessVersionString returns an allocated string that will never > be freed because you assign it to tmp that's only used to point inside > other buffers and thus is (correctly) not freed anywhere. You need to > make sure the pointer returned here is freed before chExtractVersion > exits. Understood. > > > + VIR_DEBUG("Cloud-Hypervisor version detected: %s", tmp); > > + > > if (virStringParseVersion(&version, tmp, true) < 0) { > > virReportError(VIR_ERR_INTERNAL_ERROR, > > _("Unable to parse cloud-hypervisor version: %1$s"), tmp); > > Jirka

1 0

[PATCH] manpage: virsh: Note that 'virsh undefine --remove-all-storage' doesn't remove backing images
by Peter Krempa 07 Sep '23

07 Sep '23

Removing a backing image could break other image chains as it's theoretically possible to share backing chains. As --storage/--remove-all-storage is fully implemented in virsh as a helper option, which enumerates and deletes VM's volumes. We do not plan to make it any more complicated. Document that backing chains are not removed. Signed-off-by: Peter Krempa <pkrempa(a)redhat.com> --- docs/manpages/virsh.rst | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/docs/manpages/virsh.rst b/docs/manpages/virsh.rst index 91e1d5de37..4ae3bb4d93 100644 --- a/docs/manpages/virsh.rst +++ b/docs/manpages/virsh.rst @@ -4614,12 +4614,13 @@ undefined; if not all of the requested volumes could be deleted, the error message indicates what still remains behind. If a volume path is not found in the domain definition, it's treated as if the volume was successfully deleted. Only volumes managed by libvirt in storage pools can be removed this -way. +way. Note that this also removes only the top level image of a backing chain, +any backing stores of the image are kept as they may be shared. (See ``domblklist`` for list of target names associated to a domain). Example: --storage vda,/path/to/storage.img The *--remove-all-storage* flag specifies that all of the domain's storage -volumes should be deleted. +volumes should be deleted as if they were specified via *--storage*. The *--delete-storage-volume-snapshots* (previously *--delete-snapshots*) flag specifies that any snapshots associated with -- 2.41.0

2 1

[PATCH] virsh-snapshot: Avoid passing NULL to qsort() in virshSnapshotListCollect()
by Michal Privoznik 07 Sep '23

07 Sep '23

If a domain has no snapshots and 'virsh snapshot-list' is called, this gets all the way down to virshSnapshotListCollect() which then collects all snapshots (none), and passes them to qsort() which doesn't like being called with NULL: extern void qsort (void *__base, size_t __nmemb, size_t __size, __compar_fn_t __compar) __nonnull ((1, 4)); Resolves: https://gitlab.com/libvirt/libvirt/-/issues/533 Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- tools/virsh-snapshot.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tools/virsh-snapshot.c b/tools/virsh-snapshot.c index d7889a38e4..ecb935b2b4 100644 --- a/tools/virsh-snapshot.c +++ b/tools/virsh-snapshot.c @@ -1310,9 +1310,11 @@ virshSnapshotListCollect(vshControl *ctl, virDomainPtr dom, } } } - if (!(orig_flags & VIR_DOMAIN_SNAPSHOT_LIST_TOPOLOGICAL)) + if (!(orig_flags & VIR_DOMAIN_SNAPSHOT_LIST_TOPOLOGICAL) && + snaplist->snaps) { qsort(snaplist->snaps, snaplist->nsnaps, sizeof(*snaplist->snaps), virshSnapSorter); + } snaplist->nsnaps -= deleted; ret = g_steal_pointer(&snaplist); -- 2.41.0

2 1

[PATCH] lxc: fix lxcContainerMountAllFS() DEREF_BEFORE_CHECK
by Dmitry Frolov 07 Sep '23

07 Sep '23

Reviewing the sources, I found, that in function lxcContainerMountAllFS() pointers vmDef->fss[i]->src and vmDef->fss[i]->src->path are checked for NULL after dereferencing in VIR_DEBUG() macro. Fixes: 57487085dc ("lxc: don't try to reference NULL when mounting filesystems") --- src/lxc/lxc_container.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/src/lxc/lxc_container.c b/src/lxc/lxc_container.c index 21220661f7..58a6695458 100644 --- a/src/lxc/lxc_container.c +++ b/src/lxc/lxc_container.c @@ -1467,13 +1467,15 @@ static int lxcContainerMountAllFS(virDomainDef *vmDef, if (STREQ(vmDef->fss[i]->dst, "/")) continue; + if (!(vmDef->fss[i]->src && vmDef->fss[i]->src->path)) + return -1; + VIR_DEBUG("Mounting '%s' -> '%s'", vmDef->fss[i]->src->path, vmDef->fss[i]->dst); if (lxcContainerResolveSymlinks(vmDef->fss[i], false) < 0) return -1; - if (!(vmDef->fss[i]->src && vmDef->fss[i]->src->path && - STRPREFIX(vmDef->fss[i]->src->path, vmDef->fss[i]->dst)) && + if (!STRPREFIX(vmDef->fss[i]->src->path, vmDef->fss[i]->dst) && lxcContainerUnmountSubtree(vmDef->fss[i]->dst, false) < 0) return -1; -- 2.34.1

4 4

[PATCH Libvirt v3 00/10] Support dirty page rate upper limit
by ~hyman 07 Sep '23

07 Sep '23

Hi, This is the latest version for the series, please review. Thanks, Yong v3: - rebase on master and fix conflicts - refactor the comments and update the version tag v2: The patch set introduce a new API virDomainSetVcpuDirtyLimit to allow upper Apps to set upper limits of dirty page rate for virtual CPUs, the corresponding virsh API as follows: limit-dirty-page-rate <domain> <rate> [--vcpu <number>] \ [--config] [--live] [--current] We put the dirty limit persistent info with the "vcpus" element in domain XML and extend dirtylimit statistics for domGetStats: <domain> ... <vcpu current='2'>3</vcpu> <vcpus> <vcpu id='0' hotpluggable='no' dirty_limit='10' order='1'.../> <vcpu id='1' hotpluggable='yes' dirty_limit='10' order='2'.../> </vcpus> ... </domain> If --vcpu option is not passed in the virsh command, set all virtual CPUs; if rate is set to zero, cancel the upper limit. Examples: To set the dirty page rate upper limit 10 MB/s for all virtual CPUs in c81_node1, use: # virsh limit-dirty-page-rate c81_node1 --rate 10 --live Set dirty page rate limit 10(MB/s) for all virtual CPUs successfully # virsh dumpxml c81_node1 | grep dirty_limit <vcpu id='0' enabled='yes' hotpluggable='no' order='1' dirty_limit='10'/> <vcpu id='1' enabled='yes' hotpluggable='no' order='2' dirty_limit='10'/> <vcpu id='2' enabled='yes' hotpluggable='no' order='3' dirty_limit='10'/> <vcpu id='3' enabled='no' hotpluggable='yes' dirty_limit='10'/> <vcpu id='4' enabled='no' hotpluggable='yes' dirty_limit='10'/> ...... Query the dirty limit info dynamically: # virsh domstats c81_node1 --dirtylimit Domain: 'c81_node1' dirtylimit.vcpu.0.limit=10 dirtylimit.vcpu.0.current=0 dirtylimit.vcpu.1.limit=10 dirtylimit.vcpu.1.current=0 dirtylimit.vcpu.2.limit=10 dirtylimit.vcpu.2.current=0 dirtylimit.vcpu.3.limit=10 dirtylimit.vcpu.3.current=0 dirtylimit.vcpu.4.limit=10 dirtylimit.vcpu.4.current=0 ...... To cancel the upper limit, use: # virsh limit-dirty-page-rate c81_node1 --rate 0 --live Cancel dirty page rate limit for all virtual CPUs successfully # virsh dumpxml c81_node1 | grep dirty_limit # virsh domstats c81_node1 --dirtylimit Domain: 'c81_node1' The dirty limit uses the QEMU dirty-limit feature introduced since 7.1.0, this feature allows CPU to be throttled as needed to keep their dirty page rate within the limit. It could, in some scenes, be used to provide quality-of-service in the aspect of the memory workload for virtual CPUs and QEMU itself use the feature to implement the dirty-limit throttle algorithm and apply it on the live migration, which improve responsiveness of large guests during live migration and can result in more stable read performance. The other application scenarios remain unexplored, before that, Libvirt could provide the basic API. Hyman Huang(黄勇) (10): qemu_capabilities: Introduce QEMU_CAPS_VCPU_DIRTY_LIMIT capability conf: Introduce XML for dirty limit configuration libvirt: Add virDomainSetVcpuDirtyLimit API qemu_driver: Implement qemuDomainSetVcpuDirtyLimit domain_validate: Export virDomainDefHasDirtyLimitStartupVcpus symbol qemu_process: Setup dirty limit after launching VM virsh: Introduce limit-dirty-page-rate api qemu_monitor: Implement qemuMonitorQueryVcpuDirtyLimit qemu_driver: Extend dirtlimit statistics for domGetStats virsh: Introduce command 'virsh domstats --dirtylimit' docs/formatdomain.rst | 7 +- docs/manpages/virsh.rst | 38 +++- include/libvirt/libvirt-domain.h | 5 + src/conf/domain_conf.c | 26 +++ src/conf/domain_conf.h | 8 + src/conf/domain_validate.c | 33 ++++ src/conf/domain_validate.h | 2 + src/conf/schemas/domaincommon.rng | 5 + src/driver-hypervisor.h | 7 + src/libvirt-domain.c | 68 +++++++ src/libvirt_private.syms | 1 + src/libvirt_public.syms | 1 + src/qemu/qemu_capabilities.c | 2 + src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_driver.c | 181 ++++++++++++++++++ src/qemu/qemu_monitor.c | 25 +++ src/qemu/qemu_monitor.h | 22 +++ src/qemu/qemu_monitor_json.c | 107 +++++++++++ src/qemu/qemu_monitor_json.h | 9 + src/qemu/qemu_process.c | 44 +++++ src/remote/remote_driver.c | 1 + src/remote/remote_protocol.x | 18 +- src/remote_protocol-structs | 7 + .../qemucapabilitiesdata/caps_7.1.0_ppc64.xml | 1 + .../caps_7.1.0_x86_64.xml | 1 + tests/qemucapabilitiesdata/caps_7.2.0_ppc.xml | 1 + .../caps_7.2.0_x86_64+hvf.xml | 1 + .../caps_7.2.0_x86_64.xml | 1 + .../caps_8.0.0_riscv64.xml | 1 + .../caps_8.0.0_x86_64.xml | 1 + .../qemucapabilitiesdata/caps_8.1.0_s390x.xml | 1 + .../caps_8.1.0_x86_64.xml | 1 + tools/virsh-domain-monitor.c | 7 + tools/virsh-domain.c | 109 +++++++++++ 34 files changed, 739 insertions(+), 4 deletions(-) -- 2.38.5

4 17

[libvirt PATCH v2] ch: Fix cloud-hypervisor version processing
by Praveen K Paladugu 07 Sep '23

07 Sep '23

Refactor the version processing logic in ch driver to support versions from non-release cloud-hypervisor binaries. This version also supports versions with branch prefixes in them. Signed-off-by: Praveen K Paladugu <prapal(a)linux.microsoft.com> --- src/ch/ch_conf.c | 39 ++++++++++++++++++++++++++++++++++++--- 1 file changed, 36 insertions(+), 3 deletions(-) diff --git a/src/ch/ch_conf.c b/src/ch/ch_conf.c index a8565d9537..a0849bfbd6 100644 --- a/src/ch/ch_conf.c +++ b/src/ch/ch_conf.c @@ -172,6 +172,33 @@ virCHDriverConfigDispose(void *obj) #define MIN_VERSION ((15 * 1000000) + (0 * 1000) + (0)) +/** + * chPreProcessVersionString - crop branch, commit info and return just the version + */ +static char * +chPreProcessVersionString(char *version) +{ + g_autofree char *new_version = version; + char *tmp_version; + + if ((tmp_version = strrchr(version, '/')) != NULL) { + new_version = tmp_version + 1; + } + + if (new_version[0] == 'v') { + new_version = new_version + 1; + } + // Duplicate the string in both cases. This would allow the calling method + // free the returned string in a consistent manner. + if ((tmp_version = strchr(new_version, '-')) != NULL) { + new_version = g_strndup(new_version, tmp_version - new_version); + } else{ + new_version = g_strdup(new_version); + } + + return g_steal_pointer(&new_version); + +} int chExtractVersion(virCHDriver *driver) { @@ -193,19 +220,25 @@ chExtractVersion(virCHDriver *driver) tmp = help; - /* expected format: cloud-hypervisor v<major>.<minor>.<micro> */ - if ((tmp = STRSKIP(tmp, "cloud-hypervisor v")) == NULL) { + /* Below are some example version formats and expected outputs: + * cloud-hypervisor v32.0.0 (expected: 32.0.0) + * cloud-hypervisor v33.0-104-ge0e3779e-dirty (expected: 33.0) + * cloud-hypervisor testing/v32.0.131-1-ga5d6db5c-dirty (expected: 32.0.131) + */ + if ((tmp = STRSKIP(tmp, "cloud-hypervisor ")) == NULL) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Unexpected output of cloud-hypervisor binary")); return -1; } + tmp = chPreProcessVersionString(tmp); + VIR_DEBUG("Cloud-Hypervisor version detected: %s", tmp); + if (virStringParseVersion(&version, tmp, true) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, _("Unable to parse cloud-hypervisor version: %1$s"), tmp); return -1; } - if (version < MIN_VERSION) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Cloud-Hypervisor version is too old (v15.0 is the minimum supported version)")); -- 2.41.0

3 2

[libvirt PATCH 0/2] Add cpu model for EPYC Genoa
by Tim Wiederhake 06 Sep '23

06 Sep '23

Plus one drive-by improvement to sync_qemu_features_i386.py. Tim Wiederhake (2): cpu_map: Ignore all vmx-* features in sync_qemu_features_i386.py cpu_map: Add cpu model EPYC Genoa src/cpu_map/index.xml | 1 + src/cpu_map/meson.build | 1 + src/cpu_map/sync_qemu_features_i386.py | 90 +------------- src/cpu_map/x86_EPYC-Genoa.xml | 115 ++++++++++++++++++ .../domaincapsdata/qemu_8.1.0-q35.x86_64.xml | 1 + .../domaincapsdata/qemu_8.1.0-tcg.x86_64.xml | 1 + tests/domaincapsdata/qemu_8.1.0.x86_64.xml | 1 + 7 files changed, 121 insertions(+), 89 deletions(-) create mode 100644 src/cpu_map/x86_EPYC-Genoa.xml -- 2.39.2

2 3