January 2018 - Devel - libvirt List Archives

[libvirt] [PATCH 00/10] Cache Allocation Technology support, this time for real (??)
by Martin Kletzander 25 Jan '18

25 Jan '18

So again all the reviews are incorporated. I managed to get my hands on a machine with CAT support, so I also found out some things that needed tweaking. @Eli: I would still appreciate you helping with some testing as I wasn't able to find a machine with multiple caches (sockets) and I couldn't test that properly (mainly the fact that full mask needs to be specified for caches for which we don't want any allocation to happen). Martin Kletzander (10): Rename virResctrlInfo to virResctrlInfoPerCache util: Add virResctrlInfo conf: Use virResctrlInfo in capabilities util: Remove now-unneeded resctrl functions resctrl: Add functions to work with resctrl allocations conf: Add support for cputune/cachetune tests: Add virresctrltest qemu: Add support for resctrl docs: Add CAT (resctrl) support into news.xml tests: Clean up and modify some vircaps2xmldata docs/formatdomain.html.in | 54 + docs/news.xml | 9 + docs/schemas/domaincommon.rng | 32 + po/POTFILES.in | 1 + src/Makefile.am | 2 +- src/conf/capabilities.c | 55 +- src/conf/capabilities.h | 4 +- src/conf/domain_conf.c | 295 +++- src/conf/domain_conf.h | 13 + src/libvirt_private.syms | 15 +- src/qemu/qemu_process.c | 60 +- src/util/virresctrl.c | 1542 ++++++++++++++++++-- src/util/virresctrl.h | 73 +- src/util/virresctrlpriv.h | 27 + tests/Makefile.am | 9 +- tests/genericxml2xmlindata/cachetune-cdp.xml | 36 + .../cachetune-colliding-allocs.xml | 30 + .../cachetune-colliding-tunes.xml | 32 + .../cachetune-colliding-types.xml | 30 + tests/genericxml2xmlindata/cachetune-small.xml | 29 + tests/genericxml2xmlindata/cachetune.xml | 33 + tests/genericxml2xmltest.c | 10 + .../vircaps2xmldata/linux-resctrl-cdp/resctrl/cpus | 1 - .../linux-resctrl-cdp/resctrl/tasks | 0 .../linux-resctrl-skx-twocaches/resctrl/schemata | 2 +- tests/vircaps2xmldata/linux-resctrl/resctrl/cpus | 1 - .../vircaps2xmldata/linux-resctrl/resctrl/schemata | 2 +- tests/vircaps2xmldata/linux-resctrl/resctrl/tasks | 0 tests/virresctrldata/resctrl-cdp.schemata | 2 + .../virresctrldata/resctrl-skx-twocaches.schemata | 1 + tests/virresctrldata/resctrl-skx.schemata | 1 + tests/virresctrldata/resctrl.schemata | 1 + tests/virresctrltest.c | 102 ++ 33 files changed, 2363 insertions(+), 141 deletions(-) create mode 100644 src/util/virresctrlpriv.h create mode 100644 tests/genericxml2xmlindata/cachetune-cdp.xml create mode 100644 tests/genericxml2xmlindata/cachetune-colliding-allocs.xml create mode 100644 tests/genericxml2xmlindata/cachetune-colliding-tunes.xml create mode 100644 tests/genericxml2xmlindata/cachetune-colliding-types.xml create mode 100644 tests/genericxml2xmlindata/cachetune-small.xml create mode 100644 tests/genericxml2xmlindata/cachetune.xml delete mode 100644 tests/vircaps2xmldata/linux-resctrl-cdp/resctrl/cpus delete mode 100644 tests/vircaps2xmldata/linux-resctrl-cdp/resctrl/tasks delete mode 100644 tests/vircaps2xmldata/linux-resctrl/resctrl/cpus delete mode 100644 tests/vircaps2xmldata/linux-resctrl/resctrl/tasks create mode 100644 tests/virresctrldata/resctrl-cdp.schemata create mode 100644 tests/virresctrldata/resctrl-skx-twocaches.schemata create mode 100644 tests/virresctrldata/resctrl-skx.schemata create mode 100644 tests/virresctrldata/resctrl.schemata create mode 100644 tests/virresctrltest.c -- 2.16.1

2 24

[libvirt] [PATCH] docs: document requirement to provide Signed-off-by lines for DCO
by Daniel P. Berrange 25 Jan '18

25 Jan '18

Document that contributors are required to assert compliance with the Developers Certification of Origin 1.1, by providing Signed-off-by tags for all commit messages. The DCO is formally stating what we have long implicitly expected of contributors in terms of their legal rights to make the contribution. This puts the project in a stronger position should any questions around contributions be raised going forward in the future. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- docs/hacking.html.in | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/docs/hacking.html.in b/docs/hacking.html.in index 6cadfb8343..98a24d785c 100644 --- a/docs/hacking.html.in +++ b/docs/hacking.html.in @@ -108,12 +108,18 @@ of the bug number is useful; but also summarize the issue rather than making all readers follow the link. You can use 'git shortlog -30' to get an idea of typical summary lines. - Libvirt does not currently attach any meaning to - Signed-off-by: lines, so it is up to you if you want to - include or omit them in the commit message. </li> + <li>Contributors to libvirt projects must + assert that they are in compliance with the + <a href="https://developercertificate.org/">Developers + Certificate of Origin 1.1</a>. This is achieved by adding + a "Signed-off-by" line to every commit message. The presence + of this line attests that the contributor has read the + above lined DCO and agrees with its statements. + </li> + <li>Split large changes into a series of smaller patches, self-contained if possible, with an explanation of each patch and an explanation of how the sequence of patches fits -- 2.14.3

5 6

[libvirt] [PATCH v3 0/3] nodedev: update caps before quering nodedev infos
by Wu Zongyong 25 Jan '18

25 Jan '18

Some capabilities of node devices rely on what driver they bound to, and therefore, these capabilities may change when the driver change. In current implemention, it is not consistent between real status and the status we get by invoking nodedev interfaces. So, this series of patches try to manually update devices' capabilities each time before nodedev driver interfaces invoked. Wu Zongyong (3): v2->v3: 1. split a single patch to three part 2. fix memory leak and lock problems 3. update caps before invoking nodedev driver interfaces (refer to patch 3/3) nodedev: add macro guard to node_device_udev header file nodedev: update mdev_types caps before dumpxml nodedev: update caps before invoking nodedev driver interfaces src/node_device/node_device_driver.c | 55 +++++++++++++++++++++++++++++++ src/node_device/node_device_linux_sysfs.c | 22 +++++++++++++ src/node_device/node_device_udev.c | 37 +++++++++++++++++---- src/node_device/node_device_udev.h | 18 +++++++--- 4 files changed, 121 insertions(+), 11 deletions(-) -- 1.9.1

3 9

[libvirt] with vhostuser I cannot use hugepages
by Adnan Mundres 25 Jan '18

25 Jan '18

In my setup I am using qemu-kvm without openstack. I am trying to use ovs with dpdk. In my xml file I added following lines for the dpdkvhostuser <cpu mode='host-passthrough'> <numa> <cell id="0" cpus="0-6" memory="16777216" unit='KiB' memAccess="shared"/> </numa> </cpu> I have enabled hugepages during boot time ot@mvmgptb11hyp01 hyp-1]# cat /proc/meminfo | grep -i hugeAnonHugePages: 126976 kBHugePages_Total: 100HugePages_Free: 80HugePages_Rsvd: 0HugePages_Surp: 0Hugepagesize: 1048576 kB But When I start my vm (virsh start vm1.xml) I am seeing that this vm is not using memory from hugepages, rather it is taking memory from total memory. When I checked the log file I see that it is using -object memory-backend-file,id=ram-node0,prealloc=yes,mem-path=/var/lib/libvirt/qemu/ram,share=yes It should use backend memory as following -object memory-backend-file,id=ram-node0,prealloc=yes,mem-path=/dev/hugepages/libvirt/qemu/vm1,share=yes Any idea how can I use memory from hugepages

3 7

[libvirt] [PATCH-RUBY] The first argument passed should be the disk not the size.
by Marius Rieder 25 Jan '18

25 Jan '18

Calling block_resize on a Libvirt::Domain object leads to a TypeError. The reason seems to be that ruby-libvirt tries to convert the size instead of the disk to a string to pass as the disk parameter. --- ext/libvirt/domain.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/ext/libvirt/domain.c b/ext/libvirt/domain.c index 7812450..aff1842 100644 --- a/ext/libvirt/domain.c +++ b/ext/libvirt/domain.c @@ -2882,7 +2882,7 @@ static VALUE libvirt_domain_block_resize(int argc, VALUE *argv, VALUE d) ruby_libvirt_generate_call_nil(virDomainBlockResize, ruby_libvirt_connect_get(d), ruby_libvirt_domain_get(d), - StringValueCStr(size), NUM2ULL(size), + StringValueCStr(disk), NUM2ULL(size), ruby_libvirt_value_to_uint(flags)); } #endif -- 2.7.4

1 0

[libvirt] [PATCH] qemu: get VM's ip address from the output of arp command
by Chen Hanxiao 25 Jan '18

25 Jan '18

From: Chen Hanxiao <chenhanxiao(a)gmail.com> Introduce VIR_DOMAIN_INTERFACE_ADDRESSES_SRC_ARP to get ip address of VM from the output of `arp -an` command. We can use: domifaddr f26-cloud --source arp to get the address. Signed-off-by: Chen Hanxiao <chenhanxiao(a)gmail.com> --- include/libvirt/libvirt-domain.h | 1 + src/qemu/qemu_driver.c | 102 +++++++++++++++++++++++++++++++++++++++ tools/virsh-domain-monitor.c | 2 + 3 files changed, 105 insertions(+) diff --git a/include/libvirt/libvirt-domain.h b/include/libvirt/libvirt-domain.h index 4048acf38..38e2d9a3e 100644 --- a/include/libvirt/libvirt-domain.h +++ b/include/libvirt/libvirt-domain.h @@ -4665,6 +4665,7 @@ typedef virMemoryParameter *virMemoryParameterPtr; typedef enum { VIR_DOMAIN_INTERFACE_ADDRESSES_SRC_LEASE = 0, /* Parse DHCP lease file */ VIR_DOMAIN_INTERFACE_ADDRESSES_SRC_AGENT = 1, /* Query qemu guest agent */ + VIR_DOMAIN_INTERFACE_ADDRESSES_SRC_ARP = 2, /* Query ARP tables */ # ifdef VIR_ENUM_SENTINELS VIR_DOMAIN_INTERFACE_ADDRESSES_SRC_LAST diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index a203c9297..5aaf69442 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -160,6 +160,9 @@ static int qemuGetDHCPInterfaces(virDomainPtr dom, virDomainObjPtr vm, virDomainInterfacePtr **ifaces); +static int qemuARPGetInterfaces(virDomainObjPtr vm, + virDomainInterfacePtr **ifaces); + static virQEMUDriverPtr qemu_driver; @@ -20384,6 +20387,10 @@ qemuDomainInterfaceAddresses(virDomainPtr dom, break; + case VIR_DOMAIN_INTERFACE_ADDRESSES_SRC_ARP: + ret = qemuARPGetInterfaces(vm, ifaces); + break; + default: virReportError(VIR_ERR_ARGUMENT_UNSUPPORTED, _("Unknown IP address data source %d"), @@ -20494,6 +20501,101 @@ qemuGetDHCPInterfaces(virDomainPtr dom, } +static int +qemuARPGetInterfaces(virDomainObjPtr vm, + virDomainInterfacePtr **ifaces) +{ + size_t i, j; + size_t ifaces_count = 0; + int ret = -1; + char macaddr[VIR_MAC_STRING_BUFLEN]; + virDomainInterfacePtr *ifaces_ret = NULL; + virDomainInterfacePtr iface = NULL; + virCommandPtr cmd = NULL; + char *outbuf; + char **lines = NULL; + char **matches = NULL; +#define ARP_CMD "/usr/sbin/arp" + + if (!(cmd = virCommandNewArgList(ARP_CMD, + "-an", + NULL))) + goto cleanup; + + virCommandSetOutputBuffer(cmd, &outbuf); + + if (virCommandRun(cmd, NULL) < 0) + goto cleanup; + + lines = virStringSplit(outbuf, "\n", 0); + if (lines == NULL) + goto cleanup; + + for (i = 0; i < vm->def->nnets; i++) { + if (vm->def->nets[i]->type != VIR_DOMAIN_NET_TYPE_NETWORK) + continue; + + virMacAddrFormat(&(vm->def->nets[i]->mac), macaddr); + + if (strstr(outbuf, macaddr)) { + for (j = 0; lines[j]; j++) { + const char *line = lines[j]; + if (line == NULL) + break; + if (strstr(line, macaddr) && + virStringSearch(line, "\\b([0-9]{1,3}\\.){3}[0-9]{1,3}\\b", + 1, &matches) == 1) { + + if (VIR_EXPAND_N(ifaces_ret, ifaces_count, 1) < 0) + goto error; + + if (VIR_ALLOC(ifaces_ret[ifaces_count - 1]) < 0) + goto error; + + iface = ifaces_ret[ifaces_count - 1]; + iface->naddrs = 1; + if (VIR_ALLOC_N(iface->addrs, iface->naddrs) < 0) + goto error; + + if (VIR_STRDUP(iface->name, vm->def->nets[i]->ifname) < 0) + goto cleanup; + + if (VIR_STRDUP(iface->hwaddr, macaddr) < 0) + goto cleanup; + + if (VIR_STRDUP(iface->addrs->addr, matches[0]) < 0) + goto cleanup; + } + } + } else { + VIR_DEBUG("Got nothing"); + continue; + } + } + + *ifaces = ifaces_ret; + ifaces_ret = NULL; + ret = ifaces_count; + + cleanup: + virCommandFree(cmd); + VIR_FREE(outbuf); + virStringListFree(lines); + virStringListFree(matches); + cmd = NULL; + return ret; + + error: + if (ifaces_ret) { + for (i = 0; i < ifaces_count; i++) + virDomainInterfaceFree(ifaces_ret[i]); + } + VIR_FREE(ifaces_ret); + + goto cleanup; +} + + static int qemuDomainSetUserPassword(virDomainPtr dom, const char *user, diff --git a/tools/virsh-domain-monitor.c b/tools/virsh-domain-monitor.c index 32a42707e..68da11ed5 100644 --- a/tools/virsh-domain-monitor.c +++ b/tools/virsh-domain-monitor.c @@ -2190,6 +2190,8 @@ cmdDomIfAddr(vshControl *ctl, const vshCmd *cmd) source = VIR_DOMAIN_INTERFACE_ADDRESSES_SRC_LEASE; } else if (STREQ(sourcestr, "agent")) { source = VIR_DOMAIN_INTERFACE_ADDRESSES_SRC_AGENT; + } else if (STREQ(sourcestr, "arp")) { + source = VIR_DOMAIN_INTERFACE_ADDRESSES_SRC_ARP; } else { vshError(ctl, _("Unknown data source '%s'"), sourcestr); goto cleanup; -- 2.14.3

2 2

[libvirt] [PATCH] libxl: add explicit linkage to xenstore library
by Daniel P. Berrange 24 Jan '18

24 Jan '18

The libxl driver calls a couple of xenstorage APIs, so it must explicitly link to this library rather than rely on indirect linkage via libxl or other xen libraries. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- m4/virt-driver-libxl.m4 | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/m4/virt-driver-libxl.m4 b/m4/virt-driver-libxl.m4 index e6dbfef686..2cc1c062d8 100644 --- a/m4/virt-driver-libxl.m4 +++ b/m4/virt-driver-libxl.m4 @@ -75,9 +75,9 @@ AC_DEFUN([LIBVIRT_DRIVER_CHECK_LIBXL], [ dnl (since Xen 4.7) if not then assume it is in libxenctrl dnl (as it was for 4.6 and earler) AC_CHECK_LIB([xentoollog], [xtl_createlogger_stdiostream], [ - LIBXL_LIBS="$LIBXL_LIBS -lxentoollog" + LIBXL_LIBS="$LIBXL_LIBS -lxenstore -lxentoollog" ],[ - LIBXL_LIBS="$LIBXL_LIBS -lxenctrl" + LIBXL_LIBS="$LIBXL_LIBS -lxenstore -lxenctrl" ]) dnl Check if libxl_domain_config_from_json is available for domXML to -- 2.14.3

2 1

[libvirt] [PATCH 0/8] virsh: Introduce more completers
by Michal Privoznik 24 Jan '18

24 Jan '18

It'd be nice if we could autocomplete the basic types like pools, volumes, networks, ... before the release. Michal Privoznik (8): virsh: Introduce virshStoragePoolNameCompleter virsh: Introduce virshStorageVolNameCompleter virsh: Introduce virshInterfaceNameCompleter virsh: Introduce virshNetworkNameCompleter virsh: Introduce virshNodeDeviceNameCompleter virsh: Introduce virshNWFilterNameCompleter virsh: Introduce virshSecretUUIDCompleter virsh: Introduce virshSnapshotNameCompleter tools/virsh-completer.c | 388 ++++++++++++++++++++++++++++++++++++++++++++++++ tools/virsh-completer.h | 32 ++++ tools/virsh-interface.c | 16 +- tools/virsh-network.c | 24 +-- tools/virsh-nodedev.c | 16 +- tools/virsh-nwfilter.c | 9 +- tools/virsh-pool.c | 28 ++-- tools/virsh-secret.c | 15 +- tools/virsh-snapshot.c | 21 ++- tools/virsh-volume.c | 45 +++--- tools/virsh.h | 6 +- 11 files changed, 525 insertions(+), 75 deletions(-) -- 2.13.6

2 23

[libvirt] Libvirt fails to apply security context to fd/node to USB device
by Randy Aybar 23 Jan '18

23 Jan '18

Hi, I'm attempting to attach and expose a USB device (WiFi adapter for testing) to an LXC container with SELinux enabled. But when enabling the XML snippet, the container fails to start with this error: 2018-01-12 19:24:31.914+0000: 2181: error : virSecuritySELinuxSetFileconHelper:1182 : unable to set security context 'system_u:object_r:svirt_sandbox_file_t:s0:c139,c284' on '//var/run/libvirt/lxc/lxc_0.dev/bus/usb//dev/bus/usb/002/002': No such file or directory Failure in libvirt_lxc startup: unable to set security context 'system_u:object_r:svirt_sandbox_file_t:s0:c139,c284' on '//var/run/libvirt/lxc/lxc_0.dev/bus/usb//dev/bus/usb/002/002': No such file or directory The XML snippet for attaching USB device: <hostdev mode='subsystem' type='usb' managed='yes'> <source> <vendor id='0x05ac'/> <product id='0x1006'/> </source> <address type='usb' bus='2' port='2'/> </hostdev> SELinux snippet (using the dynamic label for the moment): <seclabel type='dynamic' model='selinux' relabel='yes'/> Running it on CentOS 7.2 and I've tried the distro from the package manager (1.3.3.3), as well as downloaded and compiled the latest stable from libvirt.org (3.10?) and came to the same error each time. Did a small dive into the code after realizing that the path just doesn't seem right. Path (seems to have an unusual and incorrect concatenation of folders) : //var/run/libvirt/lxc/lxc_0.dev/bus/usb//dev/bus/usb/002/002 "vroot" seems to be declared by the LXC controller (src/lxc/lxc_controller.c) as such: if (virAsprintf(&vroot, "/%s/%s.dev/bus/usb/", LXC_STATE_DIR, vmDef->name) < 0) goto cleanup; Then upon setting up security for all of the container's attached devices, we call virUSBDeviceNew to setup the attached USB device and give us the path to apply a context to. Since vroot is present, we get this weird path when running through this (src/util/virusb.c): if (virAsprintf(&dev->path, "%s" USB_DEVFS "%03d/%03d", vroot ? vroot : "", dev->bus, dev->dev) < 0) { virUSBDeviceFree(dev); return NULL; } Note: # define USB_DEVFS "/dev/bus/usb/" Should we just be blindly appending this definition if vroot is present, making the path incorrect? If this isn't a bug, I propose the following change: if (virAsprintf(&dev->path, "%s" "%03d/%03d", vroot ? vroot : USB_DEVFS, dev->bus, dev->dev) < 0) { virUSBDeviceFree(dev); return NULL; } Would kindly appreciate any feedback on whether this is a bug or maybe I'm missing something and is the reason why it's written this way. Thanks, Randy

2 2

[libvirt] [PATCH 0/9] Yet another version of CAT stuff (no idea about the version number)
by Martin Kletzander 23 Jan '18

23 Jan '18

Added stuff that Pavel wanted, removed some testcases that I still have in my repo, but it's way to complicated to add properly, so for now I would just go with this and manual testing of starting some domains. I got no hardware to test this on, currently, so some testing is needed before pushing this. @Eli: Can you help with the testing? Martin Kletzander (9): Rename virResctrlInfo to virResctrlInfoPerCache util: Add virResctrlInfo conf: Use virResctrlInfo in capabilities util: Remove now-unneeded resctrl functions resctrl: Add functions to work with resctrl allocations conf: Add support for cputune/cachetune tests: Add virresctrltest qemu: Add support for resctrl docs: Add CAT (resctrl) support into news.xml docs/formatdomain.html.in | 54 + docs/news.xml | 9 + docs/schemas/domaincommon.rng | 32 + po/POTFILES.in | 1 + src/Makefile.am | 2 +- src/conf/capabilities.c | 55 +- src/conf/capabilities.h | 4 +- src/conf/domain_conf.c | 251 ++++ src/conf/domain_conf.h | 13 + src/libvirt_private.syms | 16 +- src/qemu/qemu_process.c | 61 +- src/util/virresctrl.c | 1380 ++++++++++++++++++-- src/util/virresctrl.h | 86 +- src/util/virresctrlpriv.h | 27 + tests/Makefile.am | 8 +- tests/genericxml2xmlindata/cachetune-cdp.xml | 36 + .../cachetune-colliding-allocs.xml | 30 + .../cachetune-colliding-tunes.xml | 32 + .../cachetune-colliding-types.xml | 30 + tests/genericxml2xmlindata/cachetune-small.xml | 29 + tests/genericxml2xmlindata/cachetune.xml | 33 + tests/genericxml2xmltest.c | 10 + tests/virresctrldata/resctrl-cdp.schemata | 2 + .../virresctrldata/resctrl-skx-twocaches.schemata | 1 + tests/virresctrldata/resctrl-skx.schemata | 1 + tests/virresctrldata/resctrl.schemata | 1 + tests/virresctrltest.c | 102 ++ 27 files changed, 2174 insertions(+), 132 deletions(-) create mode 100644 src/util/virresctrlpriv.h create mode 100644 tests/genericxml2xmlindata/cachetune-cdp.xml create mode 100644 tests/genericxml2xmlindata/cachetune-colliding-allocs.xml create mode 100644 tests/genericxml2xmlindata/cachetune-colliding-tunes.xml create mode 100644 tests/genericxml2xmlindata/cachetune-colliding-types.xml create mode 100644 tests/genericxml2xmlindata/cachetune-small.xml create mode 100644 tests/genericxml2xmlindata/cachetune.xml create mode 100644 tests/virresctrldata/resctrl-cdp.schemata create mode 100644 tests/virresctrldata/resctrl-skx-twocaches.schemata create mode 100644 tests/virresctrldata/resctrl-skx.schemata create mode 100644 tests/virresctrldata/resctrl.schemata create mode 100644 tests/virresctrltest.c -- 2.15.1

6 31