September 2017 - Devel - libvirt List Archives

[libvirt] [PATCH] qemu: Introduce a wrapper over virFileWrapperFdClose
by Michal Privoznik 19 Sep '17

19 Sep '17

https://bugzilla.redhat.com/show_bug.cgi?id=1448268 When migrating to a file (e.g. when doing 'virsh save file'), couple of things are happening in the thread that is executing the API: 1) the domain obj is locked 2) iohelper is spawned as a separate process to handle all I/O 3) the thread waits for iohelper to finish 4) the domain obj is unlocked Now, the problem is that while the thread waits in step 3 for iohelper to finish this may take ages because iohelper calls fdatasync(). And unfortunately, we are waiting the whole time with the domain locked. So if another thread wants to jump in and say copy the domain name ('virsh list' for instance), they are stuck. The solution is to unlock the domain whenever waiting for I/O and lock it back again when it finished. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/qemu/qemu_driver.c | 27 +++++++++++++++++++++++---- 1 file changed, 23 insertions(+), 4 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index b334cf20b..f81d3def4 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -3216,6 +3216,25 @@ qemuOpenFileAs(uid_t fallback_uid, gid_t fallback_gid, goto cleanup; } + +static int +qemuFileWrapperFDClose(virDomainObjPtr vm, + virFileWrapperFdPtr fd) +{ + int ret; + + /* virFileWrapperFd uses iohelper to write data onto disk. + * However, iohelper calls fdatasync() which may take ages to + * finish. Therefore, we shouldn't be waiting with the domain + * object locked. */ + + virObjectUnlock(vm); + ret = virFileWrapperFdClose(fd); + virObjectLock(vm); + return ret; +} + + /* Helper function to execute a migration to file with a correct save header * the caller needs to make sure that the processors are stopped and do all other * actions besides saving memory */ @@ -3276,7 +3295,7 @@ qemuDomainSaveMemory(virQEMUDriverPtr driver, goto cleanup; } - if (virFileWrapperFdClose(wrapperFd) < 0) + if (qemuFileWrapperFDClose(vm, wrapperFd) < 0) goto cleanup; if ((fd = qemuOpenFile(driver, vm, path, O_WRONLY, NULL, NULL)) < 0 || @@ -3827,7 +3846,7 @@ doCoreDump(virQEMUDriverPtr driver, path); goto cleanup; } - if (virFileWrapperFdClose(wrapperFd) < 0) + if (qemuFileWrapperFDClose(vm, wrapperFd) < 0) goto cleanup; ret = 0; @@ -6687,7 +6706,7 @@ qemuDomainRestoreFlags(virConnectPtr conn, ret = qemuDomainSaveImageStartVM(conn, driver, vm, &fd, data, path, false, QEMU_ASYNC_JOB_START); - if (virFileWrapperFdClose(wrapperFd) < 0) + if (qemuFileWrapperFDClose(vm, wrapperFd) < 0) VIR_WARN("Failed to close %s", path); qemuProcessEndJob(driver, vm); @@ -6962,7 +6981,7 @@ qemuDomainObjRestore(virConnectPtr conn, ret = qemuDomainSaveImageStartVM(conn, driver, vm, &fd, data, path, start_paused, asyncJob); - if (virFileWrapperFdClose(wrapperFd) < 0) + if (qemuFileWrapperFDClose(vm, wrapperFd) < 0) VIR_WARN("Failed to close %s", path); cleanup: -- 2.13.5

2 2

[libvirt] [PATCH v5 0/2] Add support for qcow2 cache
by Liu Qing 18 Sep '17

18 Sep '17

Qcow2 small IO random write performance will drop dramatically if the l2 cache table could not cover the whole disk. This will be a lot of l2 cache table RW operations if cache miss happens frequently. This patch exports the qcow2 driver parameter l2-cache-size/refcount-cache-size, first added in Qemu 2.2, and cache-clean-interval, first added in Qemu 2.5, in libvirt. change since v4: updated doc and did code adjustment based on John Ferlan's review. Liu Qing (2): conf, docs: Add qcow2 cache configuration support qemu: add capability checking for qcow2 cache configuration docs/formatdomain.html.in | 41 +++++++++ docs/schemas/domaincommon.rng | 35 ++++++++ src/conf/domain_conf.c | 96 ++++++++++++++++++++-- src/qemu/qemu_capabilities.c | 11 +++ src/qemu/qemu_capabilities.h | 5 ++ src/qemu/qemu_command.c | 33 ++++++++ src/qemu/qemu_driver.c | 5 ++ src/util/virstoragefile.c | 3 + src/util/virstoragefile.h | 6 ++ tests/qemucapabilitiesdata/caps_2.5.0.x86_64.xml | 3 + .../caps_2.6.0-gicv2.aarch64.xml | 3 + .../caps_2.6.0-gicv3.aarch64.xml | 3 + tests/qemucapabilitiesdata/caps_2.6.0.ppc64le.xml | 3 + tests/qemucapabilitiesdata/caps_2.6.0.x86_64.xml | 3 + tests/qemucapabilitiesdata/caps_2.7.0.s390x.xml | 3 + tests/qemucapabilitiesdata/caps_2.7.0.x86_64.xml | 3 + tests/qemucapabilitiesdata/caps_2.8.0.s390x.xml | 3 + tests/qemucapabilitiesdata/caps_2.8.0.x86_64.xml | 3 + tests/qemucapabilitiesdata/caps_2.9.0.ppc64le.xml | 3 + tests/qemucapabilitiesdata/caps_2.9.0.s390x.xml | 3 + tests/qemucapabilitiesdata/caps_2.9.0.x86_64.xml | 3 + .../qemuxml2argv-disk-drive-qcow2-cache.args | 28 +++++++ .../qemuxml2argv-disk-drive-qcow2-cache.xml | 43 ++++++++++ tests/qemuxml2argvtest.c | 4 + .../qemuxml2xmlout-disk-drive-qcow2-cache.xml | 43 ++++++++++ tests/qemuxml2xmltest.c | 1 + 26 files changed, 385 insertions(+), 5 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-drive-qcow2-cache.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-drive-qcow2-cache.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-disk-drive-qcow2-cache.xml -- 1.8.3.1

3 5

[libvirt] [PATCH 0/2] [V2] Implement LVM volume resize via lvresize
by apolyakov＠beget.ru 18 Sep '17

18 Sep '17

From: Alexander Polyakov <apolyakov(a)beget.com> Includes suggestions by John Ferlan & Peter Krempa Alexander Polyakov (2): logical: implement volume resize news: Document lvm volume resize docs/news.xml | 8 ++++++++ m4/virt-storage-lvm.m4 | 4 ++++ src/storage/storage_backend_logical.c | 37 +++++++++++++++++++++++++++++++++++ 3 files changed, 49 insertions(+) -- 2.13.5

2 4

[libvirt] [PATCH] apparmor: cater for new AAVMF image location
by Guido Günther 18 Sep '17

18 Sep '17

Things moved again, sigh. --- src/security/virt-aa-helper.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/security/virt-aa-helper.c b/src/security/virt-aa-helper.c index 55a686a59c..0b43c8e391 100644 --- a/src/security/virt-aa-helper.c +++ b/src/security/virt-aa-helper.c @@ -516,7 +516,8 @@ valid_path(const char *path, const bool readonly) "/usr/share/OVMF/", /* for OVMF images */ "/usr/share/ovmf/", /* for OVMF images */ "/usr/share/AAVMF/", /* for AAVMF images */ - "/usr/share/qemu-efi/" /* for AAVMF images */ + "/usr/share/qemu-efi/", /* for AAVMF images */ + "/usr/share/qemu-efi-aarch64/" /* for AAVMF images */ }; /* override the above with these */ const char * const override[] = { -- 2.14.1

4 4

[libvirt] [PATCH] apparmor: add attach_disconnected
by Guido Günther 18 Sep '17

18 Sep '17

Otherwise we fail to reconnect to /dev/net/tun opened by libvirtd like [ 8144.507756] audit: type=1400 audit(1505488162.386:38069121): apparmor="DENIED" operation="file_perm" info="Failed name lookup - disconnected path" error=-13 profile="libvirt-5dfcc8a7-b79a-4fa9-a41f-f6271651934c" name="dev/net/tun" pid=9607 comm="qemu-system-x86" requested_mask="r" denied_mask="r" fsuid=117 ouid=0 --- I do wonder why we didn't see this earlier though. examples/apparmor/TEMPLATE.lxc | 2 +- examples/apparmor/TEMPLATE.qemu | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/examples/apparmor/TEMPLATE.lxc b/examples/apparmor/TEMPLATE.lxc index 7b64885a1c..f1005dc575 100644 --- a/examples/apparmor/TEMPLATE.lxc +++ b/examples/apparmor/TEMPLATE.lxc @@ -4,7 +4,7 @@ #include <tunables/global> -profile LIBVIRT_TEMPLATE { +profile LIBVIRT_TEMPLATE flags=(attach_disconnected) { #include <abstractions/libvirt-lxc> # Globally allows everything to run under this profile diff --git a/examples/apparmor/TEMPLATE.qemu b/examples/apparmor/TEMPLATE.qemu index 008a221244..a327315d92 100644 --- a/examples/apparmor/TEMPLATE.qemu +++ b/examples/apparmor/TEMPLATE.qemu @@ -4,6 +4,6 @@ #include <tunables/global> -profile LIBVIRT_TEMPLATE { +profile LIBVIRT_TEMPLATE flags=(attach_disconnected) { #include <abstractions/libvirt-qemu> } -- 2.14.1

4 4

[libvirt] [PATCHv2] nodedev: add switchdev to NIC capabilities
by Edan David 18 Sep '17

18 Sep '17

Adding functionality to libvirt that will allow querying the interface for the availability of switchdev Offloading NIC capabilities. The switchdev mode was introduced in kernel 4.8, the iproute2-devlink command to retrieve the swtichdev NIC feature, Command example: devlink dev eswitch show pci/0000:03:00.0 This feature is needed for Openstack so we can do a scheduling decision if the NIC is in Hardware Offload (switchdev) or regular SR-IOV (legacy) mode. And select the appropriate hypervisors with the requested capability see [1]. [1] - https://specs.openstack.org/openstack/nova-specs/specs/pike/approved/enable… --- configure.ac | 13 ++ docs/formatnode.html.in | 1 + src/util/virnetdev.c | 187 +++++++++++++++++++++- src/util/virnetdev.h | 1 + tests/nodedevschemadata/net_00_13_02_b9_f9_d3.xml | 1 + tests/nodedevschemadata/net_00_15_58_2f_e9_55.xml | 1 + 6 files changed, 203 insertions(+), 1 deletion(-) diff --git a/configure.ac b/configure.ac index b12b7fa..c089798 100644 --- a/configure.ac +++ b/configure.ac @@ -627,6 +627,19 @@ if test "$with_linux" = "yes"; then AC_CHECK_HEADERS([linux/btrfs.h]) fi +dnl +dnl check for kernel headers required by devlink +dnl +if test "$with_linux" = "yes"; then + AC_CHECK_HEADERS([linux/devlink.h]) + AC_CHECK_DECLS([DEVLINK_GENL_VERSION, DEVLINK_GENL_NAME, DEVLINK_ATTR_MAX, DEVLINK_CMD_ESWITCH_GET, DEVLINK_ATTR_BUS_NAME, DEVLINK_ATTR_DEV_NAME, DEVLINK_ATTR_ESWITCH_MODE, DEVLINK_ESWITCH_MODE_SWITCHDEV], + [AC_DEFINE([HAVE_DECL_DEVLINK], + [1], + [whether devlink declarations is available])], + [], + [[#include <linux/devlink.h>]]) +fi + dnl Allow perl/python overrides AC_PATH_PROGS([PYTHON], [python2 python]) if test -z "$PYTHON"; then diff --git a/docs/formatnode.html.in b/docs/formatnode.html.in index 4d935b5..29244a8 100644 --- a/docs/formatnode.html.in +++ b/docs/formatnode.html.in @@ -227,6 +227,7 @@ <dt><code>rxhash</code></dt><dd>receive-hashing</dd> <dt><code>rdma</code></dt><dd>remote-direct-memory-access</dd> <dt><code>txudptnl</code></dt><dd>tx-udp-tunnel-segmentation</dd> + <dt><code>switchdev</code></dt><dd>kernel-forward-plane-offload</dd> </dl> </dd> <dt><code>capability</code></dt> diff --git a/src/util/virnetdev.c b/src/util/virnetdev.c index 51a6e42..fc7c961 100644 --- a/src/util/virnetdev.c +++ b/src/util/virnetdev.c @@ -59,6 +59,10 @@ # include <net/if_dl.h> #endif +#if HAVE_DECL_DEVLINK +# include <linux/devlink.h> +#endif + #ifndef IFNAMSIZ # define IFNAMSIZ 16 #endif @@ -2481,7 +2485,8 @@ VIR_ENUM_IMPL(virNetDevFeature, "ntuple", "rxhash", "rdma", - "txudptnl") + "txudptnl", + "switchdev") #ifdef __linux__ int @@ -2936,6 +2941,7 @@ int virNetDevGetRxFilter(const char *ifname, return ret; } + #if defined(SIOCETHTOOL) && defined(HAVE_STRUCT_IFREQ) /** @@ -3115,6 +3121,182 @@ virNetDevGetEthtoolFeatures(virBitmapPtr bitmap, } +#if HAVE_DECL_DEVLINK +/** + * virNetDevPutExtraHeader + * reserve and prepare room for an extra header + * This function sets to zero the room that is required to put the extra + * header after the initial Netlink header. This function also increases + * the nlmsg_len field. + * + * @nlh: pointer to Netlink header + * @size: size of the extra header that we want to put + * + * Returns pointer to the start of the extended header + */ +static void * +virNetDevPutExtraHeader(struct nlmsghdr *nlh, + size_t size) +{ + char *ptr = (char *)nlh + nlh->nlmsg_len; + size_t len = NLMSG_ALIGN(size); + nlh->nlmsg_len += len; + memset(ptr, 0, len); + return ptr; +} + + +/** + * virNetDevGetFamilyId: + * This function supplies the devlink family id + * + * @family_name: the name of the family to query + * + * Returns family id or 0 on failure. + */ +static uint32_t +virNetDevGetFamilyId(const char *family_name) +{ + struct nl_msg *nl_msg = NULL; + struct nlmsghdr *resp = NULL; + struct genlmsghdr* gmsgh = NULL; + struct nlattr *tb[CTRL_ATTR_MAX + 1] = {NULL, }; + unsigned int recvbuflen; + uint32_t family_id = 0; + + if (!(nl_msg = nlmsg_alloc_simple(GENL_ID_CTRL, + NLM_F_REQUEST | NLM_F_ACK))) { + virReportOOMError(); + goto cleanup; + } + + if (!(gmsgh = virNetDevPutExtraHeader(nlmsg_hdr(nl_msg), sizeof(struct genlmsghdr)))) + goto cleanup; + + gmsgh->cmd = CTRL_CMD_GETFAMILY; + gmsgh->version = DEVLINK_GENL_VERSION; + + if (nla_put_string(nl_msg, CTRL_ATTR_FAMILY_NAME, family_name) < 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, "%s", + _("allocated netlink buffer is too small")); + goto cleanup; + } + + if (virNetlinkCommand(nl_msg, &resp, &recvbuflen, 0, 0, NETLINK_GENERIC, 0) < 0) + goto cleanup; + + if (nlmsg_parse(resp, sizeof(struct nlmsghdr), tb, CTRL_CMD_MAX, NULL) < 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, "%s", + _("malformed netlink response message")); + goto cleanup; + } + + if (tb[CTRL_ATTR_FAMILY_ID] == NULL) + goto cleanup; + + family_id = *(uint32_t *)RTA_DATA(tb[CTRL_ATTR_FAMILY_ID]); + + cleanup: + nlmsg_free(nl_msg); + VIR_FREE(resp); + return family_id; +} + + +/** + * virNetDevSwitchdevFeature + * This function checks for the availability of Switchdev feature + * and add it to bitmap + * + * @ifname: name of the interface + * @out: add Switchdev feature if exist to bitmap + * + * Returns 0 on success, -1 on failure. + */ +static int +virNetDevSwitchdevFeature(const char *ifname, + virBitmapPtr *out) +{ + struct nl_msg *nl_msg = NULL; + struct nlmsghdr *resp = NULL; + unsigned int recvbuflen; + struct nlattr *tb[DEVLINK_ATTR_MAX + 1] = {NULL, }; + virPCIDevicePtr pci_device_ptr = NULL; + struct genlmsghdr* gmsgh = NULL; + const char *pci_name; + char *pfname = NULL; + int is_vf = -1; + int ret = -1; + uint32_t family_id; + + if ((family_id = virNetDevGetFamilyId(DEVLINK_GENL_NAME)) <= 0) + return ret; + + if ((is_vf = virNetDevIsVirtualFunction(ifname)) < 0) + return ret; + + if (is_vf == 1 && virNetDevGetPhysicalFunction(ifname, &pfname) < 0) + goto cleanup; + + if (!(nl_msg = nlmsg_alloc_simple(family_id, + NLM_F_REQUEST | NLM_F_ACK))) { + virReportOOMError(); + goto cleanup; + } + + if (!(gmsgh = virNetDevPutExtraHeader(nlmsg_hdr(nl_msg), sizeof(struct genlmsghdr)))) + goto cleanup; + + gmsgh->cmd = DEVLINK_CMD_ESWITCH_GET; + gmsgh->version = DEVLINK_GENL_VERSION; + + pci_device_ptr = pfname ? virNetDevGetPCIDevice(pfname) : + virNetDevGetPCIDevice(ifname); + if (pci_device_ptr == NULL) + goto cleanup; + + pci_name = virPCIDeviceGetName(pci_device_ptr); + + if (nla_put(nl_msg, DEVLINK_ATTR_BUS_NAME, strlen("pci")+1, "pci") < 0 || + nla_put(nl_msg, DEVLINK_ATTR_DEV_NAME, strlen(pci_name)+1, pci_name) < 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, "%s", + _("allocated netlink buffer is too small")); + goto cleanup; + } + + if (virNetlinkCommand(nl_msg, &resp, &recvbuflen, 0, 0, NETLINK_GENERIC, 0) < 0) + goto cleanup; + + if (nlmsg_parse(resp, sizeof(struct genlmsghdr), tb, DEVLINK_ATTR_MAX, NULL) < 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, "%s", + _("malformed netlink response message")); + goto cleanup; + } + + if (tb[DEVLINK_ATTR_ESWITCH_MODE] && + *(int *)RTA_DATA(tb[DEVLINK_ATTR_ESWITCH_MODE]) == DEVLINK_ESWITCH_MODE_SWITCHDEV) { + ignore_value(virBitmapSetBit(*out, VIR_NET_DEV_FEAT_SWITCHDEV)); + } + + ret = 0; + + cleanup: + nlmsg_free(nl_msg); + virPCIDeviceFree(pci_device_ptr); + VIR_FREE(resp); + VIR_FREE(pfname); + return ret; +} +#else +static int +virNetDevSwitchdevFeature(const char *ifname ATTRIBUTE_UNUSED, + virBitmapPtr *out ATTRIBUTE_UNUSED) +{ + return 0; +} +#endif + + # if HAVE_DECL_ETHTOOL_GFEATURES /** * virNetDevGFeatureAvailable @@ -3315,6 +3497,9 @@ virNetDevGetFeatures(const char *ifname, if (virNetDevRDMAFeature(ifname, out) < 0) goto cleanup; + if (virNetDevSwitchdevFeature(ifname, out) < 0) + goto cleanup; + ret = 0; cleanup: VIR_FORCE_CLOSE(fd); diff --git a/src/util/virnetdev.h b/src/util/virnetdev.h index 9205c0e..71eaf45 100644 --- a/src/util/virnetdev.h +++ b/src/util/virnetdev.h @@ -112,6 +112,7 @@ typedef enum { VIR_NET_DEV_FEAT_RXHASH, VIR_NET_DEV_FEAT_RDMA, VIR_NET_DEV_FEAT_TXUDPTNL, + VIR_NET_DEV_FEAT_SWITCHDEV, VIR_NET_DEV_FEAT_LAST } virNetDevFeature; diff --git a/tests/nodedevschemadata/net_00_13_02_b9_f9_d3.xml b/tests/nodedevschemadata/net_00_13_02_b9_f9_d3.xml index d4c96e8..88252e6 100644 --- a/tests/nodedevschemadata/net_00_13_02_b9_f9_d3.xml +++ b/tests/nodedevschemadata/net_00_13_02_b9_f9_d3.xml @@ -15,6 +15,7 @@ <feature name='rxhash'/> <feature name='rdma'/> <feature name='txudptnl'/> + <feature name='switchdev'/> <capability type='80211'/> </capability> </device> diff --git a/tests/nodedevschemadata/net_00_15_58_2f_e9_55.xml b/tests/nodedevschemadata/net_00_15_58_2f_e9_55.xml index 71bf90e..f77dfcc 100644 --- a/tests/nodedevschemadata/net_00_15_58_2f_e9_55.xml +++ b/tests/nodedevschemadata/net_00_15_58_2f_e9_55.xml @@ -15,6 +15,7 @@ <feature name='rxhash'/> <feature name='rdma'/> <feature name='txudptnl'/> + <feature name='switchdev'/> <capability type='80203'/> </capability> </device> -- 2.1.4

4 11

[libvirt] [PATCH v2 0/3] Be more selective when determining cdrom for taint messaging
by John Ferlan 18 Sep '17

18 Sep '17

v1: https://www.redhat.com/archives/libvir-list/2017-September/msg00103.html Changes since v1: Split into 3 parts... The first patch would be the bare minimum using STRPREFIX instead of STREQ type comparisons for the incoming path to be "/dev/cdrom[N]" or "/dev/srN" (or resolved to that). This would "work" for the most part, but then since it's possible to make even more checks let's check against the collected node device data. Patch 2 therefore will "tag" the already collected cdrom data with a capability. This allows patch3 to find any/all CDROM's on the host and compare the resolved path to that list of devices returning "true" if something matches a node device declared physical CDROM. I split things up mainly to make it easier to decide whether patch 1 is sufficient or not. If patch2 and patch3 are OK, I would also add a release note indicating the improvement to find CDROM by node device capability. It's a separate "improvement" on it's own as well. Whether it's truly useful or not, is a different question... John Ferlan (3): qemu: Be more selective when determining cdrom for taint messaging nodedev: Add capability bit to detect 'cdrom' devices qemu: Add inquiry to nodedev for cdrom taint checking include/libvirt/libvirt-nodedev.h | 1 + src/conf/node_device_conf.c | 6 ++- src/conf/node_device_conf.h | 5 ++- src/conf/virnodedeviceobj.c | 21 ++++++++-- src/node_device/node_device_driver.c | 1 + src/node_device/node_device_udev.c | 2 + src/qemu/qemu_domain.c | 78 +++++++++++++++++++++++++++++++++++- src/qemu/qemu_domain.h | 2 + src/qemu/qemu_driver.c | 4 +- src/qemu/qemu_process.c | 2 +- tools/virsh-nodedev.c | 3 ++ tools/virsh.pod | 2 +- 12 files changed, 116 insertions(+), 11 deletions(-) -- 2.9.5

2 5

[libvirt] [PATCH v3 0/6] Work around the kernel mdev uevent race in nodedev
by Erik Skultety 18 Sep '17

18 Sep '17

v2 here: https://www.redhat.com/archives/libvir-list/2017-July/msg01268.html Since v2: - added patch 4/6 that fixes the issue with the handler thread spamming logs with "udev_monitor_receive_device returned NULL" -> the event loop callback now disables polling on the udev monitor's fd every time there's a new event, leaving the responsibility for re-enabling it back to the handler thread once it had removed the corresponding data from the socket. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1463285 Erik Skultety (6): nodedev: Introduce udevCheckMonitorFD helper function udev: Split udevEventHandleCallback in two functions udev: Convert udevEventHandleThread to an actual thread routine nodedev: Disable/re-enable polling on the udev fd util: Introduce virFileWaitForAccess nodedev: Work around the uevent race by hooking up virFileWaitForAccess src/libvirt_private.syms | 1 + src/node_device/node_device_udev.c | 208 ++++++++++++++++++++++++++++++++----- src/util/virfile.c | 29 ++++++ src/util/virfile.h | 2 + 4 files changed, 214 insertions(+), 26 deletions(-) -- 2.13.3

2 22

[libvirt] [PATCH go-xml] Add support for domain hostdev and test code
by zhenwei.pi 18 Sep '17

18 Sep '17

Signed-off-by: zhenwei.pi <zhenwei.pi(a)youruncloud.com> --- domain.go | 36 ++++++++++++++++++++++++++++++++++++ domain_test.go | 44 ++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 80 insertions(+) diff --git a/domain.go b/domain.go index bead49a..1bcc9cc 100644 --- a/domain.go +++ b/domain.go @@ -407,6 +407,29 @@ type DomainRNG struct { Backend *DomainRNGBackend `xml:"backend"` } +type DomainHostdevAdapter struct { + Name string `xml:"name,attr,omitempty"` +} + +type DomainHostdevSource struct { + Protocol string `xml:"protocol,attr,omitempty"` + Name string `xml:"name,attr,omitempty"` + Wwpn string `xml:"wwpn,attr,omitempty"` + Adapter *DomainHostdevAdapter `xml:"adapter"` + Address *DomainAddress `xml:"address"` +} + +type DomainHostdev struct { + XMLName xml.Name `xml:"hostdev"` + Mode string `xml:"mode,attr"` + Type string `xml:"type,attr"` + Sgio string `xml:"sgio,attr,omitempty"` + Rawio string `xml:"rawio,attr,omitempty"` + Managed string `xml:"managed,attr,omitempty"` + Source *DomainHostdevSource `xml:"source"` + Address *DomainAddress `xml:"address"` +} + type DomainDeviceList struct { Emulator string `xml:"emulator,omitempty"` Controllers []DomainController `xml:"controller"` @@ -422,6 +445,7 @@ type DomainDeviceList struct { MemBalloon *DomainMemBalloon `xml:"memballoon"` Sounds []DomainSound `xml:"sound"` RNGs []DomainRNG `xml:"rng"` + Hostdevs []DomainHostdev `xml:"hostdev"` } type DomainMemory struct { @@ -794,6 +818,18 @@ func (d *DomainRNG) Marshal() (string, error) { return string(doc), nil } +func (d *DomainHostdev) Unmarshal(doc string) error { + return xml.Unmarshal([]byte(doc), d) +} + +func (d *DomainHostdev) Marshal() (string, error) { + doc, err := xml.MarshalIndent(d, "", " ") + if err != nil { + return "", err + } + return string(doc), nil +} + type HexUint uint func (h *HexUint) UnmarshalXMLAttr(attr xml.Attr) error { diff --git a/domain_test.go b/domain_test.go index d1b107d..73dd47b 100644 --- a/domain_test.go +++ b/domain_test.go @@ -37,6 +37,13 @@ type Address struct { Function HexUint } +type ScsiAddress struct { + Controller uint + Bus HexUint + Target uint + Unit uint +} + var uhciIndex uint = 0 var uhciAddr = Address{0, 0, 1, 2} @@ -46,6 +53,7 @@ var videoAddr = Address{0, 0, 5, 0} var fsAddr = Address{0, 0, 6, 0} var balloonAddr = Address{0, 0, 7, 0} var duplexAddr = Address{0, 0, 8, 0} +var hostdevScsi = ScsiAddress{0, 0, 3, 0} var serialPort uint = 0 var tabletBus HexUint = 0 @@ -1457,6 +1465,42 @@ var domainTestData = []struct { `</rng>`, }, }, + { + Object: &DomainHostdev{ + Mode: "subsystem", + Type: "scsi", + Sgio: "unfiltered", + Rawio: "yes", + Source: &DomainHostdevSource{ + Adapter: &DomainHostdevAdapter{ + Name: "scsi_host0", + }, + Address: &DomainAddress{ + Type: "scsi", + Bus: &hostdevScsi.Bus, + Target: &hostdevScsi.Target, + Unit: &hostdevScsi.Unit, + }, + }, + Address: &DomainAddress{ + Type: "drive", + Controller: &hostdevScsi.Controller, + Bus: &hostdevScsi.Bus, + Target: &hostdevScsi.Target, + Unit: &hostdevScsi.Unit, + }, + }, + + Expected: []string{ + `<hostdev mode="subsystem" type="scsi" sgio="unfiltered" rawio="yes">`, + ` <source>`, + ` <adapter name="scsi_host0"></adapter>`, + ` <address type="scsi" bus="0" target="3" unit="0"></address>`, + ` </source>`, + ` <address type="drive" controller="0" bus="0" target="3" unit="0"></address>`, + `</hostdev>`, + }, + }, } func TestDomain(t *testing.T) { -- 2.7.4

2 1

[libvirt] Questions about function virPCIDeviceIsBehindSwitchLackingACS in virpci.c
by Wuzongyong (Euler Dept) 18 Sep '17

18 Sep '17

In function virPCIDeviceIsBehindSwitchLackingACS, I noticed that(line 8): 1 if (virPCIDeviceGetParent(dev, &parent) < 0) 2 return -1; 3 if (!parent) { 4 /* if we have no parent, and this is the root bus, ACS doesn't come 5 * into play since devices on the root bus can't P2P without going 6 * through the root IOMMU. 7 */ 8 if (dev->address.bus == 0) { 9 return 0; 10 } else { 11 virReportError(VIR_ERR_INTERNAL_ERROR, 12 _("Failed to find parent device for %s"), 13 dev->name); 14 return -1; 15 } 16 } Why we just return 0 only if device's bus is 0? In my server, I can see a root bus which bus number is greater than 0, see the results(just a part) after I run lspci -t: +-[0000:80]-+-02.0-[81-83]--+-00.0 | | \-00.1 | +-05.0 | +-05.1 | +-05.2 | \-05.4 +-[0000:7f]-+-08.0 | +-08.2 | +-08.3 | + . . . | \-1f.2 \-[0000:00]-+-00.0 +-01.0-[01]----00.0 +-02.0-[02]--+-00.0 | +-00.1 | +-00.2 | \-00.3 +-02.2-[03]-- +-03.0-[04-0b]----00.0-[05-0b]--+-08.0-[06-08]----00.0 | \-10.0-[09-0b]----00.0 +-05.0 +-05.1 +-05.2 +-05.4 +-11.0 +-11.4 +-16.0 +-16.1 +-1a.0 If I assign the device 0000:81:00.0 to a VM, I get "Failed to find parent device". I think I should get no error with return value 0 just like bus number is 0, because bus 80 is the root bus as well in my case. In the <<Intel C610 Series Chipset and Intel X99 Chipset Platform Controller Hub(PCH)>> Datasheet, I found that(Chapter 9.1): For some server platforms, it may be desirable to have multiple PCHs in the system Which means some PCH's may reside on a bus greater than 0. So, is this a bug? Thanks, Zongyong Wu

1 0