- Devel - libvirt List Archives

[libvirt] [PATCH 0/6] Couple of storage driver improvements
by Michal Privoznik 06 Mar '19

06 Mar '19

*** BLURB HERE *** Michal Prívozník (6): storage_backend_iscsi_direct: Simplify vol zeroing virISCSIDirectReportLuns: Drop ClearVols storageVolWipePattern: Don't take shortcut to refreshPool() storage_driver: Introduce storagePoolRefreshImpl() storagePoolRefreshFailCleanup: Clear volumes on failed refresh virsh-pool: Offer only active pool for pool-refresh completer src/storage/storage_backend_gluster.c | 2 - src/storage/storage_backend_iscsi_direct.c | 21 ++---- src/storage/storage_backend_logical.c | 12 +--- src/storage/storage_backend_rbd.c | 4 +- src/storage/storage_driver.c | 77 ++++++++++++---------- src/storage/storage_util.c | 2 - tools/virsh-pool.c | 2 +- 7 files changed, 55 insertions(+), 65 deletions(-) -- 2.19.2

1 7

[libvirt] [PATCH] qemu: Don't set migration caps when changing postcopy bandwidth
by Jiri Denemark 06 Mar '19

06 Mar '19

The qemuMigrationParamsApply internal API was designed to apply all migration parameters and capabilities before we start to migrate a domain. The API can also be called outside migration job, e.g., via a call to qemuDomainMigrateSetMaxSpeed with VIR_DOMAIN_MIGRATE_MAX_SPEED_POSTCOPY flag, but in that case it should only apply migration parameters and ignore capabilities. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- Notes: Best viewed with --ignore-all-space src/qemu/qemu_migration_params.c | 23 ++++++++++++++++------- 1 file changed, 16 insertions(+), 7 deletions(-) diff --git a/src/qemu/qemu_migration_params.c b/src/qemu/qemu_migration_params.c index 24e5368783..623193cf6c 100644 --- a/src/qemu/qemu_migration_params.c +++ b/src/qemu/qemu_migration_params.c @@ -778,14 +778,23 @@ qemuMigrationParamsApply(virQEMUDriverPtr driver, if (qemuDomainObjEnterMonitorAsync(driver, vm, asyncJob) < 0) return -1; - if (!(caps = qemuMigrationCapsToJSON(priv->migrationCaps, migParams->caps))) - goto cleanup; - - if (virJSONValueArraySize(caps) > 0) { - rc = qemuMonitorSetMigrationCapabilities(priv->mon, caps); - caps = NULL; - if (rc < 0) + if (asyncJob == QEMU_ASYNC_JOB_NONE) { + if (!virBitmapIsAllClear(migParams->caps)) { + virReportError(VIR_ERR_INTERNAL_ERROR, "%s", + _("Migration capabilities can only be set by " + "a migration job")); goto cleanup; + } + } else { + if (!(caps = qemuMigrationCapsToJSON(priv->migrationCaps, migParams->caps))) + goto cleanup; + + if (virJSONValueArraySize(caps) > 0) { + rc = qemuMonitorSetMigrationCapabilities(priv->mon, caps); + caps = NULL; + if (rc < 0) + goto cleanup; + } } /* If QEMU is too old to support xbzrle-cache-size migration parameter, -- 2.21.0

3 5

[libvirt] [PATCH] rpm: add dep on xfsprogs-devel for reflink support
by Daniel P. Berrangé 06 Mar '19

06 Mar '19

Support for XFS reflink clone was added in: commit 8ed874b39b3c330bbcdff434e08995dbb4467285 Author: Julio Faracco <jcfaracco(a)gmail.com> Date: Fri Jul 6 10:43:01 2018 -0300 storage: Rename btrfsCloneFile to support other filesystems. commit 2e11298f938672c5430193277c98b9474a68b1fb Author: Julio Faracco <jcfaracco(a)gmail.com> Date: Fri Jul 6 10:43:00 2018 -0300 configure: Adding XFS library/headers check. But these patches missed that the xfs/xfs.h header is not installed unless you have xfsprogs-devel present. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- libvirt.spec.in | 2 ++ 1 file changed, 2 insertions(+) diff --git a/libvirt.spec.in b/libvirt.spec.in index 9beffba203..83b43f4ef6 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -326,6 +326,8 @@ BuildRequires: libiscsi-devel BuildRequires: parted-devel # For Multipath support BuildRequires: device-mapper-devel +# For XFS reflink clone support +BuildRequires: xfsprogs-devel %if %{with_storage_rbd} BuildRequires: librados2-devel BuildRequires: librbd1-devel -- 2.20.1

2 1

[libvirt] [PATCH 1/2] security: aa-helper: nvidia rules for gl devices
by Christian Ehrhardt 06 Mar '19

06 Mar '19

Further testing with different devices showed that we need more rules to drive gl backends with nvidia cards. Related denies look like: apparmor="DENIED" operation="open" name="/usr/share/egl/egl_external_platform.d/" requested_mask="r" apparmor="DENIED" operation="open" name="/proc/modules" requested_mask="r" apparmor="DENIED" operation="open" name="/proc/driver/nvidia/params" requested_mask="r" apparmor="DENIED" operation="mknod" name="/dev/nvidiactl" requested_mask="c" Fixes: https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/1817943 Signed-off-by: Christian Ehrhardt <christian.ehrhardt(a)canonical.com> --- src/security/virt-aa-helper.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/src/security/virt-aa-helper.c b/src/security/virt-aa-helper.c index e9120213ff..13b507ff69 100644 --- a/src/security/virt-aa-helper.c +++ b/src/security/virt-aa-helper.c @@ -1279,6 +1279,11 @@ get_files(vahControl * ctl) virBufferAddLit(&buf, " \"/usr/share/drirc.d/{,*.conf}\" r,\n"); virBufferAddLit(&buf, " \"/etc/glvnd/egl_vendor.d/{,*}\" r,\n"); virBufferAddLit(&buf, " \"/usr/share/glvnd/egl_vendor.d/{,*}\" r,\n"); + virBufferAddLit(&buf, " \"/usr/share/egl/egl_external_platform.d/\" r,\n"); + virBufferAddLit(&buf, " \"/usr/share/egl/egl_external_platform.d/*\" r,\n"); + virBufferAddLit(&buf, " \"/proc/modules\" r,\n"); + virBufferAddLit(&buf, " \"/proc/driver/nvidia/params\" r,\n"); + virBufferAddLit(&buf, " \"/dev/nvidiactl\" rw,\n"); virBufferAddLit(&buf, " # Probe DRI device attributes\n"); virBufferAddLit(&buf, " \"/dev/dri/\" r,\n"); virBufferAddLit(&buf, " \"/sys/devices/*/*/{uevent,vendor,device,subsystem_vendor,subsystem_device}\" r,\n"); -- 2.17.1

2 5

[libvirt] [PATCH for-4.0 v4 0/2] virtio: Provide version-specific variants of virtio PCI devices
by Eduardo Habkost 06 Mar '19

06 Mar '19

Existing modern-only device types are not being touched by v3, as they don't need separate variants. However, I plan to implement separate cleanups in the code that calls virtio_pci_force_virtio_1(), first, and then propose additional changes (e.g. deprecating disable-legacy and disable-modern in those device types). Changes v3 -> v4: * Trivial comment fixes (Cornelia Huck) * Test code update (Caio Carrara) Changes v2 -> v3: * Split into two separate patches (type registration helper and introduction of new types) * Rewrote virtio_pci_types_register() completely: * Replaced magic generation of type names with explicit fields in VirtioPCIDeviceTypeInfo * Removed modern_only field (not necessary anymore) * Don't register a separate base type unless necessary Changes v1 -> v2: * Removed *-0.9 devices. Nobody will want to use them, if transitional devices work with legacy drivers (Gerd Hoffmann, Michael S. Tsirkin) * Drop virtio version from name: rename -1.0-transitional to -transitional (Michael S. Tsirkin) * Renamed -1.0 to -non-transitional * Don't add any extra variants to modern-only device types (they don't need it) * Fix typo on TYPE_VIRTIO_INPUT_HOST_PCI (crash reported by Cornelia Huck) * No need to change cast macros for modern-only devices * Rename virtio_register_types() to virtio_pci_types_register() Original patch description: Many of the current virtio-*-pci device types actually represent 3 different types of devices: * virtio 1.0 non-transitional devices * virtio 1.0 transitional devices * virtio 0.9 ("legacy device" in virtio 1.0 terminology) That would be just an annoyance if it didn't break our device/bus compatibility QMP interfaces. With this multi-purpose device type, there's no way to tell management software that transitional devices and legacy devices require a Conventional PCI bus. The multi-purpose device types would also prevent us from telling management software what's the PCI vendor/device ID for them, because their PCI IDs change at runtime depending on the bus where they were plugged. This patch adds separate device types for each of those virtio device flavors: * virtio-*-pci: the existing multi-purpose device types * virtio-*-pci-transitional: virtio-1.0 device supporting legacy drivers * virtio-*-pci-non-transitional: modern-only Reference to previous discussion that originated this idea: https://www.mail-archive.com/qemu-devel@nongnu.org/msg558389.html Eduardo Habkost (2): virtio: Helper for registering virtio device types virtio: Provide version-specific variants of virtio PCI devices hw/virtio/virtio-pci.h | 78 +++++++-- hw/display/virtio-gpu-pci.c | 7 +- hw/display/virtio-vga.c | 7 +- hw/virtio/virtio-crypto-pci.c | 7 +- hw/virtio/virtio-pci.c | 267 ++++++++++++++++++++++------- tests/acceptance/virtio_version.py | 176 +++++++++++++++++++ 6 files changed, 452 insertions(+), 90 deletions(-) create mode 100644 tests/acceptance/virtio_version.py -- 2.18.0.rc1.1.g3f1ff2140

9 13

[libvirt] [PATCH] libxl: Add support for max_grant_frame setting
by Jim Fehlig 06 Mar '19

06 Mar '19

Xen 4.10 introduced the max_grant_frames xl config setting, which can be set globally in xl.conf(5) or per-domain in xl.cfg(5). max_grant_frames specifies the maximum number of grant frames the domain is allowed to have, which in turn controls the number of pages the domain can share. This patch adds support for setting max_grant_frames on a global level in libxl.conf. Per-domain support via domXML can be provided in a future patch. Signed-off-by: Jim Fehlig <jfehlig(a)suse.com> --- src/libxl/libvirtd_libxl.aug | 2 ++ src/libxl/libxl.conf | 8 ++++++++ src/libxl/libxl_conf.c | 9 +++++++++ src/libxl/libxl_conf.h | 2 ++ src/libxl/test_libvirtd_libxl.aug.in | 1 + 5 files changed, 22 insertions(+) diff --git a/src/libxl/libvirtd_libxl.aug b/src/libxl/libvirtd_libxl.aug index 58b9af3707..1c3e2719fc 100644 --- a/src/libxl/libvirtd_libxl.aug +++ b/src/libxl/libvirtd_libxl.aug @@ -29,6 +29,7 @@ module Libvirtd_libxl = let keepalive_interval_entry = int_entry "keepalive_interval" let keepalive_count_entry = int_entry "keepalive_count" let nested_hvm_entry = bool_entry "nested_hvm" + let max_grant_frames_entry = int_entry "max_grant_frames" (* Each entry in the config is one of the following ... *) let entry = autoballoon_entry @@ -36,6 +37,7 @@ module Libvirtd_libxl = | keepalive_interval_entry | keepalive_count_entry | nested_hvm_entry + | max_grant_frames_entry let comment = [ label "#comment" . del /#[ \t]*/ "# " . store /([^ \t\n][^\n]*)?/ . del /\n/ "\n" ] let empty = [ label "#empty" . eol ] diff --git a/src/libxl/libxl.conf b/src/libxl/libxl.conf index 72825a71c5..12d7abeda6 100644 --- a/src/libxl/libxl.conf +++ b/src/libxl/libxl.conf @@ -49,3 +49,11 @@ # element. # By default it is disabled. #nested_hvm = 0 + +# Xen domains can grant other domains access to its memory pages, which is +# most commonly used for the operation of paravirtualized devices. +# max_grant_frames was introduced in Xen 4.10 and can be used to specify a +# default value for the maximum number of grant frames each domain started +# by libvirt is allowed to have. The default value is 32. +# +#max_grant_frames = 32 diff --git a/src/libxl/libxl_conf.c b/src/libxl/libxl_conf.c index c769050ff1..f57ffe1a01 100644 --- a/src/libxl/libxl_conf.c +++ b/src/libxl/libxl_conf.c @@ -393,6 +393,12 @@ libxlMakeDomBuildInfo(virDomainDefPtr def, def->mem.cur_balloon = VIR_ROUND_UP(def->mem.cur_balloon, 1024); b_info->max_memkb = virDomainDefGetMemoryInitial(def); b_info->target_memkb = def->mem.cur_balloon; + +#ifdef LIBXL_HAVE_BUILDINFO_GRANT_LIMITS + if (cfg->max_grant_frames > 0) + b_info->max_grant_frames = cfg->max_grant_frames; +#endif + if (hvm || pvh) { if (caps && def->cpu && def->cpu->mode == (VIR_CPU_MODE_HOST_PASSTHROUGH)) { @@ -1888,6 +1894,9 @@ int libxlDriverConfigLoadFile(libxlDriverConfigPtr cfg, if (virConfGetValueBool(conf, "nested_hvm", &cfg->nested_hvm) < 0) goto cleanup; + if (virConfGetValueUInt(conf, "max_grant_frames", &cfg->max_grant_frames) < 0) + goto cleanup; + ret = 0; cleanup: diff --git a/src/libxl/libxl_conf.h b/src/libxl/libxl_conf.h index fee94241af..985524f3eb 100644 --- a/src/libxl/libxl_conf.h +++ b/src/libxl/libxl_conf.h @@ -86,6 +86,8 @@ struct _libxlDriverConfig { bool nested_hvm; + unsigned int max_grant_frames; + /* Once created, caps are immutable */ virCapsPtr caps; diff --git a/src/libxl/test_libvirtd_libxl.aug.in b/src/libxl/test_libvirtd_libxl.aug.in index 372a43f94a..754ddc10f6 100644 --- a/src/libxl/test_libvirtd_libxl.aug.in +++ b/src/libxl/test_libvirtd_libxl.aug.in @@ -7,3 +7,4 @@ module Test_libvirtd_libxl = { "keepalive_interval" = "5" } { "keepalive_count" = "5" } { "nested_hvm" = "0" } +{ "max_grant_frames" = "32" } -- 2.20.1

2 3

[libvirt] domXML modeling question
by Jim Fehlig 06 Mar '19

06 Mar '19

There have been a few requests [1][2] to support Xen's max_grant_frames setting in libvirt domXML, but I'm not quite sure how to model it. The documentation [3] on this setting states: Specify the maximum number of grant frames the domain is allowed to have. This value controls how many pages the domain is able to grant access to for other domains, needed e.g. for the operation of paravirtualized devices. The default is settable via xl.conf(5). It smells of a <memtune> setting, e.g. the amount of memory a domain can share, but doesn't map to any of the existing settings. A new subelement <shared_limit> doesn't feel right. Does anyone suggest a better way of modeling max_grant_frames? Another option I considered is setting the value based on number of PV devices, but I think that flies in the face of libvirt's policy of not dictating policy. Regardless of domain config modeling I can work on a driver-wide setting in libxl.conf, similar to Xen's xl.conf(5) global. Regards, Jim [1] https://www.redhat.com/archives/libvir-list/2018-April/msg00216.html [2] https://www.redhat.com/archives/libvirt-users/2019-January/msg00011.html [3] http://xenbits.xen.org/gitweb/?p=xen.git;a=blob;f=docs/man/xl.cfg.5.pod.in;…

3 3

[libvirt] [PATCH 0/4] Couple of patches about virsh completion
by Lin Ma 06 Mar '19

06 Mar '19

Lin Ma (4): virsh: Only return active domain names for detach-device-alias virsh: Add device name completion for target option of detach-disk command virsh-network: Introduce virshNetworkEventCallback to handle network events virsh: Add event name completion to 'network-event' command tools/virsh-completer.c | 27 +++++++++++++++++++++++++++ tools/virsh-completer.h | 4 ++++ tools/virsh-domain.c | 3 ++- tools/virsh-network.c | 18 +++++++++++++++--- tools/virsh-network.h | 8 ++++++++ 5 files changed, 56 insertions(+), 4 deletions(-) -- 2.19.0

2 10

[libvirt] [PATCH v1 00/15] Firmware auto selection
by Michal Privoznik 05 Mar '19

05 Mar '19

Libvirt allows specifying firmware for domains for quite some time now. However, problem for mgmt applications is that they do not know which firmware to chose as all they see are their paths and from that it's impossible to tell whether one of them supports say secure boot. This problem was addressed by qemu where Lazslo and Daniel created a document, specification which describes metadata for each individual firmware image. In the description (which itself is a JSON file for easy machine parsing) then it's specified whether the firmware it's describing supports secureboot, s3/s4 states, it it's bios or efi, and so on. These patches take advantage of that, and even though the description files are not picked up by that many distributions yet, it allows users to not care about putting specific firmware path into their domain XML. It's as easy as: <os firmware='efi'> <loader secure='yes'/> </os> to have libvirt pick up OVMF image with secure enabled boot (and enabled System Management Mode at the same time). The metadata specification lives under qemu.git/docs/interop/firmware.json and I highly recommend you go and read it before reviewing (unless you're Laszlo or Daniel in which case you already know what the document says). As usual, you can find my patches at my github: https://github.com/zippy2/libvirt/commits/firmware_v1 Michal Prívozník (15): virmock: Initialize both symbols in VIR_MOCK_REAL_INIT_ALT qemu_domain: Separate NVRAM VAR store file name generation qemu_capabilities: Expose qemu <-> libvirt arch translators virDomainLoaderDefParseXML: Allow loader path to be NULL conf: Introduce VIR_DOMAIN_LOADER_TYPE_NONE conf: Introduce firmware attribute to <os/> qemu: Introduce basic skeleton for parsing firmware description test: Introduce qemufirmwaretest qemu_firmware: Introduce qemuFirmwareFetchConfigs qemufirmwaretest: Test qemuFirmwareFetchConfigs() qemu_firmware: Introduce qemuFirmwareFillDomain() qemu_process: Call qemuFirmwareFillDomain qemuDomainDefValidate: Don't require SMM if automatic firmware selection enabled qemu: Enable firmware autoselection qemuxml2argvtest: Test os.firmware autoselection docs/formatdomain.html.in | 22 +- docs/schemas/domaincommon.rng | 12 +- src/conf/domain_conf.c | 113 +- src/conf/domain_conf.h | 15 +- src/libvirt_private.syms | 2 + src/qemu/Makefile.inc.am | 2 + src/qemu/qemu_capabilities.c | 4 +- src/qemu/qemu_capabilities.h | 3 + src/qemu/qemu_command.c | 1 + src/qemu/qemu_domain.c | 34 +- src/qemu/qemu_domain.h | 4 + src/qemu/qemu_firmware.c | 1285 +++++++++++++++++ src/qemu/qemu_firmware.h | 50 + src/qemu/qemu_process.c | 5 + tests/Makefile.am | 14 +- tests/domaincapsschemadata/full.xml | 1 + .../etc/qemu/firmware/40-ovmf-sb.json | 1 + .../etc/qemu/firmware/60-ovmf.json | 0 .../user/.config/qemu/firmware/10-bios.json | 0 .../usr/share/qemu/firmware/40-bios.json | 35 + .../usr/share/qemu/firmware/50-ovmf-sb.json | 36 + .../usr/share/qemu/firmware/60-ovmf.json | 35 + .../usr/share/qemu/firmware/70-aavmf.json | 35 + tests/qemufirmwaretest.c | 129 ++ ...arch64-os-firmware-efi.aarch64-latest.args | 37 + .../aarch64-os-firmware-efi.xml | 30 + .../os-firmware-bios.x86_64-latest.args | 39 + tests/qemuxml2argvdata/os-firmware-bios.xml | 68 + ...os-firmware-efi-secboot.x86_64-latest.args | 42 + .../os-firmware-efi-secboot.xml | 68 + .../os-firmware-efi.x86_64-latest.args | 42 + tests/qemuxml2argvdata/os-firmware-efi.xml | 68 + tests/qemuxml2argvtest.c | 17 + .../aarch64-os-firmware-efi.xml | 1 + tests/qemuxml2xmloutdata/os-firmware-bios.xml | 1 + .../os-firmware-efi-secboot.xml | 1 + tests/qemuxml2xmloutdata/os-firmware-efi.xml | 1 + tests/qemuxml2xmltest.c | 27 + tests/virmock.h | 5 +- 39 files changed, 2255 insertions(+), 30 deletions(-) create mode 100644 src/qemu/qemu_firmware.c create mode 100644 src/qemu/qemu_firmware.h create mode 120000 tests/qemufirmwaredata/etc/qemu/firmware/40-ovmf-sb.json create mode 100644 tests/qemufirmwaredata/etc/qemu/firmware/60-ovmf.json create mode 100644 tests/qemufirmwaredata/home/user/.config/qemu/firmware/10-bios.json create mode 100644 tests/qemufirmwaredata/usr/share/qemu/firmware/40-bios.json create mode 100644 tests/qemufirmwaredata/usr/share/qemu/firmware/50-ovmf-sb.json create mode 100644 tests/qemufirmwaredata/usr/share/qemu/firmware/60-ovmf.json create mode 100644 tests/qemufirmwaredata/usr/share/qemu/firmware/70-aavmf.json create mode 100644 tests/qemufirmwaretest.c create mode 100644 tests/qemuxml2argvdata/aarch64-os-firmware-efi.aarch64-latest.args create mode 100644 tests/qemuxml2argvdata/aarch64-os-firmware-efi.xml create mode 100644 tests/qemuxml2argvdata/os-firmware-bios.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/os-firmware-bios.xml create mode 100644 tests/qemuxml2argvdata/os-firmware-efi-secboot.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/os-firmware-efi-secboot.xml create mode 100644 tests/qemuxml2argvdata/os-firmware-efi.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/os-firmware-efi.xml create mode 120000 tests/qemuxml2xmloutdata/aarch64-os-firmware-efi.xml create mode 120000 tests/qemuxml2xmloutdata/os-firmware-bios.xml create mode 120000 tests/qemuxml2xmloutdata/os-firmware-efi-secboot.xml create mode 120000 tests/qemuxml2xmloutdata/os-firmware-efi.xml -- 2.19.2

3 37

[libvirt] [PATCH] docs: Fix version number
by Andrea Bolognani 05 Mar '19

05 Mar '19

virtio-(non-)transitional device models have been introduced in 5.2.0, not 5.1.0. Signed-off-by: Andrea Bolognani <abologna(a)redhat.com> --- Pushed as trivial. docs/formatdomain.html.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index f9574f0b24..eb00c01d96 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -4127,7 +4127,7 @@ device when plugged into a PCI Express slot, and like a <code>virtio-transitional</code> device otherwise; libvirt will pick one or the other based on the machine type. This is the best - choice when compatibility with libvirt versions older than 5.1.0 + choice when compatibility with libvirt versions older than 5.2.0 is necessary, but it's otherwise not recommended to use it. </dd> </dl> -- 2.20.1

1 0