- Devel - libvirt List Archives

[PATCH 0/2] test_driver: Implement some Security related APIs
by Luke Yue 15 Jun '21

15 Jun '21

Luke Yue (2): test_driver: Implement virNodeGetSecurityModel test_driver: Implement virDomainGetSecurityLabel src/test/test_driver.c | 63 ++++++++++++++++++++++++++++++++++++++++++ tests/virshtest.c | 4 +++ 2 files changed, 67 insertions(+) -- 2.31.1

2 4

[PATCH v2 0/2] Replace some libvirt handling function with GLib APIs
by Luke Yue 15 Jun '21

15 Jun '21

Compared to original: - Ensure the virFindFileInPath() return an absolute path (tried to use glibcompat, failed to solve some Windows related issue) - Just remove virFileAbsPath() and use g_canonicalize_filename() instead Luke Yue (2): Replace virFileAbsPath() with g_canonicalize_filename() virfile: Simplify virFindFileInPath() with g_find_program_in_path() src/libvirt-domain.c | 16 +++++----- src/libvirt_private.syms | 1 - src/util/virfile.c | 69 ++-------------------------------------- src/util/virfile.h | 3 -- src/util/virlog.c | 2 +- 5 files changed, 12 insertions(+), 79 deletions(-) -- 2.31.1

2 4

[libvirt PATCH 0/5] use confidential-guest-support if available
by Pavel Hrdina 14 Jun '21

14 Jun '21

Pavel Hrdina (5): qemu_capabilities: detect if confidential-guest-support is available qemu_command: use confidential-guest-support if available tests: add AMD SEV bits into qemu-6.1.0 replies tests: use newer machine in qemuxml2argv launch-security-sev tests: add launch-security-sev test with latest capabilities src/qemu/qemu_capabilities.c | 14 +- src/qemu/qemu_capabilities.h | 3 + src/qemu/qemu_command.c | 9 +- .../domaincapsdata/qemu_6.1.0-q35.x86_64.xml | 5 +- .../domaincapsdata/qemu_6.1.0-tcg.x86_64.xml | 5 +- tests/domaincapsdata/qemu_6.1.0.x86_64.xml | 5 +- .../caps_2.12.0.aarch64.replies | 137 ++++++++++++++-- .../caps_2.12.0.ppc64.replies | 133 +++++++++++++-- .../caps_2.12.0.s390x.replies | 137 ++++++++++++++-- .../caps_2.12.0.x86_64.replies | 153 +++++++++++++++--- .../caps_3.0.0.ppc64.replies | 133 +++++++++++++-- .../caps_3.0.0.riscv32.replies | 129 ++++++++++++++- .../caps_3.0.0.riscv64.replies | 129 ++++++++++++++- .../caps_3.0.0.s390x.replies | 137 ++++++++++++++-- .../caps_3.0.0.x86_64.replies | 153 +++++++++++++++--- .../caps_3.1.0.ppc64.replies | 133 +++++++++++++-- .../caps_3.1.0.x86_64.replies | 153 +++++++++++++++--- .../caps_4.0.0.aarch64.replies | 137 ++++++++++++++-- .../caps_4.0.0.ppc64.replies | 133 +++++++++++++-- .../caps_4.0.0.riscv32.replies | 129 ++++++++++++++- .../caps_4.0.0.riscv64.replies | 129 ++++++++++++++- .../caps_4.0.0.s390x.replies | 137 ++++++++++++++-- .../caps_4.0.0.x86_64.replies | 153 +++++++++++++++--- .../caps_4.1.0.x86_64.replies | 145 +++++++++++++++-- .../caps_4.2.0.aarch64.replies | 141 ++++++++++++++-- .../caps_4.2.0.ppc64.replies | 133 +++++++++++++-- .../caps_4.2.0.s390x.replies | 137 ++++++++++++++-- .../caps_4.2.0.x86_64.replies | 145 +++++++++++++++-- .../caps_5.0.0.aarch64.replies | 126 +++++++++++++-- .../caps_5.0.0.ppc64.replies | 118 ++++++++++++-- .../caps_5.0.0.riscv64.replies | 114 ++++++++++++- .../caps_5.0.0.x86_64.replies | 130 +++++++++++++-- .../caps_5.1.0.sparc.replies | 114 ++++++++++++- .../caps_5.1.0.x86_64.replies | 130 +++++++++++++-- .../caps_5.2.0.aarch64.replies | 121 ++++++++++++-- .../caps_5.2.0.ppc64.replies | 113 +++++++++++-- .../caps_5.2.0.riscv64.replies | 109 ++++++++++++- .../caps_5.2.0.s390x.replies | 117 ++++++++++++-- .../caps_5.2.0.x86_64.replies | 125 ++++++++++++-- .../caps_6.0.0.s390x.replies | 130 +++++++++++++-- .../qemucapabilitiesdata/caps_6.0.0.s390x.xml | 1 + .../caps_6.0.0.x86_64.replies | 138 ++++++++++++++-- .../caps_6.0.0.x86_64.xml | 1 + .../caps_6.1.0.x86_64.replies | 148 ++++++++++++++--- .../caps_6.1.0.x86_64.xml | 8 + .../launch-security-sev.x86_64-2.12.0.args | 2 +- .../launch-security-sev.x86_64-latest.args | 36 +++++ .../qemuxml2argvdata/launch-security-sev.xml | 2 +- tests/qemuxml2argvtest.c | 1 + 49 files changed, 4403 insertions(+), 468 deletions(-) create mode 100644 tests/qemuxml2argvdata/launch-security-sev.x86_64-latest.args -- 2.31.1

2 13

[PATCH 0/3] Couple of cleanups
by Michal Privoznik 14 Jun '21

14 Jun '21

I've noticed these while reviewing Tim's asan patches. Michal Prívozník (3): Don't call qsort() over NULL tests: Don't pass INT_MAX to virFileReadAll() src: Use 1U for bit shifting src/conf/capabilities.c | 6 ++++-- src/conf/domain_capabilities.h | 2 +- src/cpu/cpu_x86.c | 10 +++++----- src/security/security_manager.c | 3 ++- tests/networkxml2firewalltest.c | 2 +- tests/testutils.c | 4 ++-- tools/nss/libvirt_nss.c | 3 ++- 7 files changed, 17 insertions(+), 13 deletions(-) -- 2.31.1

5 13

[libvirt PATCH] qemu_firmware: select correct firmware for AMD SEV-ES
by Pavel Hrdina 14 Jun '21

14 Jun '21

When using firmware auto-selection and user enables AMD SEV-ES we need to pick correct firmware that actually supports it. This can be detected by having `amd-sev-es` in the firmware JSON description. Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- src/qemu/qemu_firmware.c | 28 ++++++++++++++++++++++++---- 1 file changed, 24 insertions(+), 4 deletions(-) diff --git a/src/qemu/qemu_firmware.c b/src/qemu/qemu_firmware.c index 2aeac635da..8d10497717 100644 --- a/src/qemu/qemu_firmware.c +++ b/src/qemu/qemu_firmware.c @@ -129,6 +129,7 @@ typedef enum { QEMU_FIRMWARE_FEATURE_ACPI_S3, QEMU_FIRMWARE_FEATURE_ACPI_S4, QEMU_FIRMWARE_FEATURE_AMD_SEV, + QEMU_FIRMWARE_FEATURE_AMD_SEV_ES, QEMU_FIRMWARE_FEATURE_ENROLLED_KEYS, QEMU_FIRMWARE_FEATURE_REQUIRES_SMM, QEMU_FIRMWARE_FEATURE_SECURE_BOOT, @@ -145,6 +146,7 @@ VIR_ENUM_IMPL(qemuFirmwareFeature, "acpi-s3", "acpi-s4", "amd-sev", + "amd-sev-es", "enrolled-keys", "requires-smm", "secure-boot", @@ -913,6 +915,9 @@ qemuFirmwareOSInterfaceTypeFromOsDefFirmware(int fw) } +#define VIR_QEMU_FIRMWARE_AMD_SEV_ES_POLICY (1 << 2) + + static bool qemuFirmwareMatchDomain(const virDomainDef *def, const qemuFirmware *fw, @@ -924,6 +929,7 @@ qemuFirmwareMatchDomain(const virDomainDef *def, bool supportsS4 = false; bool requiresSMM = false; bool supportsSEV = false; + bool supportsSEVES = false; bool supportsSecureBoot = false; bool hasEnrolledKeys = false; int reqSecureBoot; @@ -972,6 +978,11 @@ qemuFirmwareMatchDomain(const virDomainDef *def, case QEMU_FIRMWARE_FEATURE_AMD_SEV: supportsSEV = true; break; + + case QEMU_FIRMWARE_FEATURE_AMD_SEV_ES: + supportsSEVES = true; + break; + case QEMU_FIRMWARE_FEATURE_REQUIRES_SMM: requiresSMM = true; break; @@ -1043,10 +1054,16 @@ qemuFirmwareMatchDomain(const virDomainDef *def, } if (def->sev && - def->sev->sectype == VIR_DOMAIN_LAUNCH_SECURITY_SEV && - !supportsSEV) { - VIR_DEBUG("Domain requires SEV, firmware '%s' doesn't support it", path); - return false; + def->sev->sectype == VIR_DOMAIN_LAUNCH_SECURITY_SEV) { + if (!supportsSEV) { + VIR_DEBUG("Domain requires SEV, firmware '%s' doesn't support it", path); + return false; + } + + if (def->sev->policy & VIR_QEMU_FIRMWARE_AMD_SEV_ES_POLICY && !supportsSEVES) { + VIR_DEBUG("Domain requires SEV-ES, firmware '%s' doesn't support it", path); + return false; + } } VIR_DEBUG("Firmware '%s' matches domain requirements", path); @@ -1148,6 +1165,7 @@ qemuFirmwareEnableFeatures(virQEMUDriver *driver, case QEMU_FIRMWARE_FEATURE_ACPI_S3: case QEMU_FIRMWARE_FEATURE_ACPI_S4: case QEMU_FIRMWARE_FEATURE_AMD_SEV: + case QEMU_FIRMWARE_FEATURE_AMD_SEV_ES: case QEMU_FIRMWARE_FEATURE_ENROLLED_KEYS: case QEMU_FIRMWARE_FEATURE_SECURE_BOOT: case QEMU_FIRMWARE_FEATURE_VERBOSE_DYNAMIC: @@ -1181,6 +1199,7 @@ qemuFirmwareSanityCheck(const qemuFirmware *fw, case QEMU_FIRMWARE_FEATURE_ACPI_S3: case QEMU_FIRMWARE_FEATURE_ACPI_S4: case QEMU_FIRMWARE_FEATURE_AMD_SEV: + case QEMU_FIRMWARE_FEATURE_AMD_SEV_ES: case QEMU_FIRMWARE_FEATURE_ENROLLED_KEYS: case QEMU_FIRMWARE_FEATURE_VERBOSE_DYNAMIC: case QEMU_FIRMWARE_FEATURE_VERBOSE_STATIC: @@ -1412,6 +1431,7 @@ qemuFirmwareGetSupported(const char *machine, case QEMU_FIRMWARE_FEATURE_ACPI_S3: case QEMU_FIRMWARE_FEATURE_ACPI_S4: case QEMU_FIRMWARE_FEATURE_AMD_SEV: + case QEMU_FIRMWARE_FEATURE_AMD_SEV_ES: case QEMU_FIRMWARE_FEATURE_ENROLLED_KEYS: case QEMU_FIRMWARE_FEATURE_SECURE_BOOT: case QEMU_FIRMWARE_FEATURE_VERBOSE_DYNAMIC: -- 2.31.1

2 2

[libvirt PATCH v2 0/7] Enable sanitizers
by Tim Wiederhake 14 Jun '21

14 Jun '21

This series enables and adds AddressSanitizer and UndefinedBehaviorSanitizer builds to the CI. See: https://clang.llvm.org/docs/AddressSanitizer.html and https://clang.llvm.org/docs/UndefinedBehaviorSanitizer.html These sanitizers already found some issues in libvirt, e.g. 4eb7c621985dad4de911ec394ac628bd1a5b29ab, 1294de209cee6643511265c7e2d4283c047cf652, 8b8c91f487592c6c067847ca59dde405ca17573f, or 1c34211c22de28127a509edbf2cf2f44cb0d891e. There exist two more relevant sanitizers, ThreadSanitizer and MemorySanitizer. Unfortunately, those two require an instrumented build of all dependencies, including libc, to work correctly. Note that clang and gcc have different implementations of these sanitizers, hence the introduction of two new jobs to the CI. The latter one issues a warning about the use of LD_PRELOAD in `virTestMain`, which in this particular case can be safely ignored by setting `ASAN_OPTIONS` to verify_asan_link_order=0` for the gcc build. Changes since V1: Incorporated changes suggested by Pavel, except for #6 (now #7): The statement in https://listman.redhat.com/archives/libvir-list/2021-May/msg00070.html on the sanitizers working with Fedora 33 is wrong, I was fooled by caching. The bug described there is present in Fedora 33, 34, and Rawhide. Cheers, Tim Tim Wiederhake (7): meson: Allow larger stack frames when instrumenting meson: Allow undefined symbols when sanitizers are enabled tests: virfilemock: realpath: Allow non-null second parameter openvz: Add missing symbols to libvirt_openvz.syms tests: openvzutilstest: Remove duplicate linking with libvirt_openvz.a virt-aa-helper: Remove duplicate linking with src/datatypes.o ci: Enable address and undefined behavior sanitizers .gitlab-ci.yml | 35 +++++++++++++++++++++++++++++++++++ build-aux/syntax-check.mk | 2 +- meson.build | 14 ++++++++++---- src/libvirt_openvz.syms | 2 ++ src/security/meson.build | 1 - tests/meson.build | 2 +- tests/virfilemock.c | 20 ++++++++++++-------- 7 files changed, 61 insertions(+), 15 deletions(-) -- 2.26.3

5 16

[PATCH 0/7] qemu: Rework qemuBuildDeviceVideoStr() slightly
by Michal Privoznik 11 Jun '21

11 Jun '21

The most important patch is 7/7, IMO. The rest is just prep work. Michal Prívozník (7): qemuDomainSupportsVideoVga: Fix const correctness qemuBuildDeviceVideoStr: Separate out video module selection qemuDeviceVideoGetModel: Deduplicate a check qemu_command: Switch from VIR_ENUM_IMPL(qemuDeviceVideo) to explicit switch() qemu_command: Switch from VIR_ENUM_IMPL(qemuDeviceVideoSecondary) to explicit switch() qemuBuildDeviceVideoStr: Move logic wrapping qemuBuildVirtioDevStr() into qemuDeviceVideoGetModel() qemuBuildDeviceVideoStr: Don't overwrite @model src/qemu/qemu_command.c | 159 +++++++++++++++++++++++++--------------- src/qemu/qemu_domain.c | 2 +- src/qemu/qemu_domain.h | 2 +- 3 files changed, 102 insertions(+), 61 deletions(-) -- 2.31.1

2 11

[PATCH v2 0/6] qemu: Adapt to virtio*gl* devices
by Han Han 11 Jun '21

11 Jun '21

Diff from v1: - Leave qemuBuildVirtioDevStr() unmodified - New patch: qemu_validate: Validate virtio*gl* devices for 3d accerlation - Use DO_TEST_CAPS_LATEST instead of DO_TEST - Save video-virtio-vga-gpu-gl.x86_64-latest.xml as the sym link of tests/qemuxml2argvdata/video-virtio-vga-gpu-gl.xml v1: https://listman.redhat.com/archives/libvir-list/2021-June/msg00181.html Han Han (6): qemu_capabilities: Add QEMU_CAPS_VIRTIO_GPU_GL_PCI qemu_capabilities: Add QEMU_CAPS_VIRTIO_VGA_GL qemu_validate: Validate virtio*gl* devices for 3d accerlation qemu: Adapt to virtio-gpu-gl-pci device qemu: Adapt to virtio-vga-gl device tests: Tests for virtio-vga-gl and virtio-gpu-gl-pci device src/qemu/qemu_capabilities.c | 4 ++ src/qemu/qemu_capabilities.h | 2 + src/qemu/qemu_command.c | 8 ++++ src/qemu/qemu_validate.c | 4 +- .../caps_6.1.0.x86_64.xml | 2 + ...video-virtio-vga-gpu-gl.x86_64-latest.args | 35 ++++++++++++++ .../video-virtio-vga-gpu-gl.xml | 46 +++++++++++++++++++ tests/qemuxml2argvtest.c | 1 + .../video-virtio-vga-gpu-gl.x86_64-latest.xml | 1 + tests/qemuxml2xmltest.c | 1 + 10 files changed, 103 insertions(+), 1 deletion(-) create mode 100644 tests/qemuxml2argvdata/video-virtio-vga-gpu-gl.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/video-virtio-vga-gpu-gl.xml create mode 120000 tests/qemuxml2xmloutdata/video-virtio-vga-gpu-gl.x86_64-latest.xml -- 2.31.1

2 13

[libvirt PATCH] cpu_map: Add cpu feature avx-vnni
by Tim Wiederhake 11 Jun '21

11 Jun '21

"avx-vvni" was introduced to qemu in commit c1826ea6a052084f2e6a0bae9dd5932a727df039, adding it Cooperlake. This feature is currently not used by any libvirt CPU models, but its addition silences a warning from sync_qemu_i386.py: ``` warning: Unknown feature 'CPUID_7_1_EAX_AVX_VNNI' warning: Feature unknown to libvirt: CPUID_7_1_EAX_AVX_VNNI ``` Signed-off-by: Tim Wiederhake <twiederh(a)redhat.com> --- src/cpu_map/sync_qemu_i386.py | 1 + src/cpu_map/x86_features.xml | 3 +++ 2 files changed, 4 insertions(+) diff --git a/src/cpu_map/sync_qemu_i386.py b/src/cpu_map/sync_qemu_i386.py index 92bb58f75b..4dd9f3b84d 100755 --- a/src/cpu_map/sync_qemu_i386.py +++ b/src/cpu_map/sync_qemu_i386.py @@ -73,6 +73,7 @@ def translate_feature(name): "CPUID_7_0_EDX_SPEC_CTRL_SSBD": "ssbd", "CPUID_7_0_EDX_STIBP": "stibp", "CPUID_7_1_EAX_AVX512_BF16": "avx512-bf16", + "CPUID_7_1_EAX_AVX_VNNI": "avx-vnni", "CPUID_8000_0008_EBX_AMD_SSBD": "amd-ssbd", "CPUID_8000_0008_EBX_CLZERO": "clzero", "CPUID_8000_0008_EBX_IBPB": "ibpb", diff --git a/src/cpu_map/x86_features.xml b/src/cpu_map/x86_features.xml index e98b84f3ef..4cf3ff0804 100644 --- a/src/cpu_map/x86_features.xml +++ b/src/cpu_map/x86_features.xml @@ -363,6 +363,9 @@ <cpuid eax_in='0x07' ecx_in='0x00' edx='0x80000000'/> </feature> + <feature name='avx-vnni'> + <cpuid eax_in='0x07' ecx_in='0x01' eax='0x00000010'/> + </feature> <feature name='avx512-bf16'> <cpuid eax_in='0x07' ecx_in='0x01' eax='0x00000020'/> </feature> -- 2.31.1

2 1

[libvirt PATCH v2 0/7] Enable autostarting mediated devices
by Jonathon Jongsma 10 Jun '21

10 Jun '21

This series replaces the initial patch that was recently reverted. It implements the autostart feature using a new virNodeDeviceGet/SetAutostart() API that is consistent with how other libvirt objects handle autostart. It also adds a counterpart virsh command nodedev-autostart. In order to easily check the 'autostart' status of the device (since it is no longer part of the device xml), a new virsh command is introduced: nodedev-info. This also presents a few more basic bits of information about the device, including 'active' and 'persistent' status, which requires exposing new APIs on the node device: IsActive() and IsPersistent(). These APIs are consistent with existing libvirt objects. Changes in version 2: - Parse the autostart property from mdevctl output. - only re-sending patch 2/7, all others were already ACKED Jonathon Jongsma (7): api: add virNodeDevice(Get|Set)Autostart() nodedev: implement virNodeDevice(Get|Set)Autostart() nodedev: Add tests for mdevctl autostart command virsh: add nodedev-autostart api: add virNodeDeviceIsPersistent()/IsActive() nodedev: Implement virNodeDeviceIsPersistent()/IsActive() virsh: add nodedev-info docs/manpages/virsh.rst | 27 +++ include/libvirt/libvirt-nodedev.h | 10 + src/conf/node_device_conf.h | 1 + src/conf/virnodedeviceobj.c | 16 ++ src/conf/virnodedeviceobj.h | 6 + src/driver-nodedev.h | 18 ++ src/libvirt-nodedev.c | 141 ++++++++++++++ src/libvirt_private.syms | 2 + src/libvirt_public.syms | 6 + src/node_device/node_device_driver.c | 183 +++++++++++++++++- src/node_device/node_device_driver.h | 19 ++ src/node_device/node_device_udev.c | 30 ++- src/remote/remote_driver.c | 6 +- src/remote/remote_protocol.x | 59 +++++- src/remote_protocol-structs | 26 +++ .../nodedevmdevctldata/mdevctl-autostart.argv | 8 + tests/nodedevmdevctltest.c | 54 ++++++ tools/virsh-nodedev.c | 139 +++++++++++++ 18 files changed, 744 insertions(+), 7 deletions(-) create mode 100644 tests/nodedevmdevctldata/mdevctl-autostart.argv -- 2.31.1

1 2

[RFC QEMU PATCH] ui: Make the DisplayType enum entries conditional
by Thomas Huth 10 Jun '21

10 Jun '21

Libvirt's "domcapabilities" command has a way to state whether certain graphic frontends are available in QEMU or not. Originally, libvirt looked at the "--help" output of the QEMU binary to determine whether SDL was available or not (by looking for the "-sdl" parameter in the help text), but since libvirt stopped doing this analysis of the help text, the detection of SDL is currently broken, see: https://bugzilla.redhat.com/show_bug.cgi?id=1790902 QEMU should provide a way via the QMP interface instead. The simplest way, without introducing additional commands, is to make the DisplayType enum entries conditional, so that the enum only contains the entries if the corresponding CONFIG_xxx switches have been set. Unfortunately, this only works for sdl, cocoa and spice, since gtk, egl-headless and curses are hard-wired in the "data" section of the DisplayOptions, and thus unfortunately always have to be defined. Signed-off-by: Thomas Huth <thuth(a)redhat.com> --- qapi/ui.json | 12 +++++++++--- ui/console.c | 4 ++++ 2 files changed, 13 insertions(+), 3 deletions(-) diff --git a/qapi/ui.json b/qapi/ui.json index 1052ca9c38..c4f44cfe50 100644 --- a/qapi/ui.json +++ b/qapi/ui.json @@ -1126,9 +1126,15 @@ # ## { 'enum' : 'DisplayType', - 'data' : [ 'default', 'none', 'gtk', 'sdl', - 'egl-headless', 'curses', 'cocoa', - 'spice-app'] } + 'data' : [ + { 'name': 'default' }, + { 'name': 'none' }, + { 'name': 'gtk' }, + { 'name': 'sdl', 'if': 'defined(CONFIG_SDL)' }, + { 'name': 'egl-headless' }, + { 'name': 'curses' }, + { 'name': 'cocoa', 'if': 'defined(CONFIG_COCOA)' }, + { 'name': 'spice-app', 'if': 'defined(CONFIG_SPICE)'} ] } ## # @DisplayOptions: diff --git a/ui/console.c b/ui/console.c index 2de5f4105b..954f7162c3 100644 --- a/ui/console.c +++ b/ui/console.c @@ -2371,8 +2371,12 @@ bool qemu_display_find_default(DisplayOptions *opts) { static DisplayType prio[] = { DISPLAY_TYPE_GTK, +#if defined(CONFIG_SDL) DISPLAY_TYPE_SDL, +#endif +#if defined(CONFIG_COCOA) DISPLAY_TYPE_COCOA +#endif }; int i; -- 2.27.0

4 11

[PATCH] domain_conf: Use virXMLFormatElement() more
by Michal Privoznik 09 Jun '21

09 Jun '21

I've identified some places (mostly by looking for virBufferUse()) that can use virXMLFormatElement() instead of open coded version of it. I'm sure there are many more places that could use the same treatment. Let's cure them some other time. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- Based on Dan's series he posted earlier today: https://listman.redhat.com/archives/libvir-list/2021-June/msg00060.html src/conf/domain_conf.c | 74 +++++++++++++++--------------------------- 1 file changed, 27 insertions(+), 47 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 7b9de2e92d..139cdfc0a7 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -25177,6 +25177,7 @@ virDomainSmartcardDefFormat(virBuffer *buf, { const char *mode = virDomainSmartcardTypeToString(def->type); g_auto(virBuffer) childBuf = VIR_BUFFER_INIT_CHILD(buf); + g_auto(virBuffer) attrBuf = VIR_BUFFER_INITIALIZER; size_t i; if (!mode) { @@ -25209,19 +25210,13 @@ virDomainSmartcardDefFormat(virBuffer *buf, } virDomainDeviceInfoFormat(&childBuf, &def->info, flags); - virBufferAsprintf(buf, "<smartcard mode='%s'", mode); + virBufferAsprintf(&attrBuf, " mode='%s'", mode); if (def->type == VIR_DOMAIN_SMARTCARD_TYPE_PASSTHROUGH && - virDomainChrAttrsDefFormat(buf, def->data.passthru, false) < 0) { + virDomainChrAttrsDefFormat(&attrBuf, def->data.passthru, false) < 0) { return -1; } - if (virBufferUse(&childBuf)) { - virBufferAddLit(buf, ">\n"); - virBufferAddBuffer(buf, &childBuf); - virBufferAddLit(buf, "</smartcard>\n"); - } else { - virBufferAddLit(buf, "/>\n"); - } + virXMLFormatElement(buf, "smartcard", &attrBuf, &childBuf); return 0; } @@ -25300,6 +25295,7 @@ virDomainSoundDefFormat(virBuffer *buf, { const char *model = virDomainSoundModelTypeToString(def->model); g_auto(virBuffer) childBuf = VIR_BUFFER_INIT_CHILD(buf); + g_auto(virBuffer) attrBuf = VIR_BUFFER_INITIALIZER; size_t i; if (!model) { @@ -25316,14 +25312,9 @@ virDomainSoundDefFormat(virBuffer *buf, virDomainDeviceInfoFormat(&childBuf, &def->info, flags); - virBufferAsprintf(buf, "<sound model='%s'", model); - if (virBufferUse(&childBuf)) { - virBufferAddLit(buf, ">\n"); - virBufferAddBuffer(buf, &childBuf); - virBufferAddLit(buf, "</sound>\n"); - } else { - virBufferAddLit(buf, "/>\n"); - } + virBufferAsprintf(&attrBuf, " model='%s'", model); + + virXMLFormatElement(buf, "sound", &attrBuf, &childBuf); return 0; } @@ -25453,6 +25444,7 @@ virDomainAudioDefFormat(virBuffer *buf, virDomainAudioDef *def) { g_auto(virBuffer) childBuf = VIR_BUFFER_INIT_CHILD(buf); + g_auto(virBuffer) attrBuf = VIR_BUFFER_INITIALIZER; g_auto(virBuffer) inputBuf = VIR_BUFFER_INITIALIZER; g_auto(virBuffer) outputBuf = VIR_BUFFER_INITIALIZER; const char *type = virDomainAudioTypeTypeToString(def->type); @@ -25463,10 +25455,10 @@ virDomainAudioDefFormat(virBuffer *buf, return -1; } - virBufferAsprintf(buf, "<audio id='%d' type='%s'", def->id, type); + virBufferAsprintf(&attrBuf, " id='%d' type='%s'", def->id, type); if (def->timerPeriod) - virBufferAsprintf(buf, " timerPeriod='%u'", def->timerPeriod); + virBufferAsprintf(&attrBuf, " timerPeriod='%u'", def->timerPeriod); switch (def->type) { case VIR_DOMAIN_AUDIO_TYPE_NONE: @@ -25489,20 +25481,20 @@ virDomainAudioDefFormat(virBuffer *buf, case VIR_DOMAIN_AUDIO_TYPE_OSS: if (def->backend.oss.tryMMap) - virBufferAsprintf(buf, " tryMMap='%s'", + virBufferAsprintf(&attrBuf, " tryMMap='%s'", virTristateBoolTypeToString(def->backend.oss.tryMMap)); if (def->backend.oss.exclusive) - virBufferAsprintf(buf, " exclusive='%s'", + virBufferAsprintf(&attrBuf, " exclusive='%s'", virTristateBoolTypeToString(def->backend.oss.exclusive)); if (def->backend.oss.dspPolicySet) - virBufferAsprintf(buf, " dspPolicy='%d'", def->backend.oss.dspPolicy); + virBufferAsprintf(&attrBuf, " dspPolicy='%d'", def->backend.oss.dspPolicy); virDomainAudioOSSFormat(&def->backend.oss.input, &inputBuf); virDomainAudioOSSFormat(&def->backend.oss.output, &outputBuf); break; case VIR_DOMAIN_AUDIO_TYPE_PULSEAUDIO: - virBufferEscapeString(buf, " serverName='%s'", + virBufferEscapeString(&attrBuf, " serverName='%s'", def->backend.pulseaudio.serverName); virDomainAudioPulseAudioFormat(&def->backend.pulseaudio.input, &inputBuf); @@ -25511,7 +25503,7 @@ virDomainAudioDefFormat(virBuffer *buf, case VIR_DOMAIN_AUDIO_TYPE_SDL: if (def->backend.sdl.driver) - virBufferAsprintf(buf, " driver='%s'", + virBufferAsprintf(&attrBuf, " driver='%s'", virDomainAudioSDLDriverTypeToString( def->backend.sdl.driver)); @@ -25523,7 +25515,7 @@ virDomainAudioDefFormat(virBuffer *buf, break; case VIR_DOMAIN_AUDIO_TYPE_FILE: - virBufferEscapeString(buf, " path='%s'", def->backend.file.path); + virBufferEscapeString(&attrBuf, " path='%s'", def->backend.file.path); break; case VIR_DOMAIN_AUDIO_TYPE_LAST: @@ -25535,13 +25527,7 @@ virDomainAudioDefFormat(virBuffer *buf, virDomainAudioCommonFormat(&def->input, &childBuf, &inputBuf, "input"); virDomainAudioCommonFormat(&def->output, &childBuf, &outputBuf, "output"); - if (virBufferUse(&childBuf)) { - virBufferAddLit(buf, ">\n"); - virBufferAddBuffer(buf, &childBuf); - virBufferAddLit(buf, "</audio>\n"); - } else { - virBufferAddLit(buf, "/>\n"); - } + virXMLFormatElement(buf, "audio", &attrBuf, &childBuf); return 0; } @@ -27002,6 +26988,7 @@ virDomainCachetuneDefFormat(virBuffer *buf, unsigned int flags) { g_auto(virBuffer) childrenBuf = VIR_BUFFER_INIT_CHILD(buf); + g_auto(virBuffer) attrBuf = VIR_BUFFER_INITIALIZER; size_t i = 0; g_autofree char *vcpus = NULL; @@ -27024,19 +27011,17 @@ virDomainCachetuneDefFormat(virBuffer *buf, if (!vcpus) return -1; - virBufferAsprintf(buf, "<cachetune vcpus='%s'", vcpus); + virBufferAsprintf(&attrBuf, " vcpus='%s'", vcpus); if (!(flags & VIR_DOMAIN_DEF_FORMAT_INACTIVE)) { const char *alloc_id = virResctrlAllocGetID(resctrl->alloc); if (!alloc_id) return -1; - virBufferAsprintf(buf, " id='%s'", alloc_id); + virBufferAsprintf(&attrBuf, " id='%s'", alloc_id); } - virBufferAddLit(buf, ">\n"); - virBufferAddBuffer(buf, &childrenBuf); - virBufferAddLit(buf, "</cachetune>\n"); + virXMLFormatElement(buf, "cachetune", &attrBuf, &childrenBuf); return 0; } @@ -27062,6 +27047,7 @@ virDomainMemorytuneDefFormat(virBuffer *buf, unsigned int flags) { g_auto(virBuffer) childrenBuf = VIR_BUFFER_INIT_CHILD(buf); + g_auto(virBuffer) attrBuf = VIR_BUFFER_INITIALIZER; g_autofree char *vcpus = NULL; size_t i = 0; @@ -27084,19 +27070,17 @@ virDomainMemorytuneDefFormat(virBuffer *buf, if (!vcpus) return -1; - virBufferAsprintf(buf, "<memorytune vcpus='%s'", vcpus); + virBufferAsprintf(&attrBuf, " vcpus='%s'", vcpus); if (!(flags & VIR_DOMAIN_DEF_FORMAT_INACTIVE)) { const char *alloc_id = virResctrlAllocGetID(resctrl->alloc); if (!alloc_id) return -1; - virBufferAsprintf(buf, " id='%s'", alloc_id); + virBufferAsprintf(&attrBuf, " id='%s'", alloc_id); } - virBufferAddLit(buf, ">\n"); - virBufferAddBuffer(buf, &childrenBuf); - virBufferAddLit(buf, "</memorytune>\n"); + virXMLFormatElement(buf, "memorytune", &attrBuf, &childrenBuf); return 0; } @@ -27213,11 +27197,7 @@ virDomainCputuneDefFormat(virBuffer *buf, for (i = 0; i < def->nresctrls; i++) virDomainMemorytuneDefFormat(&childrenBuf, def->resctrls[i], flags); - if (virBufferUse(&childrenBuf)) { - virBufferAddLit(buf, "<cputune>\n"); - virBufferAddBuffer(buf, &childrenBuf); - virBufferAddLit(buf, "</cputune>\n"); - } + virXMLFormatElement(buf, "cputune", NULL, &childrenBuf); return 0; } -- 2.31.1

2 1

[PATCH] docs: switch ci status page to use Go -module repos
by Daniel P. Berrangé 09 Jun '21

09 Jun '21

Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- docs/ci.rst | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/docs/ci.rst b/docs/ci.rst index b321a67bd9..71418d74bf 100644 --- a/docs/ci.rst +++ b/docs/ci.rst @@ -50,10 +50,10 @@ Language bindings :target: https://gitlab.com/libvirt/libvirt-csharp/pipelines :alt: libvirt-csharp pipeline status - * - libvirt-go - - .. image:: https://gitlab.com/libvirt/libvirt-go/badges/master/pipeline.svg - :target: https://gitlab.com/libvirt/libvirt-go/pipelines - :alt: libvirt-go pipeline status + * - libvirt-go-module + - .. image:: https://gitlab.com/libvirt/libvirt-go-module/badges/master/pipeline.svg + :target: https://gitlab.com/libvirt/libvirt-go-module/pipelines + :alt: libvirt-go-module pipeline status * - libvirt-java - .. image:: https://gitlab.com/libvirt/libvirt-java/badges/master/pipeline.svg @@ -116,10 +116,10 @@ Object mappings :target: https://gitlab.com/libvirt/libvirt-glib/pipelines :alt: libvirt-glib pipeline status - * - libvirt-go-xml - - .. image:: https://gitlab.com/libvirt/libvirt-go-xml/badges/master/pipeline.svg - :target: https://gitlab.com/libvirt/libvirt-go-xml/pipelines - :alt: libvirt-go-xml pipeline status + * - libvirt-go-xml-module + - .. image:: https://gitlab.com/libvirt/libvirt-go-xml-module/badges/master/pipeline.svg + :target: https://gitlab.com/libvirt/libvirt-go-xml-module/pipelines + :alt: libvirt-go-xml-module pipeline status * - libvirt-snmp - .. image:: https://gitlab.com/libvirt/libvirt-snmp/badges/master/pipeline.svg -- 2.31.1

2 1

[libvirt PATCH] cpu_map: sync_qemu_i386.py: Remove superfluous semicolon
by Tim Wiederhake 09 Jun '21

09 Jun '21

The semicolon in question makes the pipeline fail over a style checker complaint. Introduced-in: 360b8eb2d2cb1b6a8c9a78fa2c5be31dd7c74487 Signed-off-by: Tim Wiederhake <twiederh(a)redhat.com> --- src/cpu_map/sync_qemu_i386.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/cpu_map/sync_qemu_i386.py b/src/cpu_map/sync_qemu_i386.py index fbf1edf5d0..92bb58f75b 100755 --- a/src/cpu_map/sync_qemu_i386.py +++ b/src/cpu_map/sync_qemu_i386.py @@ -209,7 +209,7 @@ def read_builtin_x86_defs(filename): raise RuntimeError("begin mark not found") match = begin_mark.match(line) if match: - break; + break match = shorthand.match(line) if match: # TCG definitions are irrelevant for cpu models -- 2.31.1

3 2

[PATCH] virt-host-validate: Fix IOMMU output on aarch64
by Fabiano Fidêncio 09 Jun '21

09 Jun '21

virt-host-validate should print "Checking for device assignment IOMMU support" for all architectures, not only for Intel / AMD. This is the output without the patch: ``` [fidencio@dentola libvirt]$ virt-host-validate QEMU: comprobando if device /dev/kvm exists : PASA QEMU: comprobando if device /dev/kvm is accessible : PASA QEMU: comprobando if device /dev/vhost-net exists : PASA QEMU: comprobando if device /dev/net/tun exists : PASA QEMU: comprobando for cgroup 'cpu' controller support : PASA QEMU: comprobando for cgroup 'cpuacct' controller support : PASA QEMU: comprobando for cgroup 'cpuset' controller support : PASA QEMU: comprobando for cgroup 'memory' controller support : PASA QEMU: comprobando for cgroup 'devices' controller support : ADVERTENCIA (Enable 'devices' in kernel Kconfig file or mount/enable cgroup controller in your system) QEMU: comprobando for cgroup 'blkio' controller support : PASA ADVERTENCIA (Unknown if this platform has IOMMU support) QEMU: comprobando for secure guest support : ADVERTENCIA (Unknown if this platform has Secure Guest support) ``` This is the output with the patch: ``` [fidencio@dentola libvirt]$ ./build/tools/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 : WARN (Enable 'devices' in kernel Kconfig file or mount/enable cgroup controller in your system) QEMU: Checking for cgroup 'blkio' controller support : PASS QEMU: Checking for device assignment IOMMU support : WARN (Unknown if this platform has IOMMU support) QEMU: Checking for secure guest support : WARN (Unknown if this platform has Secure Guest support) ``` Signed-off-by: Fabiano Fidêncio <fabiano(a)fidencio.org> --- This is a follow-up on https://listman.redhat.com/archives/libvir-list/2021-June/msg00190.html --- tools/virt-host-validate-common.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/virt-host-validate-common.c b/tools/virt-host-validate-common.c index 9412bb7514..ee22a88b31 100644 --- a/tools/virt-host-validate-common.c +++ b/tools/virt-host-validate-common.c @@ -335,6 +335,8 @@ int virHostValidateIOMMU(const char *hvname, struct dirent *dent; int rc; + virHostMsgCheck(hvname, "%s", _("for device assignment IOMMU support")); + flags = virHostValidateGetCPUFlags(); if (flags && virBitmapIsBitSet(flags, VIR_HOST_VALIDATE_CPU_FLAG_VMX)) @@ -345,7 +347,6 @@ int virHostValidateIOMMU(const char *hvname, virBitmapFree(flags); if (isIntel) { - virHostMsgCheck(hvname, "%s", _("for device assignment IOMMU support")); if (access("/sys/firmware/acpi/tables/DMAR", F_OK) == 0) { virHostMsgPass(); bootarg = "intel_iommu=on"; @@ -357,7 +358,6 @@ int virHostValidateIOMMU(const char *hvname, return VIR_HOST_VALIDATE_FAILURE(level); } } else if (isAMD) { - virHostMsgCheck(hvname, "%s", _("for device assignment IOMMU support")); if (access("/sys/firmware/acpi/tables/IVRS", F_OK) == 0) { virHostMsgPass(); bootarg = "iommu=pt iommu=1"; -- 2.31.1

2 3

[libvirt PATCH 0/2] Update sync_qemu_i386.py tool for changes in QEMU
by Tim Wiederhake 09 Jun '21

09 Jun '21

QEMU commit e11fd68996fb27c040552320f01a7d30a15a7cc1 changed a line that was used by our tool as marker. Commit 1 prepares for the change, commit 2 actually makes the required adjustment. Cheers, Tim Tim Wiederhake (2): cpu_map: sync_qemu_i386.py: Use regex to look for begin mark cpu_map: sync_qemu_i386.py: Allow begin mark to contain `const` src/cpu_map/sync_qemu_i386.py | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) -- 2.26.3

2 3

[libvirt PATCH 0/4] Support autostart for mediated devices
by Jonathon Jongsma 09 Jun '21

09 Jun '21

This is a short patch series based on an initial patch from Boris Fiuczynski that I massaged a little bit and added a few additional patches. This allows you to define mediated devices in libvirt which are started automatically at boot or parent device plugin. Boris Fiuczynski (1): nodedev: support auto-start property for mdevs Jonathon Jongsma (3): test: move nodedev xml2xml output to a separate dir tests: nodedevxml2xmltest: test more mdev files docs: nodedev: document mdev uuid property docs/formatnode.html.in | 14 ++++++++ docs/schemas/nodedev.rng | 11 +++++++ src/conf/node_device_conf.c | 20 ++++++++++- src/conf/node_device_conf.h | 12 +++++++ src/libvirt_private.syms | 2 ++ src/node_device/node_device_driver.c | 7 +++- .../mdevctl-list-multiple.out.xml | 4 +++ tests/nodedevschemadata/mdev_autostart.xml | 10 ++++++ tests/nodedevschemadata/mdev_manual.xml | 10 ++++++ tests/nodedevxml2xmlout/DVD_GCC_4247N.xml | 15 +++++++++ tests/nodedevxml2xmlout/DVD_with_media.xml | 18 ++++++++++ tests/nodedevxml2xmlout/ap_07_0038.xml | 9 +++++ tests/nodedevxml2xmlout/ap_card07.xml | 8 +++++ tests/nodedevxml2xmlout/ap_matrix.xml | 7 ++++ .../ap_matrix_mdev_types.xml | 14 ++++++++ tests/nodedevxml2xmlout/ccw_0_0_ffff.xml | 10 ++++++ tests/nodedevxml2xmlout/computer.xml | 16 +++++++++ .../css_0_0_fffe_mdev_types.xml | 17 ++++++++++ tests/nodedevxml2xmlout/css_0_0_ffff.xml | 10 ++++++ tests/nodedevxml2xmlout/drm_renderD129.xml | 10 ++++++ ...v_3627463d_b7f0_4fea_b468_f1da537d301b.xml | 9 +++++ tests/nodedevxml2xmlout/mdev_autostart.xml | 11 +++++++ ...v_d069d019_36ea_4111_8f0a_8c9a70e21366.xml | 10 ++++++ ...v_d2441d39_495e_4243_ad9f_beb3f14c23d9.xml | 11 +++++++ ...v_ee0b88c4_f554_4dc1_809d_b2a01e8e48ad.xml | 10 ++++++ ...v_fedc4916_1ca8_49ac_b176_871d16c13076.xml | 10 ++++++ tests/nodedevxml2xmlout/mdev_manual.xml | 11 +++++++ .../net_00_13_02_b9_f9_d3.xml | 21 ++++++++++++ .../net_00_15_58_2f_e9_55.xml | 21 ++++++++++++ .../pci_0000_00_02_0_header_type.xml | 16 +++++++++ .../pci_0000_00_1c_0_header_type.xml | 21 ++++++++++++ .../pci_0000_02_10_7_mdev_types.xml | 33 +++++++++++++++++++ .../pci_0000_02_10_7_sriov.xml | 24 ++++++++++++++ .../pci_0000_02_10_7_sriov_pf_vfs_all.xml | 29 ++++++++++++++++ ...0_02_10_7_sriov_pf_vfs_all_header_type.xml | 31 +++++++++++++++++ .../pci_0000_02_10_7_sriov_vfs.xml | 27 +++++++++++++++ ..._0000_02_10_7_sriov_zero_vfs_max_count.xml | 22 +++++++++++++ tests/nodedevxml2xmlout/pci_1002_71c4.xml | 14 ++++++++ .../pci_8086_0c0c_snd_hda_intel.xml | 17 ++++++++++ .../pci_8086_10c9_sriov_pf.xml | 18 ++++++++++ .../pci_8086_27c5_scsi_host.xml | 7 ++++ .../pci_8086_27c5_scsi_host_0.xml | 7 ++++ .../pci_8086_27c5_scsi_host_0_unique_id.xml | 8 +++++ ...i_8086_27c5_scsi_host_scsi_device_lun0.xml | 11 +++++++ .../pci_8086_27c5_scsi_host_scsi_host.xml | 7 ++++ .../pci_8086_4238_pcie_wireless.xml | 27 +++++++++++++++ tests/nodedevxml2xmlout/scsi_target0_0_0.xml | 7 ++++ tests/nodedevxml2xmlout/scsi_target1_0_0.xml | 12 +++++++ ...rial_3600c0ff000d7a2a5d463ff4902000000.xml | 19 +++++++++++ ...al_SATA_HTS721010G9SA00_MPCZ12Y0GNGWSE.xml | 14 ++++++++ .../usb_device_1d6b_1_0000_00_1d_0.xml | 10 ++++++ .../usb_device_1d6b_1_0000_00_1d_0_if0.xml | 10 ++++++ tests/nodedevxml2xmltest.c | 17 +++++++--- 53 files changed, 739 insertions(+), 7 deletions(-) create mode 100644 tests/nodedevschemadata/mdev_autostart.xml create mode 100644 tests/nodedevschemadata/mdev_manual.xml create mode 100644 tests/nodedevxml2xmlout/DVD_GCC_4247N.xml create mode 100644 tests/nodedevxml2xmlout/DVD_with_media.xml create mode 100644 tests/nodedevxml2xmlout/ap_07_0038.xml create mode 100644 tests/nodedevxml2xmlout/ap_card07.xml create mode 100644 tests/nodedevxml2xmlout/ap_matrix.xml create mode 100644 tests/nodedevxml2xmlout/ap_matrix_mdev_types.xml create mode 100644 tests/nodedevxml2xmlout/ccw_0_0_ffff.xml create mode 100644 tests/nodedevxml2xmlout/computer.xml create mode 100644 tests/nodedevxml2xmlout/css_0_0_fffe_mdev_types.xml create mode 100644 tests/nodedevxml2xmlout/css_0_0_ffff.xml create mode 100644 tests/nodedevxml2xmlout/drm_renderD129.xml create mode 100644 tests/nodedevxml2xmlout/mdev_3627463d_b7f0_4fea_b468_f1da537d301b.xml create mode 100644 tests/nodedevxml2xmlout/mdev_autostart.xml create mode 100644 tests/nodedevxml2xmlout/mdev_d069d019_36ea_4111_8f0a_8c9a70e21366.xml create mode 100644 tests/nodedevxml2xmlout/mdev_d2441d39_495e_4243_ad9f_beb3f14c23d9.xml create mode 100644 tests/nodedevxml2xmlout/mdev_ee0b88c4_f554_4dc1_809d_b2a01e8e48ad.xml create mode 100644 tests/nodedevxml2xmlout/mdev_fedc4916_1ca8_49ac_b176_871d16c13076.xml create mode 100644 tests/nodedevxml2xmlout/mdev_manual.xml create mode 100644 tests/nodedevxml2xmlout/net_00_13_02_b9_f9_d3.xml create mode 100644 tests/nodedevxml2xmlout/net_00_15_58_2f_e9_55.xml create mode 100644 tests/nodedevxml2xmlout/pci_0000_00_02_0_header_type.xml create mode 100644 tests/nodedevxml2xmlout/pci_0000_00_1c_0_header_type.xml create mode 100644 tests/nodedevxml2xmlout/pci_0000_02_10_7_mdev_types.xml create mode 100644 tests/nodedevxml2xmlout/pci_0000_02_10_7_sriov.xml create mode 100644 tests/nodedevxml2xmlout/pci_0000_02_10_7_sriov_pf_vfs_all.xml create mode 100644 tests/nodedevxml2xmlout/pci_0000_02_10_7_sriov_pf_vfs_all_header_type.xml create mode 100644 tests/nodedevxml2xmlout/pci_0000_02_10_7_sriov_vfs.xml create mode 100644 tests/nodedevxml2xmlout/pci_0000_02_10_7_sriov_zero_vfs_max_count.xml create mode 100644 tests/nodedevxml2xmlout/pci_1002_71c4.xml create mode 100644 tests/nodedevxml2xmlout/pci_8086_0c0c_snd_hda_intel.xml create mode 100644 tests/nodedevxml2xmlout/pci_8086_10c9_sriov_pf.xml create mode 100644 tests/nodedevxml2xmlout/pci_8086_27c5_scsi_host.xml create mode 100644 tests/nodedevxml2xmlout/pci_8086_27c5_scsi_host_0.xml create mode 100644 tests/nodedevxml2xmlout/pci_8086_27c5_scsi_host_0_unique_id.xml create mode 100644 tests/nodedevxml2xmlout/pci_8086_27c5_scsi_host_scsi_device_lun0.xml create mode 100644 tests/nodedevxml2xmlout/pci_8086_27c5_scsi_host_scsi_host.xml create mode 100644 tests/nodedevxml2xmlout/pci_8086_4238_pcie_wireless.xml create mode 100644 tests/nodedevxml2xmlout/scsi_target0_0_0.xml create mode 100644 tests/nodedevxml2xmlout/scsi_target1_0_0.xml create mode 100644 tests/nodedevxml2xmlout/storage_serial_3600c0ff000d7a2a5d463ff4902000000.xml create mode 100644 tests/nodedevxml2xmlout/storage_serial_SATA_HTS721010G9SA00_MPCZ12Y0GNGWSE.xml create mode 100644 tests/nodedevxml2xmlout/usb_device_1d6b_1_0000_00_1d_0.xml create mode 100644 tests/nodedevxml2xmlout/usb_device_1d6b_1_0000_00_1d_0_if0.xml -- 2.31.1

6 16

[PATCH 0/5] qemu: Adapt to virtio-gpu-gl-pci and virtio-vga-gl
by Han Han 09 Jun '21

09 Jun '21

The virgl property of virtio-gpu-pci&virtio-vga will be replaced by virtio-gpu-gl-pci&virtio-vga-gl in QEMU 6.1. Adapt to that update. Resolves: https://gitlab.com/libvirt/libvirt/-/issues/167 https://bugzilla.redhat.com/show_bug.cgi?id=1967356 Han Han (5): qemu_capabilities: Add QEMU_CAPS_VIRTIO_GPU_GL_PCI qemu: Adapt to virtio-gpu-gl-pci device when available qemu_capabilities: Add QEMU_CAPS_VIRTIO_VGA_GL qemu: Adapt to virtio-vga-gl device tests: Tests for virtio-vga-gl and virtio-gpu-gl-pci device src/qemu/qemu_capabilities.c | 4 ++ src/qemu/qemu_capabilities.h | 2 + src/qemu/qemu_command.c | 11 ++++- .../caps_6.1.0.x86_64.xml | 2 + .../video-virtio-vga-gpu-gl.args | 29 +++++++++++++ .../video-virtio-vga-gpu-gl.xml | 34 +++++++++++++++ tests/qemuxml2argvtest.c | 6 +++ .../video-virtio-vga-gpu-gl.xml | 43 +++++++++++++++++++ tests/qemuxml2xmltest.c | 6 +++ 9 files changed, 136 insertions(+), 1 deletion(-) create mode 100644 tests/qemuxml2argvdata/video-virtio-vga-gpu-gl.args create mode 100644 tests/qemuxml2argvdata/video-virtio-vga-gpu-gl.xml create mode 100644 tests/qemuxml2xmloutdata/video-virtio-vga-gpu-gl.xml -- 2.31.1

3 12

[PATCH 0/5] virth-host-validate: Couple of cleanups
by Michal Privoznik 08 Jun '21

08 Jun '21

I've noticed couple of bugs/problems while reviewing Fabiano's patch. Here are fixes. Michal Prívozník (5): virt-host-validate: Initialize the error object virt-host-validate: Report an error if failed to detect CGroups virt-host-validate: Turn failure to read /proc/cmdline into an error virt-host-validate: Call VIR_HOST_VALIDATE_FAILURE() more frequently virHostValidateSecureGuests: Drop useless 'return 0' at the end tools/virt-host-validate-common.c | 28 +++++++++++++++------------- tools/virt-host-validate.c | 6 +++++- 2 files changed, 20 insertions(+), 14 deletions(-) -- 2.31.1

3 9

[PATCH] apparmor: Add denied capabilities
by Jim Fehlig 08 Jun '21

08 Jun '21

The audit log contains the following denials from libvirtd apparmor="DENIED" operation="capable" profile="libvirtd" pid=6012 comm="daemon-init" capability=17 capname="sys_rawio" apparmor="DENIED" operation="capable" profile="libvirtd" pid=6012 comm="rpc-worker" capability=39 capname="bpf" apparmor="DENIED" operation="capable" profile="libvirtd" pid=6012 comm="rpc-worker" capability=38 capname="perfmon" Squelch the denials and allow the capabilities in the libvirtd apparmor profile. Signed-off-by: Jim Fehlig <jfehlig(a)suse.com> --- I'm not really sure when these denials first started appearing, nor have I noticed any problems they are causing. Likely I have not exercised the affected functionality. src/security/apparmor/usr.sbin.libvirtd.in | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/security/apparmor/usr.sbin.libvirtd.in b/src/security/apparmor/usr.sbin.libvirtd.in index bf4563e1e8..928782b709 100644 --- a/src/security/apparmor/usr.sbin.libvirtd.in +++ b/src/security/apparmor/usr.sbin.libvirtd.in @@ -25,6 +25,9 @@ profile libvirtd @sbindir@/libvirtd flags=(attach_disconnected) { capability fsetid, capability audit_write, capability ipc_lock, + capability sys_rawio, + capability bpf, + capability perfmon, # Needed for vfio capability sys_resource, -- 2.31.1

3 4

[PATCH] storage: Don't overwrite error in virISCSIDirectDisconnect()
by Michal Privoznik 08 Jun '21

08 Jun '21

The iscsi-direct storage pool backend works merely like this: a connection is established to the target (usually done via virStorageBackendISCSIDirectSetConnection()), intended action is executed (e.g. reporting LUNs, volume wiping), and at the end the connection is closed via virISCSIDirectDisconnect(). The problem is that virISCSIDirectDisconnect() reports its own errors which may overwrite error that occurred during LUN reporting, or volume wiping or whatever. To fix this, use virErrorPreserveLast() + virErrorRestore() combo, which either preserves previously reported error message, or is NOP if there's no error reported. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1797879 Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/storage/storage_backend_iscsi_direct.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/src/storage/storage_backend_iscsi_direct.c b/src/storage/storage_backend_iscsi_direct.c index 0bff1882b9..e4a14c3fd6 100644 --- a/src/storage/storage_backend_iscsi_direct.c +++ b/src/storage/storage_backend_iscsi_direct.c @@ -391,19 +391,28 @@ virISCSIDirectReportLuns(virStoragePoolObj *pool, static int virISCSIDirectDisconnect(struct iscsi_context *iscsi) { + virErrorPtr orig_err; + int ret = -1; + + virErrorPreserveLast(&orig_err); + if (iscsi_logout_sync(iscsi) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, _("Failed to logout: %s"), iscsi_get_error(iscsi)); - return -1; + goto cleanup; } if (iscsi_disconnect(iscsi) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, _("Failed to disconnect: %s"), iscsi_get_error(iscsi)); - return -1; + goto cleanup; } - return 0; + + ret = 0; + cleanup: + virErrorRestore(&orig_err); + return ret; } static int -- 2.31.1

2 1

[PATCH 0/2] Another round of CH driver fixes
by Michal Privoznik 08 Jun '21

08 Jun '21

*** BLURB HERE *** Michal Prívozník (2): ch_driver: Avoid driver double free virCHDriverConfig: Drop @uri member src/ch/ch_conf.c | 1 - src/ch/ch_conf.h | 1 - src/ch/ch_driver.c | 2 +- 3 files changed, 1 insertion(+), 3 deletions(-) -- 2.31.1

2 3

[PATCH] meson.build: Compile with -Walloca
by Thomas Huth 08 Jun '21

08 Jun '21

We are already compiling libvirt with -Wvla - so it does not make too much sense to still allow people to use alloca() instead. Thus put it on the list of things we want to warn about. Fortunately, there is currently no warning with this flag, so the current sources should be clean. Signed-off-by: Thomas Huth <thuth(a)redhat.com> --- meson.build | 1 + 1 file changed, 1 insertion(+) diff --git a/meson.build b/meson.build index 40c841e777..a97efdf8f4 100644 --- a/meson.build +++ b/meson.build @@ -234,6 +234,7 @@ cc_flags += [ '-Waddress-of-packed-member', '-Waggressive-loop-optimizations', '-Walloc-size-larger-than=@0@'.format(alloc_max.stdout().strip()), + '-Walloca', '-Warray-bounds=2', '-Wattribute-alias=2', '-Wattribute-warning', -- 2.27.0

3 2

[PATCH] meson.build: Remove the -Wvla-larger-then flag
by Thomas Huth 08 Jun '21

08 Jun '21

The flag has a typo in it, it's "...-than=..." and not "...-then=...", so this was in fact never used. Since we're also using -Wvla (without size), we should already get warnings about any variable length arrays anyway, so the additional "-Wvla-larger-than" does not make much sense and thus we can simply drop this. Signed-off-by: Thomas Huth <thuth(a)redhat.com> --- meson.build | 1 - 1 file changed, 1 deletion(-) diff --git a/meson.build b/meson.build index 4f23f9104e..40c841e777 100644 --- a/meson.build +++ b/meson.build @@ -390,7 +390,6 @@ cc_flags += [ '-Wvariadic-macros', '-Wvector-operation-performance', '-Wvla', - '-Wvla-larger-then=4031', '-Wvolatile-register-var', '-Wwrite-strings', ] -- 2.27.0

4 4

[PATCH 0/3] libxl: A few firmware improvments
by Jim Fehlig 08 Jun '21

08 Jun '21

Upstream ovmf will be removing support for Xen in the traditional ovmf package, moving it to a separate OvmfXen package https://bugzilla.tianocore.org/show_bug.cgi?id=2122 This motivated me to verify the impact on libvirt+ovmf+xen. Fortunately it is minimal since the new OvmfXen package is basically a reduced version of the traditional package. I.e. it should contain everything the traditional package contains and thus is binary compatible and should not affect existing guests using the traditional package. Indeed my initial testing has shown this to be true. While testing I cooked up a few improvements to the libxl driver firmware handling, including adding support for automatic firmware selection. Thanks for review/comments! Jim Fehlig (3): libxl: Introduce domain def validate callback libxl: Forbid domain definition with secure boot enabled libxl: Support firmware autoselection src/libxl/libxl_conf.c | 17 ++++++++++++-- src/libxl/libxl_domain.c | 49 ++++++++++++++++++++++++++++++++-------- 2 files changed, 54 insertions(+), 12 deletions(-) -- 2.31.1

2 4