- Devel - libvirt List Archives

[PATCH] conf: Turn virDomainDef.kvm_features into a struct
by Michal Privoznik 14 Dec '21

14 Dec '21

In future commits we will need to store not just an array of VIR_TRISTATE_SWITCH_* but also an additional integer. Follow the example of TCG and introduce a structure where both the array an integer can live. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/conf/domain_conf.c | 20 +++++++++++++------- src/conf/domain_conf.h | 7 ++++++- src/qemu/qemu_command.c | 8 ++++---- 3 files changed, 23 insertions(+), 12 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index b6249aa76f..07cbfa24bc 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -3713,6 +3713,7 @@ void virDomainDefFree(virDomainDef *def) g_free(def->emulator); g_free(def->description); g_free(def->title); + g_free(def->kvm_features); g_free(def->hyperv_vendor_id); g_free(def->tcg_features); @@ -17532,7 +17533,9 @@ static int virDomainFeaturesKVMDefParse(virDomainDef *def, xmlNodePtr node) { - def->features[VIR_DOMAIN_FEATURE_KVM] = VIR_TRISTATE_SWITCH_ON; + g_autofree virDomainFeatureKVM *kvm = NULL; + + kvm = g_new0(virDomainFeatureKVM, 1); node = xmlFirstElementChild(node); while (node) { @@ -17551,11 +17554,14 @@ virDomainFeaturesKVMDefParse(virDomainDef *def, &value) < 0) return -1; - def->kvm_features[feature] = value; + kvm->features[feature] = value; node = xmlNextElementSibling(node); } + def->features[VIR_DOMAIN_FEATURE_KVM] = VIR_TRISTATE_SWITCH_ON; + def->kvm_features = g_steal_pointer(&kvm); + return 0; } @@ -21803,13 +21809,13 @@ virDomainDefFeaturesCheckABIStability(virDomainDef *src, case VIR_DOMAIN_KVM_DEDICATED: case VIR_DOMAIN_KVM_POLLCONTROL: case VIR_DOMAIN_KVM_PVIPI: - if (src->kvm_features[i] != dst->kvm_features[i]) { + if (src->kvm_features->features[i] != dst->kvm_features->features[i]) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, _("State of KVM feature '%s' differs: " "source: '%s', destination: '%s'"), virDomainKVMTypeToString(i), - virTristateSwitchTypeToString(src->kvm_features[i]), - virTristateSwitchTypeToString(dst->kvm_features[i])); + virTristateSwitchTypeToString(src->kvm_features->features[i]), + virTristateSwitchTypeToString(dst->kvm_features->features[i])); return false; } @@ -27873,11 +27879,11 @@ virDomainDefFormatFeatures(virBuffer *buf, case VIR_DOMAIN_KVM_DEDICATED: case VIR_DOMAIN_KVM_POLLCONTROL: case VIR_DOMAIN_KVM_PVIPI: - if (def->kvm_features[j]) + if (def->kvm_features->features[j]) virBufferAsprintf(&childBuf, "<%s state='%s'/>\n", virDomainKVMTypeToString(j), virTristateSwitchTypeToString( - def->kvm_features[j])); + def->kvm_features->features[j])); break; case VIR_DOMAIN_KVM_LAST: diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index b410922f68..5613d621bb 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -2263,6 +2263,11 @@ typedef enum { VIR_ENUM_DECL(virDomainIBS); +typedef struct _virDomainFeatureKVM virDomainFeatureKVM; +struct _virDomainFeatureKVM { + int features[VIR_DOMAIN_KVM_LAST]; +}; + typedef struct _virDomainFeatureTCG virDomainFeatureTCG; struct _virDomainFeatureTCG { unsigned long long tb_cache; /* Stored in KiB */ @@ -2819,7 +2824,7 @@ struct _virDomainDef { int features[VIR_DOMAIN_FEATURE_LAST]; int caps_features[VIR_DOMAIN_PROCES_CAPS_FEATURE_LAST]; int hyperv_features[VIR_DOMAIN_HYPERV_LAST]; - int kvm_features[VIR_DOMAIN_KVM_LAST]; + virDomainFeatureKVM *kvm_features; int msrs_features[VIR_DOMAIN_MSRS_LAST]; int xen_features[VIR_DOMAIN_XEN_LAST]; virDomainXenPassthroughMode xen_passthrough_mode; diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index ce22372749..7cdec9f910 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -6765,22 +6765,22 @@ qemuBuildCpuCommandLine(virCommand *cmd, for (i = 0; i < VIR_DOMAIN_KVM_LAST; i++) { switch ((virDomainKVM) i) { case VIR_DOMAIN_KVM_HIDDEN: - if (def->kvm_features[i] == VIR_TRISTATE_SWITCH_ON) + if (def->kvm_features->features[i] == VIR_TRISTATE_SWITCH_ON) virBufferAddLit(&buf, ",kvm=off"); break; case VIR_DOMAIN_KVM_DEDICATED: - if (def->kvm_features[i] == VIR_TRISTATE_SWITCH_ON) + if (def->kvm_features->features[i] == VIR_TRISTATE_SWITCH_ON) virBufferAddLit(&buf, ",kvm-hint-dedicated=on"); break; case VIR_DOMAIN_KVM_POLLCONTROL: - if (def->kvm_features[i] == VIR_TRISTATE_SWITCH_ON) + if (def->kvm_features->features[i] == VIR_TRISTATE_SWITCH_ON) virBufferAddLit(&buf, ",kvm-poll-control=on"); break; case VIR_DOMAIN_KVM_PVIPI: - if (def->kvm_features[i] == VIR_TRISTATE_SWITCH_OFF) + if (def->kvm_features->features[i] == VIR_TRISTATE_SWITCH_OFF) virBufferAddLit(&buf, ",kvm-pv-ipi=off"); break; -- 2.32.0

2 1

[libvirt PATCH] examples: hellolibvirt: fix argc check
by Ján Tomko 14 Dec '21

14 Dec '21

https://gitlab.com/libvirt/libvirt/-/issues/255 Reported-by: Jeremy Alcim Signed-off-by: Ján Tomko <jtomko(a)redhat.com> --- Pushed as trivial. examples/c/misc/hellolibvirt.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/examples/c/misc/hellolibvirt.c b/examples/c/misc/hellolibvirt.c index a598e01be2..39cefe934c 100644 --- a/examples/c/misc/hellolibvirt.c +++ b/examples/c/misc/hellolibvirt.c @@ -107,11 +107,12 @@ main(int argc, char *argv[]) { int ret = 0; virConnectPtr conn; - char *uri; + char *uri = NULL; printf("Attempting to connect to hypervisor\n"); - uri = (argc > 0 ? argv[1] : NULL); + if (argc > 1) + uri = argv[1]; /* virConnectOpenAuth is called here with all default parameters, * except, possibly, the URI of the hypervisor. */ -- 2.31.1

1 0

[PATCH v3 0/4] Introduce TCG domain features
by Michal Privoznik 13 Dec '21

13 Dec '21

v3 of: https://listman.redhat.com/archives/libvir-list/2021-November/msg00194.html Note, patches 1/4 and 2/4 from v2 were pushed already. diff to v2: - Split feature validation into a separate patch - Validate that the size is a multiple of MiB, because that's QEMU's granularity - Remembered to write NEWS entry - Worked in Peter's comments from v2 (mostly to make code more future proof) Michal Prívozník (4): conf: Introduce TCG domain features qemu: Validate TCG feature is enabled only for TCG domains qemu: Generate command line for tb-cache feature NEWS: Document TCG features NEWS.rst | 5 ++ docs/formatdomain.rst | 11 ++++ docs/schemas/domaincommon.rng | 15 ++++- src/conf/domain_conf.c | 58 ++++++++++++++++++- src/conf/domain_conf.h | 7 +++ src/qemu/qemu_command.c | 5 ++ src/qemu/qemu_validate.c | 17 ++++++ ...efault-cpu-tcg-features.x86_64-latest.args | 36 ++++++++++++ .../x86_64-default-cpu-tcg-features.xml | 56 ++++++++++++++++++ tests/qemuxml2argvtest.c | 1 + ...default-cpu-tcg-features.x86_64-latest.xml | 1 + tests/qemuxml2xmltest.c | 1 + 12 files changed, 210 insertions(+), 3 deletions(-) create mode 100644 tests/qemuxml2argvdata/x86_64-default-cpu-tcg-features.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/x86_64-default-cpu-tcg-features.xml create mode 120000 tests/qemuxml2xmloutdata/x86_64-default-cpu-tcg-features.x86_64-latest.xml -- 2.32.0

2 7

[libvirt PATCH 00/14] use g_auto for virCommand (Episode I: The End)
by Ján Tomko 13 Dec '21

13 Dec '21

A partial cleanup. The rest is in progress (TM) Ján Tomko (14): bhyve: refactor virBhyveProbeGrubCaps bhyve: refactor bhyveProbeCapsDeviceHelper bhyve: refactor bhyveProbeCapsFromHelp bhyve: refactor bhyveConnectDomainXMLToNative bhyve: use g_auto in virBhyveProcessStartImpl bhyve: use g_auto in virBhyveProcessStop bhyve: refactor virBhyveProcessBuildBhyveCmd lxc: use g_auto in lxcContainerChild lxc: refactor virLXCProcessBuildControllerCmd lxc: use g_auto for virCommand in virLXCProcessEnsureRootFS storage: logical: use two cmd vars in GetPoolSources storage: util: steal cmd in CreateQemuImgCmdFromVol security: apparmor: use automatic cleanup in load_profile qemu: use automatic cleanup for virCommand src/bhyve/bhyve_capabilities.c | 57 ++++++++++----------------- src/bhyve/bhyve_command.c | 36 ++++++++--------- src/bhyve/bhyve_driver.c | 28 ++++++------- src/bhyve/bhyve_process.c | 25 ++++-------- src/lxc/lxc_container.c | 3 +- src/lxc/lxc_process.c | 14 ++----- src/qemu/qemu_driver.c | 3 +- src/qemu/qemu_interface.c | 3 +- src/qemu/qemu_tpm.c | 4 +- src/security/security_apparmor.c | 15 ++----- src/storage/storage_backend_logical.c | 18 ++++----- src/storage/storage_util.c | 5 +-- 12 files changed, 78 insertions(+), 133 deletions(-) -- 2.31.1

2 15

[PATCH 0/3] split daemon: Support sparse streams
by Michal Privoznik 13 Dec '21

13 Dec '21

See 3/3 for explanation. Michal Prívozník (3): virStreamInData: Allow callback to not rewind the stream rpc: Introduce virNetClientStreamInData() remote_driver: Implement virStreamInData() callback src/libvirt-stream.c | 9 ++++- src/libvirt_remote.syms | 1 + src/remote/remote_daemon_stream.c | 8 ++-- src/remote/remote_driver.c | 26 +++++++++++++ src/rpc/virnetclientstream.c | 61 +++++++++++++++++++++++++++++++ src/rpc/virnetclientstream.h | 4 ++ 6 files changed, 105 insertions(+), 4 deletions(-) -- 2.32.0

3 11

[libvirt PATCH 0/8] Use g_auto for virCaps and LXCDriverConfig (glib chronicles)
by Ján Tomko 13 Dec '21

13 Dec '21

Pipeline in progress: https://gitlab.com/janotomko/libvirt/-/pipelines/427100328 Ján Tomko (8): lxc: fix error value of lxcNodeGetSecurityModel Use g_auto for freeing virCaps Use g_auto for stealing virCaps lxc: define cleanup function for virLXCDriverConfig lxc: use g_auto for virLXCDriverGetConfig lxc: remove cleanup from lxcNodeGetSecurityModel bhyve: bhyveConnectCompareCPU: remove cleanup label Remove empty cleanup sections src/bhyve/bhyve_driver.c | 46 ++++++---------- src/ch/ch_conf.c | 14 ++--- src/ch/ch_driver.c | 3 +- src/hyperv/hyperv_driver.c | 10 ++-- src/libxl/libxl_capabilities.c | 14 ++--- src/libxl/libxl_conf.c | 11 ++-- src/libxl/xen_common.c | 8 +-- src/libxl/xen_xl.c | 32 ++++------- src/lxc/lxc_conf.c | 18 +++---- src/lxc/lxc_conf.h | 2 + src/lxc/lxc_container.c | 12 ++--- src/lxc/lxc_driver.c | 97 ++++++++++++---------------------- src/lxc/lxc_native.c | 41 +++++--------- src/lxc/lxc_process.c | 44 +++++---------- src/test/test_driver.c | 14 ++--- src/util/vircgroup.c | 12 ++--- src/util/virfirmware.c | 18 +++---- src/vmware/vmware_conf.c | 5 +- src/vz/vz_driver.c | 20 +++---- tests/qemucaps2xmltest.c | 14 ++--- tests/testutilsxen.c | 5 +- tests/vboxsnapshotxmltest.c | 8 +-- tests/virconftest.c | 24 ++++----- tests/virfiletest.c | 14 ++--- tests/virstringtest.c | 16 +++--- 25 files changed, 174 insertions(+), 328 deletions(-) -- 2.31.1

2 10

[libvirt PATCH] qemu: qemuBuildChardevCommand: use chrSourcePriv unconditionally
by Ján Tomko 13 Dec '21

13 Dec '21

In the QEMU driver, we allocate private source data unconditionally for every chardev and the rest of the function just assumes it's there. Signed-off-by: Ján Tomko <jtomko(a)redhat.com> --- Pointed out by Coverity. src/qemu/qemu_command.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 613f7a5d2a..0017cf6ca7 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -1493,7 +1493,7 @@ qemuBuildChardevCommand(virCommand *cmd, * secinfo is added only to a TCP serial device during * qemuDomainSecretChardevPrepare. Subsequently called * functions can just check the config fields */ - if (chrSourcePriv && chrSourcePriv->secinfo) { + if (chrSourcePriv->secinfo) { if (qemuBuildObjectSecretCommandLine(cmd, chrSourcePriv->secinfo, qemuCaps) < 0) -- 2.31.1

2 1

[libvirt PATCH v2 00/12] Improve AMD SEV support
by Daniel P. Berrangé 13 Dec '21

13 Dec '21

This addresses a few issues in the AMD SEV support - Neither host or domain level SEV metadata is exposed in virsh commands - The domain launch security parameters don't expose enough info to validate the measurement - Support verified direct kernel boot - Report max SEV/SEV-ES guest counts The second point was the initial purpose of my work. Per the SEV API guide to calculate the measurement we need measurement = HMAC(0x04 || API_MAJOR || API_MINOR || BUILD || GCTX.POLICY || GCTX.LD || MNONCE; GCTX.TIK) The API_MINOR, API_MAJOR, BUILD values are things that are available from 'query-sev' QMP command and libvirt does not expose this info. This patch series adds them to virDomainGetLaunchSecurityParams alongside the measurement that we already report. So now the client can fetch this info and calculate an expected measurement to compare with the actual measurement they got. They will thus know if the guest is safe to inject secrets into, which is where Jim's recent patches come into play. Daniel P. Berrangé (12): include: add new launch security parameters qemu: report error querying launch params for inactive guest qemu: add monitor APIs for query-sev qemu: report new launch security parameters tools: add 'domlaunchsecinfo' virsh command tools: add 'nodesevinfo' virsh command conf: extend domain capabilities for max SEV guest count include: define parameters for reporting SEV guest limits qemu: report max number of SEV guests conf: add support for setting SEV kernel hashes qemu: probe for sev-guest.kernel-hashes property qemu: format sev-guest.kernel-hashes property docs/formatdomain.rst | 7 +- docs/formatdomaincaps.html.in | 6 + docs/manpages/virsh.rst | 31 +++++ docs/schemas/domaincaps.rng | 6 + docs/schemas/domaincommon.rng | 5 + include/libvirt/libvirt-domain.h | 32 +++++ include/libvirt/libvirt-host.h | 16 +++ src/conf/domain_capabilities.c | 4 + src/conf/domain_capabilities.h | 2 + src/conf/domain_conf.c | 8 ++ src/conf/domain_conf.h | 1 + src/qemu/qemu_capabilities.c | 46 ++++++++ src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 7 ++ src/qemu/qemu_driver.c | 59 ++++++++-- src/qemu/qemu_monitor.c | 13 +++ src/qemu/qemu_monitor.h | 9 ++ src/qemu/qemu_monitor_json.c | 46 ++++++++ src/qemu/qemu_monitor_json.h | 9 ++ src/qemu/qemu_validate.c | 8 ++ .../domaincapsdata/qemu_2.12.0-q35.x86_64.xml | 2 + .../domaincapsdata/qemu_2.12.0-tcg.x86_64.xml | 2 + tests/domaincapsdata/qemu_2.12.0.x86_64.xml | 2 + .../domaincapsdata/qemu_6.0.0-q35.x86_64.xml | 2 + .../domaincapsdata/qemu_6.0.0-tcg.x86_64.xml | 2 + tests/domaincapsdata/qemu_6.0.0.x86_64.xml | 2 + .../domaincapsdata/qemu_6.2.0-q35.x86_64.xml | 7 +- .../domaincapsdata/qemu_6.2.0-tcg.x86_64.xml | 7 +- tests/domaincapsdata/qemu_6.2.0.x86_64.xml | 7 +- .../caps_2.12.0.x86_64.replies | 97 ++++++++++++---- .../caps_3.0.0.x86_64.replies | 97 ++++++++++++---- .../caps_3.1.0.x86_64.replies | 97 ++++++++++++---- .../caps_4.0.0.x86_64.replies | 97 ++++++++++++---- .../caps_4.1.0.x86_64.replies | 89 ++++++++++---- .../caps_4.2.0.x86_64.replies | 89 ++++++++++---- .../caps_5.0.0.x86_64.replies | 89 ++++++++++---- .../caps_5.1.0.x86_64.replies | 89 ++++++++++---- .../caps_5.2.0.x86_64.replies | 89 ++++++++++---- .../caps_6.0.0.x86_64.replies | 89 ++++++++++---- .../caps_6.1.0.x86_64.replies | 89 ++++++++++---- .../caps_6.2.0.x86_64.replies | 109 ++++++++++++++---- .../caps_6.2.0.x86_64.xml | 8 ++ tests/qemumonitorjsontest.c | 43 +++++++ ...unch-security-sev-direct.x86_64-6.2.0.args | 40 +++++++ .../launch-security-sev-direct.xml | 39 +++++++ tests/qemuxml2argvtest.c | 1 + tools/virsh-domain.c | 53 +++++++++ tools/virsh-host.c | 45 ++++++++ 48 files changed, 1427 insertions(+), 271 deletions(-) create mode 100644 tests/qemuxml2argvdata/launch-security-sev-direct.x86_64-6.2.0.args create mode 100644 tests/qemuxml2argvdata/launch-security-sev-direct.xml -- 2.33.1

2 24

[PATCH] qemuProcessPrepareHost: Create domain private dirs as early as possible
by Michal Privoznik 13 Dec '21

13 Dec '21

As of ff024b60cc3 we are opening chardevs before starting QEMU. However, we also doing that before domain private directories are created. This leaves us impossible to create guest agent socket which lives under priv->channelTargetDir. While creating the dirs can be moved just before qemuProcessPrepareHostBackendChardev() it's better to do it as the very first step so that this kind of error is prevented in future. Fixes: ff024b60cc39d5d41b1e68728a00a47e103ec4dd Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/qemu/qemu_process.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 4963ce383f..82d0af5549 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -7024,6 +7024,14 @@ qemuProcessPrepareHost(virQEMUDriver *driver, qemuDomainObjPrivate *priv = vm->privateData; g_autoptr(virQEMUDriverConfig) cfg = virQEMUDriverGetConfig(driver); + /* + * Create all per-domain directories in order to make sure domain + * with any possible seclabels can access it. + */ + if (qemuProcessMakeDir(driver, vm, priv->libDir) < 0 || + qemuProcessMakeDir(driver, vm, priv->channelTargetDir) < 0) + return -1; + if (qemuPrepareNVRAM(driver, vm) < 0) return -1; @@ -7085,14 +7093,6 @@ qemuProcessPrepareHost(virQEMUDriver *driver, return -1; } - /* - * Create all per-domain directories in order to make sure domain - * with any possible seclabels can access it. - */ - if (qemuProcessMakeDir(driver, vm, priv->libDir) < 0 || - qemuProcessMakeDir(driver, vm, priv->channelTargetDir) < 0) - return -1; - VIR_DEBUG("Write domain masterKey"); if (qemuDomainWriteMasterKeyFile(driver, vm) < 0) return -1; -- 2.32.0

4 3

[libvirt PATCH] nss: Use shared_library() for nss_libvirt_lib
by Andrea Bolognani 13 Dec '21

13 Dec '21

shared_module() is intended for shared objects that are loaded at runtime using dlopen() whereas NSS plugins need to be full-fledged shared libraries with, among other things, a proper SONAME. Meson seems to have become more strict about this recently, because libnss_libvirt.so.2 gets a SONAME when I build it with Meson 0.59.4 on Fedora 34 but doesn't when I use Meson 0.60.2 on Debian testing instead. Either way, shared_library() was always the right function to use for NSS plugins. Fixes: 36780c931900555706fd6db9fc2ce2b4cabf9045 Signed-off-by: Andrea Bolognani <abologna(a)redhat.com> --- tools/nss/meson.build | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/nss/meson.build b/tools/nss/meson.build index 198936f3d4..f77309ebca 100644 --- a/tools/nss/meson.build +++ b/tools/nss/meson.build @@ -59,7 +59,7 @@ nss_libvirt_guest_syms = '@0@@1@'.format( meson.current_source_dir() / nss_guest_sym_file, ) -nss_libvirt_lib = shared_module( +nss_libvirt_lib = shared_library( 'nss_libvirt', name_prefix: nss_prefix, name_suffix: 'so.@0@'.format(nss_so_ver), -- 2.31.1

3 2