- Devel - libvirt List Archives

[PATCH 0/2] Minor fixes to the RPM spec
by Neal Gompa 12 May '21

12 May '21

As part of the work to add openSUSE support to the spec file, I discovered a couple of small fixes worth submitting separately. Neal Gompa (2): rpm: Simplify expression of supported platforms rpm: Drop unnecessary libiscsi runtime dependency libvirt.spec.in | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) -- 2.29.2

3 8

[libvirt PATCH 0/3] spec: Fix platform check
by Andrea Bolognani 12 May '21

12 May '21

Andrea Bolognani (3): spec: Reintroduce supported_platform variable spec: Move definition of supported_platform variable spec: Simplify platform check libvirt.spec.in | 12 +++++++++--- mingw-libvirt.spec.in | 12 +++++++++--- 2 files changed, 18 insertions(+), 6 deletions(-) -- 2.26.3

2 4

[libvirt PATCH 0/3] spec: Bump min_fedora and min_rhel
by Andrea Bolognani 11 May '21

11 May '21

Plus a couple additional cleanups. Test pipeline: https://gitlab.com/abologna/libvirt/-/pipelines/297950317 Andrea Bolognani (3): spec: Don't disable LTO in Fedora 34 spec: Bump min_fedora and min_rhel spec: Drop supported_platform variable libvirt.spec.in | 121 ++++++++---------------------------------- mingw-libvirt.spec.in | 14 ++--- 2 files changed, 25 insertions(+), 110 deletions(-) -- 2.26.3

4 8

[PATCH 0/4] Introduce a new virDomainReloadTlsCertificates API
by Zheng Yan 11 May '21

11 May '21

The new virDomainReloadTlsCertificates API is used to notify domain reload its certificates without restart, and avoid service interruption. Currently, only QEMU VNC TLS certificates are supported, but parameters and flags are also reserved for subsequent scenarios. Take reload QEMU VNC TLS certificates as an example, we can call: virDomainReloadTlsCertificates(domain, VIR_DOMAIN_TLS_CERT_GRAPHICS_VNC, NULL, 0, 0); Then the specified QMP message would be send to QEMU: {"execute": "display-reload", "arguments":{"type": "vnc", "tls-certs": true}} Refers: https://gitlab.com/qemu-project/qemu/-/commit/9cc07651655ee86eca41059f5ead8… This is v3 patch, thanks for Han Han and Michal previous review. Zheng Yan (4): qemu_capabilities: Add QEMU_CAPS_DISPLAY_RELOAD qemu: add new driver API for reload TLS certs libvirt: Introduce virDomainReloadTlsCertificates API virsh: Introduce domreload-certs sub command docs/manpages/virsh.rst | 15 +++++ include/libvirt/libvirt-domain.h | 20 ++++++ src/driver-hypervisor.h | 8 +++ src/libvirt-domain.c | 57 +++++++++++++++++ src/libvirt_public.syms | 5 ++ src/qemu/qemu_capabilities.c | 2 + src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_driver.c | 40 ++++++++++++ src/qemu/qemu_hotplug.c | 17 ++++++ src/qemu/qemu_hotplug.h | 4 ++ src/qemu/qemu_monitor.c | 27 ++++++++ src/qemu/qemu_monitor.h | 3 + src/qemu/qemu_monitor_json.c | 27 ++++++++ src/qemu/qemu_monitor_json.h | 4 ++ src/remote/remote_driver.c | 1 + src/remote/remote_protocol.x | 15 ++++- src/remote_protocol-structs | 10 +++ .../caps_6.0.0.x86_64.xml | 1 + tools/virsh-domain.c | 61 +++++++++++++++++++ 19 files changed, 317 insertions(+), 1 deletion(-) -- 2.25.1

1 4

[libvirt PATCH 0/4] gitlab: improve CI speed
by Daniel P. Berrangé 11 May '21

11 May '21

In several of the jobs we are running the libvirt build twice. We also uncessarily run the syntax-check job multiple times which causes more work for contributors looking at failed jobs. Daniel P. Berrangé (4): gitlab: only print failed tests in codestyle job gitlab: remove ability to skip 'ninja dist' gitlab: run "meson" directly instead of "ninja" indirection gitlab: avoid building libvirt twice and running syntax-check twice .gitlab-ci.yml | 31 ++++++++++++++----------------- ci/cirrus/build.yml | 6 ++++-- 2 files changed, 18 insertions(+), 19 deletions(-) -- 2.31.1

2 5

[libvirt PATCH 00/10] Refactor more XML parsing boilerplate code, part IX
by Tim Wiederhake 11 May '21

11 May '21

For background, see https://listman.redhat.com/archives/libvir-list/2021-April/msg00668.html Tim Wiederhake (10): virDomainObjParseXML: Use virXMLProp* virDomainObjParseXML: Use g_autoptr virNetworkDHCPLeaseTimeDef: Make expiry unsigned long long virNetworkDHCPLeaseTimeDefParseXML: Use virXMLProp* virPCIEDeviceInfoLinkParseXML: Use virXMLProp* virPCIEDeviceInfoLinkParseXML: Remove unused parameter ctxt virNodeDevCapsDefParseXML: Use virXMLProp* conf: node_device: Register autoptr cleanup function for virNodeDevCapsDef virNodeDevCapsDefParseXML: Use g_autoptr virNodeDeviceDefParseXML: Use virXMLProp* src/conf/domain_conf.c | 64 ++++++++++--------------- src/conf/network_conf.c | 32 +++++-------- src/conf/network_conf.h | 2 +- src/conf/node_device_conf.c | 94 ++++++++----------------------------- src/conf/node_device_conf.h | 2 + src/network/bridge_driver.c | 2 +- 6 files changed, 62 insertions(+), 134 deletions(-) -- 2.26.3

2 11

[PATCH 0/5] Couple of virCapabilities cleanups
by Michal Privoznik 11 May '21

11 May '21

I'm working on reporting HMAT in virCapabilities and noticed couple of possible cleanups. I'm sending them upfront. Michal Prívozník (5): virCapabilitiesHostNUMAFormat: Swap order of arguments virCapabilitiesHostNUMAInitReal: Free @cpus properly virCapabilitiesHostNUMAAddCell: Take double pointer virCapabilitiesHostNUMAInitReal: Use g_auto* where possible virCapabilitiesHostNUMAInitReal: Bring variables into loop src/conf/capabilities.c | 75 ++++++++++++++++++---------------- src/conf/capabilities.h | 6 +-- src/libxl/libxl_capabilities.c | 4 +- src/test/test_driver.c | 5 ++- tests/testutils.c | 6 +-- 5 files changed, 50 insertions(+), 46 deletions(-) -- 2.26.3

4 8

[libvirt PATCH v2 0/2] rST-ify and rewrite a couple of block-layer docs
by Kashyap Chamarthy 11 May '21

11 May '21

I don't think the Debian "sid" failure is related to my patch: https://gitlab.com/kashyapc/libvirt/-/pipelines/300380749/failures Changes in v2: - Mention backupBegin() [Peter] - Add examples of push-mode backup using backupBegin() [Peter, Kashyap] - Use existing rST conventions to make `rst2html5` happy [Michal] - Update meson.build [Michal] - Use .contents:: [Michal] - Paragraph alignment [Michal] Kashyap Chamarthy (2): docs: kbase: Add a doc on live full disk backup docs: kbase: Add a doc on merging disk image chains docs/kbase/live_full_disk_backup.rst | 186 +++++++++++++++++++++ docs/kbase/merging_disk_image_chains.rst | 200 +++++++++++++++++++++++ docs/kbase/meson.build | 2 + 3 files changed, 388 insertions(+) create mode 100644 docs/kbase/live_full_disk_backup.rst create mode 100644 docs/kbase/merging_disk_image_chains.rst -- 2.30.2

2 7

[libvirt PATCH v2 00/10] make internal only secrets work with split daemons
by Daniel P. Berrangé 10 May '21

10 May '21

If you define a secret with private="yes", then libvirt won't let any client query the secret value after it is set. Only other libvirt drivers inside the daemon can query it by passing a special internal only flag to the virSecretGetValue API. The remote driver/daemon refuses to let this internal flag go over the wire preventing normal clients from using it This doesn't work with the split daemons because the virSecretGetValue API done by virqemud / virtstoraged has to go over the wire to reach the virsecretd. We need to come up with an alternative way to "prove" that the caller of virSecretGetValue is a libvirt daemon, as opposed to a general libvirt client. Note with if only traditional POSIX DAC permissions are in effect then we could consider it pointless trying to restrict access to clients running the same user/group as the libvirt daemon. We ought to take into account that the client might be confined by SELinux though, so the "private secret" concept isn't entirely pointless. Thus doing a simple uid of client == uid of daemon check is a bit too weak. The UID check might also not fly if the modular daemons are run inside containers with user namespaces, as the container for virtsecretd and virtqemud might have different user mappings in theory. This series adds a concept of a "token" which is known only to the libvirt daemons. The first daemon to use it writes a random hex string to /var/run/libvirt/common/system.token. Other daemons can read and compare this. Unless a MAC system is present this is still largely security theatre, but that's not really worse than the historical behaviour. When an API call is made the virIdentity by default reflects the identity of the UNIX process that initiated it. When connecting to virtproxyd, the client apps' identity is forwarded to the next virtNNNNd daemon. When libvirt drivers, however, initiate an API call we never set any identity. With monolithic libvirtd, they'd inherit the current client identity automagically since it was all in the same thread local. With modular daemons the othe driver would see the identity of the other libvirt daemon which is bad as this gives elevated privileges in the ACL check. Thus we fix the code which drivers use to open a connection to other daemons, such that it applies the current caller's identity. It does this using an "elevated" identity though, which means, we have added in the system token. Thus the virtsecretd daemon getting the call virSecretGetValue sees the virIdentity reflecting the client application which originally called the virDomainCreate() API, but with the system token set. Thus virsecretd can see that the virSecretGetValue was invoked by another daemon, not a libvirt client app. Changed in v2... We can't set the elevated identity only when opening the virConnect for the secret driver. This works for modular daemons, as the identity is passed to the virsecretd at time of opening and thus applies to the later virSecretGetValue call on that connection. For monolithic daemon, the identity present at virConnectOpen is irrelevant. The virSecretGetValue call will just directly query the current thread's identity. IOW, to work in both deployment scenarios we need to have the elevated identity set across both virConnectOpen and virSecretGetValue Daniel P. Berrangé (10): util: add virRandomToken API util: introduce concept of a system token into identities util: generate a persistent system token util: set system token for system identity util: add API for copying identity objects util: helper to temporary elevate privileges of the current identity src: add API to determine if current identity is a system identity src: set identity when opening secondary drivers src: elevate current identity privilege when fetching secret secret: rework handling of private secrets src/driver-secret.h | 9 +- src/driver.c | 27 +++ src/libvirt-secret.c | 2 +- src/libvirt_private.syms | 7 + src/libxl/libxl_conf.c | 5 + src/qemu/qemu_domain.c | 11 +- src/qemu/qemu_tpm.c | 5 + src/remote/remote_driver.c | 8 +- src/secret/secret_driver.c | 34 ++- src/storage/storage_backend_iscsi.c | 5 + src/storage/storage_backend_iscsi_direct.c | 5 + src/storage/storage_backend_rbd.c | 5 + src/storage/storage_util.c | 5 + src/util/viridentity.c | 239 +++++++++++++++++++++ src/util/viridentity.h | 11 + src/util/virrandom.c | 18 ++ src/util/virrandom.h | 1 + src/util/virsecret.c | 3 +- tests/qemuxml2argvtest.c | 3 +- 19 files changed, 375 insertions(+), 28 deletions(-) -- 2.31.1

3 15

[PATCH] Introduce a new virDomainReloadTlsCertificates API
by Zheng Yan 10 May '21

10 May '21

The new virDomainReloadTlsCertificates API is used to notify domain reload its certificates without restart, and avoid service interruption. Currently, only QEMU VNC TLS certificates are supported, but parameters and flags are also reserved for subsequent scenarios. Take reload QEMU VNC TLS certificates as an example, we can call: virDomainReloadTlsCertificates(domain, VIR_DOMAIN_TLS_CERT_GRAPHICS_VNC, NULL, 0, 0); Then the specified QMP message would be send to QEMU: {"execute": "display-reload", "arguments":{"type": "vnc", "tls-certs": true}} Refers: https://gitlab.com/qemu-project/qemu/-/commit/9cc07651655ee86eca41059f5ead8… --- include/libvirt/libvirt-domain.h | 20 +++++++++++ src/driver-hypervisor.h | 8 +++++ src/libvirt-domain.c | 58 ++++++++++++++++++++++++++++++++ src/libvirt_public.syms | 5 +++ src/qemu/qemu_driver.c | 40 ++++++++++++++++++++++ src/qemu/qemu_hotplug.c | 17 ++++++++++ src/qemu/qemu_hotplug.h | 4 +++ src/qemu/qemu_monitor.c | 27 +++++++++++++++ src/qemu/qemu_monitor.h | 3 ++ src/qemu/qemu_monitor_json.c | 27 +++++++++++++++ src/qemu/qemu_monitor_json.h | 4 +++ src/remote/remote_driver.c | 1 + src/remote/remote_protocol.x | 15 ++++++++- src/remote_protocol-structs | 10 ++++++ 14 files changed, 238 insertions(+), 1 deletion(-) diff --git a/include/libvirt/libvirt-domain.h b/include/libvirt/libvirt-domain.h index e99bfb7654..357d3598a6 100644 --- a/include/libvirt/libvirt-domain.h +++ b/include/libvirt/libvirt-domain.h @@ -5152,4 +5152,24 @@ int virDomainStartDirtyRateCalc(virDomainPtr domain, int seconds, unsigned int flags); +/** + * virDomainTlsCertificateType: + * the used scene of TLS certificates for doamin + */ +typedef enum { + VIR_DOMAIN_TLS_CERT_GRAPHICS_VNC = 0, + VIR_DOMAIN_TLS_CERT_GRAPHICS_SPICE = 1, + +# ifdef VIR_ENUM_SENTINELS + VIR_DOMAIN_TLS_CERT_LAST +# endif +} virDomainTlsCertificateType; + +int +virDomainReloadTlsCertificates(virDomainPtr domain, + unsigned int type, + virTypedParameterPtr params, + int nparams, + unsigned int flags); + #endif /* LIBVIRT_DOMAIN_H */ diff --git a/src/driver-hypervisor.h b/src/driver-hypervisor.h index d642af8a37..d0d4976441 100644 --- a/src/driver-hypervisor.h +++ b/src/driver-hypervisor.h @@ -1410,6 +1410,13 @@ typedef int int seconds, unsigned int flags); +typedef int +(*virDrvDomainReloadTlsCertificates)(virDomainPtr domain, + unsigned int type, + virTypedParameterPtr params, + int nparams, + unsigned int flags); + typedef struct _virHypervisorDriver virHypervisorDriver; /** @@ -1676,4 +1683,5 @@ struct _virHypervisorDriver { virDrvDomainAuthorizedSSHKeysSet domainAuthorizedSSHKeysSet; virDrvDomainGetMessages domainGetMessages; virDrvDomainStartDirtyRateCalc domainStartDirtyRateCalc; + virDrvDomainReloadTlsCertificates domainReloadTlsCertificates; }; diff --git a/src/libvirt-domain.c b/src/libvirt-domain.c index 42c75f6cc5..1b6889ab49 100644 --- a/src/libvirt-domain.c +++ b/src/libvirt-domain.c @@ -13218,3 +13218,61 @@ virDomainStartDirtyRateCalc(virDomainPtr domain, virDispatchError(conn); return -1; } + +/** + * virDomainReloadTlsCertificates: + * @domain: a domain object. + * @type: a value of virDomainTlsCertificateType + * @params: pointer to TLS Certs parameter objects, must be NULL if not used + * @nparams: number of TLS Certs parameter objects, must be 0 if not used + * @flags: extra flags; not used yet, so callers should always pass 0 + * + * Notify domain reload its certificates with specified 'type' + * + * Returns 0 in case of success, -1 otherwise. + */ +int +virDomainReloadTlsCertificates(virDomainPtr domain, + unsigned int type, + virTypedParameterPtr params, + int nparams, + unsigned int flags) +{ + virConnectPtr conn; + + VIR_DOMAIN_DEBUG(domain, "certificate type=%u, params=%p, nparams=%d, flags=%x", + type, params, nparams, flags); + + virResetLastError(); + + virCheckDomainReturn(domain, -1); + conn = domain->conn; + + virCheckReadOnlyGoto(conn->flags, error); + virCheckNonNegativeArgGoto(nparams, error); + + if (type >= VIR_DOMAIN_TLS_CERT_LAST) { + virReportInvalidArg(type, + _("type must be less than %d"), + VIR_DOMAIN_TLS_CERT_LAST); + goto error; + } + + if (conn->driver->domainReloadTlsCertificates) { + int ret; + ret = conn->driver->domainReloadTlsCertificates(domain, + type, + params, + nparams, + flags); + if (ret < 0) + goto error; + return ret; + } + + virReportUnsupportedError(); + + error: + virDispatchError(domain->conn); + return -1; +} diff --git a/src/libvirt_public.syms b/src/libvirt_public.syms index 5678a13cda..30ff012958 100644 --- a/src/libvirt_public.syms +++ b/src/libvirt_public.syms @@ -896,4 +896,9 @@ LIBVIRT_7.3.0 { virNodeDeviceCreate; } LIBVIRT_7.2.0; +LIBVIRT_7.4.0 { + global: + virDomainReloadTlsCertificates; +} LIBVIRT_7.3.0; + # .... define new API here using predicted next version number .... diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index c90d52edc0..422a350c65 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -20449,6 +20449,45 @@ qemuDomainStartDirtyRateCalc(virDomainPtr dom, return ret; } +static int +qemuDomainReloadTlsCertificates(virDomainPtr domain, + unsigned int type, + virTypedParameterPtr params, + int nparams, + unsigned int flags) +{ + int ret = -1; + virQEMUDriver *driver = domain->conn->privateData; + virDomainObj *vm = qemuDomObjFromDomain(domain); + + if (!vm) + goto cleanup; + + virCheckNonNullArgGoto(params, cleanup); + if (nparams != 0) { + virReportInvalidZeroArg(nparams); + goto cleanup; + } + virCheckFlagsGoto(0, cleanup); + + if (qemuDomainObjBeginJob(driver, vm, QEMU_JOB_MODIFY) < 0) + goto cleanup; + + if (!virDomainObjIsActive(vm)) { + virReportError(VIR_ERR_OPERATION_INVALID, + "%s", _("domain is not running")); + goto endjob; + } + + ret = qemuDomainReloadTLSCerts(driver, vm, type); + + endjob: + qemuDomainObjEndJob(driver, vm); + + cleanup: + virDomainObjEndAPI(&vm); + return ret; +} static virHypervisorDriver qemuHypervisorDriver = { .name = QEMU_DRIVER_NAME, @@ -20693,6 +20732,7 @@ static virHypervisorDriver qemuHypervisorDriver = { .domainAuthorizedSSHKeysSet = qemuDomainAuthorizedSSHKeysSet, /* 6.10.0 */ .domainGetMessages = qemuDomainGetMessages, /* 7.1.0 */ .domainStartDirtyRateCalc = qemuDomainStartDirtyRateCalc, /* 7.2.0 */ + .domainReloadTlsCertificates = qemuDomainReloadTlsCertificates, /* 7.4.0 */ }; diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index a64cddb9e7..34dc035d73 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -6704,3 +6704,20 @@ qemuDomainSetVcpuInternal(virQEMUDriver *driver, virBitmapFree(livevcpus); return ret; } + +int qemuDomainReloadTLSCerts(virQEMUDriverPtr driver, + virDomainObjPtr vm, + unsigned int type) +{ + int ret = -1; + qemuDomainObjPrivate *priv = vm->privateData; + + qemuDomainObjEnterMonitor(driver, vm); + + ret = qemuMonitorDisplayReloadTLSCerts(priv->mon, type); + + if (qemuDomainObjExitMonitor(driver, vm) < 0) + ret = -1; + + return ret; +} diff --git a/src/qemu/qemu_hotplug.h b/src/qemu/qemu_hotplug.h index df8f76f8d6..411741a688 100644 --- a/src/qemu/qemu_hotplug.h +++ b/src/qemu/qemu_hotplug.h @@ -160,3 +160,7 @@ int qemuHotplugAttachDBusVMState(virQEMUDriver *driver, int qemuHotplugRemoveDBusVMState(virQEMUDriver *driver, virDomainObj *vm, qemuDomainAsyncJob asyncJob); + +int qemuDomainReloadTLSCerts(virQEMUDriverPtr driver, + virDomainObjPtr vm, + unsigned int type); diff --git a/src/qemu/qemu_monitor.c b/src/qemu/qemu_monitor.c index 1e6f892e08..11f8cc8670 100644 --- a/src/qemu/qemu_monitor.c +++ b/src/qemu/qemu_monitor.c @@ -4746,3 +4746,30 @@ qemuMonitorQueryDirtyRate(qemuMonitor *mon, return qemuMonitorJSONQueryDirtyRate(mon, info); } + +static const char * +qemuMonitorTLSCertificateTypeToString(unsigned int type) +{ + switch (type) { + /* for now, only VNC is supported */ + case VIR_DOMAIN_TLS_CERT_GRAPHICS_VNC: + return "vnc"; + default: + virReportError(VIR_ERR_INVALID_ARG, + _("unsupported qemu certificate type %u"), + type); + return NULL; + } +} + +int +qemuMonitorDisplayReloadTLSCerts(qemuMonitorPtr mon, unsigned int type) +{ + const char *certType = qemuMonitorTLSCertificateTypeToString(type); + if (!certType) + return -1; + + QEMU_CHECK_MONITOR(mon); + + return qemuMonitorJSONDisplayReload(mon, certType, true); +} diff --git a/src/qemu/qemu_monitor.h b/src/qemu/qemu_monitor.h index 6a25def78b..f26f92fb51 100644 --- a/src/qemu/qemu_monitor.h +++ b/src/qemu/qemu_monitor.h @@ -1496,3 +1496,6 @@ struct _qemuMonitorDirtyRateInfo { int qemuMonitorQueryDirtyRate(qemuMonitor *mon, qemuMonitorDirtyRateInfo *info); + +int qemuMonitorDisplayReloadTLSCerts(qemuMonitorPtr mon, + unsigned int type); diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index 46aa3330a8..9934613cc2 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -9446,3 +9446,30 @@ qemuMonitorJSONQueryDirtyRate(qemuMonitor *mon, return qemuMonitorJSONExtractDirtyRateInfo(data, info); } + +int qemuMonitorJSONDisplayReload(qemuMonitorPtr mon, + const char *type, + bool tlsCerts) +{ + int ret = -1; + virJSONValuePtr reply = NULL; + virJSONValuePtr cmd = qemuMonitorJSONMakeCommand("display-reload", + "s:type", type, + "b:tls-certs", tlsCerts, + NULL); + if (!cmd) + return -1; + + if (qemuMonitorJSONCommand(mon, cmd, &reply) < 0) + goto cleanup; + + if (qemuMonitorJSONCheckError(cmd, reply) < 0) + goto cleanup; + + ret = 0; + + cleanup: + virJSONValueFree(cmd); + virJSONValueFree(reply); + return ret; +} diff --git a/src/qemu/qemu_monitor_json.h b/src/qemu/qemu_monitor_json.h index 01a3ba25f1..73761d54f8 100644 --- a/src/qemu/qemu_monitor_json.h +++ b/src/qemu/qemu_monitor_json.h @@ -706,3 +706,7 @@ qemuMonitorJSONStartDirtyRateCalc(qemuMonitor *mon, int qemuMonitorJSONQueryDirtyRate(qemuMonitor *mon, qemuMonitorDirtyRateInfo *info); + +int qemuMonitorJSONDisplayReload(qemuMonitorPtr mon, + const char *type, + bool tlsCerts); diff --git a/src/remote/remote_driver.c b/src/remote/remote_driver.c index 0c72d69933..0e6e4e3007 100644 --- a/src/remote/remote_driver.c +++ b/src/remote/remote_driver.c @@ -8566,6 +8566,7 @@ static virHypervisorDriver hypervisor_driver = { .domainAuthorizedSSHKeysSet = remoteDomainAuthorizedSSHKeysSet, /* 6.10.0 */ .domainGetMessages = remoteDomainGetMessages, /* 7.1.0 */ .domainStartDirtyRateCalc = remoteDomainStartDirtyRateCalc, /* 7.2.0 */ + .domainReloadTlsCertificates = remoteDomainReloadTlsCertificates, /* 7.4.0 */ }; static virNetworkDriver network_driver = { diff --git a/src/remote/remote_protocol.x b/src/remote/remote_protocol.x index de69704b68..c13b3e1f5c 100644 --- a/src/remote/remote_protocol.x +++ b/src/remote/remote_protocol.x @@ -286,6 +286,8 @@ const REMOTE_DOMAIN_AUTHORIZED_SSH_KEYS_MAX = 2048; /* Upper limit on number of messages */ const REMOTE_DOMAIN_MESSAGES_MAX = 2048; +/* Upper limit on list of TLS certificate parameters */ +const REMOTE_DOMAIN_RELOAD_TLS_CERT_PARAMETERS_MAX = 16; /* UUID. VIR_UUID_BUFLEN definition comes from libvirt.h */ typedef opaque remote_uuid[VIR_UUID_BUFLEN]; @@ -3836,6 +3838,12 @@ struct remote_domain_start_dirty_rate_calc_args { unsigned int flags; }; +struct remote_domain_reload_tls_certificates_args { + remote_nonnull_domain dom; + unsigned int type; + remote_typed_param params<REMOTE_DOMAIN_RELOAD_TLS_CERT_PARAMETERS_MAX>; + unsigned int flags; +}; /*----- Protocol. -----*/ @@ -6784,6 +6792,11 @@ enum remote_procedure { * @priority: high * @acl: node_device:start */ - REMOTE_PROC_NODE_DEVICE_CREATE = 430 + REMOTE_PROC_NODE_DEVICE_CREATE = 430, + /** + * @generate: both + * @acl: domain:write + */ + REMOTE_PROC_DOMAIN_RELOAD_TLS_CERTIFICATES = 431 }; diff --git a/src/remote_protocol-structs b/src/remote_protocol-structs index 6b46328adc..799a8596ea 100644 --- a/src/remote_protocol-structs +++ b/src/remote_protocol-structs @@ -3192,6 +3192,15 @@ struct remote_domain_start_dirty_rate_calc_args { int seconds; u_int flags; }; +struct remote_domain_reload_tls_certificates_args { + remote_nonnull_domain dom; + u_int type; + struct { + u_int params_len; + remote_typed_param * params_val; + } params; + u_int flags; +}; enum remote_procedure { REMOTE_PROC_CONNECT_OPEN = 1, REMOTE_PROC_CONNECT_CLOSE = 2, @@ -3623,4 +3632,5 @@ enum remote_procedure { REMOTE_PROC_NODE_DEVICE_DEFINE_XML = 428, REMOTE_PROC_NODE_DEVICE_UNDEFINE = 429, REMOTE_PROC_NODE_DEVICE_CREATE = 430, + REMOTE_PROC_DOMAIN_RELOAD_TLS_CERTIFICATES = 431, }; -- 2.25.1

2 1

[PATCH] securityselinuxhelper: Fix retval of setcon_raw() and security_disable()
by Michal Privoznik 10 May '21

10 May '21

The securityselinuxhelper is a mock that's replacing libselinux APIs with our own implementation to achieve deterministic results. Our implementation uses env vars (among other things) to hold internal state. For instance, "FAKE_SELINUX_CONTEXT" and "FAKE_SELINUX_DISABLED" variables are used. However, as we were switching from setenv() to g_setenv() we also changed the set of possible retvals from setcon_raw() and security_disable(). Previously, the retval of setenv() was used directly which returns 0 on success and -1 on error. But g_setenv() has different retval semantics: it returns 1 on success and 0 on error. This discrepancy can be observed by running viridentitytest where case #2 reports an error ("!") - because setcon_raw() returns 1. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- tests/securityselinuxhelper.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/securityselinuxhelper.c b/tests/securityselinuxhelper.c index b308ef5095..c3d6505ef2 100644 --- a/tests/securityselinuxhelper.c +++ b/tests/securityselinuxhelper.c @@ -140,7 +140,7 @@ int setcon_raw(const char *context) errno = EINVAL; return -1; } - return g_setenv("FAKE_SELINUX_CONTEXT", context, TRUE); + return g_setenv("FAKE_SELINUX_CONTEXT", context, TRUE) == TRUE ? 0 : -1; } int setcon(const char *context) @@ -219,7 +219,7 @@ int security_disable(void) return -1; } - return g_setenv("FAKE_SELINUX_DISABLED", "1", TRUE); + return g_setenv("FAKE_SELINUX_DISABLED", "1", TRUE) == TRUE ? 0 : -1; } int security_getenforce(void) -- 2.26.3

2 1

[libvirt PATCH v3 0/8] cleanup meson checks for runtime binaries
by Pavel Hrdina 10 May '21

10 May '21

Recent attempt to add a lot of meson options to specify different runtime paths motivated me enough to cleanup this from meson. Changes in v3: - some patches were already pushed - removed patch that moved virFindFileInPath from testutilsqemu.c Changes in v2: - split and rework patch 16/17 to address review comments - added a new patch to cleanup libvirt.spec.in file Pavel Hrdina (8): virfile: introduce virFindFileInPathFull() qemu_conf: use virFindFileInPathFull for runtime binaries meson: drop check for runtime binary dependencies meson: move iscsiadm check into storage_iscsi condition meson: stop setting runtime binaries defines during compilation meson: use runtime binaries to only resolve features with "auto" value meson: optional_programs should be used only for building libvirt libvirt.spec: drop no longer required build dependencies libvirt.spec.in | 31 ---- meson.build | 214 +++++++++---------------- src/bhyve/bhyve_command.c | 4 + src/libvirt_private.syms | 2 +- src/locking/lock_driver_lockd.c | 12 +- src/network/bridge_driver.c | 2 + src/node_device/node_device_driver.c | 2 + src/qemu/qemu_conf.c | 23 ++- src/storage/storage_backend_logical.c | 13 ++ src/storage/storage_backend_sheepdog.c | 2 + src/storage/storage_backend_zfs.c | 3 + src/storage/storage_util.c | 2 + src/storage/storage_util.h | 6 + src/util/virdnsmasq.c | 1 + src/util/virfile.c | 16 +- src/util/virfile.h | 5 +- src/util/virfirewall.h | 4 + src/util/viriscsi.h | 2 + src/util/virkmod.h | 3 + src/util/virnetdevbandwidth.h | 2 + src/util/virnetdevip.c | 2 + src/util/virnetdevmidonet.c | 2 + src/util/virnetdevopenvswitch.c | 2 + src/util/virnuma.c | 1 + src/util/virsysinfo.c | 1 + src/util/virutil.c | 2 + tests/testutilsqemu.c | 3 +- tests/virfirewallmock.c | 3 +- 28 files changed, 173 insertions(+), 192 deletions(-) -- 2.30.2

1 9

[PATCH v3 0/1] qemu: add support for max-ram-below-4g option
by Zhiyong Ye 10 May '21

10 May '21

For background, see: https://listman.redhat.com/archives/libvir-list/2021-April/msg01196.html V3 to V2: * Fix conflicts created by git pull V2 to V1: * Change the 'below4g' attribute to not share the 'unit' argument with actual memory size, and let its unit default to KiB * Add docs and test changes * Add check in the ABI stability checker code Zhiyong Ye (1): qemu: add support for max-ram-below-4g option docs/formatdomain.rst | 10 ++++++-- docs/schemas/domaincommon.rng | 5 ++++ src/conf/domain_conf.c | 15 ++++++++++++ src/conf/domain_conf.h | 3 +++ src/conf/domain_validate.c | 13 ++++++++++ src/qemu/qemu_command.c | 4 ++++ tests/qemuxml2argvdata/memory-below4g.args | 29 ++++++++++++++++++++++ tests/qemuxml2argvdata/memory-below4g.xml | 26 ++++++++++++++++++++ tests/qemuxml2argvtest.c | 1 + tests/qemuxml2xmloutdata/memory-below4g.xml | 37 +++++++++++++++++++++++++++++ tests/qemuxml2xmltest.c | 1 + 11 files changed, 142 insertions(+), 2 deletions(-) create mode 100644 tests/qemuxml2argvdata/memory-below4g.args create mode 100644 tests/qemuxml2argvdata/memory-below4g.xml create mode 100644 tests/qemuxml2xmloutdata/memory-below4g.xml -- 2.11.0

1 1

After blockcopy with pivot, a dirty-bitmap is added, but not been removed automatically.
by ChangLimin 10 May '21

10 May '21

Hi Peter, Since https://libvirt.org/git/?p=libvirt.git;a=commitdiff;h=7bfff40fdfe5410c446c1…, after blockcopy with pivot, a dirty-bitmap is added, but not been removed automatically. Is that something wrong? 0. start a vm, and switch on some traces # virsh start cirros # virsh qemu-monitor-command cirros --hmp "trace-event handle_qmp* on" # virsh qemu-monitor-command cirros --hmp "trace-event hbitmap* on" 1. cirros0.img is opened. # virsh domblklist cirros Target Source ------------------------------------ vda /vms/images1/cirros0.img 2. blockcopy and pivot to cirros1.img # virsh blockcopy cirros vda --dest /vms/images1/cirros1.img --pivot --transient-job Successfully pivoted 3. cirros1.img is opened # virsh domblklist cirros Target Source ------------------------------------ vda /vms/images1/cirros1.img 4. From qemu log, a dirty-bitmap is added @1620475221.389151 77998@1620475125.281656:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "human-monitor-command", "arguments": {"command-line": "trace-event qmp_job* on"}, "id": "libvirt-380"} 77998@1620475125.299950:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "human-monitor-command", "arguments": {"command-line": "trace-event mirror* on"}, "id": "libvirt-381"} 77998@1620475125.317931:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "human-monitor-command", "arguments": {"command-line": "trace-event hbitmap* on"}, "id": "libvirt-382"} 77998@1620475217.640133:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "query-named-block-nodes", "arguments": {"flat": true}, "id": "libvirt-383"} 77998@1620475217.642025:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "blockdev-add", "arguments": {"node-name": "libvirt-2-storage", "driver": "file", "filename": "/vms/images1/cirros1.img", "auto-read-only": true, "aio": "native", "cache": {"no-flush": false, "direct": true}, "discard": "unmap"}, "id": "libvirt-384"} 77998@1620475217.643434:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "blockdev-create", "arguments": {"job-id": "create-libvirt-2-format", "options": {"driver": "qcow2", "cluster-size": 65536, "size": 46137344, "file": "libvirt-2-storage"}}, "id": "libvirt-385"} 77998@1620475217.827378:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "query-jobs", "id": "libvirt-386"} 77998@1620475217.827980:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "job-dismiss", "arguments": {"id": "create-libvirt-2-format"}, "id": "libvirt-387"} 77998@1620475217.850802:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "blockdev-add", "arguments": {"backing": null, "node-name": "libvirt-2-format", "driver": "qcow2", "read-only": false, "cache": {"no-flush": false, "direct": true}, "file": "libvirt-2-storage"}, "id": "libvirt-388"} 77998@1620475217.853157:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "blockdev-mirror", "arguments": {"auto-finalize": true, "device": "libvirt-1-format", "auto-dismiss": false, "sync": "full", "job-id": "copy-vda-libvirt-1-format", "target": "libvirt-2-format"}, "id": "libvirt-389"} 77998@1620475217.871674:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "query-block-jobs", "id": "libvirt-390"} 77998@1620475221.387625:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "query-block-jobs", "id": "libvirt-396"} 77998@1620475221.389151:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "transaction", "arguments": {"actions": [{"type": "block-dirty-bitmap-add", "data": {"name": "libvirt-tmp-activewrite", "persistent": false, "node": "libvirt-2-format", "disabled": false}}]}, "id": "libvirt-397"} 77998@1620475221.389755:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "job-complete", "arguments": {"id": "copy-vda-libvirt-1-format"}, "id": "libvirt-398"} 77998@1620475221.391321:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "query-jobs", "id": "libvirt-399"} 77998@1620475221.391926:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "job-dismiss", "arguments": {"id": "copy-vda-libvirt-1-format"}, "id": "libvirt-400"} 77998@1620475221.393052:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "blockdev-del", "arguments": {"node-name": "libvirt-1-format"}, "id": "libvirt-401"} 77998@1620475221.393702:handle_qmp_command mon 0x55ebd4b176b0 req: {"execute": "blockdev-del", "arguments": {"node-name": "libvirt-1-storage"}, "id": "libvirt-402"} Also there are many hbitmap_xx events. 5. I have debuged the source, in qemuBlockJobProcessEventCompletedCopyBitmaps , QEMU_CAPS_BLOCKDEV_REOPEN cap is not supported unti the latest qemu. So the bitmap is not removed. Is the QEMU_CAPS_BLOCKDEV_REOPEN cap be must? Regards, Chang Limin

1 0

[PATCH v2 0/1] qemu: add support for max-ram-below-4g option
by Zhiyong Ye 10 May '21

10 May '21

Hi all, Sorry to bother again. Since I'm new to the libvirt community, and also this is my first time to submit patches, I'm truly sorry for the format problems that I emailed before. Could you kindly help me review it again, and let me know if there is anything I need to change. For background, see: https://listman.redhat.com/archives/libvir-list/2021-April/msg01196.html Changes since V1: * Change the 'below4g' attribute to not share the 'unit' argument with actual memory size, and let its unit default to KiB * Add docs and test changes * Add check in the ABI stability checker code Best regards! Zhiyong Ye Zhiyong Ye (1): qemu: add support for max-ram-below-4g option docs/formatdomain.rst | 10 ++++++-- docs/schemas/domaincommon.rng | 5 ++++ src/conf/domain_conf.c | 15 ++++++++++++ src/conf/domain_conf.h | 3 +++ src/conf/domain_validate.c | 13 ++++++++++ src/qemu/qemu_command.c | 4 ++++ tests/qemuxml2argvdata/memory-below4g.args | 29 ++++++++++++++++++++++ tests/qemuxml2argvdata/memory-below4g.xml | 26 ++++++++++++++++++++ tests/qemuxml2argvtest.c | 1 + tests/qemuxml2xmloutdata/memory-below4g.xml | 37 +++++++++++++++++++++++++++++ tests/qemuxml2xmltest.c | 1 + 11 files changed, 142 insertions(+), 2 deletions(-) create mode 100644 tests/qemuxml2argvdata/memory-below4g.args create mode 100644 tests/qemuxml2argvdata/memory-below4g.xml create mode 100644 tests/qemuxml2xmloutdata/memory-below4g.xml -- 2.11.0

2 3

[libvirt PATCH 0/3] cpu-data.py: Add support for kcpuid
by Tim Wiederhake 07 May '21

07 May '21

Linux 5.13 introduces "kcpuid", a tool similar to cpuid, which we depend on for cpu-data collection with tests/cputestdata/cpu-data.py. This series adds support for kcpuid to cpu-data.py. See also https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/too… /arch/x86/kcpuid and https://lore.kernel.org/lkml/1614928878-86075-1-git-send-email-feng.tang@in… l.com/ Tim Wiederhake (3): cpu-data.py: Factor out cpuid parsing cpu-data.py: Parse kcpuid output cpu-data.py: Automatically adjust command line for kcpuid tests/cputestdata/cpu-data.py | 95 +++++++++++++++++++++++++++-------- 1 file changed, 75 insertions(+), 20 deletions(-) --=20 2.26.3

2 5

[libvirt PATCH 0/9] make internal only secrets work with split daemons
by Daniel P. Berrangé 07 May '21

07 May '21

If you define a secret with private="yes", then libvirt won't let any client query the secret value after it is set. Only other libvirt drivers inside the daemon can query it by passing a special internal only flag to the virSecretGetValue API. The remote driver/daemon refuses to let this internal flag go over the wire preventing normal clients from using it This doesn't work with the split daemons because the virSecretGetValue API done by virqemud / virtstoraged has to go over the wire to reach the virsecretd. We need to come up with an alternative way to "prove" that the caller of virSecretGetValue is a libvirt daemon, as opposed to a general libvirt client. Note with if only traditional POSIX DAC permissions are in effect then we could consider it pointless trying to restrict access to clients running the same user/group as the libvirt daemon. We ought to take into account that the client might be confined by SELinux though, so the "private secret" concept isn't entirely pointless. Thus doing a simple uid of client == uid of daemon check is a bit too weak. The UID check might also not fly if the modular daemons are run inside containers with user namespaces, as the container for virtsecretd and virtqemud might have different user mappings in theory. This series adds a concept of a "token" which is known only to the libvirt daemons. The first daemon to use it writes a random hex string to /var/run/libvirt/common/system.token. Other daemons can read and compare this. Unless a MAC system is present this is still largely security theatre, but that's not really worse than the historical behaviour. When an API call is made the virIdentity by default reflects the identity of the UNIX process that initiated it. When connecting to virtproxyd, the client apps' identity is forwarded to the next virtNNNNd daemon. When libvirt drivers, however, initiate an API call we never set any identity. With monolithic libvirtd, they'd inherit the current client identity automagically since it was all in the same thread local. With modular daemons the othe driver would see the identity of the other libvirt daemon which is bad as this gives elevated privileges in the ACL check. Thus we fix the code which drivers use to open a connection to other daemons, such that it applies the current caller's identity. It does this using an "elevated" identity though, which means, we have added in the system token. Thus the virtsecretd daemon getting the call virSecretGetValue sees the virIdentity reflecting the client application which originally called the virDomainCreate() API, but with the system token set. Thus virsecretd can see that the virSecretGetValue was invoked by another daemon, not a libvirt client app. Daniel P. Berrangé (9): util: add virRandomToken API util: introduce concept of a system token into identities util: generate a persistent system token src: set system token for system identity util: add API for copying identity objects util: add method for getting the current identity with system token src: add API to determine if current identity is a system identity secret: rework handling of private secrets src: set identity when opening secondary drivers src/driver-secret.h | 9 +- src/driver.c | 27 +++++ src/libvirt-secret.c | 2 +- src/libvirt_private.syms | 5 + src/remote/remote_driver.c | 8 +- src/secret/secret_driver.c | 34 ++++-- src/util/viridentity.c | 230 +++++++++++++++++++++++++++++++++++++ src/util/viridentity.h | 7 ++ src/util/virrandom.c | 18 +++ src/util/virrandom.h | 1 + src/util/virsecret.c | 3 +- tests/qemuxml2argvtest.c | 3 +- 12 files changed, 320 insertions(+), 27 deletions(-) -- 2.31.1

2 20

[libvirt PATCH 0/2] rST-ify and rewrite a couple of block layer-related docs
by Kashyap Chamarthy 07 May '21

07 May '21

I've cleaned up these two wiki articles that I wrote a while ago and turned them into rST. https://wiki.libvirt.org/page/Live-disk-backup-with-active-blockcommit https://wiki.libvirt.org/page/Live-merge-an-entire-disk-image-chain-includi… Once this series merges, I'll make the above two wiki pages point to the kbase docs. Kashyap Chamarthy (2): docs: kbase: Add a doc on live full disk backup docs: kbase: Add a doc on merging disk image chains docs/kbase/index.rst | 6 + docs/kbase/live_full_disk_backup.rst | 111 +++++++++++++ docs/kbase/merging_disk_image_chains.rst | 197 +++++++++++++++++++++++ 3 files changed, 314 insertions(+) create mode 100644 docs/kbase/live_full_disk_backup.rst create mode 100644 docs/kbase/merging_disk_image_chains.rst -- 2.30.2

3 8

[PATCH v1] Introduce virDomainReloadTLSCertificates API
by Yanzheng (A) 07 May '21

07 May '21

Introduce a new virDomainReloadTLSCertificates API for notify domain reload its certificates without restart, and avoid service interruption. Take reload QEMU VNC TLS certificates as an example, we can call: virDomainReloadTLSCertificates(dom, VIR_DOMAIN_TLS_CERT_GRAPHICS_VNC) Then the specified QMP message would be send to QEMU: {"execute": "display-reload", "arguments":{"type": "vnc", "tls-certs": true}} Refers: https://gitlab.com/qemu-project/qemu/-/commit/9cc07651655ee86eca41059f5ead8… --- include/libvirt/libvirt-domain.h | 17 ++++++++++++++++ src/driver-hypervisor.h | 5 +++++ src/libvirt-domain.c | 33 ++++++++++++++++++++++++++++++++ src/qemu/qemu_driver.c | 11 +++++++++++ src/qemu/qemu_hotplug.c | 21 ++++++++++++++++++++ src/qemu/qemu_hotplug.h | 4 ++++ src/qemu/qemu_monitor.c | 10 ++++++++++ src/qemu/qemu_monitor.h | 3 +++ src/qemu/qemu_monitor_json.c | 22 +++++++++++++++++++++ src/qemu/qemu_monitor_json.h | 3 +++ 10 files changed, 129 insertions(+) diff --git a/include/libvirt/libvirt-domain.h b/include/libvirt/libvirt-domain.h index e99bfb7654..aeb33d69d9 100644 --- a/include/libvirt/libvirt-domain.h +++ b/include/libvirt/libvirt-domain.h @@ -5152,4 +5152,21 @@ int virDomainStartDirtyRateCalc(virDomainPtr domain, int seconds, unsigned int flags); +/** + * virDomainTLSCertificaType: + * + * the used scene of TLS certificates for doamin. + */ +typedef enum { + VIR_DOMAIN_TLS_CERT_GRAPHICS_VNC = 0, + VIR_DOMAIN_TLS_CERT_GRAPHICS_SPICE = 1, + + VIR_DOMAIN_TLS_CERT_LAST +} virDomainTLSCertificaType; + +int +virDomainReloadTLSCertificates(virDomainPtr domain, + unsigned int type); + + #endif /* LIBVIRT_DOMAIN_H */ diff --git a/src/driver-hypervisor.h b/src/driver-hypervisor.h index d642af8a37..8de2bc4137 100644 --- a/src/driver-hypervisor.h +++ b/src/driver-hypervisor.h @@ -1410,6 +1410,10 @@ typedef int int seconds, unsigned int flags); +typedef int +(*virDrvDomainReloadTLSCertificates)(virDomainPtr domain, + unsigned int type); + typedef struct _virHypervisorDriver virHypervisorDriver; /** @@ -1676,4 +1680,5 @@ struct _virHypervisorDriver { virDrvDomainAuthorizedSSHKeysSet domainAuthorizedSSHKeysSet; virDrvDomainGetMessages domainGetMessages; virDrvDomainStartDirtyRateCalc domainStartDirtyRateCalc; + virDrvDomainReloadTLSCertificates domainiReloadTLSCertificates; }; diff --git a/src/libvirt-domain.c b/src/libvirt-domain.c index 42c75f6cc5..fb9e5ec2d1 100644 --- a/src/libvirt-domain.c +++ b/src/libvirt-domain.c @@ -13218,3 +13218,36 @@ virDomainStartDirtyRateCalc(virDomainPtr domain, virDispatchError(conn); return -1; } + +/** + * virDomainReloadTLSCertificates: + * @domain: a domain object. + * @type: a value of virDomainTLSCertificaType + * + * Notify domain reload its certificates with specified 'type'. + * + * Returns 0 in case of success, -1 otherwise . + */ +int +virDomainReloadTLSCertificates(virDomainPtr domain, + unsigned int type) +{ + virConnectPtr conn; + VIR_DOMAIN_DEBUG(domain, "certificate type=%d", type); + virResetLastError(); + virCheckDomainReturn(domain, -1); + conn = domain->conn; + if (type >= VIR_DOMAIN_TLS_CERT_LAST) + goto error; + if (conn->driver->domainiReloadTLSCertificates) { + int ret; + ret = conn->driver->domainiReloadTLSCertificates(domain, type); + if (ret < 0) + goto error; + return ret; + } + virReportUnsupportedError(); + error: + virDispatchError(domain->conn); + return -1; +} diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index c90d52edc0..61cd8cfa24 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -20449,6 +20449,16 @@ qemuDomainStartDirtyRateCalc(virDomainPtr dom, return ret; } +static int +qemuDomainReloadTLSCertificates(virDomainPtr domain, + unsigned int type) +{ + virQEMUDriver *driver = domain->conn->privateData; + virDomainObj *vm = qemuDomainObjFromDomain(domain); + if (!driver || !vm) + return -1; + return qemuDomainReloadTLSCerts(driver, vm, type); +} static virHypervisorDriver qemuHypervisorDriver = { .name = QEMU_DRIVER_NAME, @@ -20693,6 +20703,7 @@ static virHypervisorDriver qemuHypervisorDriver = { .domainAuthorizedSSHKeysSet = qemuDomainAuthorizedSSHKeysSet, /* 6.10.0 */ .domainGetMessages = qemuDomainGetMessages, /* 7.1.0 */ .domainStartDirtyRateCalc = qemuDomainStartDirtyRateCalc, /* 7.2.0 */ + .domainiReloadTLSCertificates = qemuDomainReloadTLSCertificates, /* 7.2.0 */ }; diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 444d89d64a..013d8728a0 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -6704,3 +6704,24 @@ qemuDomainSetVcpuInternal(virQEMUDriver *driver, virBitmapFree(livevcpus); return ret; } + +int qemuDomainReloadTLSCerts(virQEMUDriverPtr driver, + virDomainObjPtr vm, + int type) +{ + int ret = -1; + qemuDomainObjPrivate *priv = vm->privateData; + /* for now, only VNC is supported */ + if (type != VIR_DOMAIN_TLS_CERT_GRAPHICS_VNC) + virReportError(VIR_ERR_INVALID_ARG, + _("invalid certificate type=%d, only support VNC"), + type); + return ret; + } + if (qemuDomainObjEnterMonitorAsync(driver, vm, QEMU_ASYNC_JOB_NONE) < 0) + return ret; + ret = qemuMonitorReloadTLSCerts(priv->mon, type); + if (qemuDomainObjExitMonitor(driver, vm) < 0) + ret = -1; + return ret; +} diff --git a/src/qemu/qemu_hotplug.h b/src/qemu/qemu_hotplug.h index df8f76f8d6..44afe23f0a 100644 --- a/src/qemu/qemu_hotplug.h +++ b/src/qemu/qemu_hotplug.h @@ -160,3 +160,7 @@ int qemuHotplugAttachDBusVMState(virQEMUDriver *driver, int qemuHotplugRemoveDBusVMState(virQEMUDriver *driver, virDomainObj *vm, qemuDomainAsyncJob asyncJob); + +int qemuDomainReloadTLSCerts(virQEMUDriverPtr driver, + virDomainObjPtr vm, + int type); diff --git a/src/qemu/qemu_monitor.c b/src/qemu/qemu_monitor.c index 3a7f231ce0..952ef87a6b 100644 --- a/src/qemu/qemu_monitor.c +++ b/src/qemu/qemu_monitor.c @@ -4746,3 +4746,13 @@ qemuMonitorQueryDirtyRate(qemuMonitor *mon, return qemuMonitorJSONQueryDirtyRate(mon, info); } + +int +qemuMonitorReloadTLSCerts(qemuMonitorPtr mon, int type) +{ + const char *protocol = qemuMonitorTypeToProtocol(type); + if (!protocol) + return -1; + VIR_DEBUG("protocol=%s", protocol); + return qemuMonitorJSONReloadTLSCerts(mon, protocol); +} diff --git a/src/qemu/qemu_monitor.h b/src/qemu/qemu_monitor.h index 6a25def78b..a5b702b023 100644 --- a/src/qemu/qemu_monitor.h +++ b/src/qemu/qemu_monitor.h @@ -1496,3 +1496,6 @@ struct _qemuMonitorDirtyRateInfo { int qemuMonitorQueryDirtyRate(qemuMonitor *mon, qemuMonitorDirtyRateInfo *info); + +int qemuMonitorReloadTLSCerts(qemuMonitorPtr mon, + int type); diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index 46aa3330a8..d2b06c4703 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -9446,3 +9446,25 @@ qemuMonitorJSONQueryDirtyRate(qemuMonitor *mon, return qemuMonitorJSONExtractDirtyRateInfo(data, info); } + +int qemuMonitorJSONReloadTLSCerts(qemuMonitorPtr mon, + const char *protocol) +{ + int ret = -1; + virJSONValuePtr reply = NULL; + virJSONValuePtr cmd = qemuMonitorJSONMakeCommand("display-reload", + "s:type", protocol, + "b:tls-certs", 1, + NULL); + if (!cmd) + return -1; + if (qemuMonitorJSONCommand(mon, cmd, &reply) < 0) + goto cleanup; + if (qemuMonitorJSONCheckError(cmd, reply) < 0) + goto cleanup; + ret = 0; + cleanup: + virJSONValueFree(cmd); + virJSONValueFree(reply); + return ret; +} diff --git a/src/qemu/qemu_monitor_json.h b/src/qemu/qemu_monitor_json.h index 01a3ba25f1..d9ad77e873 100644 --- a/src/qemu/qemu_monitor_json.h +++ b/src/qemu/qemu_monitor_json.h @@ -706,3 +706,6 @@ qemuMonitorJSONStartDirtyRateCalc(qemuMonitor *mon, int qemuMonitorJSONQueryDirtyRate(qemuMonitor *mon, qemuMonitorDirtyRateInfo *info); + +int qemuMonitorJSONReloadTLSCerts(qemuMonitorPtr mon, + const char *protocol); -- 2.25.1

2 2

[PATCH] virthread: Make sure virOnce() returns -1 on error
by Michal Privoznik 07 May '21

07 May '21

Since its introduction in v0.9.1~65 the virOnce() was expected to follow the usual retval logic (0 for success, a negative number for failure). However, that was never the case. On the other hand, looking into glibc and musl the pthread_once() never returns anything other than zero (uclibc-ng seems to not implement pthread_once()), therefore we never really hit any problem. But for code cleanliness (and to match POSIX documentation), let's change to code so that our retval logic is honoured. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/util/virthread.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/src/util/virthread.c b/src/util/virthread.c index 5ddbf7c49a..e89c1a09fb 100644 --- a/src/util/virthread.c +++ b/src/util/virthread.c @@ -39,7 +39,15 @@ int virOnce(virOnceControl *once, virOnceFunc init) { - return pthread_once(&once->once, init); + int ret; + + ret = pthread_once(&once->once, init); + if (ret != 0) { + errno = ret; + return -1; + } + + return 0; } -- 2.26.3

2 1

[PATCH v2] Add page_per_vq flag to the 'driver' element of virtio devices
by Gavi Teitz 07 May '21

07 May '21

https://bugzilla.redhat.com/show_bug.cgi?id=1925363 Add support for setting the page-per-vq flag, which is important for vdpa with vhost-user performance. Signed-off-by: Gavi Teitz <gavi(a)nvidia.com> --- docs/formatdomain.rst | 11 +++++- docs/schemas/domaincommon.rng | 5 +++ src/conf/domain_conf.c | 10 +++++ src/conf/domain_conf.h | 1 + src/qemu/qemu_command.c | 5 +++ src/qemu/qemu_hotplug.c | 1 + .../net-virtio-page-per-vq.x86_64-latest.args | 38 ++++++++++++++++++ tests/qemuxml2argvdata/net-virtio-page-per-vq.xml | 29 ++++++++++++++ tests/qemuxml2argvtest.c | 1 + .../net-virtio-page-per-vq.x86_64-latest.xml | 46 ++++++++++++++++++++++ tests/qemuxml2xmltest.c | 1 + 11 files changed, 147 insertions(+), 1 deletion(-) create mode 100644 tests/qemuxml2argvdata/net-virtio-page-per-vq.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/net-virtio-page-per-vq.xml create mode 100644 tests/qemuxml2xmloutdata/net-virtio-page-per-vq.x86_64-latest.xml diff --git a/docs/formatdomain.rst b/docs/formatdomain.rst index fa5c14f..ce7a537 100644 --- a/docs/formatdomain.rst +++ b/docs/formatdomain.rst @@ -5120,7 +5120,7 @@ Setting NIC driver-specific options <source network='default'/> <target dev='vnet1'/> <model type='virtio'/> - <driver name='vhost' txmode='iothread' ioeventfd='on' event_idx='off' queues='5' rx_queue_size='256' tx_queue_size='256'> + <driver name='vhost' txmode='iothread' ioeventfd='on' event_idx='off' queues='5' rx_queue_size='256' tx_queue_size='256' page_per_vq='on'> <host csum='off' gso='off' tso4='off' tso6='off' ecn='off' ufo='off' mrg_rxbuf='off'/> <guest csum='off' tso4='off' tso6='off' ecn='off' ufo='off'/> </driver> @@ -5215,6 +5215,15 @@ following attributes are available for the ``"virtio"`` NIC driver: only for ``vhostuser`` type. :since:`Since 3.7.0 (QEMU and KVM only)` **In general you should leave this option alone, unless you are very certain you know what you are doing.** +``page_per_vq`` + This optional attribute controls the layout of the notification capabilities + exposed to the guest. When enabled, each virtio queue will have a dedicated + page on the device BAR exposed to the guest. It is recommended to be used when + vDPA is enabled on the hypervisor, as it enables mapping the notification area + to the physical device, which is only supported in page granularity. The + default is determined by QEMU; as off. :since:`Since 7.4.0` + **In general you should leave this option alone, unless you are very certain + you know what you are doing.** virtio options For virtio interfaces, `Virtio-specific options <#elementsVirtio>`__ can also be set. ( :since:`Since 3.5.0` ) diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index a2e5c50..e61ad67 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -3463,6 +3463,11 @@ </attribute> </optional> <optional> + <attribute name="page_per_vq"> + <ref name="virOnOff"/> + </attribute> + </optional> + <optional> <attribute name="txmode"> <choice> <value>iothread</value> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index cb668d3..0e219ed 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -10948,6 +10948,12 @@ virDomainNetDefParseXML(virDomainXMLOption *xmlopt, def->driver.virtio.tx_queue_size = q; } + if ((tmpNode = virXPathNode("./driver", ctxt))) { + if (virXMLPropTristateSwitch(tmpNode, "page_per_vq", VIR_XML_PROP_NONE, + &def->driver.virtio.page_per_vq) < 0) + goto error; + } + if ((tmpNode = virXPathNode("./driver/host", ctxt))) { if (virXMLPropTristateSwitch(tmpNode, "csum", VIR_XML_PROP_NONE, &def->driver.virtio.host.csum) < 0) @@ -25221,6 +25227,10 @@ virDomainVirtioNetDriverFormat(virBuffer *buf, if (def->driver.virtio.tx_queue_size) virBufferAsprintf(buf, " tx_queue_size='%u'", def->driver.virtio.tx_queue_size); + if (def->driver.virtio.page_per_vq) { + virBufferAsprintf(buf, " page_per_vq='%s'", + virTristateSwitchTypeToString(def->driver.virtio.page_per_vq)); + } virDomainVirtioOptionsFormat(buf, def->virtio); } diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index 85c318d..7aab565 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -1027,6 +1027,7 @@ struct _virDomainNetDef { virDomainNetVirtioTxModeType txmode; virTristateSwitch ioeventfd; virTristateSwitch event_idx; + virTristateSwitch page_per_vq; unsigned int queues; /* Multiqueue virtio-net */ unsigned int rx_queue_size; unsigned int tx_queue_size; diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index ca2265c..3545ee9 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -3629,6 +3629,11 @@ qemuBuildNicDevStr(virDomainDef *def, if (net->driver.virtio.tx_queue_size) virBufferAsprintf(&buf, ",tx_queue_size=%u", net->driver.virtio.tx_queue_size); + if (net->driver.virtio.page_per_vq) { + virBufferAsprintf(&buf, ",page-per-vq=%s", + virTristateSwitchTypeToString(net->driver.virtio.page_per_vq)); + } + if (net->mtu) virBufferAsprintf(&buf, ",host_mtu=%u", net->mtu); diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 444d89d..a244709 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -3578,6 +3578,7 @@ qemuDomainChangeNet(virQEMUDriver *driver, olddev->driver.virtio.queues != newdev->driver.virtio.queues || olddev->driver.virtio.rx_queue_size != newdev->driver.virtio.rx_queue_size || olddev->driver.virtio.tx_queue_size != newdev->driver.virtio.tx_queue_size || + olddev->driver.virtio.page_per_vq != newdev->driver.virtio.page_per_vq || olddev->driver.virtio.host.csum != newdev->driver.virtio.host.csum || olddev->driver.virtio.host.gso != newdev->driver.virtio.host.gso || olddev->driver.virtio.host.tso4 != newdev->driver.virtio.host.tso4 || diff --git a/tests/qemuxml2argvdata/net-virtio-page-per-vq.x86_64-latest.args b/tests/qemuxml2argvdata/net-virtio-page-per-vq.x86_64-latest.args new file mode 100644 index 0000000..5a52595 --- /dev/null +++ b/tests/qemuxml2argvdata/net-virtio-page-per-vq.x86_64-latest.args @@ -0,0 +1,38 @@ +LC_ALL=C \ +PATH=/bin \ +HOME=/tmp/lib/domain--1-QEMUGuest1 \ +USER=test \ +LOGNAME=test \ +XDG_DATA_HOME=/tmp/lib/domain--1-QEMUGuest1/.local/share \ +XDG_CACHE_HOME=/tmp/lib/domain--1-QEMUGuest1/.cache \ +XDG_CONFIG_HOME=/tmp/lib/domain--1-QEMUGuest1/.config \ +/usr/bin/qemu-system-i386 \ +-name guest=QEMUGuest1,debug-threads=on \ +-S \ +-object '{"qom-type":"secret","id":"masterKey0","format":"raw","file":"/tmp/lib/domain--1-QEMUGuest1/master-key.aes"}' \ +-machine pc,accel=tcg,usb=off,dump-guest-core=off,memory-backend=pc.ram \ +-cpu qemu64 \ +-m 214 \ +-object '{"qom-type":"memory-backend-ram","id":"pc.ram","size":224395264}' \ +-overcommit mem-lock=off \ +-smp 1,sockets=1,cores=1,threads=1 \ +-uuid c7a5fdbd-edaf-9455-926a-d65c16db1809 \ +-display none \ +-no-user-config \ +-nodefaults \ +-chardev socket,id=charmonitor,fd=1729,server=on,wait=off \ +-mon chardev=charmonitor,id=monitor,mode=control \ +-rtc base=utc \ +-no-shutdown \ +-no-acpi \ +-boot strict=on \ +-device piix3-usb-uhci,id=usb,bus=pci.0,addr=0x1.0x2 \ +-blockdev '{"driver":"host_device","filename":"/dev/HostVG/QEMUGuest1","node-name":"libvirt-1-storage","auto-read-only":true,"discard":"unmap"}' \ +-blockdev '{"node-name":"libvirt-1-format","read-only":false,"driver":"raw","file":"libvirt-1-storage"}' \ +-device ide-hd,bus=ide.0,unit=0,drive=libvirt-1-format,id=ide0-0-0,bootindex=1 \ +-netdev user,id=hostnet0 \ +-device virtio-net-pci,page-per-vq=on,netdev=hostnet0,id=net0,mac=00:11:22:33:44:55,bus=pci.0,addr=0x2 \ +-audiodev id=audio1,driver=none \ +-device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x3 \ +-sandbox on,obsolete=deny,elevateprivileges=deny,spawn=deny,resourcecontrol=deny \ +-msg timestamp=on diff --git a/tests/qemuxml2argvdata/net-virtio-page-per-vq.xml b/tests/qemuxml2argvdata/net-virtio-page-per-vq.xml new file mode 100644 index 0000000..e22ecd6 --- /dev/null +++ b/tests/qemuxml2argvdata/net-virtio-page-per-vq.xml @@ -0,0 +1,29 @@ +<domain type='qemu'> + <name>QEMUGuest1</name> + <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid> + <memory unit='KiB'>219100</memory> + <currentMemory unit='KiB'>219100</currentMemory> + <vcpu placement='static'>1</vcpu> + <os> + <type arch='i686' machine='pc'>hvm</type> + <boot dev='hd'/> + </os> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <emulator>/usr/bin/qemu-system-i386</emulator> + <disk type='block' device='disk'> + <source dev='/dev/HostVG/QEMUGuest1'/> + <target dev='hda' bus='ide'/> + </disk> + <controller type='usb' index='0'/> + <interface type='user'> + <mac address='00:11:22:33:44:55'/> + <model type='virtio'/> + <driver page_per_vq='on'/> + </interface> + <memballoon model='virtio'/> + </devices> +</domain> diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index a9dafe2..3dc866a 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -1634,6 +1634,7 @@ mymain(void) QEMU_CAPS_VIRTIO_NET_TX_QUEUE_SIZE); DO_TEST_PARSE_ERROR("net-virtio-rxqueuesize-invalid-size", QEMU_CAPS_VIRTIO_NET_RX_QUEUE_SIZE); + DO_TEST_CAPS_LATEST("net-virtio-page-per-vq"); DO_TEST("net-virtio-teaming", QEMU_CAPS_VIRTIO_NET_FAILOVER, QEMU_CAPS_DEVICE_VFIO_PCI); diff --git a/tests/qemuxml2xmloutdata/net-virtio-page-per-vq.x86_64-latest.xml b/tests/qemuxml2xmloutdata/net-virtio-page-per-vq.x86_64-latest.xml new file mode 100644 index 0000000..34f7ee5 --- /dev/null +++ b/tests/qemuxml2xmloutdata/net-virtio-page-per-vq.x86_64-latest.xml @@ -0,0 +1,46 @@ +<domain type='qemu'> + <name>QEMUGuest1</name> + <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid> + <memory unit='KiB'>219100</memory> + <currentMemory unit='KiB'>219100</currentMemory> + <vcpu placement='static'>1</vcpu> + <os> + <type arch='i686' machine='pc'>hvm</type> + <boot dev='hd'/> + </os> + <cpu mode='custom' match='exact' check='none'> + <model fallback='forbid'>qemu64</model> + </cpu> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <emulator>/usr/bin/qemu-system-i386</emulator> + <disk type='block' device='disk'> + <driver name='qemu' type='raw'/> + <source dev='/dev/HostVG/QEMUGuest1'/> + <target dev='hda' bus='ide'/> + <address type='drive' controller='0' bus='0' target='0' unit='0'/> + </disk> + <controller type='usb' index='0' model='piix3-uhci'> + <address type='pci' domain='0x0000' bus='0x00' slot='0x01' function='0x2'/> + </controller> + <controller type='pci' index='0' model='pci-root'/> + <controller type='ide' index='0'> + <address type='pci' domain='0x0000' bus='0x00' slot='0x01' function='0x1'/> + </controller> + <interface type='user'> + <mac address='00:11:22:33:44:55'/> + <model type='virtio'/> + <driver page_per_vq='on'/> + <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/> + </interface> + <input type='mouse' bus='ps2'/> + <input type='keyboard' bus='ps2'/> + <audio id='1' type='none'/> + <memballoon model='virtio'> + <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> + </memballoon> + </devices> +</domain> diff --git a/tests/qemuxml2xmltest.c b/tests/qemuxml2xmltest.c index 7af6f90..93f29d6 100644 --- a/tests/qemuxml2xmltest.c +++ b/tests/qemuxml2xmltest.c @@ -444,6 +444,7 @@ mymain(void) DO_TEST("net-virtio-rxtxqueuesize", QEMU_CAPS_VIRTIO_NET_RX_QUEUE_SIZE, QEMU_CAPS_VIRTIO_NET_TX_QUEUE_SIZE); + DO_TEST_CAPS_LATEST("net-virtio-page-per-vq"); DO_TEST("net-virtio-teaming", QEMU_CAPS_VIRTIO_NET_FAILOVER, QEMU_CAPS_DEVICE_VFIO_PCI); -- 1.8.3.1

4 4

[PATCH 00/17] conf: Fix and prevent uninitialized memory use with new virXMLProp* helpers
by Peter Krempa 06 May '21

06 May '21

Compilers aren't able to see that the value passed via a pointer from the new virXMLProp helpers may be uninitialized in certain cases. Fix 3 such cases, prepare the code and then ensure that the new virXMLProp* helpers always initialize the memory. CI pipeline (once it finishes) can be viewed at: https://gitlab.com/pipo.sk/libvirt/-/pipelines/298562552 Peter Krempa (17): util: xml: Extract implementation of xml property -> enum parsing to a common helper virXMLPropULongLong: Always initialize @result virDomainVcpuParse: Assign default vcpus count based on return value of virXMLPropUInt virDomainDiskDefDriverParseXML: Fix usage of virXMLPropUInt virXMLPropUInt: Always initialize @result conf: Define autoptr func for virDomainIOThreadIDDef virDomainIOThreadIDDefParseXML: Refactor cleanup virXMLPropInt: Always initialize '@result' virDomainBackupDiskDefParseXML: Fill default backup state after parsing it virXMLPropTristateBool: Always initialize '@result' conf: domain: Don't initialize virTristateBool local variables used for virXMLPropTristateBool virXMLPropTristateSwitch: Always initialize '@result' virDomainAudioCommonParse: Fix parsing of 'format' virDomainVideoDefParseXML: Fix parsing of 'backend' util: xml: Introduce virXMLPropEnumDefault conf: domain: Convert virXMLPropEnum to virXMLPropEnumDefault where we set defaults virXMLPropEnum: Always initialize '@result' src/conf/backup_conf.c | 5 +- src/conf/domain_conf.c | 176 ++++++++++++++++++++------------------- src/conf/domain_conf.h | 1 + src/libvirt_private.syms | 1 + src/util/virxml.c | 156 +++++++++++++++++++--------------- src/util/virxml.h | 14 +++- 6 files changed, 198 insertions(+), 155 deletions(-) -- 2.30.2

2 20

[libvirt PATCH 00/10] Refactor more XML parsing boilerplate code, part VIII
by Tim Wiederhake 06 May '21

06 May '21

For background, see https://listman.redhat.com/archives/libvir-list/2021-April/msg00668.html Note that patch #1 depends on https://listman.redhat.com/archives/libvir-list/2021-April/msg01260.html from part VII. Tim Wiederhake (10): virDomainRedirFilterUSBDevDefParseXML: Use g_auto* virDomainPerfEventDefParseXML: Use virXMLProp* virDomainMemoryDefParseXML: Use virXMLProp* virDomainVcpuPinDefParseXML: Use virXMLProp* virDomainIOThreadPinDefParseXML: Use virXMLProp* virDomainSchedulerParseCommonAttrs: Use virXMLProp* virDomainDef: Change type of placement_mode to virDomainCpuPlacementMode virDomainVcpuParse: Use virXMLProp* virDomainCachetuneDefParseCache: Use virXMLProp* virDomainResctrlMonDefParse: Use virXMLProp* src/conf/domain_conf.c | 300 +++++++++-------------------------------- src/conf/domain_conf.h | 2 +- 2 files changed, 63 insertions(+), 239 deletions(-) -- 2.26.3

2 11

[libvirt PATCH 00/10] Refactor more XML parsing boilerplate code, part VII
by Tim Wiederhake 06 May '21

06 May '21

For background, see https://listman.redhat.com/archives/libvir-list/2021-April/msg00668.html Tim Wiederhake (10): virDomainAudioIOCommon: Change type of format to virDomainAudioFormat virDomainAudioCommonParse: Use virXMLProp* virDomainMemballoonDef: Change type of model to virDomainMemballoonModel virDomainMemballoonDefParseXML: Use virXMLProp* virDomainShmemDef: Change type of model to virDomainShmemModel virDomainShmemDef: Change type of role to virDomainShmemRole virDomainShmemDefParseXML: Use virXMLProp* conf: domain: Register autoptr cleanup function for virDomainDeviceDef virDomainShmemDef: Use g_auto* virDomainRedirFilterUSBDevDefParseXML: Use virXMLProp* src/conf/domain_conf.c | 268 ++++++++++----------------------- src/conf/domain_conf.h | 9 +- src/libxl/libxl_conf.c | 8 +- src/qemu/qemu_command.c | 4 +- src/qemu/qemu_domain_address.c | 2 +- src/qemu/qemu_hotplug.c | 4 +- src/qemu/qemu_monitor.c | 2 +- src/qemu/qemu_validate.c | 5 + src/security/virt-aa-helper.c | 4 + 9 files changed, 100 insertions(+), 206 deletions(-) -- 2.26.3

3 22

[libvirt PATCH v3 0/4] Refactor more XML parsing boilerplate code, part VI
by Tim Wiederhake 06 May '21

06 May '21

For background, see https://listman.redhat.com/archives/libvir-list/2021-April/msg00668.html V1: https://listman.redhat.com/archives/libvir-list/2021-April/msg01184.html V2: https://listman.redhat.com/archives/libvir-list/2021-April/msg01238.html Changes since V2: * Rebased * Prepended patches to fix issues in the function * Removed superfluous "VIR_XML_PROP_NONE" Tim Wiederhake (4): virDomainFSDefParseXML: Remove unused variables virDomainGraphicsDefParseXMLSpice: Fix variable type virDomainGraphicsDefParseXMLSpice: Make name and mode mandatory virDomainGraphicsDefParseXMLSpice: Use virXMLProp* src/conf/domain_conf.c | 195 +++++++++++++++++------------------------ 1 file changed, 80 insertions(+), 115 deletions(-) -- 2.26.3

3 6

[PATCH RFC v5 00/12] Add riscv kvm accel support
by Yifei Jiang 06 May '21

06 May '21

This series adds both riscv32 and riscv64 kvm support, and implements migration based on riscv. It is based on temporarily unaccepted kvm: https://github.com/kvm-riscv/linux (lastest version v17). This series depends on above pending changes which haven't yet been accepted, so this QEMU patch series is treated as RFC patches until that dependency has been dealt with. Several steps to use this: 1. Build emulation $ ./configure --target-list=riscv64-softmmu $ make -j$(nproc) 2. Build kernel https://github.com/kvm-riscv/linux 3. Build QEMU VM Cross built in riscv toolchain. $ PKG_CONFIG_LIBDIR=<toolchain pkgconfig path> $ export PKG_CONFIG_SYSROOT_DIR=<toolchain sysroot path> $ ./configure --target-list=riscv64-softmmu --enable-kvm \ --cross-prefix=riscv64-linux-gnu- --disable-libiscsi --disable-glusterfs \ --disable-libusb --disable-usb-redir --audio-drv-list= --disable-opengl \ --disable-libxml2 $ make -j$(nproc) 4. Start emulation $ ./qemu-system-riscv64 -M virt -m 4096M -cpu rv64,x-h=true -nographic \ -name guest=riscv-hyp,debug-threads=on \ -smp 4 \ -bios ./fw_jump.bin \ -kernel ./Image \ -drive file=./hyp.img,format=raw,id=hd0 \ -device virtio-blk-device,drive=hd0 \ -append "root=/dev/vda rw console=ttyS0 earlycon=sbi" 5. Start kvm-acceled QEMU VM in emulation $ ./qemu-system-riscv64 -M virt,accel=kvm -m 1024M -cpu host -nographic \ -name guest=riscv-guset \ -smp 2 \ -bios none \ -kernel ./Image \ -drive file=./guest.img,format=raw,id=hd0 \ -device virtio-blk-device,drive=hd0 \ -append "root=/dev/vda rw console=ttyS0 earlycon=sbi" Changes since RFC v4 - Rebase on QEMU v6.0.0-rc2 and kvm-riscv linux v17. - Remove time scaling support as software solution is incomplete. Because it will cause unacceptable performance degradation. and We will post a better solution. - Revise according to Alistair's review comments. - Remove compile time XLEN checks in kvm_riscv_reg_id - Surround TYPE_RISCV_CPU_HOST definition by CONFIG_KVM and share it between RV32 and RV64. - Add kvm-stub.c for reduce unnecessary compilation checks. - Add riscv_setup_direct_kernel() to direct boot kernel for KVM. Changes since RFC v3 - Rebase on QEMU v5.2.0-rc2 and kvm-riscv linux v15. - Add time scaling support(New patches 13, 14 and 15). - Fix the bug that guest vm can't reboot. Changes since RFC v2 - Fix checkpatch error at target/riscv/sbi_ecall_interface.h. - Add riscv migration support. Changes since RFC v1 - Add separate SBI ecall interface header. - Add riscv32 kvm accel support. Yifei Jiang (12): linux-header: Update linux/kvm.h target/riscv: Add target/riscv/kvm.c to place the public kvm interface target/riscv: Implement function kvm_arch_init_vcpu target/riscv: Implement kvm_arch_get_registers target/riscv: Implement kvm_arch_put_registers target/riscv: Support start kernel directly by KVM hw/riscv: PLIC update external interrupt by KVM when kvm enabled target/riscv: Handle KVM_EXIT_RISCV_SBI exit target/riscv: Add host cpu type target/riscv: Add kvm_riscv_get/put_regs_timer target/riscv: Implement virtual time adjusting with vm state changing target/riscv: Support virtual time context synchronization hw/intc/sifive_plic.c | 29 +- hw/riscv/boot.c | 11 + hw/riscv/virt.c | 7 + include/hw/riscv/boot.h | 1 + linux-headers/linux/kvm.h | 97 +++++ meson.build | 2 + target/riscv/cpu.c | 17 + target/riscv/cpu.h | 10 + target/riscv/kvm-stub.c | 30 ++ target/riscv/kvm.c | 605 +++++++++++++++++++++++++++++ target/riscv/kvm_riscv.h | 25 ++ target/riscv/machine.c | 14 + target/riscv/meson.build | 1 + target/riscv/sbi_ecall_interface.h | 72 ++++ 14 files changed, 912 insertions(+), 9 deletions(-) create mode 100644 target/riscv/kvm-stub.c create mode 100644 target/riscv/kvm.c create mode 100644 target/riscv/kvm_riscv.h create mode 100644 target/riscv/sbi_ecall_interface.h -- 2.19.1

4 20

[PATCH 0/2] Couple of XML parsing fixes
by Michal Privoznik 05 May '21

05 May '21

Pipeline: https://gitlab.com/MichalPrivoznik/libvirt/-/pipelines/297747756 Michal Prívozník (2): virDomainChrSourceDefParseXML: Fix attribute names for nmdm type virDomainVideoDriverDefParseXML: Allow zero value for @vgaconf src/conf/domain_conf.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) -- 2.26.3

2 4

[libvirt PATCH] gitlab: enable debug output for test suite
by Daniel P. Berrangé 05 May '21

05 May '21

When CI jobs fail on the test suite, we don't see much helpful information by default: stderr: TEST: bhyvexml2argvtest ......!!.............!......!........... 40 ........................!...... 71 FAIL Some tests failed. Run them using: VIR_TEST_DEBUG=1 VIR_TEST_RANGE=7-8,22,29,65 /tmp/cirrus-ci-build/build/meson-private/dist-build/tests/bhyvexml2argvtest Following the instructions to re-run the test with VIR_TEST_DEBUG=1 is great if running locally, but quite unfriendly for CI. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- Compare before: https://gitlab.com/libvirt/libvirt/-/jobs/1237988893 and after: https://gitlab.com/berrange/libvirt/-/jobs/1238324746 .gitlab-ci.yml | 2 ++ ci/cirrus/build.yml | 2 ++ 2 files changed, 4 insertions(+) diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml index 35ec626113..89f618e678 100644 --- a/.gitlab-ci.yml +++ b/.gitlab-ci.yml @@ -11,6 +11,8 @@ stages: export CCACHE_DIR="$CCACHE_BASEDIR/ccache" export CCACHE_MAXSIZE="500M" export PATH="$CCACHE_WRAPPERSDIR:$PATH" + export VIR_TEST_VERBOSE="1" + export VIR_TEST_DEBUG="1" # Common templates diff --git a/ci/cirrus/build.yml b/ci/cirrus/build.yml index 2d3c46a77c..f7ef5c2d13 100644 --- a/ci/cirrus/build.yml +++ b/ci/cirrus/build.yml @@ -9,6 +9,8 @@ env: PKG_CONFIG_PATH: "@PKG_CONFIG_PATH@" PYTHON: "@PYTHON@" MAKE: "@MAKE@" + VIR_TEST_VERBOSE: "1" + VIR_TEST_DEBUG: "1" build_task: install_script: -- 2.31.1

2 1

[PATCH] virDomainVideoDriverDefParseXML: Fix use of uninitialized variable
by Peter Krempa 05 May '21

05 May '21

Commit fc7e1b2f03cfe2a5f3fce737390548505382d8ed which refactored the video driver parse helper introduced a use of uninitialized variable, which caused test failure at least when compiled with clang. Pass 'def->vgaconf' directly to virXMLPropEnum. 'def' needs to be converted to use g_autofree to handle error scenarios. Signed-off-by: Peter Krempa <pkrempa(a)redhat.com> --- Pushed under the build-breaker fix rule. src/conf/domain_conf.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 64f525fc08..d5d8bd0a01 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -14313,8 +14313,7 @@ static virDomainVideoDriverDef * virDomainVideoDriverDefParseXML(xmlNodePtr node, xmlXPathContextPtr ctxt) { - virDomainVideoDriverDef *def; - unsigned int val; + g_autofree virDomainVideoDriverDef *def = NULL; xmlNodePtr driver = NULL; VIR_XPATH_NODE_AUTORESTORE(ctxt) @@ -14323,15 +14322,14 @@ virDomainVideoDriverDefParseXML(xmlNodePtr node, if (!(driver = virXPathNode("./driver", ctxt))) return NULL; + def = g_new0(virDomainVideoDriverDef, 1); + if (virXMLPropEnum(driver, "vgaconf", virDomainVideoVGAConfTypeFromString, - VIR_XML_PROP_NONZERO, &val) < 0) + VIR_XML_PROP_NONZERO, &def->vgaconf) < 0) return NULL; - def = g_new0(virDomainVideoDriverDef, 1); - def->vgaconf = val; - - return def; + return g_steal_pointer(&def); } static virDomainVideoDef * -- 2.30.2

1 0

[libvirt PATCH 0/8] ci: Refresh contents, update list of targets
by Andrea Bolognani 05 May '21

05 May '21

This brings our pipeline in sync with the current status of the libvirt-ci repository. Test pipeline: https://gitlab.com/abologna/libvirt/-/pipelines/296773363 Andrea Bolognani (8): ci: Remove Fedora Rawhide dwarf4 hack ci: Rename CentOS Stream 8 and openSUSE Leap 15.2 containers ci: Refresh contents ci: Add openSUSE Tumbleweed ci: Add FreeBSD 13 ci: Add Fedora 34 ci: Drop Fedora 32 ci: Drop CentOS 7 .gitlab-ci.yml | 88 +++++------ ci/cirrus/freebsd-12.vars | 2 +- .../{freebsd-12.vars =3D> freebsd-13.vars} | 4 +- ci/cirrus/freebsd-current.vars | 2 +- ci/cirrus/macos-11.vars | 2 +- ci/containers/centos-7.Dockerfile | 138 ------------------ ci/containers/centos-8.Dockerfile | 2 +- ....Dockerfile =3D> centos-stream-8.Dockerfile} | 10 +- .../debian-10-cross-aarch64.Dockerfile | 4 +- .../debian-10-cross-armv6l.Dockerfile | 4 +- .../debian-10-cross-armv7l.Dockerfile | 4 +- ci/containers/debian-10-cross-i686.Dockerfile | 4 +- ci/containers/debian-10-cross-mips.Dockerfile | 4 +- .../debian-10-cross-mips64el.Dockerfile | 4 +- .../debian-10-cross-mipsel.Dockerfile | 4 +- .../debian-10-cross-ppc64le.Dockerfile | 4 +- .../debian-10-cross-s390x.Dockerfile | 4 +- ci/containers/debian-10.Dockerfile | 4 +- .../debian-sid-cross-aarch64.Dockerfile | 4 +- .../debian-sid-cross-armv6l.Dockerfile | 4 +- .../debian-sid-cross-armv7l.Dockerfile | 4 +- .../debian-sid-cross-i686.Dockerfile | 4 +- .../debian-sid-cross-mips64el.Dockerfile | 4 +- .../debian-sid-cross-mipsel.Dockerfile | 4 +- .../debian-sid-cross-ppc64le.Dockerfile | 4 +- .../debian-sid-cross-s390x.Dockerfile | 4 +- ci/containers/debian-sid.Dockerfile | 4 +- ci/containers/fedora-33.Dockerfile | 4 +- ...ora-32.Dockerfile =3D> fedora-34.Dockerfile} | 8 +- .../fedora-rawhide-cross-mingw32.Dockerfile | 4 +- .../fedora-rawhide-cross-mingw64.Dockerfile | 4 +- ci/containers/fedora-rawhide.Dockerfile | 4 +- ...ockerfile =3D> opensuse-leap-152.Dockerfile} | 6 +- ...kerfile =3D> opensuse-tumbleweed.Dockerfile} | 15 +- ci/containers/ubuntu-1804.Dockerfile | 4 +- ci/containers/ubuntu-2004.Dockerfile | 4 +- 36 files changed, 117 insertions(+), 260 deletions(-) copy ci/cirrus/{freebsd-12.vars =3D> freebsd-13.vars} (79%) delete mode 100644 ci/containers/centos-7.Dockerfile rename ci/containers/{centos-stream.Dockerfile =3D> centos-stream-8.Dockerfi= le} (89%) rename ci/containers/{fedora-32.Dockerfile =3D> fedora-34.Dockerfile} (93%) copy ci/containers/{opensuse-152.Dockerfile =3D> opensuse-leap-152.Dockerfil= e} (94%) rename ci/containers/{opensuse-152.Dockerfile =3D> opensuse-tumbleweed.Docke= rfile} (88%) --=20 2.26.3

2 9

[PATCH v1 0/4] libxl: fix affinity assignment
by Olaf Hering 05 May '21

05 May '21

Olaf Hering (4): libxl: remove obsolete VIR_LIBXL_EVENT_CONST libxl: remove obsolete check for result of xc_get_max_cpus libxl: remove obsolete libxlDomainSetVcpuAffinities function libxl: set vcpu affinity during domain creation src/libxl/libxl_conf.c | 53 ++++++++++++++++++++++++++++++++++++++++ src/libxl/libxl_domain.c | 48 +----------------------------------- src/libxl/libxl_domain.h | 18 +------------- src/libxl/libxl_driver.c | 6 +---- 4 files changed, 56 insertions(+), 69 deletions(-)

2 10

[PATCH v2] docs: cputune is also supported by the xen driver
by Olaf Hering 05 May '21

05 May '21

Since commit 68c5b6fb2b5fdabce775e9f8fc761a400e16a9d3 libxl also handles a domain/cputune/vcpupin element in domU.xml. Signed-off-by: Olaf Hering <olaf(a)aepfle.de> --- docs/formatdomain.rst | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/formatdomain.rst b/docs/formatdomain.rst index 282176c4f4..fa5c14febc 100644 --- a/docs/formatdomain.rst +++ b/docs/formatdomain.rst @@ -744,8 +744,8 @@ CPU Tuning of element ``vcpu`` is not specified, the vCPU is pinned to all the physical CPUs by default. It contains two required attributes, the attribute ``vcpu`` specifies vCPU id, and the attribute ``cpuset`` is same as attribute - ``cpuset`` of element ``vcpu``. (NB: Only qemu driver support) :since:`Since - 0.9.0` + ``cpuset`` of element ``vcpu``. + :since:`QEMU driver support since 0.9.0, Xen driver support since 0.9.1` ``emulatorpin`` The optional ``emulatorpin`` element specifies which of host physical CPUs the "emulator", a subset of a domain not including vCPU or iothreads will be

2 1

[PATCH v1] docs: cputune is also supported by the xen driver
by Olaf Hering 04 May '21

04 May '21

Since commit 68c5b6fb2b5fdabce775e9f8fc761a400e16a9d3 libxl also handles a domain/cputune/vcpupin element in domU.xml. Signed-off-by: Olaf Hering <olaf(a)aepfle.de> --- docs/formatdomain.rst | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/formatdomain.rst b/docs/formatdomain.rst index 282176c4f4..0328fe488b 100644 --- a/docs/formatdomain.rst +++ b/docs/formatdomain.rst @@ -744,8 +744,8 @@ CPU Tuning of element ``vcpu`` is not specified, the vCPU is pinned to all the physical CPUs by default. It contains two required attributes, the attribute ``vcpu`` specifies vCPU id, and the attribute ``cpuset`` is same as attribute - ``cpuset`` of element ``vcpu``. (NB: Only qemu driver support) :since:`Since - 0.9.0` + ``cpuset`` of element ``vcpu``. (NB: Only qemu and xen driver support) + :since:`Since 0.9.0` ``emulatorpin`` The optional ``emulatorpin`` element specifies which of host physical CPUs the "emulator", a subset of a domain not including vCPU or iothreads will be

2 1

[PATCH] virtio-blk: drop deprecated scsi=on|off property
by Stefan Hajnoczi 04 May '21

04 May '21

The scsi=on|off property was deprecated in QEMU 5.0 and can be removed completely at this point. Drop the scsi=on|off option. It was only available on Legacy virtio-blk devices. Linux v5.6 already dropped support for it. Remove the hw_compat_2_4[] property assignment since scsi=on|off no longer exists. Old guests with Legacy virtio-blk devices no longer see the SCSI host features bit. Live migrating old guests from an old QEMU with the SCSI feature bit enabled will fail with "Features 0x... unsupported. Allowed features: 0x...". We've followed the QEMU deprecation policy so users have been warned... I have tested that libvirt still works when the property is absent. It no longer adds scsi=on|off to the command-line. Cc: Markus Armbruster <armbru(a)redhat.com> Cc: Christoph Hellwig <hch(a)lst.de> Cc: Peter Krempa <pkrempa(a)redhat.com> Cc: Dr. David Alan Gilbert <dgilbert(a)redhat.com> Signed-off-by: Stefan Hajnoczi <stefanha(a)redhat.com> --- docs/specs/tpm.rst | 2 +- docs/system/deprecated.rst | 13 --- docs/pci_expander_bridge.txt | 2 +- hw/block/virtio-blk.c | 192 +---------------------------------- hw/core/machine.c | 2 - 5 files changed, 3 insertions(+), 208 deletions(-) diff --git a/docs/specs/tpm.rst b/docs/specs/tpm.rst index 3be190343a..0ec017ab95 100644 --- a/docs/specs/tpm.rst +++ b/docs/specs/tpm.rst @@ -328,7 +328,7 @@ In case a pSeries machine is emulated, use the following command line: -tpmdev emulator,id=tpm0,chardev=chrtpm \ -device tpm-spapr,tpmdev=tpm0 \ -device spapr-vscsi,id=scsi0,reg=0x00002000 \ - -device virtio-blk-pci,scsi=off,bus=pci.0,addr=0x3,drive=drive-virtio-disk0,id=virtio-disk0 \ + -device virtio-blk-pci,bus=pci.0,addr=0x3,drive=drive-virtio-disk0,id=virtio-disk0 \ -drive file=test.img,format=raw,if=none,id=drive-virtio-disk0 In case an Arm virt machine is emulated, use the following command line: diff --git a/docs/system/deprecated.rst b/docs/system/deprecated.rst index 80cae86252..1abb64b669 100644 --- a/docs/system/deprecated.rst +++ b/docs/system/deprecated.rst @@ -248,19 +248,6 @@ machines have been renamed ``raspi2b`` and ``raspi3b``. Device options -------------- -Emulated device options -''''''''''''''''''''''' - -``-device virtio-blk,scsi=on|off`` (since 5.0.0) -^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - -The virtio-blk SCSI passthrough feature is a legacy VIRTIO feature. VIRTIO 1.0 -and later do not support it because the virtio-scsi device was introduced for -full SCSI support. Use virtio-scsi instead when SCSI passthrough is required. - -Note this also applies to ``-device virtio-blk-pci,scsi=on|off``, which is an -alias. - Block device options '''''''''''''''''''' diff --git a/docs/pci_expander_bridge.txt b/docs/pci_expander_bridge.txt index 36750273bb..540191f5e0 100644 --- a/docs/pci_expander_bridge.txt +++ b/docs/pci_expander_bridge.txt @@ -25,7 +25,7 @@ A detailed command line would be: -object memory-backend-ram,size=1024M,policy=bind,host-nodes=1,id=ram-node1 -numa node,nodeid=1,cpus=1,memdev=ram-node1 -device pxb,id=bridge1,bus=pci.0,numa_node=1,bus_nr=4 -netdev user,id=nd -device e1000,bus=bridge1,addr=0x4,netdev=nd -device pxb,id=bridge2,bus=pci.0,numa_node=0,bus_nr=8 -device e1000,bus=bridge2,addr=0x3 --device pxb,id=bridge3,bus=pci.0,bus_nr=40 -drive if=none,id=drive0,file=[img] -device virtio-blk-pci,drive=drive0,scsi=off,bus=bridge3,addr=1 +-device pxb,id=bridge3,bus=pci.0,bus_nr=40 -drive if=none,id=drive0,file=[img] -device virtio-blk-pci,drive=drive0,bus=bridge3,addr=1 Here you have: - 2 NUMA nodes for the guest, 0 and 1. (both mapped to the same NUMA node in host, but you can and should put it in different host NUMA nodes) diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c index d28979efb8..5023e046fc 100644 --- a/hw/block/virtio-blk.c +++ b/hw/block/virtio-blk.c @@ -25,10 +25,6 @@ #include "sysemu/runstate.h" #include "hw/virtio/virtio-blk.h" #include "dataplane/virtio-blk.h" -#include "scsi/constants.h" -#ifdef __linux__ -# include <scsi/sg.h> -#endif #include "hw/virtio/virtio-bus.h" #include "migration/qemu-file-types.h" #include "hw/virtio/virtio-access.h" @@ -200,59 +196,6 @@ out: aio_context_release(blk_get_aio_context(s->conf.conf.blk)); } -#ifdef __linux__ - -typedef struct { - VirtIOBlockReq *req; - struct sg_io_hdr hdr; -} VirtIOBlockIoctlReq; - -static void virtio_blk_ioctl_complete(void *opaque, int status) -{ - VirtIOBlockIoctlReq *ioctl_req = opaque; - VirtIOBlockReq *req = ioctl_req->req; - VirtIOBlock *s = req->dev; - VirtIODevice *vdev = VIRTIO_DEVICE(s); - struct virtio_scsi_inhdr *scsi; - struct sg_io_hdr *hdr; - - scsi = (void *)req->elem.in_sg[req->elem.in_num - 2].iov_base; - - if (status) { - status = VIRTIO_BLK_S_UNSUPP; - virtio_stl_p(vdev, &scsi->errors, 255); - goto out; - } - - hdr = &ioctl_req->hdr; - /* - * From SCSI-Generic-HOWTO: "Some lower level drivers (e.g. ide-scsi) - * clear the masked_status field [hence status gets cleared too, see - * block/scsi_ioctl.c] even when a CHECK_CONDITION or COMMAND_TERMINATED - * status has occurred. However they do set DRIVER_SENSE in driver_status - * field. Also a (sb_len_wr > 0) indicates there is a sense buffer. - */ - if (hdr->status == 0 && hdr->sb_len_wr > 0) { - hdr->status = CHECK_CONDITION; - } - - virtio_stl_p(vdev, &scsi->errors, - hdr->status | (hdr->msg_status << 8) | - (hdr->host_status << 16) | (hdr->driver_status << 24)); - virtio_stl_p(vdev, &scsi->residual, hdr->resid); - virtio_stl_p(vdev, &scsi->sense_len, hdr->sb_len_wr); - virtio_stl_p(vdev, &scsi->data_len, hdr->dxfer_len); - -out: - aio_context_acquire(blk_get_aio_context(s->conf.conf.blk)); - virtio_blk_req_complete(req, status); - virtio_blk_free_request(req); - aio_context_release(blk_get_aio_context(s->conf.conf.blk)); - g_free(ioctl_req); -} - -#endif - static VirtIOBlockReq *virtio_blk_get_request(VirtIOBlock *s, VirtQueue *vq) { VirtIOBlockReq *req = virtqueue_pop(vq, sizeof(VirtIOBlockReq)); @@ -263,126 +206,6 @@ static VirtIOBlockReq *virtio_blk_get_request(VirtIOBlock *s, VirtQueue *vq) return req; } -static int virtio_blk_handle_scsi_req(VirtIOBlockReq *req) -{ - int status = VIRTIO_BLK_S_OK; - struct virtio_scsi_inhdr *scsi = NULL; - VirtIOBlock *blk = req->dev; - VirtIODevice *vdev = VIRTIO_DEVICE(blk); - VirtQueueElement *elem = &req->elem; - -#ifdef __linux__ - int i; - VirtIOBlockIoctlReq *ioctl_req; - BlockAIOCB *acb; -#endif - - /* - * We require at least one output segment each for the virtio_blk_outhdr - * and the SCSI command block. - * - * We also at least require the virtio_blk_inhdr, the virtio_scsi_inhdr - * and the sense buffer pointer in the input segments. - */ - if (elem->out_num < 2 || elem->in_num < 3) { - status = VIRTIO_BLK_S_IOERR; - goto fail; - } - - /* - * The scsi inhdr is placed in the second-to-last input segment, just - * before the regular inhdr. - */ - scsi = (void *)elem->in_sg[elem->in_num - 2].iov_base; - - if (!virtio_has_feature(blk->host_features, VIRTIO_BLK_F_SCSI)) { - status = VIRTIO_BLK_S_UNSUPP; - goto fail; - } - - /* - * No support for bidirection commands yet. - */ - if (elem->out_num > 2 && elem->in_num > 3) { - status = VIRTIO_BLK_S_UNSUPP; - goto fail; - } - -#ifdef __linux__ - ioctl_req = g_new0(VirtIOBlockIoctlReq, 1); - ioctl_req->req = req; - ioctl_req->hdr.interface_id = 'S'; - ioctl_req->hdr.cmd_len = elem->out_sg[1].iov_len; - ioctl_req->hdr.cmdp = elem->out_sg[1].iov_base; - ioctl_req->hdr.dxfer_len = 0; - - if (elem->out_num > 2) { - /* - * If there are more than the minimally required 2 output segments - * there is write payload starting from the third iovec. - */ - ioctl_req->hdr.dxfer_direction = SG_DXFER_TO_DEV; - ioctl_req->hdr.iovec_count = elem->out_num - 2; - - for (i = 0; i < ioctl_req->hdr.iovec_count; i++) { - ioctl_req->hdr.dxfer_len += elem->out_sg[i + 2].iov_len; - } - - ioctl_req->hdr.dxferp = elem->out_sg + 2; - - } else if (elem->in_num > 3) { - /* - * If we have more than 3 input segments the guest wants to actually - * read data. - */ - ioctl_req->hdr.dxfer_direction = SG_DXFER_FROM_DEV; - ioctl_req->hdr.iovec_count = elem->in_num - 3; - for (i = 0; i < ioctl_req->hdr.iovec_count; i++) { - ioctl_req->hdr.dxfer_len += elem->in_sg[i].iov_len; - } - - ioctl_req->hdr.dxferp = elem->in_sg; - } else { - /* - * Some SCSI commands don't actually transfer any data. - */ - ioctl_req->hdr.dxfer_direction = SG_DXFER_NONE; - } - - ioctl_req->hdr.sbp = elem->in_sg[elem->in_num - 3].iov_base; - ioctl_req->hdr.mx_sb_len = elem->in_sg[elem->in_num - 3].iov_len; - - acb = blk_aio_ioctl(blk->blk, SG_IO, &ioctl_req->hdr, - virtio_blk_ioctl_complete, ioctl_req); - if (!acb) { - g_free(ioctl_req); - status = VIRTIO_BLK_S_UNSUPP; - goto fail; - } - return -EINPROGRESS; -#else - abort(); -#endif - -fail: - /* Just put anything nonzero so that the ioctl fails in the guest. */ - if (scsi) { - virtio_stl_p(vdev, &scsi->errors, 255); - } - return status; -} - -static void virtio_blk_handle_scsi(VirtIOBlockReq *req) -{ - int status; - - status = virtio_blk_handle_scsi_req(req); - if (status != -EINPROGRESS) { - virtio_blk_req_complete(req, status); - virtio_blk_free_request(req); - } -} - static inline void submit_requests(BlockBackend *blk, MultiReqBuffer *mrb, int start, int num_reqs, int niov) { @@ -699,9 +522,6 @@ static int virtio_blk_handle_request(VirtIOBlockReq *req, MultiReqBuffer *mrb) case VIRTIO_BLK_T_FLUSH: virtio_blk_handle_flush(req, mrb); break; - case VIRTIO_BLK_T_SCSI_CMD: - virtio_blk_handle_scsi(req); - break; case VIRTIO_BLK_T_GET_ID: { /* @@ -1010,14 +830,8 @@ static uint64_t virtio_blk_get_features(VirtIODevice *vdev, uint64_t features, virtio_add_feature(&features, VIRTIO_BLK_F_GEOMETRY); virtio_add_feature(&features, VIRTIO_BLK_F_TOPOLOGY); virtio_add_feature(&features, VIRTIO_BLK_F_BLK_SIZE); - if (virtio_has_feature(features, VIRTIO_F_VERSION_1)) { - if (virtio_has_feature(s->host_features, VIRTIO_BLK_F_SCSI)) { - error_setg(errp, "Please set scsi=off for virtio-blk devices in order to use virtio 1.0"); - return 0; - } - } else { + if (!virtio_has_feature(features, VIRTIO_F_VERSION_1)) { virtio_clear_feature(&features, VIRTIO_F_ANY_LAYOUT); - virtio_add_feature(&features, VIRTIO_BLK_F_SCSI); } if (blk_enable_write_cache(s->blk) || @@ -1289,10 +1103,6 @@ static Property virtio_blk_properties[] = { DEFINE_PROP_STRING("serial", VirtIOBlock, conf.serial), DEFINE_PROP_BIT64("config-wce", VirtIOBlock, host_features, VIRTIO_BLK_F_CONFIG_WCE, true), -#ifdef __linux__ - DEFINE_PROP_BIT64("scsi", VirtIOBlock, host_features, - VIRTIO_BLK_F_SCSI, false), -#endif DEFINE_PROP_BIT("request-merging", VirtIOBlock, conf.request_merging, 0, true), DEFINE_PROP_UINT16("num-queues", VirtIOBlock, conf.num_queues, diff --git a/hw/core/machine.c b/hw/core/machine.c index 40def78183..286f18ec6d 100644 --- a/hw/core/machine.c +++ b/hw/core/machine.c @@ -194,8 +194,6 @@ GlobalProperty hw_compat_2_5[] = { const size_t hw_compat_2_5_len = G_N_ELEMENTS(hw_compat_2_5); GlobalProperty hw_compat_2_4[] = { - /* Optional because the 'scsi' property is Linux-only */ - { "virtio-blk-device", "scsi", "true", .optional = true }, { "e1000", "extra_mac_registers", "off" }, { "virtio-pci", "x-disable-pcie", "on" }, { "virtio-pci", "migrate-extra", "off" }, -- 2.30.2

6 10

Re: Let's remove some deprecated stuff
by Eric Blake 04 May '21

04 May '21

On 4/29/21 4:59 AM, Markus Armbruster wrote: > If you're cc'ed, you added a section to docs/system/deprecated.rst that > is old enough to permit removal. This is *not* a demand to remove, it's > a polite request to consider whether the time for removal has come. > Extra points for telling us in a reply. "We should remove, but I can't > do it myself right now" is a valid answer. Let's review the file: > [adjusting cc for this response] > > Eric Blake: > > qemu-img amend to adjust backing file (since 5.1) > ''''''''''''''''''''''''''''''''''''''''''''''''' > > The use of ``qemu-img amend`` to modify the name or format of a qcow2 > backing image is deprecated; this functionality was never fully > documented or tested, and interferes with other amend operations that > need access to the original backing image (such as deciding whether a > v3 zero cluster may be left unallocated when converting to a v2 > image). Rather, any changes to the backing chain should be performed > with ``qemu-img rebase -u`` either before or after the remaining > changes being performed by amend, as appropriate. > > qemu-img backing file without format (since 5.1) > '''''''''''''''''''''''''''''''''''''''''''''''' > > The use of ``qemu-img create``, ``qemu-img rebase``, or ``qemu-img > convert`` to create or modify an image that depends on a backing file > now recommends that an explicit backing format be provided. This is > for safety: if QEMU probes a different format than what you thought, > the data presented to the guest will be corrupt; similarly, presenting > a raw image to a guest allows a potential security exploit if a future > probe sees a non-raw image based on guest writes. > > To avoid the warning message, or even future refusal to create an > unsafe image, you must pass ``-o backing_fmt=`` (or the shorthand > ``-F`` during create) to specify the intended backing format. You may > use ``qemu-img rebase -u`` to retroactively add a backing format to an > existing image. However, be aware that there are already potential > security risks to blindly using ``qemu-img info`` to probe the format > of an untrusted backing image, when deciding what format to add into > an existing image. I'm not sure how widely used these were, but I'm game for writing a patch to drop them. I'm fairly certain libvirt is not using them. > > Kevin Wolf: > > ``nbd-server-add`` and ``nbd-server-remove`` (since 5.2) > '''''''''''''''''''''''''''''''''''''''''''''''''''''''' > > Use the more generic commands ``block-export-add`` and ``block-export-del`` > instead. As part of this deprecation, where ``nbd-server-add`` used a > single ``bitmap``, the new ``block-export-add`` uses a list of ``bitmaps``. Peter, is libvirt good for this one to go? -- Eric Blake, Principal Software Engineer Red Hat, Inc. +1-919-301-3226 Virtualization: qemu.org | libvirt.org

2 1

[libvirt PATCH 0/6] Enable sanitizers
by Tim Wiederhake 04 May '21

04 May '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. Cheers, Tim Tim Wiederhake (6): meson: Allow larger stack frames when instrumenting meson: Allow undefined symbols when sanitizers are enabled tests: virfilemock: realpath: Allow non-null second parameter 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 | 8 +++++++- src/libvirt_openvz.syms | 2 ++ src/security/meson.build | 1 - tests/meson.build | 2 +- tests/virfilemock.c | 12 ++++++------ 7 files changed, 52 insertions(+), 10 deletions(-) -- 2.26.3

3 14

[PATCH] Add page_per_vq flag to the 'driver' element of virtio devices
by Gavi Teitz 04 May '21

04 May '21

https://bugzilla.redhat.com/show_bug.cgi?id=1925363 Add support for setting the page-per-vq flag, which is important for vdpa with vhost-user performance. Signed-off-by: Gavi Teitz <gavi(a)nvidia.com> --- docs/formatdomain.rst | 11 +++++- docs/schemas/domaincommon.rng | 5 +++ src/conf/domain_conf.c | 15 ++++++++ src/conf/domain_conf.h | 1 + src/qemu/qemu_capabilities.c | 2 + src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 5 +++ src/qemu/qemu_hotplug.c | 1 + tests/qemuxml2argvdata/net-virtio-page-per-vq.args | 31 ++++++++++++++++ tests/qemuxml2argvdata/net-virtio-page-per-vq.xml | 29 +++++++++++++++ tests/qemuxml2argvtest.c | 2 + .../qemuxml2xmloutdata/net-virtio-page-per-vq.xml | 43 ++++++++++++++++++++++ tests/qemuxml2xmltest.c | 2 + 13 files changed, 147 insertions(+), 1 deletion(-) create mode 100644 tests/qemuxml2argvdata/net-virtio-page-per-vq.args create mode 100644 tests/qemuxml2argvdata/net-virtio-page-per-vq.xml create mode 100644 tests/qemuxml2xmloutdata/net-virtio-page-per-vq.xml diff --git a/docs/formatdomain.rst b/docs/formatdomain.rst index 282176c..f5a9c70 100644 --- a/docs/formatdomain.rst +++ b/docs/formatdomain.rst @@ -5120,7 +5120,7 @@ Setting NIC driver-specific options <source network='default'/> <target dev='vnet1'/> <model type='virtio'/> - <driver name='vhost' txmode='iothread' ioeventfd='on' event_idx='off' queues='5' rx_queue_size='256' tx_queue_size='256'> + <driver name='vhost' txmode='iothread' ioeventfd='on' event_idx='off' queues='5' rx_queue_size='256' tx_queue_size='256' page_per_vq='on'> <host csum='off' gso='off' tso4='off' tso6='off' ecn='off' ufo='off' mrg_rxbuf='off'/> <guest csum='off' tso4='off' tso6='off' ecn='off' ufo='off'/> </driver> @@ -5215,6 +5215,15 @@ following attributes are available for the ``"virtio"`` NIC driver: only for ``vhostuser`` type. :since:`Since 3.7.0 (QEMU and KVM only)` **In general you should leave this option alone, unless you are very certain you know what you are doing.** +``page_per_vq`` + This optional attribute controls the layout of the notification capabilities + exposed to the guest. When enabled, each virtio queue will have a dedicated + page on the device BAR exposed to the guest. It is recommended to be used when + vDPA is enabled on the hypervisor, as it enables mapping the notification area + to the physical device, which is only supported in page granularity. The + default is determined by QEMU; as off. :since:`Since 2.8.0 (QEMU only)` + **In general you should leave this option alone, unless you are very certain + you know what you are doing.** virtio options For virtio interfaces, `Virtio-specific options <#elementsVirtio>`__ can also be set. ( :since:`Since 3.5.0` ) diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index a2e5c50..e61ad67 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -3463,6 +3463,11 @@ </attribute> </optional> <optional> + <attribute name="page_per_vq"> + <ref name="virOnOff"/> + </attribute> + </optional> + <optional> <attribute name="txmode"> <choice> <value>iothread</value> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 9d98f48..0350fde 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -10446,6 +10446,7 @@ virDomainNetDefParseXML(virDomainXMLOption *xmlopt, g_autofree char *queues = NULL; g_autofree char *rx_queue_size = NULL; g_autofree char *tx_queue_size = NULL; + g_autofree char *page_per_vq = NULL; g_autofree char *filter = NULL; g_autofree char *internal = NULL; g_autofree char *mode = NULL; @@ -10615,6 +10616,7 @@ virDomainNetDefParseXML(virDomainXMLOption *xmlopt, queues = virXMLPropString(cur, "queues"); rx_queue_size = virXMLPropString(cur, "rx_queue_size"); tx_queue_size = virXMLPropString(cur, "tx_queue_size"); + page_per_vq = virXMLPropString(cur, "page_per_vq"); if (virDomainVirtioOptionsParseXML(cur, &def->virtio) < 0) goto error; @@ -11041,6 +11043,15 @@ virDomainNetDefParseXML(virDomainXMLOption *xmlopt, } def->driver.virtio.tx_queue_size = q; } + if (page_per_vq) { + if ((val = virTristateSwitchTypeFromString(page_per_vq)) <= 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("unknown page_per_vq mode '%s'"), + page_per_vq); + goto error; + } + def->driver.virtio.page_per_vq = val; + } if ((tmpNode = virXPathNode("./driver/host", ctxt))) { if (virXMLPropTristateSwitch(tmpNode, "csum", VIR_XML_PROP_NONE, @@ -25453,6 +25464,10 @@ virDomainVirtioNetDriverFormat(virBuffer *buf, if (def->driver.virtio.tx_queue_size) virBufferAsprintf(buf, " tx_queue_size='%u'", def->driver.virtio.tx_queue_size); + if (def->driver.virtio.page_per_vq) { + virBufferAsprintf(buf, " page_per_vq='%s'", + virTristateSwitchTypeToString(def->driver.virtio.page_per_vq)); + } virDomainVirtioOptionsFormat(buf, def->virtio); } diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index 85c318d..7aab565 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -1027,6 +1027,7 @@ struct _virDomainNetDef { virDomainNetVirtioTxModeType txmode; virTristateSwitch ioeventfd; virTristateSwitch event_idx; + virTristateSwitch page_per_vq; unsigned int queues; /* Multiqueue virtio-net */ unsigned int rx_queue_size; unsigned int tx_queue_size; diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 7971a9c..e5646cb 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -629,6 +629,7 @@ VIR_ENUM_IMPL(virQEMUCaps, /* 400 */ "compat-deprecated", "acpi-index", + "virtio-net.page_per_vq", ); @@ -1405,6 +1406,7 @@ static struct virQEMUCapsDevicePropsFlags virQEMUCapsDevicePropsVirtioNet[] = { { "event_idx", QEMU_CAPS_VIRTIO_NET_EVENT_IDX, NULL }, { "rx_queue_size", QEMU_CAPS_VIRTIO_NET_RX_QUEUE_SIZE, NULL }, { "tx_queue_size", QEMU_CAPS_VIRTIO_NET_TX_QUEUE_SIZE, NULL }, + { "page_per_vq", QEMU_CAPS_VIRTIO_NET_PAGE_PER_VQ, NULL }, { "host_mtu", QEMU_CAPS_VIRTIO_NET_HOST_MTU, NULL }, { "disable-legacy", QEMU_CAPS_VIRTIO_PCI_DISABLE_LEGACY, NULL }, { "iommu_platform", QEMU_CAPS_VIRTIO_PCI_IOMMU_PLATFORM, NULL }, diff --git a/src/qemu/qemu_capabilities.h b/src/qemu/qemu_capabilities.h index f54aad5..75feed4 100644 --- a/src/qemu/qemu_capabilities.h +++ b/src/qemu/qemu_capabilities.h @@ -609,6 +609,7 @@ typedef enum { /* virQEMUCapsFlags grouping marker for syntax-check */ /* 400 */ QEMU_CAPS_COMPAT_DEPRECATED, /* -compat deprecated-(input|output) is supported */ QEMU_CAPS_ACPI_INDEX, /* PCI device 'acpi-index' property */ + QEMU_CAPS_VIRTIO_NET_PAGE_PER_VQ, /* virtio-net-*.page_per_vq */ QEMU_CAPS_LAST /* this must always be the last item */ } virQEMUCapsFlags; diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index d7f1c71..7b5834f 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -3630,6 +3630,11 @@ qemuBuildNicDevStr(virDomainDef *def, if (net->driver.virtio.tx_queue_size) virBufferAsprintf(&buf, ",tx_queue_size=%u", net->driver.virtio.tx_queue_size); + if (net->driver.virtio.page_per_vq) { + virBufferAsprintf(&buf, ",page-per-vq=%s", + virTristateSwitchTypeToString(net->driver.virtio.page_per_vq)); + } + if (net->mtu) virBufferAsprintf(&buf, ",host_mtu=%u", net->mtu); diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 4344edc..f79c3d8 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -3578,6 +3578,7 @@ qemuDomainChangeNet(virQEMUDriver *driver, olddev->driver.virtio.queues != newdev->driver.virtio.queues || olddev->driver.virtio.rx_queue_size != newdev->driver.virtio.rx_queue_size || olddev->driver.virtio.tx_queue_size != newdev->driver.virtio.tx_queue_size || + olddev->driver.virtio.page_per_vq != newdev->driver.virtio.page_per_vq || olddev->driver.virtio.host.csum != newdev->driver.virtio.host.csum || olddev->driver.virtio.host.gso != newdev->driver.virtio.host.gso || olddev->driver.virtio.host.tso4 != newdev->driver.virtio.host.tso4 || diff --git a/tests/qemuxml2argvdata/net-virtio-page-per-vq.args b/tests/qemuxml2argvdata/net-virtio-page-per-vq.args new file mode 100644 index 0000000..d7db71f --- /dev/null +++ b/tests/qemuxml2argvdata/net-virtio-page-per-vq.args @@ -0,0 +1,31 @@ +LC_ALL=C \ +PATH=/bin \ +HOME=/tmp/lib/domain--1-QEMUGuest1 \ +USER=test \ +LOGNAME=test \ +XDG_DATA_HOME=/tmp/lib/domain--1-QEMUGuest1/.local/share \ +XDG_CACHE_HOME=/tmp/lib/domain--1-QEMUGuest1/.cache \ +XDG_CONFIG_HOME=/tmp/lib/domain--1-QEMUGuest1/.config \ +QEMU_AUDIO_DRV=none \ +/usr/bin/qemu-system-i386 \ +-name QEMUGuest1 \ +-S \ +-machine pc,accel=tcg,usb=off,dump-guest-core=off \ +-m 214 \ +-realtime mlock=off \ +-smp 1,sockets=1,cores=1,threads=1 \ +-uuid c7a5fdbd-edaf-9455-926a-d65c16db1809 \ +-display none \ +-no-user-config \ +-nodefaults \ +-chardev socket,id=charmonitor,path=/tmp/lib/domain--1-QEMUGuest1/monitor.sock,server=on,wait=off \ +-mon chardev=charmonitor,id=monitor,mode=control \ +-rtc base=utc \ +-no-shutdown \ +-no-acpi \ +-usb \ +-drive file=/dev/HostVG/QEMUGuest1,format=raw,if=none,id=drive-ide0-0-0 \ +-device ide-hd,bus=ide.0,unit=0,drive=drive-ide0-0-0,id=ide0-0-0,bootindex=1 \ +-netdev user,id=hostnet0 \ +-device virtio-net-pci,page-per-vq=on,netdev=hostnet0,id=net0,mac=00:11:22:33:44:55,bus=pci.0,addr=0x3 \ +-device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x4 diff --git a/tests/qemuxml2argvdata/net-virtio-page-per-vq.xml b/tests/qemuxml2argvdata/net-virtio-page-per-vq.xml new file mode 100644 index 0000000..e22ecd6 --- /dev/null +++ b/tests/qemuxml2argvdata/net-virtio-page-per-vq.xml @@ -0,0 +1,29 @@ +<domain type='qemu'> + <name>QEMUGuest1</name> + <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid> + <memory unit='KiB'>219100</memory> + <currentMemory unit='KiB'>219100</currentMemory> + <vcpu placement='static'>1</vcpu> + <os> + <type arch='i686' machine='pc'>hvm</type> + <boot dev='hd'/> + </os> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <emulator>/usr/bin/qemu-system-i386</emulator> + <disk type='block' device='disk'> + <source dev='/dev/HostVG/QEMUGuest1'/> + <target dev='hda' bus='ide'/> + </disk> + <controller type='usb' index='0'/> + <interface type='user'> + <mac address='00:11:22:33:44:55'/> + <model type='virtio'/> + <driver page_per_vq='on'/> + </interface> + <memballoon model='virtio'/> + </devices> +</domain> diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index f0efe98..f6b8f34 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -1654,6 +1654,8 @@ mymain(void) QEMU_CAPS_VIRTIO_NET_TX_QUEUE_SIZE); DO_TEST_PARSE_ERROR("net-virtio-rxqueuesize-invalid-size", QEMU_CAPS_VIRTIO_NET_RX_QUEUE_SIZE); + DO_TEST("net-virtio-page-per-vq", + QEMU_CAPS_VIRTIO_NET_PAGE_PER_VQ); DO_TEST("net-virtio-teaming", QEMU_CAPS_VIRTIO_NET_FAILOVER, QEMU_CAPS_DEVICE_VFIO_PCI); diff --git a/tests/qemuxml2xmloutdata/net-virtio-page-per-vq.xml b/tests/qemuxml2xmloutdata/net-virtio-page-per-vq.xml new file mode 100644 index 0000000..a35ed8a --- /dev/null +++ b/tests/qemuxml2xmloutdata/net-virtio-page-per-vq.xml @@ -0,0 +1,43 @@ +<domain type='qemu'> + <name>QEMUGuest1</name> + <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid> + <memory unit='KiB'>219100</memory> + <currentMemory unit='KiB'>219100</currentMemory> + <vcpu placement='static'>1</vcpu> + <os> + <type arch='i686' machine='pc'>hvm</type> + <boot dev='hd'/> + </os> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <emulator>/usr/bin/qemu-system-i386</emulator> + <disk type='block' device='disk'> + <driver name='qemu' type='raw'/> + <source dev='/dev/HostVG/QEMUGuest1'/> + <target dev='hda' bus='ide'/> + <address type='drive' controller='0' bus='0' target='0' unit='0'/> + </disk> + <controller type='usb' index='0'> + <address type='pci' domain='0x0000' bus='0x00' slot='0x01' function='0x2'/> + </controller> + <controller type='pci' index='0' model='pci-root'/> + <controller type='ide' index='0'> + <address type='pci' domain='0x0000' bus='0x00' slot='0x01' function='0x1'/> + </controller> + <interface type='user'> + <mac address='00:11:22:33:44:55'/> + <model type='virtio'/> + <driver page_per_vq='on'/> + <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> + </interface> + <input type='mouse' bus='ps2'/> + <input type='keyboard' bus='ps2'/> + <audio id='1' type='none'/> + <memballoon model='virtio'> + <address type='pci' domain='0x0000' bus='0x00' slot='0x04' function='0x0'/> + </memballoon> + </devices> +</domain> diff --git a/tests/qemuxml2xmltest.c b/tests/qemuxml2xmltest.c index c37de0c..2bd6eea 100644 --- a/tests/qemuxml2xmltest.c +++ b/tests/qemuxml2xmltest.c @@ -450,6 +450,8 @@ mymain(void) DO_TEST("net-virtio-rxtxqueuesize", QEMU_CAPS_VIRTIO_NET_RX_QUEUE_SIZE, QEMU_CAPS_VIRTIO_NET_TX_QUEUE_SIZE); + DO_TEST("net-virtio-page-per-vq", + QEMU_CAPS_VIRTIO_NET_PAGE_PER_VQ); DO_TEST("net-virtio-teaming", QEMU_CAPS_VIRTIO_NET_FAILOVER, QEMU_CAPS_DEVICE_VFIO_PCI); -- 1.8.3.1

3 4

[PATCH v3 RESEND for v7.4.0 00/14] Introduce virtio-mem <memory/> model
by Michal Privoznik 04 May '21

04 May '21

This is a rebased version of v3 I've sent about a month ago: https://listman.redhat.com/archives/libvir-list/2021-March/msg00823.html v2 here: https://listman.redhat.com/archives/libvir-list/2021-February/msg00961.html My intent is to merge these only after the upcoming release so that we have the biggest test window possible. diff to v2: - Dropped code that forbade use of virtio-mem and memballoon at the same time; - This meant that I had to adjust memory accounting, qemuDomainSetMemoryFlags() - see patches 11/15 and 12/15 which are new. - Fixed small nits raised by Peter in his review of v2 Michal Prívozník (14): virhostmem: Introduce virHostMemGetTHPSize() qemu_process: Deduplicate code in qemuProcessNeedHugepagesPath() qemu_process: Drop needless check in qemuProcessNeedMemoryBackingPath() qemu_capabilities: Introduce QEMU_CAPS_DEVICE_VIRTIO_MEM_PCI conf: Introduce virtio-mem <memory/> model qemu: Build command line for virtio-mem qemu: Wire up <memory/> live update qemu: Wire up MEMORY_DEVICE_SIZE_CHANGE event qemu: Refresh the actual size of virtio-mem on monitor reconnect qemu: Account for both memballoon and virtio-mem qemuDomainSetMemoryFlags: Take virtio-mem into consideration virsh: Introduce update-memory-device command news: document recent virtio memory addition kbase: Document virtio-mem NEWS.rst | 16 ++ docs/formatdomain.rst | 45 +++- docs/kbase/index.rst | 4 + docs/kbase/memorydevices.rst | 150 +++++++++++ docs/kbase/meson.build | 1 + docs/manpages/virsh.rst | 30 +++ docs/schemas/domaincommon.rng | 16 ++ examples/c/misc/event-test.c | 17 ++ include/libvirt/libvirt-domain.h | 23 ++ src/conf/domain_conf.c | 115 ++++++++- src/conf/domain_conf.h | 15 ++ src/conf/domain_event.c | 84 +++++++ src/conf/domain_event.h | 10 + src/conf/domain_validate.c | 39 +++ src/libvirt_private.syms | 5 + src/qemu/qemu_alias.c | 10 +- src/qemu/qemu_capabilities.c | 2 + src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 13 +- src/qemu/qemu_domain.c | 50 +++- src/qemu/qemu_domain.h | 1 + src/qemu/qemu_domain_address.c | 38 ++- src/qemu/qemu_driver.c | 233 +++++++++++++++++- src/qemu/qemu_hotplug.c | 18 ++ src/qemu/qemu_hotplug.h | 5 + src/qemu/qemu_monitor.c | 37 +++ src/qemu/qemu_monitor.h | 28 +++ src/qemu/qemu_monitor_json.c | 97 ++++++-- src/qemu/qemu_monitor_json.h | 5 + src/qemu/qemu_process.c | 118 ++++++++- src/qemu/qemu_validate.c | 8 + src/remote/remote_daemon_dispatch.c | 30 +++ src/remote/remote_driver.c | 32 +++ src/remote/remote_protocol.x | 14 +- src/remote_protocol-structs | 7 + src/security/security_apparmor.c | 1 + src/security/security_dac.c | 2 + src/security/security_selinux.c | 2 + src/util/virhostmem.c | 63 +++++ src/util/virhostmem.h | 3 + tests/domaincapsmock.c | 9 + .../caps_5.1.0.x86_64.xml | 1 + .../caps_5.2.0.x86_64.xml | 1 + .../caps_6.0.0.x86_64.xml | 1 + ...mory-hotplug-virtio-mem.x86_64-latest.args | 41 +++ .../memory-hotplug-virtio-mem.xml | 67 +++++ tests/qemuxml2argvtest.c | 1 + ...emory-hotplug-virtio-mem.x86_64-latest.xml | 1 + tests/qemuxml2xmltest.c | 1 + tools/virsh-domain.c | 169 +++++++++++++ 50 files changed, 1613 insertions(+), 67 deletions(-) create mode 100644 docs/kbase/memorydevices.rst create mode 100644 tests/qemuxml2argvdata/memory-hotplug-virtio-mem.x86_64-latest.args create mode 100644 tests/qemuxml2argvdata/memory-hotplug-virtio-mem.xml create mode 120000 tests/qemuxml2xmloutdata/memory-hotplug-virtio-mem.x86_64-latest.xml -- 2.26.3

4 32

[PATCH 1/2] qcow2: Prohibit backing file changes in 'qemu-img amend'
by Eric Blake 03 May '21

03 May '21

This was deprecated back in bc5ee6da7 (qcow2: Deprecate use of qemu-img amend to change backing file), and no one in the meantime has given any reasons why it should be supported. Time to make change attempts a hard error (but for convenience, specifying the _same_ backing chain is not forbidden). Update a couple of iotests to match. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- docs/system/deprecated.rst | 12 ------------ docs/system/removed-features.rst | 12 ++++++++++++ block/qcow2.c | 13 ++++--------- tests/qemu-iotests/061 | 3 +++ tests/qemu-iotests/061.out | 3 ++- tests/qemu-iotests/082.out | 6 ++++-- 6 files changed, 25 insertions(+), 24 deletions(-) diff --git a/docs/system/deprecated.rst b/docs/system/deprecated.rst index 80cae862528a..9ec1a9d0e03e 100644 --- a/docs/system/deprecated.rst +++ b/docs/system/deprecated.rst @@ -315,18 +315,6 @@ this CPU is also deprecated. Related binaries ---------------- -qemu-img amend to adjust backing file (since 5.1) -''''''''''''''''''''''''''''''''''''''''''''''''' - -The use of ``qemu-img amend`` to modify the name or format of a qcow2 -backing image is deprecated; this functionality was never fully -documented or tested, and interferes with other amend operations that -need access to the original backing image (such as deciding whether a -v3 zero cluster may be left unallocated when converting to a v2 -image). Rather, any changes to the backing chain should be performed -with ``qemu-img rebase -u`` either before or after the remaining -changes being performed by amend, as appropriate. - qemu-img backing file without format (since 5.1) '''''''''''''''''''''''''''''''''''''''''''''''' diff --git a/docs/system/removed-features.rst b/docs/system/removed-features.rst index 29e90601a51a..28b5df757d35 100644 --- a/docs/system/removed-features.rst +++ b/docs/system/removed-features.rst @@ -454,6 +454,18 @@ topologies described with -smp include all possible cpus, i.e. The ``enforce-config-section`` property was replaced by the ``-global migration.send-configuration={on|off}`` option. +qemu-img amend to adjust backing file (removed in 6.1) +'''''''''''''''''''''''''''''''''''''''''''''''''''''' + +The use of ``qemu-img amend`` to modify the name or format of a qcow2 +backing image was never fully documented or tested, and interferes +with other amend operations that need access to the original backing +image (such as deciding whether a v3 zero cluster may be left +unallocated when converting to a v2 image). Any changes to the +backing chain should be performed with ``qemu-img rebase -u`` either +before or after the remaining changes being performed by amend, as +appropriate. + Block devices ------------- diff --git a/block/qcow2.c b/block/qcow2.c index 9727ae8fe341..898c4fc4464c 100644 --- a/block/qcow2.c +++ b/block/qcow2.c @@ -5620,15 +5620,10 @@ static int qcow2_amend_options(BlockDriverState *bs, QemuOpts *opts, if (backing_file || backing_format) { if (g_strcmp0(backing_file, s->image_backing_file) || g_strcmp0(backing_format, s->image_backing_format)) { - warn_report("Deprecated use of amend to alter the backing file; " - "use qemu-img rebase instead"); - } - ret = qcow2_change_backing_file(bs, - backing_file ?: s->image_backing_file, - backing_format ?: s->image_backing_format); - if (ret < 0) { - error_setg_errno(errp, -ret, "Failed to change the backing file"); - return ret; + error_setg(errp, "Cannot amend the backing file"); + error_append_hint(errp, + "You can use 'qemu-img rebase' instead.\n"); + return -EINVAL; } } diff --git a/tests/qemu-iotests/061 b/tests/qemu-iotests/061 index e26d94a0df31..9507c223bda4 100755 --- a/tests/qemu-iotests/061 +++ b/tests/qemu-iotests/061 @@ -167,6 +167,9 @@ _make_test_img -o "compat=1.1" 64M TEST_IMG="$TEST_IMG.base" _make_test_img -o "compat=1.1" 64M $QEMU_IO -c "write -P 0x2a 0 128k" "$TEST_IMG.base" | _filter_qemu_io $QEMU_IO -c "read -P 0 0 128k" "$TEST_IMG" | _filter_qemu_io +$QEMU_IMG amend -o "backing_file=$TEST_IMG.base,backing_fmt=qcow2" \ + "$TEST_IMG" && echo "unexpected pass" +$QEMU_IMG rebase -u -b "$TEST_IMG.base" -F qcow2 "$TEST_IMG" $QEMU_IMG amend -o "backing_file=$TEST_IMG.base,backing_fmt=qcow2" "$TEST_IMG" $QEMU_IO -c "read -P 0x2a 0 128k" "$TEST_IMG" | _filter_qemu_io _check_test_img diff --git a/tests/qemu-iotests/061.out b/tests/qemu-iotests/061.out index ee30da266514..7ecbd4dea875 100644 --- a/tests/qemu-iotests/061.out +++ b/tests/qemu-iotests/061.out @@ -370,7 +370,8 @@ wrote 131072/131072 bytes at offset 0 128 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) read 131072/131072 bytes at offset 0 128 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) -qemu-img: warning: Deprecated use of amend to alter the backing file; use qemu-img rebase instead +qemu-img: Cannot amend the backing file +You can use 'qemu-img rebase' instead. read 131072/131072 bytes at offset 0 128 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) No errors were found on the image. diff --git a/tests/qemu-iotests/082.out b/tests/qemu-iotests/082.out index b70c12c1393d..077ed0f2c7d9 100644 --- a/tests/qemu-iotests/082.out +++ b/tests/qemu-iotests/082.out @@ -808,12 +808,14 @@ Amend options for 'qcow2': size=<size> - Virtual disk size Testing: amend -f qcow2 -o backing_file=TEST_DIR/t.qcow2,,help TEST_DIR/t.qcow2 -qemu-img: warning: Deprecated use of amend to alter the backing file; use qemu-img rebase instead +qemu-img: Cannot amend the backing file +You can use 'qemu-img rebase' instead. Testing: rebase -u -b -f qcow2 TEST_DIR/t.qcow2 Testing: amend -f qcow2 -o backing_file=TEST_DIR/t.qcow2,,? TEST_DIR/t.qcow2 -qemu-img: warning: Deprecated use of amend to alter the backing file; use qemu-img rebase instead +qemu-img: Cannot amend the backing file +You can use 'qemu-img rebase' instead. Testing: rebase -u -b -f qcow2 TEST_DIR/t.qcow2 -- 2.31.1

1 0

[PATCH] Deprecate pmem=on with non-DAX capable backend file
by Igor Mammedov 03 May '21

03 May '21

It is not safe to pretend that emulated NVDIMM supports persistence while backend actually failed to enable it and used non-persistent mapping as fall back. Instead of falling-back, QEMU should be more strict and error out with clear message that it's not supported. So if user asks for persistence (pmem=on), they should store backing file on NVDIMM. Signed-off-by: Igor Mammedov <imammedo(a)redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd(a)redhat.com> --- v2: rephrase deprecation comment andwarning message (Philippe Mathieu-Daudé <philmd(a)redhat.com>) --- docs/system/deprecated.rst | 17 +++++++++++++++++ util/mmap-alloc.c | 3 +++ 2 files changed, 20 insertions(+) diff --git a/docs/system/deprecated.rst b/docs/system/deprecated.rst index bacd76d7a5..e79fb02b3a 100644 --- a/docs/system/deprecated.rst +++ b/docs/system/deprecated.rst @@ -327,6 +327,23 @@ The Raspberry Pi machines come in various models (A, A+, B, B+). To be able to distinguish which model QEMU is implementing, the ``raspi2`` and ``raspi3`` machines have been renamed ``raspi2b`` and ``raspi3b``. +Backend options +--------------- + +Using non-persistent backing file with pmem=on (since 6.0) +'''''''''''''''''''''''''''''''''''''''''''''''''''''''''' + +This option is used when ``memory-backend-file`` is consumed by emulated NVDIMM +device. However enabling ``memory-backend-file.pmem`` option, when backing file +is (a) not DAX capable or (b) not on a filesystem that support direct mapping +of persistent memory, is not safe and may lead to data loss or corruption in case +of host crash. +Options are: + - modify VM configuration to set ``pmem=off`` to continue using fake NVDIMM + (without persistence guaranties) with backing file on non DAX storage + - move backing file to NVDIMM storage and keep ``pmem=on`` + (to have NVDIMM with persistence guaranties). + Device options -------------- diff --git a/util/mmap-alloc.c b/util/mmap-alloc.c index 27dcccd8ec..0388cc3be2 100644 --- a/util/mmap-alloc.c +++ b/util/mmap-alloc.c @@ -20,6 +20,7 @@ #include "qemu/osdep.h" #include "qemu/mmap-alloc.h" #include "qemu/host-utils.h" +#include "qemu/error-report.h" #define HUGETLBFS_MAGIC 0x958458f6 @@ -166,6 +167,8 @@ void *qemu_ram_mmap(int fd, "crash.\n", file_name); g_free(proc_link); g_free(file_name); + warn_report("Using non DAX backing file with 'pmem=on' option" + " is deprecated"); } /* * if map failed with MAP_SHARED_VALIDATE | MAP_SYNC, -- 2.27.0

3 5

[PATCH v2] meson: Declare GLIB_VERSION_* macros at configure
by Michal Privoznik 03 May '21

03 May '21

So far we have three places where glib version is recorded: meson.build and then in config.h. The latter is so well hidden that it's easy to miss when bumping minimal glib version in the former. With a bit of python^Wmeson string magic GLIB_VERSION_MIN_REQUIRED and GLIB_VERSION_MAX_ALLOWED macros can be defined to match glib_version from meson.build. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- v2 of: https://listman.redhat.com/archives/libvir-list/2021-April/msg01300.html diff to v1: - Introduced new glib_version_str variable per Pavel's suggestion config.h | 10 ---------- meson.build | 8 ++++++++ 2 files changed, 8 insertions(+), 10 deletions(-) diff --git a/config.h b/config.h index ca6720f37d..0eacfd139d 100644 --- a/config.h +++ b/config.h @@ -51,13 +51,3 @@ #else # error You either need at least GCC 4.8 or Clang 3.4 or XCode Clang 5.1 to compile libvirt #endif - -/* Ask for warnings for anything that was marked deprecated in - * the defined version, or before. It is a candidate for rewrite. - */ -#define GLIB_VERSION_MIN_REQUIRED GLIB_VERSION_2_48 - -/* Ask for warnings if code tries to use function that did not - * exist in the defined version. These risk breaking builds - */ -#define GLIB_VERSION_MAX_ALLOWED GLIB_VERSION_2_48 diff --git a/meson.build b/meson.build index 951da67896..ecbfdb1208 100644 --- a/meson.build +++ b/meson.build @@ -949,6 +949,14 @@ endif glib_dep = declare_dependency( dependencies: [ glib_dep, gobject_dep, gio_dep ], ) +glib_version_arr = glib_version.split('.') +glib_version_str = 'GLIB_VERSION_@0@_@1@'.format(glib_version_arr[0], glib_version_arr[1]) +# Ask for warnings for anything that was marked deprecated in +# the defined version, or before. It is a candidate for rewrite. +conf.set('GLIB_VERSION_MIN_REQUIRED', glib_version_str) +# Ask for warnings if code tries to use function that did not +# exist in the defined version. These risk breaking builds +conf.set('GLIB_VERSION_MAX_ALLOWED', glib_version_str) glusterfs_version = '3.4.1' glusterfs_dep = dependency('glusterfs-api', version: '>=' + glusterfs_version, required: get_option('glusterfs')) -- 2.26.3

2 1

[libvirt PATCH v2 0/6] Refactor more XML parsing boilerplate code, part VI
by Tim Wiederhake 03 May '21

03 May '21

For background, see https://listman.redhat.com/archives/libvir-list/2021-April/msg00668.html V1: https://listman.redhat.com/archives/libvir-list/2021-April/msg01184.html Changes since V1: * Split up "virDomainFeaturesDefParse" * Added missing VIR_XML_PROP_NONZERO * Some line break changes * Mentioned printf %s NULL fix in commit message * Improved justification for parser strictening Tim Wiederhake (6): virDomainGraphicsDefParseXMLSpice: Use virXMLProp* virDomainSoundDefParseXML: Use virXMLProp* virDomainFeaturesDefParse: Use virXMLPropUInt virDomainFeaturesDefParse: Use virXMLPropTristateSwitch virDomainFeaturesDefParse: Use virXMLPropEnum virDomainAudioSDLParse: Use virXMLProp* src/conf/domain_conf.c | 609 +++++------------- .../qemuxml2argvdata/aarch64-gic-invalid.err | 2 +- 2 files changed, 170 insertions(+), 441 deletions(-) -- 2.26.3

2 13

[libvirt PATCH v2 00/11] Refactor more XML parsing boilerplate code, part V
by Tim Wiederhake 03 May '21

03 May '21

For background, see https://listman.redhat.com/archives/libvir-list/2021-April/msg00668.html Changes since V1: * Split up patch for virDomainControllerDefParseXML * Code style fixes Tim Wiederhake (11): virXMLPropEnum: Fix return value virDomainControllerDef: Change type of ioeventfd to virTristateSwitch virDomainPCIControllerOpts: Change type of modelName to virDomainControllerPCIModelName virDomainControllerDefParseXML: Use virXMLProp* virDomainControllerDefParseXML: Cosmetic changes virDomainActualNetDef: Change type of type to virDomainNetType virDomainActualNetDefParseXML: Use virXMLProp* virDomainNetDefParseXML: Use virXMLProp* virDomainGraphicsListenDefParseXML: Use virXMLProp* virDomainGraphicsDef: Change type of sharePolicy to virDomainGraphicsVNCSharePolicy virDomainGraphicsDefParseXMLVNC: Use virXMLProp* src/conf/domain_conf.c | 643 +++++++++++---------------------- src/conf/domain_conf.h | 8 +- src/qemu/qemu_domain_address.c | 2 +- src/util/virxml.c | 2 +- 4 files changed, 227 insertions(+), 428 deletions(-) -- 2.26.3

2 13

Release of libvirt-7.3.0
by Jiri Denemark 03 May '21

03 May '21

The 7.3.0 release of both libvirt and libvirt-python is tagged and signed tarballs and source RPMs are available at https://libvirt.org/sources/ https://libvirt.org/sources/python/ Thanks everybody who helped with this release by sending patches, reviewing, testing, or providing any other feedback. Your work is greatly appreciated. * New features * xen: Support domains with more than 4TB The xen driver now supports domains with more than 4TB of memory with xen >= 4.13. * qemu: add socket for virtiofs filesystems Libvirt now supports ``filesystem`` devices that connect to a ``virtiofsd`` daemon launched outside of libvirtd, via the ``socket`` attribute of the ``source`` element. * nodedev: Add ability to manage persistent mediated devices Persistent mediated devices can now be managed with libvirt. ``virNodeDeviceDefineXML()`` defines a new device, ``virNodeDeviceUndefine()`` removes an existing definition, and ``virNodeDeviceCreate()`` starts a device definition that is currently inactive. Corresponding virsh commands ``nodedev-define``, ``nodedev-undefine``, and ``nodedev-start`` were also added. ``nodedev-list`` only lists active devices by default. Inactive device definitions can be shown with the new ``--inactive`` and ``--all`` flags. * qemu: Allow use of qemu's ``-compat`` option Curious developers or testers now can enable certain ``-compat`` modes which allow to notice use of deprecated commands and options as qemu will use the selected method to notify the user. The new behaviour can be requested using either the ``deprecation_behavior`` option in ``qemu.conf`` for all VMs or using ``<qemu:deprecation behavior='...'/>`` in the VM XML. * Improvements * virsh: Improve errors with ``virsh snapshot-create-as`` The XML document constructed by virsh was forced through XML schema validation which yielded unintelligible error messages in cases such as when the path to the new image did not start with a slash. XML documents are no longer validated as the XML parser actually has better error messages which allow users to figure the problem out quickly. * qemu: Terminate backing store when doing a full-chain block pull When pulling everything into the overlay image the chain can be terminated since we know that it won't depend on any backing image and thus can prevent attempts to probe the backing chain. * qemu: Expose disk serial in virDomainGetGuestInfo() The ``virDomainGetGuestInfo()`` reports disk serial number among with other disk information. * Bug fixes * qemu: Fix crash of libvirt on full block pull of a disk When the persistent definition contains a compatible disk (meaning the definition of the running and persistent config match) a block pull job would leave a dangling pointer in the config definition which resulted in a crash. * qemu: Use proper job cancelling command Libvirt's API contract for aborting a block copy job in 'ready' state declares that the destination image of the copy will contain a consistent image of the disk from the time when the block job was aborted. This requires that libvirt uses the proper cancelling qemu command to ensure that the data is consistent which was not the case. * qemu: Don't attempt storage migration when there are no migratable disks Due to a logic bug introduced in the previous release libvirt would attempt to migrate disks in case when no disks are selected/eligible for migration. * qemu: Fix very rare race when two block job 'ready' events are delivered In certain high-load scenarios, qemu might deliver the 'ready' event twice and if it's delivered when pivoting to the destination during a block copy job, libvirt would get confused and execute the code as if the job were aborted. * lxc: Fix container destroy with CGroupsV2 When an LXC container was started and the host used CGroupsV2 it might have had created nested controllers under the container's scope. Libvirt was unaware and thus destroying the container failed with a cryptic error: ``failed to get cgroup backend for 'pathOfController'``. The CGroup removal code was reworked and is now capable of dealing with such scenario. * bash-completion: Fix argument passing to $1 Due to a bug in bash completion script, the auto completion did not work properly when a connection URI or read only flag were specified on ``virsh`` or ``virt-admin`` command line. Enjoy. Jirka

1 0

[PATCH] block: Drop the sheepdog block driver
by Markus Armbruster 03 May '21

03 May '21

It was deprecated in commit e1c4269763, v5.2.0. See that commit message for rationale. Signed-off-by: Markus Armbruster <armbru(a)redhat.com> --- docs/system/deprecated.rst | 9 - docs/system/device-url-syntax.rst.inc | 18 - docs/system/qemu-block-drivers.rst.inc | 69 - docs/system/removed-features.rst | 7 + configure | 10 - meson.build | 1 - qapi/block-core.json | 93 +- qapi/transaction.json | 8 +- block/sheepdog.c | 3356 ------------------------ .gitlab-ci.yml | 1 - MAINTAINERS | 6 - block/meson.build | 1 - block/trace-events | 14 - tests/qemu-iotests/005 | 5 - tests/qemu-iotests/025 | 2 +- tests/qemu-iotests/check | 3 +- tests/qemu-iotests/common.rc | 4 - 17 files changed, 14 insertions(+), 3593 deletions(-) delete mode 100644 block/sheepdog.c diff --git a/docs/system/deprecated.rst b/docs/system/deprecated.rst index 80cae86252..519c49b4a2 100644 --- a/docs/system/deprecated.rst +++ b/docs/system/deprecated.rst @@ -286,15 +286,6 @@ The above, converted to the current supported format:: json:{"file.driver":"rbd", "file.pool":"rbd", "file.image":"name"} -``sheepdog`` driver (since 5.2.0) -^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ - -The ``sheepdog`` block device driver is deprecated. The corresponding upstream -server project is no longer actively maintained. Users are recommended to switch -to an alternative distributed block device driver such as RBD. The -``qemu-img convert`` command can be used to liberate existing data by moving -it out of sheepdog volumes into an alternative storage backend. - linux-user mode CPUs -------------------- diff --git a/docs/system/device-url-syntax.rst.inc b/docs/system/device-url-syntax.rst.inc index 6f6ec8366b..d15a021508 100644 --- a/docs/system/device-url-syntax.rst.inc +++ b/docs/system/device-url-syntax.rst.inc @@ -85,24 +85,6 @@ These are specified using a special URL syntax. Currently authentication must be done using ssh-agent. Other authentication methods may be supported in future. -``Sheepdog`` - Sheepdog is a distributed storage system for QEMU. QEMU supports - using either local sheepdog devices or remote networked devices. - - Syntax for specifying a sheepdog device - - :: - - sheepdog[+tcp|+unix]://[host:port]/vdiname[?socket=path][#snapid|#tag] - - Example - - .. parsed-literal:: - - |qemu_system| --drive file=sheepdog://192.0.2.1:30000/MyVirtualMachine - - See also https://sheepdog.github.io/sheepdog/. - ``GlusterFS`` GlusterFS is a user space distributed file system. QEMU supports the use of GlusterFS volumes for hosting VM disk images using TCP, Unix diff --git a/docs/system/qemu-block-drivers.rst.inc b/docs/system/qemu-block-drivers.rst.inc index 60a064b232..16225710eb 100644 --- a/docs/system/qemu-block-drivers.rst.inc +++ b/docs/system/qemu-block-drivers.rst.inc @@ -547,75 +547,6 @@ also available. Here are some example of the older syntax: |qemu_system| linux2.img -hdb nbd:unix:/tmp/my_socket |qemu_system| -cdrom nbd:localhost:10809:exportname=debian-500-ppc-netinst - - -Sheepdog disk images -~~~~~~~~~~~~~~~~~~~~ - -Sheepdog is a distributed storage system for QEMU. It provides highly -available block level storage volumes that can be attached to -QEMU-based virtual machines. - -You can create a Sheepdog disk image with the command: - -.. parsed-literal:: - - qemu-img create sheepdog:///IMAGE SIZE - -where *IMAGE* is the Sheepdog image name and *SIZE* is its -size. - -To import the existing *FILENAME* to Sheepdog, you can use a -convert command. - -.. parsed-literal:: - - qemu-img convert FILENAME sheepdog:///IMAGE - -You can boot from the Sheepdog disk image with the command: - -.. parsed-literal:: - - |qemu_system| sheepdog:///IMAGE - -You can also create a snapshot of the Sheepdog image like qcow2. - -.. parsed-literal:: - - qemu-img snapshot -c TAG sheepdog:///IMAGE - -where *TAG* is a tag name of the newly created snapshot. - -To boot from the Sheepdog snapshot, specify the tag name of the -snapshot. - -.. parsed-literal:: - - |qemu_system| sheepdog:///IMAGE#TAG - -You can create a cloned image from the existing snapshot. - -.. parsed-literal:: - - qemu-img create -b sheepdog:///BASE#TAG sheepdog:///IMAGE - -where *BASE* is an image name of the source snapshot and *TAG* -is its tag name. - -You can use an unix socket instead of an inet socket: - -.. parsed-literal:: - - |qemu_system| sheepdog+unix:///IMAGE?socket=PATH - -If the Sheepdog daemon doesn't run on the local host, you need to -specify one of the Sheepdog servers to connect to. - -.. parsed-literal:: - - qemu-img create sheepdog://HOSTNAME:PORT/IMAGE SIZE - |qemu_system| sheepdog://HOSTNAME:PORT/IMAGE - iSCSI LUNs ~~~~~~~~~~ diff --git a/docs/system/removed-features.rst b/docs/system/removed-features.rst index 29e90601a5..7ce91133da 100644 --- a/docs/system/removed-features.rst +++ b/docs/system/removed-features.rst @@ -461,3 +461,10 @@ VXHS backend (removed in 5.1) ''''''''''''''''''''''''''''' The VXHS code did not compile since v2.12.0. It was removed in 5.1. + +``sheepdog`` driver (removed in 6.0) +'''''''''''''''''''''''''''''''''''' + +The corresponding upstream server project is no longer maintained. +Users are recommended to switch to an alternative distributed block +device driver such as RBD. diff --git a/configure b/configure index 4f374b4889..5676203373 100755 --- a/configure +++ b/configure @@ -446,7 +446,6 @@ vdi=${default_feature:-yes} vvfat=${default_feature:-yes} qed=${default_feature:-yes} parallels=${default_feature:-yes} -sheepdog="no" libxml2="$default_feature" debug_mutex="no" libpmem="$default_feature" @@ -1474,10 +1473,6 @@ for opt do ;; --enable-parallels) parallels="yes" ;; - --disable-sheepdog) sheepdog="no" - ;; - --enable-sheepdog) sheepdog="yes" - ;; --disable-vhost-user) vhost_user="no" ;; --enable-vhost-user) vhost_user="yes" @@ -1907,7 +1902,6 @@ disabled with --disable-FEATURE, default is enabled if available vvfat vvfat image format support qed qed image format support parallels parallels image format support - sheepdog sheepdog block driver support (deprecated) crypto-afalg Linux AF_ALG crypto backend driver capstone capstone disassembler support debug-mutex mutex debugging support @@ -6073,10 +6067,6 @@ fi if test "$parallels" = "yes" ; then echo "CONFIG_PARALLELS=y" >> $config_host_mak fi -if test "$sheepdog" = "yes" ; then - add_to deprecated_features "sheepdog" - echo "CONFIG_SHEEPDOG=y" >> $config_host_mak -fi if test "$have_mlockall" = "yes" ; then echo "HAVE_MLOCKALL=y" >> $config_host_mak fi diff --git a/meson.build b/meson.build index c6f4b0cf5e..f32c0e8bce 100644 --- a/meson.build +++ b/meson.build @@ -2616,7 +2616,6 @@ if have_block summary_info += {'vvfat support': config_host.has_key('CONFIG_VVFAT')} summary_info += {'qed support': config_host.has_key('CONFIG_QED')} summary_info += {'parallels support': config_host.has_key('CONFIG_PARALLELS')} - summary_info += {'sheepdog support': config_host.has_key('CONFIG_SHEEPDOG')} summary_info += {'FUSE exports': fuse.found()} endif summary(summary_info, bool_yn: true, section: 'Block layer support') diff --git a/qapi/block-core.json b/qapi/block-core.json index 6d227924d0..2ea294129e 100644 --- a/qapi/block-core.json +++ b/qapi/block-core.json @@ -2818,7 +2818,6 @@ 'luks', 'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', 'parallels', 'preallocate', 'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd', { 'name': 'replication', 'if': 'defined(CONFIG_REPLICATION)' }, - 'sheepdog', 'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat' ] } ## @@ -3651,26 +3650,6 @@ '*key-secret': 'str', '*server': ['InetSocketAddressBase'] } } -## -# @BlockdevOptionsSheepdog: -# -# Driver specific block device options for sheepdog -# -# @vdi: Virtual disk image name -# @server: The Sheepdog server to connect to -# @snap-id: Snapshot ID -# @tag: Snapshot tag name -# -# Only one of @snap-id and @tag may be present. -# -# Since: 2.9 -## -{ 'struct': 'BlockdevOptionsSheepdog', - 'data': { 'server': 'SocketAddress', - 'vdi': 'str', - '*snap-id': 'uint32', - '*tag': 'str' } } - ## # @ReplicationMode: # @@ -4037,7 +4016,6 @@ 'rbd': 'BlockdevOptionsRbd', 'replication': { 'type': 'BlockdevOptionsReplication', 'if': 'defined(CONFIG_REPLICATION)' }, - 'sheepdog': 'BlockdevOptionsSheepdog', 'ssh': 'BlockdevOptionsSsh', 'throttle': 'BlockdevOptionsThrottle', 'vdi': 'BlockdevOptionsGenericFormat', @@ -4496,74 +4474,6 @@ '*zeroed-grain': 'bool' } } -## -# @SheepdogRedundancyType: -# -# @full: Create a fully replicated vdi with x copies -# @erasure-coded: Create an erasure coded vdi with x data strips and -# y parity strips -# -# Since: 2.12 -## -{ 'enum': 'SheepdogRedundancyType', - 'data': [ 'full', 'erasure-coded' ] } - -## -# @SheepdogRedundancyFull: -# -# @copies: Number of copies to use (between 1 and 31) -# -# Since: 2.12 -## -{ 'struct': 'SheepdogRedundancyFull', - 'data': { 'copies': 'int' }} - -## -# @SheepdogRedundancyErasureCoded: -# -# @data-strips: Number of data strips to use (one of {2,4,8,16}) -# @parity-strips: Number of parity strips to use (between 1 and 15) -# -# Since: 2.12 -## -{ 'struct': 'SheepdogRedundancyErasureCoded', - 'data': { 'data-strips': 'int', - 'parity-strips': 'int' }} - -## -# @SheepdogRedundancy: -# -# Since: 2.12 -## -{ 'union': 'SheepdogRedundancy', - 'base': { 'type': 'SheepdogRedundancyType' }, - 'discriminator': 'type', - 'data': { 'full': 'SheepdogRedundancyFull', - 'erasure-coded': 'SheepdogRedundancyErasureCoded' } } - -## -# @BlockdevCreateOptionsSheepdog: -# -# Driver specific image creation options for Sheepdog. -# -# @location: Where to store the new image file -# @size: Size of the virtual disk in bytes -# @backing-file: File name of a base image -# @preallocation: Preallocation mode for the new image (default: off; -# allowed values: off, full) -# @redundancy: Redundancy of the image -# @object-size: Object size of the image -# -# Since: 2.12 -## -{ 'struct': 'BlockdevCreateOptionsSheepdog', - 'data': { 'location': 'BlockdevOptionsSheepdog', - 'size': 'size', - '*backing-file': 'str', - '*preallocation': 'PreallocMode', - '*redundancy': 'SheepdogRedundancy', - '*object-size': 'size' } } - ## # @BlockdevCreateOptionsSsh: # @@ -4687,7 +4597,6 @@ 'qcow2': 'BlockdevCreateOptionsQcow2', 'qed': 'BlockdevCreateOptionsQed', 'rbd': 'BlockdevCreateOptionsRbd', - 'sheepdog': 'BlockdevCreateOptionsSheepdog', 'ssh': 'BlockdevCreateOptionsSsh', 'vdi': 'BlockdevCreateOptionsVdi', 'vhdx': 'BlockdevCreateOptionsVhdx', @@ -5322,7 +5231,7 @@ # # Notes: In transaction, if @name is empty, or any snapshot matching @name # exists, the operation will fail. Only some image formats support it, -# for example, qcow2, rbd, and sheepdog. +# for example, qcow2, and rbd. # # Since: 1.7 ## diff --git a/qapi/transaction.json b/qapi/transaction.json index 15ddebdbc3..894258d9e2 100644 --- a/qapi/transaction.json +++ b/qapi/transaction.json @@ -112,10 +112,10 @@ # # On failure, the original disks pre-snapshot attempt will be used. # -# For internal snapshots, the dictionary contains the device and the snapshot's -# name. If an internal snapshot matching name already exists, the request will -# be rejected. Only some image formats support it, for example, qcow2, rbd, -# and sheepdog. +# For internal snapshots, the dictionary contains the device and the +# snapshot's name. If an internal snapshot matching name already exists, +# the request will be rejected. Only some image formats support it, for +# example, qcow2, and rbd, # # On failure, qemu will try delete the newly created internal snapshot in the # transaction. When an I/O error occurs during deletion, the user needs to fix diff --git a/block/sheepdog.c b/block/sheepdog.c deleted file mode 100644 index a45c73826d..0000000000 --- a/block/sheepdog.c +++ /dev/null @@ -1,3356 +0,0 @@ -/* - * Copyright (C) 2009-2010 Nippon Telegraph and Telephone Corporation. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License version - * 2 as published by the Free Software Foundation. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. - * - * Contributions after 2012-01-13 are licensed under the terms of the - * GNU GPL, version 2 or (at your option) any later version. - */ - -#include "qemu/osdep.h" -#include "qemu-common.h" -#include "qapi/error.h" -#include "qapi/qapi-visit-sockets.h" -#include "qapi/qapi-visit-block-core.h" -#include "qapi/qmp/qdict.h" -#include "qapi/qobject-input-visitor.h" -#include "qapi/qobject-output-visitor.h" -#include "qemu/uri.h" -#include "qemu/error-report.h" -#include "qemu/main-loop.h" -#include "qemu/module.h" -#include "qemu/option.h" -#include "qemu/sockets.h" -#include "block/block_int.h" -#include "block/qdict.h" -#include "sysemu/block-backend.h" -#include "qemu/bitops.h" -#include "qemu/cutils.h" -#include "trace.h" - -#define SD_PROTO_VER 0x01 - -#define SD_DEFAULT_ADDR "localhost" -#define SD_DEFAULT_PORT 7000 - -#define SD_OP_CREATE_AND_WRITE_OBJ 0x01 -#define SD_OP_READ_OBJ 0x02 -#define SD_OP_WRITE_OBJ 0x03 -/* 0x04 is used internally by Sheepdog */ - -#define SD_OP_NEW_VDI 0x11 -#define SD_OP_LOCK_VDI 0x12 -#define SD_OP_RELEASE_VDI 0x13 -#define SD_OP_GET_VDI_INFO 0x14 -#define SD_OP_READ_VDIS 0x15 -#define SD_OP_FLUSH_VDI 0x16 -#define SD_OP_DEL_VDI 0x17 -#define SD_OP_GET_CLUSTER_DEFAULT 0x18 - -#define SD_FLAG_CMD_WRITE 0x01 -#define SD_FLAG_CMD_COW 0x02 -#define SD_FLAG_CMD_CACHE 0x04 /* Writeback mode for cache */ -#define SD_FLAG_CMD_DIRECT 0x08 /* Don't use cache */ - -#define SD_RES_SUCCESS 0x00 /* Success */ -#define SD_RES_UNKNOWN 0x01 /* Unknown error */ -#define SD_RES_NO_OBJ 0x02 /* No object found */ -#define SD_RES_EIO 0x03 /* I/O error */ -#define SD_RES_VDI_EXIST 0x04 /* Vdi exists already */ -#define SD_RES_INVALID_PARMS 0x05 /* Invalid parameters */ -#define SD_RES_SYSTEM_ERROR 0x06 /* System error */ -#define SD_RES_VDI_LOCKED 0x07 /* Vdi is locked */ -#define SD_RES_NO_VDI 0x08 /* No vdi found */ -#define SD_RES_NO_BASE_VDI 0x09 /* No base vdi found */ -#define SD_RES_VDI_READ 0x0A /* Cannot read requested vdi */ -#define SD_RES_VDI_WRITE 0x0B /* Cannot write requested vdi */ -#define SD_RES_BASE_VDI_READ 0x0C /* Cannot read base vdi */ -#define SD_RES_BASE_VDI_WRITE 0x0D /* Cannot write base vdi */ -#define SD_RES_NO_TAG 0x0E /* Requested tag is not found */ -#define SD_RES_STARTUP 0x0F /* Sheepdog is on starting up */ -#define SD_RES_VDI_NOT_LOCKED 0x10 /* Vdi is not locked */ -#define SD_RES_SHUTDOWN 0x11 /* Sheepdog is shutting down */ -#define SD_RES_NO_MEM 0x12 /* Cannot allocate memory */ -#define SD_RES_FULL_VDI 0x13 /* we already have the maximum vdis */ -#define SD_RES_VER_MISMATCH 0x14 /* Protocol version mismatch */ -#define SD_RES_NO_SPACE 0x15 /* Server has no room for new objects */ -#define SD_RES_WAIT_FOR_FORMAT 0x16 /* Waiting for a format operation */ -#define SD_RES_WAIT_FOR_JOIN 0x17 /* Waiting for other nodes joining */ -#define SD_RES_JOIN_FAILED 0x18 /* Target node had failed to join sheepdog */ -#define SD_RES_HALT 0x19 /* Sheepdog is stopped serving IO request */ -#define SD_RES_READONLY 0x1A /* Object is read-only */ - -/* - * Object ID rules - * - * 0 - 19 (20 bits): data object space - * 20 - 31 (12 bits): reserved data object space - * 32 - 55 (24 bits): vdi object space - * 56 - 59 ( 4 bits): reserved vdi object space - * 60 - 63 ( 4 bits): object type identifier space - */ - -#define VDI_SPACE_SHIFT 32 -#define VDI_BIT (UINT64_C(1) << 63) -#define VMSTATE_BIT (UINT64_C(1) << 62) -#define MAX_DATA_OBJS (UINT64_C(1) << 20) -#define MAX_CHILDREN 1024 -#define SD_MAX_VDI_LEN 256 -#define SD_MAX_VDI_TAG_LEN 256 -#define SD_NR_VDIS (1U << 24) -#define SD_DATA_OBJ_SIZE (UINT64_C(1) << 22) -#define SD_MAX_VDI_SIZE (SD_DATA_OBJ_SIZE * MAX_DATA_OBJS) -#define SD_DEFAULT_BLOCK_SIZE_SHIFT 22 -/* - * For erasure coding, we use at most SD_EC_MAX_STRIP for data strips and - * (SD_EC_MAX_STRIP - 1) for parity strips - * - * SD_MAX_COPIES is sum of number of data strips and parity strips. - */ -#define SD_EC_MAX_STRIP 16 -#define SD_MAX_COPIES (SD_EC_MAX_STRIP * 2 - 1) - -#define SD_INODE_SIZE (sizeof(SheepdogInode)) -#define CURRENT_VDI_ID 0 - -#define LOCK_TYPE_NORMAL 0 -#define LOCK_TYPE_SHARED 1 /* for iSCSI multipath */ - -typedef struct SheepdogReq { - uint8_t proto_ver; - uint8_t opcode; - uint16_t flags; - uint32_t epoch; - uint32_t id; - uint32_t data_length; - uint32_t opcode_specific[8]; -} SheepdogReq; - -typedef struct SheepdogRsp { - uint8_t proto_ver; - uint8_t opcode; - uint16_t flags; - uint32_t epoch; - uint32_t id; - uint32_t data_length; - uint32_t result; - uint32_t opcode_specific[7]; -} SheepdogRsp; - -typedef struct SheepdogObjReq { - uint8_t proto_ver; - uint8_t opcode; - uint16_t flags; - uint32_t epoch; - uint32_t id; - uint32_t data_length; - uint64_t oid; - uint64_t cow_oid; - uint8_t copies; - uint8_t copy_policy; - uint8_t reserved[6]; - uint64_t offset; -} SheepdogObjReq; - -typedef struct SheepdogObjRsp { - uint8_t proto_ver; - uint8_t opcode; - uint16_t flags; - uint32_t epoch; - uint32_t id; - uint32_t data_length; - uint32_t result; - uint8_t copies; - uint8_t copy_policy; - uint8_t reserved[2]; - uint32_t pad[6]; -} SheepdogObjRsp; - -typedef struct SheepdogVdiReq { - uint8_t proto_ver; - uint8_t opcode; - uint16_t flags; - uint32_t epoch; - uint32_t id; - uint32_t data_length; - uint64_t vdi_size; - uint32_t base_vdi_id; - uint8_t copies; - uint8_t copy_policy; - uint8_t store_policy; - uint8_t block_size_shift; - uint32_t snapid; - uint32_t type; - uint32_t pad[2]; -} SheepdogVdiReq; - -typedef struct SheepdogVdiRsp { - uint8_t proto_ver; - uint8_t opcode; - uint16_t flags; - uint32_t epoch; - uint32_t id; - uint32_t data_length; - uint32_t result; - uint32_t rsvd; - uint32_t vdi_id; - uint32_t pad[5]; -} SheepdogVdiRsp; - -typedef struct SheepdogClusterRsp { - uint8_t proto_ver; - uint8_t opcode; - uint16_t flags; - uint32_t epoch; - uint32_t id; - uint32_t data_length; - uint32_t result; - uint8_t nr_copies; - uint8_t copy_policy; - uint8_t block_size_shift; - uint8_t __pad1; - uint32_t __pad2[6]; -} SheepdogClusterRsp; - -typedef struct SheepdogInode { - char name[SD_MAX_VDI_LEN]; - char tag[SD_MAX_VDI_TAG_LEN]; - uint64_t ctime; - uint64_t snap_ctime; - uint64_t vm_clock_nsec; - uint64_t vdi_size; - uint64_t vm_state_size; - uint16_t copy_policy; - uint8_t nr_copies; - uint8_t block_size_shift; - uint32_t snap_id; - uint32_t vdi_id; - uint32_t parent_vdi_id; - uint32_t child_vdi_id[MAX_CHILDREN]; - uint32_t data_vdi_id[MAX_DATA_OBJS]; -} SheepdogInode; - -#define SD_INODE_HEADER_SIZE offsetof(SheepdogInode, data_vdi_id) - -/* - * 64 bit FNV-1a non-zero initial basis - */ -#define FNV1A_64_INIT ((uint64_t)0xcbf29ce484222325ULL) - -static void deprecation_warning(void) -{ - static bool warned; - - if (!warned) { - warn_report("the sheepdog block driver is deprecated"); - warned = true; - } -} - -/* - * 64 bit Fowler/Noll/Vo FNV-1a hash code - */ -static inline uint64_t fnv_64a_buf(void *buf, size_t len, uint64_t hval) -{ - unsigned char *bp = buf; - unsigned char *be = bp + len; - while (bp < be) { - hval ^= (uint64_t) *bp++; - hval += (hval << 1) + (hval << 4) + (hval << 5) + - (hval << 7) + (hval << 8) + (hval << 40); - } - return hval; -} - -static inline bool is_data_obj_writable(SheepdogInode *inode, unsigned int idx) -{ - return inode->vdi_id == inode->data_vdi_id[idx]; -} - -static inline bool is_data_obj(uint64_t oid) -{ - return !(VDI_BIT & oid); -} - -static inline uint64_t data_oid_to_idx(uint64_t oid) -{ - return oid & (MAX_DATA_OBJS - 1); -} - -static inline uint32_t oid_to_vid(uint64_t oid) -{ - return (oid & ~VDI_BIT) >> VDI_SPACE_SHIFT; -} - -static inline uint64_t vid_to_vdi_oid(uint32_t vid) -{ - return VDI_BIT | ((uint64_t)vid << VDI_SPACE_SHIFT); -} - -static inline uint64_t vid_to_vmstate_oid(uint32_t vid, uint32_t idx) -{ - return VMSTATE_BIT | ((uint64_t)vid << VDI_SPACE_SHIFT) | idx; -} - -static inline uint64_t vid_to_data_oid(uint32_t vid, uint32_t idx) -{ - return ((uint64_t)vid << VDI_SPACE_SHIFT) | idx; -} - -static inline bool is_snapshot(struct SheepdogInode *inode) -{ - return !!inode->snap_ctime; -} - -static inline size_t count_data_objs(const struct SheepdogInode *inode) -{ - return DIV_ROUND_UP(inode->vdi_size, - (1UL << inode->block_size_shift)); -} - -typedef struct SheepdogAIOCB SheepdogAIOCB; -typedef struct BDRVSheepdogState BDRVSheepdogState; - -typedef struct AIOReq { - SheepdogAIOCB *aiocb; - unsigned int iov_offset; - - uint64_t oid; - uint64_t base_oid; - uint64_t offset; - unsigned int data_len; - uint8_t flags; - uint32_t id; - bool create; - - QLIST_ENTRY(AIOReq) aio_siblings; -} AIOReq; - -enum AIOCBState { - AIOCB_WRITE_UDATA, - AIOCB_READ_UDATA, - AIOCB_FLUSH_CACHE, - AIOCB_DISCARD_OBJ, -}; - -#define AIOCBOverlapping(x, y) \ - (!(x->max_affect_data_idx < y->min_affect_data_idx \ - || y->max_affect_data_idx < x->min_affect_data_idx)) - -struct SheepdogAIOCB { - BDRVSheepdogState *s; - - QEMUIOVector *qiov; - - int64_t sector_num; - int nb_sectors; - - int ret; - enum AIOCBState aiocb_type; - - Coroutine *coroutine; - int nr_pending; - - uint32_t min_affect_data_idx; - uint32_t max_affect_data_idx; - - /* - * The difference between affect_data_idx and dirty_data_idx: - * affect_data_idx represents range of index of all request types. - * dirty_data_idx represents range of index updated by COW requests. - * dirty_data_idx is used for updating an inode object. - */ - uint32_t min_dirty_data_idx; - uint32_t max_dirty_data_idx; - - QLIST_ENTRY(SheepdogAIOCB) aiocb_siblings; -}; - -struct BDRVSheepdogState { - BlockDriverState *bs; - AioContext *aio_context; - - SheepdogInode inode; - - char name[SD_MAX_VDI_LEN]; - bool is_snapshot; - uint32_t cache_flags; - bool discard_supported; - - SocketAddress *addr; - int fd; - - CoMutex lock; - Coroutine *co_send; - Coroutine *co_recv; - - uint32_t aioreq_seq_num; - - /* Every aio request must be linked to either of these queues. */ - QLIST_HEAD(, AIOReq) inflight_aio_head; - QLIST_HEAD(, AIOReq) failed_aio_head; - - CoMutex queue_lock; - CoQueue overlapping_queue; - QLIST_HEAD(, SheepdogAIOCB) inflight_aiocb_head; -}; - -typedef struct BDRVSheepdogReopenState { - int fd; - int cache_flags; -} BDRVSheepdogReopenState; - -static const char *sd_strerror(int err) -{ - int i; - - static const struct { - int err; - const char *desc; - } errors[] = { - {SD_RES_SUCCESS, "Success"}, - {SD_RES_UNKNOWN, "Unknown error"}, - {SD_RES_NO_OBJ, "No object found"}, - {SD_RES_EIO, "I/O error"}, - {SD_RES_VDI_EXIST, "VDI exists already"}, - {SD_RES_INVALID_PARMS, "Invalid parameters"}, - {SD_RES_SYSTEM_ERROR, "System error"}, - {SD_RES_VDI_LOCKED, "VDI is already locked"}, - {SD_RES_NO_VDI, "No vdi found"}, - {SD_RES_NO_BASE_VDI, "No base VDI found"}, - {SD_RES_VDI_READ, "Failed read the requested VDI"}, - {SD_RES_VDI_WRITE, "Failed to write the requested VDI"}, - {SD_RES_BASE_VDI_READ, "Failed to read the base VDI"}, - {SD_RES_BASE_VDI_WRITE, "Failed to write the base VDI"}, - {SD_RES_NO_TAG, "Failed to find the requested tag"}, - {SD_RES_STARTUP, "The system is still booting"}, - {SD_RES_VDI_NOT_LOCKED, "VDI isn't locked"}, - {SD_RES_SHUTDOWN, "The system is shutting down"}, - {SD_RES_NO_MEM, "Out of memory on the server"}, - {SD_RES_FULL_VDI, "We already have the maximum vdis"}, - {SD_RES_VER_MISMATCH, "Protocol version mismatch"}, - {SD_RES_NO_SPACE, "Server has no space for new objects"}, - {SD_RES_WAIT_FOR_FORMAT, "Sheepdog is waiting for a format operation"}, - {SD_RES_WAIT_FOR_JOIN, "Sheepdog is waiting for other nodes joining"}, - {SD_RES_JOIN_FAILED, "Target node had failed to join sheepdog"}, - {SD_RES_HALT, "Sheepdog is stopped serving IO request"}, - {SD_RES_READONLY, "Object is read-only"}, - }; - - for (i = 0; i < ARRAY_SIZE(errors); ++i) { - if (errors[i].err == err) { - return errors[i].desc; - } - } - - return "Invalid error code"; -} - -/* - * Sheepdog I/O handling: - * - * 1. In sd_co_rw_vector, we send the I/O requests to the server and - * link the requests to the inflight_list in the - * BDRVSheepdogState. The function yields while waiting for - * receiving the response. - * - * 2. We receive the response in aio_read_response, the fd handler to - * the sheepdog connection. We switch back to sd_co_readv/sd_writev - * after all the requests belonging to the AIOCB are finished. If - * needed, sd_co_writev will send another requests for the vdi object. - */ - -static inline AIOReq *alloc_aio_req(BDRVSheepdogState *s, SheepdogAIOCB *acb, - uint64_t oid, unsigned int data_len, - uint64_t offset, uint8_t flags, bool create, - uint64_t base_oid, unsigned int iov_offset) -{ - AIOReq *aio_req; - - aio_req = g_malloc(sizeof(*aio_req)); - aio_req->aiocb = acb; - aio_req->iov_offset = iov_offset; - aio_req->oid = oid; - aio_req->base_oid = base_oid; - aio_req->offset = offset; - aio_req->data_len = data_len; - aio_req->flags = flags; - aio_req->id = s->aioreq_seq_num++; - aio_req->create = create; - - acb->nr_pending++; - return aio_req; -} - -static void wait_for_overlapping_aiocb(BDRVSheepdogState *s, SheepdogAIOCB *acb) -{ - SheepdogAIOCB *cb; - -retry: - QLIST_FOREACH(cb, &s->inflight_aiocb_head, aiocb_siblings) { - if (AIOCBOverlapping(acb, cb)) { - qemu_co_queue_wait(&s->overlapping_queue, &s->queue_lock); - goto retry; - } - } -} - -static void sd_aio_setup(SheepdogAIOCB *acb, BDRVSheepdogState *s, - QEMUIOVector *qiov, int64_t sector_num, int nb_sectors, - int type) -{ - uint32_t object_size; - - object_size = (UINT32_C(1) << s->inode.block_size_shift); - - acb->s = s; - - acb->qiov = qiov; - - acb->sector_num = sector_num; - acb->nb_sectors = nb_sectors; - - acb->coroutine = qemu_coroutine_self(); - acb->ret = 0; - acb->nr_pending = 0; - - acb->min_affect_data_idx = acb->sector_num * BDRV_SECTOR_SIZE / object_size; - acb->max_affect_data_idx = (acb->sector_num * BDRV_SECTOR_SIZE + - acb->nb_sectors * BDRV_SECTOR_SIZE) / object_size; - - acb->min_dirty_data_idx = UINT32_MAX; - acb->max_dirty_data_idx = 0; - acb->aiocb_type = type; - - if (type == AIOCB_FLUSH_CACHE) { - return; - } - - qemu_co_mutex_lock(&s->queue_lock); - wait_for_overlapping_aiocb(s, acb); - QLIST_INSERT_HEAD(&s->inflight_aiocb_head, acb, aiocb_siblings); - qemu_co_mutex_unlock(&s->queue_lock); -} - -static SocketAddress *sd_server_config(QDict *options, Error **errp) -{ - QDict *server = NULL; - Visitor *iv = NULL; - SocketAddress *saddr = NULL; - - qdict_extract_subqdict(options, &server, "server."); - - iv = qobject_input_visitor_new_flat_confused(server, errp); - if (!iv) { - goto done; - } - - if (!visit_type_SocketAddress(iv, NULL, &saddr, errp)) { - goto done; - } - -done: - visit_free(iv); - qobject_unref(server); - return saddr; -} - -/* Return -EIO in case of error, file descriptor on success */ -static int connect_to_sdog(BDRVSheepdogState *s, Error **errp) -{ - int fd; - - fd = socket_connect(s->addr, errp); - - if (s->addr->type == SOCKET_ADDRESS_TYPE_INET && fd >= 0) { - int ret = socket_set_nodelay(fd); - if (ret < 0) { - warn_report("can't set TCP_NODELAY: %s", strerror(errno)); - } - } - - if (fd >= 0) { - qemu_set_nonblock(fd); - } else { - fd = -EIO; - } - - return fd; -} - -/* Return 0 on success and -errno in case of error */ -static coroutine_fn int send_co_req(int sockfd, SheepdogReq *hdr, void *data, - unsigned int *wlen) -{ - int ret; - - ret = qemu_co_send(sockfd, hdr, sizeof(*hdr)); - if (ret != sizeof(*hdr)) { - error_report("failed to send a req, %s", strerror(errno)); - return -errno; - } - - ret = qemu_co_send(sockfd, data, *wlen); - if (ret != *wlen) { - error_report("failed to send a req, %s", strerror(errno)); - return -errno; - } - - return ret; -} - -typedef struct SheepdogReqCo { - int sockfd; - BlockDriverState *bs; - AioContext *aio_context; - SheepdogReq *hdr; - void *data; - unsigned int *wlen; - unsigned int *rlen; - int ret; - bool finished; - Coroutine *co; -} SheepdogReqCo; - -static void restart_co_req(void *opaque) -{ - SheepdogReqCo *srco = opaque; - - aio_co_wake(srco->co); -} - -static coroutine_fn void do_co_req(void *opaque) -{ - int ret; - SheepdogReqCo *srco = opaque; - int sockfd = srco->sockfd; - SheepdogReq *hdr = srco->hdr; - void *data = srco->data; - unsigned int *wlen = srco->wlen; - unsigned int *rlen = srco->rlen; - - srco->co = qemu_coroutine_self(); - aio_set_fd_handler(srco->aio_context, sockfd, false, - NULL, restart_co_req, NULL, srco); - - ret = send_co_req(sockfd, hdr, data, wlen); - if (ret < 0) { - goto out; - } - - aio_set_fd_handler(srco->aio_context, sockfd, false, - restart_co_req, NULL, NULL, srco); - - ret = qemu_co_recv(sockfd, hdr, sizeof(*hdr)); - if (ret != sizeof(*hdr)) { - error_report("failed to get a rsp, %s", strerror(errno)); - ret = -errno; - goto out; - } - - if (*rlen > hdr->data_length) { - *rlen = hdr->data_length; - } - - if (*rlen) { - ret = qemu_co_recv(sockfd, data, *rlen); - if (ret != *rlen) { - error_report("failed to get the data, %s", strerror(errno)); - ret = -errno; - goto out; - } - } - ret = 0; -out: - /* there is at most one request for this sockfd, so it is safe to - * set each handler to NULL. */ - aio_set_fd_handler(srco->aio_context, sockfd, false, - NULL, NULL, NULL, NULL); - - srco->co = NULL; - srco->ret = ret; - /* Set srco->finished before reading bs->wakeup. */ - qatomic_mb_set(&srco->finished, true); - if (srco->bs) { - bdrv_wakeup(srco->bs); - } -} - -/* - * Send the request to the sheep in a synchronous manner. - * - * Return 0 on success, -errno in case of error. - */ -static int do_req(int sockfd, BlockDriverState *bs, SheepdogReq *hdr, - void *data, unsigned int *wlen, unsigned int *rlen) -{ - Coroutine *co; - SheepdogReqCo srco = { - .sockfd = sockfd, - .aio_context = bs ? bdrv_get_aio_context(bs) : qemu_get_aio_context(), - .bs = bs, - .hdr = hdr, - .data = data, - .wlen = wlen, - .rlen = rlen, - .ret = 0, - .finished = false, - }; - - if (qemu_in_coroutine()) { - do_co_req(&srco); - } else { - co = qemu_coroutine_create(do_co_req, &srco); - if (bs) { - bdrv_coroutine_enter(bs, co); - BDRV_POLL_WHILE(bs, !srco.finished); - } else { - qemu_coroutine_enter(co); - while (!srco.finished) { - aio_poll(qemu_get_aio_context(), true); - } - } - } - - return srco.ret; -} - -static void coroutine_fn add_aio_request(BDRVSheepdogState *s, AIOReq *aio_req, - struct iovec *iov, int niov, - enum AIOCBState aiocb_type); -static void coroutine_fn resend_aioreq(BDRVSheepdogState *s, AIOReq *aio_req); -static int reload_inode(BDRVSheepdogState *s, uint32_t snapid, const char *tag); -static int get_sheep_fd(BDRVSheepdogState *s, Error **errp); -static void co_write_request(void *opaque); - -static coroutine_fn void reconnect_to_sdog(void *opaque) -{ - BDRVSheepdogState *s = opaque; - AIOReq *aio_req, *next; - - aio_set_fd_handler(s->aio_context, s->fd, false, NULL, - NULL, NULL, NULL); - close(s->fd); - s->fd = -1; - - /* Wait for outstanding write requests to be completed. */ - while (s->co_send != NULL) { - co_write_request(opaque); - } - - /* Try to reconnect the sheepdog server every one second. */ - while (s->fd < 0) { - Error *local_err = NULL; - s->fd = get_sheep_fd(s, &local_err); - if (s->fd < 0) { - trace_sheepdog_reconnect_to_sdog(); - error_report_err(local_err); - qemu_co_sleep_ns(QEMU_CLOCK_REALTIME, NANOSECONDS_PER_SECOND); - } - }; - - /* - * Now we have to resend all the request in the inflight queue. However, - * resend_aioreq() can yield and newly created requests can be added to the - * inflight queue before the coroutine is resumed. To avoid mixing them, we - * have to move all the inflight requests to the failed queue before - * resend_aioreq() is called. - */ - qemu_co_mutex_lock(&s->queue_lock); - QLIST_FOREACH_SAFE(aio_req, &s->inflight_aio_head, aio_siblings, next) { - QLIST_REMOVE(aio_req, aio_siblings); - QLIST_INSERT_HEAD(&s->failed_aio_head, aio_req, aio_siblings); - } - - /* Resend all the failed aio requests. */ - while (!QLIST_EMPTY(&s->failed_aio_head)) { - aio_req = QLIST_FIRST(&s->failed_aio_head); - QLIST_REMOVE(aio_req, aio_siblings); - qemu_co_mutex_unlock(&s->queue_lock); - resend_aioreq(s, aio_req); - qemu_co_mutex_lock(&s->queue_lock); - } - qemu_co_mutex_unlock(&s->queue_lock); -} - -/* - * Receive responses of the I/O requests. - * - * This function is registered as a fd handler, and called from the - * main loop when s->fd is ready for reading responses. - */ -static void coroutine_fn aio_read_response(void *opaque) -{ - SheepdogObjRsp rsp; - BDRVSheepdogState *s = opaque; - int fd = s->fd; - int ret; - AIOReq *aio_req = NULL; - SheepdogAIOCB *acb; - uint64_t idx; - - /* read a header */ - ret = qemu_co_recv(fd, &rsp, sizeof(rsp)); - if (ret != sizeof(rsp)) { - error_report("failed to get the header, %s", strerror(errno)); - goto err; - } - - /* find the right aio_req from the inflight aio list */ - QLIST_FOREACH(aio_req, &s->inflight_aio_head, aio_siblings) { - if (aio_req->id == rsp.id) { - break; - } - } - if (!aio_req) { - error_report("cannot find aio_req %x", rsp.id); - goto err; - } - - acb = aio_req->aiocb; - - switch (acb->aiocb_type) { - case AIOCB_WRITE_UDATA: - if (!is_data_obj(aio_req->oid)) { - break; - } - idx = data_oid_to_idx(aio_req->oid); - - if (aio_req->create) { - /* - * If the object is newly created one, we need to update - * the vdi object (metadata object). min_dirty_data_idx - * and max_dirty_data_idx are changed to include updated - * index between them. - */ - if (rsp.result == SD_RES_SUCCESS) { - s->inode.data_vdi_id[idx] = s->inode.vdi_id; - acb->max_dirty_data_idx = MAX(idx, acb->max_dirty_data_idx); - acb->min_dirty_data_idx = MIN(idx, acb->min_dirty_data_idx); - } - } - break; - case AIOCB_READ_UDATA: - ret = qemu_co_recvv(fd, acb->qiov->iov, acb->qiov->niov, - aio_req->iov_offset, rsp.data_length); - if (ret != rsp.data_length) { - error_report("failed to get the data, %s", strerror(errno)); - goto err; - } - break; - case AIOCB_FLUSH_CACHE: - if (rsp.result == SD_RES_INVALID_PARMS) { - trace_sheepdog_aio_read_response(); - s->cache_flags = SD_FLAG_CMD_DIRECT; - rsp.result = SD_RES_SUCCESS; - } - break; - case AIOCB_DISCARD_OBJ: - switch (rsp.result) { - case SD_RES_INVALID_PARMS: - error_report("server doesn't support discard command"); - rsp.result = SD_RES_SUCCESS; - s->discard_supported = false; - break; - default: - break; - } - } - - /* No more data for this aio_req (reload_inode below uses its own file - * descriptor handler which doesn't use co_recv). - */ - s->co_recv = NULL; - - qemu_co_mutex_lock(&s->queue_lock); - QLIST_REMOVE(aio_req, aio_siblings); - qemu_co_mutex_unlock(&s->queue_lock); - - switch (rsp.result) { - case SD_RES_SUCCESS: - break; - case SD_RES_READONLY: - if (s->inode.vdi_id == oid_to_vid(aio_req->oid)) { - ret = reload_inode(s, 0, ""); - if (ret < 0) { - goto err; - } - } - if (is_data_obj(aio_req->oid)) { - aio_req->oid = vid_to_data_oid(s->inode.vdi_id, - data_oid_to_idx(aio_req->oid)); - } else { - aio_req->oid = vid_to_vdi_oid(s->inode.vdi_id); - } - resend_aioreq(s, aio_req); - return; - default: - acb->ret = -EIO; - error_report("%s", sd_strerror(rsp.result)); - break; - } - - g_free(aio_req); - - if (!--acb->nr_pending) { - /* - * We've finished all requests which belong to the AIOCB, so - * we can switch back to sd_co_readv/writev now. - */ - aio_co_wake(acb->coroutine); - } - - return; - -err: - reconnect_to_sdog(opaque); -} - -static void co_read_response(void *opaque) -{ - BDRVSheepdogState *s = opaque; - - if (!s->co_recv) { - s->co_recv = qemu_coroutine_create(aio_read_response, opaque); - } - - aio_co_enter(s->aio_context, s->co_recv); -} - -static void co_write_request(void *opaque) -{ - BDRVSheepdogState *s = opaque; - - aio_co_wake(s->co_send); -} - -/* - * Return a socket descriptor to read/write objects. - * - * We cannot use this descriptor for other operations because - * the block driver may be on waiting response from the server. - */ -static int get_sheep_fd(BDRVSheepdogState *s, Error **errp) -{ - int fd; - - fd = connect_to_sdog(s, errp); - if (fd < 0) { - return fd; - } - - aio_set_fd_handler(s->aio_context, fd, false, - co_read_response, NULL, NULL, s); - return fd; -} - -/* - * Parse numeric snapshot ID in @str - * If @str can't be parsed as number, return false. - * Else, if the number is zero or too large, set *@snapid to zero and - * return true. - * Else, set *@snapid to the number and return true. - */ -static bool sd_parse_snapid(const char *str, uint32_t *snapid) -{ - unsigned long ul; - int ret; - - ret = qemu_strtoul(str, NULL, 10, &ul); - if (ret == -ERANGE) { - ul = ret = 0; - } - if (ret) { - return false; - } - if (ul > UINT32_MAX) { - ul = 0; - } - - *snapid = ul; - return true; -} - -static bool sd_parse_snapid_or_tag(const char *str, - uint32_t *snapid, char tag[]) -{ - if (!sd_parse_snapid(str, snapid)) { - *snapid = 0; - if (g_strlcpy(tag, str, SD_MAX_VDI_TAG_LEN) >= SD_MAX_VDI_TAG_LEN) { - return false; - } - } else if (!*snapid) { - return false; - } else { - tag[0] = 0; - } - return true; -} - -typedef struct { - const char *path; /* non-null iff transport is tcp */ - const char *host; /* valid when transport is tcp */ - int port; /* valid when transport is tcp */ - char vdi[SD_MAX_VDI_LEN]; - char tag[SD_MAX_VDI_TAG_LEN]; - uint32_t snap_id; - /* Remainder is only for sd_config_done() */ - URI *uri; - QueryParams *qp; -} SheepdogConfig; - -static void sd_config_done(SheepdogConfig *cfg) -{ - if (cfg->qp) { - query_params_free(cfg->qp); - } - uri_free(cfg->uri); -} - -static void sd_parse_uri(SheepdogConfig *cfg, const char *filename, - Error **errp) -{ - Error *err = NULL; - QueryParams *qp = NULL; - bool is_unix; - URI *uri; - - memset(cfg, 0, sizeof(*cfg)); - - cfg->uri = uri = uri_parse(filename); - if (!uri) { - error_setg(&err, "invalid URI '%s'", filename); - goto out; - } - - /* transport */ - if (!g_strcmp0(uri->scheme, "sheepdog")) { - is_unix = false; - } else if (!g_strcmp0(uri->scheme, "sheepdog+tcp")) { - is_unix = false; - } else if (!g_strcmp0(uri->scheme, "sheepdog+unix")) { - is_unix = true; - } else { - error_setg(&err, "URI scheme must be 'sheepdog', 'sheepdog+tcp'," - " or 'sheepdog+unix'"); - goto out; - } - - if (uri->path == NULL || !strcmp(uri->path, "/")) { - error_setg(&err, "missing file path in URI"); - goto out; - } - if (g_strlcpy(cfg->vdi, uri->path + 1, SD_MAX_VDI_LEN) - >= SD_MAX_VDI_LEN) { - error_setg(&err, "VDI name is too long"); - goto out; - } - - cfg->qp = qp = query_params_parse(uri->query); - - if (is_unix) { - /* sheepdog+unix:///vdiname?socket=path */ - if (uri->server || uri->port) { - error_setg(&err, "URI scheme %s doesn't accept a server address", - uri->scheme); - goto out; - } - if (!qp->n) { - error_setg(&err, - "URI scheme %s requires query parameter 'socket'", - uri->scheme); - goto out; - } - if (qp->n != 1 || strcmp(qp->p[0].name, "socket")) { - error_setg(&err, "unexpected query parameters"); - goto out; - } - cfg->path = qp->p[0].value; - } else { - /* sheepdog[+tcp]://[host:port]/vdiname */ - if (qp->n) { - error_setg(&err, "unexpected query parameters"); - goto out; - } - cfg->host = uri->server; - cfg->port = uri->port; - } - - /* snapshot tag */ - if (uri->fragment) { - if (!sd_parse_snapid_or_tag(uri->fragment, - &cfg->snap_id, cfg->tag)) { - error_setg(&err, "'%s' is not a valid snapshot ID", - uri->fragment); - goto out; - } - } else { - cfg->snap_id = CURRENT_VDI_ID; /* search current vdi */ - } - -out: - if (err) { - error_propagate(errp, err); - sd_config_done(cfg); - } -} - -/* - * Parse a filename (old syntax) - * - * filename must be one of the following formats: - * 1. [vdiname] - * 2. [vdiname]:[snapid] - * 3. [vdiname]:[tag] - * 4. [hostname]:[port]:[vdiname] - * 5. [hostname]:[port]:[vdiname]:[snapid] - * 6. [hostname]:[port]:[vdiname]:[tag] - * - * You can boot from the snapshot images by specifying `snapid` or - * `tag'. - * - * You can run VMs outside the Sheepdog cluster by specifying - * `hostname' and `port' (experimental). - */ -static void parse_vdiname(SheepdogConfig *cfg, const char *filename, - Error **errp) -{ - Error *err = NULL; - char *p, *q, *uri; - const char *host_spec, *vdi_spec; - int nr_sep; - - strstart(filename, "sheepdog:", &filename); - p = q = g_strdup(filename); - - /* count the number of separators */ - nr_sep = 0; - while (*p) { - if (*p == ':') { - nr_sep++; - } - p++; - } - p = q; - - /* use the first two tokens as host_spec. */ - if (nr_sep >= 2) { - host_spec = p; - p = strchr(p, ':'); - p++; - p = strchr(p, ':'); - *p++ = '\0'; - } else { - host_spec = ""; - } - - vdi_spec = p; - - p = strchr(vdi_spec, ':'); - if (p) { - *p++ = '#'; - } - - uri = g_strdup_printf("sheepdog://%s/%s", host_spec, vdi_spec); - - /* - * FIXME We to escape URI meta-characters, e.g. "x?y=z" - * produces "sheepdog://x?y=z". Because of that ... - */ - sd_parse_uri(cfg, uri, &err); - if (err) { - /* - * ... this can fail, but the error message is misleading. - * Replace it by the traditional useless one until the - * escaping is fixed. - */ - error_free(err); - error_setg(errp, "Can't parse filename"); - } - - g_free(q); - g_free(uri); -} - -static void sd_parse_filename(const char *filename, QDict *options, - Error **errp) -{ - Error *err = NULL; - SheepdogConfig cfg; - char buf[32]; - - if (strstr(filename, "://")) { - sd_parse_uri(&cfg, filename, &err); - } else { - parse_vdiname(&cfg, filename, &err); - } - if (err) { - error_propagate(errp, err); - return; - } - - if (cfg.path) { - qdict_set_default_str(options, "server.path", cfg.path); - qdict_set_default_str(options, "server.type", "unix"); - } else { - qdict_set_default_str(options, "server.type", "inet"); - qdict_set_default_str(options, "server.host", - cfg.host ?: SD_DEFAULT_ADDR); - snprintf(buf, sizeof(buf), "%d", cfg.port ?: SD_DEFAULT_PORT); - qdict_set_default_str(options, "server.port", buf); - } - qdict_set_default_str(options, "vdi", cfg.vdi); - qdict_set_default_str(options, "tag", cfg.tag); - if (cfg.snap_id) { - snprintf(buf, sizeof(buf), "%d", cfg.snap_id); - qdict_set_default_str(options, "snap-id", buf); - } - - sd_config_done(&cfg); -} - -static int find_vdi_name(BDRVSheepdogState *s, const char *filename, - uint32_t snapid, const char *tag, uint32_t *vid, - bool lock, Error **errp) -{ - int ret, fd; - SheepdogVdiReq hdr; - SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr; - unsigned int wlen, rlen = 0; - char buf[SD_MAX_VDI_LEN + SD_MAX_VDI_TAG_LEN] QEMU_NONSTRING; - - fd = connect_to_sdog(s, errp); - if (fd < 0) { - return fd; - } - - /* This pair of strncpy calls ensures that the buffer is zero-filled, - * which is desirable since we'll soon be sending those bytes, and - * don't want the send_req to read uninitialized data. - */ - strncpy(buf, filename, SD_MAX_VDI_LEN); - strncpy(buf + SD_MAX_VDI_LEN, tag, SD_MAX_VDI_TAG_LEN); - - memset(&hdr, 0, sizeof(hdr)); - if (lock) { - hdr.opcode = SD_OP_LOCK_VDI; - hdr.type = LOCK_TYPE_NORMAL; - } else { - hdr.opcode = SD_OP_GET_VDI_INFO; - } - wlen = SD_MAX_VDI_LEN + SD_MAX_VDI_TAG_LEN; - hdr.proto_ver = SD_PROTO_VER; - hdr.data_length = wlen; - hdr.snapid = snapid; - hdr.flags = SD_FLAG_CMD_WRITE; - - ret = do_req(fd, s->bs, (SheepdogReq *)&hdr, buf, &wlen, &rlen); - if (ret) { - error_setg_errno(errp, -ret, "cannot get vdi info"); - goto out; - } - - if (rsp->result != SD_RES_SUCCESS) { - error_setg(errp, "cannot get vdi info, %s, %s %" PRIu32 " %s", - sd_strerror(rsp->result), filename, snapid, tag); - if (rsp->result == SD_RES_NO_VDI) { - ret = -ENOENT; - } else if (rsp->result == SD_RES_VDI_LOCKED) { - ret = -EBUSY; - } else { - ret = -EIO; - } - goto out; - } - *vid = rsp->vdi_id; - - ret = 0; -out: - closesocket(fd); - return ret; -} - -static void coroutine_fn add_aio_request(BDRVSheepdogState *s, AIOReq *aio_req, - struct iovec *iov, int niov, - enum AIOCBState aiocb_type) -{ - int nr_copies = s->inode.nr_copies; - SheepdogObjReq hdr; - unsigned int wlen = 0; - int ret; - uint64_t oid = aio_req->oid; - unsigned int datalen = aio_req->data_len; - uint64_t offset = aio_req->offset; - uint8_t flags = aio_req->flags; - uint64_t old_oid = aio_req->base_oid; - bool create = aio_req->create; - - qemu_co_mutex_lock(&s->queue_lock); - QLIST_INSERT_HEAD(&s->inflight_aio_head, aio_req, aio_siblings); - qemu_co_mutex_unlock(&s->queue_lock); - - if (!nr_copies) { - error_report("bug"); - } - - memset(&hdr, 0, sizeof(hdr)); - - switch (aiocb_type) { - case AIOCB_FLUSH_CACHE: - hdr.opcode = SD_OP_FLUSH_VDI; - break; - case AIOCB_READ_UDATA: - hdr.opcode = SD_OP_READ_OBJ; - hdr.flags = flags; - break; - case AIOCB_WRITE_UDATA: - if (create) { - hdr.opcode = SD_OP_CREATE_AND_WRITE_OBJ; - } else { - hdr.opcode = SD_OP_WRITE_OBJ; - } - wlen = datalen; - hdr.flags = SD_FLAG_CMD_WRITE | flags; - break; - case AIOCB_DISCARD_OBJ: - hdr.opcode = SD_OP_WRITE_OBJ; - hdr.flags = SD_FLAG_CMD_WRITE | flags; - s->inode.data_vdi_id[data_oid_to_idx(oid)] = 0; - offset = offsetof(SheepdogInode, - data_vdi_id[data_oid_to_idx(oid)]); - oid = vid_to_vdi_oid(s->inode.vdi_id); - wlen = datalen = sizeof(uint32_t); - break; - } - - if (s->cache_flags) { - hdr.flags |= s->cache_flags; - } - - hdr.oid = oid; - hdr.cow_oid = old_oid; - hdr.copies = s->inode.nr_copies; - - hdr.data_length = datalen; - hdr.offset = offset; - - hdr.id = aio_req->id; - - qemu_co_mutex_lock(&s->lock); - s->co_send = qemu_coroutine_self(); - aio_set_fd_handler(s->aio_context, s->fd, false, - co_read_response, co_write_request, NULL, s); - socket_set_cork(s->fd, 1); - - /* send a header */ - ret = qemu_co_send(s->fd, &hdr, sizeof(hdr)); - if (ret != sizeof(hdr)) { - error_report("failed to send a req, %s", strerror(errno)); - goto out; - } - - if (wlen) { - ret = qemu_co_sendv(s->fd, iov, niov, aio_req->iov_offset, wlen); - if (ret != wlen) { - error_report("failed to send a data, %s", strerror(errno)); - } - } -out: - socket_set_cork(s->fd, 0); - aio_set_fd_handler(s->aio_context, s->fd, false, - co_read_response, NULL, NULL, s); - s->co_send = NULL; - qemu_co_mutex_unlock(&s->lock); -} - -static int read_write_object(int fd, BlockDriverState *bs, char *buf, - uint64_t oid, uint8_t copies, - unsigned int datalen, uint64_t offset, - bool write, bool create, uint32_t cache_flags) -{ - SheepdogObjReq hdr; - SheepdogObjRsp *rsp = (SheepdogObjRsp *)&hdr; - unsigned int wlen, rlen; - int ret; - - memset(&hdr, 0, sizeof(hdr)); - - if (write) { - wlen = datalen; - rlen = 0; - hdr.flags = SD_FLAG_CMD_WRITE; - if (create) { - hdr.opcode = SD_OP_CREATE_AND_WRITE_OBJ; - } else { - hdr.opcode = SD_OP_WRITE_OBJ; - } - } else { - wlen = 0; - rlen = datalen; - hdr.opcode = SD_OP_READ_OBJ; - } - - hdr.flags |= cache_flags; - - hdr.oid = oid; - hdr.data_length = datalen; - hdr.offset = offset; - hdr.copies = copies; - - ret = do_req(fd, bs, (SheepdogReq *)&hdr, buf, &wlen, &rlen); - if (ret) { - error_report("failed to send a request to the sheep"); - return ret; - } - - switch (rsp->result) { - case SD_RES_SUCCESS: - return 0; - default: - error_report("%s", sd_strerror(rsp->result)); - return -EIO; - } -} - -static int read_object(int fd, BlockDriverState *bs, char *buf, - uint64_t oid, uint8_t copies, - unsigned int datalen, uint64_t offset, - uint32_t cache_flags) -{ - return read_write_object(fd, bs, buf, oid, copies, - datalen, offset, false, - false, cache_flags); -} - -static int write_object(int fd, BlockDriverState *bs, char *buf, - uint64_t oid, uint8_t copies, - unsigned int datalen, uint64_t offset, bool create, - uint32_t cache_flags) -{ - return read_write_object(fd, bs, buf, oid, copies, - datalen, offset, true, - create, cache_flags); -} - -/* update inode with the latest state */ -static int reload_inode(BDRVSheepdogState *s, uint32_t snapid, const char *tag) -{ - Error *local_err = NULL; - SheepdogInode *inode; - int ret = 0, fd; - uint32_t vid = 0; - - fd = connect_to_sdog(s, &local_err); - if (fd < 0) { - error_report_err(local_err); - return -EIO; - } - - inode = g_malloc(SD_INODE_HEADER_SIZE); - - ret = find_vdi_name(s, s->name, snapid, tag, &vid, false, &local_err); - if (ret) { - error_report_err(local_err); - goto out; - } - - ret = read_object(fd, s->bs, (char *)inode, vid_to_vdi_oid(vid), - s->inode.nr_copies, SD_INODE_HEADER_SIZE, 0, - s->cache_flags); - if (ret < 0) { - goto out; - } - - if (inode->vdi_id != s->inode.vdi_id) { - memcpy(&s->inode, inode, SD_INODE_HEADER_SIZE); - } - -out: - g_free(inode); - closesocket(fd); - - return ret; -} - -static void coroutine_fn resend_aioreq(BDRVSheepdogState *s, AIOReq *aio_req) -{ - SheepdogAIOCB *acb = aio_req->aiocb; - - aio_req->create = false; - - /* check whether this request becomes a CoW one */ - if (acb->aiocb_type == AIOCB_WRITE_UDATA && is_data_obj(aio_req->oid)) { - int idx = data_oid_to_idx(aio_req->oid); - - if (is_data_obj_writable(&s->inode, idx)) { - goto out; - } - - if (s->inode.data_vdi_id[idx]) { - aio_req->base_oid = vid_to_data_oid(s->inode.data_vdi_id[idx], idx); - aio_req->flags |= SD_FLAG_CMD_COW; - } - aio_req->create = true; - } -out: - if (is_data_obj(aio_req->oid)) { - add_aio_request(s, aio_req, acb->qiov->iov, acb->qiov->niov, - acb->aiocb_type); - } else { - struct iovec iov; - iov.iov_base = &s->inode; - iov.iov_len = sizeof(s->inode); - add_aio_request(s, aio_req, &iov, 1, AIOCB_WRITE_UDATA); - } -} - -static void sd_detach_aio_context(BlockDriverState *bs) -{ - BDRVSheepdogState *s = bs->opaque; - - aio_set_fd_handler(s->aio_context, s->fd, false, NULL, - NULL, NULL, NULL); -} - -static void sd_attach_aio_context(BlockDriverState *bs, - AioContext *new_context) -{ - BDRVSheepdogState *s = bs->opaque; - - s->aio_context = new_context; - aio_set_fd_handler(new_context, s->fd, false, - co_read_response, NULL, NULL, s); -} - -static QemuOptsList runtime_opts = { - .name = "sheepdog", - .head = QTAILQ_HEAD_INITIALIZER(runtime_opts.head), - .desc = { - { - .name = "vdi", - .type = QEMU_OPT_STRING, - }, - { - .name = "snap-id", - .type = QEMU_OPT_NUMBER, - }, - { - .name = "tag", - .type = QEMU_OPT_STRING, - }, - { /* end of list */ } - }, -}; - -static int sd_open(BlockDriverState *bs, QDict *options, int flags, - Error **errp) -{ - int ret, fd; - uint32_t vid = 0; - BDRVSheepdogState *s = bs->opaque; - const char *vdi, *snap_id_str, *tag; - uint64_t snap_id; - char *buf = NULL; - QemuOpts *opts; - - deprecation_warning(); - - s->bs = bs; - s->aio_context = bdrv_get_aio_context(bs); - - opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort); - if (!qemu_opts_absorb_qdict(opts, options, errp)) { - ret = -EINVAL; - goto err_no_fd; - } - - s->addr = sd_server_config(options, errp); - if (!s->addr) { - ret = -EINVAL; - goto err_no_fd; - } - - vdi = qemu_opt_get(opts, "vdi"); - snap_id_str = qemu_opt_get(opts, "snap-id"); - snap_id = qemu_opt_get_number(opts, "snap-id", CURRENT_VDI_ID); - tag = qemu_opt_get(opts, "tag"); - - if (!vdi) { - error_setg(errp, "parameter 'vdi' is missing"); - ret = -EINVAL; - goto err_no_fd; - } - if (strlen(vdi) >= SD_MAX_VDI_LEN) { - error_setg(errp, "value of parameter 'vdi' is too long"); - ret = -EINVAL; - goto err_no_fd; - } - - if (snap_id > UINT32_MAX) { - snap_id = 0; - } - if (snap_id_str && !snap_id) { - error_setg(errp, "'snap-id=%s' is not a valid snapshot ID", - snap_id_str); - ret = -EINVAL; - goto err_no_fd; - } - - if (!tag) { - tag = ""; - } - if (strlen(tag) >= SD_MAX_VDI_TAG_LEN) { - error_setg(errp, "value of parameter 'tag' is too long"); - ret = -EINVAL; - goto err_no_fd; - } - - QLIST_INIT(&s->inflight_aio_head); - QLIST_INIT(&s->failed_aio_head); - QLIST_INIT(&s->inflight_aiocb_head); - - s->fd = get_sheep_fd(s, errp); - if (s->fd < 0) { - ret = s->fd; - goto err_no_fd; - } - - ret = find_vdi_name(s, vdi, (uint32_t)snap_id, tag, &vid, true, errp); - if (ret) { - goto err; - } - - /* - * QEMU block layer emulates writethrough cache as 'writeback + flush', so - * we always set SD_FLAG_CMD_CACHE (writeback cache) as default. - */ - s->cache_flags = SD_FLAG_CMD_CACHE; - if (flags & BDRV_O_NOCACHE) { - s->cache_flags = SD_FLAG_CMD_DIRECT; - } - s->discard_supported = true; - - if (snap_id || tag[0]) { - trace_sheepdog_open(vid); - s->is_snapshot = true; - } - - fd = connect_to_sdog(s, errp); - if (fd < 0) { - ret = fd; - goto err; - } - - buf = g_malloc(SD_INODE_SIZE); - ret = read_object(fd, s->bs, buf, vid_to_vdi_oid(vid), - 0, SD_INODE_SIZE, 0, s->cache_flags); - - closesocket(fd); - - if (ret) { - error_setg(errp, "Can't read snapshot inode"); - goto err; - } - - memcpy(&s->inode, buf, sizeof(s->inode)); - - bs->total_sectors = s->inode.vdi_size / BDRV_SECTOR_SIZE; - bs->supported_truncate_flags = BDRV_REQ_ZERO_WRITE; - pstrcpy(s->name, sizeof(s->name), vdi); - qemu_co_mutex_init(&s->lock); - qemu_co_mutex_init(&s->queue_lock); - qemu_co_queue_init(&s->overlapping_queue); - qemu_opts_del(opts); - g_free(buf); - return 0; - -err: - aio_set_fd_handler(bdrv_get_aio_context(bs), s->fd, - false, NULL, NULL, NULL, NULL); - closesocket(s->fd); -err_no_fd: - qemu_opts_del(opts); - g_free(buf); - return ret; -} - -static int sd_reopen_prepare(BDRVReopenState *state, BlockReopenQueue *queue, - Error **errp) -{ - BDRVSheepdogState *s = state->bs->opaque; - BDRVSheepdogReopenState *re_s; - int ret = 0; - - re_s = state->opaque = g_new0(BDRVSheepdogReopenState, 1); - - re_s->cache_flags = SD_FLAG_CMD_CACHE; - if (state->flags & BDRV_O_NOCACHE) { - re_s->cache_flags = SD_FLAG_CMD_DIRECT; - } - - re_s->fd = get_sheep_fd(s, errp); - if (re_s->fd < 0) { - ret = re_s->fd; - return ret; - } - - return ret; -} - -static void sd_reopen_commit(BDRVReopenState *state) -{ - BDRVSheepdogReopenState *re_s = state->opaque; - BDRVSheepdogState *s = state->bs->opaque; - - if (s->fd) { - aio_set_fd_handler(s->aio_context, s->fd, false, - NULL, NULL, NULL, NULL); - closesocket(s->fd); - } - - s->fd = re_s->fd; - s->cache_flags = re_s->cache_flags; - - g_free(state->opaque); - state->opaque = NULL; - - return; -} - -static void sd_reopen_abort(BDRVReopenState *state) -{ - BDRVSheepdogReopenState *re_s = state->opaque; - BDRVSheepdogState *s = state->bs->opaque; - - if (re_s == NULL) { - return; - } - - if (re_s->fd) { - aio_set_fd_handler(s->aio_context, re_s->fd, false, - NULL, NULL, NULL, NULL); - closesocket(re_s->fd); - } - - g_free(state->opaque); - state->opaque = NULL; - - return; -} - -static int do_sd_create(BDRVSheepdogState *s, uint32_t *vdi_id, int snapshot, - Error **errp) -{ - SheepdogVdiReq hdr; - SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr; - int fd, ret; - unsigned int wlen, rlen = 0; - char buf[SD_MAX_VDI_LEN]; - - fd = connect_to_sdog(s, errp); - if (fd < 0) { - return fd; - } - - /* FIXME: would it be better to fail (e.g., return -EIO) when filename - * does not fit in buf? For now, just truncate and avoid buffer overrun. - */ - memset(buf, 0, sizeof(buf)); - pstrcpy(buf, sizeof(buf), s->name); - - memset(&hdr, 0, sizeof(hdr)); - hdr.opcode = SD_OP_NEW_VDI; - hdr.base_vdi_id = s->inode.vdi_id; - - wlen = SD_MAX_VDI_LEN; - - hdr.flags = SD_FLAG_CMD_WRITE; - hdr.snapid = snapshot; - - hdr.data_length = wlen; - hdr.vdi_size = s->inode.vdi_size; - hdr.copy_policy = s->inode.copy_policy; - hdr.copies = s->inode.nr_copies; - hdr.block_size_shift = s->inode.block_size_shift; - - ret = do_req(fd, NULL, (SheepdogReq *)&hdr, buf, &wlen, &rlen); - - closesocket(fd); - - if (ret) { - error_setg_errno(errp, -ret, "create failed"); - return ret; - } - - if (rsp->result != SD_RES_SUCCESS) { - error_setg(errp, "%s, %s", sd_strerror(rsp->result), s->inode.name); - return -EIO; - } - - if (vdi_id) { - *vdi_id = rsp->vdi_id; - } - - return 0; -} - -static int sd_prealloc(BlockDriverState *bs, int64_t old_size, int64_t new_size, - Error **errp) -{ - BlockBackend *blk = NULL; - BDRVSheepdogState *base = bs->opaque; - unsigned long buf_size; - uint32_t idx, max_idx; - uint32_t object_size; - void *buf = NULL; - int ret; - - blk = blk_new_with_bs(bs, - BLK_PERM_CONSISTENT_READ | BLK_PERM_WRITE | BLK_PERM_RESIZE, - BLK_PERM_ALL, errp); - - if (!blk) { - ret = -EPERM; - goto out_with_err_set; - } - - blk_set_allow_write_beyond_eof(blk, true); - - object_size = (UINT32_C(1) << base->inode.block_size_shift); - buf_size = MIN(object_size, SD_DATA_OBJ_SIZE); - buf = g_malloc0(buf_size); - - max_idx = DIV_ROUND_UP(new_size, buf_size); - - for (idx = old_size / buf_size; idx < max_idx; idx++) { - /* - * The created image can be a cloned image, so we need to read - * a data from the source image. - */ - ret = blk_pread(blk, idx * buf_size, buf, buf_size); - if (ret < 0) { - goto out; - } - ret = blk_pwrite(blk, idx * buf_size, buf, buf_size, 0); - if (ret < 0) { - goto out; - } - } - - ret = 0; -out: - if (ret < 0) { - error_setg_errno(errp, -ret, "Can't pre-allocate"); - } -out_with_err_set: - blk_unref(blk); - g_free(buf); - - return ret; -} - -static int sd_create_prealloc(BlockdevOptionsSheepdog *location, int64_t size, - Error **errp) -{ - BlockDriverState *bs; - Visitor *v; - QObject *obj = NULL; - QDict *qdict; - int ret; - - v = qobject_output_visitor_new(&obj); - visit_type_BlockdevOptionsSheepdog(v, NULL, &location, &error_abort); - visit_free(v); - - qdict = qobject_to(QDict, obj); - qdict_flatten(qdict); - - qdict_put_str(qdict, "driver", "sheepdog"); - - bs = bdrv_open(NULL, NULL, qdict, BDRV_O_PROTOCOL | BDRV_O_RDWR, errp); - if (bs == NULL) { - ret = -EIO; - goto fail; - } - - ret = sd_prealloc(bs, 0, size, errp); -fail: - bdrv_unref(bs); - qobject_unref(qdict); - return ret; -} - -static int parse_redundancy(BDRVSheepdogState *s, SheepdogRedundancy *opt) -{ - struct SheepdogInode *inode = &s->inode; - - switch (opt->type) { - case SHEEPDOG_REDUNDANCY_TYPE_FULL: - if (opt->u.full.copies > SD_MAX_COPIES || opt->u.full.copies < 1) { - return -EINVAL; - } - inode->copy_policy = 0; - inode->nr_copies = opt->u.full.copies; - return 0; - - case SHEEPDOG_REDUNDANCY_TYPE_ERASURE_CODED: - { - int64_t copy = opt->u.erasure_coded.data_strips; - int64_t parity = opt->u.erasure_coded.parity_strips; - - if (copy != 2 && copy != 4 && copy != 8 && copy != 16) { - return -EINVAL; - } - - if (parity >= SD_EC_MAX_STRIP || parity < 1) { - return -EINVAL; - } - - /* - * 4 bits for parity and 4 bits for data. - * We have to compress upper data bits because it can't represent 16 - */ - inode->copy_policy = ((copy / 2) << 4) + parity; - inode->nr_copies = copy + parity; - return 0; - } - - default: - g_assert_not_reached(); - } - - return -EINVAL; -} - -/* - * Sheepdog support two kinds of redundancy, full replication and erasure - * coding. - * - * # create a fully replicated vdi with x copies - * -o redundancy=x (1 <= x <= SD_MAX_COPIES) - * - * # create a erasure coded vdi with x data strips and y parity strips - * -o redundancy=x:y (x must be one of {2,4,8,16} and 1 <= y < SD_EC_MAX_STRIP) - */ -static SheepdogRedundancy *parse_redundancy_str(const char *opt) -{ - SheepdogRedundancy *redundancy; - const char *n1, *n2; - long copy, parity; - char p[10]; - int ret; - - pstrcpy(p, sizeof(p), opt); - n1 = strtok(p, ":"); - n2 = strtok(NULL, ":"); - - if (!n1) { - return NULL; - } - - ret = qemu_strtol(n1, NULL, 10, &copy); - if (ret < 0) { - return NULL; - } - - redundancy = g_new0(SheepdogRedundancy, 1); - if (!n2) { - *redundancy = (SheepdogRedundancy) { - .type = SHEEPDOG_REDUNDANCY_TYPE_FULL, - .u.full.copies = copy, - }; - } else { - ret = qemu_strtol(n2, NULL, 10, &parity); - if (ret < 0) { - g_free(redundancy); - return NULL; - } - - *redundancy = (SheepdogRedundancy) { - .type = SHEEPDOG_REDUNDANCY_TYPE_ERASURE_CODED, - .u.erasure_coded = { - .data_strips = copy, - .parity_strips = parity, - }, - }; - } - - return redundancy; -} - -static int parse_block_size_shift(BDRVSheepdogState *s, - BlockdevCreateOptionsSheepdog *opts) -{ - struct SheepdogInode *inode = &s->inode; - uint64_t object_size; - int obj_order; - - if (opts->has_object_size) { - object_size = opts->object_size; - - if ((object_size - 1) & object_size) { /* not a power of 2? */ - return -EINVAL; - } - obj_order = ctz32(object_size); - if (obj_order < 20 || obj_order > 31) { - return -EINVAL; - } - inode->block_size_shift = (uint8_t)obj_order; - } - - return 0; -} - -static int sd_co_create(BlockdevCreateOptions *options, Error **errp) -{ - BlockdevCreateOptionsSheepdog *opts = &options->u.sheepdog; - int ret = 0; - uint32_t vid = 0; - char *backing_file = NULL; - char *buf = NULL; - BDRVSheepdogState *s; - uint64_t max_vdi_size; - bool prealloc = false; - - assert(options->driver == BLOCKDEV_DRIVER_SHEEPDOG); - - deprecation_warning(); - - s = g_new0(BDRVSheepdogState, 1); - - /* Steal SocketAddress from QAPI, set NULL to prevent double free */ - s->addr = opts->location->server; - opts->location->server = NULL; - - if (strlen(opts->location->vdi) >= sizeof(s->name)) { - error_setg(errp, "'vdi' string too long"); - ret = -EINVAL; - goto out; - } - pstrcpy(s->name, sizeof(s->name), opts->location->vdi); - - s->inode.vdi_size = opts->size; - backing_file = opts->backing_file; - - if (!opts->has_preallocation) { - opts->preallocation = PREALLOC_MODE_OFF; - } - switch (opts->preallocation) { - case PREALLOC_MODE_OFF: - prealloc = false; - break; - case PREALLOC_MODE_FULL: - prealloc = true; - break; - default: - error_setg(errp, "Preallocation mode not supported for Sheepdog"); - ret = -EINVAL; - goto out; - } - - if (opts->has_redundancy) { - ret = parse_redundancy(s, opts->redundancy); - if (ret < 0) { - error_setg(errp, "Invalid redundancy mode"); - goto out; - } - } - ret = parse_block_size_shift(s, opts); - if (ret < 0) { - error_setg(errp, "Invalid object_size." - " obect_size needs to be power of 2" - " and be limited from 2^20 to 2^31"); - goto out; - } - - if (opts->has_backing_file) { - BlockBackend *blk; - BDRVSheepdogState *base; - BlockDriver *drv; - - /* Currently, only Sheepdog backing image is supported. */ - drv = bdrv_find_protocol(opts->backing_file, true, NULL); - if (!drv || strcmp(drv->protocol_name, "sheepdog") != 0) { - error_setg(errp, "backing_file must be a sheepdog image"); - ret = -EINVAL; - goto out; - } - - blk = blk_new_open(opts->backing_file, NULL, NULL, - BDRV_O_PROTOCOL, errp); - if (blk == NULL) { - ret = -EIO; - goto out; - } - - base = blk_bs(blk)->opaque; - - if (!is_snapshot(&base->inode)) { - error_setg(errp, "cannot clone from a non snapshot vdi"); - blk_unref(blk); - ret = -EINVAL; - goto out; - } - s->inode.vdi_id = base->inode.vdi_id; - blk_unref(blk); - } - - s->aio_context = qemu_get_aio_context(); - - /* if block_size_shift is not specified, get cluster default value */ - if (s->inode.block_size_shift == 0) { - SheepdogVdiReq hdr; - SheepdogClusterRsp *rsp = (SheepdogClusterRsp *)&hdr; - int fd; - unsigned int wlen = 0, rlen = 0; - - fd = connect_to_sdog(s, errp); - if (fd < 0) { - ret = fd; - goto out; - } - - memset(&hdr, 0, sizeof(hdr)); - hdr.opcode = SD_OP_GET_CLUSTER_DEFAULT; - hdr.proto_ver = SD_PROTO_VER; - - ret = do_req(fd, NULL, (SheepdogReq *)&hdr, - NULL, &wlen, &rlen); - closesocket(fd); - if (ret) { - error_setg_errno(errp, -ret, "failed to get cluster default"); - goto out; - } - if (rsp->result == SD_RES_SUCCESS) { - s->inode.block_size_shift = rsp->block_size_shift; - } else { - s->inode.block_size_shift = SD_DEFAULT_BLOCK_SIZE_SHIFT; - } - } - - max_vdi_size = (UINT64_C(1) << s->inode.block_size_shift) * MAX_DATA_OBJS; - - if (s->inode.vdi_size > max_vdi_size) { - error_setg(errp, "An image is too large." - " The maximum image size is %"PRIu64 "GB", - max_vdi_size / 1024 / 1024 / 1024); - ret = -EINVAL; - goto out; - } - - ret = do_sd_create(s, &vid, 0, errp); - if (ret) { - goto out; - } - - if (prealloc) { - ret = sd_create_prealloc(opts->location, opts->size, errp); - } -out: - g_free(backing_file); - g_free(buf); - g_free(s->addr); - g_free(s); - return ret; -} - -static int coroutine_fn sd_co_create_opts(BlockDriver *drv, - const char *filename, - QemuOpts *opts, - Error **errp) -{ - BlockdevCreateOptions *create_options = NULL; - QDict *qdict = NULL, *location_qdict; - Visitor *v; - char *redundancy = NULL; - Error *local_err = NULL; - int ret; - char *backing_fmt = NULL; - - redundancy = qemu_opt_get_del(opts, BLOCK_OPT_REDUNDANCY); - backing_fmt = qemu_opt_get_del(opts, BLOCK_OPT_BACKING_FMT); - - if (backing_fmt && strcmp(backing_fmt, "sheepdog") != 0) { - error_setg(errp, "backing_file must be a sheepdog image"); - ret = -EINVAL; - goto fail; - } - - qdict = qemu_opts_to_qdict(opts, NULL); - qdict_put_str(qdict, "driver", "sheepdog"); - - location_qdict = qdict_new(); - qdict_put(qdict, "location", location_qdict); - - sd_parse_filename(filename, location_qdict, &local_err); - if (local_err) { - error_propagate(errp, local_err); - ret = -EINVAL; - goto fail; - } - - qdict_flatten(qdict); - - /* Change legacy command line options into QMP ones */ - static const QDictRenames opt_renames[] = { - { BLOCK_OPT_BACKING_FILE, "backing-file" }, - { BLOCK_OPT_OBJECT_SIZE, "object-size" }, - { NULL, NULL }, - }; - - if (!qdict_rename_keys(qdict, opt_renames, errp)) { - ret = -EINVAL; - goto fail; - } - - /* Get the QAPI object */ - v = qobject_input_visitor_new_flat_confused(qdict, errp); - if (!v) { - ret = -EINVAL; - goto fail; - } - - visit_type_BlockdevCreateOptions(v, NULL, &create_options, errp); - visit_free(v); - if (!create_options) { - ret = -EINVAL; - goto fail; - } - - assert(create_options->driver == BLOCKDEV_DRIVER_SHEEPDOG); - create_options->u.sheepdog.size = - ROUND_UP(create_options->u.sheepdog.size, BDRV_SECTOR_SIZE); - - if (redundancy) { - create_options->u.sheepdog.has_redundancy = true; - create_options->u.sheepdog.redundancy = - parse_redundancy_str(redundancy); - if (create_options->u.sheepdog.redundancy == NULL) { - error_setg(errp, "Invalid redundancy mode"); - ret = -EINVAL; - goto fail; - } - } - - ret = sd_co_create(create_options, errp); -fail: - qapi_free_BlockdevCreateOptions(create_options); - qobject_unref(qdict); - g_free(redundancy); - g_free(backing_fmt); - return ret; -} - -static void sd_close(BlockDriverState *bs) -{ - Error *local_err = NULL; - BDRVSheepdogState *s = bs->opaque; - SheepdogVdiReq hdr; - SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr; - unsigned int wlen, rlen = 0; - int fd, ret; - - trace_sheepdog_close(s->name); - - fd = connect_to_sdog(s, &local_err); - if (fd < 0) { - error_report_err(local_err); - return; - } - - memset(&hdr, 0, sizeof(hdr)); - - hdr.opcode = SD_OP_RELEASE_VDI; - hdr.type = LOCK_TYPE_NORMAL; - hdr.base_vdi_id = s->inode.vdi_id; - wlen = strlen(s->name) + 1; - hdr.data_length = wlen; - hdr.flags = SD_FLAG_CMD_WRITE; - - ret = do_req(fd, s->bs, (SheepdogReq *)&hdr, - s->name, &wlen, &rlen); - - closesocket(fd); - - if (!ret && rsp->result != SD_RES_SUCCESS && - rsp->result != SD_RES_VDI_NOT_LOCKED) { - error_report("%s, %s", sd_strerror(rsp->result), s->name); - } - - aio_set_fd_handler(bdrv_get_aio_context(bs), s->fd, - false, NULL, NULL, NULL, NULL); - closesocket(s->fd); - qapi_free_SocketAddress(s->addr); -} - -static int64_t sd_getlength(BlockDriverState *bs) -{ - BDRVSheepdogState *s = bs->opaque; - - return s->inode.vdi_size; -} - -static int coroutine_fn sd_co_truncate(BlockDriverState *bs, int64_t offset, - bool exact, PreallocMode prealloc, - BdrvRequestFlags flags, Error **errp) -{ - BDRVSheepdogState *s = bs->opaque; - int ret, fd; - unsigned int datalen; - uint64_t max_vdi_size; - int64_t old_size = s->inode.vdi_size; - - if (prealloc != PREALLOC_MODE_OFF && prealloc != PREALLOC_MODE_FULL) { - error_setg(errp, "Unsupported preallocation mode '%s'", - PreallocMode_str(prealloc)); - return -ENOTSUP; - } - - max_vdi_size = (UINT64_C(1) << s->inode.block_size_shift) * MAX_DATA_OBJS; - if (offset < old_size) { - error_setg(errp, "shrinking is not supported"); - return -EINVAL; - } else if (offset > max_vdi_size) { - error_setg(errp, "too big image size"); - return -EINVAL; - } - - fd = connect_to_sdog(s, errp); - if (fd < 0) { - return fd; - } - - /* we don't need to update entire object */ - datalen = SD_INODE_HEADER_SIZE; - s->inode.vdi_size = offset; - ret = write_object(fd, s->bs, (char *)&s->inode, - vid_to_vdi_oid(s->inode.vdi_id), s->inode.nr_copies, - datalen, 0, false, s->cache_flags); - close(fd); - - if (ret < 0) { - error_setg_errno(errp, -ret, "failed to update an inode"); - return ret; - } - - if (prealloc == PREALLOC_MODE_FULL) { - ret = sd_prealloc(bs, old_size, offset, errp); - if (ret < 0) { - return ret; - } - } - - return 0; -} - -/* - * This function is called after writing data objects. If we need to - * update metadata, this sends a write request to the vdi object. - */ -static void coroutine_fn sd_write_done(SheepdogAIOCB *acb) -{ - BDRVSheepdogState *s = acb->s; - struct iovec iov; - AIOReq *aio_req; - uint32_t offset, data_len, mn, mx; - - mn = acb->min_dirty_data_idx; - mx = acb->max_dirty_data_idx; - if (mn <= mx) { - /* we need to update the vdi object. */ - ++acb->nr_pending; - offset = sizeof(s->inode) - sizeof(s->inode.data_vdi_id) + - mn * sizeof(s->inode.data_vdi_id[0]); - data_len = (mx - mn + 1) * sizeof(s->inode.data_vdi_id[0]); - - acb->min_dirty_data_idx = UINT32_MAX; - acb->max_dirty_data_idx = 0; - - iov.iov_base = &s->inode; - iov.iov_len = sizeof(s->inode); - aio_req = alloc_aio_req(s, acb, vid_to_vdi_oid(s->inode.vdi_id), - data_len, offset, 0, false, 0, offset); - add_aio_request(s, aio_req, &iov, 1, AIOCB_WRITE_UDATA); - if (--acb->nr_pending) { - qemu_coroutine_yield(); - } - } -} - -/* Delete current working VDI on the snapshot chain */ -static bool sd_delete(BDRVSheepdogState *s) -{ - Error *local_err = NULL; - unsigned int wlen = SD_MAX_VDI_LEN, rlen = 0; - SheepdogVdiReq hdr = { - .opcode = SD_OP_DEL_VDI, - .base_vdi_id = s->inode.vdi_id, - .data_length = wlen, - .flags = SD_FLAG_CMD_WRITE, - }; - SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr; - int fd, ret; - - fd = connect_to_sdog(s, &local_err); - if (fd < 0) { - error_report_err(local_err); - return false; - } - - ret = do_req(fd, s->bs, (SheepdogReq *)&hdr, - s->name, &wlen, &rlen); - closesocket(fd); - if (ret) { - return false; - } - switch (rsp->result) { - case SD_RES_NO_VDI: - error_report("%s was already deleted", s->name); - /* fall through */ - case SD_RES_SUCCESS: - break; - default: - error_report("%s, %s", sd_strerror(rsp->result), s->name); - return false; - } - - return true; -} - -/* - * Create a writable VDI from a snapshot - */ -static int sd_create_branch(BDRVSheepdogState *s) -{ - Error *local_err = NULL; - int ret, fd; - uint32_t vid; - char *buf; - bool deleted; - - trace_sheepdog_create_branch_snapshot(s->inode.vdi_id); - - buf = g_malloc(SD_INODE_SIZE); - - /* - * Even If deletion fails, we will just create extra snapshot based on - * the working VDI which was supposed to be deleted. So no need to - * false bail out. - */ - deleted = sd_delete(s); - ret = do_sd_create(s, &vid, !deleted, &local_err); - if (ret) { - error_report_err(local_err); - goto out; - } - - trace_sheepdog_create_branch_created(vid); - - fd = connect_to_sdog(s, &local_err); - if (fd < 0) { - error_report_err(local_err); - ret = fd; - goto out; - } - - ret = read_object(fd, s->bs, buf, vid_to_vdi_oid(vid), - s->inode.nr_copies, SD_INODE_SIZE, 0, s->cache_flags); - - closesocket(fd); - - if (ret < 0) { - goto out; - } - - memcpy(&s->inode, buf, sizeof(s->inode)); - - s->is_snapshot = false; - ret = 0; - trace_sheepdog_create_branch_new(s->inode.vdi_id); - -out: - g_free(buf); - - return ret; -} - -/* - * Send I/O requests to the server. - * - * This function sends requests to the server, links the requests to - * the inflight_list in BDRVSheepdogState, and exits without - * waiting the response. The responses are received in the - * `aio_read_response' function which is called from the main loop as - * a fd handler. - * - * Returns 1 when we need to wait a response, 0 when there is no sent - * request and -errno in error cases. - */ -static void coroutine_fn sd_co_rw_vector(SheepdogAIOCB *acb) -{ - int ret = 0; - unsigned long len, done = 0, total = acb->nb_sectors * BDRV_SECTOR_SIZE; - unsigned long idx; - uint32_t object_size; - uint64_t oid; - uint64_t offset; - BDRVSheepdogState *s = acb->s; - SheepdogInode *inode = &s->inode; - AIOReq *aio_req; - - if (acb->aiocb_type == AIOCB_WRITE_UDATA && s->is_snapshot) { - /* - * In the case we open the snapshot VDI, Sheepdog creates the - * writable VDI when we do a write operation first. - */ - ret = sd_create_branch(s); - if (ret) { - acb->ret = -EIO; - return; - } - } - - object_size = (UINT32_C(1) << inode->block_size_shift); - idx = acb->sector_num * BDRV_SECTOR_SIZE / object_size; - offset = (acb->sector_num * BDRV_SECTOR_SIZE) % object_size; - - /* - * Make sure we don't free the aiocb before we are done with all requests. - * This additional reference is dropped at the end of this function. - */ - acb->nr_pending++; - - while (done != total) { - uint8_t flags = 0; - uint64_t old_oid = 0; - bool create = false; - - oid = vid_to_data_oid(inode->data_vdi_id[idx], idx); - - len = MIN(total - done, object_size - offset); - - switch (acb->aiocb_type) { - case AIOCB_READ_UDATA: - if (!inode->data_vdi_id[idx]) { - qemu_iovec_memset(acb->qiov, done, 0, len); - goto done; - } - break; - case AIOCB_WRITE_UDATA: - if (!inode->data_vdi_id[idx]) { - create = true; - } else if (!is_data_obj_writable(inode, idx)) { - /* Copy-On-Write */ - create = true; - old_oid = oid; - flags = SD_FLAG_CMD_COW; - } - break; - case AIOCB_DISCARD_OBJ: - /* - * We discard the object only when the whole object is - * 1) allocated 2) trimmed. Otherwise, simply skip it. - */ - if (len != object_size || inode->data_vdi_id[idx] == 0) { - goto done; - } - break; - default: - break; - } - - if (create) { - trace_sheepdog_co_rw_vector_update(inode->vdi_id, oid, - vid_to_data_oid(inode->data_vdi_id[idx], idx), - idx); - oid = vid_to_data_oid(inode->vdi_id, idx); - trace_sheepdog_co_rw_vector_new(oid); - } - - aio_req = alloc_aio_req(s, acb, oid, len, offset, flags, create, - old_oid, - acb->aiocb_type == AIOCB_DISCARD_OBJ ? - 0 : done); - add_aio_request(s, aio_req, acb->qiov->iov, acb->qiov->niov, - acb->aiocb_type); - done: - offset = 0; - idx++; - done += len; - } - if (--acb->nr_pending) { - qemu_coroutine_yield(); - } -} - -static void sd_aio_complete(SheepdogAIOCB *acb) -{ - BDRVSheepdogState *s; - if (acb->aiocb_type == AIOCB_FLUSH_CACHE) { - return; - } - - s = acb->s; - qemu_co_mutex_lock(&s->queue_lock); - QLIST_REMOVE(acb, aiocb_siblings); - qemu_co_queue_restart_all(&s->overlapping_queue); - qemu_co_mutex_unlock(&s->queue_lock); -} - -static coroutine_fn int sd_co_writev(BlockDriverState *bs, int64_t sector_num, - int nb_sectors, QEMUIOVector *qiov, - int flags) -{ - SheepdogAIOCB acb; - int ret; - int64_t offset = (sector_num + nb_sectors) * BDRV_SECTOR_SIZE; - BDRVSheepdogState *s = bs->opaque; - - assert(!flags); - if (offset > s->inode.vdi_size) { - ret = sd_co_truncate(bs, offset, false, PREALLOC_MODE_OFF, 0, NULL); - if (ret < 0) { - return ret; - } - } - - sd_aio_setup(&acb, s, qiov, sector_num, nb_sectors, AIOCB_WRITE_UDATA); - sd_co_rw_vector(&acb); - sd_write_done(&acb); - sd_aio_complete(&acb); - - return acb.ret; -} - -static coroutine_fn int sd_co_readv(BlockDriverState *bs, int64_t sector_num, - int nb_sectors, QEMUIOVector *qiov) -{ - SheepdogAIOCB acb; - BDRVSheepdogState *s = bs->opaque; - - sd_aio_setup(&acb, s, qiov, sector_num, nb_sectors, AIOCB_READ_UDATA); - sd_co_rw_vector(&acb); - sd_aio_complete(&acb); - - return acb.ret; -} - -static int coroutine_fn sd_co_flush_to_disk(BlockDriverState *bs) -{ - BDRVSheepdogState *s = bs->opaque; - SheepdogAIOCB acb; - AIOReq *aio_req; - - if (s->cache_flags != SD_FLAG_CMD_CACHE) { - return 0; - } - - sd_aio_setup(&acb, s, NULL, 0, 0, AIOCB_FLUSH_CACHE); - - acb.nr_pending++; - aio_req = alloc_aio_req(s, &acb, vid_to_vdi_oid(s->inode.vdi_id), - 0, 0, 0, false, 0, 0); - add_aio_request(s, aio_req, NULL, 0, acb.aiocb_type); - - if (--acb.nr_pending) { - qemu_coroutine_yield(); - } - - sd_aio_complete(&acb); - return acb.ret; -} - -static int sd_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info) -{ - Error *local_err = NULL; - BDRVSheepdogState *s = bs->opaque; - int ret, fd; - uint32_t new_vid; - SheepdogInode *inode; - unsigned int datalen; - - trace_sheepdog_snapshot_create_info(sn_info->name, sn_info->id_str, s->name, - sn_info->vm_state_size, s->is_snapshot); - - if (s->is_snapshot) { - error_report("You can't create a snapshot of a snapshot VDI, " - "%s (%" PRIu32 ").", s->name, s->inode.vdi_id); - - return -EINVAL; - } - - trace_sheepdog_snapshot_create(sn_info->name, sn_info->id_str); - - s->inode.vm_state_size = sn_info->vm_state_size; - s->inode.vm_clock_nsec = sn_info->vm_clock_nsec; - /* It appears that inode.tag does not require a NUL terminator, - * which means this use of strncpy is ok. - */ - strncpy(s->inode.tag, sn_info->name, sizeof(s->inode.tag)); - /* we don't need to update entire object */ - datalen = SD_INODE_HEADER_SIZE; - inode = g_malloc(datalen); - - /* refresh inode. */ - fd = connect_to_sdog(s, &local_err); - if (fd < 0) { - error_report_err(local_err); - ret = fd; - goto cleanup; - } - - ret = write_object(fd, s->bs, (char *)&s->inode, - vid_to_vdi_oid(s->inode.vdi_id), s->inode.nr_copies, - datalen, 0, false, s->cache_flags); - if (ret < 0) { - error_report("failed to write snapshot's inode."); - goto cleanup; - } - - ret = do_sd_create(s, &new_vid, 1, &local_err); - if (ret < 0) { - error_reportf_err(local_err, - "failed to create inode for snapshot: "); - goto cleanup; - } - - ret = read_object(fd, s->bs, (char *)inode, - vid_to_vdi_oid(new_vid), s->inode.nr_copies, datalen, 0, - s->cache_flags); - - if (ret < 0) { - error_report("failed to read new inode info. %s", strerror(errno)); - goto cleanup; - } - - memcpy(&s->inode, inode, datalen); - trace_sheepdog_snapshot_create_inode(s->inode.name, s->inode.snap_id, - s->inode.vdi_id); - -cleanup: - g_free(inode); - closesocket(fd); - return ret; -} - -/* - * We implement rollback(loadvm) operation to the specified snapshot by - * 1) switch to the snapshot - * 2) rely on sd_create_branch to delete working VDI and - * 3) create a new working VDI based on the specified snapshot - */ -static int sd_snapshot_goto(BlockDriverState *bs, const char *snapshot_id) -{ - BDRVSheepdogState *s = bs->opaque; - BDRVSheepdogState *old_s; - char tag[SD_MAX_VDI_TAG_LEN]; - uint32_t snapid = 0; - int ret; - - if (!sd_parse_snapid_or_tag(snapshot_id, &snapid, tag)) { - return -EINVAL; - } - - old_s = g_new(BDRVSheepdogState, 1); - - memcpy(old_s, s, sizeof(BDRVSheepdogState)); - - ret = reload_inode(s, snapid, tag); - if (ret) { - goto out; - } - - ret = sd_create_branch(s); - if (ret) { - goto out; - } - - g_free(old_s); - - return 0; -out: - /* recover bdrv_sd_state */ - memcpy(s, old_s, sizeof(BDRVSheepdogState)); - g_free(old_s); - - error_report("failed to open. recover old bdrv_sd_state."); - - return ret; -} - -#define NR_BATCHED_DISCARD 128 - -static int remove_objects(BDRVSheepdogState *s, Error **errp) -{ - int fd, i = 0, nr_objs = 0; - int ret; - SheepdogInode *inode = &s->inode; - - fd = connect_to_sdog(s, errp); - if (fd < 0) { - return fd; - } - - nr_objs = count_data_objs(inode); - while (i < nr_objs) { - int start_idx, nr_filled_idx; - - while (i < nr_objs && !inode->data_vdi_id[i]) { - i++; - } - start_idx = i; - - nr_filled_idx = 0; - while (i < nr_objs && nr_filled_idx < NR_BATCHED_DISCARD) { - if (inode->data_vdi_id[i]) { - inode->data_vdi_id[i] = 0; - nr_filled_idx++; - } - - i++; - } - - ret = write_object(fd, s->bs, - (char *)&inode->data_vdi_id[start_idx], - vid_to_vdi_oid(s->inode.vdi_id), inode->nr_copies, - (i - start_idx) * sizeof(uint32_t), - offsetof(struct SheepdogInode, - data_vdi_id[start_idx]), - false, s->cache_flags); - if (ret < 0) { - error_setg(errp, "Failed to discard snapshot inode"); - goto out; - } - } - - ret = 0; -out: - closesocket(fd); - return ret; -} - -static int sd_snapshot_delete(BlockDriverState *bs, - const char *snapshot_id, - const char *name, - Error **errp) -{ - /* - * FIXME should delete the snapshot matching both @snapshot_id and - * @name, but @name not used here - */ - unsigned long snap_id = 0; - char snap_tag[SD_MAX_VDI_TAG_LEN]; - int fd, ret; - char buf[SD_MAX_VDI_LEN + SD_MAX_VDI_TAG_LEN]; - BDRVSheepdogState *s = bs->opaque; - unsigned int wlen = SD_MAX_VDI_LEN + SD_MAX_VDI_TAG_LEN, rlen = 0; - uint32_t vid; - SheepdogVdiReq hdr = { - .opcode = SD_OP_DEL_VDI, - .data_length = wlen, - .flags = SD_FLAG_CMD_WRITE, - }; - SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr; - - ret = remove_objects(s, errp); - if (ret) { - return ret; - } - - memset(buf, 0, sizeof(buf)); - memset(snap_tag, 0, sizeof(snap_tag)); - pstrcpy(buf, SD_MAX_VDI_LEN, s->name); - /* TODO Use sd_parse_snapid() once this mess is cleaned up */ - ret = qemu_strtoul(snapshot_id, NULL, 10, &snap_id); - if (ret || snap_id > UINT32_MAX) { - /* - * FIXME Since qemu_strtoul() returns -EINVAL when - * @snapshot_id is null, @snapshot_id is mandatory. Correct - * would be to require at least one of @snapshot_id and @name. - */ - error_setg(errp, "Invalid snapshot ID: %s", - snapshot_id ? snapshot_id : "<null>"); - return -EINVAL; - } - - if (snap_id) { - hdr.snapid = (uint32_t) snap_id; - } else { - /* FIXME I suspect we should use @name here */ - /* FIXME don't truncate silently */ - pstrcpy(snap_tag, sizeof(snap_tag), snapshot_id); - pstrcpy(buf + SD_MAX_VDI_LEN, SD_MAX_VDI_TAG_LEN, snap_tag); - } - - ret = find_vdi_name(s, s->name, snap_id, snap_tag, &vid, true, errp); - if (ret) { - return ret; - } - - fd = connect_to_sdog(s, errp); - if (fd < 0) { - return fd; - } - - ret = do_req(fd, s->bs, (SheepdogReq *)&hdr, - buf, &wlen, &rlen); - closesocket(fd); - if (ret) { - error_setg_errno(errp, -ret, "Couldn't send request to server"); - return ret; - } - - switch (rsp->result) { - case SD_RES_NO_VDI: - error_setg(errp, "Can't find the snapshot"); - return -ENOENT; - case SD_RES_SUCCESS: - break; - default: - error_setg(errp, "%s", sd_strerror(rsp->result)); - return -EIO; - } - - return 0; -} - -static int sd_snapshot_list(BlockDriverState *bs, QEMUSnapshotInfo **psn_tab) -{ - Error *local_err = NULL; - BDRVSheepdogState *s = bs->opaque; - SheepdogReq req; - int fd, nr = 1024, ret, max = BITS_TO_LONGS(SD_NR_VDIS) * sizeof(long); - QEMUSnapshotInfo *sn_tab = NULL; - unsigned wlen, rlen; - int found = 0; - SheepdogInode *inode; - unsigned long *vdi_inuse; - unsigned int start_nr; - uint64_t hval; - uint32_t vid; - - vdi_inuse = g_malloc(max); - inode = g_malloc(SD_INODE_HEADER_SIZE); - - fd = connect_to_sdog(s, &local_err); - if (fd < 0) { - error_report_err(local_err); - ret = fd; - goto out; - } - - rlen = max; - wlen = 0; - - memset(&req, 0, sizeof(req)); - - req.opcode = SD_OP_READ_VDIS; - req.data_length = max; - - ret = do_req(fd, s->bs, &req, vdi_inuse, &wlen, &rlen); - - closesocket(fd); - if (ret) { - goto out; - } - - sn_tab = g_new0(QEMUSnapshotInfo, nr); - - /* calculate a vdi id with hash function */ - hval = fnv_64a_buf(s->name, strlen(s->name), FNV1A_64_INIT); - start_nr = hval & (SD_NR_VDIS - 1); - - fd = connect_to_sdog(s, &local_err); - if (fd < 0) { - error_report_err(local_err); - ret = fd; - goto out; - } - - for (vid = start_nr; found < nr; vid = (vid + 1) % SD_NR_VDIS) { - if (!test_bit(vid, vdi_inuse)) { - break; - } - - /* we don't need to read entire object */ - ret = read_object(fd, s->bs, (char *)inode, - vid_to_vdi_oid(vid), - 0, SD_INODE_HEADER_SIZE, 0, - s->cache_flags); - - if (ret) { - continue; - } - - if (!strcmp(inode->name, s->name) && is_snapshot(inode)) { - sn_tab[found].date_sec = inode->snap_ctime >> 32; - sn_tab[found].date_nsec = inode->snap_ctime & 0xffffffff; - sn_tab[found].vm_state_size = inode->vm_state_size; - sn_tab[found].vm_clock_nsec = inode->vm_clock_nsec; - - snprintf(sn_tab[found].id_str, sizeof(sn_tab[found].id_str), - "%" PRIu32, inode->snap_id); - pstrcpy(sn_tab[found].name, - MIN(sizeof(sn_tab[found].name), sizeof(inode->tag)), - inode->tag); - found++; - } - } - - closesocket(fd); -out: - *psn_tab = sn_tab; - - g_free(vdi_inuse); - g_free(inode); - - if (ret < 0) { - return ret; - } - - return found; -} - -static int do_load_save_vmstate(BDRVSheepdogState *s, uint8_t *data, - int64_t pos, int size, int load) -{ - Error *local_err = NULL; - bool create; - int fd, ret = 0, remaining = size; - unsigned int data_len; - uint64_t vmstate_oid; - uint64_t offset; - uint32_t vdi_index; - uint32_t vdi_id = load ? s->inode.parent_vdi_id : s->inode.vdi_id; - uint32_t object_size = (UINT32_C(1) << s->inode.block_size_shift); - - fd = connect_to_sdog(s, &local_err); - if (fd < 0) { - error_report_err(local_err); - return fd; - } - - while (remaining) { - vdi_index = pos / object_size; - offset = pos % object_size; - - data_len = MIN(remaining, object_size - offset); - - vmstate_oid = vid_to_vmstate_oid(vdi_id, vdi_index); - - create = (offset == 0); - if (load) { - ret = read_object(fd, s->bs, (char *)data, vmstate_oid, - s->inode.nr_copies, data_len, offset, - s->cache_flags); - } else { - ret = write_object(fd, s->bs, (char *)data, vmstate_oid, - s->inode.nr_copies, data_len, offset, create, - s->cache_flags); - } - - if (ret < 0) { - error_report("failed to save vmstate %s", strerror(errno)); - goto cleanup; - } - - pos += data_len; - data += data_len; - remaining -= data_len; - } - ret = size; -cleanup: - closesocket(fd); - return ret; -} - -static int sd_save_vmstate(BlockDriverState *bs, QEMUIOVector *qiov, - int64_t pos) -{ - BDRVSheepdogState *s = bs->opaque; - void *buf; - int ret; - - buf = qemu_blockalign(bs, qiov->size); - qemu_iovec_to_buf(qiov, 0, buf, qiov->size); - ret = do_load_save_vmstate(s, (uint8_t *) buf, pos, qiov->size, 0); - qemu_vfree(buf); - - return ret; -} - -static int sd_load_vmstate(BlockDriverState *bs, QEMUIOVector *qiov, - int64_t pos) -{ - BDRVSheepdogState *s = bs->opaque; - void *buf; - int ret; - - buf = qemu_blockalign(bs, qiov->size); - ret = do_load_save_vmstate(s, buf, pos, qiov->size, 1); - qemu_iovec_from_buf(qiov, 0, buf, qiov->size); - qemu_vfree(buf); - - return ret; -} - - -static coroutine_fn int sd_co_pdiscard(BlockDriverState *bs, int64_t offset, - int bytes) -{ - SheepdogAIOCB acb; - BDRVSheepdogState *s = bs->opaque; - QEMUIOVector discard_iov; - struct iovec iov; - uint32_t zero = 0; - - if (!s->discard_supported) { - return 0; - } - - memset(&discard_iov, 0, sizeof(discard_iov)); - memset(&iov, 0, sizeof(iov)); - iov.iov_base = &zero; - iov.iov_len = sizeof(zero); - discard_iov.iov = &iov; - discard_iov.niov = 1; - if (!QEMU_IS_ALIGNED(offset | bytes, BDRV_SECTOR_SIZE)) { - return -ENOTSUP; - } - sd_aio_setup(&acb, s, &discard_iov, offset >> BDRV_SECTOR_BITS, - bytes >> BDRV_SECTOR_BITS, AIOCB_DISCARD_OBJ); - sd_co_rw_vector(&acb); - sd_aio_complete(&acb); - - return acb.ret; -} - -static coroutine_fn int -sd_co_block_status(BlockDriverState *bs, bool want_zero, int64_t offset, - int64_t bytes, int64_t *pnum, int64_t *map, - BlockDriverState **file) -{ - BDRVSheepdogState *s = bs->opaque; - SheepdogInode *inode = &s->inode; - uint32_t object_size = (UINT32_C(1) << inode->block_size_shift); - unsigned long start = offset / object_size, - end = DIV_ROUND_UP(offset + bytes, object_size); - unsigned long idx; - *map = offset; - int ret = BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID; - - for (idx = start; idx < end; idx++) { - if (inode->data_vdi_id[idx] == 0) { - break; - } - } - if (idx == start) { - /* Get the longest length of unallocated sectors */ - ret = 0; - for (idx = start + 1; idx < end; idx++) { - if (inode->data_vdi_id[idx] != 0) { - break; - } - } - } - - *pnum = (idx - start) * object_size; - if (*pnum > bytes) { - *pnum = bytes; - } - if (ret > 0 && ret & BDRV_BLOCK_OFFSET_VALID) { - *file = bs; - } - return ret; -} - -static int64_t sd_get_allocated_file_size(BlockDriverState *bs) -{ - BDRVSheepdogState *s = bs->opaque; - SheepdogInode *inode = &s->inode; - uint32_t object_size = (UINT32_C(1) << inode->block_size_shift); - unsigned long i, last = DIV_ROUND_UP(inode->vdi_size, object_size); - uint64_t size = 0; - - for (i = 0; i < last; i++) { - if (inode->data_vdi_id[i] == 0) { - continue; - } - size += object_size; - } - return size; -} - -static QemuOptsList sd_create_opts = { - .name = "sheepdog-create-opts", - .head = QTAILQ_HEAD_INITIALIZER(sd_create_opts.head), - .desc = { - { - .name = BLOCK_OPT_SIZE, - .type = QEMU_OPT_SIZE, - .help = "Virtual disk size" - }, - { - .name = BLOCK_OPT_BACKING_FILE, - .type = QEMU_OPT_STRING, - .help = "File name of a base image" - }, - { - .name = BLOCK_OPT_BACKING_FMT, - .type = QEMU_OPT_STRING, - .help = "Must be 'sheepdog' if present", - }, - { - .name = BLOCK_OPT_PREALLOC, - .type = QEMU_OPT_STRING, - .help = "Preallocation mode (allowed values: off, full)" - }, - { - .name = BLOCK_OPT_REDUNDANCY, - .type = QEMU_OPT_STRING, - .help = "Redundancy of the image" - }, - { - .name = BLOCK_OPT_OBJECT_SIZE, - .type = QEMU_OPT_SIZE, - .help = "Object size of the image" - }, - { /* end of list */ } - } -}; - -static const char *const sd_strong_runtime_opts[] = { - "vdi", - "snap-id", - "tag", - "server.", - - NULL -}; - -static BlockDriver bdrv_sheepdog = { - .format_name = "sheepdog", - .protocol_name = "sheepdog", - .instance_size = sizeof(BDRVSheepdogState), - .bdrv_parse_filename = sd_parse_filename, - .bdrv_file_open = sd_open, - .bdrv_reopen_prepare = sd_reopen_prepare, - .bdrv_reopen_commit = sd_reopen_commit, - .bdrv_reopen_abort = sd_reopen_abort, - .bdrv_close = sd_close, - .bdrv_co_create = sd_co_create, - .bdrv_co_create_opts = sd_co_create_opts, - .bdrv_has_zero_init = bdrv_has_zero_init_1, - .bdrv_getlength = sd_getlength, - .bdrv_get_allocated_file_size = sd_get_allocated_file_size, - .bdrv_co_truncate = sd_co_truncate, - - .bdrv_co_readv = sd_co_readv, - .bdrv_co_writev = sd_co_writev, - .bdrv_co_flush_to_disk = sd_co_flush_to_disk, - .bdrv_co_pdiscard = sd_co_pdiscard, - .bdrv_co_block_status = sd_co_block_status, - - .bdrv_snapshot_create = sd_snapshot_create, - .bdrv_snapshot_goto = sd_snapshot_goto, - .bdrv_snapshot_delete = sd_snapshot_delete, - .bdrv_snapshot_list = sd_snapshot_list, - - .bdrv_save_vmstate = sd_save_vmstate, - .bdrv_load_vmstate = sd_load_vmstate, - - .bdrv_detach_aio_context = sd_detach_aio_context, - .bdrv_attach_aio_context = sd_attach_aio_context, - - .create_opts = &sd_create_opts, - .strong_runtime_opts = sd_strong_runtime_opts, -}; - -static BlockDriver bdrv_sheepdog_tcp = { - .format_name = "sheepdog", - .protocol_name = "sheepdog+tcp", - .instance_size = sizeof(BDRVSheepdogState), - .bdrv_parse_filename = sd_parse_filename, - .bdrv_file_open = sd_open, - .bdrv_reopen_prepare = sd_reopen_prepare, - .bdrv_reopen_commit = sd_reopen_commit, - .bdrv_reopen_abort = sd_reopen_abort, - .bdrv_close = sd_close, - .bdrv_co_create = sd_co_create, - .bdrv_co_create_opts = sd_co_create_opts, - .bdrv_has_zero_init = bdrv_has_zero_init_1, - .bdrv_getlength = sd_getlength, - .bdrv_get_allocated_file_size = sd_get_allocated_file_size, - .bdrv_co_truncate = sd_co_truncate, - - .bdrv_co_readv = sd_co_readv, - .bdrv_co_writev = sd_co_writev, - .bdrv_co_flush_to_disk = sd_co_flush_to_disk, - .bdrv_co_pdiscard = sd_co_pdiscard, - .bdrv_co_block_status = sd_co_block_status, - - .bdrv_snapshot_create = sd_snapshot_create, - .bdrv_snapshot_goto = sd_snapshot_goto, - .bdrv_snapshot_delete = sd_snapshot_delete, - .bdrv_snapshot_list = sd_snapshot_list, - - .bdrv_save_vmstate = sd_save_vmstate, - .bdrv_load_vmstate = sd_load_vmstate, - - .bdrv_detach_aio_context = sd_detach_aio_context, - .bdrv_attach_aio_context = sd_attach_aio_context, - - .create_opts = &sd_create_opts, - .strong_runtime_opts = sd_strong_runtime_opts, -}; - -static BlockDriver bdrv_sheepdog_unix = { - .format_name = "sheepdog", - .protocol_name = "sheepdog+unix", - .instance_size = sizeof(BDRVSheepdogState), - .bdrv_parse_filename = sd_parse_filename, - .bdrv_file_open = sd_open, - .bdrv_reopen_prepare = sd_reopen_prepare, - .bdrv_reopen_commit = sd_reopen_commit, - .bdrv_reopen_abort = sd_reopen_abort, - .bdrv_close = sd_close, - .bdrv_co_create = sd_co_create, - .bdrv_co_create_opts = sd_co_create_opts, - .bdrv_has_zero_init = bdrv_has_zero_init_1, - .bdrv_getlength = sd_getlength, - .bdrv_get_allocated_file_size = sd_get_allocated_file_size, - .bdrv_co_truncate = sd_co_truncate, - - .bdrv_co_readv = sd_co_readv, - .bdrv_co_writev = sd_co_writev, - .bdrv_co_flush_to_disk = sd_co_flush_to_disk, - .bdrv_co_pdiscard = sd_co_pdiscard, - .bdrv_co_block_status = sd_co_block_status, - - .bdrv_snapshot_create = sd_snapshot_create, - .bdrv_snapshot_goto = sd_snapshot_goto, - .bdrv_snapshot_delete = sd_snapshot_delete, - .bdrv_snapshot_list = sd_snapshot_list, - - .bdrv_save_vmstate = sd_save_vmstate, - .bdrv_load_vmstate = sd_load_vmstate, - - .bdrv_detach_aio_context = sd_detach_aio_context, - .bdrv_attach_aio_context = sd_attach_aio_context, - - .create_opts = &sd_create_opts, - .strong_runtime_opts = sd_strong_runtime_opts, -}; - -static void bdrv_sheepdog_init(void) -{ - bdrv_register(&bdrv_sheepdog); - bdrv_register(&bdrv_sheepdog_tcp); - bdrv_register(&bdrv_sheepdog_unix); -} -block_init(bdrv_sheepdog_init); diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml index 52d65d6c04..aead19ae2c 100644 --- a/.gitlab-ci.yml +++ b/.gitlab-ci.yml @@ -338,7 +338,6 @@ build-disabled: --disable-replication --disable-sdl --disable-seccomp - --disable-sheepdog --disable-slirp --disable-smartcard --disable-snappy diff --git a/MAINTAINERS b/MAINTAINERS index 4c05ff8bba..cb7ef272a2 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3056,12 +3056,6 @@ L: qemu-block(a)nongnu.org S: Supported F: block/rbd.c -Sheepdog -M: Liu Yuan <namei.unix(a)gmail.com> -L: qemu-block(a)nongnu.org -S: Odd Fixes -F: block/sheepdog.c - VHDX M: Jeff Cody <codyprime(a)gmail.com> L: qemu-block(a)nongnu.org diff --git a/block/meson.build b/block/meson.build index d21990ec95..e687c54dbc 100644 --- a/block/meson.build +++ b/block/meson.build @@ -64,7 +64,6 @@ block_ss.add(when: 'CONFIG_POSIX', if_true: [files('file-posix.c'), coref, iokit block_ss.add(when: libiscsi, if_true: files('iscsi-opts.c')) block_ss.add(when: 'CONFIG_LINUX', if_true: files('nvme.c')) block_ss.add(when: 'CONFIG_REPLICATION', if_true: files('replication.c')) -block_ss.add(when: 'CONFIG_SHEEPDOG', if_true: files('sheepdog.c')) block_ss.add(when: ['CONFIG_LINUX_AIO', libaio], if_true: files('linux-aio.c')) block_ss.add(when: ['CONFIG_LINUX_IO_URING', linux_io_uring], if_true: files('io_uring.c')) diff --git a/block/trace-events b/block/trace-events index 1a12d634e2..31062ed437 100644 --- a/block/trace-events +++ b/block/trace-events @@ -207,19 +207,5 @@ file_FindEjectableOpticalMedia(const char *media) "Matching using %s" file_setup_cdrom(const char *partition) "Using %s as optical disc" file_hdev_is_sg(int type, int version) "SG device found: type=%d, version=%d" -# sheepdog.c -sheepdog_reconnect_to_sdog(void) "Wait for connection to be established" -sheepdog_aio_read_response(void) "disable cache since the server doesn't support it" -sheepdog_open(uint32_t vid) "0x%" PRIx32 " snapshot inode was open" -sheepdog_close(const char *name) "%s" -sheepdog_create_branch_snapshot(uint32_t vdi) "0x%" PRIx32 " is snapshot" -sheepdog_create_branch_created(uint32_t vdi) "0x%" PRIx32 " is created" -sheepdog_create_branch_new(uint32_t vdi) "0x%" PRIx32 " was newly created" -sheepdog_co_rw_vector_update(uint32_t vdi, uint64_t oid, uint64_t data, long idx) "update ino (%" PRIu32 ") %" PRIu64 " %" PRIu64 " %ld" -sheepdog_co_rw_vector_new(uint64_t oid) "new oid 0x%" PRIx64 -sheepdog_snapshot_create_info(const char *sn_name, const char *id, const char *name, int64_t size, int is_snapshot) "sn_info: name %s id_str %s s: name %s vm_state_size %" PRId64 " " "is_snapshot %d" -sheepdog_snapshot_create(const char *sn_name, const char *id) "%s %s" -sheepdog_snapshot_create_inode(const char *name, uint32_t snap, uint32_t vdi) "s->inode: name %s snap_id 0x%" PRIx32 " vdi 0x%" PRIx32 - # ssh.c sftp_error(const char *op, const char *ssh_err, int ssh_err_code, int sftp_err_code) "%s failed: %s (libssh error code: %d, sftp error code: %d)" diff --git a/tests/qemu-iotests/005 b/tests/qemu-iotests/005 index 40e64a9a8f..ba377543b0 100755 --- a/tests/qemu-iotests/005 +++ b/tests/qemu-iotests/005 @@ -52,11 +52,6 @@ if [ "$IMGFMT" = "vpc" ]; then _notrun "image format $IMGFMT does not support large image sizes" fi -# sheepdog image is limited to 4TB, so we can't test it here -if [ "$IMGPROTO" = "sheepdog" ]; then - _notrun "image protocol $IMGPROTO does not support large image sizes" -fi - # Sanity check: For raw, we require a file system that permits the creation # of a HUGE (but very sparse) file. Check we can create it before continuing. if [ "$IMGFMT" = "raw" ]; then diff --git a/tests/qemu-iotests/025 b/tests/qemu-iotests/025 index da77ed3154..80686a30d5 100755 --- a/tests/qemu-iotests/025 +++ b/tests/qemu-iotests/025 @@ -39,7 +39,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15 . ./common.pattern _supported_fmt raw qcow2 qed luks -_supported_proto file sheepdog rbd nfs fuse +_supported_proto file rbd nfs fuse echo "=== Creating image" echo diff --git a/tests/qemu-iotests/check b/tests/qemu-iotests/check index d1c87ceaf1..08f51366f1 100755 --- a/tests/qemu-iotests/check +++ b/tests/qemu-iotests/check @@ -65,8 +65,7 @@ def make_argparser() -> argparse.ArgumentParser: mg.add_argument('-' + fmt, dest='imgfmt', action='store_const', const=fmt, help=f'test {fmt}') - protocol_list = ['file', 'rbd', 'sheepdog', 'nbd', 'ssh', 'nfs', - 'fuse'] + protocol_list = ['file', 'rbd', 'nbd', 'ssh', 'nfs', 'fuse'] g_prt = p.add_argument_group( ' image protocol options', 'The following options set the IMGPROTO environment variable. ' diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc index 7f49c9716d..cbbf6d7c7f 100644 --- a/tests/qemu-iotests/common.rc +++ b/tests/qemu-iotests/common.rc @@ -641,10 +641,6 @@ _cleanup_test_img() rbd --no-progress rm "$TEST_DIR/t.$IMGFMT" > /dev/null ;; - sheepdog) - collie vdi delete "$TEST_DIR/t.$IMGFMT" - ;; - esac } -- 2.26.3

2 1

[PATCH 0/2] meson: Declare GLIB_VERSION_* macros at configure
by Michal Privoznik 03 May '21

03 May '21

I was playing around with bumped glib version and found couple of issues. Michal Prívozník (2): qemu_domainjob: Drop 'const' from strings in _qemuDomainJobObj meson: Declare GLIB_VERSION_* macros at configure config.h | 10 ---------- meson.build | 7 +++++++ src/qemu/qemu_domainjob.h | 6 +++--- 3 files changed, 10 insertions(+), 13 deletions(-) -- 2.26.3

3 5

[libvirt PATCH] meson: Fix compatibility with Meson 0.58
by Andrea Bolognani 03 May '21

03 May '21

Builds failed with tests/meson.build:690:0: ERROR: List item must be one of <class 'str'>, not <class 'list'> before this change. https://gitlab.com/libvirt/libvirt/-/issues/158 Signed-off-by: Andrea Bolognani <abologna(a)redhat.com> --- Test pipeline: https://gitlab.com/abologna/libvirt/-/pipelines/296072074 (hasn't completed yet) Should we consider this for 7.3.0? It would be great if people who have a very new version of Meson would be able to build the new version of libvirt out of the box. tests/meson.build | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/tests/meson.build b/tests/meson.build index 05c3e90195..9900983d0c 100644 --- a/tests/meson.build +++ b/tests/meson.build @@ -687,12 +687,12 @@ foreach name : test_scripts test(name, script, env: tests_env) endforeach +testenv = runutf8 +testenv += 'VIR_TEST_FILE_ACCESS=1' + add_test_setup( 'access', - env: [ - 'VIR_TEST_FILE_ACCESS=1', - runutf8, - ], + env: testenv, exe_wrapper: [ python3_prog, check_file_access_prog.path() ], ) -- 2.26.3

3 2

[PATCH for 7.3] conf: Fix heap corruption when hot-adding a lease
by Peter Krempa 03 May '21

03 May '21

Commit 28a86993162f7d2f ( v6.9.0-179-g28a8699316 ) incorrectly replaced VIR_EXPAND_N by g_renew. VIR_EXPAND_N has these two extra effects apart from reallocating memory: 1) The newly allocated memory is zeroed out 2) The number of elements in the array which is passed to VIR_EXPAND_N is increased. This comes into play when used with virDomainLeaseInsertPreAlloced, which expects that the array element count already includes the space for the added 'lease', by plainly just assigning to 'leases[nleases - 1' Since g_renew does not increase the number of elements in the array any existing code which calls virDomainLeaseInsertPreAlloced thus either overwrites a lease definition or corrupts the heap if there are no leases to start with. To preserve existing functionality we revert the code back to using VIR_EXPAND_N which at this point doesn't return any value, so other commits don't need to be reverted. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1953577 Signed-off-by: Peter Krempa <pkrempa(a)redhat.com> --- src/conf/domain_conf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 9d98f487ea..84570c001c 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -16837,7 +16837,7 @@ int virDomainLeaseIndex(virDomainDef *def, void virDomainLeaseInsertPreAlloc(virDomainDef *def) { - def->leases = g_renew(virDomainLeaseDef *, def->leases, def->nleases + 1); + VIR_EXPAND_N(def->leases, def->nleases, 1); } void virDomainLeaseInsert(virDomainDef *def, virDomainLeaseDef *lease) -- 2.30.2

2 1

[libvirt PATCH] virnetdevbridge: Ignore EEXIST when adding an entry to fdb
by Jiri Denemark 03 May '21

03 May '21

When updating entries in a bridge forwarding database (i.e., when macTableManager='libvirt' is configured for the bridge), we may end up in a situation when the entry we want to add is already present. Let's just ignore the error in such a case. This fixes an error to resume a domain when fdb entries were not properly removed when the domain was paused: virsh # resume test error: Failed to resume domain test error: error adding fdb entry for vnet2: File exists For some reason, fdb entries are only removed when libvirt explicitly stops CPUs, but nothing happens when we just get STOP event from QEMU. An alternative approach would be to make sure we always remove the entries regardless on why a domain was paused (e.g., during migration), but that would be a significantly more disruptive change with possible side effects. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/util/virnetdevbridge.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/src/util/virnetdevbridge.c b/src/util/virnetdevbridge.c index 7b5ea4fe1d..4fe84cc162 100644 --- a/src/util/virnetdevbridge.c +++ b/src/util/virnetdevbridge.c @@ -1063,9 +1063,13 @@ virNetDevBridgeFDBAddDel(const virMacAddr *mac, const char *ifname, if (resp->nlmsg_len < NLMSG_LENGTH(sizeof(*err))) goto malformed_resp; if (err->error) { - virReportSystemError(-err->error, - _("error adding fdb entry for %s"), ifname); - return -1; + if (isAdd && -err->error == EEXIST) { + VIR_DEBUG("fdb entry for %s already exists", ifname); + } else { + virReportSystemError(-err->error, + _("error adding fdb entry for %s"), ifname); + return -1; + } } break; case NLMSG_DONE: -- 2.31.1

2 1

[PATCH 0/3] Introduce <input type='linux'>
by Kristina Hanicova 30 Apr '21

30 Apr '21

4 6