- Devel - libvirt List Archives

[libvirt PATCHv2 0/3] docs: qemu: add socket for virtiofs filesystems
by Ján Tomko 27 Apr '21

27 Apr '21

Ján Tomko (3): docs: document new socket attribute for virtiofs docs: virtiofs: add section about externally-launched virtiofsd NEWS: qemu: add socket for virtiofs filesystems NEWS.rst | 6 ++++++ docs/formatdomain.rst | 10 +++++++++- docs/kbase/virtiofs.rst | 15 +++++++++++++++ 3 files changed, 30 insertions(+), 1 deletion(-) -- 2.29.2

2 4

[PATCH] qapi: deprecate drive-backup
by Vladimir Sementsov-Ogievskiy 27 Apr '21

27 Apr '21

Modern way is using blockdev-add + blockdev-backup, which provides a lot more control on how target is opened. As example of drive-backup problems consider the following: User of drive-backup expects that target will be opened in the same cache and aio mode as source. Corresponding logic is in drive_backup_prepare(), where we take bs->open_flags of source. It works rather bad if source was added by blockdev-add. Assume source is qcow2 image. On blockdev-add we should specify aio and cache options for file child of qcow2 node. What happens next: drive_backup_prepare() looks at bs->open_flags of qcow2 source node. But there no BDRV_O_NOCAHE neither BDRV_O_NATIVE_AIO: BDRV_O_NOCAHE is places in bs->file->bs->open_flags, and BDRV_O_NATIVE_AIO is nowhere, as file-posix parse options and simply set s->use_linux_aio. Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov(a)virtuozzo.com> --- Hi all! I remember, I suggested to deprecate drive-backup some time ago, and nobody complain.. But that old patch was inside the series with other more questionable deprecations and it did not landed. Let's finally deprecate what should be deprecated long ago. We now faced a problem in our downstream, described in commit message. In downstream I've fixed it by simply enabling O_DIRECT and linux_aio unconditionally for drive_backup target. But actually this just shows that using drive-backup in blockdev era is a bad idea. So let's motivate everyone (including Virtuozzo of course) to move to new interfaces and avoid problems with all that outdated option inheritance. docs/system/deprecated.rst | 5 +++++ qapi/block-core.json | 5 ++++- 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/docs/system/deprecated.rst b/docs/system/deprecated.rst index 80cae86252..b6f5766e17 100644 --- a/docs/system/deprecated.rst +++ b/docs/system/deprecated.rst @@ -186,6 +186,11 @@ 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``. +``drive-backup`` (since 6.0) +'''''''''''''''''''''''''''' + +Use ``blockdev-backup`` in pair with ``blockdev-add`` instead. + System accelerators ------------------- diff --git a/qapi/block-core.json b/qapi/block-core.json index 6d227924d0..8e2c6e1622 100644 --- a/qapi/block-core.json +++ b/qapi/block-core.json @@ -1642,6 +1642,9 @@ # The operation can be stopped before it has completed using the # block-job-cancel command. # +# Features: +# @deprecated: This command is deprecated. Use @blockdev-backup instead. +# # Returns: - nothing on success # - If @device is not a valid block device, GenericError # @@ -1657,7 +1660,7 @@ # ## { 'command': 'drive-backup', 'boxed': true, - 'data': 'DriveBackup' } + 'data': 'DriveBackup', 'features': ['deprecated'] } ## # @blockdev-backup: -- 2.29.2

5 10

[PATCH] qemu: add support for max-ram-below-4g option
by Zhiyong Ye 27 Apr '21

27 Apr '21

Limit the amount of ram below 4G. This can increase the address space used by PCI devices below 4G and it can be used by adding attributes in XML like this: <domain> ... <memory unit="MiB" below4g="2048">4096</memory> ... </domain> Signed-off-by: Zhiyong Ye <yezhiyong(a)bytedance.com> Signed-off-by: zhenwei pi <pizhenwei(a)bytedance.com> Signed-off-by: zhangruien <zhangruien(a)bytedance.com> --- src/conf/domain_conf.c | 19 +++++++++++++++++++ src/conf/domain_conf.h | 3 +++ src/qemu/qemu_command.c | 4 ++++ 3 files changed, 26 insertions(+) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index a72d58f488..c211a69ed1 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -4720,6 +4720,19 @@ virDomainDefPostParseMemory(virDomainDef *def, return -1; } + if (def->mem.max_ram_below_4g && + def->mem.max_ram_below_4g < (1ULL << 10)) { + virReportError(VIR_ERR_XML_ERROR, "%s", + _("maximum memory size below the 4GiB boundary is too small, " + "BIOS may not work with less than 1MiB")); + return -1; + } else if (def->mem.max_ram_below_4g > (1ULL << 22)) { + virReportError(VIR_ERR_XML_ERROR, "%s", + _("maximum memory size below the 4GiB boundary must be " + "less than or equal to 4GiB")); + return -1; + } + return 0; } @@ -19786,6 +19799,10 @@ virDomainDefParseMemory(virDomainDef *def, &def->mem.max_memory, false, false) < 0) goto error; + if (virDomainParseMemory("./memory[1]/@below4g", "./memory[1]/@unit", ctxt, + &def->mem.max_ram_below_4g, false, true) < 0) + goto error; + if (virXPathUInt("string(./maxMemory[1]/@slots)", ctxt, &def->mem.memory_slots) == -2) { virReportError(VIR_ERR_XML_ERROR, "%s", _("Failed to parse memory slot count")); @@ -28844,6 +28861,8 @@ virDomainDefFormatInternalSetRootName(virDomainDef *def, if (def->mem.dump_core) virBufferAsprintf(buf, " dumpCore='%s'", virTristateSwitchTypeToString(def->mem.dump_core)); + if (def->mem.max_ram_below_4g > 0) + virBufferAsprintf(buf, " below4g='%llu'", def->mem.max_ram_below_4g); virBufferAsprintf(buf, " unit='KiB'>%llu</memory>\n", virDomainDefGetMemoryTotal(def)); diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index 4838687edf..a939d43e93 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -2597,6 +2597,9 @@ struct _virDomainMemtune { unsigned long long max_memory; /* in kibibytes */ unsigned int memory_slots; /* maximum count of RAM memory slots */ + /* maximum memory below the 4GiB boundary (32bit boundary) */ + unsigned long long max_ram_below_4g; /* in kibibytes */ + bool nosharepages; bool locked; int dump_core; /* enum virTristateSwitch */ diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index be93182092..c69ad781e6 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -6961,6 +6961,10 @@ qemuBuildMachineCommandLine(virCommand *cmd, cfg->dumpGuestCore ? "on" : "off"); } + if (def->mem.max_ram_below_4g > 0) + virBufferAsprintf(&buf, ",max-ram-below-4g=%llu", + def->mem.max_ram_below_4g * 1024); + if (def->mem.nosharepages) virBufferAddLit(&buf, ",mem-merge=off"); -- 2.24.3 (Apple Git-128)

3 5

[libvirt PATCH 00/11] ci: helper: Improve cross-building support and more
by Andrea Bolognani 26 Apr '21

26 Apr '21

Mainly, this switches the output of the list-images action to be more compact and changes all actions that work on containers so that the cross-building architecture is provided as a separate argument. A few additional cleanups / improvements are thrown in as well. Andrea Bolognani (11): ci: util: Drop documentation for api_version parameter ci: util: Replace get_image_distro() with get_image_info() ci: util: Document that get_registry_images() returns raw data ci: util: Add _raw suffix to get_registry_images() ci: util: Add new get_registry_images() function ci: helper: Use get_registry_images() ci: helper: Improve output for list-images action ci: helper: Add _make_run_action() function ci: helper: Accept --cross-arch argument ci: helper: Enable gitlabparser for container actions ci: helper: Perform some validation on containers ci/helper | 58 +++++++++++++++++++++++++++---------------- ci/util.py | 73 ++++++++++++++++++++++++++++++++++++++++++------------ 2 files changed, 93 insertions(+), 38 deletions(-) -- 2.26.3

2 17

[libvirt] how about add a new API for set display parameters
by Yanzheng (A) 26 Apr '21

26 Apr '21

Hi team, struct _virHypervisorDriver now have more than 260 APIs to management various hypervisor drivers, including hardware resources and lifecycle, etc. But, there is no API for (re)set display devices and its parameters. Take the following uses the QEMU driver as an example. To update the VNC TLS certificates without restart virtual machine, run the following command: {"execute": "display-reload", "arguments":{"type": "vnc", "tls-certs": true}} Refers: https://gitlab.com/qemu-project/qemu/-/commit/9cc07651655ee86eca41059f5ead8… Can we add a new libvirt API to set the hypervisor driver display configurations ? If the committer team thinks it's worth doing, I can submit patches to implement the function. e,g. The preceding QMP commands 'display-reload' can be encapsulated in this way: index 42c75f6cc5..ab5ae0a3e6 100644 --- a/src/libvirt-domain.c +++ b/src/libvirt-domain.c @@ -13218,3 +13218,16 @@ virDomainStartDirtyRateCalc(virDomainPtr domain, virDispatchError(conn); return -1; } + +int +virDomainSetDisplayParameters(virDomainPtr domain, + virTypedParameterPtr params, + int nparams, + unsigned int flags) +{ + ... + if (conn->driver->domainSetDisplayParameters) { + conn->driver->domainSetDisplayParameters(domain, params, nparams, flags); + } + ... +} index d642af8a37..197da82556 100644 --- a/src/driver-hypervisor.h +++ b/src/driver-hypervisor.h @@ -1410,6 +1410,12 @@ typedef int int seconds, unsigned int flags); +typedef int +(*virDrvDomainSetDisplayParameters)(virDomainPtr domain. + virTypedParameterPtr params, + int nparams, + unsigned int flags); + typedef struct _virHypervisorDriver virHypervisorDriver; /** @@ -1676,4 +1682,5 @@ struct _virHypervisorDriver { virDrvDomainAuthorizedSSHKeysSet domainAuthorizedSSHKeysSet; virDrvDomainGetMessages domainGetMessages; virDrvDomainStartDirtyRateCalc domainStartDirtyRateCalc; + virDrvDomainSetDisplayParameters domainSetDisplayParameters; }; --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -20454,6 +20454,21 @@ qemuDomainStartDirtyRateCalc(virDomainPtr dom, return ret; } +static int +qemuDomainSetDisplayParameters(virDomainPtr domain, + virTypedParameterPtr params, + int nparams, + unsigned int flags) +{ + ... + if (params has 'display-reload') + virDomainQemuMonitorCommand(domain, strJsonCmd, &reply, + VIR_DOMAIN_QEMU_MONITOR_COMMAND_DEFAULT); + else if (...) { + /* other display parameters setting */ + } + ... +} static virHypervisorDriver qemuHypervisorDriver = { .name = QEMU_DRIVER_NAME, @@ -20698,6 +20713,7 @@ static virHypervisorDriver qemuHypervisorDriver = { .domainAuthorizedSSHKeysSet = qemuDomainAuthorizedSSHKeysSet, /* 6.10.0 */ .domainGetMessages = qemuDomainGetMessages, /* 7.1.0 */ .domainStartDirtyRateCalc = qemuDomainStartDirtyRateCalc, /* 7.2.0 */ + .domainSetDisplayParameters = qemuDomainSetDisplayParameters; /* 7.2.0 */ }; Regards :)

1 0

[libvirt PATCH 00/10] Refactor more XML parsing boilerplate code, part VI
by Tim Wiederhake 26 Apr '21

26 Apr '21

For background, see https://listman.redhat.com/archives/libvir-list/2021-April/msg00668.html Tim Wiederhake (10): virDomainGraphicsDefParseXMLSDL: Use virXMLProp* virDomainGraphicsDefParseXMLDesktop: Use virXMLProp* virDomainGraphicsDefParseXMLSpice: Use virXMLProp* virDomainGraphicsDefParseXML: Use virXMLProp* virDomainSoundDef: Change type of model to virDomainSoundModel virDomainSoundDefParseXML: Use virXMLProp* virDomainDef: Change type of hyperv_stimer_direct to virTristateSwitch virDomainDef: Change type of xen_passthrough_mode to virDomainXenPassthroughMode virDomainFeaturesDefParse: Use virXMLProp* virDomainAudioSDLParse: Use virXMLProp* src/conf/domain_conf.c | 683 +++++------------- src/conf/domain_conf.h | 6 +- src/qemu/qemu_command.c | 2 +- src/qemu/qemu_domain_address.c | 2 +- src/qemu/qemu_validate.c | 2 +- .../qemuxml2argvdata/aarch64-gic-invalid.err | 2 +- 6 files changed, 206 insertions(+), 491 deletions(-) -- 2.26.3

3 17

[PATCH] security: fix virSecurityManagerGetNested access illegal address
by gongwei＠smartx.com 26 Apr '21

26 Apr '21

When stop libvirtd is used, libvirtd exits the eventloop and cleans up the driverState first. Then release threadPool. If the workers thread is still executing at this time, it needs to access driverState. If the value in driverState is not judged at this time, direct access will cause an abnormal exit and damage the cache file of libvirt. In our example, the migration task is in progress at this time, the source is waiting for the target libvirtd dstFinish to return, the source libvirtd is stopped, and a crash occurs. After start libvirtd, the corresponding virtual machine process cannot be managed by libvirt stack: #0 virSecurityManagerGetNested (mgr=0x7f76141143c0) at security/security_manager.c:1033 1033 if (STREQ("stack", mgr->drv->name)) (gdb) bt #0 virSecurityManagerGetNested (mgr=0x7f76141143c0) at security/security_manager.c:1033 #1 0x00007f761c31660e in virQEMUDriverCreateCapabilities (driver=driver@entry=0x7f7614111060) at qemu/qemu_conf.c:1043 #2 0x00007f761c3168b3 in virQEMUDriverGetCapabilities (driver=0x7f7614111060, refresh=<optimized out>) at qemu/qemu_conf.c:1103 #3 0x00007f761c334d16 in qemuMigrationCookieXMLParse (flags=32, ctxt=0x7f76040040c0, doc=0x7f76040425c0, driver=0x7f7614111060, mig=0x7f760400ee10) at qemu/qemu_migration_cookie.c:1209 #4 qemuMigrationCookieXMLParseStr (flags=32, xml=0x7f7604004580 "<qemu-migration>\n <name>519ed304-375a-4819-a2d5-2f0ba662b9bc</name> 049152ab-efdf-4aaf-ab08-b57ac1816351</uuid> <hostname>gongwei-nestedcluster-20210330042359-1</me> <hostuuid>41d69"..., driver=0x7f7614111060, mig=0x7f760400ee10) at qemu/qemu_migration_cookie.c:1404 #5 qemuMigrationEatCookie (driver=driver@entry=0x7f7614111060, dom=dom@entry=0x7f7604001ac0, cookiein=cookiein@entry=0x7f7604004580 "<qemu-migration> <name>519ed304-375a-4819-a2d5-2f09bc</name> <uuid>049152ab-efdf-4aaf-ab08-b57ac1816351</uuid> <hostname>gongwei-nestedcluste0330042359-1</hostname> <hostuuid>41d69"..., cookieinlen=cookieinlen@entry=1410, flags=flags@entry=32) at qemu/qemu_migration_cookie.c:1501 #6 0x00007f761c3291d5 in qemuMigrationSrcConfirmPhase (driver=driver@entry=0x7f7614111060, vm=vm@entry=0x7f7604001ac0, cookiein=0x7f7604004580 "<qemu-migration> <name>519ed304-375a-4819-a2d5-2f0ba662b9bc</nameuuid>049152ab-efdf-4aaf-ab08-b57ac1816351</uuid> <hostname>gongwei-nestedcluster-2021033004235ostname> <hostuuid>41d69"..., cookieinlen=1410, flags=14875, retcode=retcode@entry=0) at qemu/qemu_migration.c:2805 #7 0x00007f761c331539 in qemuMigrationSrcPerformPeer2Peer3 (flags=14875, useParams=true, bandwidth=0, migParams=0x7f760400f070, nbdPort=0, migrate_disks=<optimized out>, nmigrate_disks=0, listenAddress=<optimized out>, graphicsuri=<optimized out>, uri=<optimized out>, dname=0x0, persist_xml=0x0, xmlin=<optimized out>, vm=0x7f7604001ac0, dconnuri=0x7f7604000df0 "qemu+tcp://10.181.177.170/system", dconn=0x7f7604021680, sconn=0x7f7608001410, driver=0x7f7614111060) at qemu/qemu_migration.c:4202 (gdb) frame 1 #1 0x00007f761c31660e in virQEMUDriverCreateCapabilities (driver=driver@entry=0x7f7614111060) at qemu/qemu_conf.c:1043 1043 if (!(sec_managers = qemuSecurityGetNested(driver->securityManager))) (gdb) p *(driver->securityManager) $2 = {parent = {parent = {u = {dummy_align1 = 140145119544368, dummy_align2 = 0x7f7614114430, s = magic = 336675888, refs = 32630}}, klass = 0xdeadbeef}, lock = {lock = {__data = { __lock = 0, __count = 0, __owner = 0, __nusers = 0, __kind = 0, __spins = 0, __elision __list = {__prev = 0x0, __next = 0x0}}, __size = '\000' <repeats 39 times>, __align = 0 drv = 0x0, flags = 0, virtDriver = 0x0, privateData = 0x0} if (STREQ("stack", mgr->drv->name) mgr->drv is 0x0 Signed-off-by: gongwei <gongwei(a)smartx.com> --- src/security/security_manager.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/security/security_manager.c b/src/security/security_manager.c index d8b84e2861..96ca9ee861 100644 --- a/src/security/security_manager.c +++ b/src/security/security_manager.c @@ -1030,6 +1030,9 @@ virSecurityManagerGetNested(virSecurityManager *mgr) { virSecurityManager ** list = NULL; + if (mgr == NULL || mgr->drv == NULL) + return NULL; + if (STREQ("stack", mgr->drv->name)) return virSecurityStackGetNested(mgr); -- 2.24.1

2 1

[PATCH v2] formatdomain.rst: update slice xml in "Hard drives" part
by MeinaLi 25 Apr '21

25 Apr '21

The disk type=’lun’ doesn't support storage slice, which was introduced in: 5d72c3ce28 qemu: domain: Forbid slice/encryption/copy_on_read with disk type='lun' Signed-off-by: Meina Li <meili(a)redhat.com> --- docs/formatdomain.rst | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/docs/formatdomain.rst b/docs/formatdomain.rst index 1b9b221611..7bd5e685e8 100644 --- a/docs/formatdomain.rst +++ b/docs/formatdomain.rst @@ -2381,9 +2381,6 @@ paravirtualized driver is specified via the ``disk`` element. <disk type='block' device='lun'> <driver name='qemu' type='raw'/> <source dev='/dev/sda'> - <slices> - <slice type='storage' offset='12345' size='123'/> - </slices> <reservations managed='no'> <source type='unix' path='/path/to/qemu-pr-helper' mode='client'/> </reservations> @@ -2393,7 +2390,11 @@ paravirtualized driver is specified via the ``disk`` element. </disk> <disk type='block' device='disk'> <driver name='qemu' type='raw'/> - <source dev='/dev/sda'/> + <source dev='/dev/sda'> + <slices> + <slice type='storage' offset='12345' size='123'/> + </slices> + </source> <geometry cyls='16383' heads='16' secs='63' trans='lba'/> <blockio logical_block_size='512' physical_block_size='4096'/> <target dev='hdj' bus='ide'/> -- 2.27.0

1 0

[libvirt PATCH 00/10] Refactor more XML parsing boilerplate code, part V
by Tim Wiederhake 23 Apr '21

23 Apr '21

For background, see https://listman.redhat.com/archives/libvir-list/2021-April/msg00668.html Tim Wiederhake (10): virXMLPropEnum: Fix return value virDomainControllerDef: Change type of ioeventfd to virTristateSwitch virDomainPCIControllerOpts: Change type of modelName to virDomainControllerPCIModelName virDomainControllerDefParseXML: Use virXMLProp* 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 22

[PATCH] qemu: Don't double free @node_cpus in qemuProcessSetupPid()
by Michal Privoznik 23 Apr '21

23 Apr '21

When placing vCPUs into CGroups the qemuProcessSetupPid() is called which then enters a for() loop (around its middle) where it calls virDomainNumaGetNodeCpumask() for each guest NUMA node. But the latter returns only a pointer not new reference/copy and thus the caller must not free it. But the variable is decorated with g_autoptr() which leads to a double free. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/qemu/qemu_process.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 05104e3098..92c1f0ab74 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -2741,8 +2741,7 @@ qemuProcessSetupPid(virDomainObj *vm, /* Look for the guest NUMA node of this vCPU */ for (i = 0; i < virDomainNumaGetNodeCount(numatune); i++) { - g_autoptr(virBitmap) node_cpus = NULL; - node_cpus = virDomainNumaGetNodeCpumask(numatune, i); + virBitmap *node_cpus = virDomainNumaGetNodeCpumask(numatune, i); if (!virBitmapIsBitSet(node_cpus, id)) continue; -- 2.26.3

2 2

[PATCH] rpc: libssh2: Enable EC host keys
by Bastian Germann 22 Apr '21

22 Apr '21

libssh2 has ECDSA and ED25519 support beginning with v1.9.0. libvirt cannot make use of those because it will handle them as unknown key types. Add support for those host key types. Signed-off-by: Bastian Germann <bastiangermann(a)fishpost.de> --- src/rpc/virnetsshsession.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) diff --git a/src/rpc/virnetsshsession.c b/src/rpc/virnetsshsession.c index fe77594..c311e90 100644 --- a/src/rpc/virnetsshsession.c +++ b/src/rpc/virnetsshsession.c @@ -389,7 +389,21 @@ virNetSSHCheckHostKey(virNetSSHSessionPtr sess) case LIBSSH2_HOSTKEY_TYPE_DSS: keyType = LIBSSH2_KNOWNHOST_KEY_SSHDSS; break; - +#ifdef LIBSSH2_HOSTKEY_TYPE_ED25519 + /* defs from libssh2 v1.9.0 or later */ + case LIBSSH2_HOSTKEY_TYPE_ECDSA_256: + keyType = LIBSSH2_KNOWNHOST_KEY_ECDSA_256; + break; + case LIBSSH2_HOSTKEY_TYPE_ECDSA_384: + keyType = LIBSSH2_KNOWNHOST_KEY_ECDSA_384; + break; + case LIBSSH2_HOSTKEY_TYPE_ECDSA_521: + keyType = LIBSSH2_KNOWNHOST_KEY_ECDSA_521; + break; + case LIBSSH2_HOSTKEY_TYPE_ED25519: + keyType = LIBSSH2_KNOWNHOST_KEY_ED25519; + break; +#endif case LIBSSH2_HOSTKEY_TYPE_UNKNOWN: default: virReportError(VIR_ERR_SSH, "%s", -- 2.31.0

3 2

[libvirt PATCH 0/3] docs: qemu: add socket for virtiofs filesystems
by Ján Tomko 22 Apr '21

22 Apr '21

Add documentation for the new attribute. Ján Tomko (3): docs: document new socket attribute for virtiofs docs: virtiofs: add section about externally-launched virtiofsd NEWS: qemu: add socket for virtiofs filesystems NEWS.rst | 6 ++++++ docs/formatdomain.rst | 10 +++++++++- docs/kbase/virtiofs.rst | 14 ++++++++++++++ 3 files changed, 29 insertions(+), 1 deletion(-) -- 2.29.2

2 4

[libvirt PATCH 00/10] Refactor more XML parsing boilerplate code, part IV
by Tim Wiederhake 22 Apr '21

22 Apr '21

For background, see https://listman.redhat.com/archives/libvir-list/2021-April/msg00668.html Tim Wiederhake (10): virDomainDiskDef: Change type of cachemode to virDomainDiskCache virDomainDiskDef: Change type of error_policy to virDomainDiskErrorPolicy virDomainDiskDef: Change type of rerror_policy to virDomainDiskErrorPolicy virDomainDiskDef: Change type of iomode to virDomainDiskInfo virDomainDiskDef: Change type of discard to virDomainDiskDiscard virDomainDiskDef: Change type of detect_zeroes to virDomainDiskDetectZeroes virDomainDiskDefDriverParseXML: Use virXMLProp* domain_conf: Introduce function virDomainChrSouceModeTypeFromString domain_conf: Remove function virDomainChrSourceDefParseMode virDomainChrSourceDefParseTCP: Use virXMLProp* src/conf/domain_conf.c | 176 +++++++++++++---------------------------- src/conf/domain_conf.h | 12 +-- 2 files changed, 61 insertions(+), 127 deletions(-) -- 2.26.3

2 11

[libvirt PATCH] vahDeinit: Fix memory leak
by Tim Wiederhake 22 Apr '21

22 Apr '21

Calling VIR_FREE on a virDomainDef* does not free its various contained pointers. Signed-off-by: Tim Wiederhake <twiederh(a)redhat.com> --- src/security/virt-aa-helper.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/security/virt-aa-helper.c b/src/security/virt-aa-helper.c index 68ac817f47..2331cc6648 100644 --- a/src/security/virt-aa-helper.c +++ b/src/security/virt-aa-helper.c @@ -79,7 +79,7 @@ vahDeinit(vahControl * ctl) if (ctl == NULL) return -1; - VIR_FREE(ctl->def); + virDomainDefFree(ctl->def); virObjectUnref(ctl->caps); virObjectUnref(ctl->xmlopt); VIR_FREE(ctl->files); -- 2.26.3

2 1

[libvirt] [PATCH 0/2] virsh completer fixes
by Lin Ma 22 Apr '21

22 Apr '21

Lin Ma (2): virsh: Add mountpoint completion to domfsfreeze/domfsthaw command virsh: Fix completion logic to guestvcpus command tools/virsh-completer-domain.c | 104 +++++++++++++++++++++++++++------ tools/virsh-completer-domain.h | 4 ++ tools/virsh-domain.c | 2 + 3 files changed, 93 insertions(+), 17 deletions(-) -- 2.26.2

2 4

[PATCH 0/4] lxc: Let the driver detect CGroups earlier
by Michal Privoznik 22 Apr '21

22 Apr '21

See 4/4 for explanation. Michal Prívozník (4): lxc_controller: Initialize ctrl->handshakeFd properly lxc_controller: Move closing of handshakeFd out of virLXCControllerDaemonHandshake() lxc: Pass another pipe to lxc_controller lxc: Let the driver detect CGroups earlier src/lxc/lxc_controller.c | 65 +++++++++++++++++++++++++++++++--------- src/lxc/lxc_process.c | 41 ++++++++++++++++++++----- 2 files changed, 84 insertions(+), 22 deletions(-) -- 2.26.3

2 8

[libvirt PATCH 00/10] Refactor more XML parsing boilerplate code, part III
by Tim Wiederhake 22 Apr '21

22 Apr '21

For background, see https://listman.redhat.com/archives/libvir-list/2021-April/msg00668.html Tim Wiederhake (10): virDomainKeyWrapCipherDefParseXML: Use virXMLProp* virDomainKeyWrapDef: Make members virTristateSwitch qemuAppendKeyWrapMachineParm: Stricten parameter types virxml: Add virXMLPropULongLong virDomainDeviceDimmAddressParseXML: Use virXMLProp* virDomainHostdevSubsysSCSIHostDefParseXML: Use virXMLProp* virStorageEncryptionInfoParseCipher: Use virXMLProp* virDomainDeviceInfoParseXML: Use virXMLProp* virDomainDiskSourceNetworkParse: Use virXMLProp* virDomainMemorytuneDefParseMemory: Use virXMLProp* src/conf/domain_conf.c | 214 ++++++++--------------------- src/conf/domain_conf.h | 4 +- src/conf/storage_encryption_conf.c | 16 +-- src/libvirt_private.syms | 1 + src/qemu/qemu_command.c | 3 +- src/util/virxml.c | 56 ++++++++ src/util/virxml.h | 8 ++ 7 files changed, 125 insertions(+), 177 deletions(-) -- 2.26.3

2 11

[PATCH] formatdomain.rst: update slice xml in "Hard drives" part
by “Meina 22 Apr '21

22 Apr '21

Update slice xml from block lun disk because it doesn't support storage slice Signed-off-by: Meina Li <meili(a)redhat.com> --- docs/formatdomain.rst | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/docs/formatdomain.rst b/docs/formatdomain.rst index 1b9b221611..44aee9fa77 100644 --- a/docs/formatdomain.rst +++ b/docs/formatdomain.rst @@ -2381,9 +2381,6 @@ paravirtualized driver is specified via the ``disk`` element. <disk type='block' device='lun'> <driver name='qemu' type='raw'/> <source dev='/dev/sda'> - <slices> - <slice type='storage' offset='12345' size='123'/> - </slices> <reservations managed='no'> <source type='unix' path='/path/to/qemu-pr-helper' mode='client'/> </reservations> @@ -2393,10 +2390,14 @@ paravirtualized driver is specified via the ``disk`` element. </disk> <disk type='block' device='disk'> <driver name='qemu' type='raw'/> - <source dev='/dev/sda'/> + <source dev='/dev/sda'> + <slices> + <slice type='storage' offset='12345' size='123'/> + </slices> + </source> <geometry cyls='16383' heads='16' secs='63' trans='lba'/> <blockio logical_block_size='512' physical_block_size='4096'/> - <target dev='hdj' bus='ide'/> + <target dev='sda' bus='scsi'/> </disk> <disk type='volume' device='disk'> <driver name='qemu' type='raw'/> -- 2.27.0

2 2

[PATCH] bash-completion: Fix argument passing to $1
by Michal Privoznik 22 Apr '21

22 Apr '21

Our vsh bash completion string is merely just a wrapper over virsh/virt-admin complete (cmdComplete) - a hidden command that uses internal readline completion to generate list of candidates. But this means that we have to pass some additional arguments to the helper process: e.g. connection URI and R/O flag. Candidates are printed on a separate line each (and can contain space), which means that when bash is reading the helper's output into an array, it needs to split items on '\n' char - hence the IFS=$'\n' prefix on the line executing the helper. This was introduced in b889594a70. But this introduced a regression - those extra arguments we might pass are stored in a string and previously were split on a space character (because $IFS was kept untouched and by default contains space). But now, after the fix that's no longer the case and thus virsh/virt-admin sees ' -r -c URI' as one argument. The solution is to take $IFS out of the picture by storing the extra arguments in an array instead of string. Fixes: b889594a7092440dc916e3f43eeeaca2684571ee Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- tools/bash-completion/vsh | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/tools/bash-completion/vsh b/tools/bash-completion/vsh index bbb25fc3eb..402945a8c5 100644 --- a/tools/bash-completion/vsh +++ b/tools/bash-completion/vsh @@ -30,12 +30,12 @@ _vsh_complete() c=$((++c)) done - CMDLINE= + CMDLINE=( ) if [ -n "${RO}" ]; then - CMDLINE="${CMDLINE} -r" + CMDLINE+=("-r") fi if [ -n "${URI}" ]; then - CMDLINE="${CMDLINE} -c ${URI}" + CMDLINE+=("-c" "${URI}") fi INPUT=( "${COMP_WORDS[@]:$i:$COMP_CWORD}" ) @@ -56,7 +56,7 @@ _vsh_complete() # the name of the command whose arguments are being # completed. # Therefore, we might just run $1. - IFS=$'\n' A=($($1 ${CMDLINE} complete -- "${INPUT[@]}" 2>/dev/null)) + IFS=$'\n' A=($($1 ${CMDLINE[@]} complete -- "${INPUT[@]}" 2>/dev/null)) COMPREPLY=($(compgen -W "${A[*]%--}" -- ${cur})) __ltrim_colon_completions "$cur" -- 2.26.3

2 1

[PATCH 0/3] tests: qemuxml2argv negative tests for hugepages and numatune
by Han Han 22 Apr '21

22 Apr '21

Han Han (3): tests: negative test for hugepages with anonymous memory backing tests: negative test for hugepages with ondemand mode tests: negative test of binding memnode cellid to an nonexist guest cell .../qemuxml2argvdata/hugepages-anonymous.err | 1 + .../qemuxml2argvdata/hugepages-anonymous.xml | 25 +++++++++++++++ tests/qemuxml2argvdata/hugepages-ondemand.err | 1 + tests/qemuxml2argvdata/hugepages-ondemand.xml | 25 +++++++++++++++ ...une-memnode-cellid-nonexist-guest-cell.err | 1 + ...une-memnode-cellid-nonexist-guest-cell.xml | 31 +++++++++++++++++++ tests/qemuxml2argvtest.c | 3 ++ 7 files changed, 87 insertions(+) create mode 100644 tests/qemuxml2argvdata/hugepages-anonymous.err create mode 100644 tests/qemuxml2argvdata/hugepages-anonymous.xml create mode 100644 tests/qemuxml2argvdata/hugepages-ondemand.err create mode 100644 tests/qemuxml2argvdata/hugepages-ondemand.xml create mode 100644 tests/qemuxml2argvdata/numatune-memnode-cellid-nonexist-guest-cell.err create mode 100644 tests/qemuxml2argvdata/numatune-memnode-cellid-nonexist-guest-cell.xml -- 2.30.2

2 5

[PATCH 0/1] replace the method in several files
by Luke Yue 22 Apr '21

22 Apr '21

Using `git grep -n "\[0\].*'/'"` to find files that may need the treatment, then judge and replace. There may be some files replaced by mistake though. Luke Yue (1): Replace AbsPath judgement method with g_path_is_absolute() src/conf/backup_conf.c | 2 +- src/conf/snapshot_conf.c | 2 +- src/conf/storage_source_conf.c | 2 +- src/lxc/lxc_native.c | 2 +- src/qemu/qemu_block.c | 2 +- src/storage_file/storage_source.c | 2 +- src/util/vircommand.c | 2 +- src/vbox/vbox_snapshot_conf.c | 2 +- src/vmware/vmware_conf.c | 2 +- src/vmware/vmware_driver.c | 2 +- tests/virtestmock.c | 2 +- tools/virt-login-shell-helper.c | 2 +- 12 files changed, 12 insertions(+), 12 deletions(-) -- 2.31.1

2 3

[libvirt PATCH v2 1/3] nodedev: fix potential leak of command
by Jonathon Jongsma 21 Apr '21

21 Apr '21

When returning early due to errors, cmd will be leaked. Use an autoptr to handle these early returns without leaking memory. Signed-off-by: Jonathon Jongsma <jjongsma(a)redhat.com> Reviewed-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/node_device/node_device_driver.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/node_device/node_device_driver.c b/src/node_device/node_device_driver.c index b0cc63ed42..e565cc29ec 100644 --- a/src/node_device/node_device_driver.c +++ b/src/node_device/node_device_driver.c @@ -724,7 +724,7 @@ nodeDeviceGetMdevctlCommand(virNodeDeviceDef *def, char **errbuf) { g_autofree char *parent_addr = NULL; - virCommand *cmd = NULL; + g_autoptr(virCommand) cmd = NULL; const char *subcommand = virMdevctlCommandTypeToString(cmd_type); g_autofree char *inbuf = NULL; @@ -787,7 +787,7 @@ nodeDeviceGetMdevctlCommand(virNodeDeviceDef *def, virCommandSetErrorBuffer(cmd, errbuf); - return cmd; + return g_steal_pointer(&cmd); } -- 2.26.3

2 3

[PATCH 00/10] qemu: Fix block job cancelling
by Peter Krempa 21 Apr '21

21 Apr '21

Since -blockdev support was introduced qemuDomainBlockJobAbort was using the wrong API to terminate the blockjob since we care about sync finishing semantics. Additionally for cases where we don't care we can force-finish the jobs such as when cancelling an migration with NBD disk copy. Peter Krempa (10): qemumonitorjsontest: Add test for 'qemuMonitorJSONBlockJobCancel' qemuMonitorJSONBlockJobCancel: Refactor cleanup qemu: monitor: Add 'force' argument for 'block-job-cancel' QMP command qemuDomainBlockJobAbort: Don't use 'job-cancel' instead of 'block-job-cancel' qemuBackupJobCancelBlockjobs: Replace qemuMonitorJobCancel by qemuMonitorBlockJobCancel qemuBlockJobRefreshJobs: Replace qemuMonitorJobCancel by qemuMonitorBlockJobCancel qemuMigrationSrcNBDCopyCancel*: Rename 'check' to 'abortMigration' qemuMigrationSrcNBDCopyCancelOne: Force-cancel disk copy jobs when aborting migration qemuMigrationSrcNBDCopyCancelled: Use do-while loop instead of jumping back qemu: monitor: Remove qemuMonitorJobCancel src/qemu/qemu_backup.c | 2 +- src/qemu/qemu_blockjob.c | 2 +- src/qemu/qemu_driver.c | 7 +-- src/qemu/qemu_migration.c | 106 ++++++++++++++++++----------------- src/qemu/qemu_monitor.c | 20 ++----- src/qemu/qemu_monitor.h | 8 +-- src/qemu/qemu_monitor_json.c | 48 +++------------- src/qemu/qemu_monitor_json.h | 8 +-- tests/qemumonitorjsontest.c | 4 +- 9 files changed, 75 insertions(+), 130 deletions(-) -- 2.30.2

2 11

[libvirt PATCH v2] storage_backend_fs: use MKFS ony if WITH_STORAGE_FS is defined
by Pavel Hrdina 21 Apr '21

21 Apr '21

The code in storage_backend_fs is used for storage_dir and storage_fs drivers so some parts need to be guarded by checking for WITH_STORAGE_FS. Fixes: 16c69e7aaed4cabfd8e8c19cc326854d4c480437 Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- GitLab CI faild on MacOS and FreeBSD where storage_fs is disabled. Changes in v2: - guard only virFindFileInPath instead of the whole function src/storage/storage_backend_fs.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/src/storage/storage_backend_fs.c b/src/storage/storage_backend_fs.c index af645ea9de..be3c3c6839 100644 --- a/src/storage/storage_backend_fs.c +++ b/src/storage/storage_backend_fs.c @@ -402,7 +402,11 @@ virStorageBackendExecuteMKFS(const char *device, const char *format) { g_autoptr(virCommand) cmd = NULL; - g_autofree char *mkfs = virFindFileInPath(MKFS); + g_autofree char *mkfs = NULL; + +#if WITH_STORAGE_FS + mkfs = virFindFileInPath(MKFS); +#endif /* WITH_STORAGE_FS */ if (!mkfs) { virReportError(VIR_ERR_INTERNAL_ERROR, @@ -412,7 +416,7 @@ virStorageBackendExecuteMKFS(const char *device, return -1; } - cmd = virCommandNewArgList(MKFS, "-t", format, NULL); + cmd = virCommandNewArgList(mkfs, "-t", format, NULL); /* use the force, otherwise mkfs.xfs won't overwrite existing fs. * Similarly mkfs.ext2, mkfs.ext3, and mkfs.ext4 require supplying -F -- 2.30.2

2 1

[libvirt PATCH] storage_backend_fs: use MKFS ony if WITH_STORAGE_FS is defined
by Pavel Hrdina 21 Apr '21

21 Apr '21

The code in storage_backend_fs is used for storage_dir and storage_fs drivers so some parts need to be guarded by checking for WITH_STORAGE_FS. Fixes: 16c69e7aaed4cabfd8e8c19cc326854d4c480437 Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- GitLab CI faild on MacOS and FreeBSD where storage_fs is disabled. src/storage/storage_backend_fs.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/storage/storage_backend_fs.c b/src/storage/storage_backend_fs.c index af645ea9de..e6a348521d 100644 --- a/src/storage/storage_backend_fs.c +++ b/src/storage/storage_backend_fs.c @@ -401,6 +401,7 @@ static int virStorageBackendExecuteMKFS(const char *device, const char *format) { +#if WITH_STORAGE_FS g_autoptr(virCommand) cmd = NULL; g_autofree char *mkfs = virFindFileInPath(MKFS); @@ -431,6 +432,7 @@ virStorageBackendExecuteMKFS(const char *device, if (virCommandRun(cmd, NULL) < 0) return -1; +#endif /* WITH_STORAGE_FS */ return 0; } -- 2.30.2

2 4