- Devel - libvirt List Archives

[libvirt] [PATCH 0/3] add option to keep nvram file on undefine
by Nikolay Shirokovskiy 13 Sep '16

13 Sep '16

There is already a patch [1] on this topic with a different approach - keep nvram file by default. There is also some discussion there. To sum up keeping nvram on undefine could be useful in some usecases so there should be an option to do it. On the other hand there is a danger of leaving domain assets after its undefine and unsing them unintentionally on defining domain with the same name. AFAIU keeping nvram by default was motivated by domain disks behaviour. I think there is a difference as libvirt never create disks for domain as opposed to nvram and managed save and without disks domain will not start so user is quite aware of disks files. On the other hand one can start using nvram file solely putting <nvram> in config and managed save is created on daemon shutdown. So user is much less aware of nvram and managed save existence. Thus one can easily mess up by unaware define $name/using/undefine/define $name again usecase. Thus I vote for keeping said assets only if it is specified explicitly so user knows what he is doing. Adding option to undefine is best solution I come up with. The other options are add checks on define or start and both are impossible. Such a check should be done without any extra flags for it to be useful but this way we break existing users. As this a proof of concept this series does not add extra flag for managed save. [1] https://www.redhat.com/archives/libvir-list/2015-February/msg00915.html Nikolay Shirokovskiy (3): api: add VIR_DOMAIN_UNDEFINE_KEEP_NVRAM flag qemu: add VIR_DOMAIN_UNDEFINE_KEEP_NVRAM support virsh: add --keep-nvram option to undefine command include/libvirt/libvirt-domain.h | 1 + src/qemu/qemu_driver.c | 26 +++++++++++++++++--------- tools/virsh-domain.c | 8 ++++++++ tools/virsh.pod | 6 +++--- 4 files changed, 29 insertions(+), 12 deletions(-) -- 1.8.3.1

5 10

[libvirt] [PATCH] conf: allow hotplugging "legacy PCI" device to manually addressed PCIe slot
by Laine Stump 12 Sep '16

12 Sep '16

In a full domain config, libvirt allows overriding the normal PCI vs. PCI Express rules when a device address is explicitly provided (so, e.g., you can force a legacy PCI device to plug into a PCIe port, although libvirt would never do that on its own). However, due to a bug libvirt doesn't give this same leeway when hotplugging devices. On top of that, current libvirt assumes that *all* devices are legacy PCI. The result of all this is that it's impossible to hotplug a device into a PCIe port, even if you manually add the PCI address. This can all be traced to the function virDomainPCIAddressEnsureAddr(), and the fact that it calls virDomainPCIaddressReserveSlot() for manually set addresses, and that function hardcodes the argument "fromConfig" to false (meaning "this address was auto-assigned, so it should be subject to stricter validation"). Since virDomainPCIAddressReserveSlot() is just a one line simple wrapper around virDomainPCIAddressReserveAddr() (adding in a hardcoded reserveEntireSlot = true and fromConfig = false), all that's needed to solve the problem with no unwanted side effects is to replace that call for virDomainPCIAddressReserveSlot() with a direct call to virDomainPCIAddressReserveAddr(), but with reserveEntireSlot = true, fromConfig = true. That's what this patch does. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1337490 --- src/conf/domain_addr.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/conf/domain_addr.c b/src/conf/domain_addr.c index 79c024f..93026c2 100644 --- a/src/conf/domain_addr.c +++ b/src/conf/domain_addr.c @@ -507,7 +507,8 @@ virDomainPCIAddressEnsureAddr(virDomainPCIAddressSetPtr addrs, addrStr, flags, true)) goto cleanup; - ret = virDomainPCIAddressReserveSlot(addrs, &dev->addr.pci, flags); + ret = virDomainPCIAddressReserveAddr(addrs, &dev->addr.pci, flags, + true, true); } else { ret = virDomainPCIAddressReserveNextSlot(addrs, dev, flags); } -- 2.7.4

2 2

[libvirt] [PATCH] audit: Audit information about shmem devices
by Martin Kletzander 12 Sep '16

12 Sep '16

Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1218603 Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- docs/auditlog.html.in | 19 +++++++++++++++++++ src/conf/domain_audit.c | 43 +++++++++++++++++++++++++++++++++++++++++++ src/conf/domain_audit.h | 4 ++++ src/libvirt_private.syms | 1 + 4 files changed, 67 insertions(+) diff --git a/docs/auditlog.html.in b/docs/auditlog.html.in index 012d0680a87b..7348948feb7a 100644 --- a/docs/auditlog.html.in +++ b/docs/auditlog.html.in @@ -352,5 +352,24 @@ <dd>The name of the cgroup controller</dd> </dl> + + <h4><a name="typeresourceshmem">console/serial/parallel/channel</a></h4> + <p> + The <code>msg</code> field will include the following sub-fields + </p> + + <dl> + <dt><code>resrc</code></dt> + <dd>The type of resource assigned. Set to <code>shmem</code></dd> + <dt><code>reason</code></dt> + <dd>The reason which caused the resource to be assigned to happen</dd> + <dt><code>size</code></dt> + <dd>The size of the shared memory region</dd> + <dt><code>shmem</code></dt> + <dd>Name of the shared memory region</dd> + <dt><code>source</code></dt> + <dd>Path of the backing character device for given emulated device</dd> + </dl> + </body> </html> diff --git a/src/conf/domain_audit.c b/src/conf/domain_audit.c index 53a58ac4c88c..cf9c03dee749 100644 --- a/src/conf/domain_audit.c +++ b/src/conf/domain_audit.c @@ -889,6 +889,9 @@ virDomainAuditStart(virDomainObjPtr vm, const char *reason, bool success) if (vm->def->tpm) virDomainAuditTPM(vm, vm->def->tpm, "start", true); + for (i = 0; i < vm->def->nshmems; i++) + virDomainAuditShmem(vm, vm->def->shmems[i], "start", true); + virDomainAuditMemory(vm, 0, virDomainDefGetMemoryTotal(vm->def), "start", true); virDomainAuditVcpu(vm, 0, virDomainDefGetVcpus(vm->def), "start", true); @@ -963,3 +966,43 @@ virDomainAuditSecurityLabel(virDomainObjPtr vm, bool success) VIR_FREE(vmname); } + +void +virDomainAuditShmem(virDomainObjPtr vm, + virDomainShmemDefPtr def, + const char *reason, bool success) +{ + char uuidstr[VIR_UUID_STRING_BUFLEN]; + char *vmname = virAuditEncode("vm", vm->def->name); + const char *srcpath = virDomainAuditChardevPath(&def->server.chr); + char *src = virAuditEncode("server", VIR_AUDIT_STR(srcpath)); + char *shmem = virAuditEncode("shmem", VIR_AUDIT_STR(def->name)); + const char *virt = virDomainVirtTypeToString(vm->def->virtType); + char *size = NULL; + + ignore_value(virAsprintfQuiet(&size, "%llu", def->size)); + virUUIDFormat(vm->def->uuid, uuidstr); + if (!vmname || !src || !size || !shmem || !size) + goto no_memory; + + if (!virt) { + VIR_WARN("Unexpected virt type %d while encoding audit message", + vm->def->virtType); + virt = "?"; + } + + VIR_AUDIT(VIR_AUDIT_RECORD_RESOURCE, success, + "virt=%s resrc=shmem reason=%s %s uuid=%s size=%s %s %s", + virt, reason, vmname, uuidstr, size ?: "?", shmem, src); + + cleanup: + VIR_FREE(vmname); + VIR_FREE(src); + VIR_FREE(size); + VIR_FREE(shmem); + return; + + no_memory: + VIR_WARN("OOM while encoding audit message"); + goto cleanup; +} diff --git a/src/conf/domain_audit.h b/src/conf/domain_audit.h index 97dadca5aae7..8cb585dc7d96 100644 --- a/src/conf/domain_audit.h +++ b/src/conf/domain_audit.h @@ -129,6 +129,10 @@ void virDomainAuditRNG(virDomainObjPtr vm, const char *reason, bool success) ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(4); +void virDomainAuditShmem(virDomainObjPtr vm, + virDomainShmemDefPtr def, + const char *reason, bool success) + ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(2) ATTRIBUTE_NONNULL(3); #endif /* __VIR_DOMAIN_AUDIT_H__ */ diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index a5fa30562f0c..1555f60ffda4 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -146,6 +146,7 @@ virDomainAuditNetDevice; virDomainAuditRedirdev; virDomainAuditRNG; virDomainAuditSecurityLabel; +virDomainAuditShmem; virDomainAuditStart; virDomainAuditStop; virDomainAuditVcpu; -- 2.10.0

2 2

[libvirt] [PATCH] qemu: fix improper initialization of cgroupControllers bitmap
by Laine Stump 12 Sep '16

12 Sep '16

virQEMUDriverConfigNew() always initializes the bitmap in its cgroupControllers member to -1 (i.e. all 1's). Prior to commit a9331394, if qemu.conf had a line with "cgroup_controllers", cgroupControllers would get reset to 0 before going through a loop setting a bit for each named cgroup controller. commit a9331394 left out the "reset to 0" part, so cgroupControllers would always be -1; if you didn't want a controller included, there was no longer a way to make that happen. This was discovered by users who were using qemu commandline passthrough to use the "input-linux" method of directing keyboard/mouse input to a virtual machine: https://www.redhat.com/archives/vfio-users/2016-April/msg00105.html Here's the first report I found of the problem encountered after upgrading libvirt beyond v2.0.0: https://www.redhat.com/archives/vfio-users/2016-August/msg00053.html Thanks to sL1pKn07 SpinFlo <sl1pkn07(a)gmail.com> for bringing the problem up in IRC, and then taking the time to do a git bisect and find the patch that started the problem. --- src/qemu/qemu_conf.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c index e7b2d8d..dad8334 100644 --- a/src/qemu/qemu_conf.c +++ b/src/qemu/qemu_conf.c @@ -613,15 +613,18 @@ int virQEMUDriverConfigLoadFile(virQEMUDriverConfigPtr cfg, &controllers) < 0) goto cleanup; - for (i = 0; controllers != NULL && controllers[i] != NULL; i++) { - int ctl; - if ((ctl = virCgroupControllerTypeFromString(controllers[i])) < 0) { - virReportError(VIR_ERR_CONF_SYNTAX, - _("Unknown cgroup controller '%s'"), - controllers[i]); - goto cleanup; + if (controllers) { + cfg-> cgroupControllers = 0; + for (i = 0; controllers[i] != NULL; i++) { + int ctl; + if ((ctl = virCgroupControllerTypeFromString(controllers[i])) < 0) { + virReportError(VIR_ERR_CONF_SYNTAX, + _("Unknown cgroup controller '%s'"), + controllers[i]); + goto cleanup; + } + cfg->cgroupControllers |= (1 << ctl); } - cfg->cgroupControllers |= (1 << ctl); } if (virConfGetValueStringList(conf, "cgroup_device_acl", false, -- 2.7.4

2 1

[libvirt] [PATCH] qemu: avoid parameter named 'listen'
by Daniel P. Berrange 12 Sep '16

12 Sep '16

previous commit: commit 2c3223785c121eafbdd503f11549d7964461934f Author: John Ferlan <jferlan(a)redhat.com> Date: Mon Jun 13 12:30:34 2016 -0400 qemu: Add the ability to hotplug the TLS X.509 environment added a parameter "bool listen" in some methods. This unfortunately clashes with the listen() method, causing compile failures on certain platforms (RHEL-6 for example) Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- Pushed as a CI build fix src/qemu/qemu_command.c | 8 ++++---- src/qemu/qemu_command.h | 2 +- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index e0b0401..3a61863 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -692,7 +692,7 @@ qemuBuildRBDSecinfoURI(virBufferPtr buf, */ int qemuBuildTLSx509BackendProps(const char *tlspath, - bool listen, + bool isListen, bool verifypeer, virQEMUCapsPtr qemuCaps, virJSONValuePtr *propsret) @@ -714,7 +714,7 @@ qemuBuildTLSx509BackendProps(const char *tlspath, if (virJSONValueObjectCreate(propsret, "s:dir", path, - "s:endpoint", (listen ? "server": "client"), + "s:endpoint", (isListen ? "server": "client"), "b:verify-peer", verifypeer, NULL) < 0) goto cleanup; @@ -743,7 +743,7 @@ qemuBuildTLSx509BackendProps(const char *tlspath, static int qemuBuildTLSx509CommandLine(virCommandPtr cmd, const char *tlspath, - bool listen, + bool isListen, bool verifypeer, const char *inalias, virQEMUCapsPtr qemuCaps) @@ -753,7 +753,7 @@ qemuBuildTLSx509CommandLine(virCommandPtr cmd, virJSONValuePtr props = NULL; char *tmp = NULL; - if (qemuBuildTLSx509BackendProps(tlspath, listen, verifypeer, + if (qemuBuildTLSx509BackendProps(tlspath, isListen, verifypeer, qemuCaps, &props) < 0) return -1; diff --git a/src/qemu/qemu_command.h b/src/qemu/qemu_command.h index b7c59bb..0c0b8f1 100644 --- a/src/qemu/qemu_command.h +++ b/src/qemu/qemu_command.h @@ -67,7 +67,7 @@ int qemuBuildSecretInfoProps(qemuDomainSecretInfoPtr secinfo, /* Generate the object properties for a tls-creds-x509 */ int qemuBuildTLSx509BackendProps(const char *tlspath, - bool listen, + bool isListen, bool verifypeer, virQEMUCapsPtr qemuCaps, virJSONValuePtr *propsret); -- 2.7.4

1 0

[libvirt] [PATCH] log: fix false oom error message
by Nikolay Shirokovskiy 12 Sep '16

12 Sep '16

In case filter is new one an oom error message is reported. This bug is introduced in 51b2606f. However this error does not get into the log in the default configuration. To trigger it one need to specify log outputs in LIBVIRT_LOG_OUTPUTS variable. --- src/util/virlog.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/util/virlog.c b/src/util/virlog.c index 06f9a60..f5b88b6 100644 --- a/src/util/virlog.c +++ b/src/util/virlog.c @@ -335,11 +335,12 @@ virLogDefineFilter(const char *match, goto cleanup; virLogFiltersSerial++; + ret = virLogNbFilters - 1; cleanup: virLogUnlock(); if (ret < 0) virReportOOMError(); - return virLogNbFilters; + return ret; } /** -- 1.8.3.1

2 1

[libvirt] [PATCH] qemu: Don't use query-migrate on destination
by Jiri Denemark 12 Sep '16

12 Sep '16

When migration fails, we need to poke QEMU monitor to check for a reason of the failure. We did this using query-migrate QMP command, which is not supposed to return any meaningful result on the destination side. Thus if the monitor was still functional when we detected the migration failure, parsing the answer from query-migrate always failed with the following error message: "info migration reply was missing return status" This irrelevant message was then used as the reason for the migration failure replacing any message we might have had. Let's use harmless query-status for poking the monitor to make sure we only get an error if the monitor connection is broken. https://bugzilla.redhat.com/show_bug.cgi?id=1374613 Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/qemu/qemu_domain.c | 20 ++++++++++++++++++++ src/qemu/qemu_domain.h | 4 ++++ src/qemu/qemu_migration.c | 6 +----- src/qemu/qemu_monitor.c | 8 ++++++++ src/qemu/qemu_monitor.h | 1 + 5 files changed, 34 insertions(+), 5 deletions(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index fd79390..fb766d0 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -6127,3 +6127,23 @@ qemuDomainVcpuPersistOrder(virDomainDefPtr def) } } } + + +int +qemuDomainCheckMonitor(virQEMUDriverPtr driver, + virDomainObjPtr vm, + qemuDomainAsyncJob asyncJob) +{ + qemuDomainObjPrivatePtr priv = vm->privateData; + int ret; + + if (qemuDomainObjEnterMonitorAsync(driver, vm, asyncJob) < 0) + return -1; + + ret = qemuMonitorCheck(priv->mon); + + if (qemuDomainObjExitMonitor(driver, vm) < 0) + return -1; + + return ret; +} diff --git a/src/qemu/qemu_domain.h b/src/qemu/qemu_domain.h index 13c0372..a1404d0 100644 --- a/src/qemu/qemu_domain.h +++ b/src/qemu/qemu_domain.h @@ -728,4 +728,8 @@ bool qemuDomainVcpuHotplugIsInOrder(virDomainDefPtr def) void qemuDomainVcpuPersistOrder(virDomainDefPtr def) ATTRIBUTE_NONNULL(1); +int qemuDomainCheckMonitor(virQEMUDriverPtr driver, + virDomainObjPtr vm, + qemuDomainAsyncJob asyncJob); + #endif /* __QEMU_DOMAIN_H__ */ diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index 07f18db..e734816 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -6208,14 +6208,10 @@ qemuMigrationFinish(virQEMUDriverPtr driver, } if (retcode != 0) { - qemuDomainJobInfo info; - /* Check for a possible error on the monitor in case Finish was called * earlier than monitor EOF handler got a chance to process the error */ - qemuMigrationFetchJobStatus(driver, vm, - QEMU_ASYNC_JOB_MIGRATION_IN, - &info); + qemuDomainCheckMonitor(driver, vm, QEMU_ASYNC_JOB_MIGRATION_IN); goto endjob; } diff --git a/src/qemu/qemu_monitor.c b/src/qemu/qemu_monitor.c index 4171914..1fdee3a 100644 --- a/src/qemu/qemu_monitor.c +++ b/src/qemu/qemu_monitor.c @@ -1617,6 +1617,14 @@ qemuMonitorStopCPUs(qemuMonitorPtr mon) int +qemuMonitorCheck(qemuMonitorPtr mon) +{ + bool running; + return qemuMonitorGetStatus(mon, &running, NULL); +} + + +int qemuMonitorGetStatus(qemuMonitorPtr mon, bool *running, virDomainPausedReason *reason) diff --git a/src/qemu/qemu_monitor.h b/src/qemu/qemu_monitor.h index b838725..255fff2 100644 --- a/src/qemu/qemu_monitor.h +++ b/src/qemu/qemu_monitor.h @@ -382,6 +382,7 @@ typedef enum { VIR_ENUM_DECL(qemuMonitorVMStatus) int qemuMonitorVMStatusToPausedReason(const char *status); +int qemuMonitorCheck(qemuMonitorPtr mon); int qemuMonitorGetStatus(qemuMonitorPtr mon, bool *running, virDomainPausedReason *reason) -- 2.10.0

2 1

[libvirt] [PATCH 0/2] A couple of volume encryption patches
by John Ferlan 12 Sep '16

12 Sep '16

I'm assume there will be a couple of bz's on these... Patch 1 fixes a problem where the vol-dumpxml would not list the <secret> element from a volume in a pool after the initial creation and a pool refresh (or libvirtd restart). This issue was missed due to rewriting the LUKS support to not use it's own "new" secret type (key), but rather use the existing "volume" secret type. Patch 2 fixes a problem where QCOW2 (or QCOW1) encrypted volume would lose the <encryption> and <secret> after a volume refresh. Details are in the patch. The issue is rooted in proper detection of the volume type during the virStorageFileGetMetadataFromBuf call. John Ferlan (2): storage: Need to refresh secret for luks volume after volume refresh storage: Need to properly read the crypt offset value src/storage/storage_backend_fs.c | 16 ++++++++++------ src/util/virstoragefile.c | 7 +++++-- 2 files changed, 15 insertions(+), 8 deletions(-) -- 2.7.4

2 3

[libvirt] [PATCH 0/2] try harder to get dest qemu errors on migation
by Nikolay Shirokovskiy 12 Sep '16

12 Sep '16

Hi, all. In case migration fails due to destination qemu exits unexpectedly user recevies the qemu log in the error message. Unfortunately log is truncated and the most interesting part is missed (below is the example of such a log [1]). Actually for the most cases the first patch will be enough to fix the issue. Originally I thought the problem is qemu logging and reading the log are not in sync (which is true) so I tried to fix it as well in the second patch. [1] log example: CPU Reset (CPU 0) EAX=00000000 EBX=00000000 ECX=00000000 EDX=00000000 ESI=00000000 EDI=00000000 EBP=00000000 ESP=00000000 EIP=00000000 EFL=00000000 [-------] CPL=0 II=0 A20=0 SMM=0 HLT=0 ES =0000 00000000 00000000 00000000 CS =0000 00000000 00000000 00000000 SS =0000 00000000 00000000 00000000 DS =0000 00000000 00000000 00000000 FS =0000 00000000 00000000 00000000 GS =0000 00000000 00000000 00000000 LDT=0000 00000000 00000000 00000000 TR =0000 00000000 00000000 00000000 GDT= 00000000 00000000 IDT= 00000000 00000000 CR0=00000000 CR2=00000000 CR3=00000000 CR4=00000000 DR0=0000000000000000 DR1=0000000000000000 DR2=0000000000000000 DR3=0000000000000000 DR6=0000000000000000 DR7=0000000000000000 CCS=00000000 CCD=00000000 CCO=DYNAMIC EFER=0000000000000000 FCW=0000 FSW=0000 [ST=0] FTW=ff MXCSR=00000000 FPR0=0000000000000000 0000 FPR1=0000000000000000 0000 FPR2=0000000000000000 0000 FPR3=0000000000000000 0000 FPR4=0000000000000000 0000 FPR5=0000000000000000 0000 FPR6=0000000000000000 0000 FPR7=0000000000000000 0000 XMM00=00000000000000000000000000000000 XMM01=00000000000000000000000000000000 XMM02=00000000000000000000000000000000 XMM03=00000000000000000000000000000000 XMM04=00000000000000000000000000000000 XMM05=00000000000000000000000000000000 XMM06=00000000000000000000000000000000 XMM07=00000000000000000000000000000000 CPU Reset (CPU 1) EAX=00000000 EBX=00000000 ECX=00000000 EDX=000206a1 ESI=00000000 EDI=00000000 EBP=00000000 ESP=00000000 EIP=0000fff0 EFL=00000002 [-------] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES =0000 00000000 0000ffff 00009300 CS =f000 ffff0000 0000ffff 00009b00 SS =0000 00000000 0000ffff 00009300 DS =0000 00000000 0000ffff 00009300 FS =0000 00000000 0000ffff 00009300 GS =0000 00000000 0000ffff 00009300 LDT=0000 00000000 0000ffff 00008200 TR =0000 00000000 0000ffff 00008b00 GDT= 00000000 0000ffff IDT= 00000000 0000ffff CR0=60000010 CR2=00000000 CR3=00000000 CR4=00000000 DR0=0000000000000000 DR1=0000000000000000 DR2=0000000000000000 DR3=0000000000000000 DR6=00000000ffff0ff0 DR7=0000000000000400 CCS=00000000 CCD=00000000 CCO=DYNAMIC EFER=0000000000000000 FCW=037f FSW=0000 [ST=0] FTW=00 MXCSR=00001f80 FPR0=0000000000000000 0000 FPR1=0000000000000000 0000 FPR2=0000000000000000 0000 FPR3=0000000000000000 0000 FPR4=0000000000000000 0000 FPR5=0000000000000000 0000 FPR6=0000000000000000 0000 FPR7=0000000000000000 0000 XMM00=00000000000000000000000000000000 XMM01=00000000000000000000000000000000 XMM02=00000000000000000000000000000000 XMM03=00000000000000000000000000000000 XMM04=00000000000000000000000000000000 XMM05=00000000000000000000000000000000 XMM06=00000000000000000000000000000000 XMM07=000 qemu: terminating on signal 15 from pid 168133 Nikolay Shirokovskiy (2): log: remove 1k limit for error messages log: drain log of exiting qemu process carefully src/logging/log_handler.c | 39 +++++++++++++++++++++++++++++++++++++-- src/qemu/qemu_domain.c | 4 ++++ src/qemu/qemu_domain.h | 1 + src/qemu/qemu_process.c | 2 ++ src/util/virerror.c | 9 ++++----- 5 files changed, 48 insertions(+), 7 deletions(-) -- 1.8.3.1

2 3

[libvirt] [PATCH v6 0/5] Add native TLS encrypted chardev TCP support
by John Ferlan 12 Sep '16

12 Sep '16

v5: http://www.redhat.com/archives/libvir-list/2016-August/msg00282.html Patches 1-5 from that series already pushed Patch 6 from that series is removed Patch 7 is untouched and is patch 3 in this series Patches 8-9 modified for new paradigm (patches 4-5 of this series) Patch 1 [NEW] From patch 6 review comment - provide a way to allow disabling the TLS for a specific guest and serial TCP chardev. A new test was added which shows that the xml2argv output does not add the TLS object within the same "frame" as the test that would add it. Patch 2 [NEW] Rather than use a <secret> object for the serial TCP chardev, create a qemu.conf option as a host wide setting. The patch implements a 'chardev_tls_x509_secret_uuid' value which will be used by patches 4-5 rather than the serial TCP specific setting. Patch 3 is unchanged Patches 4-5 were adjusted in order to use the chardevTCPtlsx509secretUUID rather than the chardev TCP specific 'chardev->source.data.tcp.seclookupdef' value sourced from a <secret>. John Ferlan (5): domain: Add optional 'disableTLS' attribute for TCP chardev conf: Introduce {default|chardev}_tls_x509_secret_uuid qemu: Introduce qemuDomainChardevPrivatePtr qemu: Add a secret object to/for a chardev tcp with secret qemu: Add the ability to hotplug a secret object for TCP chardev TLS docs/formatdomain.html.in | 20 +++ docs/schemas/domaincommon.rng | 5 + src/conf/domain_conf.c | 51 ++++++-- src/conf/domain_conf.h | 5 +- src/libxl/libxl_domain.c | 2 +- src/lxc/lxc_native.c | 2 +- src/qemu/libvirtd_qemu.aug | 2 + src/qemu/qemu.conf | 24 ++++ src/qemu/qemu_command.c | 34 ++++- src/qemu/qemu_command.h | 1 + src/qemu/qemu_conf.c | 22 ++++ src/qemu/qemu_conf.h | 3 + src/qemu/qemu_domain.c | 143 ++++++++++++++++++++- src/qemu/qemu_domain.h | 30 ++++- src/qemu/qemu_driver.c | 2 +- src/qemu/qemu_hotplug.c | 63 ++++++++- src/qemu/qemu_hotplug.h | 3 +- src/qemu/qemu_parse_command.c | 4 +- src/qemu/qemu_process.c | 6 +- src/qemu/test_libvirtd_qemu.aug.in | 2 + src/vz/vz_sdk.c | 2 +- src/xenconfig/xen_sxpr.c | 2 +- tests/qemuhotplugtest.c | 2 +- ...argv-serial-tcp-tlsx509-chardev-disableTLS.args | 30 +++++ ...2argv-serial-tcp-tlsx509-chardev-disableTLS.xml | 50 +++++++ ...xml2argv-serial-tcp-tlsx509-secret-chardev.args | 38 ++++++ ...uxml2argv-serial-tcp-tlsx509-secret-chardev.xml | 50 +++++++ tests/qemuxml2argvtest.c | 22 ++++ ...mlout-serial-tcp-tlsx509-chardev-disableTLS.xml | 1 + tests/qemuxml2xmltest.c | 1 + 30 files changed, 588 insertions(+), 34 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-serial-tcp-tlsx509-chardev-disableTLS.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-serial-tcp-tlsx509-chardev-disableTLS.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-serial-tcp-tlsx509-secret-chardev.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-serial-tcp-tlsx509-secret-chardev.xml create mode 120000 tests/qemuxml2xmloutdata/qemuxml2xmlout-serial-tcp-tlsx509-chardev-disableTLS.xml -- 2.7.4

2 7

[libvirt] [PATCH 0/2] try harder to get dest qemu errors on migation
by Nikolay Shirokovskiy 12 Sep '16

12 Sep '16

Hi, all. In case migration fails due to destination qemu exits unexpectedly user recevies the qemu log in the error message. Unfortunately log is truncated and the most interesting part is missed (below is the example of such a log [1]). Actually for the most cases the first patch will be enough to fix the issue. Originally I thought the problem is qemu logging and reading the log are not in sync (which is true) so I tried to fix it as well in the second patch. [1] log example: CPU Reset (CPU 0) EAX=00000000 EBX=00000000 ECX=00000000 EDX=00000000 ESI=00000000 EDI=00000000 EBP=00000000 ESP=00000000 EIP=00000000 EFL=00000000 [-------] CPL=0 II=0 A20=0 SMM=0 HLT=0 ES =0000 00000000 00000000 00000000 CS =0000 00000000 00000000 00000000 SS =0000 00000000 00000000 00000000 DS =0000 00000000 00000000 00000000 FS =0000 00000000 00000000 00000000 GS =0000 00000000 00000000 00000000 LDT=0000 00000000 00000000 00000000 TR =0000 00000000 00000000 00000000 GDT= 00000000 00000000 IDT= 00000000 00000000 CR0=00000000 CR2=00000000 CR3=00000000 CR4=00000000 DR0=0000000000000000 DR1=0000000000000000 DR2=0000000000000000 DR3=0000000000000000 DR6=0000000000000000 DR7=0000000000000000 CCS=00000000 CCD=00000000 CCO=DYNAMIC EFER=0000000000000000 FCW=0000 FSW=0000 [ST=0] FTW=ff MXCSR=00000000 FPR0=0000000000000000 0000 FPR1=0000000000000000 0000 FPR2=0000000000000000 0000 FPR3=0000000000000000 0000 FPR4=0000000000000000 0000 FPR5=0000000000000000 0000 FPR6=0000000000000000 0000 FPR7=0000000000000000 0000 XMM00=00000000000000000000000000000000 XMM01=00000000000000000000000000000000 XMM02=00000000000000000000000000000000 XMM03=00000000000000000000000000000000 XMM04=00000000000000000000000000000000 XMM05=00000000000000000000000000000000 XMM06=00000000000000000000000000000000 XMM07=00000000000000000000000000000000 CPU Reset (CPU 1) EAX=00000000 EBX=00000000 ECX=00000000 EDX=000206a1 ESI=00000000 EDI=00000000 EBP=00000000 ESP=00000000 EIP=0000fff0 EFL=00000002 [-------] CPL=0 II=0 A20=1 SMM=0 HLT=0 ES =0000 00000000 0000ffff 00009300 CS =f000 ffff0000 0000ffff 00009b00 SS =0000 00000000 0000ffff 00009300 DS =0000 00000000 0000ffff 00009300 FS =0000 00000000 0000ffff 00009300 GS =0000 00000000 0000ffff 00009300 LDT=0000 00000000 0000ffff 00008200 TR =0000 00000000 0000ffff 00008b00 GDT= 00000000 0000ffff IDT= 00000000 0000ffff CR0=60000010 CR2=00000000 CR3=00000000 CR4=00000000 DR0=0000000000000000 DR1=0000000000000000 DR2=0000000000000000 DR3=0000000000000000 DR6=00000000ffff0ff0 DR7=0000000000000400 CCS=00000000 CCD=00000000 CCO=DYNAMIC EFER=0000000000000000 FCW=037f FSW=0000 [ST=0] FTW=00 MXCSR=00001f80 FPR0=0000000000000000 0000 FPR1=0000000000000000 0000 FPR2=0000000000000000 0000 FPR3=0000000000000000 0000 FPR4=0000000000000000 0000 FPR5=0000000000000000 0000 FPR6=0000000000000000 0000 FPR7=0000000000000000 0000 XMM00=00000000000000000000000000000000 XMM01=00000000000000000000000000000000 XMM02=00000000000000000000000000000000 XMM03=00000000000000000000000000000000 XMM04=00000000000000000000000000000000 XMM05=00000000000000000000000000000000 XMM06=00000000000000000000000000000000 XMM07=000 qemu: terminating on signal 15 from pid 168133 Nikolay Shirokovskiy (2): log: remove 1k limit for error messages log: drain log of exiting qemu process carefully src/logging/log_handler.c | 39 +++++++++++++++++++++++++++++++++++++-- src/qemu/qemu_domain.c | 4 ++++ src/qemu/qemu_domain.h | 1 + src/qemu/qemu_process.c | 2 ++ src/util/virerror.c | 9 ++++----- 5 files changed, 48 insertions(+), 7 deletions(-) -- 1.8.3.1

1 0

[libvirt] [gconfig v2 0/6] Add support for changing listen address
by Visarion Alexandru 12 Sep '16

12 Sep '16

From: Visarion Alexandru <viorel.visarion(a)gmail.com> libvirt-domain-graphics-listen defines an abstract class used to represent one listen node. libvirt-domain-graphics-liste-address defines a class that represents a listen node of type 'address', for the graphics device. We have to first remove the 'listen' attribute of the graphics device before adding child 'listen' nodes, to avoid inconsistencies check. I added a test for the new API which uses a SPICE graphics (I noticed that generally APIs aren't tested on both SPICE and VNC) Visarion Alexandru (6): libvirt-gconfig: Include GIO dependency Introduce libvirt-domain-graphics-listen Introduce libvirt-gconfig-domain-graphics-listen-address config: Add vnc listen getter/setter config: Add spice listen getter/setter tests: Add test for GVirConfigDomainGraphicsListenAddress configure.ac | 2 + libvirt-gconfig-1.0.pc.in | 2 +- libvirt-gconfig/Makefile.am | 10 +- ...ibvirt-gconfig-domain-graphics-listen-address.c | 128 +++++++++++++++++++++ ...ibvirt-gconfig-domain-graphics-listen-address.h | 78 +++++++++++++ .../libvirt-gconfig-domain-graphics-listen.c | 49 ++++++++ .../libvirt-gconfig-domain-graphics-listen.h | 65 +++++++++++ .../libvirt-gconfig-domain-graphics-spice.c | 84 ++++++++++++++ .../libvirt-gconfig-domain-graphics-spice.h | 4 + .../libvirt-gconfig-domain-graphics-vnc.c | 84 ++++++++++++++ .../libvirt-gconfig-domain-graphics-vnc.h | 4 + libvirt-gconfig/libvirt-gconfig.h | 2 + libvirt-gconfig/libvirt-gconfig.sym | 12 ++ tests/test-gconfig.c | 36 ++++++ .../xml/gconfig-domain-device-graphics-listen.xml | 7 ++ vapi/Makefile.am | 1 + 16 files changed, 565 insertions(+), 3 deletions(-) create mode 100644 libvirt-gconfig/libvirt-gconfig-domain-graphics-listen-address.c create mode 100644 libvirt-gconfig/libvirt-gconfig-domain-graphics-listen-address.h create mode 100644 libvirt-gconfig/libvirt-gconfig-domain-graphics-listen.c create mode 100644 libvirt-gconfig/libvirt-gconfig-domain-graphics-listen.h create mode 100644 tests/xml/gconfig-domain-device-graphics-listen.xml -- 2.7.4

3 14

[libvirt] [PATCH] util: Quiet the logging if perf file doesn't exist
by John Ferlan 12 Sep '16

12 Sep '16

Commit id 'b00d7f29' shifted the opening of the /sys/devices/intel_cqm/type file from event enable to perf event initialization. If the file did not exist, then an error would be written to the domain log: 2016-09-06 20:51:21.677+0000: 7310: error : virFileReadAll:1360 : Failed to open file '/sys/devices/intel_cqm/type': No such file or directory Since the error is now handled in virPerfEventEnable by checking if the event_attr->attrType == 0 for CMT, MBML, and MBMT events - we can just use the Quiet API in order to not log the error we're going to throw away. Additionally, rather than using virReportSystemError, use virReportError and VIR_ERR_ARGUMENT_UNSUPPORTED in order to signify that support isn't there for that type of perf event - adjust the error message as well. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- src/util/virperf.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/src/util/virperf.c b/src/util/virperf.c index cd66b05..a65a4bf 100644 --- a/src/util/virperf.c +++ b/src/util/virperf.c @@ -116,7 +116,7 @@ virPerfRdtAttrInit(void) char *tmp = NULL; unsigned int attr_type = 0; - if (virFileReadAll("/sys/devices/intel_cqm/type", 10, &buf) < 0) + if (virFileReadAllQuiet("/sys/devices/intel_cqm/type", 10, &buf) < 0) goto error; if ((tmp = strchr(buf, '\n'))) @@ -174,9 +174,9 @@ virPerfEventEnable(virPerfPtr perf, if (event_attr->attrType == 0 && (type == VIR_PERF_EVENT_CMT || type == VIR_PERF_EVENT_MBMT || type == VIR_PERF_EVENT_MBML)) { - virReportSystemError(errno, - _("Unable to open perf event for %s"), - virPerfEventTypeToString(event->type)); + virReportError(VIR_ERR_ARGUMENT_UNSUPPORTED, + _("unable to enable perf event for %s"), + virPerfEventTypeToString(event->type)); return -1; } -- 2.7.4

2 1

[libvirt] [PATCH 0/2] qemu: hotplug: Fix two issues with drives made empty by using starup policy
by Peter Krempa 12 Sep '16

12 Sep '16

If a cdrom drive is emptied by 'requisite' startup policy the policy setting would not be cleared and due to a bug in qemu the cdrom image could not be changed. The first patch clears the startup policy setting and the second one fixes ejection of cdroms if --force is specified. Also I've reported the issue in qemu https://bugzilla.redhat.com/show_bug.cgi?id=1373264 Peter Krempa (2): qemu: domain: Clear startup policy for dropped removable media qemu: hotplug: Don't wait if cdrom tray is opened forcibly src/qemu/qemu_domain.c | 2 ++ src/qemu/qemu_hotplug.c | 9 ++++----- 2 files changed, 6 insertions(+), 5 deletions(-) -- 2.9.2

2 3

[libvirt] [PATCH v2 00/10] Introduce NVDIMM support
by Michal Privoznik 09 Sep '16

09 Sep '16

This is v2 of: https://www.redhat.com/archives/libvir-list/2016-August/msg00055.html diff to v1: - Rebase to the latest HEAD - Introduced relabeling for host side of NVDIMM Michal Privoznik (10): Introduce NVDIMM memory model qemu: Introduce QEMU_CAPS_DEVICE_NVDIMM qemu: Implement NVDIMM conf: Introduce memAccess to <memory/> qemu: Implement memAccess for <memory/> banks security_dac: Label host side of NVDIMM security_selinux: Label host side of NVDIMM security: Introduce internal APIs for memdev labelling secdrivers: Implement memdev relabel APIs qemu_hotplug: Relabel memdev docs/formatdomain.html.in | 41 ++++++-- docs/schemas/domaincommon.rng | 51 ++++++---- src/conf/domain_conf.c | 112 ++++++++++++++++----- src/conf/domain_conf.h | 4 + src/libvirt_private.syms | 4 + src/qemu/qemu_alias.c | 12 ++- src/qemu/qemu_capabilities.c | 2 + src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 87 +++++++++++----- src/qemu/qemu_command.h | 2 + src/qemu/qemu_domain.c | 29 ++++-- src/qemu/qemu_hotplug.c | 17 +++- src/security/security_dac.c | 76 ++++++++++++++ src/security/security_driver.h | 9 ++ src/security/security_manager.c | 56 +++++++++++ src/security/security_manager.h | 7 ++ src/security/security_nop.c | 19 ++++ src/security/security_selinux.c | 69 +++++++++++++ src/security/security_stack.c | 38 +++++++ tests/qemucapabilitiesdata/caps_2.6.0.x86_64.xml | 1 + tests/qemucapabilitiesdata/caps_2.7.0.x86_64.xml | 1 + .../qemuxml2argv-hugepages-numa.args | 5 +- .../qemuxml2argv-hugepages-pages.args | 24 ++--- .../qemuxml2argv-hugepages-pages2.args | 8 +- .../qemuxml2argv-hugepages-pages3.args | 4 +- .../qemuxml2argv-hugepages-shared.args | 22 ++-- .../qemuxml2argv-memory-hotplug-dimm-addr.args | 5 +- .../qemuxml2argv-memory-hotplug-dimm.args | 5 +- ...muxml2argv-memory-hotplug-nvdimm-memAccess.args | 26 +++++ ...emuxml2argv-memory-hotplug-nvdimm-memAccess.xml | 49 +++++++++ .../qemuxml2argv-memory-hotplug-nvdimm.args | 25 +++++ .../qemuxml2argv-memory-hotplug-nvdimm.xml | 49 +++++++++ tests/qemuxml2argvtest.c | 6 +- 33 files changed, 743 insertions(+), 123 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-memory-hotplug-nvdimm-memAccess.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-memory-hotplug-nvdimm-memAccess.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-memory-hotplug-nvdimm.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-memory-hotplug-nvdimm.xml -- 2.8.4

4 22

[libvirt] [PATCH v5 0/9] Add native TLS encrypted chardev TCP support
by John Ferlan 09 Sep '16

09 Sep '16

v4: http://www.redhat.com/archives/libvir-list/2016-June/msg01709.html Since I have it on a branch and have been updating, I figured I'd post the most recent stuff. Patches 1-4 were "partially" ACK'd in v2 of this series, but there's been changes to the conf handling upstream. Patch 5 adds a new secret type 'tls'. Previous incarnations of these changes borrowed a common secret type, but this one is specific. It's more or less what got removed for LUKS with the names changed to protect the innocent (reference Dragnet). Patches 6-9 is what was mostly missing in the earlier series. Differences to v4... mostly updates/merges with the numerous changes to master since that time. I lost track of whether the desire was to have /etc/pki/libvirt-%s or /etc/pki/qemu-%s directories... I think we've been using the libvirt-%s for vnc/spice so far, so I just followed that for chardev although there was a comment at one time to use qemu-chardev during review of patch 2 of the v2 series: http://www.redhat.com/archives/libvir-list/2016-June/msg01072.html John Ferlan (9): conf: Add new default TLS X.509 certificate default directory conf: Introduce chartcp_tls_x509_cert_dir qemu: Add support for TLS X.509 path to TCP chardev backend qemu: Add the ability to hotplug the TLS X.509 environment conf: Add new secret type "tls" conf: Add new secret element for tcp chardev qemu: Introduce qemuDomainChardevPrivatePtr qemu: Add a secret object to/for a chardev tcp with secret qemu: Add the ability to hotplug a secret object for TCP chardev TLS docs/aclpolkit.html.in | 4 + docs/formatdomain.html.in | 29 +++++ docs/formatsecret.html.in | 59 ++++++++- docs/schemas/domaincommon.rng | 21 +++ docs/schemas/secret.rng | 10 ++ include/libvirt/libvirt-secret.h | 1 + src/access/viraccessdriverpolkit.c | 13 ++ src/conf/domain_conf.c | 64 ++++++++-- src/conf/domain_conf.h | 8 +- src/conf/secret_conf.c | 23 +++- src/conf/secret_conf.h | 1 + src/conf/virsecretobj.c | 5 + src/libxl/libxl_domain.c | 2 +- src/lxc/lxc_native.c | 2 +- src/qemu/libvirtd_qemu.aug | 11 +- src/qemu/qemu.conf | 83 +++++++++--- src/qemu/qemu_alias.c | 16 +++ src/qemu/qemu_alias.h | 3 + src/qemu/qemu_command.c | 141 ++++++++++++++++++++- src/qemu/qemu_command.h | 9 ++ src/qemu/qemu_conf.c | 57 ++++++++- src/qemu/qemu_conf.h | 7 + src/qemu/qemu_domain.c | 124 +++++++++++++++++- src/qemu/qemu_domain.h | 22 ++++ src/qemu/qemu_driver.c | 2 +- src/qemu/qemu_hotplug.c | 104 ++++++++++++++- src/qemu/qemu_hotplug.h | 3 +- src/qemu/qemu_monitor_json.c | 9 ++ src/qemu/qemu_parse_command.c | 4 +- src/qemu/qemu_process.c | 2 +- src/qemu/test_libvirtd_qemu.aug.in | 5 + src/vz/vz_sdk.c | 2 +- src/xenconfig/xen_sxpr.c | 2 +- tests/qemuhotplugtest.c | 2 +- .../qemuxml2argv-serial-tcp-tlsx509-chardev.args | 33 +++++ .../qemuxml2argv-serial-tcp-tlsx509-chardev.xml | 41 ++++++ ...xml2argv-serial-tcp-tlsx509-secret-chardev.args | 38 ++++++ ...uxml2argv-serial-tcp-tlsx509-secret-chardev.xml | 51 ++++++++ tests/qemuxml2argvtest.c | 21 +++ .../qemuxml2xmlout-serial-tcp-tlsx509-chardev.xml | 50 ++++++++ ...ml2xmlout-serial-tcp-tlsx509-secret-chardev.xml | 1 + tests/qemuxml2xmltest.c | 2 + tests/secretxml2xmlin/usage-tls.xml | 7 + tests/secretxml2xmltest.c | 1 + 44 files changed, 1038 insertions(+), 57 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-serial-tcp-tlsx509-chardev.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-serial-tcp-tlsx509-chardev.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-serial-tcp-tlsx509-secret-chardev.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-serial-tcp-tlsx509-secret-chardev.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-serial-tcp-tlsx509-chardev.xml create mode 120000 tests/qemuxml2xmloutdata/qemuxml2xmlout-serial-tcp-tlsx509-secret-chardev.xml create mode 100644 tests/secretxml2xmlin/usage-tls.xml -- 2.7.4

2 18

Re: [libvirt] [PATCH v7 0/4] Add Mediated device support
by Alex Williamson 08 Sep '16

08 Sep '16

Hi folks, At KVM Forum we had a BoF session primarily around the mediated device sysfs interface. I'd like to share what I think we agreed on and the "problem areas" that still need some work so we can get the thoughts and ideas from those who weren't able to attend. DanPB expressed some concern about the mdev_supported_types sysfs interface, which exposes a flat csv file with fields like "type", "number of instance", "vendor string", and then a bunch of type specific fields like "framebuffer size", "resolution", "frame rate limit", etc. This is not entirely machine parsing friendly and sort of abuses the sysfs concept of one value per file. Example output taken from Neo's libvirt RFC: cat /sys/bus/pci/devices/0000:86:00.0/mdev_supported_types # vgpu_type_id, vgpu_type, max_instance, num_heads, frl_config, framebuffer, max_resolution 11 ,"GRID M60-0B", 16, 2, 45, 512M, 2560x1600 12 ,"GRID M60-0Q", 16, 2, 60, 512M, 2560x1600 13 ,"GRID M60-1B", 8, 2, 45, 1024M, 2560x1600 14 ,"GRID M60-1Q", 8, 2, 60, 1024M, 2560x1600 15 ,"GRID M60-2B", 4, 2, 45, 2048M, 2560x1600 16 ,"GRID M60-2Q", 4, 4, 60, 2048M, 2560x1600 17 ,"GRID M60-4Q", 2, 4, 60, 4096M, 3840x2160 18 ,"GRID M60-8Q", 1, 4, 60, 8192M, 3840x2160 The create/destroy then looks like this: echo "$mdev_UUID:vendor_specific_argument_list" > /sys/bus/pci/devices/.../mdev_create echo "$mdev_UUID:vendor_specific_argument_list" > /sys/bus/pci/devices/.../mdev_destroy "vendor_specific_argument_list" is nebulous. So the idea to fix this is to explode this into a directory structure, something like: ├── mdev_destroy └── mdev_supported_types ├── 11 │ ├── create │ ├── description │ └── max_instances ├── 12 │ ├── create │ ├── description │ └── max_instances └── 13 ├── create ├── description └── max_instances Note that I'm only exposing the minimal attributes here for simplicity, the other attributes would be included in separate files and we would require vendors to create standard attributes for common device classes. For vGPUs like NVIDIA where we don't support multiple types concurrently, this directory structure would update as mdev devices are created, removing no longer available types. I carried forward max_instances here, but perhaps we really want to copy SR-IOV and report a max and current allocation. Creation and deletion is simplified as we can simply "echo $UUID > create" per type. I don't understand why destroy had a parameter list, so here I imagine we can simply do the same... in fact, I'd actually rather see a "remove" sysfs entry under each mdev device, so we remove it at the device rather than in some central location (any objections?). We discussed how this might look with Intel devices which do allow mixed vGPU types concurrently. We believe, but need confirmation, that the vendor driver could still make a finite set of supported types, perhaps with additional module options to the vendor driver to enable more "exotic" types. So for instance if IGD vGPUs are based on power-of-2 portions of the framebuffer size, then the vendor driver could list types with 32MB, 64MB, 128MB, etc in useful and popular sizes. As vGPUs are allocated, the larger sizes may become unavailable. We still don't have any way for the admin to learn in advance how the available supported types will change once mdev devices start to be created. I'm not sure how we can create a specification for this, so probing by creating devices may be the most flexible model. The other issue is the start/stop requirement, which was revealed to setup peer-to-peer resources between vGPUs which is a limited hardware resource. We'd really like to have these happen automatically on the first open of a vfio mdev device file and final release. So we brainstormed how the open/release callbacks could know the other mdev devices for a given user. This is where the instance number came into play previously. This is an area that needs work. There was a thought that perhaps on open() the vendor driver could look at the user pid and use that to associate with other devices, but the problem here is that we open and begin access to each device, so devices do this discovery serially rather than in parallel as desired. (we might not fault in mmio space yet though, so I wonder if open() could set the association of mdev to pid, then the first mmio fault would trigger the resource allocation? Then all the "magic" would live in the vendor driver. open() could fail if the pid already has running mdev devices and the vendor driver chooses not to support hotplug) One comment was that for a GPU that only supports homogeneous vGPUs, libvirt may choose to create all the vGPUs in advance and handle them as we do SR-IOV VFs. The UUID+instance model would preclude such a use case. We also considered whether iommu groups could be (ab)used for this use case, peer-to-peer would in fact be an iommu grouping constraint afterall. This would have the same UUID+instance constraint as above though and would require some sort of sysfs interface for the user to be able to create multiple mdevs within a group. Everyone was given homework to think about this on their flights home, so I expect plenty of ideas by now ;) Overall I think mediated devices were well received by the community, so let's keep up the development and discussion to bring it to fruition. Thanks, Alex

11 48

[libvirt] [PATCH] usb: allow host devices to be specified by port
by Thomas Hebb 08 Sep '16

08 Sep '16

Previously, only VID/PID and bus/device matching were supported. Neither of these provide a stable and persistent way of assigning a guest a specific host device out of several with the same VID and PID, as device numbers change on every enumeration. Add a third method of matching, bus/port, which allows a specific port on the host to be specified using the dotted notation found in Linux's "devpath" sysfs attribute. --- src/conf/domain_conf.c | 36 ++++- src/conf/domain_conf.h | 2 + src/libvirt_private.syms | 2 - src/util/virhostdev.c | 126 ++++++++-------- src/util/virusb.c | 160 ++++++++------------- src/util/virusb.h | 23 ++- tests/virusbtest.c | 124 ++++++++++------ .../sys_bus_usb/devices/1-1.5.3.1/devpath | 1 + .../sys_bus_usb/devices/1-1.5.3.3/devpath | 1 + .../sys_bus_usb/devices/1-1.5.3/devpath | 1 + .../sys_bus_usb/devices/1-1.5.4/devpath | 1 + .../sys_bus_usb/devices/1-1.5.5/devpath | 1 + .../sys_bus_usb/devices/1-1.5.6/devpath | 1 + .../sys_bus_usb/devices/1-1.5/devpath | 1 + .../sys_bus_usb/devices/1-1.6/devpath | 1 + .../virusbtestdata/sys_bus_usb/devices/1-1/devpath | 1 + .../sys_bus_usb/devices/2-1.2/devpath | 1 + .../virusbtestdata/sys_bus_usb/devices/2-1/devpath | 1 + .../sys_bus_usb/devices/usb1/devpath | 1 + .../sys_bus_usb/devices/usb2/devpath | 1 + .../sys_bus_usb/devices/usb3/devpath | 1 + .../sys_bus_usb/devices/usb4/devpath | 1 + 22 files changed, 266 insertions(+), 222 deletions(-) create mode 100644 tests/virusbtestdata/sys_bus_usb/devices/1-1.5.3.1/devpath create mode 100644 tests/virusbtestdata/sys_bus_usb/devices/1-1.5.3.3/devpath create mode 100644 tests/virusbtestdata/sys_bus_usb/devices/1-1.5.3/devpath create mode 100644 tests/virusbtestdata/sys_bus_usb/devices/1-1.5.4/devpath create mode 100644 tests/virusbtestdata/sys_bus_usb/devices/1-1.5.5/devpath create mode 100644 tests/virusbtestdata/sys_bus_usb/devices/1-1.5.6/devpath create mode 100644 tests/virusbtestdata/sys_bus_usb/devices/1-1.5/devpath create mode 100644 tests/virusbtestdata/sys_bus_usb/devices/1-1.6/devpath create mode 100644 tests/virusbtestdata/sys_bus_usb/devices/1-1/devpath create mode 100644 tests/virusbtestdata/sys_bus_usb/devices/2-1.2/devpath create mode 100644 tests/virusbtestdata/sys_bus_usb/devices/2-1/devpath create mode 100644 tests/virusbtestdata/sys_bus_usb/devices/usb1/devpath create mode 100644 tests/virusbtestdata/sys_bus_usb/devices/usb2/devpath create mode 100644 tests/virusbtestdata/sys_bus_usb/devices/usb3/devpath create mode 100644 tests/virusbtestdata/sys_bus_usb/devices/usb4/devpath diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 9f7b906..a587cc2 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -5531,7 +5531,8 @@ virDomainHostdevSubsysUSBDefParseXML(xmlNodePtr node, goto out; } } else if (xmlStrEqual(cur->name, BAD_CAST "address")) { - char *bus, *device; + char *bus, *device, *port; + bool dev_or_port = false; bus = virXMLPropString(cur, "bus"); if (bus) { @@ -5557,10 +5558,24 @@ virDomainHostdevSubsysUSBDefParseXML(xmlNodePtr node, VIR_FREE(device); goto out; } + dev_or_port = true; VIR_FREE(device); - } else { + } + + port = virXMLPropString(cur, "port"); + if (port) { + if (*port) { + usbsrc->port = port; + dev_or_port = true; + } else { + VIR_FREE(port); + } + } + + if (!dev_or_port) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", - _("usb address needs device id")); + _("usb address needs either device id " + "or port")); goto out; } } else { @@ -20326,10 +20341,17 @@ virDomainHostdevDefFormatSubsys(virBufferPtr buf, virBufferAsprintf(buf, "<vendor id='0x%.4x'/>\n", usbsrc->vendor); virBufferAsprintf(buf, "<product id='0x%.4x'/>\n", usbsrc->product); } - if (usbsrc->bus || usbsrc->device) { - virBufferAsprintf(buf, "<address %sbus='%d' device='%d'/>\n", - includeTypeInAddr ? "type='usb' " : "", - usbsrc->bus, usbsrc->device); + if (usbsrc->bus || usbsrc->device || usbsrc->port) { + virBufferAddLit(buf, "<address"); + if (includeTypeInAddr) + virBufferAddLit(buf, " type='usb'"); + if (usbsrc->bus) + virBufferAsprintf(buf, " bus='%u'", usbsrc->bus); + if (usbsrc->device) + virBufferAsprintf(buf, " device='%u'", usbsrc->device); + if (usbsrc->port) + virBufferAsprintf(buf, " port='%s'", usbsrc->port); + virBufferAddLit(buf, "/>\n"); } break; case VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI: diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index ba0ad5f..91bd621 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -317,6 +317,8 @@ VIR_ENUM_DECL(virDomainHostdevSubsysSCSIProtocol) typedef struct _virDomainHostdevSubsysUSB virDomainHostdevSubsysUSB; typedef virDomainHostdevSubsysUSB *virDomainHostdevSubsysUSBPtr; struct _virDomainHostdevSubsysUSB { + char *port; + bool autoAddress; /* bus/device were filled automatically based on vendor/product */ unsigned bus; diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index ccb4c5e..d412eac 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -2511,8 +2511,6 @@ virURIResolveAlias; # util/virusb.h virUSBDeviceFileIterate; virUSBDeviceFind; -virUSBDeviceFindByBus; -virUSBDeviceFindByVendor; virUSBDeviceFree; virUSBDeviceGetBus; virUSBDeviceGetDevno; diff --git a/src/util/virhostdev.c b/src/util/virhostdev.c index 9b5ca6f..e809c97 100644 --- a/src/util/virhostdev.c +++ b/src/util/virhostdev.c @@ -1179,89 +1179,95 @@ virHostdevMarkUSBDevices(virHostdevManagerPtr mgr, static int +virHostdevFindUSBDeviceWithFlags(virDomainHostdevDefPtr hostdev, + bool mandatory, + unsigned int flags, + virUSBDevicePtr *usb) +{ + virDomainHostdevSubsysUSBPtr usbsrc = &hostdev->source.subsys.u.usb; + unsigned vendor = usbsrc->vendor; + unsigned product = usbsrc->product; + unsigned bus = usbsrc->bus; + char *port = usbsrc->port; + unsigned device = usbsrc->device; + virUSBDeviceListPtr devs; + int rc; + + rc = virUSBDeviceFind(vendor, product, bus, device, port, NULL, + mandatory, flags, &devs); + if (rc < 0) + return -1; + + if (rc == 1) { + *usb = virUSBDeviceListGet(devs, 0); + virUSBDeviceListSteal(devs, *usb); + } + virObjectUnref(devs); + + if (rc > 1) { + virReportError(VIR_ERR_OPERATION_FAILED, + _("Multiple USB devices for %x:%x, " + "use <address> to specify one"), + vendor, product); + return -1; + } + + return rc; +} + +static int virHostdevFindUSBDevice(virDomainHostdevDefPtr hostdev, bool mandatory, virUSBDevicePtr *usb) { virDomainHostdevSubsysUSBPtr usbsrc = &hostdev->source.subsys.u.usb; unsigned vendor = usbsrc->vendor; - unsigned product = usbsrc->product; unsigned bus = usbsrc->bus; unsigned device = usbsrc->device; + char *port = usbsrc->port; bool autoAddress = usbsrc->autoAddress; + unsigned int flags = 0; int rc; *usb = NULL; - if (vendor && bus) { - rc = virUSBDeviceFind(vendor, product, bus, device, - NULL, - autoAddress ? false : mandatory, - usb); - if (rc < 0) { - return -1; - } else if (!autoAddress) { - goto out; - } else { - VIR_INFO("USB device %x:%x could not be found at previous" - " address (bus:%u device:%u)", - vendor, product, bus, device); - } - } + /* First attempt, matching on all known fields. */ + if (vendor) + flags |= USB_DEVICE_FIND_BY_VENDOR; + if (device) + flags |= USB_DEVICE_FIND_BY_DEVICE; + if (port) + flags |= USB_DEVICE_FIND_BY_PORT; + + rc = virHostdevFindUSBDeviceWithFlags(hostdev, + autoAddress ? false : mandatory, + flags, usb); + if (rc < 0) + return -1; - /* When vendor is specified, its USB address is either unspecified or the - * device could not be found at the USB device where it had been - * automatically found before. - */ - if (vendor) { - virUSBDeviceListPtr devs; + if (rc != 1 && autoAddress) { + VIR_INFO("USB device could not be found at previous address " + "(bus:%u device:%u)", bus, device); - rc = virUSBDeviceFindByVendor(vendor, product, NULL, mandatory, &devs); - if (rc < 0) - return -1; + /* Second attempt, for when the device number has changed. */ + flags &= ~((unsigned int) USB_DEVICE_FIND_BY_DEVICE); + usbsrc->device = 0; - if (rc == 1) { - *usb = virUSBDeviceListGet(devs, 0); - virUSBDeviceListSteal(devs, *usb); - } - virObjectUnref(devs); - - if (rc == 0) { - goto out; - } else if (rc > 1) { - if (autoAddress) { - virReportError(VIR_ERR_OPERATION_FAILED, - _("Multiple USB devices for %x:%x were found," - " but none of them is at bus:%u device:%u"), - vendor, product, bus, device); - } else { - virReportError(VIR_ERR_OPERATION_FAILED, - _("Multiple USB devices for %x:%x, " - "use <address> to specify one"), - vendor, product); - } + rc = virHostdevFindUSBDeviceWithFlags(hostdev, mandatory, + flags, usb); + + if (rc < 0) return -1; - } + } + if (!*usb) { + hostdev->missing = true; + } else if (!usbsrc->bus || !usbsrc->device) { usbsrc->bus = virUSBDeviceGetBus(*usb); usbsrc->device = virUSBDeviceGetDevno(*usb); usbsrc->autoAddress = true; - - if (autoAddress) { - VIR_INFO("USB device %x:%x found at bus:%u device:%u (moved" - " from bus:%u device:%u)", - vendor, product, - usbsrc->bus, usbsrc->device, - bus, device); - } - } else if (!vendor && bus) { - if (virUSBDeviceFindByBus(bus, device, NULL, mandatory, usb) < 0) - return -1; } - out: - if (!*usb) - hostdev->missing = true; return 0; } diff --git a/src/util/virusb.c b/src/util/virusb.c index 6a001a7..2e15849 100644 --- a/src/util/virusb.c +++ b/src/util/virusb.c @@ -68,12 +68,6 @@ struct _virUSBDeviceList { virUSBDevicePtr *devs; }; -typedef enum { - USB_DEVICE_ALL = 0, - USB_DEVICE_FIND_BY_VENDOR = 1 << 0, - USB_DEVICE_FIND_BY_BUS = 1 << 1, -} virUSBDeviceFindFlags; - static virClassPtr virUSBDeviceListClass; static void virUSBDeviceListDispose(void *obj); @@ -119,11 +113,33 @@ static int virUSBSysReadFile(const char *f_name, const char *d_name, return ret; } +static int virUSBSysReadFileStr(const char *f_name, const char *d_name, + char **value) +{ + int ret = -1, tmp; + char *buf = NULL; + char *filename = NULL; + + tmp = virAsprintf(&filename, USB_SYSFS "/devices/%s/%s", d_name, f_name); + if (tmp < 0) + goto cleanup; + + if (virFileReadAll(filename, 1024, &buf) < 0) + goto cleanup; + + *value = buf; + ret = 0; + cleanup: + VIR_FREE(filename); + return ret; +} + static virUSBDeviceListPtr virUSBDeviceSearch(unsigned int vendor, unsigned int product, unsigned int bus, unsigned int devno, + const char *port, const char *vroot, unsigned int flags) { @@ -143,6 +159,8 @@ virUSBDeviceSearch(unsigned int vendor, while ((direrr = virDirRead(dir, &de, USB_SYSFS "/devices")) > 0) { unsigned int found_prod, found_vend, found_bus, found_devno; + char *found_port; + bool port_matches; char *tmpstr = de->d_name; if (strchr(de->d_name, ':')) @@ -170,16 +188,31 @@ virUSBDeviceSearch(unsigned int vendor, 10, &found_devno) < 0) goto cleanup; + if (virUSBSysReadFileStr("devpath", de->d_name, + &found_port) < 0) { + goto cleanup; + } else { + found_port[strlen(found_port) - 1] = '\0'; /* remove newline */ + port_matches = STREQ_NULLABLE(found_port, port); + VIR_FREE(found_port); + } + if ((flags & USB_DEVICE_FIND_BY_VENDOR) && (found_prod != product || found_vend != vendor)) continue; - if (flags & USB_DEVICE_FIND_BY_BUS) { + if (flags & USB_DEVICE_FIND_BY_DEVICE) { if (found_bus != bus || found_devno != devno) continue; found = true; } + if (flags & USB_DEVICE_FIND_BY_PORT) { + if (found_bus != bus || !port_matches) + continue; + found = true; + } + usb = virUSBDeviceNew(found_bus, found_devno, vroot); if (!usb) goto cleanup; @@ -204,36 +237,43 @@ virUSBDeviceSearch(unsigned int vendor, } int -virUSBDeviceFindByVendor(unsigned int vendor, - unsigned int product, - const char *vroot, - bool mandatory, - virUSBDeviceListPtr *devices) +virUSBDeviceFind(unsigned int vendor, + unsigned int product, + unsigned int bus, + unsigned int devno, + const char *port, + const char *vroot, + bool mandatory, + unsigned int flags, + virUSBDeviceListPtr *devices) { virUSBDeviceListPtr list; int count; - if (!(list = virUSBDeviceSearch(vendor, product, 0, 0, - vroot, - USB_DEVICE_FIND_BY_VENDOR))) + if (!(list = virUSBDeviceSearch(vendor, product, bus, devno, port, + vroot, flags))) return -1; - if (list->count == 0) { + count = list->count; + if (count == 0) { virObjectUnref(list); if (!mandatory) { - VIR_DEBUG("Did not find USB device %x:%x", - vendor, product); if (devices) *devices = NULL; return 0; } virReportError(VIR_ERR_INTERNAL_ERROR, - _("Did not find USB device %x:%x"), vendor, product); + _("Did not find a matching USB device (matching " + "on%s%s%s all others ignored): vid:%04x, pid:%04x, " + "bus:%u, device:%u, port:%s"), + flags & USB_DEVICE_FIND_BY_VENDOR ? " vid/pid," : "", + flags & USB_DEVICE_FIND_BY_DEVICE ? " bus/device," : "", + flags & USB_DEVICE_FIND_BY_PORT ? " bus/port," : "", + vendor, product, bus, devno, port ? port : "(null)"); return -1; } - count = list->count; if (devices) *devices = list; else @@ -242,86 +282,6 @@ virUSBDeviceFindByVendor(unsigned int vendor, return count; } -int -virUSBDeviceFindByBus(unsigned int bus, - unsigned int devno, - const char *vroot, - bool mandatory, - virUSBDevicePtr *usb) -{ - virUSBDeviceListPtr list; - - if (!(list = virUSBDeviceSearch(0, 0, bus, devno, - vroot, - USB_DEVICE_FIND_BY_BUS))) - return -1; - - if (list->count == 0) { - virObjectUnref(list); - if (!mandatory) { - VIR_DEBUG("Did not find USB device bus:%u device:%u", - bus, devno); - if (usb) - *usb = NULL; - return 0; - } - - virReportError(VIR_ERR_INTERNAL_ERROR, - _("Did not find USB device bus:%u device:%u"), - bus, devno); - return -1; - } - - if (usb) { - *usb = virUSBDeviceListGet(list, 0); - virUSBDeviceListSteal(list, *usb); - } - virObjectUnref(list); - - return 0; -} - -int -virUSBDeviceFind(unsigned int vendor, - unsigned int product, - unsigned int bus, - unsigned int devno, - const char *vroot, - bool mandatory, - virUSBDevicePtr *usb) -{ - virUSBDeviceListPtr list; - - unsigned int flags = USB_DEVICE_FIND_BY_VENDOR|USB_DEVICE_FIND_BY_BUS; - if (!(list = virUSBDeviceSearch(vendor, product, bus, devno, - vroot, flags))) - return -1; - - if (list->count == 0) { - virObjectUnref(list); - if (!mandatory) { - VIR_DEBUG("Did not find USB device %x:%x bus:%u device:%u", - vendor, product, bus, devno); - if (usb) - *usb = NULL; - return 0; - } - - virReportError(VIR_ERR_INTERNAL_ERROR, - _("Did not find USB device %x:%x bus:%u device:%u"), - vendor, product, bus, devno); - return -1; - } - - if (usb) { - *usb = virUSBDeviceListGet(list, 0); - virUSBDeviceListSteal(list, *usb); - } - virObjectUnref(list); - - return 0; -} - virUSBDevicePtr virUSBDeviceNew(unsigned int bus, unsigned int devno, diff --git a/src/util/virusb.h b/src/util/virusb.h index f98ea21..0834cd4 100644 --- a/src/util/virusb.h +++ b/src/util/virusb.h @@ -35,29 +35,26 @@ typedef virUSBDevice *virUSBDevicePtr; typedef struct _virUSBDeviceList virUSBDeviceList; typedef virUSBDeviceList *virUSBDeviceListPtr; +typedef enum { + USB_DEVICE_ALL = 0, + USB_DEVICE_FIND_BY_VENDOR = 1 << 0, + USB_DEVICE_FIND_BY_DEVICE = 1 << 1, + USB_DEVICE_FIND_BY_PORT = 1 << 2, +} virUSBDeviceFindFlags; + virUSBDevicePtr virUSBDeviceNew(unsigned int bus, unsigned int devno, const char *vroot); -int virUSBDeviceFindByBus(unsigned int bus, - unsigned int devno, - const char *vroot, - bool mandatory, - virUSBDevicePtr *usb); - -int virUSBDeviceFindByVendor(unsigned int vendor, - unsigned int product, - const char *vroot, - bool mandatory, - virUSBDeviceListPtr *devices); - int virUSBDeviceFind(unsigned int vendor, unsigned int product, unsigned int bus, unsigned int devno, + const char *port, const char *vroot, bool mandatory, - virUSBDevicePtr *usb); + unsigned int flags, + virUSBDeviceListPtr *devices); void virUSBDeviceFree(virUSBDevicePtr dev); int virUSBDeviceSetUsedBy(virUSBDevicePtr dev, diff --git a/tests/virusbtest.c b/tests/virusbtest.c index 4bbfe4a..0782b73 100644 --- a/tests/virusbtest.c +++ b/tests/virusbtest.c @@ -33,7 +33,8 @@ typedef enum { FIND_BY_ALL, FIND_BY_VENDOR, - FIND_BY_BUS + FIND_BY_DEVICE, + FIND_BY_PORT } testUSBFindFlags; struct findTestInfo { @@ -42,6 +43,7 @@ struct findTestInfo { unsigned int product; unsigned int bus; unsigned int devno; + const char *port; const char *vroot; bool mandatory; int how; @@ -78,25 +80,31 @@ static int testDeviceFind(const void *opaque) virUSBDeviceListPtr devs = NULL; int rv = 0; size_t i, ndevs = 0; + unsigned int flags = 0; switch (info->how) { case FIND_BY_ALL: - rv = virUSBDeviceFind(info->vendor, info->product, - info->bus, info->devno, - info->vroot, info->mandatory, &dev); + flags = USB_DEVICE_FIND_BY_VENDOR | + USB_DEVICE_FIND_BY_DEVICE | + USB_DEVICE_FIND_BY_PORT; break; case FIND_BY_VENDOR: - rv = virUSBDeviceFindByVendor(info->vendor, info->product, - info->vroot, info->mandatory, &devs); + flags = USB_DEVICE_FIND_BY_VENDOR; break; - case FIND_BY_BUS: - rv = virUSBDeviceFindByBus(info->bus, info->devno, - info->vroot, info->mandatory, &dev); + case FIND_BY_DEVICE: + flags = USB_DEVICE_FIND_BY_DEVICE; + break; + case FIND_BY_PORT: + flags = USB_DEVICE_FIND_BY_PORT; break; } + rv = virUSBDeviceFind(info->vendor, info->product, + info->bus, info->devno, info->port, + info->vroot, info->mandatory, flags, &devs); + if (info->expectFailure) { - if (rv == 0) { + if (rv >= 0) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", "unexpected success"); } else { @@ -107,9 +115,18 @@ static int testDeviceFind(const void *opaque) goto cleanup; } + if (info->how != FIND_BY_VENDOR) { + if (rv == 1) { + dev = virUSBDeviceListGet(devs, 0); + virUSBDeviceListSteal(devs, dev); + } else { + goto cleanup; + } + } + switch (info->how) { case FIND_BY_ALL: - case FIND_BY_BUS: + case FIND_BY_DEVICE: if (virUSBDeviceFileIterate(dev, testDeviceFileActor, NULL) < 0) goto cleanup; break; @@ -155,14 +172,17 @@ testUSBList(const void *opaque ATTRIBUTE_UNUSED) virUSBDeviceListPtr list = NULL; virUSBDeviceListPtr devlist = NULL; virUSBDevicePtr dev = NULL; + virUSBDeviceListPtr devs = NULL; int ret = -1; + int rv; size_t i, ndevs; if (!(list = virUSBDeviceListNew())) goto cleanup; #define EXPECTED_NDEVS_ONE 3 - if (virUSBDeviceFindByVendor(0x1d6b, 0x0002, NULL, true, &devlist) < 0) + if (virUSBDeviceFind(0x1d6b, 0x0002, 0, 0, NULL, NULL, true, + USB_DEVICE_FIND_BY_VENDOR, &devlist) < 0) goto cleanup; ndevs = virUSBDeviceListCount(devlist); @@ -186,7 +206,8 @@ testUSBList(const void *opaque ATTRIBUTE_UNUSED) goto cleanup; #define EXPECTED_NDEVS_TWO 3 - if (virUSBDeviceFindByVendor(0x18d1, 0x4e22, NULL, true, &devlist) < 0) + if (virUSBDeviceFind(0x18d1, 0x4e22, 0, 0, NULL, NULL, true, + USB_DEVICE_FIND_BY_VENDOR, &devlist) < 0) goto cleanup; ndevs = virUSBDeviceListCount(devlist); @@ -206,8 +227,17 @@ testUSBList(const void *opaque ATTRIBUTE_UNUSED) EXPECTED_NDEVS_ONE + EXPECTED_NDEVS_TWO) < 0) goto cleanup; - if (virUSBDeviceFind(0x18d1, 0x4e22, 1, 20, NULL, true, &dev) < 0) + rv = virUSBDeviceFind(0x18d1, 0x4e22, 1, 20, "1.5.6", NULL, true, + USB_DEVICE_FIND_BY_VENDOR | + USB_DEVICE_FIND_BY_DEVICE | + USB_DEVICE_FIND_BY_PORT, &devs); + if (rv != 1) { goto cleanup; + } else { + dev = virUSBDeviceListGet(devs, 0); + virUSBDeviceListSteal(devs, dev); + } + virObjectUnref(devs); if (!virUSBDeviceListFind(list, dev)) { virReportError(VIR_ERR_INTERNAL_ERROR, @@ -239,49 +269,63 @@ mymain(void) { int rv = 0; -#define DO_TEST_FIND_FULL(name, vend, prod, bus, devno, vroot, mand, how, fail) \ - do { \ - struct findTestInfo data = { name, vend, prod, bus, \ - devno, vroot, mand, how, fail \ - }; \ - if (virTestRun("USBDeviceFind " name, testDeviceFind, &data) < 0) \ - rv = -1; \ +#define DO_TEST_FIND_FULL(name, vend, prod, bus, devno, \ + port, vroot, mand, how, fail) \ + do { \ + struct findTestInfo data = { name, vend, prod, bus, \ + devno, port, vroot, mand, how, fail \ + }; \ + if (virTestRun("USBDeviceFind " name, testDeviceFind, &data) < 0) \ + rv = -1; \ } while (0) -#define DO_TEST_FIND(name, vend, prod, bus, devno) \ - DO_TEST_FIND_FULL(name, vend, prod, bus, devno, NULL, true, \ +#define DO_TEST_FIND(name, vend, prod, bus, devno, port) \ + DO_TEST_FIND_FULL(name, vend, prod, bus, devno, port, NULL, true, \ FIND_BY_ALL, false) -#define DO_TEST_FIND_FAIL(name, vend, prod, bus, devno) \ - DO_TEST_FIND_FULL(name, vend, prod, bus, devno, NULL, true, \ +#define DO_TEST_FIND_FAIL(name, vend, prod, bus, devno, port) \ + DO_TEST_FIND_FULL(name, vend, prod, bus, devno, port, NULL, true, \ FIND_BY_ALL, true) -#define DO_TEST_FIND_BY_BUS(name, bus, devno) \ - DO_TEST_FIND_FULL(name, 101, 202, bus, devno, NULL, true, \ - FIND_BY_BUS, false) -#define DO_TEST_FIND_BY_BUS_FAIL(name, bus, devno) \ - DO_TEST_FIND_FULL(name, 101, 202, bus, devno, NULL, true, \ - FIND_BY_BUS, true) +#define DO_TEST_FIND_BY_DEVICE(name, bus, devno) \ + DO_TEST_FIND_FULL(name, 101, 202, bus, devno, NULL, NULL, true, \ + FIND_BY_DEVICE, false) +#define DO_TEST_FIND_BY_DEVICE_FAIL(name, bus, devno) \ + DO_TEST_FIND_FULL(name, 101, 202, bus, devno, NULL, NULL, true, \ + FIND_BY_DEVICE, true) #define DO_TEST_FIND_BY_VENDOR(name, vend, prod) \ - DO_TEST_FIND_FULL(name, vend, prod, 123, 456, NULL, true, \ + DO_TEST_FIND_FULL(name, vend, prod, 123, 456, NULL, NULL, true, \ FIND_BY_VENDOR, false) #define DO_TEST_FIND_BY_VENDOR_FAIL(name, vend, prod) \ - DO_TEST_FIND_FULL(name, vend, prod, 123, 456, NULL, true, \ + DO_TEST_FIND_FULL(name, vend, prod, 123, 456, NULL, NULL, true, \ FIND_BY_VENDOR, true) - DO_TEST_FIND("Nexus", 0x18d1, 0x4e22, 1, 20); - DO_TEST_FIND_FAIL("Nexus wrong devnum", 0x18d1, 0x4e22, 1, 25); - DO_TEST_FIND_FAIL("Bogus", 0xf00d, 0xbeef, 1024, 768); +#define DO_TEST_FIND_BY_PORT(name, bus, port) \ + DO_TEST_FIND_FULL(name, 101, 202, bus, 456, port, NULL, true, \ + FIND_BY_PORT, false) +#define DO_TEST_FIND_BY_PORT_FAIL(name, bus, port) \ + DO_TEST_FIND_FULL(name, 101, 202, bus, 456, port, NULL, true, \ + FIND_BY_PORT, true) + + DO_TEST_FIND("Nexus", 0x18d1, 0x4e22, 1, 20, "1.5.6"); + DO_TEST_FIND_FAIL("Nexus wrong devnum", 0x18d1, 0x4e22, 1, 25, "1.5.6"); + DO_TEST_FIND_FAIL("Nexus wrong port", 0x18d1, 0x4e22, 1, 25, "1.5.4"); + DO_TEST_FIND_FAIL("Bogus", 0xf00d, 0xbeef, 1024, 768, "1.2.3.4"); - DO_TEST_FIND_BY_BUS("integrated camera", 1, 5); - DO_TEST_FIND_BY_BUS_FAIL("wrong bus/devno combination", 2, 20); - DO_TEST_FIND_BY_BUS_FAIL("missing bus", 5, 20); - DO_TEST_FIND_BY_BUS_FAIL("missing devnum", 1, 158); + DO_TEST_FIND_BY_DEVICE("integrated camera", 1, 5); + DO_TEST_FIND_BY_DEVICE_FAIL("wrong bus/devno combination", 2, 20); + DO_TEST_FIND_BY_DEVICE_FAIL("missing bus", 5, 20); + DO_TEST_FIND_BY_DEVICE_FAIL("missing devnum", 1, 158); DO_TEST_FIND_BY_VENDOR("Nexus (multiple results)", 0x18d1, 0x4e22); DO_TEST_FIND_BY_VENDOR_FAIL("Bogus vendor and product", 0xf00d, 0xbeef); DO_TEST_FIND_BY_VENDOR_FAIL("Valid vendor", 0x1d6b, 0xbeef); + DO_TEST_FIND_BY_PORT("Logitech mouse", 1, "1.5.3.3"); + DO_TEST_FIND_BY_PORT_FAIL("wrong bus/port combination", 2, "1.5.3.3"); + DO_TEST_FIND_BY_PORT_FAIL("missing bus", 5, "1.5.3.3"); + DO_TEST_FIND_BY_PORT_FAIL("missing port", 1, "8.2.5"); + if (virTestRun("USB List test", testUSBList, NULL) < 0) rv = -1; diff --git a/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.3.1/devpath b/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.3.1/devpath new file mode 100644 index 0000000..02a7fbe --- /dev/null +++ b/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.3.1/devpath @@ -0,0 +1 @@ +1.5.3.1 diff --git a/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.3.3/devpath b/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.3.3/devpath new file mode 100644 index 0000000..23ca863 --- /dev/null +++ b/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.3.3/devpath @@ -0,0 +1 @@ +1.5.3.3 diff --git a/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.3/devpath b/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.3/devpath new file mode 100644 index 0000000..8af85be --- /dev/null +++ b/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.3/devpath @@ -0,0 +1 @@ +1.5.3 diff --git a/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.4/devpath b/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.4/devpath new file mode 100644 index 0000000..94fe62c --- /dev/null +++ b/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.4/devpath @@ -0,0 +1 @@ +1.5.4 diff --git a/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.5/devpath b/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.5/devpath new file mode 100644 index 0000000..9075be4 --- /dev/null +++ b/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.5/devpath @@ -0,0 +1 @@ +1.5.5 diff --git a/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.6/devpath b/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.6/devpath new file mode 100644 index 0000000..eac1e0a --- /dev/null +++ b/tests/virusbtestdata/sys_bus_usb/devices/1-1.5.6/devpath @@ -0,0 +1 @@ +1.5.6 diff --git a/tests/virusbtestdata/sys_bus_usb/devices/1-1.5/devpath b/tests/virusbtestdata/sys_bus_usb/devices/1-1.5/devpath new file mode 100644 index 0000000..c239c60 --- /dev/null +++ b/tests/virusbtestdata/sys_bus_usb/devices/1-1.5/devpath @@ -0,0 +1 @@ +1.5 diff --git a/tests/virusbtestdata/sys_bus_usb/devices/1-1.6/devpath b/tests/virusbtestdata/sys_bus_usb/devices/1-1.6/devpath new file mode 100644 index 0000000..810ee4e --- /dev/null +++ b/tests/virusbtestdata/sys_bus_usb/devices/1-1.6/devpath @@ -0,0 +1 @@ +1.6 diff --git a/tests/virusbtestdata/sys_bus_usb/devices/1-1/devpath b/tests/virusbtestdata/sys_bus_usb/devices/1-1/devpath new file mode 100644 index 0000000..d00491f --- /dev/null +++ b/tests/virusbtestdata/sys_bus_usb/devices/1-1/devpath @@ -0,0 +1 @@ +1 diff --git a/tests/virusbtestdata/sys_bus_usb/devices/2-1.2/devpath b/tests/virusbtestdata/sys_bus_usb/devices/2-1.2/devpath new file mode 100644 index 0000000..5625e59 --- /dev/null +++ b/tests/virusbtestdata/sys_bus_usb/devices/2-1.2/devpath @@ -0,0 +1 @@ +1.2 diff --git a/tests/virusbtestdata/sys_bus_usb/devices/2-1/devpath b/tests/virusbtestdata/sys_bus_usb/devices/2-1/devpath new file mode 100644 index 0000000..d00491f --- /dev/null +++ b/tests/virusbtestdata/sys_bus_usb/devices/2-1/devpath @@ -0,0 +1 @@ +1 diff --git a/tests/virusbtestdata/sys_bus_usb/devices/usb1/devpath b/tests/virusbtestdata/sys_bus_usb/devices/usb1/devpath new file mode 100644 index 0000000..573541a --- /dev/null +++ b/tests/virusbtestdata/sys_bus_usb/devices/usb1/devpath @@ -0,0 +1 @@ +0 diff --git a/tests/virusbtestdata/sys_bus_usb/devices/usb2/devpath b/tests/virusbtestdata/sys_bus_usb/devices/usb2/devpath new file mode 100644 index 0000000..573541a --- /dev/null +++ b/tests/virusbtestdata/sys_bus_usb/devices/usb2/devpath @@ -0,0 +1 @@ +0 diff --git a/tests/virusbtestdata/sys_bus_usb/devices/usb3/devpath b/tests/virusbtestdata/sys_bus_usb/devices/usb3/devpath new file mode 100644 index 0000000..573541a --- /dev/null +++ b/tests/virusbtestdata/sys_bus_usb/devices/usb3/devpath @@ -0,0 +1 @@ +0 diff --git a/tests/virusbtestdata/sys_bus_usb/devices/usb4/devpath b/tests/virusbtestdata/sys_bus_usb/devices/usb4/devpath new file mode 100644 index 0000000..573541a --- /dev/null +++ b/tests/virusbtestdata/sys_bus_usb/devices/usb4/devpath @@ -0,0 +1 @@ +0 -- 2.7.4

2 1

Re: [libvirt] [PATCH 00/16] Hyper-V: Improve 2008, Introduce 2012
by Fernando Casas Schössow 08 Sep '16

08 Sep '16

Hi there, I was wondering if any of these patches actually made it into libvirt code. I'm really looking forward hyper-v support improvement and features like shutdown VM (not destroy/turn off but clean shutdown) implemented. Especially interested in 2012/2012 R2 since 2008 is getting old and 2016 is right around the corner. I'm already using libvirt 2.1.0 with some patches I found in https://github.com/vikhyath/libvirt-hyperv-r2-2012 but it would be great to have the code reviewed and merged into libvirt releases. Thanks. Fer > The following patches include work originally done by Yves Vinter back > in 2014. The last patch introduces support for Hyper-V 2012, while > still > supporting 2008. I am not sure that the method I used to include the > 2012 > support is the best approach, mainly due to code duplication, but I am > open to suggestions on how to do this better. > > Jason Miesionczek (16): > hyperv: additional server 2008 wmi classes > hyperv: add cim types support to code generator > hyperv: add get capabilities > hyperv: implement connectGetVersion > hyperv: implement vcpu functions > hyperv: implement nodeGetFreeMemory > hyperv: implement ability to send xml soap requests > hyperv: introduce basic network driver > hyperv: add domain shutdown function > hyperv: add get scheduler functions > hyperv: add set memory functions > hyperv: set vpcu functions > hyperv: domain undefine functions > hyperv: domain define and associated functions > hyperv: network list functions > hyperv: introduce 2012 support > > src/Makefile.am | 2 + > src/hyperv/hyperv_driver.c | 1989 ++++++++++++++++++++++++++++++++- > src/hyperv/hyperv_driver_2012.c | 299 +++++ > src/hyperv/hyperv_driver_2012.h | 55 + > src/hyperv/hyperv_network_driver.c | 280 +++++ > src/hyperv/hyperv_network_driver.h | 30 + > src/hyperv/hyperv_private.h | 8 + > src/hyperv/hyperv_wmi.c | 709 +++++++++++- > src/hyperv/hyperv_wmi.h | 78 ++ > src/hyperv/hyperv_wmi_generator.input | 518 ++++++++- > src/hyperv/hyperv_wmi_generator.py | 68 +- > src/hyperv/openwsman.h | 4 + > 12 files changed, 3989 insertions(+), 51 deletions(-) > create mode 100644 src/hyperv/hyperv_driver_2012.c > create mode 100644 src/hyperv/hyperv_driver_2012.h > create mode 100644 src/hyperv/hyperv_network_driver.c > create mode 100644 src/hyperv/hyperv_network_driver.h

3 7

[libvirt] [PATCH] security: Fixing wrong label in virt-aa-helper.c.
by Julio Faracco 08 Sep '16

08 Sep '16

There is an issue with a wrong label inside vah_add_path(). The compilation fails with the error: make[3]: Entering directory '/tmp/libvirt/src' CC security/virt_aa_helper-virt-aa-helper.o security/virt-aa-helper.c: In function 'vah_add_path': security/virt-aa-helper.c:769:9: error: label 'clean' used but not defined goto clean; This patch moves 'clean' label to 'cleanup' label. Signed-off-by: Julio Faracco <jcfaracco(a)gmail.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 2aa6e28..fc01076 100644 --- a/src/security/virt-aa-helper.c +++ b/src/security/virt-aa-helper.c @@ -766,7 +766,7 @@ vah_add_path(virBufferPtr buf, const char *path, const char *perms, bool recursi } if (VIR_STRDUP_QUIET(perms_new, perms) < 0) - goto clean; + goto cleanup; if (strchr(perms_new, 'w') != NULL) { readonly = false; -- 2.7.4

2 2

[libvirt] [PATCH 0/2] qemu: Remove stale transient def when migration fails
by Jiri Denemark 08 Sep '16

08 Sep '16

If a migration of a domain which is already defined on the destination host failed early (before we tried to start QEMU), we would forget to remove the incoming transient definition. Later on when someone starts the domain on the destination host, we will use the stale incoming definition and the persistent definition will just be ignored. https://bugzilla.redhat.com/show_bug.cgi?id=1368774 Jiri Denemark (2): Add helper for removing transient definition qemu: Remove stale transient def when migration fails src/conf/domain_conf.c | 19 +++++++++++++++++++ src/conf/domain_conf.h | 1 + src/libvirt_private.syms | 1 + src/libxl/libxl_domain.c | 8 +------- src/lxc/lxc_process.c | 7 +------ src/qemu/qemu_migration.c | 4 ++++ src/qemu/qemu_process.c | 7 +------ src/test/test_driver.c | 9 ++------- src/uml/uml_driver.c | 14 ++------------ 9 files changed, 32 insertions(+), 38 deletions(-) -- 2.10.0

2 4

[libvirt] Opening a read-only connection - with SASL auth.
by Yaniv Kaul 08 Sep '16

08 Sep '16

Looking at the collectd virt plugin, it opens a read only connection with libvirt[1]. We'd like to use it with SASL auth. (which oVirt configures libvirt to use by default, with a simple username/password) - is that possible? TIA, Y. [1] https://github.com/collectd/collectd/blob/d782a871aca61ac966ee8d928930719bd…

1 0

[libvirt] [PATCH] Allow address lease query with inactive domain
by Amador Pahim 08 Sep '16

08 Sep '16

The <drive>DomainInterfaceAddresses, when using the source type LEASE, does not need the domain to be running. It only checks the dhcp lease file and gets the address information from a valid lease, if any. This patch removes the virDomainObjIsActive(vm) check from the LEASE query type on both qemu and lxc drivers, keeping it only for query type AGENT on qemu driver (since lxc does not support AGENT type). Signed-off-by: Amador Pahim <amador(a)pahim.org> --- src/libxl/libxl_driver.c | 6 ------ src/qemu/qemu_driver.c | 12 ++++++------ 2 files changed, 6 insertions(+), 12 deletions(-) diff --git a/src/libxl/libxl_driver.c b/src/libxl/libxl_driver.c index 480fef9..9f8c36a 100644 --- a/src/libxl/libxl_driver.c +++ b/src/libxl/libxl_driver.c @@ -6251,12 +6251,6 @@ libxlDomainInterfaceAddresses(virDomainPtr dom, if (virDomainInterfaceAddressesEnsureACL(dom->conn, vm->def) < 0) goto cleanup; - if (!virDomainObjIsActive(vm)) { - virReportError(VIR_ERR_OPERATION_INVALID, "%s", - _("domain is not running")); - goto cleanup; - } - switch (source) { case VIR_DOMAIN_INTERFACE_ADDRESSES_SRC_LEASE: ret = libxlGetDHCPInterfaces(dom, vm, ifaces); diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 807e06d..a8dfa2f 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -19600,18 +19600,18 @@ qemuDomainInterfaceAddresses(virDomainPtr dom, if (virDomainInterfaceAddressesEnsureACL(dom->conn, vm->def) < 0) goto cleanup; - if (!virDomainObjIsActive(vm)) { - virReportError(VIR_ERR_OPERATION_INVALID, "%s", - _("domain is not running")); - goto cleanup; - } - switch (source) { case VIR_DOMAIN_INTERFACE_ADDRESSES_SRC_LEASE: ret = qemuGetDHCPInterfaces(dom, vm, ifaces); break; case VIR_DOMAIN_INTERFACE_ADDRESSES_SRC_AGENT: + if (!virDomainObjIsActive(vm)) { + virReportError(VIR_ERR_OPERATION_INVALID, "%s", + _("domain is not running")); + goto cleanup; + } + if (qemuDomainObjBeginJob(driver, vm, QEMU_JOB_QUERY) < 0) goto cleanup; -- 2.7.4

3 4

[libvirt] [PATCH] virt-admin: Document srv-list command in virt-admin's man page
by Erik Skultety 08 Sep '16

08 Sep '16

srv-list command was missing in the man page. Signed-off-by: Erik Skultety <eskultet(a)redhat.com> --- tools/virt-admin.pod | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/tools/virt-admin.pod b/tools/virt-admin.pod index 2199038..c74707a 100644 --- a/tools/virt-admin.pod +++ b/tools/virt-admin.pod @@ -143,6 +143,20 @@ libvirtd:///system is used. =back +=head1 DAEMON COMMANDS + +Following commands allow to monitor daemon's state as well as directly change +its internal configuration. + +=over 4 + +=item B<srv-list> + +Lists all manageable servers on daemon the client is currently connected to +through administration interface. + +=back + =head1 SERVER COMMANDS Following commands manipulate daemon's server internal configuration. -- 2.5.5

2 2

[libvirt] [PATCH] virt-admin: Fix some English grammar in the manpage
by Erik Skultety 08 Sep '16

08 Sep '16

Signed-off-by: Erik Skultety <eskultet(a)redhat.com> --- tools/virt-admin.pod | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) Pushed as trivial. diff --git a/tools/virt-admin.pod b/tools/virt-admin.pod index 0195001..020eb51 100644 --- a/tools/virt-admin.pod +++ b/tools/virt-admin.pod @@ -159,7 +159,7 @@ currently connected to. =head1 SERVER COMMANDS -Following commands manipulate daemon's server internal configuration. +The following commands manipulate daemon's server internal configuration. The I<server> is specified by its name. =over 4 @@ -281,7 +281,7 @@ I<--max-clients>. =head1 CLIENT COMMANDS -Following commands provide management and monitoring of clients connected to +The following commands provide management and monitoring of clients connected to one of daemon's available servers. Clients are specified by their numeric ID which is obtained by listing all clients connected to a specified server (see command B<srv-clients-list>). -- 2.5.5

1 0

[libvirt] [libvirt-php][PATCH 0/8] Couple of php fixes
by Michal Privoznik 08 Sep '16

08 Sep '16

Pushed under php-autopush rule. Michal Privoznik (8): libvirt_domain_get_network_info: Don't overwrite retval libvirt_storagepool_get_autostart: Rework translate_counter_type: Rework libvirt_domain_block_commit: Reindent libvirt_domain_qemu_agent_command: Reindent libvirt_domain_disk_add: Reindent libvirt_domain_disk_remove: Reindent libvirt_list_storagepools: Reindent src/libvirt-php.c | 129 ++++++++++++++++++++++++++---------------------------- 1 file changed, 61 insertions(+), 68 deletions(-) -- 2.8.4

1 8

[libvirt] [libvirt-php][PATCH] Don't leak @mime in libvirt_domain_get_screenshot_api
by Michal Privoznik 08 Sep '16

08 Sep '16

When getting a screenshot of a domain via virDomainScreenshot() the MIME type of a image is returned as a string which needs to be freed by caller. However, we are not doing that anywhere. ==24748== 48 bytes in 2 blocks are definitely lost in loss record 123 of 171 ==24748== at 0x4C29F80: malloc (vg_replace_malloc.c:296) ==24748== by 0x90674AB: xdr_string (xdr.c:790) ==24748== by 0xCD3AF1D: xdr_remote_nonnull_string (in /usr/lib64/libvirt.so.0.2003.0) ==24748== by 0x9067B34: xdr_reference (xdr_ref.c:84) ==24748== by 0x9067C50: xdr_pointer (xdr_ref.c:135) ==24748== by 0xCD3B284: xdr_remote_string (in /usr/lib64/libvirt.so.0.2003.0) ==24748== by 0xCD3D288: xdr_remote_domain_screenshot_ret (in /usr/lib64/libvirt.so.0.2003.0) ==24748== by 0xCD5168E: virNetMessageDecodePayload (in /usr/lib64/libvirt.so.0.2003.0) ==24748== by 0xCD47E91: virNetClientProgramCall (in /usr/lib64/libvirt.so.0.2003.0) ==24748== by 0xCD1EBD3: callFull.isra.3 (in /usr/lib64/libvirt.so.0.2003.0) ==24748== by 0xCD29973: remoteDomainScreenshot (in /usr/lib64/libvirt.so.0.2003.0) ==24748== by 0xCCDC24C: virDomainScreenshot (in /usr/lib64/libvirt.so.0.2003.0) Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- Pushed under php-autopush rule. src/libvirt-php.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/src/libvirt-php.c b/src/libvirt-php.c index 8edcb10..57ef1a0 100644 --- a/src/libvirt-php.c +++ b/src/libvirt-php.c @@ -4767,8 +4767,10 @@ PHP_FUNCTION(libvirt_domain_get_screenshot_api) } #ifndef EXTWIN - if (mkstemp(file) == 0) + if (mkstemp(file) == 0) { + free(mime); RETURN_FALSE; + } #endif if ((fd = open(file, O_WRONLY|O_CREAT|O_EXCL, 0666)) < 0) { @@ -4776,6 +4778,7 @@ PHP_FUNCTION(libvirt_domain_get_screenshot_api) (fd = open(file, O_WRONLY|O_TRUNC, 0666)) < 0) { virStreamFree(st); set_error_if_unset("Cannot get create file to save domain screenshot" TSRMLS_CC); + free(mime); RETURN_FALSE; } } @@ -4783,6 +4786,7 @@ PHP_FUNCTION(libvirt_domain_get_screenshot_api) if (virStreamRecvAll(st, streamSink, &fd) < 0) { virStreamFree(st); set_error_if_unset("Cannot receive screenshot data" TSRMLS_CC); + free(mime); RETURN_FALSE; } @@ -4791,6 +4795,7 @@ PHP_FUNCTION(libvirt_domain_get_screenshot_api) if (virStreamFinish(st) < 0) { virStreamFree(st); set_error_if_unset("Cannot close stream for domain" TSRMLS_CC); + free(mime); RETURN_FALSE; } @@ -4816,6 +4821,8 @@ PHP_FUNCTION(libvirt_domain_get_screenshot_api) VIRT_ADD_ASSOC_STRING(return_value, "file", file); VIRT_ADD_ASSOC_STRING(return_value, "mime", mime); } + + free(mime); } /* -- 2.8.4

1 0

[libvirt] [PATCH] Avoid segfault in virt-aa-helper when handling read-only mount filesystems
by rufo 07 Sep '16

07 Sep '16

This patch fixes a segfault in virt-aa-helper caused by attempting to modify a string literal in situ. It is triggered when a domain has a <filesystem> with type='mount' configured readonly, and libvirt is using the AppArmor security driver for sVirt confinement. --- src/security/virt-aa-helper.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/src/security/virt-aa-helper.c b/src/security/virt-aa-helper.c index 49e12b9..c22aa66 100644 --- a/src/security/virt-aa-helper.c +++ b/src/security/virt-aa-helper.c @@ -740,6 +740,7 @@ vah_add_path(virBufferPtr buf, const char *path, const char *perms, bool recursi bool readonly = true; bool explicit_deny_rule = true; char *sub = NULL; + char *perms_new = strdup(perms); if (path == NULL) return rc; @@ -764,12 +765,12 @@ vah_add_path(virBufferPtr buf, const char *path, const char *perms, bool recursi return rc; } - if (strchr(perms, 'w') != NULL) { + if (strchr(perms_new, 'w') != NULL) { readonly = false; explicit_deny_rule = false; } - if ((sub = strchr(perms, 'R')) != NULL) { + if ((sub = strchr(perms_new, 'R')) != NULL) { /* Don't write the invalid R permission, replace it with 'r' */ sub[0] = 'r'; explicit_deny_rule = false; @@ -787,7 +788,7 @@ vah_add_path(virBufferPtr buf, const char *path, const char *perms, bool recursi if (tmp[strlen(tmp) - 1] == '/') tmp[strlen(tmp) - 1] = '\0'; - virBufferAsprintf(buf, " \"%s%s\" %s,\n", tmp, recursive ? "/**" : "", perms); + virBufferAsprintf(buf, " \"%s%s\" %s,\n", tmp, recursive ? "/**" : "", perms_new); if (explicit_deny_rule) { virBufferAddLit(buf, " # don't audit writes to readonly files\n"); virBufferAsprintf(buf, " deny \"%s%s\" w,\n", tmp, recursive ? "/**" : ""); -- 2.9.3

4 5

[libvirt] [PATCH] Delete extra wrap after vol-reisize error
by Yanqiu Zhang 07 Sep '16

07 Sep '16

This patch is to delete the extra wrap "\n" after failed vol-reisize error, both "Failed to change size of volume to" and "Failed to change size of volume by". For error with wrap, there will be an extra wrap between two errors, such as: (1)# virsh vol-resize --pool default --vol vol-test 5M error: Failed to change size of volume 'vol-test' to 5M error: invalid argument: Can't shrink capacity below current capacity unless shrink flag explicitly specified (2)# virsh vol-resize /var/lib/libvirt/images/volds --shrink --delta 10M error: Failed to change size of volume 'volds' by 10M error: invalid argument: can't shrink capacity below existing allocation --- tools/virsh-volume.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/virsh-volume.c b/tools/virsh-volume.c index d35fee1..4b3b4d7 100644 --- a/tools/virsh-volume.c +++ b/tools/virsh-volume.c @@ -1128,8 +1128,8 @@ cmdVolResize(vshControl *ctl, const vshCmd *cmd) ret = true; } else { vshError(ctl, - delta ? _("Failed to change size of volume '%s' by %s\n") - : _("Failed to change size of volume '%s' to %s\n"), + delta ? _("Failed to change size of volume '%s' by %s") + : _("Failed to change size of volume '%s' to %s"), virStorageVolGetName(vol), capacityStr); ret = false; } -- 1.8.3.1

2 1

[libvirt] [libvirt-php][PATCH] libvirt_list_active_domains: Don't free domain before strduping its name
by Michal Privoznik 07 Sep '16

07 Sep '16

There's a bug in the implementation of libvirt_list_active_domains(). We try to return an array of names of active domains. So we iterate over array of domains as returned by libvirt and get each domain's name. But we don't obtain a duplicate of that name rather just a pointer into virDomain object. The very next thing we do is we free the virDomain object and then try to copy what's now invalid pointer. Reverse the order of those two actions. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/libvirt-php.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) diff --git a/src/libvirt-php.c b/src/libvirt-php.c index 2045c59..8edcb10 100644 --- a/src/libvirt-php.c +++ b/src/libvirt-php.c @@ -9128,18 +9128,15 @@ PHP_FUNCTION(libvirt_list_active_domains) domain=virDomainLookupByID(conn->conn,ids[i]); if (domain!=NULL) { - name=virDomainGetName(domain); - - if (virDomainFree (domain)) - resource_change_counter(INT_RESOURCE_DOMAIN, conn->conn, domain, 0 TSRMLS_CC); - - if (name==NULL) - { - efree (ids); + if (!(name = virDomainGetName(domain))) { + efree(ids); RETURN_FALSE; } VIRT_ADD_NEXT_INDEX_STRING(return_value, name); + + if (virDomainFree(domain)) + resource_change_counter(INT_RESOURCE_DOMAIN, conn->conn, domain, 0 TSRMLS_CC); } } efree(ids); -- 2.8.4

2 1

[libvirt] [libvirt-php] libvirt_list_active_domains returns garbage querying a Hyper-V host
by Fernando Casas Schössow 07 Sep '16

07 Sep '16

Hi all, I'm using libvirt-php to manage a Hyper-V 2012 R2 host. When trying to retrieve the active VMs using the function libvirt_list_active_domains() I get an array with the correct amount of items but with garbage instead of the domain names: Array ( [0] => P’*µ [1] => +µ [2] => `„*µ [3] => °Ø*µ [4] => Ð„*µ [5] => `†*µ [6] => [7] => `†*µ [8] => [9] => p+µ [10] => p"+µ [11] => P#+µ [12] => ß*µ ) On the other hand libvirt_list_inactive_domains() and libvirt_list_domains() work as expected returning the right amount of items and the domain names correctly: libvirt_list_inactive_domains() output: Array ( [0] => DCHOMELAB01 [1] => VMFCSW7 [2] => REMOTEAPP01 [3] => MPSSD01 [4] => CS16SVR ) Sofware versions: OS: CentOS Linux release 7.2.1511 (Core) Kernel: 3.10.0-327.28.2.el7.x86_64 PHP: 5.4.16-36.3 libvirt: 2.1.0 libvirt-php: 0.5.2 Node: Hyper-V 2012 R2 Thanks in advance for your replies. Fer

3 5

[libvirt] [PATCH] qemu_driver: protect existed save image
by Chen Hanxiao 07 Sep '16

07 Sep '16

From: Chen Hanxiao <chenhanxiao(a)gmail.com> We may overwrite existed save image by 'virsh save' command. This patch will check qemu save headers before trying to rewrite it. Signed-off-by: Chen Hanxiao <chenhanxiao(a)gmail.com> --- src/qemu/qemu_driver.c | 141 +++++++++++++++++++++++++++++++------------------ 1 file changed, 89 insertions(+), 52 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index a0ac2ef..56aa079 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -2806,6 +2806,73 @@ bswap_header(virQEMUSaveHeaderPtr hdr) } +static int +qemuSaveHeaderCheck(int fd, + virQEMUSaveHeaderPtr header, + const char *path, + bool unlink_corrupt) +{ + int ret = -1; + + if (saferead(fd, header, sizeof(*header)) != sizeof(*header)) { + if (unlink_corrupt) { + if (VIR_CLOSE(fd) < 0 || unlink(path) < 0) { + virReportSystemError(errno, + _("cannot remove corrupt file: %s"), + path); + goto error; + } + return -3; + } + virReportError(VIR_ERR_OPERATION_FAILED, + "%s", _("failed to read qemu header")); + goto error; + } + + if (memcmp((*header).magic, QEMU_SAVE_MAGIC, sizeof((*header).magic)) != 0) { + const char *msg = _("image magic is incorrect"); + + if (memcmp((*header).magic, QEMU_SAVE_PARTIAL, + sizeof((*header).magic)) == 0) { + msg = _("save image is incomplete"); + if (unlink_corrupt) { + if (VIR_CLOSE(fd) < 0 || unlink(path) < 0) { + virReportSystemError(errno, + _("cannot remove corrupt file: %s"), + path); + goto error; + } + return -3; + } + } + virReportError(VIR_ERR_OPERATION_FAILED, "%s", msg); + goto error; + } + + if ((*header).version > QEMU_SAVE_VERSION) { + /* convert endianess and try again */ + bswap_header(header); + } + + if ((*header).version > QEMU_SAVE_VERSION) { + virReportError(VIR_ERR_OPERATION_FAILED, + _("image version is not supported (%d > %d)"), + (*header).version, QEMU_SAVE_VERSION); + goto error; + } + + if ((*header).xml_len <= 0) { + virReportError(VIR_ERR_OPERATION_FAILED, + _("invalid XML length: %d"), (*header).xml_len); + goto error; + } + + ret = 0; + + error: + return ret; +} + /* return -errno on failure, or 0 on success */ static int qemuDomainSaveHeader(int fd, const char *path, const char *xml, @@ -2827,6 +2894,7 @@ qemuDomainSaveHeader(int fd, const char *path, const char *xml, _("failed to write xml to '%s'"), path); goto endjob; } + endjob: return ret; } @@ -3067,6 +3135,24 @@ qemuDomainSaveMemory(virQEMUDriverPtr driver, goto cleanup; } } + + /* don't overwrite exsited save files */ + if (virFileExists(path)) { + virQEMUSaveHeader header_tmp; + fd = qemuOpenFile(driver, NULL, path, O_RDONLY, NULL, NULL); + if (fd < 0) + goto cleanup; + + if (qemuSaveHeaderCheck(fd, &header_tmp, path, false) >= 0) { + ret = -EEXIST; + virReportError(VIR_ERR_OPERATION_FAILED, + _("failed to write existed image file '%s'"), + path); + goto cleanup; + } + VIR_FORCE_CLOSE(fd); + } + fd = qemuOpenFile(driver, vm, path, O_WRONLY | O_TRUNC | O_CREAT | directFlag, &needUnlink, &bypassSecurityDriver); @@ -6327,6 +6413,7 @@ qemuDomainSaveImageOpen(virQEMUDriverPtr driver, bool unlink_corrupt) { int fd = -1; + int ret = -1; virQEMUSaveHeader header; char *xml = NULL; virDomainDefPtr def = NULL; @@ -6353,58 +6440,8 @@ qemuDomainSaveImageOpen(virQEMUDriverPtr driver, VIR_FILE_WRAPPER_BYPASS_CACHE))) goto error; - if (saferead(fd, &header, sizeof(header)) != sizeof(header)) { - if (unlink_corrupt) { - if (VIR_CLOSE(fd) < 0 || unlink(path) < 0) { - virReportSystemError(errno, - _("cannot remove corrupt file: %s"), - path); - goto error; - } - return -3; - } - virReportError(VIR_ERR_OPERATION_FAILED, - "%s", _("failed to read qemu header")); + if ((ret = qemuSaveHeaderCheck(fd, &header, path, unlink_corrupt)) < 0) goto error; - } - - if (memcmp(header.magic, QEMU_SAVE_MAGIC, sizeof(header.magic)) != 0) { - const char *msg = _("image magic is incorrect"); - - if (memcmp(header.magic, QEMU_SAVE_PARTIAL, - sizeof(header.magic)) == 0) { - msg = _("save image is incomplete"); - if (unlink_corrupt) { - if (VIR_CLOSE(fd) < 0 || unlink(path) < 0) { - virReportSystemError(errno, - _("cannot remove corrupt file: %s"), - path); - goto error; - } - return -3; - } - } - virReportError(VIR_ERR_OPERATION_FAILED, "%s", msg); - goto error; - } - - if (header.version > QEMU_SAVE_VERSION) { - /* convert endianess and try again */ - bswap_header(&header); - } - - if (header.version > QEMU_SAVE_VERSION) { - virReportError(VIR_ERR_OPERATION_FAILED, - _("image version is not supported (%d > %d)"), - header.version, QEMU_SAVE_VERSION); - goto error; - } - - if (header.xml_len <= 0) { - virReportError(VIR_ERR_OPERATION_FAILED, - _("invalid XML length: %d"), header.xml_len); - goto error; - } if (VIR_ALLOC_N(xml, header.xml_len) < 0) goto error; @@ -6439,7 +6476,7 @@ qemuDomainSaveImageOpen(virQEMUDriverPtr driver, VIR_FORCE_CLOSE(fd); virObjectUnref(caps); - return -1; + return ret; } static int ATTRIBUTE_NONNULL(4) ATTRIBUTE_NONNULL(5) ATTRIBUTE_NONNULL(6) -- 1.8.3.1

2 2

[libvirt] NPIV storage pools do not map to same LUN units across hosts.
by Nitesh Konkar 07 Sep '16

07 Sep '16

Link: http://wiki.libvirt.org/page/NPIV_in_libvirt Topic: Virtual machine configuration change to use vHBA LUN There is a NPIV storage pool defined on two hosts and pool contains a total of 8 volumes, allocated from a storage device. Source: # virsh vol-list poolvhba0 Name Path ------------------------------------------------------------------------------ unit:0:0:0 /dev/disk/by-id/wwn-0x6005076802818bda3000000000000366 unit:0:0:1 /dev/disk/by-id/wwn-0x6005076802818bda3000000000000367 unit:0:0:2 /dev/disk/by-id/wwn-0x6005076802818bda3000000000000368 unit:0:0:3 /dev/disk/by-id/wwn-0x6005076802818bda3000000000000369 unit:0:0:4 /dev/disk/by-id/wwn-0x6005076802818bda300000000000036a unit:0:0:5 /dev/disk/by-id/wwn-0x6005076802818bda3000000000000380 unit:0:0:6 /dev/disk/by-id/wwn-0x6005076802818bda3000000000000381 unit:0:0:7 /dev/disk/by-id/wwn-0x6005076802818bda3000000000000382 -------------------------------------------------------------------- Destination: -------------------------------------------------------------------- # virsh vol-list poolvhba0 Name Path ------------------------------------------------------------------------------ unit:0:0:0 /dev/disk/by-id/wwn-0x6005076802818bda3000000000000380 unit:0:0:1 /dev/disk/by-id/wwn-0x6005076802818bda3000000000000381 unit:0:0:2 /dev/disk/by-id/wwn-0x6005076802818bda3000000000000382 unit:0:0:3 /dev/disk/by-id/wwn-0x6005076802818bda3000000000000367 unit:0:0:4 /dev/disk/by-id/wwn-0x6005076802818bda3000000000000368 unit:0:0:5 /dev/disk/by-id/wwn-0x6005076802818bda3000000000000366 unit:0:0:6 /dev/disk/by-id/wwn-0x6005076802818bda300000000000036a unit:0:0:7 /dev/disk/by-id/wwn-0x6005076802818bda3000000000000369 -------------------------------------------------------------------- As you can see in the above output,the same set of eight LUNs from the storage server have been mapped, but the order that the LUNs are probed on each host is different, resulting in different unit names on the two different hosts . If the the guest XMLs is referencing its storage by "unit" number then is it safe to migrate such guests because the "unit number" is assigned by the driver according to the specific way it probes the storage and hence when you migrate these guests , it results in different unit names on the destination hosts. Thus the migrated guest gets mapped to the wrong LUNs and is given the wrong disks. The problem is that the LUN numbers on the destination host and source host do not agree. Example, LUN 0 on source_host, for example, may be LUN 5 on destination_host. When the guest is given the wrong disk, it suffers a fatal I/O error. (This is manifested as fatal I/O errors since the guest has no idea that its disks just changed out under it.)The migration does not take into account that the unit numbers do match on on the source and destination sides. So, should libvirt make sure that the guest domains reference NPIV pool volumes by their globally-unique wwn instead of by "unit" numbers? The guest XML references its storage by "unit" number. Eg:- <disk type='volume' device='lun'> <driver name='qemu' type='raw' cache='none'/> <source pool='poolvhba0' volume='unit:0:0:0'/> <backingStore/> <target dev='vdb' bus='virtio'/> <alias name='virtio-disk1'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x0'/> </disk> I am planning to write a patch for it. Any comments on the above observation/approach would be appreciated. Thanks, Nitesh.

2 1

[libvirt] [PATCH] virt-admin: Output srv-clients-set data as unsigned int rather than signed
by Erik Skultety 07 Sep '16

07 Sep '16

Unfortunately, commit a8962f70 only fixed first half of the reported issue of virt-admin outputting negative values where unsigned int is expected by BZ below, so this commit represents the other missing half of the fix. resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1356769 Signed-off-by: Erik Skultety <eskultet(a)redhat.com> --- Pushed under trivial rule. Erik tools/virt-admin.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/virt-admin.c b/tools/virt-admin.c index 513054b..12ec057 100644 --- a/tools/virt-admin.c +++ b/tools/virt-admin.c @@ -857,7 +857,7 @@ cmdSrvClientsInfo(vshControl *ctl, const vshCmd *cmd) } for (i = 0; i < nparams; i++) - vshPrint(ctl, "%-20s: %d\n", params[i].field, params[i].value.ui); + vshPrint(ctl, "%-20s: %u\n", params[i].field, params[i].value.ui); ret = true; -- 2.5.5

1 0

[libvirt] [libvirt-php] libvirt_connect not reading out credential info on 0.5.2
by Fernando Casas Schössow 07 Sep '16

07 Sep '16

I'm running libvirt-php 0.5.2 on CentOS 7 with libvirt 2.1.0. Using virsh I'm able to connect to hyper-v hosts correctly but using libvirt-php it fails during authentication since it seems that the credentials are not being passed along. This is the php code I'm using: <?php $logfile = 'test.log'; unlink($logfile); if (!libvirt_logfile_set($logfile)) die('Cannot set the log file'); $connstr = 'hyperv://user@hyperv-host/?transport=http'; $credentials = array(VIR_CRED_AUTHNAME=>'Administrator',VIR_CRED_PASSPHRASE=>'somepass'); $conn = libvirt_connect($connstr, false, $credentials); echo libvirt_get_last_error(); unset($conn); $fp = fopen($logfile, 'r'); $str = fread($fp, filesize($logfile)); fclose($fp); echo '<pre>'; echo $str; print_r($credentials); echo '</pre>'; ?> And this is the output: authentication failed: Password request failed [2016-09-02 11:10:02 libvirt-php/core ]: libvirt_connect: Found 0 elements for credentials [2016-09-02 11:10:02 libvirt-php/core ]: libvirt_virConnectAuthCallback: cred 0, type 5, prompt Enter Administrator's password for hyperv-host challenge hyperv-host [2016-09-02 11:10:02 libvirt-php/core ]: libvirt_virConnectAuthCallback: result (null) (0) [2016-09-02 11:10:02 libvirt-php/core ]: libvirt_connect: Cannot establish connection to hyperv://Administrator@hyperv-host/?transport=http Array ( [2] => Administrator [5] => somepass ) Note the "Found 0 elements for credentials". After doing some googling I found an email thread "[libvirt] [libvirt-php PATCH 0/3] Fix PHP5 compatibilty issues." that I think is related to my problem, especially patch 2/3. I also checked github project and saw that those patches are already merged in the code so I went ahead and apply them to 0.5.2 source code and rebuild: use VIRT_FOREACH macros everywhere - Commit: f4b760d libvirt_connect: use loop macros to read cred info - Commit: d704106 Define macros for looping php hash tables - Commit: 673a0bf The build went fine but now when I'm trying to connect to the hyper-v host using libvirt-php I get an internal server error (500) so it seems PHP is crashing. From the system log I see: kernel: traps: php-cgi[43452] general protection ip:7f63907f74a0 sp:7fff12d61f68 error:0 in libvirt.so.0.2001.0[7f6390735000+385000] Is there any other patch that I'm missing? Any ideas on how to fix the libvirt_connect credentials issue without applying the patches above? Sofware versions: OS: CentOS Linux release 7.2.1511 (Core) Kernel: 3.10.0-327.28.2.el7.x86_64 PHP: 5.4.16-36.3.el7_2.x86_64 libvirt: 2.1.0-4.1.x86_64 libvirt-php: 0.5.2 Thanks in advance for your replies. Fer

2 12

[libvirt] [PATCH] util: fix crash in virClassIsDerivedFrom for CloseCallbacks objects
by Maxim Nestratov 07 Sep '16

07 Sep '16

There is a possibility that qemu driver frees by unreferencing its closeCallbacks pointer as it has the only reference to the object, while in fact not all users of CloseCallbacks called thier virCloseCallbacksUnset. Backtrace is the following: Thread #1: 0 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0 1 in virCondWait (c=<optimized out>, m=<optimized out>) at util/virthread.c:154 2 in virThreadPoolFree (pool=0x7f0810110b50) at util/virthreadpool.c:266 3 in qemuStateCleanup () at qemu/qemu_driver.c:1116 4 in virStateCleanup () at libvirt.c:808 5 in main (argc=<optimized out>, argv=<optimized out>) at libvirtd.c:1660 Thread #2: 0 in virClassIsDerivedFrom (klass=0xdeadbeef, parent=0x7f0837c694d0) at util/virobject.c:169 1 in virObjectIsClass (anyobj=anyobj@entry=0x7f08101d4760, klass=<optimized out>) at util/virobject.c:365 2 in virObjectLock (anyobj=0x7f08101d4760) at util/virobject.c:317 3 in virCloseCallbacksUnset (closeCallbacks=0x7f08101d4760, vm=vm@entry=0x7f08101d47b0, cb=cb@entry=0x7f081d078fc0 <qemuProcessAutoDestroy>) at util/virclosecallbacks.c:163 4 in qemuProcessAutoDestroyRemove (driver=driver@entry=0x7f081018be50, vm=vm@entry=0x7f08101d47b0) at qemu/qemu_process.c:6368 5 in qemuProcessStop (driver=driver@entry=0x7f081018be50, vm=vm@entry=0x7f08101d47b0, reason=reason@entry=VIR_DOMAIN_SHUTOFF_SHUTDOWN, asyncJob=asyncJob@entry=QEMU_ASYNC_JOB_NONE, flags=flags@entry=0) at qemu/qemu_process.c:5854 6 in processMonitorEOFEvent (vm=0x7f08101d47b0, driver=0x7f081018be50) at qemu/qemu_driver.c:4585 7 qemuProcessEventHandler (data=<optimized out>, opaque=0x7f081018be50) at qemu/qemu_driver.c:4629 8 in virThreadPoolWorker (opaque=opaque@entry=0x7f0837c4f820) at util/virthreadpool.c:145 9 in virThreadHelper (data=<optimized out>) at util/virthread.c:206 10 in start_thread () from /lib64/libpthread.so.0 Let's reference CloseCallbacks object in virCloseCallbacksSet and unreference in virCloseCallbacksUnset. Signed-off-by: Maxim Nestratov <mnestratov(a)virtuozzo.com> --- src/util/virclosecallbacks.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/util/virclosecallbacks.c b/src/util/virclosecallbacks.c index 82d4071..891a92b 100644 --- a/src/util/virclosecallbacks.c +++ b/src/util/virclosecallbacks.c @@ -141,6 +141,7 @@ virCloseCallbacksSet(virCloseCallbacksPtr closeCallbacks, virObjectRef(vm); } + virObjectRef(closeCallbacks); ret = 0; cleanup: virObjectUnlock(closeCallbacks); @@ -180,6 +181,8 @@ virCloseCallbacksUnset(virCloseCallbacksPtr closeCallbacks, ret = 0; cleanup: virObjectUnlock(closeCallbacks); + if (!ret) + virObjectUnref(closeCallbacks); return ret; } -- 2.4.11

2 4

[libvirt] [PATCH] qemu: guest agent: introduce new error code VIR_ERR_AGENT_UNSYNCED
by Maxim Nestratov 07 Sep '16

07 Sep '16

From: Yuri Pudgorodskiy <yur(a)virtuozzo.com> A separate error code will help recognize real failures from necessity to try again Signed-off-by: Maxim Nestratov <mnestratov(a)virtuozzo.com> --- include/libvirt/virterror.h | 2 ++ src/qemu/qemu_agent.c | 6 +++--- src/util/virerror.c | 6 ++++++ 3 files changed, 11 insertions(+), 3 deletions(-) diff --git a/include/libvirt/virterror.h b/include/libvirt/virterror.h index 2ec560e..efe83aa 100644 --- a/include/libvirt/virterror.h +++ b/include/libvirt/virterror.h @@ -315,6 +315,8 @@ typedef enum { VIR_ERR_AUTH_UNAVAILABLE = 94, /* authentication unavailable */ VIR_ERR_NO_SERVER = 95, /* Server was not found */ VIR_ERR_NO_CLIENT = 96, /* Client was not found */ + VIR_ERR_AGENT_UNSYNCED = 97, /* guest agent replies with wrong id + to guest-sync command */ } virErrorNumber; /** diff --git a/src/qemu/qemu_agent.c b/src/qemu/qemu_agent.c index eeede6b..fdc4fed 100644 --- a/src/qemu/qemu_agent.c +++ b/src/qemu/qemu_agent.c @@ -966,21 +966,21 @@ qemuAgentGuestSync(qemuAgentPtr mon) goto cleanup; if (!sync_msg.rxObject) { - virReportError(VIR_ERR_INTERNAL_ERROR, "%s", + virReportError(VIR_ERR_AGENT_UNSYNCED, "%s", _("Missing monitor reply object")); goto cleanup; } if (virJSONValueObjectGetNumberUlong(sync_msg.rxObject, "return", &id_ret) < 0) { - virReportError(VIR_ERR_INTERNAL_ERROR, "%s", + virReportError(VIR_ERR_AGENT_UNSYNCED, "%s", _("Malformed return value")); goto cleanup; } VIR_DEBUG("Guest returned ID: %llu", id_ret); if (id_ret != id) { - virReportError(VIR_ERR_INTERNAL_ERROR, + virReportError(VIR_ERR_AGENT_UNSYNCED, _("Guest agent returned ID: %llu instead of %llu"), id_ret, id); goto cleanup; diff --git a/src/util/virerror.c b/src/util/virerror.c index 1177570..2958308 100644 --- a/src/util/virerror.c +++ b/src/util/virerror.c @@ -1394,6 +1394,12 @@ virErrorMsg(virErrorNumber error, const char *info) else errmsg = _("Client not found: %s"); break; + case VIR_ERR_AGENT_UNSYNCED: + if (info == NULL) + errmsg = _("guest agent replied with wrong id to guest-sync command"); + else + errmsg = _("guest agent replied with wrong id to guest-sync command: %s"); + break; } return errmsg; } -- 2.4.11

3 4

[libvirt] [PATCH] virsh: use virConnectGetDomainCapabilities with maxvcpus
by Shivaprasad G Bhat 07 Sep '16

07 Sep '16

virsh maxvcpus --type kvm output is useless on PPC. Also, in commit e6806d79 we documented not rely on virConnectGetMaxVcpus output. Fix the maxvcpus to use virConnectGetDomainCapabilities now to make it useful. The call is made to use the default emulator binary and to check for the host machine and arch which is what the command intends to show anyway. Signed-off-by: Shivaprasad G Bhat <sbhat(a)linux.vnet.ibm.com> --- tools/virsh-host.c | 28 +++++++++++++++++++++++++--- 1 file changed, 25 insertions(+), 3 deletions(-) diff --git a/tools/virsh-host.c b/tools/virsh-host.c index 57f0c0e..505cfbb 100644 --- a/tools/virsh-host.c +++ b/tools/virsh-host.c @@ -607,16 +607,38 @@ cmdMaxvcpus(vshControl *ctl, const vshCmd *cmd) { const char *type = NULL; int vcpus; + char *caps = NULL; + const unsigned int flags = 0; /* No flags so far */ + xmlDocPtr xml = NULL; + xmlXPathContextPtr ctxt = NULL; virshControlPtr priv = ctl->privData; if (vshCommandOptStringReq(ctl, cmd, "type", &type) < 0) return false; - if ((vcpus = virConnectGetMaxVcpus(priv->conn, type)) < 0) - return false; + caps = virConnectGetDomainCapabilities(priv->conn, NULL, NULL, NULL, type, flags); + if (caps) { + xml = virXMLParseStringCtxt(caps, _("(domainCapabilities)"), &ctxt); + if (!xml) { + VIR_FREE(caps); + return false; + } - vshPrint(ctl, "%d\n", vcpus); + virXPathInt("string(./vcpu[1]/@max)", ctxt, &vcpus); + xmlXPathFreeContext(ctxt); + xmlFreeDoc(xml); + VIR_FREE(caps); + } else { + if (last_error && last_error->code != VIR_ERR_NO_SUPPORT) + return false; + + vshResetLibvirtError(); + if ((vcpus = virConnectGetMaxVcpus(priv->conn, type)) < 0) + return false; + } + + vshPrint(ctl, "%d\n", vcpus); return true; }

2 2

[libvirt] [PATCH] tools: Pass opaque data in vshCompleter and introduce autoCompleteOpaque
by Nishith Shah 06 Sep '16

06 Sep '16

This patch changes the signature of vshCompleters, allowing to pass along some data that we might want to along with the completers; for example, we might want to pass the autocomplete vshControl along with the completer, in case the completer requires a connection to libvirtd. Signed-off-by: Nishith Shah <nishithshah.2211(a)gmail.com> --- tools/vsh.c | 10 +++++++++- tools/vsh.h | 2 +- 2 files changed, 10 insertions(+), 2 deletions(-) diff --git a/tools/vsh.c b/tools/vsh.c index 3157922..3cc03ec 100644 --- a/tools/vsh.c +++ b/tools/vsh.c @@ -64,6 +64,11 @@ # define SA_SIGINFO 0 #endif +#ifdef WITH_READLINE +/* For autocompletion */ +void *autoCompleteOpaque; +#endif + /* NOTE: It would be much nicer to have these two as part of vshControl * structure, unfortunately readline doesn't support passing opaque data * and only relies on static data accessible from the user-side callback @@ -2808,7 +2813,8 @@ vshReadlineParse(const char *text, int state) res = vshReadlineOptionsGenerator(sanitized_text, state, cmd); } else if (non_bool_opt_exists && data_complete && opt->completer) { if (!completed_list) - completed_list = opt->completer(opt->completer_flags); + completed_list = opt->completer(autoCompleteOpaque, + opt->completer_flags); if (completed_list) { while ((completed_name = completed_list[completed_list_index])) { completed_list_index++; @@ -2858,6 +2864,8 @@ vshReadlineInit(vshControl *ctl) char *histsize_env = NULL; const char *histsize_str = NULL; + /* Opaque data for autocomplete callbacks and connections to libvirtd */ + autoCompleteOpaque = ctl; /* Allow conditional parsing of the ~/.inputrc file. * Work around ancient readline 4.1 (hello Mac OS X), * which declared it as 'char *' instead of 'const char *'. diff --git a/tools/vsh.h b/tools/vsh.h index 7f43055..e53910f 100644 --- a/tools/vsh.h +++ b/tools/vsh.h @@ -123,7 +123,7 @@ typedef struct _vshCmdOpt vshCmdOpt; typedef struct _vshCmdOptDef vshCmdOptDef; typedef struct _vshControl vshControl; -typedef char **(*vshCompleter)(unsigned int flags); +typedef char **(*vshCompleter)(void *opaque, unsigned int flags); /* * vshCmdInfo -- name/value pair for information about command -- 2.1.4

2 1

[libvirt] [PATCH] util: hostcpu: improve CPU freq code for FreeBSD
by Roman Bogorodskiy 06 Sep '16

06 Sep '16

Current implementation uses the dev.cpu.0.freq sysctl that is provided by the cpufreq(4) framework and returns the actual CPU frequency. However, there are environment where it's not available, e.g. when running nested in KVM. In this case fall back to hw.clockrate that reports CPU frequency at the boot time. Resolves (hopefully): https://bugzilla.redhat.com/show_bug.cgi?id=1369964 --- src/util/virhostcpu.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/src/util/virhostcpu.c b/src/util/virhostcpu.c index 0f03ff8..b5a37a8 100644 --- a/src/util/virhostcpu.c +++ b/src/util/virhostcpu.c @@ -994,9 +994,16 @@ virHostCPUGetInfo(virArch hostarch ATTRIBUTE_UNUSED, *threads = 1; # ifdef __FreeBSD__ + /* dev.cpu.%d.freq reports current active CPU frequency. It is provided by + * the cpufreq(4) framework. However, it might be disabled or no driver + * available. In this case fallback to "hw.clockrate" which reports boot time + * CPU frequency. */ + if (sysctlbyname("dev.cpu.0.freq", &cpu_freq, &cpu_freq_len, NULL, 0) < 0) { - virReportSystemError(errno, "%s", _("cannot obtain CPU freq")); - return -1; + if (sysctlbyname("hw.clockrate", &cpu_freq, &cpu_freq_len, NULL, 0) < 0) { + virReportSystemError(errno, "%s", _("cannot obtain CPU freq")); + return -1; + } } *mhz = cpu_freq; -- 2.7.4

2 2

[libvirt] [PATCH] virsh: Fix *-event error string
by Christophe Fergeau 06 Sep '16

06 Sep '16

When using virsh net-event non-existing-net the error message says that 'either --list or event type is required' This is misleading as 'virsh net-event $valid-event-type' is not going to work either. What is expected is 'virsh net-event --event $valid-event-type' This commit fixes the string in pool-event, nodedev-event, event, and net-event. --- tools/virsh-domain.c | 2 +- tools/virsh-network.c | 2 +- tools/virsh-nodedev.c | 2 +- tools/virsh-pool.c | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index a614512..702a8bd8 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -12694,7 +12694,7 @@ cmdEvent(vshControl *ctl, const vshCmd *cmd) } } else if (!all) { vshError(ctl, "%s", - _("one of --list, --all, or event type is required")); + _("one of --list, --all, or --event type is required")); return false; } diff --git a/tools/virsh-network.c b/tools/virsh-network.c index eec7faf..c6bd132 100644 --- a/tools/virsh-network.c +++ b/tools/virsh-network.c @@ -1238,7 +1238,7 @@ cmdNetworkEvent(vshControl *ctl, const vshCmd *cmd) if (vshCommandOptStringReq(ctl, cmd, "event", &eventName) < 0) return false; if (!eventName) { - vshError(ctl, "%s", _("either --list or event type is required")); + vshError(ctl, "%s", _("either --list or --event type is required")); return false; } if ((event = virshNetworkEventIdTypeFromString(eventName)) < 0) { diff --git a/tools/virsh-nodedev.c b/tools/virsh-nodedev.c index 805c0ff..0e695b9 100644 --- a/tools/virsh-nodedev.c +++ b/tools/virsh-nodedev.c @@ -903,7 +903,7 @@ cmdNodeDeviceEvent(vshControl *ctl, const vshCmd *cmd) if (vshCommandOptStringReq(ctl, cmd, "event", &eventName) < 0) return false; if (!eventName) { - vshError(ctl, "%s", _("either --list or event type is required")); + vshError(ctl, "%s", _("either --list or --event type is required")); return false; } diff --git a/tools/virsh-pool.c b/tools/virsh-pool.c index d25851e..70b2bdd 100644 --- a/tools/virsh-pool.c +++ b/tools/virsh-pool.c @@ -2057,7 +2057,7 @@ cmdPoolEvent(vshControl *ctl, const vshCmd *cmd) if (vshCommandOptStringReq(ctl, cmd, "event", &eventName) < 0) return false; if (!eventName) { - vshError(ctl, "%s", _("either --list or event type is required")); + vshError(ctl, "%s", _("either --list or --event type is required")); return false; } -- 2.7.4

2 2

[libvirt] qapi DEVICE_DELETED event issued *before* instance_finalize?!
by Alex Williamson 06 Sep '16

06 Sep '16

Hey, I'm out of my QOM depth, so I'll just beg for help in advance. I noticed in testing vfio-pci hotunplug that the host seems to be trying to reclaim the device before QEMU is actually done with it, there's a very short race where libvirt has seen the DEVICE_DELETED event and tries to unbind the physical device from vfio-pci, the use count is clearly non-zero because the host driver tries to send a device request, but that event channel has already been torn down. Nearly immediately after, QEMU finally releases the device, but we can't do a proper reset due to some issues with device references in the kernel. When I run gdb on QEMU with breakpoints at qapi_event_send_device_deleted() and vfio_instance_finalize(), the QAPI even happens first. Clearly this is horribly wrong, right? I can't unmap my references to the vfio device file until my instance_finalize is called, so I'm always going to have that open when libvirt takes the DEVICE_DELETED event as a cue to return the device to host drivers. The call chains look like this: #0 qapi_event_send_device_deleted (has_device=true, device=0x7f5ca3e36fb0 "hostdev0", path=0x7f5c89e84fe0 "/machine/peripheral/hostdev0", errp=0x7f5ca241f9e8 <error_abort>) at qapi-event.c:412 #1 0x00007f5ca1701608 in device_unparent (obj=0x7f5ca43ffc00) at hw/core/qdev.c:1115 #2 0x00007f5ca18b7891 in object_finalize_child_property (obj=0x7f5ca380f500, name=0x7f5ca3f21da0 "hostdev0", opaque=0x7f5ca43ffc00) at qom/object.c:1362 #3 0x00007f5ca18b56b2 in object_property_del_child (obj=0x7f5ca380f500, child=0x7f5ca43ffc00, errp=0x0) at qom/object.c:422 #4 0x00007f5ca18b5790 in object_unparent (obj=0x7f5ca43ffc00) at qom/object.c:441 #5 0x00007f5ca16c1f31 in acpi_pcihp_eject_slot (s=0x7f5ca4c41268, bsel=0, slots=4) at hw/acpi/pcihp.c:139 #0 vfio_instance_finalize (obj=0x7f5ca43ffc00) at /net/gimli/home/alwillia/Work/qemu.git/hw/vfio/pci.c:2731 #1 0x00007f5ca18b57c0 in object_deinit (obj=0x7f5ca43ffc00, type=0x7f5ca376f490) at qom/object.c:448 #2 0x00007f5ca18b5831 in object_finalize (data=0x7f5ca43ffc00) at qom/object.c:462 #3 0x00007f5ca18b6782 in object_unref (obj=0x7f5ca43ffc00) at qom/object.c:896 #4 0x00007f5ca1550cc0 in memory_region_unref (mr=0x7f5ca43fff00) at /net/gimli/home/alwillia/Work/qemu.git/memory.c:1476 #5 0x00007f5ca1553886 in do_address_space_destroy (as=0x7f5ca43ffe10) at /net/gimli/home/alwillia/Work/qemu.git/memory.c:2272 It appears that DEVICE_DELETED only means the VM is done with the device but libvirt is interpreting it as QEMU is done with the device. Which is correct? Do we need a new event or do we need to fix the ordering of this event? An ordering fix would be more compatible with existing libvirt. Thanks, Alex

5 8

[libvirt] virt-admin commands aliases
by Erik Skultety 06 Sep '16

06 Sep '16

Hi there, after my presentation at KVM Forum, it was pointed out from the audience that we might think about doing something about the naming of the virt-admin's comands, since there is some sort of inconsistency: srv- vs. client- vs. dmn- (not merged yet). When I sent patches to upstream I already knew that the naming was not optimal, but I didn't come up with anything better so I hoped that the reviewer might think of something better which unfortunately did not happen. Anyway, there are multiple options how this can be approached but I'm not 100% satisfied with neither of them: 1) rename the commands completely Although clean, obviously this is the non-preferred option because this would break any backwards compatibility however, I think there is a fair chance that people haven't actually started using it yet (although that might change between 7.3 and 7.4). 2) create aliases for non-abbreviated forms of the commands That way, srv- would become server- and dmn- would become daemon-. However, by doing this we'll end up with 6 almost identical entries in the commands structure which might be error-prone once we decide to add/create&add a flag to the command primitive, since the flag would have to be added both to the alias and to the original (unlikely, but possible that someone might forget about that) 3) abbreviate client- to something like clnt- Identical to the above except for the amount of duplicate entries which would be reduced to 2 4) leave it as is if such a consensus is reached and accepted I guess this does no need any additional comments. Thanks for any ideas. Erik

4 6

[libvirt] [PATCH] qemu: Don't warn about missing device in DEVICE_DELETED event
by Jiri Denemark 06 Sep '16

06 Sep '16

Debug priority is good enough for this. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/qemu/qemu_monitor_json.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index 3d0a120..8384314 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -941,7 +941,7 @@ qemuMonitorJSONHandleDeviceDeleted(qemuMonitorPtr mon, virJSONValuePtr data) const char *device; if (!(device = virJSONValueObjectGetString(data, "device"))) { - VIR_WARN("missing device in device deleted event"); + VIR_DEBUG("missing device in device deleted event"); return; } -- 2.10.0

2 1

[libvirt] [PATCH 0/2] util: storage: Fixes for the JSON pseudo protocol parser
by Peter Krempa 06 Sep '16

06 Sep '16

Gluster protocol type was not set properly and the RBD protocol was missing. Peter Krempa (2): util: storage: Properly set protocol type when parsing gluster json string util: storage: Add json pseudo protocol support for legacy RBD strings src/util/virstoragefile.c | 26 ++++++++++++++++++++++++++ tests/virstoragetest.c | 10 ++++++++-- 2 files changed, 34 insertions(+), 2 deletions(-) -- 2.9.2

2 3

[libvirt] [PATCH] docs: Add libvirt-go Go bindings to binding page
by Roman Mohr 06 Sep '16

06 Sep '16

Signed-off-by: Roman Mohr <rmohr(a)redhat.com> --- docs/bindings.html.in | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/docs/bindings.html.in b/docs/bindings.html.in index 95cfe25..7fe26df 100644 --- a/docs/bindings.html.in +++ b/docs/bindings.html.in @@ -14,6 +14,10 @@ <strong>C#</strong>: Arnaud Champion develops <a href="csharp.html">C# bindings</a>. </li> + <li> + <strong>Go</strong>: Kyle Kelley et al. are developing + <a href="https://github.com/rgbkrk/libvirt-go">Go bindings</a>. + </li> <li> <strong>Java</strong>: Daniel Veillard develops <a href="java.html">Java bindings</a>. -- 2.5.5

2 1

[libvirt] [PATCH 0/2] Improve handling of QEMU core dumping
by Daniel P. Berrange 06 Sep '16

06 Sep '16

Daniel P. Berrange (2): qemu: add a max_core setting to qemu.conf for core dump size qemu: allow turning off QEMU guest RAM dump globally src/libvirt_private.syms | 2 ++ src/qemu/libvirtd_qemu.aug | 5 +++++ src/qemu/qemu.conf | 31 +++++++++++++++++++++++++++++++ src/qemu/qemu_command.c | 18 ++++++++++++------ src/qemu/qemu_conf.c | 20 ++++++++++++++++++++ src/qemu/qemu_conf.h | 2 ++ src/qemu/qemu_process.c | 1 + src/qemu/test_libvirtd_qemu.aug.in | 2 ++ src/util/vircommand.c | 14 ++++++++++++++ src/util/vircommand.h | 1 + src/util/virprocess.c | 36 ++++++++++++++++++++++++++++++++++++ src/util/virprocess.h | 1 + tests/qemuxml2argvtest.c | 4 ++++ 13 files changed, 131 insertions(+), 6 deletions(-) -- 2.7.4

3 6

[libvirt] [PATCH] Prefix major/minor with gnu_dev_
by Michal Privoznik 06 Sep '16

06 Sep '16

In the latest glibc, major() and minor() functions are marked as deprecated (glibc commit dbab6577): CC util/libvirt_util_la-vircgroup.lo util/vircgroup.c: In function 'virCgroupGetBlockDevString': util/vircgroup.c:768:5: error: '__major_from_sys_types' is deprecated: In the GNU C Library, `major' is defined by <sys/sysmacros.h>. For historical compatibility, it is currently defined by <sys/types.h> as well, but we plan to remove this soon. To use `major', include <sys/sysmacros.h> directly. If you did not intend to use a system-defined macro `major', you should #undef it after including <sys/types.h>. [-Werror=deprecated-declarations] if (virAsprintf(&ret, "%d:%d ", major(sb.st_rdev), minor(sb.st_rdev)) < 0) ^~ In file included from /usr/include/features.h:397:0, from /usr/include/bits/libc-header-start.h:33, from /usr/include/stdio.h:28, from ../gnulib/lib/stdio.h:43, from util/vircgroup.c:26: /usr/include/sys/sysmacros.h:87:1: note: declared here __SYSMACROS_DEFINE_MAJOR (__SYSMACROS_FST_IMPL_TEMPL) ^ Applications are supposed to use gnu_dev_major() or gnu_dev_minor() respectively. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/conf/domain_audit.c | 4 ++-- src/lxc/lxc_controller.c | 10 +++++----- src/lxc/lxc_driver.c | 26 +++++++++++++------------- src/util/vircgroup.c | 10 +++++----- src/util/virutil.c | 6 +++--- tests/vircgroupmock.c | 16 ++++++++-------- 6 files changed, 36 insertions(+), 36 deletions(-) diff --git a/src/conf/domain_audit.c b/src/conf/domain_audit.c index 53a58ac..d0bfa58 100644 --- a/src/conf/domain_audit.c +++ b/src/conf/domain_audit.c @@ -52,8 +52,8 @@ virDomainAuditGetRdev(const char *path) if (stat(path, &sb) == 0 && (S_ISCHR(sb.st_mode) || S_ISBLK(sb.st_mode))) { - int maj = major(sb.st_rdev); - int min = minor(sb.st_rdev); + int maj = gnu_dev_major(sb.st_rdev); + int min = gnu_dev_minor(sb.st_rdev); ignore_value(virAsprintfQuiet(&ret, "%02X:%02X", maj, min)); } return ret; diff --git a/src/lxc/lxc_controller.c b/src/lxc/lxc_controller.c index 825b4d4..3bf7eb2 100644 --- a/src/lxc/lxc_controller.c +++ b/src/lxc/lxc_controller.c @@ -1520,7 +1520,7 @@ static int virLXCControllerPopulateDevices(virLXCControllerPtr ctrl) LXC_STATE_DIR, ctrl->def->name, devs[i].path) < 0) goto cleanup; - dev_t dev = makedev(devs[i].maj, devs[i].min); + dev_t dev = gnu_dev_makedev(devs[i].maj, devs[i].min); if (mknod(path, S_IFCHR, dev) < 0 || chmod(path, devs[i].mode)) { virReportSystemError(errno, @@ -1592,7 +1592,7 @@ virLXCControllerSetupHostdevSubsysUSB(virDomainDefPtr vmDef, } VIR_DEBUG("Creating dev %s (%d,%d)", - dstfile, major(sb.st_rdev), minor(sb.st_rdev)); + dstfile, gnu_dev_major(sb.st_rdev), gnu_dev_minor(sb.st_rdev)); if (mknod(dstfile, mode, sb.st_rdev) < 0) { virReportSystemError(errno, _("Unable to create device %s"), @@ -1672,7 +1672,7 @@ virLXCControllerSetupHostdevCapsStorage(virDomainDefPtr vmDef, mode = 0700 | S_IFBLK; VIR_DEBUG("Creating dev %s (%d,%d)", dst, - major(sb.st_rdev), minor(sb.st_rdev)); + gnu_dev_major(sb.st_rdev), gnu_dev_minor(sb.st_rdev)); if (mknod(dst, mode, sb.st_rdev) < 0) { virReportSystemError(errno, _("Unable to create device %s"), @@ -1751,7 +1751,7 @@ virLXCControllerSetupHostdevCapsMisc(virDomainDefPtr vmDef, mode = 0700 | S_IFCHR; VIR_DEBUG("Creating dev %s (%d,%d)", dst, - major(sb.st_rdev), minor(sb.st_rdev)); + gnu_dev_major(sb.st_rdev), gnu_dev_minor(sb.st_rdev)); if (mknod(dst, mode, sb.st_rdev) < 0) { virReportSystemError(errno, _("Unable to create device %s"), @@ -1911,7 +1911,7 @@ static int virLXCControllerSetupDisk(virLXCControllerPtr ctrl, * to that normally implied by the device name */ VIR_DEBUG("Creating dev %s (%d,%d) from %s", - dst, major(sb.st_rdev), minor(sb.st_rdev), tmpsrc); + dst, gnu_dev_major(sb.st_rdev), gnu_dev_minor(sb.st_rdev), tmpsrc); if (mknod(dst, mode, sb.st_rdev) < 0) { virReportSystemError(errno, _("Unable to create device %s"), diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index da98b38..3fe79fd 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -3714,7 +3714,7 @@ lxcDomainAttachDeviceMknodHelper(pid_t pid ATTRIBUTE_UNUSED, * to that normally implied by the device name */ VIR_DEBUG("Creating dev %s (%d,%d)", - data->file, major(data->dev), minor(data->dev)); + data->file, gnu_dev_major(data->dev), gnu_dev_minor(data->dev)); if (mknod(data->file, data->mode, data->dev) < 0) { virReportSystemError(errno, _("Unable to create device %s"), @@ -3895,8 +3895,8 @@ lxcDomainAttachDeviceDiskLive(virLXCDriverPtr driver, if (virCgroupAllowDevice(priv->cgroup, 'b', - major(sb.st_rdev), - minor(sb.st_rdev), + gnu_dev_major(sb.st_rdev), + gnu_dev_minor(sb.st_rdev), perms) < 0) goto cleanup; @@ -3915,8 +3915,8 @@ lxcDomainAttachDeviceDiskLive(virLXCDriverPtr driver, file) < 0) { if (virCgroupDenyDevice(priv->cgroup, 'b', - major(sb.st_rdev), - minor(sb.st_rdev), + gnu_dev_major(sb.st_rdev), + gnu_dev_minor(sb.st_rdev), perms) < 0) VIR_WARN("cannot deny device %s for domain %s", src, vm->def->name); @@ -4152,8 +4152,8 @@ lxcDomainAttachDeviceHostdevStorageLive(virLXCDriverPtr driver, if (virCgroupAllowDevice(priv->cgroup, 'b', - major(sb.st_rdev), - minor(sb.st_rdev), + gnu_dev_major(sb.st_rdev), + gnu_dev_minor(sb.st_rdev), VIR_CGROUP_DEVICE_RWM) < 0) goto cleanup; @@ -4165,8 +4165,8 @@ lxcDomainAttachDeviceHostdevStorageLive(virLXCDriverPtr driver, def->source.caps.u.storage.block) < 0) { if (virCgroupDenyDevice(priv->cgroup, 'b', - major(sb.st_rdev), - minor(sb.st_rdev), + gnu_dev_major(sb.st_rdev), + gnu_dev_minor(sb.st_rdev), VIR_CGROUP_DEVICE_RWM) < 0) VIR_WARN("cannot deny device %s for domain %s", def->source.caps.u.storage.block, vm->def->name); @@ -4221,8 +4221,8 @@ lxcDomainAttachDeviceHostdevMiscLive(virLXCDriverPtr driver, if (virCgroupAllowDevice(priv->cgroup, 'c', - major(sb.st_rdev), - minor(sb.st_rdev), + gnu_dev_major(sb.st_rdev), + gnu_dev_minor(sb.st_rdev), VIR_CGROUP_DEVICE_RWM) < 0) goto cleanup; @@ -4237,8 +4237,8 @@ lxcDomainAttachDeviceHostdevMiscLive(virLXCDriverPtr driver, def->source.caps.u.misc.chardev) < 0) { if (virCgroupDenyDevice(priv->cgroup, 'c', - major(sb.st_rdev), - minor(sb.st_rdev), + gnu_dev_major(sb.st_rdev), + gnu_dev_minor(sb.st_rdev), VIR_CGROUP_DEVICE_RWM) < 0) VIR_WARN("cannot deny device %s for domain %s", def->source.caps.u.storage.block, vm->def->name); diff --git a/src/util/vircgroup.c b/src/util/vircgroup.c index f2477d5..f4f6a01 100644 --- a/src/util/vircgroup.c +++ b/src/util/vircgroup.c @@ -765,7 +765,7 @@ virCgroupGetBlockDevString(const char *path) /* Automatically append space after the string since all callers * use it anyway */ - if (virAsprintf(&ret, "%d:%d ", major(sb.st_rdev), minor(sb.st_rdev)) < 0) + if (virAsprintf(&ret, "%d:%d ", gnu_dev_major(sb.st_rdev), gnu_dev_minor(sb.st_rdev)) < 0) return NULL; return ret; @@ -2941,8 +2941,8 @@ virCgroupAllowDevicePath(virCgroupPtr group, return virCgroupAllowDevice(group, S_ISCHR(sb.st_mode) ? 'c' : 'b', - major(sb.st_rdev), - minor(sb.st_rdev), + gnu_dev_major(sb.st_rdev), + gnu_dev_minor(sb.st_rdev), perms); } @@ -3032,8 +3032,8 @@ virCgroupDenyDevicePath(virCgroupPtr group, return virCgroupDenyDevice(group, S_ISCHR(sb.st_mode) ? 'c' : 'b', - major(sb.st_rdev), - minor(sb.st_rdev), + gnu_dev_major(sb.st_rdev), + gnu_dev_minor(sb.st_rdev), perms); } diff --git a/src/util/virutil.c b/src/util/virutil.c index 170dd59..b0770bc 100644 --- a/src/util/virutil.c +++ b/src/util/virutil.c @@ -1616,7 +1616,7 @@ virIsDevMapperDevice(const char *dev_name) if (!stat(dev_name, &buf) && S_ISBLK(buf.st_mode) && - dm_is_dm_major(major(buf.st_rdev))) + dm_is_dm_major(gnu_dev_major(buf.st_rdev))) return true; return false; @@ -1664,9 +1664,9 @@ virGetDeviceID(const char *path, int *maj, int *min) return -EINVAL; if (maj) - *maj = major(sb.st_rdev); + *maj = gnu_dev_major(sb.st_rdev); if (min) - *min = minor(sb.st_rdev); + *min = gnu_dev_minor(sb.st_rdev); return 0; } diff --git a/tests/vircgroupmock.c b/tests/vircgroupmock.c index ce6fd46..8f69047 100644 --- a/tests/vircgroupmock.c +++ b/tests/vircgroupmock.c @@ -562,11 +562,11 @@ int __lxstat(int ver, const char *path, struct stat *sb) free(newpath); } else if (STRPREFIX(path, fakedevicedir0)) { sb->st_mode = S_IFBLK; - sb->st_rdev = makedev(8, 0); + sb->st_rdev = gnu_dev_makedev(8, 0); return 0; } else if (STRPREFIX(path, fakedevicedir1)) { sb->st_mode = S_IFBLK; - sb->st_rdev = makedev(9, 0); + sb->st_rdev = gnu_dev_makedev(9, 0); return 0; } else { ret = real___lxstat(ver, path, sb); @@ -593,11 +593,11 @@ int lstat(const char *path, struct stat *sb) free(newpath); } else if (STRPREFIX(path, fakedevicedir0)) { sb->st_mode = S_IFBLK; - sb->st_rdev = makedev(8, 0); + sb->st_rdev = gnu_dev_makedev(8, 0); return 0; } else if (STRPREFIX(path, fakedevicedir1)) { sb->st_mode = S_IFBLK; - sb->st_rdev = makedev(9, 0); + sb->st_rdev = gnu_dev_makedev(9, 0); return 0; } else { ret = real_lstat(path, sb); @@ -624,11 +624,11 @@ int __xstat(int ver, const char *path, struct stat *sb) free(newpath); } else if (STRPREFIX(path, fakedevicedir0)) { sb->st_mode = S_IFBLK; - sb->st_rdev = makedev(8, 0); + sb->st_rdev = gnu_dev_makedev(8, 0); return 0; } else if (STRPREFIX(path, fakedevicedir1)) { sb->st_mode = S_IFBLK; - sb->st_rdev = makedev(9, 0); + sb->st_rdev = gnu_dev_makedev(9, 0); return 0; } else { ret = real___xstat(ver, path, sb); @@ -661,11 +661,11 @@ int stat(const char *path, struct stat *sb) } } else if (STRPREFIX(path, fakedevicedir0)) { sb->st_mode = S_IFBLK; - sb->st_rdev = makedev(8, 0); + sb->st_rdev = gnu_dev_makedev(8, 0); return 0; } else if (STRPREFIX(path, fakedevicedir1)) { sb->st_mode = S_IFBLK; - sb->st_rdev = makedev(9, 0); + sb->st_rdev = gnu_dev_makedev(9, 0); return 0; } else { if (!(newpath = strdup(path))) -- 2.8.4

3 2

[libvirt] Nice golang bindings for libvirt
by Roman Mohr 06 Sep '16

06 Sep '16

Hi all, I have been using these pretty nice libvirt-binding library for golang [1]. It is under active development, pretty well maintained and has a very nice travis setup to run tests against libvirt [2]. I was wondering if the bindings could be added to [3] on libvirt.org. Best Regards, Roman [1] https://github.com/rgbkrk [2] https://github.com/rgbkrk/libvirt-go/blob/master/.travis.yml [3] https://libvirt.org/bindings.html

2 2

[libvirt] [libvirt-glib/libvirt-gconfig 00/17] Graphics: Introduce the new Remote and Local classes (and also implement a few missing methods).
by Fabiano Fidêncio 06 Sep '16

06 Sep '16

While trying to use libvirt-gobject and libvirt-gconfig for accessing VMs and looking at their config, instead of using libvirt and parsing XML directly, I found out that a few methods have been missing and that libvirt-gconfig is not exactly thought for the "reading their config" use case (see more explanations on the 10th and 14th commits. This series, unfortunately, introduces an ABI breakage. Fabiano Fidêncio (17): gconfig: Implement gvir_config_domain_graphics_vnc_get_autoport() gconfig: Implement gvir_config_domain_graphics_spice_get_autoport() gconfig: Implement gvir_config_domain_graphics_rdp_get_autoport() gconfig: Implement gvir_config_domain_graphics_sdl_get_display() gconfig: Implement gvir_config_domain_graphics_sdl_get_fullscreen() gconfig: Implement gvir_config_domain_graphics_spice_get_tls_port() gconfig: Implement gvir_config_domain_graphics_spice_{get,set}_host() gconfig: Implement gvir_config_domain_graphics_vnc_{get,set}_host() gconfig: Implement gvir_config_domain_graphics_rdp_{get,set}_host() gconfig: Add GVirCofigDomainGraphicsRemote class gconfig: Adapt GVirConfigDomainGraphicsSpice to GVirConfigDomainGraphicsRemote gconfig: Adapt GVirConfigDomainGraphicsRdp to GVirConfigDomainGraphicsRemote gconfig: Adapt GVirConfigDomainGraphicsVnc to GVirConfigDomainGraphicsRemote gconfig: Add GVirCofigDomainGraphicsLocal class gconfig: Adapt GVirConfigDomainGraphicsSdl to GVirConfigDomainGraphicsLocal gconfig: Adapt GVirConfigDomainGraphicsDesktop to GVirConfigDomainGraphicsLocal gconfig,graphics: Avoid crash when gvir_config_object_new_from_xml() returns NULL libvirt-gconfig/Makefile.am | 4 + .../libvirt-gconfig-domain-graphics-desktop.c | 14 ++- .../libvirt-gconfig-domain-graphics-desktop.h | 4 +- .../libvirt-gconfig-domain-graphics-local.c | 97 +++++++++++++++++++ .../libvirt-gconfig-domain-graphics-local.h | 68 ++++++++++++++ .../libvirt-gconfig-domain-graphics-rdp.c | 32 ++++++- .../libvirt-gconfig-domain-graphics-rdp.h | 9 +- .../libvirt-gconfig-domain-graphics-remote.c | 103 +++++++++++++++++++++ .../libvirt-gconfig-domain-graphics-remote.h | 70 ++++++++++++++ .../libvirt-gconfig-domain-graphics-sdl.c | 19 +++- .../libvirt-gconfig-domain-graphics-sdl.h | 6 +- .../libvirt-gconfig-domain-graphics-spice.c | 40 +++++++- .../libvirt-gconfig-domain-graphics-spice.h | 10 +- .../libvirt-gconfig-domain-graphics-vnc.c | 32 ++++++- .../libvirt-gconfig-domain-graphics-vnc.h | 9 +- libvirt-gconfig/libvirt-gconfig.h | 2 + libvirt-gconfig/libvirt-gconfig.sym | 20 ++++ po/POTFILES.in | 2 + 18 files changed, 513 insertions(+), 28 deletions(-) create mode 100644 libvirt-gconfig/libvirt-gconfig-domain-graphics-local.c create mode 100644 libvirt-gconfig/libvirt-gconfig-domain-graphics-local.h create mode 100644 libvirt-gconfig/libvirt-gconfig-domain-graphics-remote.c create mode 100644 libvirt-gconfig/libvirt-gconfig-domain-graphics-remote.h -- 2.5.0

2 30