- Devel - libvirt List Archives

[libvirt] [PATCH] libxl: Avoid a variable named 'stat'
by George Dunlap 24 Aug '17

24 Aug '17

Using a variable named 'stat' clashes with the system function 'stat()' causing compiler warnings on some platforms: libxl/libxl_driver.c: In function 'libxlDomainBlockStatsVBD': libxl/libxl_driver.c:5387: error: declaration of 'stat' shadows a global declaration [-Wshadow] /usr/include/sys/stat.h:455: error: shadowed declaration is here [-Wshadow] Signed-off-by: George Dunlap <george.dunlap(a)citrix.com> --- CC: Daniel P. Berrange <berrange(a)redhat.com> CC: Jim Fehlig <jfehlig(a)suse.com> CC: xen-devel(a)xenproject.org CC: libvir-list(a)redhat.com --- src/libxl/libxl_driver.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/src/libxl/libxl_driver.c b/src/libxl/libxl_driver.c index c261e2155..4c7775358 100644 --- a/src/libxl/libxl_driver.c +++ b/src/libxl/libxl_driver.c @@ -5374,7 +5374,7 @@ libxlDomainBlockStatsVBD(virDomainObjPtr vm, int devno = libxlDiskPathToID(dev); int size; char *path, *name, *val; - unsigned long long stat; + unsigned long long status; path = name = val = NULL; if (devno < 0) { @@ -5401,12 +5401,12 @@ libxlDomainBlockStatsVBD(virDomainObjPtr vm, # define LIBXL_SET_VBDSTAT(FIELD, VAR, MUL) \ if ((virAsprintf(&name, "%s/"FIELD, path) < 0) || \ (virFileReadAll(name, 256, &val) < 0) || \ - (sscanf(val, "%llu", &stat) != 1)) { \ + (sscanf(val, "%llu", &status) != 1)) { \ virReportError(VIR_ERR_OPERATION_FAILED, \ _("cannot read %s"), name); \ goto cleanup; \ } \ - VAR += (stat * MUL); \ + VAR += (status * MUL); \ VIR_FREE(name); \ VIR_FREE(val); -- 2.14.1

3 2

[libvirt] [PATCH] qemu: Report error on failure to set isolation group
by Andrea Bolognani 24 Aug '17

24 Aug '17

This is more user-friendly because the error will be displayed directly instead of being buried in the log. Signed-off-by: Andrea Bolognani <abologna(a)redhat.com> --- src/qemu/qemu_domain_address.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) diff --git a/src/qemu/qemu_domain_address.c b/src/qemu/qemu_domain_address.c index 8066ed123..16bf0cdf9 100644 --- a/src/qemu/qemu_domain_address.c +++ b/src/qemu/qemu_domain_address.c @@ -1012,10 +1012,11 @@ qemuDomainFillDeviceIsolationGroup(virDomainDefPtr def, tmp = virPCIDeviceAddressGetIOMMUGroupNum(hostAddr); if (tmp < 0) { - VIR_WARN("Can't look up isolation group for host device " - "%04x:%02x:%02x.%x", - hostAddr->domain, hostAddr->bus, - hostAddr->slot, hostAddr->function); + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Can't look up isolation group for host device " + "%04x:%02x:%02x.%x"), + hostAddr->domain, hostAddr->bus, + hostAddr->slot, hostAddr->function); goto cleanup; } @@ -1056,9 +1057,11 @@ qemuDomainFillDeviceIsolationGroup(virDomainDefPtr def, tmp = qemuDomainFindUnusedIsolationGroup(def); if (tmp == 0) { - VIR_WARN("Can't obtain usable isolation group for interface " - "configured to use hostdev-backed network '%s'", - iface->data.network.name); + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Can't obtain usable isolation group for " + "interface configured to use hostdev-backed " + "network '%s'"), + iface->data.network.name); goto cleanup; } -- 2.13.5

2 1

[libvirt] Question about "f16c", "rdrand" and "vme" in cpu_map.xml
by Jay Zhou 24 Aug '17

24 Aug '17

Hi all, "f16c", "rdrand" and "vme" are all standard features, the IvyBridge CPU model has these three features in the file of src/cpu/cpu_map.xml, but the higher model of Haswell-noTSX, Haswell, Broadwell-noTSX and Broadwell do not, which makes it incompatible. BTW, I noticed that "f16c", "rdrand" and "vme" are a part of features of Haswell-noTSX, Haswell, Broadwell-noTSX and Broadwell in QEMU. The commit 732d3d147adecb867f62c67ab3743c0aefd87e50(which introduces the Haswell-noTSX CPU model in libvirt) said that: "Inheritance among CPU model is cool but it makes reviewing CPU model definitions and comparing them to CPU models from QEMU rather hard and unpleasant." I do not get the point "it makes reviewing CPU model definitions and comparing them to CPU models from QEMU rather hard and unpleasant" clearly. Could anyone explain it more explicitly? Looking forward to your replay. Thanks, Jay

2 1

[libvirt] [PATCH] maint: Update to latest gnulib
by Erik Skultety 24 Aug '17

24 Aug '17

This pulls in, among other new things, vc-list-files fix to make syntax-check work with git worktrees. Signed-off-by: Erik Skultety <eskultet(a)redhat.com> --- .gnulib | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.gnulib b/.gnulib index ce4ee4cbb..47c634c05 160000 --- a/.gnulib +++ b/.gnulib @@ -1 +1 @@ -Subproject commit ce4ee4cbb596a9d7de2786cf8c48cf62a4edede7 +Subproject commit 47c634c05a946651e2834dcf0cd7cd0dc7a21d3c -- 2.13.3

2 2

[libvirt] [PATCH v5 00/15] virObject adjustments for common object
by John Ferlan 24 Aug '17

24 Aug '17

v4: https://www.redhat.com/archives/libvir-list/2017-August/msg00537.html Changes since v4... I've removed the virObjectLookupKeys completely relying on LookupHash completely. The adjusted LookupHash object uses named "objsUUID" and "objsName" as well as API's to specifically Search or ForEach by Name or UUID. The New, Add, and Remove API's were adjusted in order to handle creating things properly. Even if that adjustment isn't considered what's desired, at least look at patches 1, 10, and 14 separately. John Ferlan (15): util: Use VIR_ERROR instead of VIR_WARN util: Introduce virObjectLookupHash util: Introduce virObjectLookupHash{Add|Remove} util: Introduce virObjectLookupHashFind[Locked] util: Introduce virObjectLookupHashForEach{UUID|Name} util: Introduce virObjectLookupHashSearch{UUID|Name}[Locked] nodedev: Use virObjectLookupHash secret: Use virObjectLookupHash util: Introduce virObjectLookupHashClone Revert "interface: Consume @def in virInterfaceObjNew" interface: Use virObjectLookupHash test: Clean up test driver Interface interactions util: Introduce virObjectLookupHashPrune network: Fix virNetworkObjBridgeInUse return type network: Use virObjectLookupHash src/conf/virinterfaceobj.c | 289 ++++++++++--------- src/conf/virnetworkobj.c | 280 ++++++------------ src/conf/virnetworkobj.h | 2 +- src/conf/virnodedeviceobj.c | 273 +++++------------- src/conf/virsecretobj.c | 251 +++++----------- src/libvirt_private.syms | 14 + src/test/test_driver.c | 55 +--- src/util/virobject.c | 675 +++++++++++++++++++++++++++++++++++++++++++- src/util/virobject.h | 123 ++++++++ 9 files changed, 1189 insertions(+), 773 deletions(-) -- 2.9.5

2 17

[libvirt] [PATCH] vz: build fix
by Nikolay Shirokovskiy 24 Aug '17

24 Aug '17

6e6faf6d changed vzDomObjAlloc signature in source but not in header file. --- src/vz/vz_utils.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/vz/vz_utils.h b/src/vz/vz_utils.h index ea2e2da..d213183 100644 --- a/src/vz/vz_utils.h +++ b/src/vz/vz_utils.h @@ -116,7 +116,7 @@ struct vzDomObj { typedef struct vzDomObj *vzDomObjPtr; -void* vzDomObjAlloc(void); +void* vzDomObjAlloc(void *opaque); void vzDomObjFree(void *p); virDomainObjPtr vzDomObjFromDomain(virDomainPtr domain); -- 1.8.3.1

2 1

[libvirt] [PATCH 0/2] conf: check address type for USB hostdevs
by Ján Tomko 23 Aug '17

23 Aug '17

*** BRB HERE *** Ján Tomko (2): conf: move hostdev address validation to virDomainHostdevDefValidate conf: check address type for USB hostdevs src/conf/domain_conf.c | 83 ++++++++++++++++++++++++++++++++++---------------- 1 file changed, 56 insertions(+), 27 deletions(-) -- 2.13.0

2 3

[libvirt] [PATCH] vmx: do not treat controllers as implicit devices
by Ján Tomko 23 Aug '17

23 Aug '17

When parsing the config, we look for the SCSI controllers one by one, remembering their models, then let virDomainDefAddImplicitDevices add them if any SCSI disk is using them. Since these controllers are not really implicit (they are present in the source config), add them explicitly. This patch maintains the behavior of not adding a controller if it was present in the config, but no disk was using it. This also resolves the memory leak of virVMXParseConfig overwriting the video device added by calling virDomainDefAddImplicitDevices before the parsing is finished. Reported-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/vmx/vmx.c | 32 ++++++++++++-------------------- 1 file changed, 12 insertions(+), 20 deletions(-) diff --git a/src/vmx/vmx.c b/src/vmx/vmx.c index 3e2f4c3e1..849cfc6b1 100644 --- a/src/vmx/vmx.c +++ b/src/vmx/vmx.c @@ -1650,6 +1650,18 @@ virVMXParseConfig(virVMXContext *ctx, if (def->disks[def->ndisks] != NULL) ++def->ndisks; } + + } + + /* add all the SCSI controllers we've seen, up until the last one that is + * currently used by a disk */ + if (def->ndisks != 0) { + virDomainDeviceInfoPtr info = &def->disks[def->ndisks - 1]->info; + for (controller = 0; controller <= info->addr.drive.controller; controller++) { + if (virDomainDefAddController(def, VIR_DOMAIN_CONTROLLER_TYPE_SCSI, + controller, scsi_virtualDev[controller]) < 0) + goto cleanup; + } } /* def:disks (ide) */ @@ -1689,26 +1701,6 @@ virVMXParseConfig(virVMXContext *ctx, ++def->ndisks; } - /* def:controllers */ - if (virDomainDefAddImplicitDevices(def) < 0) { - virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Could not add controllers")); - goto cleanup; - } - - for (controller = 0; controller < def->ncontrollers; ++controller) { - if (def->controllers[controller]->type == VIR_DOMAIN_CONTROLLER_TYPE_SCSI) { - if (def->controllers[controller]->idx > 3) { - virReportError(VIR_ERR_INTERNAL_ERROR, - _("SCSI controller index %d out of [0..3] range"), - def->controllers[controller]->idx); - goto cleanup; - } - - def->controllers[controller]->model = - scsi_virtualDev[def->controllers[controller]->idx]; - } - } - /* def:fss */ if (virVMXGetConfigBoolean(conf, "isolation.tools.hgfs.disable", &hgfs_disabled, true, true) < 0) { -- 2.13.0

2 1

[libvirt] [PATCH] tests: Fix leak in securityselinuxtest
by John Ferlan 23 Aug '17

23 Aug '17

If we jump to the error: label and @secbuf is allocated, then it's not free'd at all. Found by Coverity Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- tests/securityselinuxtest.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/securityselinuxtest.c b/tests/securityselinuxtest.c index f6bc07a..0ac2828 100644 --- a/tests/securityselinuxtest.c +++ b/tests/securityselinuxtest.c @@ -68,7 +68,7 @@ testBuildDomainDef(bool dynamic, const char *baselabel) { virDomainDefPtr def; - virSecurityLabelDefPtr secdef; + virSecurityLabelDefPtr secdef = NULL; if (!(def = virDomainDefNew())) goto error; @@ -98,6 +98,7 @@ testBuildDomainDef(bool dynamic, error: virDomainDefFree(def); + virSecurityLabelDefFree(secdef); return NULL; } -- 2.9.5

2 1

[libvirt] [PATCH 0/2] Yet another namespace fix, kinda
by Martin Kletzander 23 Aug '17

23 Aug '17

Fixing easy problem (patch 1) lead me to finding out that there is yet another problem that needs fixing (patch 2). For more information read the code. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1443434 Martin Kletzander (2): qemu: Don't mangle the storage format for type='dir' qemu: Also treat directories properly when using namespaces src/qemu/qemu_domain.c | 38 +++++++++++++++++++--- src/storage/storage_source.c | 5 +++ .../qemuxml2argv-floppy-drive-noformat.args | 24 ++++++++++++++ .../qemuxml2argv-floppy-drive-noformat.xml | 31 ++++++++++++++++++ tests/qemuxml2argvtest.c | 2 ++ 5 files changed, 95 insertions(+), 5 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-floppy-drive-noformat.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-floppy-drive-noformat.xml -- 2.14.1

2 5