- Devel - libvirt List Archives

[libvirt] [PATCH] storage: logical: drop useless if
by Roman Bogorodskiy 21 Jul '14

21 Jul '14

virStorageBackendLogicalCreateVol contains a piece like: if (vol->target.path != NULL) { /* A target path passed to CreateVol has no meaning */ VIR_FREE(vol->target.path); } The 'if' is useless here, but 'syntax-check' doesn't catch that because of the comment, so drop the 'if'. --- src/storage/storage_backend_logical.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/src/storage/storage_backend_logical.c b/src/storage/storage_backend_logical.c index ed62c2f..4959985 100644 --- a/src/storage/storage_backend_logical.c +++ b/src/storage/storage_backend_logical.c @@ -742,10 +742,8 @@ virStorageBackendLogicalCreateVol(virConnectPtr conn, vol->type = VIR_STORAGE_VOL_BLOCK; - if (vol->target.path != NULL) { - /* A target path passed to CreateVol has no meaning */ - VIR_FREE(vol->target.path); - } + /* A target path passed to CreateVol has no meaning */ + VIR_FREE(vol->target.path); if (virAsprintf(&vol->target.path, "%s/%s", pool->def->target.path, -- 1.9.0

2 2

[libvirt] [PATCH] Fix build on non-Linux platforms
by Roman Bogorodskiy 21 Jul '14

21 Jul '14

Commit ef48a1b introduced virFindSCSIHostByPCI for Linux and a stub for other platforms that returns -1 while the function should return 'char *', so use 'return NULL' instead. Commit fbd91d4 introduced virReadSCSIUniqueId with the third argument 'int *result', however the stub for non-Linux patform uses 'unsigned int *result', so change it to 'int *result'. Pushed under the build breaker rule. --- src/util/virutil.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/util/virutil.c b/src/util/virutil.c index 1c6d261..20e9f0e 100644 --- a/src/util/virutil.c +++ b/src/util/virutil.c @@ -2176,7 +2176,7 @@ virFindFCHostCapableVport(const char *sysfs_prefix) int virReadSCSIUniqueId(const char *sysfs_prefix ATTRIBUTE_UNUSED, int host ATTRIBUTE_UNUSED, - unsigned int *result ATTRIBUTE_UNUSED) + int *result ATTRIBUTE_UNUSED) { virReportSystemError(ENOSYS, "%s", _("Not supported on this platform")); return -1; @@ -2188,7 +2188,7 @@ virFindSCSIHostByPCI(const char *sysfs_prefix ATTRIBUTE_UNUSED, unsigned int unique_id ATTRIBUTE_UNUSED) { virReportSystemError(ENOSYS, "%s", _("Not supported on this platform")); - return -1; + return NULL; } int -- 1.9.0

2 2

[libvirt] [PATCH] build: fix build without numactl
by Eric Blake 21 Jul '14

21 Jul '14

Under ./configure --without-numactl but with numactl-devel installed, the build fails with: ../../src/util/virnuma.c: In function 'virNumaNodeIsAvailable': ../../src/util/virnuma.c:407:5: error: implicit declaration of function 'numa_bitmask_isbitset' [-Werror=implicit-function-declaration] return numa_bitmask_isbitset(numa_nodes_ptr, node); ^ and other failures, all because the configure results for particular functions were used without regard to whether libnuma was even being linked in. * src/util/virnuma.c (virNumaGetPages): Fix message typo. (virNumaNodeIsAvailable): Correct build when not using numactl. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- Pushing under the build-breaker rule. src/util/virnuma.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/util/virnuma.c b/src/util/virnuma.c index 46f48d2..7a11a3b 100644 --- a/src/util/virnuma.c +++ b/src/util/virnuma.c @@ -390,7 +390,7 @@ virNumaGetMaxCPUs(void) } -#ifdef HAVE_NUMA_BITMASK_ISBITSET +#if WITH_NUMACTL && HAVE_NUMA_BITMASK_ISBITSET /** * virNumaNodeIsAvailable: * @node: node to check @@ -485,7 +485,7 @@ virNumaGetDistances(int node ATTRIBUTE_UNUSED, { *distances = NULL; *ndistances = 0; - VIR_DEBUG("NUMA distance information isn't availble on this host"); + VIR_DEBUG("NUMA distance information isn't available on this host"); return 0; } #endif -- 1.9.3

1 0

[libvirt] [REPOST 0/8] storage_scsi: Stable SCSI host addressing
by John Ferlan 21 Jul '14

21 Jul '14

Reposting of a series from last month changing only the span version in the docs from 1.2.6 to 1.2.7. Previous posting here: http://www.redhat.com/archives/libvir-list/2014-June/msg00448.html The concept still remains the same - rather than rely on the hostNN numbers for the scsi_host to remain stable and unique across host reboots and/or kernel rebuilds, allow use a combination of the scsi_host's PCI address and the value from the hostNN's 'unique_id' file. John Ferlan (6): getAdapterName: check for SCSI_HOST scsi_backend: Use existing LINUX_SYSFS_SCSI_HOST_PREFIX definition virutil: Introduce virReadSCSIUniqueId Add unique_id to nodedev output scsi_host: Introduce virFindSCSIHostByPCI getAdapterName: Lookup stable scsi_host Osier Yang (2): virStoragePoolSourceAdapter: Refine the SCSI_HOST adapter name storage: Introduce parentaddr into virStoragePoolSourceAdapter docs/formatnode.html.in | 11 + docs/formatstorage.html.in | 143 ++++++++-- docs/schemas/basictypes.rng | 24 +- docs/schemas/nodedev.rng | 6 + src/conf/node_device_conf.c | 23 +- src/conf/node_device_conf.h | 1 + src/conf/storage_conf.c | 111 +++++++- src/conf/storage_conf.h | 8 +- src/libvirt_private.syms | 2 + src/node_device/node_device_linux_sysfs.c | 6 + src/phyp/phyp_driver.c | 8 +- src/storage/storage_backend_scsi.c | 53 +++- src/test/test_driver.c | 5 +- src/util/virutil.c | 154 +++++++++++ src/util/virutil.h | 8 + tests/Makefile.am | 7 + .../pci_8086_27c5_scsi_host_0_unique_id.xml | 8 + tests/nodedevxml2xmltest.c | 1 + tests/scsihosttest.c | 308 +++++++++++++++++++++ .../pool-scsi-type-scsi-host-stable.xml | 19 ++ .../pool-scsi-type-scsi-host-stable.xml | 22 ++ tests/storagepoolxml2xmltest.c | 1 + 22 files changed, 868 insertions(+), 61 deletions(-) create mode 100644 tests/nodedevschemadata/pci_8086_27c5_scsi_host_0_unique_id.xml create mode 100644 tests/scsihosttest.c create mode 100644 tests/storagepoolxml2xmlin/pool-scsi-type-scsi-host-stable.xml create mode 100644 tests/storagepoolxml2xmlout/pool-scsi-type-scsi-host-stable.xml -- 1.9.3

2 13

[libvirt] [PATCH] schema: pool: netfs: Don't enforce slash in glusterfs pool source
by Peter Krempa 21 Jul '14

21 Jul '14

Gluster volumes don't start with a leading slash. Our schema for netfs gluster pools enforces it though. Luckily mount.glusterfs skips it. Allow a slashless volume name for glusterfs netfs mounts in the schema. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1101999 --- docs/schemas/basictypes.rng | 6 +++ docs/schemas/storagepool.rng | 44 +++++++++++++++++----- .../pool-netfs-gluster-without-slash.xml | 12 ++++++ 3 files changed, 53 insertions(+), 9 deletions(-) create mode 100644 tests/storagepoolxml2xmlin/pool-netfs-gluster-without-slash.xml diff --git a/docs/schemas/basictypes.rng b/docs/schemas/basictypes.rng index 5fe3a97..9c9419f 100644 --- a/docs/schemas/basictypes.rng +++ b/docs/schemas/basictypes.rng @@ -231,6 +231,12 @@ </data> </define> + <define name="dirPath"> + <data type="string"> + <param name="pattern">[a-zA-Z0-9_\.\+\-\\&"'<>/%]*</param> + </data> + </define> + <define name="absFilePath"> <data type="string"> <param name="pattern">/[a-zA-Z0-9_\.\+\-\\&"'<>/%,]+</param> diff --git a/docs/schemas/storagepool.rng b/docs/schemas/storagepool.rng index 8d7a94d..b2d1473 100644 --- a/docs/schemas/storagepool.rng +++ b/docs/schemas/storagepool.rng @@ -327,6 +327,15 @@ </element> </define> + <define name='sourceinfonetfsgluster'> + <element name='dir'> + <attribute name='path'> + <ref name='dirPath'/> + </attribute> + <empty/> + </element> + </define> + <define name='sourceinfoname'> <element name='name'> <text/> @@ -394,7 +403,6 @@ <value>auto</value> <value>nfs</value> <value>cifs</value> - <value>glusterfs</value> </choice> </attribute> </element> @@ -468,14 +476,32 @@ <define name='sourcenetfs'> <element name='source'> - <interleave> - <ref name='sourceinfohost'/> - <ref name='sourceinfodir'/> - <ref name='sourcefmtnetfs'/> - <optional> - <ref name='sourceinfovendor'/> - </optional> - </interleave> + <choice> + <group> + <interleave> + <ref name='sourceinfohost'/> + <ref name='sourceinfodir'/> + <ref name='sourcefmtnetfs'/> + <optional> + <ref name='sourceinfovendor'/> + </optional> + </interleave> + </group> + <group> + <interleave> + <ref name='sourceinfohost'/> + <ref name='sourceinfonetfsgluster'/> + <element name='format'> + <attribute name='type'> + <value>glusterfs</value> + </attribute> + </element> + <optional> + <ref name='sourceinfovendor'/> + </optional> + </interleave> + </group> + </choice> </element> </define> diff --git a/tests/storagepoolxml2xmlin/pool-netfs-gluster-without-slash.xml b/tests/storagepoolxml2xmlin/pool-netfs-gluster-without-slash.xml new file mode 100644 index 0000000..69a2c6d --- /dev/null +++ b/tests/storagepoolxml2xmlin/pool-netfs-gluster-without-slash.xml @@ -0,0 +1,12 @@ +<pool type='netfs'> + <source> + <host name='example.com'/> + <format type='glusterfs'/> + <dir path='volume'/> + </source> + <name>netfs-gluster</name> + <uuid>d5609ced-94b1-489e-b218-eff35c30336a</uuid> + <target> + <path>/mnt/gluster</path> + </target> +</pool> -- 2.0.0

2 3

[libvirt] [PATCH] doc: Explicitly specify how to override spice channel mode
by Peter Krempa 21 Jul '14

21 Jul '14

Be more clear that the "<channel mode=" attribute overrides the default set by "defaultMode". Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1033704 --- docs/formatdomain.html.in | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index bb6f710..8950959 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -4191,7 +4191,13 @@ qemu-kvm -net nic,model=? /dev/null configured, it can be desirable to restrict what channels can be run on each port. This is achieved by adding one or more <channel> elements inside the - main <graphics> element. Valid channel names + main <graphics> element and setting the <code>mode</code> + attribute to either <code>secure</code> or <code>insecure</code>. + Setting the mode attribute overrides the default value as set + by the <code>defaultMode</code> attribute. (Note that specifying + <code>any</code> as mode discards the entry as the channel would + inherit the default mode anyways) + Valid channel names include <code>main</code>, <code>display</code>, <code>inputs</code>, <code>cursor</code>, <code>playback</code>, <code>record</code> -- 2.0.0

2 2

[libvirt] [PATCH] domtop: Fix build on mingw
by Michal Privoznik 21 Jul '14

21 Jul '14

Firstly, there's no sigaction() nor struct sigaction on mingw. We have to use the one implemented by gnulib (and hence link with gnulib). Then, for some reason one header file from windows defines ERROR symbol. Yes it does. Sigh. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- examples/domtop/Makefile.am | 6 ++++-- examples/domtop/domtop.c | 17 +++++++++++++++++ 2 files changed, 21 insertions(+), 2 deletions(-) diff --git a/examples/domtop/Makefile.am b/examples/domtop/Makefile.am index c5cb6c7..dbebb46 100644 --- a/examples/domtop/Makefile.am +++ b/examples/domtop/Makefile.am @@ -16,9 +16,11 @@ ## License along with this library. If not, see ## <http://www.gnu.org/licenses/>. -INCLUDES = -I$(top_builddir)/include -I$(top_srcdir)/include +INCLUDES = -I$(top_builddir)/include -I$(top_srcdir)/include \ + -I$(top_builddir)/gnulib/lib -I$(top_srcdir)/gnulib/lib \ + -I$(top_srcdir) LDADDS = $(STATIC_BINARIES) $(WARN_CFLAGS) $(top_builddir)/src/libvirt.la \ - $(COVERAGE_LDFLAGS) + $(top_builddir)/gnulib/lib/libgnu.la $(COVERAGE_LDFLAGS) noinst_PROGRAMS=domtop diff --git a/examples/domtop/domtop.c b/examples/domtop/domtop.c index 4ac7889..af5da46 100644 --- a/examples/domtop/domtop.c +++ b/examples/domtop/domtop.c @@ -20,6 +20,8 @@ * Author: Michal Privoznik <mprivozn(a)redhat.com> */ +#include <config.h> + #include <errno.h> #include <getopt.h> #include <libvirt/libvirt.h> @@ -35,6 +37,21 @@ static bool debug; static bool run_top; +/* On mingw, there's a header file that poisons the well: + * + * + * CC domtop.o + *domtop.c:40:0: warning: "ERROR" redefined [enabled by default] + * #define ERROR(...) \ + * ^ + *In file included from /usr/i686-w64-mingw32/sys-root/mingw/include/windows.h:71:0, + * from /usr/i686-w64-mingw32/sys-root/mingw/include/winsock2.h:23, + * from ../../gnulib/lib/unistd.h:48, + * from domtop.c:35: + * /usr/i686-w64-mingw32/sys-root/mingw/include/wingdi.h:75:0: note: this is the location of the previous definition + * #define ERROR 0 + */ +#undef ERROR #define ERROR(...) \ do { \ fprintf(stderr, "ERROR %s:%d : ", __FUNCTION__, __LINE__); \ -- 1.8.5.5

2 1

[libvirt] [PATCH] node_device: HAL: Ignore return value of virStrToLong_ui
by Peter Krempa 21 Jul '14

21 Jul '14

Commit 5df813177c3b609a8cf5db26ae94b26d4a40063d forgot to adjust a few callers of virStrToLong_ui to ignore the returned value in some ancient parts of the code. --- Notes: Pushed under build-braker rule as HAL enabled hosts fail to compile. src/node_device/node_device_hal.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/src/node_device/node_device_hal.c b/src/node_device/node_device_hal.c index 8656b5d..cd7d399 100644 --- a/src/node_device/node_device_hal.c +++ b/src/node_device/node_device_hal.c @@ -151,10 +151,10 @@ gather_pci_cap(LibHalContext *ctx, const char *udi, if (get_str_prop(ctx, udi, "pci.linux.sysfs_path", &sysfs_path) == 0) { char *p = strrchr(sysfs_path, '/'); if (p) { - (void)virStrToLong_ui(p+1, &p, 16, &d->pci_dev.domain); - (void)virStrToLong_ui(p+1, &p, 16, &d->pci_dev.bus); - (void)virStrToLong_ui(p+1, &p, 16, &d->pci_dev.slot); - (void)virStrToLong_ui(p+1, &p, 16, &d->pci_dev.function); + ignore_value(virStrToLong_ui(p+1, &p, 16, &d->pci_dev.domain)); + ignore_value(virStrToLong_ui(p+1, &p, 16, &d->pci_dev.bus)); + ignore_value(virStrToLong_ui(p+1, &p, 16, &d->pci_dev.slot)); + ignore_value(virStrToLong_ui(p+1, &p, 16, &d->pci_dev.function)); } if (!virPCIGetPhysicalFunction(sysfs_path, -- 2.0.0

1 0

[libvirt] [PATCH 0/7] Fix domain state handling when reverting snapshots in PMSUSPENDED state
by Peter Krempa 21 Jul '14

21 Jul '14

Unfortunately qemu behaves strangely so it probably can't be fully verified :/ Peter Krempa (7): event: Add transition reason for PMSUSPENDED state from snapshot ops lib: Add reason for a domain reaching the PMSUSPENDED state qemu: snapshot: Convert if-else switch to switch statement qemu: snapshot: Reject revertion from clearly bad states qemu: snapshot: Add helper to generate lifecycle events qemu: snapshot: Refactor event creation when reverting snapshots qemu: snapshot: Correctly revert snapshots in PMSUSPENDED state examples/object-events/event-test.c | 3 + include/libvirt/libvirt.h.in | 6 + src/conf/domain_conf.c | 3 +- src/qemu/qemu_driver.c | 301 +++++++++++++++++++++++++----------- tools/virsh-domain-monitor.c | 3 +- tools/virsh-domain.c | 3 +- 6 files changed, 228 insertions(+), 91 deletions(-) -- 2.0.0

2 15

[libvirt] [PATCH 0/3] Resolve a few coverity issues
by Peter Krempa 21 Jul '14

21 Jul '14

Mostly introduced by the "domtop" program. Peter Krempa (3): examples: domtop: Fix uninitialized variable use examples: domtop: Avoid leaking memory util: Check return value from virStrToLong* functions examples/domtop/domtop.c | 10 +++------- src/util/virsexpr.c | 6 +++--- src/util/virstring.h | 30 ++++++++++++++++++++---------- src/vbox/vbox_tmpl.c | 4 ++-- src/xen/xs_internal.c | 2 +- src/xenxs/xen_xm.c | 9 +++++++-- 6 files changed, 36 insertions(+), 25 deletions(-) -- 2.0.0

2 5